gdb
[deliverable/binutils-gdb.git] / gdb / doc / gdb.texinfo
1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1998,
3 @c 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
4 @c Free Software Foundation, Inc.
5 @c
6 @c %**start of header
7 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
8 @c of @set vars. However, you can override filename with makeinfo -o.
9 @setfilename gdb.info
10 @c
11 @include gdb-cfg.texi
12 @c
13 @settitle Debugging with @value{GDBN}
14 @setchapternewpage odd
15 @c %**end of header
16
17 @iftex
18 @c @smallbook
19 @c @cropmarks
20 @end iftex
21
22 @finalout
23 @syncodeindex ky cp
24 @syncodeindex tp cp
25
26 @c readline appendices use @vindex, @findex and @ftable,
27 @c annotate.texi and gdbmi use @findex.
28 @syncodeindex vr cp
29 @syncodeindex fn cp
30
31 @c !!set GDB manual's edition---not the same as GDB version!
32 @c This is updated by GNU Press.
33 @set EDITION Ninth
34
35 @c !!set GDB edit command default editor
36 @set EDITOR /bin/ex
37
38 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
39
40 @c This is a dir.info fragment to support semi-automated addition of
41 @c manuals to an info tree.
42 @dircategory Software development
43 @direntry
44 * Gdb: (gdb). The GNU debugger.
45 @end direntry
46
47 @copying
48 Copyright @copyright{} 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996,
49 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
50 Free Software Foundation, Inc.
51
52 Permission is granted to copy, distribute and/or modify this document
53 under the terms of the GNU Free Documentation License, Version 1.3 or
54 any later version published by the Free Software Foundation; with the
55 Invariant Sections being ``Free Software'' and ``Free Software Needs
56 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
57 and with the Back-Cover Texts as in (a) below.
58
59 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
60 this GNU Manual. Buying copies from GNU Press supports the FSF in
61 developing GNU and promoting software freedom.''
62 @end copying
63
64 @ifnottex
65 This file documents the @sc{gnu} debugger @value{GDBN}.
66
67 This is the @value{EDITION} Edition, of @cite{Debugging with
68 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
69 @ifset VERSION_PACKAGE
70 @value{VERSION_PACKAGE}
71 @end ifset
72 Version @value{GDBVN}.
73
74 @insertcopying
75 @end ifnottex
76
77 @titlepage
78 @title Debugging with @value{GDBN}
79 @subtitle The @sc{gnu} Source-Level Debugger
80 @sp 1
81 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
82 @ifset VERSION_PACKAGE
83 @sp 1
84 @subtitle @value{VERSION_PACKAGE}
85 @end ifset
86 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
87 @page
88 @tex
89 {\parskip=0pt
90 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
91 \hfill {\it Debugging with @value{GDBN}}\par
92 \hfill \TeX{}info \texinfoversion\par
93 }
94 @end tex
95
96 @vskip 0pt plus 1filll
97 Published by the Free Software Foundation @*
98 51 Franklin Street, Fifth Floor,
99 Boston, MA 02110-1301, USA@*
100 ISBN 1-882114-77-9 @*
101
102 @insertcopying
103 @page
104 This edition of the GDB manual is dedicated to the memory of Fred
105 Fish. Fred was a long-standing contributor to GDB and to Free
106 software in general. We will miss him.
107 @end titlepage
108 @page
109
110 @ifnottex
111 @node Top, Summary, (dir), (dir)
112
113 @top Debugging with @value{GDBN}
114
115 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
116
117 This is the @value{EDITION} Edition, for @value{GDBN}
118 @ifset VERSION_PACKAGE
119 @value{VERSION_PACKAGE}
120 @end ifset
121 Version @value{GDBVN}.
122
123 Copyright (C) 1988-2010 Free Software Foundation, Inc.
124
125 This edition of the GDB manual is dedicated to the memory of Fred
126 Fish. Fred was a long-standing contributor to GDB and to Free
127 software in general. We will miss him.
128
129 @menu
130 * Summary:: Summary of @value{GDBN}
131 * Sample Session:: A sample @value{GDBN} session
132
133 * Invocation:: Getting in and out of @value{GDBN}
134 * Commands:: @value{GDBN} commands
135 * Running:: Running programs under @value{GDBN}
136 * Stopping:: Stopping and continuing
137 * Reverse Execution:: Running programs backward
138 * Process Record and Replay:: Recording inferior's execution and replaying it
139 * Stack:: Examining the stack
140 * Source:: Examining source files
141 * Data:: Examining data
142 * Optimized Code:: Debugging optimized code
143 * Macros:: Preprocessor Macros
144 * Tracepoints:: Debugging remote targets non-intrusively
145 * Overlays:: Debugging programs that use overlays
146
147 * Languages:: Using @value{GDBN} with different languages
148
149 * Symbols:: Examining the symbol table
150 * Altering:: Altering execution
151 * GDB Files:: @value{GDBN} files
152 * Targets:: Specifying a debugging target
153 * Remote Debugging:: Debugging remote programs
154 * Configurations:: Configuration-specific information
155 * Controlling GDB:: Controlling @value{GDBN}
156 * Extending GDB:: Extending @value{GDBN}
157 * Interpreters:: Command Interpreters
158 * TUI:: @value{GDBN} Text User Interface
159 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
160 * GDB/MI:: @value{GDBN}'s Machine Interface.
161 * Annotations:: @value{GDBN}'s annotation interface.
162 * JIT Interface:: Using the JIT debugging interface.
163
164 * GDB Bugs:: Reporting bugs in @value{GDBN}
165
166 * Command Line Editing:: Command Line Editing
167 * Using History Interactively:: Using History Interactively
168 * Formatting Documentation:: How to format and print @value{GDBN} documentation
169 * Installing GDB:: Installing GDB
170 * Maintenance Commands:: Maintenance Commands
171 * Remote Protocol:: GDB Remote Serial Protocol
172 * Agent Expressions:: The GDB Agent Expression Mechanism
173 * Target Descriptions:: How targets can describe themselves to
174 @value{GDBN}
175 * Operating System Information:: Getting additional information from
176 the operating system
177 * Trace File Format:: GDB trace file format
178 * Copying:: GNU General Public License says
179 how you can copy and share GDB
180 * GNU Free Documentation License:: The license for this documentation
181 * Index:: Index
182 @end menu
183
184 @end ifnottex
185
186 @contents
187
188 @node Summary
189 @unnumbered Summary of @value{GDBN}
190
191 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
192 going on ``inside'' another program while it executes---or what another
193 program was doing at the moment it crashed.
194
195 @value{GDBN} can do four main kinds of things (plus other things in support of
196 these) to help you catch bugs in the act:
197
198 @itemize @bullet
199 @item
200 Start your program, specifying anything that might affect its behavior.
201
202 @item
203 Make your program stop on specified conditions.
204
205 @item
206 Examine what has happened, when your program has stopped.
207
208 @item
209 Change things in your program, so you can experiment with correcting the
210 effects of one bug and go on to learn about another.
211 @end itemize
212
213 You can use @value{GDBN} to debug programs written in C and C@t{++}.
214 For more information, see @ref{Supported Languages,,Supported Languages}.
215 For more information, see @ref{C,,C and C++}.
216
217 Support for D is partial. For information on D, see
218 @ref{D,,D}.
219
220 @cindex Modula-2
221 Support for Modula-2 is partial. For information on Modula-2, see
222 @ref{Modula-2,,Modula-2}.
223
224 @cindex Pascal
225 Debugging Pascal programs which use sets, subranges, file variables, or
226 nested functions does not currently work. @value{GDBN} does not support
227 entering expressions, printing values, or similar features using Pascal
228 syntax.
229
230 @cindex Fortran
231 @value{GDBN} can be used to debug programs written in Fortran, although
232 it may be necessary to refer to some variables with a trailing
233 underscore.
234
235 @value{GDBN} can be used to debug programs written in Objective-C,
236 using either the Apple/NeXT or the GNU Objective-C runtime.
237
238 @menu
239 * Free Software:: Freely redistributable software
240 * Contributors:: Contributors to GDB
241 @end menu
242
243 @node Free Software
244 @unnumberedsec Free Software
245
246 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
247 General Public License
248 (GPL). The GPL gives you the freedom to copy or adapt a licensed
249 program---but every person getting a copy also gets with it the
250 freedom to modify that copy (which means that they must get access to
251 the source code), and the freedom to distribute further copies.
252 Typical software companies use copyrights to limit your freedoms; the
253 Free Software Foundation uses the GPL to preserve these freedoms.
254
255 Fundamentally, the General Public License is a license which says that
256 you have these freedoms and that you cannot take these freedoms away
257 from anyone else.
258
259 @unnumberedsec Free Software Needs Free Documentation
260
261 The biggest deficiency in the free software community today is not in
262 the software---it is the lack of good free documentation that we can
263 include with the free software. Many of our most important
264 programs do not come with free reference manuals and free introductory
265 texts. Documentation is an essential part of any software package;
266 when an important free software package does not come with a free
267 manual and a free tutorial, that is a major gap. We have many such
268 gaps today.
269
270 Consider Perl, for instance. The tutorial manuals that people
271 normally use are non-free. How did this come about? Because the
272 authors of those manuals published them with restrictive terms---no
273 copying, no modification, source files not available---which exclude
274 them from the free software world.
275
276 That wasn't the first time this sort of thing happened, and it was far
277 from the last. Many times we have heard a GNU user eagerly describe a
278 manual that he is writing, his intended contribution to the community,
279 only to learn that he had ruined everything by signing a publication
280 contract to make it non-free.
281
282 Free documentation, like free software, is a matter of freedom, not
283 price. The problem with the non-free manual is not that publishers
284 charge a price for printed copies---that in itself is fine. (The Free
285 Software Foundation sells printed copies of manuals, too.) The
286 problem is the restrictions on the use of the manual. Free manuals
287 are available in source code form, and give you permission to copy and
288 modify. Non-free manuals do not allow this.
289
290 The criteria of freedom for a free manual are roughly the same as for
291 free software. Redistribution (including the normal kinds of
292 commercial redistribution) must be permitted, so that the manual can
293 accompany every copy of the program, both on-line and on paper.
294
295 Permission for modification of the technical content is crucial too.
296 When people modify the software, adding or changing features, if they
297 are conscientious they will change the manual too---so they can
298 provide accurate and clear documentation for the modified program. A
299 manual that leaves you no choice but to write a new manual to document
300 a changed version of the program is not really available to our
301 community.
302
303 Some kinds of limits on the way modification is handled are
304 acceptable. For example, requirements to preserve the original
305 author's copyright notice, the distribution terms, or the list of
306 authors, are ok. It is also no problem to require modified versions
307 to include notice that they were modified. Even entire sections that
308 may not be deleted or changed are acceptable, as long as they deal
309 with nontechnical topics (like this one). These kinds of restrictions
310 are acceptable because they don't obstruct the community's normal use
311 of the manual.
312
313 However, it must be possible to modify all the @emph{technical}
314 content of the manual, and then distribute the result in all the usual
315 media, through all the usual channels. Otherwise, the restrictions
316 obstruct the use of the manual, it is not free, and we need another
317 manual to replace it.
318
319 Please spread the word about this issue. Our community continues to
320 lose manuals to proprietary publishing. If we spread the word that
321 free software needs free reference manuals and free tutorials, perhaps
322 the next person who wants to contribute by writing documentation will
323 realize, before it is too late, that only free manuals contribute to
324 the free software community.
325
326 If you are writing documentation, please insist on publishing it under
327 the GNU Free Documentation License or another free documentation
328 license. Remember that this decision requires your approval---you
329 don't have to let the publisher decide. Some commercial publishers
330 will use a free license if you insist, but they will not propose the
331 option; it is up to you to raise the issue and say firmly that this is
332 what you want. If the publisher you are dealing with refuses, please
333 try other publishers. If you're not sure whether a proposed license
334 is free, write to @email{licensing@@gnu.org}.
335
336 You can encourage commercial publishers to sell more free, copylefted
337 manuals and tutorials by buying them, and particularly by buying
338 copies from the publishers that paid for their writing or for major
339 improvements. Meanwhile, try to avoid buying non-free documentation
340 at all. Check the distribution terms of a manual before you buy it,
341 and insist that whoever seeks your business must respect your freedom.
342 Check the history of the book, and try to reward the publishers that
343 have paid or pay the authors to work on it.
344
345 The Free Software Foundation maintains a list of free documentation
346 published by other publishers, at
347 @url{http://www.fsf.org/doc/other-free-books.html}.
348
349 @node Contributors
350 @unnumberedsec Contributors to @value{GDBN}
351
352 Richard Stallman was the original author of @value{GDBN}, and of many
353 other @sc{gnu} programs. Many others have contributed to its
354 development. This section attempts to credit major contributors. One
355 of the virtues of free software is that everyone is free to contribute
356 to it; with regret, we cannot actually acknowledge everyone here. The
357 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
358 blow-by-blow account.
359
360 Changes much prior to version 2.0 are lost in the mists of time.
361
362 @quotation
363 @emph{Plea:} Additions to this section are particularly welcome. If you
364 or your friends (or enemies, to be evenhanded) have been unfairly
365 omitted from this list, we would like to add your names!
366 @end quotation
367
368 So that they may not regard their many labors as thankless, we
369 particularly thank those who shepherded @value{GDBN} through major
370 releases:
371 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
372 Jim Blandy (release 4.18);
373 Jason Molenda (release 4.17);
374 Stan Shebs (release 4.14);
375 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
376 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
377 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
378 Jim Kingdon (releases 3.5, 3.4, and 3.3);
379 and Randy Smith (releases 3.2, 3.1, and 3.0).
380
381 Richard Stallman, assisted at various times by Peter TerMaat, Chris
382 Hanson, and Richard Mlynarik, handled releases through 2.8.
383
384 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
385 in @value{GDBN}, with significant additional contributions from Per
386 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
387 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
388 much general update work leading to release 3.0).
389
390 @value{GDBN} uses the BFD subroutine library to examine multiple
391 object-file formats; BFD was a joint project of David V.
392 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
393
394 David Johnson wrote the original COFF support; Pace Willison did
395 the original support for encapsulated COFF.
396
397 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
398
399 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
400 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
401 support.
402 Jean-Daniel Fekete contributed Sun 386i support.
403 Chris Hanson improved the HP9000 support.
404 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
405 David Johnson contributed Encore Umax support.
406 Jyrki Kuoppala contributed Altos 3068 support.
407 Jeff Law contributed HP PA and SOM support.
408 Keith Packard contributed NS32K support.
409 Doug Rabson contributed Acorn Risc Machine support.
410 Bob Rusk contributed Harris Nighthawk CX-UX support.
411 Chris Smith contributed Convex support (and Fortran debugging).
412 Jonathan Stone contributed Pyramid support.
413 Michael Tiemann contributed SPARC support.
414 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
415 Pace Willison contributed Intel 386 support.
416 Jay Vosburgh contributed Symmetry support.
417 Marko Mlinar contributed OpenRISC 1000 support.
418
419 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
420
421 Rich Schaefer and Peter Schauer helped with support of SunOS shared
422 libraries.
423
424 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
425 about several machine instruction sets.
426
427 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
428 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
429 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
430 and RDI targets, respectively.
431
432 Brian Fox is the author of the readline libraries providing
433 command-line editing and command history.
434
435 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
436 Modula-2 support, and contributed the Languages chapter of this manual.
437
438 Fred Fish wrote most of the support for Unix System Vr4.
439 He also enhanced the command-completion support to cover C@t{++} overloaded
440 symbols.
441
442 Hitachi America (now Renesas America), Ltd. sponsored the support for
443 H8/300, H8/500, and Super-H processors.
444
445 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
446
447 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
448 processors.
449
450 Toshiba sponsored the support for the TX39 Mips processor.
451
452 Matsushita sponsored the support for the MN10200 and MN10300 processors.
453
454 Fujitsu sponsored the support for SPARClite and FR30 processors.
455
456 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
457 watchpoints.
458
459 Michael Snyder added support for tracepoints.
460
461 Stu Grossman wrote gdbserver.
462
463 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
464 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
465
466 The following people at the Hewlett-Packard Company contributed
467 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
468 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
469 compiler, and the Text User Interface (nee Terminal User Interface):
470 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
471 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
472 provided HP-specific information in this manual.
473
474 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
475 Robert Hoehne made significant contributions to the DJGPP port.
476
477 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
478 development since 1991. Cygnus engineers who have worked on @value{GDBN}
479 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
480 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
481 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
482 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
483 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
484 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
485 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
486 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
487 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
488 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
489 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
490 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
491 Zuhn have made contributions both large and small.
492
493 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
494 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
495
496 Jim Blandy added support for preprocessor macros, while working for Red
497 Hat.
498
499 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
500 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
501 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
502 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
503 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
504 with the migration of old architectures to this new framework.
505
506 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
507 unwinder framework, this consisting of a fresh new design featuring
508 frame IDs, independent frame sniffers, and the sentinel frame. Mark
509 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
510 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
511 trad unwinders. The architecture-specific changes, each involving a
512 complete rewrite of the architecture's frame code, were carried out by
513 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
514 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
515 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
516 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
517 Weigand.
518
519 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
520 Tensilica, Inc.@: contributed support for Xtensa processors. Others
521 who have worked on the Xtensa port of @value{GDBN} in the past include
522 Steve Tjiang, John Newlin, and Scott Foehner.
523
524 Michael Eager and staff of Xilinx, Inc., contributed support for the
525 Xilinx MicroBlaze architecture.
526
527 @node Sample Session
528 @chapter A Sample @value{GDBN} Session
529
530 You can use this manual at your leisure to read all about @value{GDBN}.
531 However, a handful of commands are enough to get started using the
532 debugger. This chapter illustrates those commands.
533
534 @iftex
535 In this sample session, we emphasize user input like this: @b{input},
536 to make it easier to pick out from the surrounding output.
537 @end iftex
538
539 @c FIXME: this example may not be appropriate for some configs, where
540 @c FIXME...primary interest is in remote use.
541
542 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
543 processor) exhibits the following bug: sometimes, when we change its
544 quote strings from the default, the commands used to capture one macro
545 definition within another stop working. In the following short @code{m4}
546 session, we define a macro @code{foo} which expands to @code{0000}; we
547 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
548 same thing. However, when we change the open quote string to
549 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
550 procedure fails to define a new synonym @code{baz}:
551
552 @smallexample
553 $ @b{cd gnu/m4}
554 $ @b{./m4}
555 @b{define(foo,0000)}
556
557 @b{foo}
558 0000
559 @b{define(bar,defn(`foo'))}
560
561 @b{bar}
562 0000
563 @b{changequote(<QUOTE>,<UNQUOTE>)}
564
565 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
566 @b{baz}
567 @b{Ctrl-d}
568 m4: End of input: 0: fatal error: EOF in string
569 @end smallexample
570
571 @noindent
572 Let us use @value{GDBN} to try to see what is going on.
573
574 @smallexample
575 $ @b{@value{GDBP} m4}
576 @c FIXME: this falsifies the exact text played out, to permit smallbook
577 @c FIXME... format to come out better.
578 @value{GDBN} is free software and you are welcome to distribute copies
579 of it under certain conditions; type "show copying" to see
580 the conditions.
581 There is absolutely no warranty for @value{GDBN}; type "show warranty"
582 for details.
583
584 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
585 (@value{GDBP})
586 @end smallexample
587
588 @noindent
589 @value{GDBN} reads only enough symbol data to know where to find the
590 rest when needed; as a result, the first prompt comes up very quickly.
591 We now tell @value{GDBN} to use a narrower display width than usual, so
592 that examples fit in this manual.
593
594 @smallexample
595 (@value{GDBP}) @b{set width 70}
596 @end smallexample
597
598 @noindent
599 We need to see how the @code{m4} built-in @code{changequote} works.
600 Having looked at the source, we know the relevant subroutine is
601 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
602 @code{break} command.
603
604 @smallexample
605 (@value{GDBP}) @b{break m4_changequote}
606 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
607 @end smallexample
608
609 @noindent
610 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
611 control; as long as control does not reach the @code{m4_changequote}
612 subroutine, the program runs as usual:
613
614 @smallexample
615 (@value{GDBP}) @b{run}
616 Starting program: /work/Editorial/gdb/gnu/m4/m4
617 @b{define(foo,0000)}
618
619 @b{foo}
620 0000
621 @end smallexample
622
623 @noindent
624 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
625 suspends execution of @code{m4}, displaying information about the
626 context where it stops.
627
628 @smallexample
629 @b{changequote(<QUOTE>,<UNQUOTE>)}
630
631 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
632 at builtin.c:879
633 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
634 @end smallexample
635
636 @noindent
637 Now we use the command @code{n} (@code{next}) to advance execution to
638 the next line of the current function.
639
640 @smallexample
641 (@value{GDBP}) @b{n}
642 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
643 : nil,
644 @end smallexample
645
646 @noindent
647 @code{set_quotes} looks like a promising subroutine. We can go into it
648 by using the command @code{s} (@code{step}) instead of @code{next}.
649 @code{step} goes to the next line to be executed in @emph{any}
650 subroutine, so it steps into @code{set_quotes}.
651
652 @smallexample
653 (@value{GDBP}) @b{s}
654 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
655 at input.c:530
656 530 if (lquote != def_lquote)
657 @end smallexample
658
659 @noindent
660 The display that shows the subroutine where @code{m4} is now
661 suspended (and its arguments) is called a stack frame display. It
662 shows a summary of the stack. We can use the @code{backtrace}
663 command (which can also be spelled @code{bt}), to see where we are
664 in the stack as a whole: the @code{backtrace} command displays a
665 stack frame for each active subroutine.
666
667 @smallexample
668 (@value{GDBP}) @b{bt}
669 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
670 at input.c:530
671 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
672 at builtin.c:882
673 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
674 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
675 at macro.c:71
676 #4 0x79dc in expand_input () at macro.c:40
677 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
678 @end smallexample
679
680 @noindent
681 We step through a few more lines to see what happens. The first two
682 times, we can use @samp{s}; the next two times we use @code{n} to avoid
683 falling into the @code{xstrdup} subroutine.
684
685 @smallexample
686 (@value{GDBP}) @b{s}
687 0x3b5c 532 if (rquote != def_rquote)
688 (@value{GDBP}) @b{s}
689 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
690 def_lquote : xstrdup(lq);
691 (@value{GDBP}) @b{n}
692 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
693 : xstrdup(rq);
694 (@value{GDBP}) @b{n}
695 538 len_lquote = strlen(rquote);
696 @end smallexample
697
698 @noindent
699 The last line displayed looks a little odd; we can examine the variables
700 @code{lquote} and @code{rquote} to see if they are in fact the new left
701 and right quotes we specified. We use the command @code{p}
702 (@code{print}) to see their values.
703
704 @smallexample
705 (@value{GDBP}) @b{p lquote}
706 $1 = 0x35d40 "<QUOTE>"
707 (@value{GDBP}) @b{p rquote}
708 $2 = 0x35d50 "<UNQUOTE>"
709 @end smallexample
710
711 @noindent
712 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
713 To look at some context, we can display ten lines of source
714 surrounding the current line with the @code{l} (@code{list}) command.
715
716 @smallexample
717 (@value{GDBP}) @b{l}
718 533 xfree(rquote);
719 534
720 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
721 : xstrdup (lq);
722 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
723 : xstrdup (rq);
724 537
725 538 len_lquote = strlen(rquote);
726 539 len_rquote = strlen(lquote);
727 540 @}
728 541
729 542 void
730 @end smallexample
731
732 @noindent
733 Let us step past the two lines that set @code{len_lquote} and
734 @code{len_rquote}, and then examine the values of those variables.
735
736 @smallexample
737 (@value{GDBP}) @b{n}
738 539 len_rquote = strlen(lquote);
739 (@value{GDBP}) @b{n}
740 540 @}
741 (@value{GDBP}) @b{p len_lquote}
742 $3 = 9
743 (@value{GDBP}) @b{p len_rquote}
744 $4 = 7
745 @end smallexample
746
747 @noindent
748 That certainly looks wrong, assuming @code{len_lquote} and
749 @code{len_rquote} are meant to be the lengths of @code{lquote} and
750 @code{rquote} respectively. We can set them to better values using
751 the @code{p} command, since it can print the value of
752 any expression---and that expression can include subroutine calls and
753 assignments.
754
755 @smallexample
756 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
757 $5 = 7
758 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
759 $6 = 9
760 @end smallexample
761
762 @noindent
763 Is that enough to fix the problem of using the new quotes with the
764 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
765 executing with the @code{c} (@code{continue}) command, and then try the
766 example that caused trouble initially:
767
768 @smallexample
769 (@value{GDBP}) @b{c}
770 Continuing.
771
772 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
773
774 baz
775 0000
776 @end smallexample
777
778 @noindent
779 Success! The new quotes now work just as well as the default ones. The
780 problem seems to have been just the two typos defining the wrong
781 lengths. We allow @code{m4} exit by giving it an EOF as input:
782
783 @smallexample
784 @b{Ctrl-d}
785 Program exited normally.
786 @end smallexample
787
788 @noindent
789 The message @samp{Program exited normally.} is from @value{GDBN}; it
790 indicates @code{m4} has finished executing. We can end our @value{GDBN}
791 session with the @value{GDBN} @code{quit} command.
792
793 @smallexample
794 (@value{GDBP}) @b{quit}
795 @end smallexample
796
797 @node Invocation
798 @chapter Getting In and Out of @value{GDBN}
799
800 This chapter discusses how to start @value{GDBN}, and how to get out of it.
801 The essentials are:
802 @itemize @bullet
803 @item
804 type @samp{@value{GDBP}} to start @value{GDBN}.
805 @item
806 type @kbd{quit} or @kbd{Ctrl-d} to exit.
807 @end itemize
808
809 @menu
810 * Invoking GDB:: How to start @value{GDBN}
811 * Quitting GDB:: How to quit @value{GDBN}
812 * Shell Commands:: How to use shell commands inside @value{GDBN}
813 * Logging Output:: How to log @value{GDBN}'s output to a file
814 @end menu
815
816 @node Invoking GDB
817 @section Invoking @value{GDBN}
818
819 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
820 @value{GDBN} reads commands from the terminal until you tell it to exit.
821
822 You can also run @code{@value{GDBP}} with a variety of arguments and options,
823 to specify more of your debugging environment at the outset.
824
825 The command-line options described here are designed
826 to cover a variety of situations; in some environments, some of these
827 options may effectively be unavailable.
828
829 The most usual way to start @value{GDBN} is with one argument,
830 specifying an executable program:
831
832 @smallexample
833 @value{GDBP} @var{program}
834 @end smallexample
835
836 @noindent
837 You can also start with both an executable program and a core file
838 specified:
839
840 @smallexample
841 @value{GDBP} @var{program} @var{core}
842 @end smallexample
843
844 You can, instead, specify a process ID as a second argument, if you want
845 to debug a running process:
846
847 @smallexample
848 @value{GDBP} @var{program} 1234
849 @end smallexample
850
851 @noindent
852 would attach @value{GDBN} to process @code{1234} (unless you also have a file
853 named @file{1234}; @value{GDBN} does check for a core file first).
854
855 Taking advantage of the second command-line argument requires a fairly
856 complete operating system; when you use @value{GDBN} as a remote
857 debugger attached to a bare board, there may not be any notion of
858 ``process'', and there is often no way to get a core dump. @value{GDBN}
859 will warn you if it is unable to attach or to read core dumps.
860
861 You can optionally have @code{@value{GDBP}} pass any arguments after the
862 executable file to the inferior using @code{--args}. This option stops
863 option processing.
864 @smallexample
865 @value{GDBP} --args gcc -O2 -c foo.c
866 @end smallexample
867 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
868 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
869
870 You can run @code{@value{GDBP}} without printing the front material, which describes
871 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
872
873 @smallexample
874 @value{GDBP} -silent
875 @end smallexample
876
877 @noindent
878 You can further control how @value{GDBN} starts up by using command-line
879 options. @value{GDBN} itself can remind you of the options available.
880
881 @noindent
882 Type
883
884 @smallexample
885 @value{GDBP} -help
886 @end smallexample
887
888 @noindent
889 to display all available options and briefly describe their use
890 (@samp{@value{GDBP} -h} is a shorter equivalent).
891
892 All options and command line arguments you give are processed
893 in sequential order. The order makes a difference when the
894 @samp{-x} option is used.
895
896
897 @menu
898 * File Options:: Choosing files
899 * Mode Options:: Choosing modes
900 * Startup:: What @value{GDBN} does during startup
901 @end menu
902
903 @node File Options
904 @subsection Choosing Files
905
906 When @value{GDBN} starts, it reads any arguments other than options as
907 specifying an executable file and core file (or process ID). This is
908 the same as if the arguments were specified by the @samp{-se} and
909 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
910 first argument that does not have an associated option flag as
911 equivalent to the @samp{-se} option followed by that argument; and the
912 second argument that does not have an associated option flag, if any, as
913 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
914 If the second argument begins with a decimal digit, @value{GDBN} will
915 first attempt to attach to it as a process, and if that fails, attempt
916 to open it as a corefile. If you have a corefile whose name begins with
917 a digit, you can prevent @value{GDBN} from treating it as a pid by
918 prefixing it with @file{./}, e.g.@: @file{./12345}.
919
920 If @value{GDBN} has not been configured to included core file support,
921 such as for most embedded targets, then it will complain about a second
922 argument and ignore it.
923
924 Many options have both long and short forms; both are shown in the
925 following list. @value{GDBN} also recognizes the long forms if you truncate
926 them, so long as enough of the option is present to be unambiguous.
927 (If you prefer, you can flag option arguments with @samp{--} rather
928 than @samp{-}, though we illustrate the more usual convention.)
929
930 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
931 @c way, both those who look for -foo and --foo in the index, will find
932 @c it.
933
934 @table @code
935 @item -symbols @var{file}
936 @itemx -s @var{file}
937 @cindex @code{--symbols}
938 @cindex @code{-s}
939 Read symbol table from file @var{file}.
940
941 @item -exec @var{file}
942 @itemx -e @var{file}
943 @cindex @code{--exec}
944 @cindex @code{-e}
945 Use file @var{file} as the executable file to execute when appropriate,
946 and for examining pure data in conjunction with a core dump.
947
948 @item -se @var{file}
949 @cindex @code{--se}
950 Read symbol table from file @var{file} and use it as the executable
951 file.
952
953 @item -core @var{file}
954 @itemx -c @var{file}
955 @cindex @code{--core}
956 @cindex @code{-c}
957 Use file @var{file} as a core dump to examine.
958
959 @item -pid @var{number}
960 @itemx -p @var{number}
961 @cindex @code{--pid}
962 @cindex @code{-p}
963 Connect to process ID @var{number}, as with the @code{attach} command.
964
965 @item -command @var{file}
966 @itemx -x @var{file}
967 @cindex @code{--command}
968 @cindex @code{-x}
969 Execute commands from file @var{file}. The contents of this file is
970 evaluated exactly as the @code{source} command would.
971 @xref{Command Files,, Command files}.
972
973 @item -eval-command @var{command}
974 @itemx -ex @var{command}
975 @cindex @code{--eval-command}
976 @cindex @code{-ex}
977 Execute a single @value{GDBN} command.
978
979 This option may be used multiple times to call multiple commands. It may
980 also be interleaved with @samp{-command} as required.
981
982 @smallexample
983 @value{GDBP} -ex 'target sim' -ex 'load' \
984 -x setbreakpoints -ex 'run' a.out
985 @end smallexample
986
987 @item -directory @var{directory}
988 @itemx -d @var{directory}
989 @cindex @code{--directory}
990 @cindex @code{-d}
991 Add @var{directory} to the path to search for source and script files.
992
993 @item -r
994 @itemx -readnow
995 @cindex @code{--readnow}
996 @cindex @code{-r}
997 Read each symbol file's entire symbol table immediately, rather than
998 the default, which is to read it incrementally as it is needed.
999 This makes startup slower, but makes future operations faster.
1000
1001 @end table
1002
1003 @node Mode Options
1004 @subsection Choosing Modes
1005
1006 You can run @value{GDBN} in various alternative modes---for example, in
1007 batch mode or quiet mode.
1008
1009 @table @code
1010 @item -nx
1011 @itemx -n
1012 @cindex @code{--nx}
1013 @cindex @code{-n}
1014 Do not execute commands found in any initialization files. Normally,
1015 @value{GDBN} executes the commands in these files after all the command
1016 options and arguments have been processed. @xref{Command Files,,Command
1017 Files}.
1018
1019 @item -quiet
1020 @itemx -silent
1021 @itemx -q
1022 @cindex @code{--quiet}
1023 @cindex @code{--silent}
1024 @cindex @code{-q}
1025 ``Quiet''. Do not print the introductory and copyright messages. These
1026 messages are also suppressed in batch mode.
1027
1028 @item -batch
1029 @cindex @code{--batch}
1030 Run in batch mode. Exit with status @code{0} after processing all the
1031 command files specified with @samp{-x} (and all commands from
1032 initialization files, if not inhibited with @samp{-n}). Exit with
1033 nonzero status if an error occurs in executing the @value{GDBN} commands
1034 in the command files. Batch mode also disables pagination, sets unlimited
1035 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1036 off} were in effect (@pxref{Messages/Warnings}).
1037
1038 Batch mode may be useful for running @value{GDBN} as a filter, for
1039 example to download and run a program on another computer; in order to
1040 make this more useful, the message
1041
1042 @smallexample
1043 Program exited normally.
1044 @end smallexample
1045
1046 @noindent
1047 (which is ordinarily issued whenever a program running under
1048 @value{GDBN} control terminates) is not issued when running in batch
1049 mode.
1050
1051 @item -batch-silent
1052 @cindex @code{--batch-silent}
1053 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1054 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1055 unaffected). This is much quieter than @samp{-silent} and would be useless
1056 for an interactive session.
1057
1058 This is particularly useful when using targets that give @samp{Loading section}
1059 messages, for example.
1060
1061 Note that targets that give their output via @value{GDBN}, as opposed to
1062 writing directly to @code{stdout}, will also be made silent.
1063
1064 @item -return-child-result
1065 @cindex @code{--return-child-result}
1066 The return code from @value{GDBN} will be the return code from the child
1067 process (the process being debugged), with the following exceptions:
1068
1069 @itemize @bullet
1070 @item
1071 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1072 internal error. In this case the exit code is the same as it would have been
1073 without @samp{-return-child-result}.
1074 @item
1075 The user quits with an explicit value. E.g., @samp{quit 1}.
1076 @item
1077 The child process never runs, or is not allowed to terminate, in which case
1078 the exit code will be -1.
1079 @end itemize
1080
1081 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1082 when @value{GDBN} is being used as a remote program loader or simulator
1083 interface.
1084
1085 @item -nowindows
1086 @itemx -nw
1087 @cindex @code{--nowindows}
1088 @cindex @code{-nw}
1089 ``No windows''. If @value{GDBN} comes with a graphical user interface
1090 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1091 interface. If no GUI is available, this option has no effect.
1092
1093 @item -windows
1094 @itemx -w
1095 @cindex @code{--windows}
1096 @cindex @code{-w}
1097 If @value{GDBN} includes a GUI, then this option requires it to be
1098 used if possible.
1099
1100 @item -cd @var{directory}
1101 @cindex @code{--cd}
1102 Run @value{GDBN} using @var{directory} as its working directory,
1103 instead of the current directory.
1104
1105 @item -fullname
1106 @itemx -f
1107 @cindex @code{--fullname}
1108 @cindex @code{-f}
1109 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1110 subprocess. It tells @value{GDBN} to output the full file name and line
1111 number in a standard, recognizable fashion each time a stack frame is
1112 displayed (which includes each time your program stops). This
1113 recognizable format looks like two @samp{\032} characters, followed by
1114 the file name, line number and character position separated by colons,
1115 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1116 @samp{\032} characters as a signal to display the source code for the
1117 frame.
1118
1119 @item -epoch
1120 @cindex @code{--epoch}
1121 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1122 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1123 routines so as to allow Epoch to display values of expressions in a
1124 separate window.
1125
1126 @item -annotate @var{level}
1127 @cindex @code{--annotate}
1128 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1129 effect is identical to using @samp{set annotate @var{level}}
1130 (@pxref{Annotations}). The annotation @var{level} controls how much
1131 information @value{GDBN} prints together with its prompt, values of
1132 expressions, source lines, and other types of output. Level 0 is the
1133 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1134 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1135 that control @value{GDBN}, and level 2 has been deprecated.
1136
1137 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1138 (@pxref{GDB/MI}).
1139
1140 @item --args
1141 @cindex @code{--args}
1142 Change interpretation of command line so that arguments following the
1143 executable file are passed as command line arguments to the inferior.
1144 This option stops option processing.
1145
1146 @item -baud @var{bps}
1147 @itemx -b @var{bps}
1148 @cindex @code{--baud}
1149 @cindex @code{-b}
1150 Set the line speed (baud rate or bits per second) of any serial
1151 interface used by @value{GDBN} for remote debugging.
1152
1153 @item -l @var{timeout}
1154 @cindex @code{-l}
1155 Set the timeout (in seconds) of any communication used by @value{GDBN}
1156 for remote debugging.
1157
1158 @item -tty @var{device}
1159 @itemx -t @var{device}
1160 @cindex @code{--tty}
1161 @cindex @code{-t}
1162 Run using @var{device} for your program's standard input and output.
1163 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1164
1165 @c resolve the situation of these eventually
1166 @item -tui
1167 @cindex @code{--tui}
1168 Activate the @dfn{Text User Interface} when starting. The Text User
1169 Interface manages several text windows on the terminal, showing
1170 source, assembly, registers and @value{GDBN} command outputs
1171 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Alternatively, the
1172 Text User Interface can be enabled by invoking the program
1173 @samp{@value{GDBTUI}}. Do not use this option if you run @value{GDBN} from
1174 Emacs (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1175
1176 @c @item -xdb
1177 @c @cindex @code{--xdb}
1178 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1179 @c For information, see the file @file{xdb_trans.html}, which is usually
1180 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1181 @c systems.
1182
1183 @item -interpreter @var{interp}
1184 @cindex @code{--interpreter}
1185 Use the interpreter @var{interp} for interface with the controlling
1186 program or device. This option is meant to be set by programs which
1187 communicate with @value{GDBN} using it as a back end.
1188 @xref{Interpreters, , Command Interpreters}.
1189
1190 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1191 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1192 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1193 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1194 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1195 @sc{gdb/mi} interfaces are no longer supported.
1196
1197 @item -write
1198 @cindex @code{--write}
1199 Open the executable and core files for both reading and writing. This
1200 is equivalent to the @samp{set write on} command inside @value{GDBN}
1201 (@pxref{Patching}).
1202
1203 @item -statistics
1204 @cindex @code{--statistics}
1205 This option causes @value{GDBN} to print statistics about time and
1206 memory usage after it completes each command and returns to the prompt.
1207
1208 @item -version
1209 @cindex @code{--version}
1210 This option causes @value{GDBN} to print its version number and
1211 no-warranty blurb, and exit.
1212
1213 @end table
1214
1215 @node Startup
1216 @subsection What @value{GDBN} Does During Startup
1217 @cindex @value{GDBN} startup
1218
1219 Here's the description of what @value{GDBN} does during session startup:
1220
1221 @enumerate
1222 @item
1223 Sets up the command interpreter as specified by the command line
1224 (@pxref{Mode Options, interpreter}).
1225
1226 @item
1227 @cindex init file
1228 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1229 used when building @value{GDBN}; @pxref{System-wide configuration,
1230 ,System-wide configuration and settings}) and executes all the commands in
1231 that file.
1232
1233 @item
1234 Reads the init file (if any) in your home directory@footnote{On
1235 DOS/Windows systems, the home directory is the one pointed to by the
1236 @code{HOME} environment variable.} and executes all the commands in
1237 that file.
1238
1239 @item
1240 Processes command line options and operands.
1241
1242 @item
1243 Reads and executes the commands from init file (if any) in the current
1244 working directory. This is only done if the current directory is
1245 different from your home directory. Thus, you can have more than one
1246 init file, one generic in your home directory, and another, specific
1247 to the program you are debugging, in the directory where you invoke
1248 @value{GDBN}.
1249
1250 @item
1251 Reads command files specified by the @samp{-x} option. @xref{Command
1252 Files}, for more details about @value{GDBN} command files.
1253
1254 @item
1255 Reads the command history recorded in the @dfn{history file}.
1256 @xref{Command History}, for more details about the command history and the
1257 files where @value{GDBN} records it.
1258 @end enumerate
1259
1260 Init files use the same syntax as @dfn{command files} (@pxref{Command
1261 Files}) and are processed by @value{GDBN} in the same way. The init
1262 file in your home directory can set options (such as @samp{set
1263 complaints}) that affect subsequent processing of command line options
1264 and operands. Init files are not executed if you use the @samp{-nx}
1265 option (@pxref{Mode Options, ,Choosing Modes}).
1266
1267 To display the list of init files loaded by gdb at startup, you
1268 can use @kbd{gdb --help}.
1269
1270 @cindex init file name
1271 @cindex @file{.gdbinit}
1272 @cindex @file{gdb.ini}
1273 The @value{GDBN} init files are normally called @file{.gdbinit}.
1274 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1275 the limitations of file names imposed by DOS filesystems. The Windows
1276 ports of @value{GDBN} use the standard name, but if they find a
1277 @file{gdb.ini} file, they warn you about that and suggest to rename
1278 the file to the standard name.
1279
1280
1281 @node Quitting GDB
1282 @section Quitting @value{GDBN}
1283 @cindex exiting @value{GDBN}
1284 @cindex leaving @value{GDBN}
1285
1286 @table @code
1287 @kindex quit @r{[}@var{expression}@r{]}
1288 @kindex q @r{(@code{quit})}
1289 @item quit @r{[}@var{expression}@r{]}
1290 @itemx q
1291 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1292 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1293 do not supply @var{expression}, @value{GDBN} will terminate normally;
1294 otherwise it will terminate using the result of @var{expression} as the
1295 error code.
1296 @end table
1297
1298 @cindex interrupt
1299 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1300 terminates the action of any @value{GDBN} command that is in progress and
1301 returns to @value{GDBN} command level. It is safe to type the interrupt
1302 character at any time because @value{GDBN} does not allow it to take effect
1303 until a time when it is safe.
1304
1305 If you have been using @value{GDBN} to control an attached process or
1306 device, you can release it with the @code{detach} command
1307 (@pxref{Attach, ,Debugging an Already-running Process}).
1308
1309 @node Shell Commands
1310 @section Shell Commands
1311
1312 If you need to execute occasional shell commands during your
1313 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1314 just use the @code{shell} command.
1315
1316 @table @code
1317 @kindex shell
1318 @cindex shell escape
1319 @item shell @var{command string}
1320 Invoke a standard shell to execute @var{command string}.
1321 If it exists, the environment variable @code{SHELL} determines which
1322 shell to run. Otherwise @value{GDBN} uses the default shell
1323 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1324 @end table
1325
1326 The utility @code{make} is often needed in development environments.
1327 You do not have to use the @code{shell} command for this purpose in
1328 @value{GDBN}:
1329
1330 @table @code
1331 @kindex make
1332 @cindex calling make
1333 @item make @var{make-args}
1334 Execute the @code{make} program with the specified
1335 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1336 @end table
1337
1338 @node Logging Output
1339 @section Logging Output
1340 @cindex logging @value{GDBN} output
1341 @cindex save @value{GDBN} output to a file
1342
1343 You may want to save the output of @value{GDBN} commands to a file.
1344 There are several commands to control @value{GDBN}'s logging.
1345
1346 @table @code
1347 @kindex set logging
1348 @item set logging on
1349 Enable logging.
1350 @item set logging off
1351 Disable logging.
1352 @cindex logging file name
1353 @item set logging file @var{file}
1354 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1355 @item set logging overwrite [on|off]
1356 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1357 you want @code{set logging on} to overwrite the logfile instead.
1358 @item set logging redirect [on|off]
1359 By default, @value{GDBN} output will go to both the terminal and the logfile.
1360 Set @code{redirect} if you want output to go only to the log file.
1361 @kindex show logging
1362 @item show logging
1363 Show the current values of the logging settings.
1364 @end table
1365
1366 @node Commands
1367 @chapter @value{GDBN} Commands
1368
1369 You can abbreviate a @value{GDBN} command to the first few letters of the command
1370 name, if that abbreviation is unambiguous; and you can repeat certain
1371 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1372 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1373 show you the alternatives available, if there is more than one possibility).
1374
1375 @menu
1376 * Command Syntax:: How to give commands to @value{GDBN}
1377 * Completion:: Command completion
1378 * Help:: How to ask @value{GDBN} for help
1379 @end menu
1380
1381 @node Command Syntax
1382 @section Command Syntax
1383
1384 A @value{GDBN} command is a single line of input. There is no limit on
1385 how long it can be. It starts with a command name, which is followed by
1386 arguments whose meaning depends on the command name. For example, the
1387 command @code{step} accepts an argument which is the number of times to
1388 step, as in @samp{step 5}. You can also use the @code{step} command
1389 with no arguments. Some commands do not allow any arguments.
1390
1391 @cindex abbreviation
1392 @value{GDBN} command names may always be truncated if that abbreviation is
1393 unambiguous. Other possible command abbreviations are listed in the
1394 documentation for individual commands. In some cases, even ambiguous
1395 abbreviations are allowed; for example, @code{s} is specially defined as
1396 equivalent to @code{step} even though there are other commands whose
1397 names start with @code{s}. You can test abbreviations by using them as
1398 arguments to the @code{help} command.
1399
1400 @cindex repeating commands
1401 @kindex RET @r{(repeat last command)}
1402 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1403 repeat the previous command. Certain commands (for example, @code{run})
1404 will not repeat this way; these are commands whose unintentional
1405 repetition might cause trouble and which you are unlikely to want to
1406 repeat. User-defined commands can disable this feature; see
1407 @ref{Define, dont-repeat}.
1408
1409 The @code{list} and @code{x} commands, when you repeat them with
1410 @key{RET}, construct new arguments rather than repeating
1411 exactly as typed. This permits easy scanning of source or memory.
1412
1413 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1414 output, in a way similar to the common utility @code{more}
1415 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1416 @key{RET} too many in this situation, @value{GDBN} disables command
1417 repetition after any command that generates this sort of display.
1418
1419 @kindex # @r{(a comment)}
1420 @cindex comment
1421 Any text from a @kbd{#} to the end of the line is a comment; it does
1422 nothing. This is useful mainly in command files (@pxref{Command
1423 Files,,Command Files}).
1424
1425 @cindex repeating command sequences
1426 @kindex Ctrl-o @r{(operate-and-get-next)}
1427 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1428 commands. This command accepts the current line, like @key{RET}, and
1429 then fetches the next line relative to the current line from the history
1430 for editing.
1431
1432 @node Completion
1433 @section Command Completion
1434
1435 @cindex completion
1436 @cindex word completion
1437 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1438 only one possibility; it can also show you what the valid possibilities
1439 are for the next word in a command, at any time. This works for @value{GDBN}
1440 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1441
1442 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1443 of a word. If there is only one possibility, @value{GDBN} fills in the
1444 word, and waits for you to finish the command (or press @key{RET} to
1445 enter it). For example, if you type
1446
1447 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1448 @c complete accuracy in these examples; space introduced for clarity.
1449 @c If texinfo enhancements make it unnecessary, it would be nice to
1450 @c replace " @key" by "@key" in the following...
1451 @smallexample
1452 (@value{GDBP}) info bre @key{TAB}
1453 @end smallexample
1454
1455 @noindent
1456 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1457 the only @code{info} subcommand beginning with @samp{bre}:
1458
1459 @smallexample
1460 (@value{GDBP}) info breakpoints
1461 @end smallexample
1462
1463 @noindent
1464 You can either press @key{RET} at this point, to run the @code{info
1465 breakpoints} command, or backspace and enter something else, if
1466 @samp{breakpoints} does not look like the command you expected. (If you
1467 were sure you wanted @code{info breakpoints} in the first place, you
1468 might as well just type @key{RET} immediately after @samp{info bre},
1469 to exploit command abbreviations rather than command completion).
1470
1471 If there is more than one possibility for the next word when you press
1472 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1473 characters and try again, or just press @key{TAB} a second time;
1474 @value{GDBN} displays all the possible completions for that word. For
1475 example, you might want to set a breakpoint on a subroutine whose name
1476 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1477 just sounds the bell. Typing @key{TAB} again displays all the
1478 function names in your program that begin with those characters, for
1479 example:
1480
1481 @smallexample
1482 (@value{GDBP}) b make_ @key{TAB}
1483 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1484 make_a_section_from_file make_environ
1485 make_abs_section make_function_type
1486 make_blockvector make_pointer_type
1487 make_cleanup make_reference_type
1488 make_command make_symbol_completion_list
1489 (@value{GDBP}) b make_
1490 @end smallexample
1491
1492 @noindent
1493 After displaying the available possibilities, @value{GDBN} copies your
1494 partial input (@samp{b make_} in the example) so you can finish the
1495 command.
1496
1497 If you just want to see the list of alternatives in the first place, you
1498 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1499 means @kbd{@key{META} ?}. You can type this either by holding down a
1500 key designated as the @key{META} shift on your keyboard (if there is
1501 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1502
1503 @cindex quotes in commands
1504 @cindex completion of quoted strings
1505 Sometimes the string you need, while logically a ``word'', may contain
1506 parentheses or other characters that @value{GDBN} normally excludes from
1507 its notion of a word. To permit word completion to work in this
1508 situation, you may enclose words in @code{'} (single quote marks) in
1509 @value{GDBN} commands.
1510
1511 The most likely situation where you might need this is in typing the
1512 name of a C@t{++} function. This is because C@t{++} allows function
1513 overloading (multiple definitions of the same function, distinguished
1514 by argument type). For example, when you want to set a breakpoint you
1515 may need to distinguish whether you mean the version of @code{name}
1516 that takes an @code{int} parameter, @code{name(int)}, or the version
1517 that takes a @code{float} parameter, @code{name(float)}. To use the
1518 word-completion facilities in this situation, type a single quote
1519 @code{'} at the beginning of the function name. This alerts
1520 @value{GDBN} that it may need to consider more information than usual
1521 when you press @key{TAB} or @kbd{M-?} to request word completion:
1522
1523 @smallexample
1524 (@value{GDBP}) b 'bubble( @kbd{M-?}
1525 bubble(double,double) bubble(int,int)
1526 (@value{GDBP}) b 'bubble(
1527 @end smallexample
1528
1529 In some cases, @value{GDBN} can tell that completing a name requires using
1530 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1531 completing as much as it can) if you do not type the quote in the first
1532 place:
1533
1534 @smallexample
1535 (@value{GDBP}) b bub @key{TAB}
1536 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1537 (@value{GDBP}) b 'bubble(
1538 @end smallexample
1539
1540 @noindent
1541 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1542 you have not yet started typing the argument list when you ask for
1543 completion on an overloaded symbol.
1544
1545 For more information about overloaded functions, see @ref{C Plus Plus
1546 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1547 overload-resolution off} to disable overload resolution;
1548 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1549
1550 @cindex completion of structure field names
1551 @cindex structure field name completion
1552 @cindex completion of union field names
1553 @cindex union field name completion
1554 When completing in an expression which looks up a field in a
1555 structure, @value{GDBN} also tries@footnote{The completer can be
1556 confused by certain kinds of invalid expressions. Also, it only
1557 examines the static type of the expression, not the dynamic type.} to
1558 limit completions to the field names available in the type of the
1559 left-hand-side:
1560
1561 @smallexample
1562 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1563 magic to_delete to_fputs to_put to_rewind
1564 to_data to_flush to_isatty to_read to_write
1565 @end smallexample
1566
1567 @noindent
1568 This is because the @code{gdb_stdout} is a variable of the type
1569 @code{struct ui_file} that is defined in @value{GDBN} sources as
1570 follows:
1571
1572 @smallexample
1573 struct ui_file
1574 @{
1575 int *magic;
1576 ui_file_flush_ftype *to_flush;
1577 ui_file_write_ftype *to_write;
1578 ui_file_fputs_ftype *to_fputs;
1579 ui_file_read_ftype *to_read;
1580 ui_file_delete_ftype *to_delete;
1581 ui_file_isatty_ftype *to_isatty;
1582 ui_file_rewind_ftype *to_rewind;
1583 ui_file_put_ftype *to_put;
1584 void *to_data;
1585 @}
1586 @end smallexample
1587
1588
1589 @node Help
1590 @section Getting Help
1591 @cindex online documentation
1592 @kindex help
1593
1594 You can always ask @value{GDBN} itself for information on its commands,
1595 using the command @code{help}.
1596
1597 @table @code
1598 @kindex h @r{(@code{help})}
1599 @item help
1600 @itemx h
1601 You can use @code{help} (abbreviated @code{h}) with no arguments to
1602 display a short list of named classes of commands:
1603
1604 @smallexample
1605 (@value{GDBP}) help
1606 List of classes of commands:
1607
1608 aliases -- Aliases of other commands
1609 breakpoints -- Making program stop at certain points
1610 data -- Examining data
1611 files -- Specifying and examining files
1612 internals -- Maintenance commands
1613 obscure -- Obscure features
1614 running -- Running the program
1615 stack -- Examining the stack
1616 status -- Status inquiries
1617 support -- Support facilities
1618 tracepoints -- Tracing of program execution without
1619 stopping the program
1620 user-defined -- User-defined commands
1621
1622 Type "help" followed by a class name for a list of
1623 commands in that class.
1624 Type "help" followed by command name for full
1625 documentation.
1626 Command name abbreviations are allowed if unambiguous.
1627 (@value{GDBP})
1628 @end smallexample
1629 @c the above line break eliminates huge line overfull...
1630
1631 @item help @var{class}
1632 Using one of the general help classes as an argument, you can get a
1633 list of the individual commands in that class. For example, here is the
1634 help display for the class @code{status}:
1635
1636 @smallexample
1637 (@value{GDBP}) help status
1638 Status inquiries.
1639
1640 List of commands:
1641
1642 @c Line break in "show" line falsifies real output, but needed
1643 @c to fit in smallbook page size.
1644 info -- Generic command for showing things
1645 about the program being debugged
1646 show -- Generic command for showing things
1647 about the debugger
1648
1649 Type "help" followed by command name for full
1650 documentation.
1651 Command name abbreviations are allowed if unambiguous.
1652 (@value{GDBP})
1653 @end smallexample
1654
1655 @item help @var{command}
1656 With a command name as @code{help} argument, @value{GDBN} displays a
1657 short paragraph on how to use that command.
1658
1659 @kindex apropos
1660 @item apropos @var{args}
1661 The @code{apropos} command searches through all of the @value{GDBN}
1662 commands, and their documentation, for the regular expression specified in
1663 @var{args}. It prints out all matches found. For example:
1664
1665 @smallexample
1666 apropos reload
1667 @end smallexample
1668
1669 @noindent
1670 results in:
1671
1672 @smallexample
1673 @c @group
1674 set symbol-reloading -- Set dynamic symbol table reloading
1675 multiple times in one run
1676 show symbol-reloading -- Show dynamic symbol table reloading
1677 multiple times in one run
1678 @c @end group
1679 @end smallexample
1680
1681 @kindex complete
1682 @item complete @var{args}
1683 The @code{complete @var{args}} command lists all the possible completions
1684 for the beginning of a command. Use @var{args} to specify the beginning of the
1685 command you want completed. For example:
1686
1687 @smallexample
1688 complete i
1689 @end smallexample
1690
1691 @noindent results in:
1692
1693 @smallexample
1694 @group
1695 if
1696 ignore
1697 info
1698 inspect
1699 @end group
1700 @end smallexample
1701
1702 @noindent This is intended for use by @sc{gnu} Emacs.
1703 @end table
1704
1705 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1706 and @code{show} to inquire about the state of your program, or the state
1707 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1708 manual introduces each of them in the appropriate context. The listings
1709 under @code{info} and under @code{show} in the Index point to
1710 all the sub-commands. @xref{Index}.
1711
1712 @c @group
1713 @table @code
1714 @kindex info
1715 @kindex i @r{(@code{info})}
1716 @item info
1717 This command (abbreviated @code{i}) is for describing the state of your
1718 program. For example, you can show the arguments passed to a function
1719 with @code{info args}, list the registers currently in use with @code{info
1720 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1721 You can get a complete list of the @code{info} sub-commands with
1722 @w{@code{help info}}.
1723
1724 @kindex set
1725 @item set
1726 You can assign the result of an expression to an environment variable with
1727 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1728 @code{set prompt $}.
1729
1730 @kindex show
1731 @item show
1732 In contrast to @code{info}, @code{show} is for describing the state of
1733 @value{GDBN} itself.
1734 You can change most of the things you can @code{show}, by using the
1735 related command @code{set}; for example, you can control what number
1736 system is used for displays with @code{set radix}, or simply inquire
1737 which is currently in use with @code{show radix}.
1738
1739 @kindex info set
1740 To display all the settable parameters and their current
1741 values, you can use @code{show} with no arguments; you may also use
1742 @code{info set}. Both commands produce the same display.
1743 @c FIXME: "info set" violates the rule that "info" is for state of
1744 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1745 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1746 @end table
1747 @c @end group
1748
1749 Here are three miscellaneous @code{show} subcommands, all of which are
1750 exceptional in lacking corresponding @code{set} commands:
1751
1752 @table @code
1753 @kindex show version
1754 @cindex @value{GDBN} version number
1755 @item show version
1756 Show what version of @value{GDBN} is running. You should include this
1757 information in @value{GDBN} bug-reports. If multiple versions of
1758 @value{GDBN} are in use at your site, you may need to determine which
1759 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1760 commands are introduced, and old ones may wither away. Also, many
1761 system vendors ship variant versions of @value{GDBN}, and there are
1762 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1763 The version number is the same as the one announced when you start
1764 @value{GDBN}.
1765
1766 @kindex show copying
1767 @kindex info copying
1768 @cindex display @value{GDBN} copyright
1769 @item show copying
1770 @itemx info copying
1771 Display information about permission for copying @value{GDBN}.
1772
1773 @kindex show warranty
1774 @kindex info warranty
1775 @item show warranty
1776 @itemx info warranty
1777 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1778 if your version of @value{GDBN} comes with one.
1779
1780 @end table
1781
1782 @node Running
1783 @chapter Running Programs Under @value{GDBN}
1784
1785 When you run a program under @value{GDBN}, you must first generate
1786 debugging information when you compile it.
1787
1788 You may start @value{GDBN} with its arguments, if any, in an environment
1789 of your choice. If you are doing native debugging, you may redirect
1790 your program's input and output, debug an already running process, or
1791 kill a child process.
1792
1793 @menu
1794 * Compilation:: Compiling for debugging
1795 * Starting:: Starting your program
1796 * Arguments:: Your program's arguments
1797 * Environment:: Your program's environment
1798
1799 * Working Directory:: Your program's working directory
1800 * Input/Output:: Your program's input and output
1801 * Attach:: Debugging an already-running process
1802 * Kill Process:: Killing the child process
1803
1804 * Inferiors and Programs:: Debugging multiple inferiors and programs
1805 * Threads:: Debugging programs with multiple threads
1806 * Forks:: Debugging forks
1807 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1808 @end menu
1809
1810 @node Compilation
1811 @section Compiling for Debugging
1812
1813 In order to debug a program effectively, you need to generate
1814 debugging information when you compile it. This debugging information
1815 is stored in the object file; it describes the data type of each
1816 variable or function and the correspondence between source line numbers
1817 and addresses in the executable code.
1818
1819 To request debugging information, specify the @samp{-g} option when you run
1820 the compiler.
1821
1822 Programs that are to be shipped to your customers are compiled with
1823 optimizations, using the @samp{-O} compiler option. However, some
1824 compilers are unable to handle the @samp{-g} and @samp{-O} options
1825 together. Using those compilers, you cannot generate optimized
1826 executables containing debugging information.
1827
1828 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1829 without @samp{-O}, making it possible to debug optimized code. We
1830 recommend that you @emph{always} use @samp{-g} whenever you compile a
1831 program. You may think your program is correct, but there is no sense
1832 in pushing your luck. For more information, see @ref{Optimized Code}.
1833
1834 Older versions of the @sc{gnu} C compiler permitted a variant option
1835 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1836 format; if your @sc{gnu} C compiler has this option, do not use it.
1837
1838 @value{GDBN} knows about preprocessor macros and can show you their
1839 expansion (@pxref{Macros}). Most compilers do not include information
1840 about preprocessor macros in the debugging information if you specify
1841 the @option{-g} flag alone, because this information is rather large.
1842 Version 3.1 and later of @value{NGCC}, the @sc{gnu} C compiler,
1843 provides macro information if you specify the options
1844 @option{-gdwarf-2} and @option{-g3}; the former option requests
1845 debugging information in the Dwarf 2 format, and the latter requests
1846 ``extra information''. In the future, we hope to find more compact
1847 ways to represent macro information, so that it can be included with
1848 @option{-g} alone.
1849
1850 @need 2000
1851 @node Starting
1852 @section Starting your Program
1853 @cindex starting
1854 @cindex running
1855
1856 @table @code
1857 @kindex run
1858 @kindex r @r{(@code{run})}
1859 @item run
1860 @itemx r
1861 Use the @code{run} command to start your program under @value{GDBN}.
1862 You must first specify the program name (except on VxWorks) with an
1863 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1864 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1865 (@pxref{Files, ,Commands to Specify Files}).
1866
1867 @end table
1868
1869 If you are running your program in an execution environment that
1870 supports processes, @code{run} creates an inferior process and makes
1871 that process run your program. In some environments without processes,
1872 @code{run} jumps to the start of your program. Other targets,
1873 like @samp{remote}, are always running. If you get an error
1874 message like this one:
1875
1876 @smallexample
1877 The "remote" target does not support "run".
1878 Try "help target" or "continue".
1879 @end smallexample
1880
1881 @noindent
1882 then use @code{continue} to run your program. You may need @code{load}
1883 first (@pxref{load}).
1884
1885 The execution of a program is affected by certain information it
1886 receives from its superior. @value{GDBN} provides ways to specify this
1887 information, which you must do @emph{before} starting your program. (You
1888 can change it after starting your program, but such changes only affect
1889 your program the next time you start it.) This information may be
1890 divided into four categories:
1891
1892 @table @asis
1893 @item The @emph{arguments.}
1894 Specify the arguments to give your program as the arguments of the
1895 @code{run} command. If a shell is available on your target, the shell
1896 is used to pass the arguments, so that you may use normal conventions
1897 (such as wildcard expansion or variable substitution) in describing
1898 the arguments.
1899 In Unix systems, you can control which shell is used with the
1900 @code{SHELL} environment variable.
1901 @xref{Arguments, ,Your Program's Arguments}.
1902
1903 @item The @emph{environment.}
1904 Your program normally inherits its environment from @value{GDBN}, but you can
1905 use the @value{GDBN} commands @code{set environment} and @code{unset
1906 environment} to change parts of the environment that affect
1907 your program. @xref{Environment, ,Your Program's Environment}.
1908
1909 @item The @emph{working directory.}
1910 Your program inherits its working directory from @value{GDBN}. You can set
1911 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1912 @xref{Working Directory, ,Your Program's Working Directory}.
1913
1914 @item The @emph{standard input and output.}
1915 Your program normally uses the same device for standard input and
1916 standard output as @value{GDBN} is using. You can redirect input and output
1917 in the @code{run} command line, or you can use the @code{tty} command to
1918 set a different device for your program.
1919 @xref{Input/Output, ,Your Program's Input and Output}.
1920
1921 @cindex pipes
1922 @emph{Warning:} While input and output redirection work, you cannot use
1923 pipes to pass the output of the program you are debugging to another
1924 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1925 wrong program.
1926 @end table
1927
1928 When you issue the @code{run} command, your program begins to execute
1929 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
1930 of how to arrange for your program to stop. Once your program has
1931 stopped, you may call functions in your program, using the @code{print}
1932 or @code{call} commands. @xref{Data, ,Examining Data}.
1933
1934 If the modification time of your symbol file has changed since the last
1935 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1936 table, and reads it again. When it does this, @value{GDBN} tries to retain
1937 your current breakpoints.
1938
1939 @table @code
1940 @kindex start
1941 @item start
1942 @cindex run to main procedure
1943 The name of the main procedure can vary from language to language.
1944 With C or C@t{++}, the main procedure name is always @code{main}, but
1945 other languages such as Ada do not require a specific name for their
1946 main procedure. The debugger provides a convenient way to start the
1947 execution of the program and to stop at the beginning of the main
1948 procedure, depending on the language used.
1949
1950 The @samp{start} command does the equivalent of setting a temporary
1951 breakpoint at the beginning of the main procedure and then invoking
1952 the @samp{run} command.
1953
1954 @cindex elaboration phase
1955 Some programs contain an @dfn{elaboration} phase where some startup code is
1956 executed before the main procedure is called. This depends on the
1957 languages used to write your program. In C@t{++}, for instance,
1958 constructors for static and global objects are executed before
1959 @code{main} is called. It is therefore possible that the debugger stops
1960 before reaching the main procedure. However, the temporary breakpoint
1961 will remain to halt execution.
1962
1963 Specify the arguments to give to your program as arguments to the
1964 @samp{start} command. These arguments will be given verbatim to the
1965 underlying @samp{run} command. Note that the same arguments will be
1966 reused if no argument is provided during subsequent calls to
1967 @samp{start} or @samp{run}.
1968
1969 It is sometimes necessary to debug the program during elaboration. In
1970 these cases, using the @code{start} command would stop the execution of
1971 your program too late, as the program would have already completed the
1972 elaboration phase. Under these circumstances, insert breakpoints in your
1973 elaboration code before running your program.
1974
1975 @kindex set exec-wrapper
1976 @item set exec-wrapper @var{wrapper}
1977 @itemx show exec-wrapper
1978 @itemx unset exec-wrapper
1979 When @samp{exec-wrapper} is set, the specified wrapper is used to
1980 launch programs for debugging. @value{GDBN} starts your program
1981 with a shell command of the form @kbd{exec @var{wrapper}
1982 @var{program}}. Quoting is added to @var{program} and its
1983 arguments, but not to @var{wrapper}, so you should add quotes if
1984 appropriate for your shell. The wrapper runs until it executes
1985 your program, and then @value{GDBN} takes control.
1986
1987 You can use any program that eventually calls @code{execve} with
1988 its arguments as a wrapper. Several standard Unix utilities do
1989 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
1990 with @code{exec "$@@"} will also work.
1991
1992 For example, you can use @code{env} to pass an environment variable to
1993 the debugged program, without setting the variable in your shell's
1994 environment:
1995
1996 @smallexample
1997 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
1998 (@value{GDBP}) run
1999 @end smallexample
2000
2001 This command is available when debugging locally on most targets, excluding
2002 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2003
2004 @kindex set disable-randomization
2005 @item set disable-randomization
2006 @itemx set disable-randomization on
2007 This option (enabled by default in @value{GDBN}) will turn off the native
2008 randomization of the virtual address space of the started program. This option
2009 is useful for multiple debugging sessions to make the execution better
2010 reproducible and memory addresses reusable across debugging sessions.
2011
2012 This feature is implemented only on @sc{gnu}/Linux. You can get the same
2013 behavior using
2014
2015 @smallexample
2016 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2017 @end smallexample
2018
2019 @item set disable-randomization off
2020 Leave the behavior of the started executable unchanged. Some bugs rear their
2021 ugly heads only when the program is loaded at certain addresses. If your bug
2022 disappears when you run the program under @value{GDBN}, that might be because
2023 @value{GDBN} by default disables the address randomization on platforms, such
2024 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2025 disable-randomization off} to try to reproduce such elusive bugs.
2026
2027 The virtual address space randomization is implemented only on @sc{gnu}/Linux.
2028 It protects the programs against some kinds of security attacks. In these
2029 cases the attacker needs to know the exact location of a concrete executable
2030 code. Randomizing its location makes it impossible to inject jumps misusing
2031 a code at its expected addresses.
2032
2033 Prelinking shared libraries provides a startup performance advantage but it
2034 makes addresses in these libraries predictable for privileged processes by
2035 having just unprivileged access at the target system. Reading the shared
2036 library binary gives enough information for assembling the malicious code
2037 misusing it. Still even a prelinked shared library can get loaded at a new
2038 random address just requiring the regular relocation process during the
2039 startup. Shared libraries not already prelinked are always loaded at
2040 a randomly chosen address.
2041
2042 Position independent executables (PIE) contain position independent code
2043 similar to the shared libraries and therefore such executables get loaded at
2044 a randomly chosen address upon startup. PIE executables always load even
2045 already prelinked shared libraries at a random address. You can build such
2046 executable using @command{gcc -fPIE -pie}.
2047
2048 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2049 (as long as the randomization is enabled).
2050
2051 @item show disable-randomization
2052 Show the current setting of the explicit disable of the native randomization of
2053 the virtual address space of the started program.
2054
2055 @end table
2056
2057 @node Arguments
2058 @section Your Program's Arguments
2059
2060 @cindex arguments (to your program)
2061 The arguments to your program can be specified by the arguments of the
2062 @code{run} command.
2063 They are passed to a shell, which expands wildcard characters and
2064 performs redirection of I/O, and thence to your program. Your
2065 @code{SHELL} environment variable (if it exists) specifies what shell
2066 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2067 the default shell (@file{/bin/sh} on Unix).
2068
2069 On non-Unix systems, the program is usually invoked directly by
2070 @value{GDBN}, which emulates I/O redirection via the appropriate system
2071 calls, and the wildcard characters are expanded by the startup code of
2072 the program, not by the shell.
2073
2074 @code{run} with no arguments uses the same arguments used by the previous
2075 @code{run}, or those set by the @code{set args} command.
2076
2077 @table @code
2078 @kindex set args
2079 @item set args
2080 Specify the arguments to be used the next time your program is run. If
2081 @code{set args} has no arguments, @code{run} executes your program
2082 with no arguments. Once you have run your program with arguments,
2083 using @code{set args} before the next @code{run} is the only way to run
2084 it again without arguments.
2085
2086 @kindex show args
2087 @item show args
2088 Show the arguments to give your program when it is started.
2089 @end table
2090
2091 @node Environment
2092 @section Your Program's Environment
2093
2094 @cindex environment (of your program)
2095 The @dfn{environment} consists of a set of environment variables and
2096 their values. Environment variables conventionally record such things as
2097 your user name, your home directory, your terminal type, and your search
2098 path for programs to run. Usually you set up environment variables with
2099 the shell and they are inherited by all the other programs you run. When
2100 debugging, it can be useful to try running your program with a modified
2101 environment without having to start @value{GDBN} over again.
2102
2103 @table @code
2104 @kindex path
2105 @item path @var{directory}
2106 Add @var{directory} to the front of the @code{PATH} environment variable
2107 (the search path for executables) that will be passed to your program.
2108 The value of @code{PATH} used by @value{GDBN} does not change.
2109 You may specify several directory names, separated by whitespace or by a
2110 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2111 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2112 is moved to the front, so it is searched sooner.
2113
2114 You can use the string @samp{$cwd} to refer to whatever is the current
2115 working directory at the time @value{GDBN} searches the path. If you
2116 use @samp{.} instead, it refers to the directory where you executed the
2117 @code{path} command. @value{GDBN} replaces @samp{.} in the
2118 @var{directory} argument (with the current path) before adding
2119 @var{directory} to the search path.
2120 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2121 @c document that, since repeating it would be a no-op.
2122
2123 @kindex show paths
2124 @item show paths
2125 Display the list of search paths for executables (the @code{PATH}
2126 environment variable).
2127
2128 @kindex show environment
2129 @item show environment @r{[}@var{varname}@r{]}
2130 Print the value of environment variable @var{varname} to be given to
2131 your program when it starts. If you do not supply @var{varname},
2132 print the names and values of all environment variables to be given to
2133 your program. You can abbreviate @code{environment} as @code{env}.
2134
2135 @kindex set environment
2136 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2137 Set environment variable @var{varname} to @var{value}. The value
2138 changes for your program only, not for @value{GDBN} itself. @var{value} may
2139 be any string; the values of environment variables are just strings, and
2140 any interpretation is supplied by your program itself. The @var{value}
2141 parameter is optional; if it is eliminated, the variable is set to a
2142 null value.
2143 @c "any string" here does not include leading, trailing
2144 @c blanks. Gnu asks: does anyone care?
2145
2146 For example, this command:
2147
2148 @smallexample
2149 set env USER = foo
2150 @end smallexample
2151
2152 @noindent
2153 tells the debugged program, when subsequently run, that its user is named
2154 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2155 are not actually required.)
2156
2157 @kindex unset environment
2158 @item unset environment @var{varname}
2159 Remove variable @var{varname} from the environment to be passed to your
2160 program. This is different from @samp{set env @var{varname} =};
2161 @code{unset environment} removes the variable from the environment,
2162 rather than assigning it an empty value.
2163 @end table
2164
2165 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2166 the shell indicated
2167 by your @code{SHELL} environment variable if it exists (or
2168 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2169 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2170 @file{.bashrc} for BASH---any variables you set in that file affect
2171 your program. You may wish to move setting of environment variables to
2172 files that are only run when you sign on, such as @file{.login} or
2173 @file{.profile}.
2174
2175 @node Working Directory
2176 @section Your Program's Working Directory
2177
2178 @cindex working directory (of your program)
2179 Each time you start your program with @code{run}, it inherits its
2180 working directory from the current working directory of @value{GDBN}.
2181 The @value{GDBN} working directory is initially whatever it inherited
2182 from its parent process (typically the shell), but you can specify a new
2183 working directory in @value{GDBN} with the @code{cd} command.
2184
2185 The @value{GDBN} working directory also serves as a default for the commands
2186 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2187 Specify Files}.
2188
2189 @table @code
2190 @kindex cd
2191 @cindex change working directory
2192 @item cd @var{directory}
2193 Set the @value{GDBN} working directory to @var{directory}.
2194
2195 @kindex pwd
2196 @item pwd
2197 Print the @value{GDBN} working directory.
2198 @end table
2199
2200 It is generally impossible to find the current working directory of
2201 the process being debugged (since a program can change its directory
2202 during its run). If you work on a system where @value{GDBN} is
2203 configured with the @file{/proc} support, you can use the @code{info
2204 proc} command (@pxref{SVR4 Process Information}) to find out the
2205 current working directory of the debuggee.
2206
2207 @node Input/Output
2208 @section Your Program's Input and Output
2209
2210 @cindex redirection
2211 @cindex i/o
2212 @cindex terminal
2213 By default, the program you run under @value{GDBN} does input and output to
2214 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2215 to its own terminal modes to interact with you, but it records the terminal
2216 modes your program was using and switches back to them when you continue
2217 running your program.
2218
2219 @table @code
2220 @kindex info terminal
2221 @item info terminal
2222 Displays information recorded by @value{GDBN} about the terminal modes your
2223 program is using.
2224 @end table
2225
2226 You can redirect your program's input and/or output using shell
2227 redirection with the @code{run} command. For example,
2228
2229 @smallexample
2230 run > outfile
2231 @end smallexample
2232
2233 @noindent
2234 starts your program, diverting its output to the file @file{outfile}.
2235
2236 @kindex tty
2237 @cindex controlling terminal
2238 Another way to specify where your program should do input and output is
2239 with the @code{tty} command. This command accepts a file name as
2240 argument, and causes this file to be the default for future @code{run}
2241 commands. It also resets the controlling terminal for the child
2242 process, for future @code{run} commands. For example,
2243
2244 @smallexample
2245 tty /dev/ttyb
2246 @end smallexample
2247
2248 @noindent
2249 directs that processes started with subsequent @code{run} commands
2250 default to do input and output on the terminal @file{/dev/ttyb} and have
2251 that as their controlling terminal.
2252
2253 An explicit redirection in @code{run} overrides the @code{tty} command's
2254 effect on the input/output device, but not its effect on the controlling
2255 terminal.
2256
2257 When you use the @code{tty} command or redirect input in the @code{run}
2258 command, only the input @emph{for your program} is affected. The input
2259 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2260 for @code{set inferior-tty}.
2261
2262 @cindex inferior tty
2263 @cindex set inferior controlling terminal
2264 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2265 display the name of the terminal that will be used for future runs of your
2266 program.
2267
2268 @table @code
2269 @item set inferior-tty /dev/ttyb
2270 @kindex set inferior-tty
2271 Set the tty for the program being debugged to /dev/ttyb.
2272
2273 @item show inferior-tty
2274 @kindex show inferior-tty
2275 Show the current tty for the program being debugged.
2276 @end table
2277
2278 @node Attach
2279 @section Debugging an Already-running Process
2280 @kindex attach
2281 @cindex attach
2282
2283 @table @code
2284 @item attach @var{process-id}
2285 This command attaches to a running process---one that was started
2286 outside @value{GDBN}. (@code{info files} shows your active
2287 targets.) The command takes as argument a process ID. The usual way to
2288 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2289 or with the @samp{jobs -l} shell command.
2290
2291 @code{attach} does not repeat if you press @key{RET} a second time after
2292 executing the command.
2293 @end table
2294
2295 To use @code{attach}, your program must be running in an environment
2296 which supports processes; for example, @code{attach} does not work for
2297 programs on bare-board targets that lack an operating system. You must
2298 also have permission to send the process a signal.
2299
2300 When you use @code{attach}, the debugger finds the program running in
2301 the process first by looking in the current working directory, then (if
2302 the program is not found) by using the source file search path
2303 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2304 the @code{file} command to load the program. @xref{Files, ,Commands to
2305 Specify Files}.
2306
2307 The first thing @value{GDBN} does after arranging to debug the specified
2308 process is to stop it. You can examine and modify an attached process
2309 with all the @value{GDBN} commands that are ordinarily available when
2310 you start processes with @code{run}. You can insert breakpoints; you
2311 can step and continue; you can modify storage. If you would rather the
2312 process continue running, you may use the @code{continue} command after
2313 attaching @value{GDBN} to the process.
2314
2315 @table @code
2316 @kindex detach
2317 @item detach
2318 When you have finished debugging the attached process, you can use the
2319 @code{detach} command to release it from @value{GDBN} control. Detaching
2320 the process continues its execution. After the @code{detach} command,
2321 that process and @value{GDBN} become completely independent once more, and you
2322 are ready to @code{attach} another process or start one with @code{run}.
2323 @code{detach} does not repeat if you press @key{RET} again after
2324 executing the command.
2325 @end table
2326
2327 If you exit @value{GDBN} while you have an attached process, you detach
2328 that process. If you use the @code{run} command, you kill that process.
2329 By default, @value{GDBN} asks for confirmation if you try to do either of these
2330 things; you can control whether or not you need to confirm by using the
2331 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2332 Messages}).
2333
2334 @node Kill Process
2335 @section Killing the Child Process
2336
2337 @table @code
2338 @kindex kill
2339 @item kill
2340 Kill the child process in which your program is running under @value{GDBN}.
2341 @end table
2342
2343 This command is useful if you wish to debug a core dump instead of a
2344 running process. @value{GDBN} ignores any core dump file while your program
2345 is running.
2346
2347 On some operating systems, a program cannot be executed outside @value{GDBN}
2348 while you have breakpoints set on it inside @value{GDBN}. You can use the
2349 @code{kill} command in this situation to permit running your program
2350 outside the debugger.
2351
2352 The @code{kill} command is also useful if you wish to recompile and
2353 relink your program, since on many systems it is impossible to modify an
2354 executable file while it is running in a process. In this case, when you
2355 next type @code{run}, @value{GDBN} notices that the file has changed, and
2356 reads the symbol table again (while trying to preserve your current
2357 breakpoint settings).
2358
2359 @node Inferiors and Programs
2360 @section Debugging Multiple Inferiors and Programs
2361
2362 @value{GDBN} lets you run and debug multiple programs in a single
2363 session. In addition, @value{GDBN} on some systems may let you run
2364 several programs simultaneously (otherwise you have to exit from one
2365 before starting another). In the most general case, you can have
2366 multiple threads of execution in each of multiple processes, launched
2367 from multiple executables.
2368
2369 @cindex inferior
2370 @value{GDBN} represents the state of each program execution with an
2371 object called an @dfn{inferior}. An inferior typically corresponds to
2372 a process, but is more general and applies also to targets that do not
2373 have processes. Inferiors may be created before a process runs, and
2374 may be retained after a process exits. Inferiors have unique
2375 identifiers that are different from process ids. Usually each
2376 inferior will also have its own distinct address space, although some
2377 embedded targets may have several inferiors running in different parts
2378 of a single address space. Each inferior may in turn have multiple
2379 threads running in it.
2380
2381 To find out what inferiors exist at any moment, use @w{@code{info
2382 inferiors}}:
2383
2384 @table @code
2385 @kindex info inferiors
2386 @item info inferiors
2387 Print a list of all inferiors currently being managed by @value{GDBN}.
2388
2389 @value{GDBN} displays for each inferior (in this order):
2390
2391 @enumerate
2392 @item
2393 the inferior number assigned by @value{GDBN}
2394
2395 @item
2396 the target system's inferior identifier
2397
2398 @item
2399 the name of the executable the inferior is running.
2400
2401 @end enumerate
2402
2403 @noindent
2404 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2405 indicates the current inferior.
2406
2407 For example,
2408 @end table
2409 @c end table here to get a little more width for example
2410
2411 @smallexample
2412 (@value{GDBP}) info inferiors
2413 Num Description Executable
2414 2 process 2307 hello
2415 * 1 process 3401 goodbye
2416 @end smallexample
2417
2418 To switch focus between inferiors, use the @code{inferior} command:
2419
2420 @table @code
2421 @kindex inferior @var{infno}
2422 @item inferior @var{infno}
2423 Make inferior number @var{infno} the current inferior. The argument
2424 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2425 in the first field of the @samp{info inferiors} display.
2426 @end table
2427
2428
2429 You can get multiple executables into a debugging session via the
2430 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2431 systems @value{GDBN} can add inferiors to the debug session
2432 automatically by following calls to @code{fork} and @code{exec}. To
2433 remove inferiors from the debugging session use the
2434 @w{@code{remove-inferior}} command.
2435
2436 @table @code
2437 @kindex add-inferior
2438 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2439 Adds @var{n} inferiors to be run using @var{executable} as the
2440 executable. @var{n} defaults to 1. If no executable is specified,
2441 the inferiors begins empty, with no program. You can still assign or
2442 change the program assigned to the inferior at any time by using the
2443 @code{file} command with the executable name as its argument.
2444
2445 @kindex clone-inferior
2446 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2447 Adds @var{n} inferiors ready to execute the same program as inferior
2448 @var{infno}. @var{n} defaults to 1. @var{infno} defaults to the
2449 number of the current inferior. This is a convenient command when you
2450 want to run another instance of the inferior you are debugging.
2451
2452 @smallexample
2453 (@value{GDBP}) info inferiors
2454 Num Description Executable
2455 * 1 process 29964 helloworld
2456 (@value{GDBP}) clone-inferior
2457 Added inferior 2.
2458 1 inferiors added.
2459 (@value{GDBP}) info inferiors
2460 Num Description Executable
2461 2 <null> helloworld
2462 * 1 process 29964 helloworld
2463 @end smallexample
2464
2465 You can now simply switch focus to inferior 2 and run it.
2466
2467 @kindex remove-inferior
2468 @item remove-inferior @var{infno}
2469 Removes the inferior @var{infno}. It is not possible to remove an
2470 inferior that is running with this command. For those, use the
2471 @code{kill} or @code{detach} command first.
2472
2473 @end table
2474
2475 To quit debugging one of the running inferiors that is not the current
2476 inferior, you can either detach from it by using the @w{@code{detach
2477 inferior}} command (allowing it to run independently), or kill it
2478 using the @w{@code{kill inferior}} command:
2479
2480 @table @code
2481 @kindex detach inferior @var{infno}
2482 @item detach inferior @var{infno}
2483 Detach from the inferior identified by @value{GDBN} inferior number
2484 @var{infno}. Note that the inferior's entry still stays on the list
2485 of inferiors shown by @code{info inferiors}, but its Description will
2486 show @samp{<null>}.
2487
2488 @kindex kill inferior @var{infno}
2489 @item kill inferior @var{infno}
2490 Kill the inferior identified by @value{GDBN} inferior number
2491 @var{infno}. Note that the inferior's entry still stays on the list
2492 of inferiors shown by @code{info inferiors}, but its Description will
2493 show @samp{<null>}.
2494 @end table
2495
2496 After the successful completion of a command such as @code{detach},
2497 @code{detach inferior}, @code{kill} or @code{kill inferior}, or after
2498 a normal process exit, the inferior is still valid and listed with
2499 @code{info inferiors}, ready to be restarted.
2500
2501
2502 To be notified when inferiors are started or exit under @value{GDBN}'s
2503 control use @w{@code{set print inferior-events}}:
2504
2505 @table @code
2506 @kindex set print inferior-events
2507 @cindex print messages on inferior start and exit
2508 @item set print inferior-events
2509 @itemx set print inferior-events on
2510 @itemx set print inferior-events off
2511 The @code{set print inferior-events} command allows you to enable or
2512 disable printing of messages when @value{GDBN} notices that new
2513 inferiors have started or that inferiors have exited or have been
2514 detached. By default, these messages will not be printed.
2515
2516 @kindex show print inferior-events
2517 @item show print inferior-events
2518 Show whether messages will be printed when @value{GDBN} detects that
2519 inferiors have started, exited or have been detached.
2520 @end table
2521
2522 Many commands will work the same with multiple programs as with a
2523 single program: e.g., @code{print myglobal} will simply display the
2524 value of @code{myglobal} in the current inferior.
2525
2526
2527 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2528 get more info about the relationship of inferiors, programs, address
2529 spaces in a debug session. You can do that with the @w{@code{maint
2530 info program-spaces}} command.
2531
2532 @table @code
2533 @kindex maint info program-spaces
2534 @item maint info program-spaces
2535 Print a list of all program spaces currently being managed by
2536 @value{GDBN}.
2537
2538 @value{GDBN} displays for each program space (in this order):
2539
2540 @enumerate
2541 @item
2542 the program space number assigned by @value{GDBN}
2543
2544 @item
2545 the name of the executable loaded into the program space, with e.g.,
2546 the @code{file} command.
2547
2548 @end enumerate
2549
2550 @noindent
2551 An asterisk @samp{*} preceding the @value{GDBN} program space number
2552 indicates the current program space.
2553
2554 In addition, below each program space line, @value{GDBN} prints extra
2555 information that isn't suitable to display in tabular form. For
2556 example, the list of inferiors bound to the program space.
2557
2558 @smallexample
2559 (@value{GDBP}) maint info program-spaces
2560 Id Executable
2561 2 goodbye
2562 Bound inferiors: ID 1 (process 21561)
2563 * 1 hello
2564 @end smallexample
2565
2566 Here we can see that no inferior is running the program @code{hello},
2567 while @code{process 21561} is running the program @code{goodbye}. On
2568 some targets, it is possible that multiple inferiors are bound to the
2569 same program space. The most common example is that of debugging both
2570 the parent and child processes of a @code{vfork} call. For example,
2571
2572 @smallexample
2573 (@value{GDBP}) maint info program-spaces
2574 Id Executable
2575 * 1 vfork-test
2576 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2577 @end smallexample
2578
2579 Here, both inferior 2 and inferior 1 are running in the same program
2580 space as a result of inferior 1 having executed a @code{vfork} call.
2581 @end table
2582
2583 @node Threads
2584 @section Debugging Programs with Multiple Threads
2585
2586 @cindex threads of execution
2587 @cindex multiple threads
2588 @cindex switching threads
2589 In some operating systems, such as HP-UX and Solaris, a single program
2590 may have more than one @dfn{thread} of execution. The precise semantics
2591 of threads differ from one operating system to another, but in general
2592 the threads of a single program are akin to multiple processes---except
2593 that they share one address space (that is, they can all examine and
2594 modify the same variables). On the other hand, each thread has its own
2595 registers and execution stack, and perhaps private memory.
2596
2597 @value{GDBN} provides these facilities for debugging multi-thread
2598 programs:
2599
2600 @itemize @bullet
2601 @item automatic notification of new threads
2602 @item @samp{thread @var{threadno}}, a command to switch among threads
2603 @item @samp{info threads}, a command to inquire about existing threads
2604 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2605 a command to apply a command to a list of threads
2606 @item thread-specific breakpoints
2607 @item @samp{set print thread-events}, which controls printing of
2608 messages on thread start and exit.
2609 @item @samp{set libthread-db-search-path @var{path}}, which lets
2610 the user specify which @code{libthread_db} to use if the default choice
2611 isn't compatible with the program.
2612 @end itemize
2613
2614 @quotation
2615 @emph{Warning:} These facilities are not yet available on every
2616 @value{GDBN} configuration where the operating system supports threads.
2617 If your @value{GDBN} does not support threads, these commands have no
2618 effect. For example, a system without thread support shows no output
2619 from @samp{info threads}, and always rejects the @code{thread} command,
2620 like this:
2621
2622 @smallexample
2623 (@value{GDBP}) info threads
2624 (@value{GDBP}) thread 1
2625 Thread ID 1 not known. Use the "info threads" command to
2626 see the IDs of currently known threads.
2627 @end smallexample
2628 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2629 @c doesn't support threads"?
2630 @end quotation
2631
2632 @cindex focus of debugging
2633 @cindex current thread
2634 The @value{GDBN} thread debugging facility allows you to observe all
2635 threads while your program runs---but whenever @value{GDBN} takes
2636 control, one thread in particular is always the focus of debugging.
2637 This thread is called the @dfn{current thread}. Debugging commands show
2638 program information from the perspective of the current thread.
2639
2640 @cindex @code{New} @var{systag} message
2641 @cindex thread identifier (system)
2642 @c FIXME-implementors!! It would be more helpful if the [New...] message
2643 @c included GDB's numeric thread handle, so you could just go to that
2644 @c thread without first checking `info threads'.
2645 Whenever @value{GDBN} detects a new thread in your program, it displays
2646 the target system's identification for the thread with a message in the
2647 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2648 whose form varies depending on the particular system. For example, on
2649 @sc{gnu}/Linux, you might see
2650
2651 @smallexample
2652 [New Thread 46912507313328 (LWP 25582)]
2653 @end smallexample
2654
2655 @noindent
2656 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2657 the @var{systag} is simply something like @samp{process 368}, with no
2658 further qualifier.
2659
2660 @c FIXME!! (1) Does the [New...] message appear even for the very first
2661 @c thread of a program, or does it only appear for the
2662 @c second---i.e.@: when it becomes obvious we have a multithread
2663 @c program?
2664 @c (2) *Is* there necessarily a first thread always? Or do some
2665 @c multithread systems permit starting a program with multiple
2666 @c threads ab initio?
2667
2668 @cindex thread number
2669 @cindex thread identifier (GDB)
2670 For debugging purposes, @value{GDBN} associates its own thread
2671 number---always a single integer---with each thread in your program.
2672
2673 @table @code
2674 @kindex info threads
2675 @item info threads
2676 Display a summary of all threads currently in your
2677 program. @value{GDBN} displays for each thread (in this order):
2678
2679 @enumerate
2680 @item
2681 the thread number assigned by @value{GDBN}
2682
2683 @item
2684 the target system's thread identifier (@var{systag})
2685
2686 @item
2687 the current stack frame summary for that thread
2688 @end enumerate
2689
2690 @noindent
2691 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2692 indicates the current thread.
2693
2694 For example,
2695 @end table
2696 @c end table here to get a little more width for example
2697
2698 @smallexample
2699 (@value{GDBP}) info threads
2700 3 process 35 thread 27 0x34e5 in sigpause ()
2701 2 process 35 thread 23 0x34e5 in sigpause ()
2702 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2703 at threadtest.c:68
2704 @end smallexample
2705
2706 On HP-UX systems:
2707
2708 @cindex debugging multithreaded programs (on HP-UX)
2709 @cindex thread identifier (GDB), on HP-UX
2710 For debugging purposes, @value{GDBN} associates its own thread
2711 number---a small integer assigned in thread-creation order---with each
2712 thread in your program.
2713
2714 @cindex @code{New} @var{systag} message, on HP-UX
2715 @cindex thread identifier (system), on HP-UX
2716 @c FIXME-implementors!! It would be more helpful if the [New...] message
2717 @c included GDB's numeric thread handle, so you could just go to that
2718 @c thread without first checking `info threads'.
2719 Whenever @value{GDBN} detects a new thread in your program, it displays
2720 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2721 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2722 whose form varies depending on the particular system. For example, on
2723 HP-UX, you see
2724
2725 @smallexample
2726 [New thread 2 (system thread 26594)]
2727 @end smallexample
2728
2729 @noindent
2730 when @value{GDBN} notices a new thread.
2731
2732 @table @code
2733 @kindex info threads (HP-UX)
2734 @item info threads
2735 Display a summary of all threads currently in your
2736 program. @value{GDBN} displays for each thread (in this order):
2737
2738 @enumerate
2739 @item the thread number assigned by @value{GDBN}
2740
2741 @item the target system's thread identifier (@var{systag})
2742
2743 @item the current stack frame summary for that thread
2744 @end enumerate
2745
2746 @noindent
2747 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2748 indicates the current thread.
2749
2750 For example,
2751 @end table
2752 @c end table here to get a little more width for example
2753
2754 @smallexample
2755 (@value{GDBP}) info threads
2756 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2757 at quicksort.c:137
2758 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2759 from /usr/lib/libc.2
2760 1 system thread 27905 0x7b003498 in _brk () \@*
2761 from /usr/lib/libc.2
2762 @end smallexample
2763
2764 On Solaris, you can display more information about user threads with a
2765 Solaris-specific command:
2766
2767 @table @code
2768 @item maint info sol-threads
2769 @kindex maint info sol-threads
2770 @cindex thread info (Solaris)
2771 Display info on Solaris user threads.
2772 @end table
2773
2774 @table @code
2775 @kindex thread @var{threadno}
2776 @item thread @var{threadno}
2777 Make thread number @var{threadno} the current thread. The command
2778 argument @var{threadno} is the internal @value{GDBN} thread number, as
2779 shown in the first field of the @samp{info threads} display.
2780 @value{GDBN} responds by displaying the system identifier of the thread
2781 you selected, and its current stack frame summary:
2782
2783 @smallexample
2784 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2785 (@value{GDBP}) thread 2
2786 [Switching to process 35 thread 23]
2787 0x34e5 in sigpause ()
2788 @end smallexample
2789
2790 @noindent
2791 As with the @samp{[New @dots{}]} message, the form of the text after
2792 @samp{Switching to} depends on your system's conventions for identifying
2793 threads.
2794
2795 @vindex $_thread@r{, convenience variable}
2796 The debugger convenience variable @samp{$_thread} contains the number
2797 of the current thread. You may find this useful in writing breakpoint
2798 conditional expressions, command scripts, and so forth. See
2799 @xref{Convenience Vars,, Convenience Variables}, for general
2800 information on convenience variables.
2801
2802 @kindex thread apply
2803 @cindex apply command to several threads
2804 @item thread apply [@var{threadno}] [@var{all}] @var{command}
2805 The @code{thread apply} command allows you to apply the named
2806 @var{command} to one or more threads. Specify the numbers of the
2807 threads that you want affected with the command argument
2808 @var{threadno}. It can be a single thread number, one of the numbers
2809 shown in the first field of the @samp{info threads} display; or it
2810 could be a range of thread numbers, as in @code{2-4}. To apply a
2811 command to all threads, type @kbd{thread apply all @var{command}}.
2812
2813 @kindex set print thread-events
2814 @cindex print messages on thread start and exit
2815 @item set print thread-events
2816 @itemx set print thread-events on
2817 @itemx set print thread-events off
2818 The @code{set print thread-events} command allows you to enable or
2819 disable printing of messages when @value{GDBN} notices that new threads have
2820 started or that threads have exited. By default, these messages will
2821 be printed if detection of these events is supported by the target.
2822 Note that these messages cannot be disabled on all targets.
2823
2824 @kindex show print thread-events
2825 @item show print thread-events
2826 Show whether messages will be printed when @value{GDBN} detects that threads
2827 have started and exited.
2828 @end table
2829
2830 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2831 more information about how @value{GDBN} behaves when you stop and start
2832 programs with multiple threads.
2833
2834 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2835 watchpoints in programs with multiple threads.
2836
2837 @table @code
2838 @kindex set libthread-db-search-path
2839 @cindex search path for @code{libthread_db}
2840 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2841 If this variable is set, @var{path} is a colon-separated list of
2842 directories @value{GDBN} will use to search for @code{libthread_db}.
2843 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2844 an empty list.
2845
2846 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2847 @code{libthread_db} library to obtain information about threads in the
2848 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2849 to find @code{libthread_db}. If that fails, @value{GDBN} will continue
2850 with default system shared library directories, and finally the directory
2851 from which @code{libpthread} was loaded in the inferior process.
2852
2853 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2854 @value{GDBN} attempts to initialize it with the current inferior process.
2855 If this initialization fails (which could happen because of a version
2856 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2857 will unload @code{libthread_db}, and continue with the next directory.
2858 If none of @code{libthread_db} libraries initialize successfully,
2859 @value{GDBN} will issue a warning and thread debugging will be disabled.
2860
2861 Setting @code{libthread-db-search-path} is currently implemented
2862 only on some platforms.
2863
2864 @kindex show libthread-db-search-path
2865 @item show libthread-db-search-path
2866 Display current libthread_db search path.
2867
2868 @kindex set debug libthread-db
2869 @kindex show debug libthread-db
2870 @cindex debugging @code{libthread_db}
2871 @item set debug libthread-db
2872 @itemx show debug libthread-db
2873 Turns on or off display of @code{libthread_db}-related events.
2874 Use @code{1} to enable, @code{0} to disable.
2875 @end table
2876
2877 @node Forks
2878 @section Debugging Forks
2879
2880 @cindex fork, debugging programs which call
2881 @cindex multiple processes
2882 @cindex processes, multiple
2883 On most systems, @value{GDBN} has no special support for debugging
2884 programs which create additional processes using the @code{fork}
2885 function. When a program forks, @value{GDBN} will continue to debug the
2886 parent process and the child process will run unimpeded. If you have
2887 set a breakpoint in any code which the child then executes, the child
2888 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2889 will cause it to terminate.
2890
2891 However, if you want to debug the child process there is a workaround
2892 which isn't too painful. Put a call to @code{sleep} in the code which
2893 the child process executes after the fork. It may be useful to sleep
2894 only if a certain environment variable is set, or a certain file exists,
2895 so that the delay need not occur when you don't want to run @value{GDBN}
2896 on the child. While the child is sleeping, use the @code{ps} program to
2897 get its process ID. Then tell @value{GDBN} (a new invocation of
2898 @value{GDBN} if you are also debugging the parent process) to attach to
2899 the child process (@pxref{Attach}). From that point on you can debug
2900 the child process just like any other process which you attached to.
2901
2902 On some systems, @value{GDBN} provides support for debugging programs that
2903 create additional processes using the @code{fork} or @code{vfork} functions.
2904 Currently, the only platforms with this feature are HP-UX (11.x and later
2905 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2906
2907 By default, when a program forks, @value{GDBN} will continue to debug
2908 the parent process and the child process will run unimpeded.
2909
2910 If you want to follow the child process instead of the parent process,
2911 use the command @w{@code{set follow-fork-mode}}.
2912
2913 @table @code
2914 @kindex set follow-fork-mode
2915 @item set follow-fork-mode @var{mode}
2916 Set the debugger response to a program call of @code{fork} or
2917 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2918 process. The @var{mode} argument can be:
2919
2920 @table @code
2921 @item parent
2922 The original process is debugged after a fork. The child process runs
2923 unimpeded. This is the default.
2924
2925 @item child
2926 The new process is debugged after a fork. The parent process runs
2927 unimpeded.
2928
2929 @end table
2930
2931 @kindex show follow-fork-mode
2932 @item show follow-fork-mode
2933 Display the current debugger response to a @code{fork} or @code{vfork} call.
2934 @end table
2935
2936 @cindex debugging multiple processes
2937 On Linux, if you want to debug both the parent and child processes, use the
2938 command @w{@code{set detach-on-fork}}.
2939
2940 @table @code
2941 @kindex set detach-on-fork
2942 @item set detach-on-fork @var{mode}
2943 Tells gdb whether to detach one of the processes after a fork, or
2944 retain debugger control over them both.
2945
2946 @table @code
2947 @item on
2948 The child process (or parent process, depending on the value of
2949 @code{follow-fork-mode}) will be detached and allowed to run
2950 independently. This is the default.
2951
2952 @item off
2953 Both processes will be held under the control of @value{GDBN}.
2954 One process (child or parent, depending on the value of
2955 @code{follow-fork-mode}) is debugged as usual, while the other
2956 is held suspended.
2957
2958 @end table
2959
2960 @kindex show detach-on-fork
2961 @item show detach-on-fork
2962 Show whether detach-on-fork mode is on/off.
2963 @end table
2964
2965 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
2966 will retain control of all forked processes (including nested forks).
2967 You can list the forked processes under the control of @value{GDBN} by
2968 using the @w{@code{info inferiors}} command, and switch from one fork
2969 to another by using the @code{inferior} command (@pxref{Inferiors and
2970 Programs, ,Debugging Multiple Inferiors and Programs}).
2971
2972 To quit debugging one of the forked processes, you can either detach
2973 from it by using the @w{@code{detach inferior}} command (allowing it
2974 to run independently), or kill it using the @w{@code{kill inferior}}
2975 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
2976 and Programs}.
2977
2978 If you ask to debug a child process and a @code{vfork} is followed by an
2979 @code{exec}, @value{GDBN} executes the new target up to the first
2980 breakpoint in the new target. If you have a breakpoint set on
2981 @code{main} in your original program, the breakpoint will also be set on
2982 the child process's @code{main}.
2983
2984 On some systems, when a child process is spawned by @code{vfork}, you
2985 cannot debug the child or parent until an @code{exec} call completes.
2986
2987 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2988 call executes, the new target restarts. To restart the parent
2989 process, use the @code{file} command with the parent executable name
2990 as its argument. By default, after an @code{exec} call executes,
2991 @value{GDBN} discards the symbols of the previous executable image.
2992 You can change this behaviour with the @w{@code{set follow-exec-mode}}
2993 command.
2994
2995 @table @code
2996 @kindex set follow-exec-mode
2997 @item set follow-exec-mode @var{mode}
2998
2999 Set debugger response to a program call of @code{exec}. An
3000 @code{exec} call replaces the program image of a process.
3001
3002 @code{follow-exec-mode} can be:
3003
3004 @table @code
3005 @item new
3006 @value{GDBN} creates a new inferior and rebinds the process to this
3007 new inferior. The program the process was running before the
3008 @code{exec} call can be restarted afterwards by restarting the
3009 original inferior.
3010
3011 For example:
3012
3013 @smallexample
3014 (@value{GDBP}) info inferiors
3015 (gdb) info inferior
3016 Id Description Executable
3017 * 1 <null> prog1
3018 (@value{GDBP}) run
3019 process 12020 is executing new program: prog2
3020 Program exited normally.
3021 (@value{GDBP}) info inferiors
3022 Id Description Executable
3023 * 2 <null> prog2
3024 1 <null> prog1
3025 @end smallexample
3026
3027 @item same
3028 @value{GDBN} keeps the process bound to the same inferior. The new
3029 executable image replaces the previous executable loaded in the
3030 inferior. Restarting the inferior after the @code{exec} call, with
3031 e.g., the @code{run} command, restarts the executable the process was
3032 running after the @code{exec} call. This is the default mode.
3033
3034 For example:
3035
3036 @smallexample
3037 (@value{GDBP}) info inferiors
3038 Id Description Executable
3039 * 1 <null> prog1
3040 (@value{GDBP}) run
3041 process 12020 is executing new program: prog2
3042 Program exited normally.
3043 (@value{GDBP}) info inferiors
3044 Id Description Executable
3045 * 1 <null> prog2
3046 @end smallexample
3047
3048 @end table
3049 @end table
3050
3051 You can use the @code{catch} command to make @value{GDBN} stop whenever
3052 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3053 Catchpoints, ,Setting Catchpoints}.
3054
3055 @node Checkpoint/Restart
3056 @section Setting a @emph{Bookmark} to Return to Later
3057
3058 @cindex checkpoint
3059 @cindex restart
3060 @cindex bookmark
3061 @cindex snapshot of a process
3062 @cindex rewind program state
3063
3064 On certain operating systems@footnote{Currently, only
3065 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3066 program's state, called a @dfn{checkpoint}, and come back to it
3067 later.
3068
3069 Returning to a checkpoint effectively undoes everything that has
3070 happened in the program since the @code{checkpoint} was saved. This
3071 includes changes in memory, registers, and even (within some limits)
3072 system state. Effectively, it is like going back in time to the
3073 moment when the checkpoint was saved.
3074
3075 Thus, if you're stepping thru a program and you think you're
3076 getting close to the point where things go wrong, you can save
3077 a checkpoint. Then, if you accidentally go too far and miss
3078 the critical statement, instead of having to restart your program
3079 from the beginning, you can just go back to the checkpoint and
3080 start again from there.
3081
3082 This can be especially useful if it takes a lot of time or
3083 steps to reach the point where you think the bug occurs.
3084
3085 To use the @code{checkpoint}/@code{restart} method of debugging:
3086
3087 @table @code
3088 @kindex checkpoint
3089 @item checkpoint
3090 Save a snapshot of the debugged program's current execution state.
3091 The @code{checkpoint} command takes no arguments, but each checkpoint
3092 is assigned a small integer id, similar to a breakpoint id.
3093
3094 @kindex info checkpoints
3095 @item info checkpoints
3096 List the checkpoints that have been saved in the current debugging
3097 session. For each checkpoint, the following information will be
3098 listed:
3099
3100 @table @code
3101 @item Checkpoint ID
3102 @item Process ID
3103 @item Code Address
3104 @item Source line, or label
3105 @end table
3106
3107 @kindex restart @var{checkpoint-id}
3108 @item restart @var{checkpoint-id}
3109 Restore the program state that was saved as checkpoint number
3110 @var{checkpoint-id}. All program variables, registers, stack frames
3111 etc.@: will be returned to the values that they had when the checkpoint
3112 was saved. In essence, gdb will ``wind back the clock'' to the point
3113 in time when the checkpoint was saved.
3114
3115 Note that breakpoints, @value{GDBN} variables, command history etc.
3116 are not affected by restoring a checkpoint. In general, a checkpoint
3117 only restores things that reside in the program being debugged, not in
3118 the debugger.
3119
3120 @kindex delete checkpoint @var{checkpoint-id}
3121 @item delete checkpoint @var{checkpoint-id}
3122 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3123
3124 @end table
3125
3126 Returning to a previously saved checkpoint will restore the user state
3127 of the program being debugged, plus a significant subset of the system
3128 (OS) state, including file pointers. It won't ``un-write'' data from
3129 a file, but it will rewind the file pointer to the previous location,
3130 so that the previously written data can be overwritten. For files
3131 opened in read mode, the pointer will also be restored so that the
3132 previously read data can be read again.
3133
3134 Of course, characters that have been sent to a printer (or other
3135 external device) cannot be ``snatched back'', and characters received
3136 from eg.@: a serial device can be removed from internal program buffers,
3137 but they cannot be ``pushed back'' into the serial pipeline, ready to
3138 be received again. Similarly, the actual contents of files that have
3139 been changed cannot be restored (at this time).
3140
3141 However, within those constraints, you actually can ``rewind'' your
3142 program to a previously saved point in time, and begin debugging it
3143 again --- and you can change the course of events so as to debug a
3144 different execution path this time.
3145
3146 @cindex checkpoints and process id
3147 Finally, there is one bit of internal program state that will be
3148 different when you return to a checkpoint --- the program's process
3149 id. Each checkpoint will have a unique process id (or @var{pid}),
3150 and each will be different from the program's original @var{pid}.
3151 If your program has saved a local copy of its process id, this could
3152 potentially pose a problem.
3153
3154 @subsection A Non-obvious Benefit of Using Checkpoints
3155
3156 On some systems such as @sc{gnu}/Linux, address space randomization
3157 is performed on new processes for security reasons. This makes it
3158 difficult or impossible to set a breakpoint, or watchpoint, on an
3159 absolute address if you have to restart the program, since the
3160 absolute location of a symbol will change from one execution to the
3161 next.
3162
3163 A checkpoint, however, is an @emph{identical} copy of a process.
3164 Therefore if you create a checkpoint at (eg.@:) the start of main,
3165 and simply return to that checkpoint instead of restarting the
3166 process, you can avoid the effects of address randomization and
3167 your symbols will all stay in the same place.
3168
3169 @node Stopping
3170 @chapter Stopping and Continuing
3171
3172 The principal purposes of using a debugger are so that you can stop your
3173 program before it terminates; or so that, if your program runs into
3174 trouble, you can investigate and find out why.
3175
3176 Inside @value{GDBN}, your program may stop for any of several reasons,
3177 such as a signal, a breakpoint, or reaching a new line after a
3178 @value{GDBN} command such as @code{step}. You may then examine and
3179 change variables, set new breakpoints or remove old ones, and then
3180 continue execution. Usually, the messages shown by @value{GDBN} provide
3181 ample explanation of the status of your program---but you can also
3182 explicitly request this information at any time.
3183
3184 @table @code
3185 @kindex info program
3186 @item info program
3187 Display information about the status of your program: whether it is
3188 running or not, what process it is, and why it stopped.
3189 @end table
3190
3191 @menu
3192 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3193 * Continuing and Stepping:: Resuming execution
3194 * Signals:: Signals
3195 * Thread Stops:: Stopping and starting multi-thread programs
3196 @end menu
3197
3198 @node Breakpoints
3199 @section Breakpoints, Watchpoints, and Catchpoints
3200
3201 @cindex breakpoints
3202 A @dfn{breakpoint} makes your program stop whenever a certain point in
3203 the program is reached. For each breakpoint, you can add conditions to
3204 control in finer detail whether your program stops. You can set
3205 breakpoints with the @code{break} command and its variants (@pxref{Set
3206 Breaks, ,Setting Breakpoints}), to specify the place where your program
3207 should stop by line number, function name or exact address in the
3208 program.
3209
3210 On some systems, you can set breakpoints in shared libraries before
3211 the executable is run. There is a minor limitation on HP-UX systems:
3212 you must wait until the executable is run in order to set breakpoints
3213 in shared library routines that are not called directly by the program
3214 (for example, routines that are arguments in a @code{pthread_create}
3215 call).
3216
3217 @cindex watchpoints
3218 @cindex data breakpoints
3219 @cindex memory tracing
3220 @cindex breakpoint on memory address
3221 @cindex breakpoint on variable modification
3222 A @dfn{watchpoint} is a special breakpoint that stops your program
3223 when the value of an expression changes. The expression may be a value
3224 of a variable, or it could involve values of one or more variables
3225 combined by operators, such as @samp{a + b}. This is sometimes called
3226 @dfn{data breakpoints}. You must use a different command to set
3227 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3228 from that, you can manage a watchpoint like any other breakpoint: you
3229 enable, disable, and delete both breakpoints and watchpoints using the
3230 same commands.
3231
3232 You can arrange to have values from your program displayed automatically
3233 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3234 Automatic Display}.
3235
3236 @cindex catchpoints
3237 @cindex breakpoint on events
3238 A @dfn{catchpoint} is another special breakpoint that stops your program
3239 when a certain kind of event occurs, such as the throwing of a C@t{++}
3240 exception or the loading of a library. As with watchpoints, you use a
3241 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3242 Catchpoints}), but aside from that, you can manage a catchpoint like any
3243 other breakpoint. (To stop when your program receives a signal, use the
3244 @code{handle} command; see @ref{Signals, ,Signals}.)
3245
3246 @cindex breakpoint numbers
3247 @cindex numbers for breakpoints
3248 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3249 catchpoint when you create it; these numbers are successive integers
3250 starting with one. In many of the commands for controlling various
3251 features of breakpoints you use the breakpoint number to say which
3252 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3253 @dfn{disabled}; if disabled, it has no effect on your program until you
3254 enable it again.
3255
3256 @cindex breakpoint ranges
3257 @cindex ranges of breakpoints
3258 Some @value{GDBN} commands accept a range of breakpoints on which to
3259 operate. A breakpoint range is either a single breakpoint number, like
3260 @samp{5}, or two such numbers, in increasing order, separated by a
3261 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3262 all breakpoints in that range are operated on.
3263
3264 @menu
3265 * Set Breaks:: Setting breakpoints
3266 * Set Watchpoints:: Setting watchpoints
3267 * Set Catchpoints:: Setting catchpoints
3268 * Delete Breaks:: Deleting breakpoints
3269 * Disabling:: Disabling breakpoints
3270 * Conditions:: Break conditions
3271 * Break Commands:: Breakpoint command lists
3272 * Save Breakpoints:: How to save breakpoints in a file
3273 * Error in Breakpoints:: ``Cannot insert breakpoints''
3274 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3275 @end menu
3276
3277 @node Set Breaks
3278 @subsection Setting Breakpoints
3279
3280 @c FIXME LMB what does GDB do if no code on line of breakpt?
3281 @c consider in particular declaration with/without initialization.
3282 @c
3283 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3284
3285 @kindex break
3286 @kindex b @r{(@code{break})}
3287 @vindex $bpnum@r{, convenience variable}
3288 @cindex latest breakpoint
3289 Breakpoints are set with the @code{break} command (abbreviated
3290 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3291 number of the breakpoint you've set most recently; see @ref{Convenience
3292 Vars,, Convenience Variables}, for a discussion of what you can do with
3293 convenience variables.
3294
3295 @table @code
3296 @item break @var{location}
3297 Set a breakpoint at the given @var{location}, which can specify a
3298 function name, a line number, or an address of an instruction.
3299 (@xref{Specify Location}, for a list of all the possible ways to
3300 specify a @var{location}.) The breakpoint will stop your program just
3301 before it executes any of the code in the specified @var{location}.
3302
3303 When using source languages that permit overloading of symbols, such as
3304 C@t{++}, a function name may refer to more than one possible place to break.
3305 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3306 that situation.
3307
3308 It is also possible to insert a breakpoint that will stop the program
3309 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3310 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3311
3312 @item break
3313 When called without any arguments, @code{break} sets a breakpoint at
3314 the next instruction to be executed in the selected stack frame
3315 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3316 innermost, this makes your program stop as soon as control
3317 returns to that frame. This is similar to the effect of a
3318 @code{finish} command in the frame inside the selected frame---except
3319 that @code{finish} does not leave an active breakpoint. If you use
3320 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3321 the next time it reaches the current location; this may be useful
3322 inside loops.
3323
3324 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3325 least one instruction has been executed. If it did not do this, you
3326 would be unable to proceed past a breakpoint without first disabling the
3327 breakpoint. This rule applies whether or not the breakpoint already
3328 existed when your program stopped.
3329
3330 @item break @dots{} if @var{cond}
3331 Set a breakpoint with condition @var{cond}; evaluate the expression
3332 @var{cond} each time the breakpoint is reached, and stop only if the
3333 value is nonzero---that is, if @var{cond} evaluates as true.
3334 @samp{@dots{}} stands for one of the possible arguments described
3335 above (or no argument) specifying where to break. @xref{Conditions,
3336 ,Break Conditions}, for more information on breakpoint conditions.
3337
3338 @kindex tbreak
3339 @item tbreak @var{args}
3340 Set a breakpoint enabled only for one stop. @var{args} are the
3341 same as for the @code{break} command, and the breakpoint is set in the same
3342 way, but the breakpoint is automatically deleted after the first time your
3343 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3344
3345 @kindex hbreak
3346 @cindex hardware breakpoints
3347 @item hbreak @var{args}
3348 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3349 @code{break} command and the breakpoint is set in the same way, but the
3350 breakpoint requires hardware support and some target hardware may not
3351 have this support. The main purpose of this is EPROM/ROM code
3352 debugging, so you can set a breakpoint at an instruction without
3353 changing the instruction. This can be used with the new trap-generation
3354 provided by SPARClite DSU and most x86-based targets. These targets
3355 will generate traps when a program accesses some data or instruction
3356 address that is assigned to the debug registers. However the hardware
3357 breakpoint registers can take a limited number of breakpoints. For
3358 example, on the DSU, only two data breakpoints can be set at a time, and
3359 @value{GDBN} will reject this command if more than two are used. Delete
3360 or disable unused hardware breakpoints before setting new ones
3361 (@pxref{Disabling, ,Disabling Breakpoints}).
3362 @xref{Conditions, ,Break Conditions}.
3363 For remote targets, you can restrict the number of hardware
3364 breakpoints @value{GDBN} will use, see @ref{set remote
3365 hardware-breakpoint-limit}.
3366
3367 @kindex thbreak
3368 @item thbreak @var{args}
3369 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3370 are the same as for the @code{hbreak} command and the breakpoint is set in
3371 the same way. However, like the @code{tbreak} command,
3372 the breakpoint is automatically deleted after the
3373 first time your program stops there. Also, like the @code{hbreak}
3374 command, the breakpoint requires hardware support and some target hardware
3375 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3376 See also @ref{Conditions, ,Break Conditions}.
3377
3378 @kindex rbreak
3379 @cindex regular expression
3380 @cindex breakpoints at functions matching a regexp
3381 @cindex set breakpoints in many functions
3382 @item rbreak @var{regex}
3383 Set breakpoints on all functions matching the regular expression
3384 @var{regex}. This command sets an unconditional breakpoint on all
3385 matches, printing a list of all breakpoints it set. Once these
3386 breakpoints are set, they are treated just like the breakpoints set with
3387 the @code{break} command. You can delete them, disable them, or make
3388 them conditional the same way as any other breakpoint.
3389
3390 The syntax of the regular expression is the standard one used with tools
3391 like @file{grep}. Note that this is different from the syntax used by
3392 shells, so for instance @code{foo*} matches all functions that include
3393 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3394 @code{.*} leading and trailing the regular expression you supply, so to
3395 match only functions that begin with @code{foo}, use @code{^foo}.
3396
3397 @cindex non-member C@t{++} functions, set breakpoint in
3398 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3399 breakpoints on overloaded functions that are not members of any special
3400 classes.
3401
3402 @cindex set breakpoints on all functions
3403 The @code{rbreak} command can be used to set breakpoints in
3404 @strong{all} the functions in a program, like this:
3405
3406 @smallexample
3407 (@value{GDBP}) rbreak .
3408 @end smallexample
3409
3410 @item rbreak @var{file}:@var{regex}
3411 If @code{rbreak} is called with a filename qualification, it limits
3412 the search for functions matching the given regular expression to the
3413 specified @var{file}. This can be used, for example, to set breakpoints on
3414 every function in a given file:
3415
3416 @smallexample
3417 (@value{GDBP}) rbreak file.c:.
3418 @end smallexample
3419
3420 The colon separating the filename qualifier from the regex may
3421 optionally be surrounded by spaces.
3422
3423 @kindex info breakpoints
3424 @cindex @code{$_} and @code{info breakpoints}
3425 @item info breakpoints @r{[}@var{n}@r{]}
3426 @itemx info break @r{[}@var{n}@r{]}
3427 Print a table of all breakpoints, watchpoints, and catchpoints set and
3428 not deleted. Optional argument @var{n} means print information only
3429 about the specified breakpoint (or watchpoint or catchpoint). For
3430 each breakpoint, following columns are printed:
3431
3432 @table @emph
3433 @item Breakpoint Numbers
3434 @item Type
3435 Breakpoint, watchpoint, or catchpoint.
3436 @item Disposition
3437 Whether the breakpoint is marked to be disabled or deleted when hit.
3438 @item Enabled or Disabled
3439 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3440 that are not enabled.
3441 @item Address
3442 Where the breakpoint is in your program, as a memory address. For a
3443 pending breakpoint whose address is not yet known, this field will
3444 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3445 library that has the symbol or line referred by breakpoint is loaded.
3446 See below for details. A breakpoint with several locations will
3447 have @samp{<MULTIPLE>} in this field---see below for details.
3448 @item What
3449 Where the breakpoint is in the source for your program, as a file and
3450 line number. For a pending breakpoint, the original string passed to
3451 the breakpoint command will be listed as it cannot be resolved until
3452 the appropriate shared library is loaded in the future.
3453 @end table
3454
3455 @noindent
3456 If a breakpoint is conditional, @code{info break} shows the condition on
3457 the line following the affected breakpoint; breakpoint commands, if any,
3458 are listed after that. A pending breakpoint is allowed to have a condition
3459 specified for it. The condition is not parsed for validity until a shared
3460 library is loaded that allows the pending breakpoint to resolve to a
3461 valid location.
3462
3463 @noindent
3464 @code{info break} with a breakpoint
3465 number @var{n} as argument lists only that breakpoint. The
3466 convenience variable @code{$_} and the default examining-address for
3467 the @code{x} command are set to the address of the last breakpoint
3468 listed (@pxref{Memory, ,Examining Memory}).
3469
3470 @noindent
3471 @code{info break} displays a count of the number of times the breakpoint
3472 has been hit. This is especially useful in conjunction with the
3473 @code{ignore} command. You can ignore a large number of breakpoint
3474 hits, look at the breakpoint info to see how many times the breakpoint
3475 was hit, and then run again, ignoring one less than that number. This
3476 will get you quickly to the last hit of that breakpoint.
3477 @end table
3478
3479 @value{GDBN} allows you to set any number of breakpoints at the same place in
3480 your program. There is nothing silly or meaningless about this. When
3481 the breakpoints are conditional, this is even useful
3482 (@pxref{Conditions, ,Break Conditions}).
3483
3484 @cindex multiple locations, breakpoints
3485 @cindex breakpoints, multiple locations
3486 It is possible that a breakpoint corresponds to several locations
3487 in your program. Examples of this situation are:
3488
3489 @itemize @bullet
3490 @item
3491 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3492 instances of the function body, used in different cases.
3493
3494 @item
3495 For a C@t{++} template function, a given line in the function can
3496 correspond to any number of instantiations.
3497
3498 @item
3499 For an inlined function, a given source line can correspond to
3500 several places where that function is inlined.
3501 @end itemize
3502
3503 In all those cases, @value{GDBN} will insert a breakpoint at all
3504 the relevant locations@footnote{
3505 As of this writing, multiple-location breakpoints work only if there's
3506 line number information for all the locations. This means that they
3507 will generally not work in system libraries, unless you have debug
3508 info with line numbers for them.}.
3509
3510 A breakpoint with multiple locations is displayed in the breakpoint
3511 table using several rows---one header row, followed by one row for
3512 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3513 address column. The rows for individual locations contain the actual
3514 addresses for locations, and show the functions to which those
3515 locations belong. The number column for a location is of the form
3516 @var{breakpoint-number}.@var{location-number}.
3517
3518 For example:
3519
3520 @smallexample
3521 Num Type Disp Enb Address What
3522 1 breakpoint keep y <MULTIPLE>
3523 stop only if i==1
3524 breakpoint already hit 1 time
3525 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3526 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3527 @end smallexample
3528
3529 Each location can be individually enabled or disabled by passing
3530 @var{breakpoint-number}.@var{location-number} as argument to the
3531 @code{enable} and @code{disable} commands. Note that you cannot
3532 delete the individual locations from the list, you can only delete the
3533 entire list of locations that belong to their parent breakpoint (with
3534 the @kbd{delete @var{num}} command, where @var{num} is the number of
3535 the parent breakpoint, 1 in the above example). Disabling or enabling
3536 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3537 that belong to that breakpoint.
3538
3539 @cindex pending breakpoints
3540 It's quite common to have a breakpoint inside a shared library.
3541 Shared libraries can be loaded and unloaded explicitly,
3542 and possibly repeatedly, as the program is executed. To support
3543 this use case, @value{GDBN} updates breakpoint locations whenever
3544 any shared library is loaded or unloaded. Typically, you would
3545 set a breakpoint in a shared library at the beginning of your
3546 debugging session, when the library is not loaded, and when the
3547 symbols from the library are not available. When you try to set
3548 breakpoint, @value{GDBN} will ask you if you want to set
3549 a so called @dfn{pending breakpoint}---breakpoint whose address
3550 is not yet resolved.
3551
3552 After the program is run, whenever a new shared library is loaded,
3553 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3554 shared library contains the symbol or line referred to by some
3555 pending breakpoint, that breakpoint is resolved and becomes an
3556 ordinary breakpoint. When a library is unloaded, all breakpoints
3557 that refer to its symbols or source lines become pending again.
3558
3559 This logic works for breakpoints with multiple locations, too. For
3560 example, if you have a breakpoint in a C@t{++} template function, and
3561 a newly loaded shared library has an instantiation of that template,
3562 a new location is added to the list of locations for the breakpoint.
3563
3564 Except for having unresolved address, pending breakpoints do not
3565 differ from regular breakpoints. You can set conditions or commands,
3566 enable and disable them and perform other breakpoint operations.
3567
3568 @value{GDBN} provides some additional commands for controlling what
3569 happens when the @samp{break} command cannot resolve breakpoint
3570 address specification to an address:
3571
3572 @kindex set breakpoint pending
3573 @kindex show breakpoint pending
3574 @table @code
3575 @item set breakpoint pending auto
3576 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3577 location, it queries you whether a pending breakpoint should be created.
3578
3579 @item set breakpoint pending on
3580 This indicates that an unrecognized breakpoint location should automatically
3581 result in a pending breakpoint being created.
3582
3583 @item set breakpoint pending off
3584 This indicates that pending breakpoints are not to be created. Any
3585 unrecognized breakpoint location results in an error. This setting does
3586 not affect any pending breakpoints previously created.
3587
3588 @item show breakpoint pending
3589 Show the current behavior setting for creating pending breakpoints.
3590 @end table
3591
3592 The settings above only affect the @code{break} command and its
3593 variants. Once breakpoint is set, it will be automatically updated
3594 as shared libraries are loaded and unloaded.
3595
3596 @cindex automatic hardware breakpoints
3597 For some targets, @value{GDBN} can automatically decide if hardware or
3598 software breakpoints should be used, depending on whether the
3599 breakpoint address is read-only or read-write. This applies to
3600 breakpoints set with the @code{break} command as well as to internal
3601 breakpoints set by commands like @code{next} and @code{finish}. For
3602 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3603 breakpoints.
3604
3605 You can control this automatic behaviour with the following commands::
3606
3607 @kindex set breakpoint auto-hw
3608 @kindex show breakpoint auto-hw
3609 @table @code
3610 @item set breakpoint auto-hw on
3611 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3612 will try to use the target memory map to decide if software or hardware
3613 breakpoint must be used.
3614
3615 @item set breakpoint auto-hw off
3616 This indicates @value{GDBN} should not automatically select breakpoint
3617 type. If the target provides a memory map, @value{GDBN} will warn when
3618 trying to set software breakpoint at a read-only address.
3619 @end table
3620
3621 @value{GDBN} normally implements breakpoints by replacing the program code
3622 at the breakpoint address with a special instruction, which, when
3623 executed, given control to the debugger. By default, the program
3624 code is so modified only when the program is resumed. As soon as
3625 the program stops, @value{GDBN} restores the original instructions. This
3626 behaviour guards against leaving breakpoints inserted in the
3627 target should gdb abrubptly disconnect. However, with slow remote
3628 targets, inserting and removing breakpoint can reduce the performance.
3629 This behavior can be controlled with the following commands::
3630
3631 @kindex set breakpoint always-inserted
3632 @kindex show breakpoint always-inserted
3633 @table @code
3634 @item set breakpoint always-inserted off
3635 All breakpoints, including newly added by the user, are inserted in
3636 the target only when the target is resumed. All breakpoints are
3637 removed from the target when it stops.
3638
3639 @item set breakpoint always-inserted on
3640 Causes all breakpoints to be inserted in the target at all times. If
3641 the user adds a new breakpoint, or changes an existing breakpoint, the
3642 breakpoints in the target are updated immediately. A breakpoint is
3643 removed from the target only when breakpoint itself is removed.
3644
3645 @cindex non-stop mode, and @code{breakpoint always-inserted}
3646 @item set breakpoint always-inserted auto
3647 This is the default mode. If @value{GDBN} is controlling the inferior
3648 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3649 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3650 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3651 @code{breakpoint always-inserted} mode is off.
3652 @end table
3653
3654 @cindex negative breakpoint numbers
3655 @cindex internal @value{GDBN} breakpoints
3656 @value{GDBN} itself sometimes sets breakpoints in your program for
3657 special purposes, such as proper handling of @code{longjmp} (in C
3658 programs). These internal breakpoints are assigned negative numbers,
3659 starting with @code{-1}; @samp{info breakpoints} does not display them.
3660 You can see these breakpoints with the @value{GDBN} maintenance command
3661 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3662
3663
3664 @node Set Watchpoints
3665 @subsection Setting Watchpoints
3666
3667 @cindex setting watchpoints
3668 You can use a watchpoint to stop execution whenever the value of an
3669 expression changes, without having to predict a particular place where
3670 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3671 The expression may be as simple as the value of a single variable, or
3672 as complex as many variables combined by operators. Examples include:
3673
3674 @itemize @bullet
3675 @item
3676 A reference to the value of a single variable.
3677
3678 @item
3679 An address cast to an appropriate data type. For example,
3680 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3681 address (assuming an @code{int} occupies 4 bytes).
3682
3683 @item
3684 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3685 expression can use any operators valid in the program's native
3686 language (@pxref{Languages}).
3687 @end itemize
3688
3689 You can set a watchpoint on an expression even if the expression can
3690 not be evaluated yet. For instance, you can set a watchpoint on
3691 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3692 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3693 the expression produces a valid value. If the expression becomes
3694 valid in some other way than changing a variable (e.g.@: if the memory
3695 pointed to by @samp{*global_ptr} becomes readable as the result of a
3696 @code{malloc} call), @value{GDBN} may not stop until the next time
3697 the expression changes.
3698
3699 @cindex software watchpoints
3700 @cindex hardware watchpoints
3701 Depending on your system, watchpoints may be implemented in software or
3702 hardware. @value{GDBN} does software watchpointing by single-stepping your
3703 program and testing the variable's value each time, which is hundreds of
3704 times slower than normal execution. (But this may still be worth it, to
3705 catch errors where you have no clue what part of your program is the
3706 culprit.)
3707
3708 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3709 x86-based targets, @value{GDBN} includes support for hardware
3710 watchpoints, which do not slow down the running of your program.
3711
3712 @table @code
3713 @kindex watch
3714 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]}
3715 Set a watchpoint for an expression. @value{GDBN} will break when the
3716 expression @var{expr} is written into by the program and its value
3717 changes. The simplest (and the most popular) use of this command is
3718 to watch the value of a single variable:
3719
3720 @smallexample
3721 (@value{GDBP}) watch foo
3722 @end smallexample
3723
3724 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3725 clause, @value{GDBN} breaks only when the thread identified by
3726 @var{threadnum} changes the value of @var{expr}. If any other threads
3727 change the value of @var{expr}, @value{GDBN} will not break. Note
3728 that watchpoints restricted to a single thread in this way only work
3729 with Hardware Watchpoints.
3730
3731 Ordinarily a watchpoint respects the scope of variables in @var{expr}
3732 (see below). The @code{-location} argument tells @value{GDBN} to
3733 instead watch the memory referred to by @var{expr}. In this case,
3734 @value{GDBN} will evaluate @var{expr}, take the address of the result,
3735 and watch the memory at that address. The type of the result is used
3736 to determine the size of the watched memory. If the expression's
3737 result does not have an address, then @value{GDBN} will print an
3738 error.
3739
3740 @kindex rwatch
3741 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]}
3742 Set a watchpoint that will break when the value of @var{expr} is read
3743 by the program.
3744
3745 @kindex awatch
3746 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]}
3747 Set a watchpoint that will break when @var{expr} is either read from
3748 or written into by the program.
3749
3750 @kindex info watchpoints @r{[}@var{n}@r{]}
3751 @item info watchpoints
3752 This command prints a list of watchpoints, using the same format as
3753 @code{info break} (@pxref{Set Breaks}).
3754 @end table
3755
3756 If you watch for a change in a numerically entered address you need to
3757 dereference it, as the address itself is just a constant number which will
3758 never change. @value{GDBN} refuses to create a watchpoint that watches
3759 a never-changing value:
3760
3761 @smallexample
3762 (@value{GDBP}) watch 0x600850
3763 Cannot watch constant value 0x600850.
3764 (@value{GDBP}) watch *(int *) 0x600850
3765 Watchpoint 1: *(int *) 6293584
3766 @end smallexample
3767
3768 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3769 watchpoints execute very quickly, and the debugger reports a change in
3770 value at the exact instruction where the change occurs. If @value{GDBN}
3771 cannot set a hardware watchpoint, it sets a software watchpoint, which
3772 executes more slowly and reports the change in value at the next
3773 @emph{statement}, not the instruction, after the change occurs.
3774
3775 @cindex use only software watchpoints
3776 You can force @value{GDBN} to use only software watchpoints with the
3777 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3778 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3779 the underlying system supports them. (Note that hardware-assisted
3780 watchpoints that were set @emph{before} setting
3781 @code{can-use-hw-watchpoints} to zero will still use the hardware
3782 mechanism of watching expression values.)
3783
3784 @table @code
3785 @item set can-use-hw-watchpoints
3786 @kindex set can-use-hw-watchpoints
3787 Set whether or not to use hardware watchpoints.
3788
3789 @item show can-use-hw-watchpoints
3790 @kindex show can-use-hw-watchpoints
3791 Show the current mode of using hardware watchpoints.
3792 @end table
3793
3794 For remote targets, you can restrict the number of hardware
3795 watchpoints @value{GDBN} will use, see @ref{set remote
3796 hardware-breakpoint-limit}.
3797
3798 When you issue the @code{watch} command, @value{GDBN} reports
3799
3800 @smallexample
3801 Hardware watchpoint @var{num}: @var{expr}
3802 @end smallexample
3803
3804 @noindent
3805 if it was able to set a hardware watchpoint.
3806
3807 Currently, the @code{awatch} and @code{rwatch} commands can only set
3808 hardware watchpoints, because accesses to data that don't change the
3809 value of the watched expression cannot be detected without examining
3810 every instruction as it is being executed, and @value{GDBN} does not do
3811 that currently. If @value{GDBN} finds that it is unable to set a
3812 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3813 will print a message like this:
3814
3815 @smallexample
3816 Expression cannot be implemented with read/access watchpoint.
3817 @end smallexample
3818
3819 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3820 data type of the watched expression is wider than what a hardware
3821 watchpoint on the target machine can handle. For example, some systems
3822 can only watch regions that are up to 4 bytes wide; on such systems you
3823 cannot set hardware watchpoints for an expression that yields a
3824 double-precision floating-point number (which is typically 8 bytes
3825 wide). As a work-around, it might be possible to break the large region
3826 into a series of smaller ones and watch them with separate watchpoints.
3827
3828 If you set too many hardware watchpoints, @value{GDBN} might be unable
3829 to insert all of them when you resume the execution of your program.
3830 Since the precise number of active watchpoints is unknown until such
3831 time as the program is about to be resumed, @value{GDBN} might not be
3832 able to warn you about this when you set the watchpoints, and the
3833 warning will be printed only when the program is resumed:
3834
3835 @smallexample
3836 Hardware watchpoint @var{num}: Could not insert watchpoint
3837 @end smallexample
3838
3839 @noindent
3840 If this happens, delete or disable some of the watchpoints.
3841
3842 Watching complex expressions that reference many variables can also
3843 exhaust the resources available for hardware-assisted watchpoints.
3844 That's because @value{GDBN} needs to watch every variable in the
3845 expression with separately allocated resources.
3846
3847 If you call a function interactively using @code{print} or @code{call},
3848 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3849 kind of breakpoint or the call completes.
3850
3851 @value{GDBN} automatically deletes watchpoints that watch local
3852 (automatic) variables, or expressions that involve such variables, when
3853 they go out of scope, that is, when the execution leaves the block in
3854 which these variables were defined. In particular, when the program
3855 being debugged terminates, @emph{all} local variables go out of scope,
3856 and so only watchpoints that watch global variables remain set. If you
3857 rerun the program, you will need to set all such watchpoints again. One
3858 way of doing that would be to set a code breakpoint at the entry to the
3859 @code{main} function and when it breaks, set all the watchpoints.
3860
3861 @cindex watchpoints and threads
3862 @cindex threads and watchpoints
3863 In multi-threaded programs, watchpoints will detect changes to the
3864 watched expression from every thread.
3865
3866 @quotation
3867 @emph{Warning:} In multi-threaded programs, software watchpoints
3868 have only limited usefulness. If @value{GDBN} creates a software
3869 watchpoint, it can only watch the value of an expression @emph{in a
3870 single thread}. If you are confident that the expression can only
3871 change due to the current thread's activity (and if you are also
3872 confident that no other thread can become current), then you can use
3873 software watchpoints as usual. However, @value{GDBN} may not notice
3874 when a non-current thread's activity changes the expression. (Hardware
3875 watchpoints, in contrast, watch an expression in all threads.)
3876 @end quotation
3877
3878 @xref{set remote hardware-watchpoint-limit}.
3879
3880 @node Set Catchpoints
3881 @subsection Setting Catchpoints
3882 @cindex catchpoints, setting
3883 @cindex exception handlers
3884 @cindex event handling
3885
3886 You can use @dfn{catchpoints} to cause the debugger to stop for certain
3887 kinds of program events, such as C@t{++} exceptions or the loading of a
3888 shared library. Use the @code{catch} command to set a catchpoint.
3889
3890 @table @code
3891 @kindex catch
3892 @item catch @var{event}
3893 Stop when @var{event} occurs. @var{event} can be any of the following:
3894 @table @code
3895 @item throw
3896 @cindex stop on C@t{++} exceptions
3897 The throwing of a C@t{++} exception.
3898
3899 @item catch
3900 The catching of a C@t{++} exception.
3901
3902 @item exception
3903 @cindex Ada exception catching
3904 @cindex catch Ada exceptions
3905 An Ada exception being raised. If an exception name is specified
3906 at the end of the command (eg @code{catch exception Program_Error}),
3907 the debugger will stop only when this specific exception is raised.
3908 Otherwise, the debugger stops execution when any Ada exception is raised.
3909
3910 When inserting an exception catchpoint on a user-defined exception whose
3911 name is identical to one of the exceptions defined by the language, the
3912 fully qualified name must be used as the exception name. Otherwise,
3913 @value{GDBN} will assume that it should stop on the pre-defined exception
3914 rather than the user-defined one. For instance, assuming an exception
3915 called @code{Constraint_Error} is defined in package @code{Pck}, then
3916 the command to use to catch such exceptions is @kbd{catch exception
3917 Pck.Constraint_Error}.
3918
3919 @item exception unhandled
3920 An exception that was raised but is not handled by the program.
3921
3922 @item assert
3923 A failed Ada assertion.
3924
3925 @item exec
3926 @cindex break on fork/exec
3927 A call to @code{exec}. This is currently only available for HP-UX
3928 and @sc{gnu}/Linux.
3929
3930 @item syscall
3931 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @dots{}
3932 @cindex break on a system call.
3933 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
3934 syscall is a mechanism for application programs to request a service
3935 from the operating system (OS) or one of the OS system services.
3936 @value{GDBN} can catch some or all of the syscalls issued by the
3937 debuggee, and show the related information for each syscall. If no
3938 argument is specified, calls to and returns from all system calls
3939 will be caught.
3940
3941 @var{name} can be any system call name that is valid for the
3942 underlying OS. Just what syscalls are valid depends on the OS. On
3943 GNU and Unix systems, you can find the full list of valid syscall
3944 names on @file{/usr/include/asm/unistd.h}.
3945
3946 @c For MS-Windows, the syscall names and the corresponding numbers
3947 @c can be found, e.g., on this URL:
3948 @c http://www.metasploit.com/users/opcode/syscalls.html
3949 @c but we don't support Windows syscalls yet.
3950
3951 Normally, @value{GDBN} knows in advance which syscalls are valid for
3952 each OS, so you can use the @value{GDBN} command-line completion
3953 facilities (@pxref{Completion,, command completion}) to list the
3954 available choices.
3955
3956 You may also specify the system call numerically. A syscall's
3957 number is the value passed to the OS's syscall dispatcher to
3958 identify the requested service. When you specify the syscall by its
3959 name, @value{GDBN} uses its database of syscalls to convert the name
3960 into the corresponding numeric code, but using the number directly
3961 may be useful if @value{GDBN}'s database does not have the complete
3962 list of syscalls on your system (e.g., because @value{GDBN} lags
3963 behind the OS upgrades).
3964
3965 The example below illustrates how this command works if you don't provide
3966 arguments to it:
3967
3968 @smallexample
3969 (@value{GDBP}) catch syscall
3970 Catchpoint 1 (syscall)
3971 (@value{GDBP}) r
3972 Starting program: /tmp/catch-syscall
3973
3974 Catchpoint 1 (call to syscall 'close'), \
3975 0xffffe424 in __kernel_vsyscall ()
3976 (@value{GDBP}) c
3977 Continuing.
3978
3979 Catchpoint 1 (returned from syscall 'close'), \
3980 0xffffe424 in __kernel_vsyscall ()
3981 (@value{GDBP})
3982 @end smallexample
3983
3984 Here is an example of catching a system call by name:
3985
3986 @smallexample
3987 (@value{GDBP}) catch syscall chroot
3988 Catchpoint 1 (syscall 'chroot' [61])
3989 (@value{GDBP}) r
3990 Starting program: /tmp/catch-syscall
3991
3992 Catchpoint 1 (call to syscall 'chroot'), \
3993 0xffffe424 in __kernel_vsyscall ()
3994 (@value{GDBP}) c
3995 Continuing.
3996
3997 Catchpoint 1 (returned from syscall 'chroot'), \
3998 0xffffe424 in __kernel_vsyscall ()
3999 (@value{GDBP})
4000 @end smallexample
4001
4002 An example of specifying a system call numerically. In the case
4003 below, the syscall number has a corresponding entry in the XML
4004 file, so @value{GDBN} finds its name and prints it:
4005
4006 @smallexample
4007 (@value{GDBP}) catch syscall 252
4008 Catchpoint 1 (syscall(s) 'exit_group')
4009 (@value{GDBP}) r
4010 Starting program: /tmp/catch-syscall
4011
4012 Catchpoint 1 (call to syscall 'exit_group'), \
4013 0xffffe424 in __kernel_vsyscall ()
4014 (@value{GDBP}) c
4015 Continuing.
4016
4017 Program exited normally.
4018 (@value{GDBP})
4019 @end smallexample
4020
4021 However, there can be situations when there is no corresponding name
4022 in XML file for that syscall number. In this case, @value{GDBN} prints
4023 a warning message saying that it was not able to find the syscall name,
4024 but the catchpoint will be set anyway. See the example below:
4025
4026 @smallexample
4027 (@value{GDBP}) catch syscall 764
4028 warning: The number '764' does not represent a known syscall.
4029 Catchpoint 2 (syscall 764)
4030 (@value{GDBP})
4031 @end smallexample
4032
4033 If you configure @value{GDBN} using the @samp{--without-expat} option,
4034 it will not be able to display syscall names. Also, if your
4035 architecture does not have an XML file describing its system calls,
4036 you will not be able to see the syscall names. It is important to
4037 notice that these two features are used for accessing the syscall
4038 name database. In either case, you will see a warning like this:
4039
4040 @smallexample
4041 (@value{GDBP}) catch syscall
4042 warning: Could not open "syscalls/i386-linux.xml"
4043 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4044 GDB will not be able to display syscall names.
4045 Catchpoint 1 (syscall)
4046 (@value{GDBP})
4047 @end smallexample
4048
4049 Of course, the file name will change depending on your architecture and system.
4050
4051 Still using the example above, you can also try to catch a syscall by its
4052 number. In this case, you would see something like:
4053
4054 @smallexample
4055 (@value{GDBP}) catch syscall 252
4056 Catchpoint 1 (syscall(s) 252)
4057 @end smallexample
4058
4059 Again, in this case @value{GDBN} would not be able to display syscall's names.
4060
4061 @item fork
4062 A call to @code{fork}. This is currently only available for HP-UX
4063 and @sc{gnu}/Linux.
4064
4065 @item vfork
4066 A call to @code{vfork}. This is currently only available for HP-UX
4067 and @sc{gnu}/Linux.
4068
4069 @end table
4070
4071 @item tcatch @var{event}
4072 Set a catchpoint that is enabled only for one stop. The catchpoint is
4073 automatically deleted after the first time the event is caught.
4074
4075 @end table
4076
4077 Use the @code{info break} command to list the current catchpoints.
4078
4079 There are currently some limitations to C@t{++} exception handling
4080 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
4081
4082 @itemize @bullet
4083 @item
4084 If you call a function interactively, @value{GDBN} normally returns
4085 control to you when the function has finished executing. If the call
4086 raises an exception, however, the call may bypass the mechanism that
4087 returns control to you and cause your program either to abort or to
4088 simply continue running until it hits a breakpoint, catches a signal
4089 that @value{GDBN} is listening for, or exits. This is the case even if
4090 you set a catchpoint for the exception; catchpoints on exceptions are
4091 disabled within interactive calls.
4092
4093 @item
4094 You cannot raise an exception interactively.
4095
4096 @item
4097 You cannot install an exception handler interactively.
4098 @end itemize
4099
4100 @cindex raise exceptions
4101 Sometimes @code{catch} is not the best way to debug exception handling:
4102 if you need to know exactly where an exception is raised, it is better to
4103 stop @emph{before} the exception handler is called, since that way you
4104 can see the stack before any unwinding takes place. If you set a
4105 breakpoint in an exception handler instead, it may not be easy to find
4106 out where the exception was raised.
4107
4108 To stop just before an exception handler is called, you need some
4109 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
4110 raised by calling a library function named @code{__raise_exception}
4111 which has the following ANSI C interface:
4112
4113 @smallexample
4114 /* @var{addr} is where the exception identifier is stored.
4115 @var{id} is the exception identifier. */
4116 void __raise_exception (void **addr, void *id);
4117 @end smallexample
4118
4119 @noindent
4120 To make the debugger catch all exceptions before any stack
4121 unwinding takes place, set a breakpoint on @code{__raise_exception}
4122 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
4123
4124 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
4125 that depends on the value of @var{id}, you can stop your program when
4126 a specific exception is raised. You can use multiple conditional
4127 breakpoints to stop your program when any of a number of exceptions are
4128 raised.
4129
4130
4131 @node Delete Breaks
4132 @subsection Deleting Breakpoints
4133
4134 @cindex clearing breakpoints, watchpoints, catchpoints
4135 @cindex deleting breakpoints, watchpoints, catchpoints
4136 It is often necessary to eliminate a breakpoint, watchpoint, or
4137 catchpoint once it has done its job and you no longer want your program
4138 to stop there. This is called @dfn{deleting} the breakpoint. A
4139 breakpoint that has been deleted no longer exists; it is forgotten.
4140
4141 With the @code{clear} command you can delete breakpoints according to
4142 where they are in your program. With the @code{delete} command you can
4143 delete individual breakpoints, watchpoints, or catchpoints by specifying
4144 their breakpoint numbers.
4145
4146 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4147 automatically ignores breakpoints on the first instruction to be executed
4148 when you continue execution without changing the execution address.
4149
4150 @table @code
4151 @kindex clear
4152 @item clear
4153 Delete any breakpoints at the next instruction to be executed in the
4154 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4155 the innermost frame is selected, this is a good way to delete a
4156 breakpoint where your program just stopped.
4157
4158 @item clear @var{location}
4159 Delete any breakpoints set at the specified @var{location}.
4160 @xref{Specify Location}, for the various forms of @var{location}; the
4161 most useful ones are listed below:
4162
4163 @table @code
4164 @item clear @var{function}
4165 @itemx clear @var{filename}:@var{function}
4166 Delete any breakpoints set at entry to the named @var{function}.
4167
4168 @item clear @var{linenum}
4169 @itemx clear @var{filename}:@var{linenum}
4170 Delete any breakpoints set at or within the code of the specified
4171 @var{linenum} of the specified @var{filename}.
4172 @end table
4173
4174 @cindex delete breakpoints
4175 @kindex delete
4176 @kindex d @r{(@code{delete})}
4177 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4178 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4179 ranges specified as arguments. If no argument is specified, delete all
4180 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4181 confirm off}). You can abbreviate this command as @code{d}.
4182 @end table
4183
4184 @node Disabling
4185 @subsection Disabling Breakpoints
4186
4187 @cindex enable/disable a breakpoint
4188 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4189 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4190 it had been deleted, but remembers the information on the breakpoint so
4191 that you can @dfn{enable} it again later.
4192
4193 You disable and enable breakpoints, watchpoints, and catchpoints with
4194 the @code{enable} and @code{disable} commands, optionally specifying
4195 one or more breakpoint numbers as arguments. Use @code{info break} to
4196 print a list of all breakpoints, watchpoints, and catchpoints if you
4197 do not know which numbers to use.
4198
4199 Disabling and enabling a breakpoint that has multiple locations
4200 affects all of its locations.
4201
4202 A breakpoint, watchpoint, or catchpoint can have any of four different
4203 states of enablement:
4204
4205 @itemize @bullet
4206 @item
4207 Enabled. The breakpoint stops your program. A breakpoint set
4208 with the @code{break} command starts out in this state.
4209 @item
4210 Disabled. The breakpoint has no effect on your program.
4211 @item
4212 Enabled once. The breakpoint stops your program, but then becomes
4213 disabled.
4214 @item
4215 Enabled for deletion. The breakpoint stops your program, but
4216 immediately after it does so it is deleted permanently. A breakpoint
4217 set with the @code{tbreak} command starts out in this state.
4218 @end itemize
4219
4220 You can use the following commands to enable or disable breakpoints,
4221 watchpoints, and catchpoints:
4222
4223 @table @code
4224 @kindex disable
4225 @kindex dis @r{(@code{disable})}
4226 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4227 Disable the specified breakpoints---or all breakpoints, if none are
4228 listed. A disabled breakpoint has no effect but is not forgotten. All
4229 options such as ignore-counts, conditions and commands are remembered in
4230 case the breakpoint is enabled again later. You may abbreviate
4231 @code{disable} as @code{dis}.
4232
4233 @kindex enable
4234 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4235 Enable the specified breakpoints (or all defined breakpoints). They
4236 become effective once again in stopping your program.
4237
4238 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4239 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4240 of these breakpoints immediately after stopping your program.
4241
4242 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4243 Enable the specified breakpoints to work once, then die. @value{GDBN}
4244 deletes any of these breakpoints as soon as your program stops there.
4245 Breakpoints set by the @code{tbreak} command start out in this state.
4246 @end table
4247
4248 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4249 @c confusing: tbreak is also initially enabled.
4250 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4251 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4252 subsequently, they become disabled or enabled only when you use one of
4253 the commands above. (The command @code{until} can set and delete a
4254 breakpoint of its own, but it does not change the state of your other
4255 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4256 Stepping}.)
4257
4258 @node Conditions
4259 @subsection Break Conditions
4260 @cindex conditional breakpoints
4261 @cindex breakpoint conditions
4262
4263 @c FIXME what is scope of break condition expr? Context where wanted?
4264 @c in particular for a watchpoint?
4265 The simplest sort of breakpoint breaks every time your program reaches a
4266 specified place. You can also specify a @dfn{condition} for a
4267 breakpoint. A condition is just a Boolean expression in your
4268 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4269 a condition evaluates the expression each time your program reaches it,
4270 and your program stops only if the condition is @emph{true}.
4271
4272 This is the converse of using assertions for program validation; in that
4273 situation, you want to stop when the assertion is violated---that is,
4274 when the condition is false. In C, if you want to test an assertion expressed
4275 by the condition @var{assert}, you should set the condition
4276 @samp{! @var{assert}} on the appropriate breakpoint.
4277
4278 Conditions are also accepted for watchpoints; you may not need them,
4279 since a watchpoint is inspecting the value of an expression anyhow---but
4280 it might be simpler, say, to just set a watchpoint on a variable name,
4281 and specify a condition that tests whether the new value is an interesting
4282 one.
4283
4284 Break conditions can have side effects, and may even call functions in
4285 your program. This can be useful, for example, to activate functions
4286 that log program progress, or to use your own print functions to
4287 format special data structures. The effects are completely predictable
4288 unless there is another enabled breakpoint at the same address. (In
4289 that case, @value{GDBN} might see the other breakpoint first and stop your
4290 program without checking the condition of this one.) Note that
4291 breakpoint commands are usually more convenient and flexible than break
4292 conditions for the
4293 purpose of performing side effects when a breakpoint is reached
4294 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4295
4296 Break conditions can be specified when a breakpoint is set, by using
4297 @samp{if} in the arguments to the @code{break} command. @xref{Set
4298 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4299 with the @code{condition} command.
4300
4301 You can also use the @code{if} keyword with the @code{watch} command.
4302 The @code{catch} command does not recognize the @code{if} keyword;
4303 @code{condition} is the only way to impose a further condition on a
4304 catchpoint.
4305
4306 @table @code
4307 @kindex condition
4308 @item condition @var{bnum} @var{expression}
4309 Specify @var{expression} as the break condition for breakpoint,
4310 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4311 breakpoint @var{bnum} stops your program only if the value of
4312 @var{expression} is true (nonzero, in C). When you use
4313 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4314 syntactic correctness, and to determine whether symbols in it have
4315 referents in the context of your breakpoint. If @var{expression} uses
4316 symbols not referenced in the context of the breakpoint, @value{GDBN}
4317 prints an error message:
4318
4319 @smallexample
4320 No symbol "foo" in current context.
4321 @end smallexample
4322
4323 @noindent
4324 @value{GDBN} does
4325 not actually evaluate @var{expression} at the time the @code{condition}
4326 command (or a command that sets a breakpoint with a condition, like
4327 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4328
4329 @item condition @var{bnum}
4330 Remove the condition from breakpoint number @var{bnum}. It becomes
4331 an ordinary unconditional breakpoint.
4332 @end table
4333
4334 @cindex ignore count (of breakpoint)
4335 A special case of a breakpoint condition is to stop only when the
4336 breakpoint has been reached a certain number of times. This is so
4337 useful that there is a special way to do it, using the @dfn{ignore
4338 count} of the breakpoint. Every breakpoint has an ignore count, which
4339 is an integer. Most of the time, the ignore count is zero, and
4340 therefore has no effect. But if your program reaches a breakpoint whose
4341 ignore count is positive, then instead of stopping, it just decrements
4342 the ignore count by one and continues. As a result, if the ignore count
4343 value is @var{n}, the breakpoint does not stop the next @var{n} times
4344 your program reaches it.
4345
4346 @table @code
4347 @kindex ignore
4348 @item ignore @var{bnum} @var{count}
4349 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4350 The next @var{count} times the breakpoint is reached, your program's
4351 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4352 takes no action.
4353
4354 To make the breakpoint stop the next time it is reached, specify
4355 a count of zero.
4356
4357 When you use @code{continue} to resume execution of your program from a
4358 breakpoint, you can specify an ignore count directly as an argument to
4359 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4360 Stepping,,Continuing and Stepping}.
4361
4362 If a breakpoint has a positive ignore count and a condition, the
4363 condition is not checked. Once the ignore count reaches zero,
4364 @value{GDBN} resumes checking the condition.
4365
4366 You could achieve the effect of the ignore count with a condition such
4367 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4368 is decremented each time. @xref{Convenience Vars, ,Convenience
4369 Variables}.
4370 @end table
4371
4372 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4373
4374
4375 @node Break Commands
4376 @subsection Breakpoint Command Lists
4377
4378 @cindex breakpoint commands
4379 You can give any breakpoint (or watchpoint or catchpoint) a series of
4380 commands to execute when your program stops due to that breakpoint. For
4381 example, you might want to print the values of certain expressions, or
4382 enable other breakpoints.
4383
4384 @table @code
4385 @kindex commands
4386 @kindex end@r{ (breakpoint commands)}
4387 @item commands @r{[}@var{range}@dots{}@r{]}
4388 @itemx @dots{} @var{command-list} @dots{}
4389 @itemx end
4390 Specify a list of commands for the given breakpoints. The commands
4391 themselves appear on the following lines. Type a line containing just
4392 @code{end} to terminate the commands.
4393
4394 To remove all commands from a breakpoint, type @code{commands} and
4395 follow it immediately with @code{end}; that is, give no commands.
4396
4397 With no argument, @code{commands} refers to the last breakpoint,
4398 watchpoint, or catchpoint set (not to the breakpoint most recently
4399 encountered). If the most recent breakpoints were set with a single
4400 command, then the @code{commands} will apply to all the breakpoints
4401 set by that command. This applies to breakpoints set by
4402 @code{rbreak}, and also applies when a single @code{break} command
4403 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4404 Expressions}).
4405 @end table
4406
4407 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4408 disabled within a @var{command-list}.
4409
4410 You can use breakpoint commands to start your program up again. Simply
4411 use the @code{continue} command, or @code{step}, or any other command
4412 that resumes execution.
4413
4414 Any other commands in the command list, after a command that resumes
4415 execution, are ignored. This is because any time you resume execution
4416 (even with a simple @code{next} or @code{step}), you may encounter
4417 another breakpoint---which could have its own command list, leading to
4418 ambiguities about which list to execute.
4419
4420 @kindex silent
4421 If the first command you specify in a command list is @code{silent}, the
4422 usual message about stopping at a breakpoint is not printed. This may
4423 be desirable for breakpoints that are to print a specific message and
4424 then continue. If none of the remaining commands print anything, you
4425 see no sign that the breakpoint was reached. @code{silent} is
4426 meaningful only at the beginning of a breakpoint command list.
4427
4428 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4429 print precisely controlled output, and are often useful in silent
4430 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4431
4432 For example, here is how you could use breakpoint commands to print the
4433 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4434
4435 @smallexample
4436 break foo if x>0
4437 commands
4438 silent
4439 printf "x is %d\n",x
4440 cont
4441 end
4442 @end smallexample
4443
4444 One application for breakpoint commands is to compensate for one bug so
4445 you can test for another. Put a breakpoint just after the erroneous line
4446 of code, give it a condition to detect the case in which something
4447 erroneous has been done, and give it commands to assign correct values
4448 to any variables that need them. End with the @code{continue} command
4449 so that your program does not stop, and start with the @code{silent}
4450 command so that no output is produced. Here is an example:
4451
4452 @smallexample
4453 break 403
4454 commands
4455 silent
4456 set x = y + 4
4457 cont
4458 end
4459 @end smallexample
4460
4461 @node Save Breakpoints
4462 @subsection How to save breakpoints to a file
4463
4464 To save breakpoint definitions to a file use the @w{@code{save
4465 breakpoints}} command.
4466
4467 @table @code
4468 @kindex save breakpoints
4469 @cindex save breakpoints to a file for future sessions
4470 @item save breakpoints [@var{filename}]
4471 This command saves all current breakpoint definitions together with
4472 their commands and ignore counts, into a file @file{@var{filename}}
4473 suitable for use in a later debugging session. This includes all
4474 types of breakpoints (breakpoints, watchpoints, catchpoints,
4475 tracepoints). To read the saved breakpoint definitions, use the
4476 @code{source} command (@pxref{Command Files}). Note that watchpoints
4477 with expressions involving local variables may fail to be recreated
4478 because it may not be possible to access the context where the
4479 watchpoint is valid anymore. Because the saved breakpoint definitions
4480 are simply a sequence of @value{GDBN} commands that recreate the
4481 breakpoints, you can edit the file in your favorite editing program,
4482 and remove the breakpoint definitions you're not interested in, or
4483 that can no longer be recreated.
4484 @end table
4485
4486 @c @ifclear BARETARGET
4487 @node Error in Breakpoints
4488 @subsection ``Cannot insert breakpoints''
4489
4490 If you request too many active hardware-assisted breakpoints and
4491 watchpoints, you will see this error message:
4492
4493 @c FIXME: the precise wording of this message may change; the relevant
4494 @c source change is not committed yet (Sep 3, 1999).
4495 @smallexample
4496 Stopped; cannot insert breakpoints.
4497 You may have requested too many hardware breakpoints and watchpoints.
4498 @end smallexample
4499
4500 @noindent
4501 This message is printed when you attempt to resume the program, since
4502 only then @value{GDBN} knows exactly how many hardware breakpoints and
4503 watchpoints it needs to insert.
4504
4505 When this message is printed, you need to disable or remove some of the
4506 hardware-assisted breakpoints and watchpoints, and then continue.
4507
4508 @node Breakpoint-related Warnings
4509 @subsection ``Breakpoint address adjusted...''
4510 @cindex breakpoint address adjusted
4511
4512 Some processor architectures place constraints on the addresses at
4513 which breakpoints may be placed. For architectures thus constrained,
4514 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4515 with the constraints dictated by the architecture.
4516
4517 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4518 a VLIW architecture in which a number of RISC-like instructions may be
4519 bundled together for parallel execution. The FR-V architecture
4520 constrains the location of a breakpoint instruction within such a
4521 bundle to the instruction with the lowest address. @value{GDBN}
4522 honors this constraint by adjusting a breakpoint's address to the
4523 first in the bundle.
4524
4525 It is not uncommon for optimized code to have bundles which contain
4526 instructions from different source statements, thus it may happen that
4527 a breakpoint's address will be adjusted from one source statement to
4528 another. Since this adjustment may significantly alter @value{GDBN}'s
4529 breakpoint related behavior from what the user expects, a warning is
4530 printed when the breakpoint is first set and also when the breakpoint
4531 is hit.
4532
4533 A warning like the one below is printed when setting a breakpoint
4534 that's been subject to address adjustment:
4535
4536 @smallexample
4537 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4538 @end smallexample
4539
4540 Such warnings are printed both for user settable and @value{GDBN}'s
4541 internal breakpoints. If you see one of these warnings, you should
4542 verify that a breakpoint set at the adjusted address will have the
4543 desired affect. If not, the breakpoint in question may be removed and
4544 other breakpoints may be set which will have the desired behavior.
4545 E.g., it may be sufficient to place the breakpoint at a later
4546 instruction. A conditional breakpoint may also be useful in some
4547 cases to prevent the breakpoint from triggering too often.
4548
4549 @value{GDBN} will also issue a warning when stopping at one of these
4550 adjusted breakpoints:
4551
4552 @smallexample
4553 warning: Breakpoint 1 address previously adjusted from 0x00010414
4554 to 0x00010410.
4555 @end smallexample
4556
4557 When this warning is encountered, it may be too late to take remedial
4558 action except in cases where the breakpoint is hit earlier or more
4559 frequently than expected.
4560
4561 @node Continuing and Stepping
4562 @section Continuing and Stepping
4563
4564 @cindex stepping
4565 @cindex continuing
4566 @cindex resuming execution
4567 @dfn{Continuing} means resuming program execution until your program
4568 completes normally. In contrast, @dfn{stepping} means executing just
4569 one more ``step'' of your program, where ``step'' may mean either one
4570 line of source code, or one machine instruction (depending on what
4571 particular command you use). Either when continuing or when stepping,
4572 your program may stop even sooner, due to a breakpoint or a signal. (If
4573 it stops due to a signal, you may want to use @code{handle}, or use
4574 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4575
4576 @table @code
4577 @kindex continue
4578 @kindex c @r{(@code{continue})}
4579 @kindex fg @r{(resume foreground execution)}
4580 @item continue @r{[}@var{ignore-count}@r{]}
4581 @itemx c @r{[}@var{ignore-count}@r{]}
4582 @itemx fg @r{[}@var{ignore-count}@r{]}
4583 Resume program execution, at the address where your program last stopped;
4584 any breakpoints set at that address are bypassed. The optional argument
4585 @var{ignore-count} allows you to specify a further number of times to
4586 ignore a breakpoint at this location; its effect is like that of
4587 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4588
4589 The argument @var{ignore-count} is meaningful only when your program
4590 stopped due to a breakpoint. At other times, the argument to
4591 @code{continue} is ignored.
4592
4593 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4594 debugged program is deemed to be the foreground program) are provided
4595 purely for convenience, and have exactly the same behavior as
4596 @code{continue}.
4597 @end table
4598
4599 To resume execution at a different place, you can use @code{return}
4600 (@pxref{Returning, ,Returning from a Function}) to go back to the
4601 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4602 Different Address}) to go to an arbitrary location in your program.
4603
4604 A typical technique for using stepping is to set a breakpoint
4605 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4606 beginning of the function or the section of your program where a problem
4607 is believed to lie, run your program until it stops at that breakpoint,
4608 and then step through the suspect area, examining the variables that are
4609 interesting, until you see the problem happen.
4610
4611 @table @code
4612 @kindex step
4613 @kindex s @r{(@code{step})}
4614 @item step
4615 Continue running your program until control reaches a different source
4616 line, then stop it and return control to @value{GDBN}. This command is
4617 abbreviated @code{s}.
4618
4619 @quotation
4620 @c "without debugging information" is imprecise; actually "without line
4621 @c numbers in the debugging information". (gcc -g1 has debugging info but
4622 @c not line numbers). But it seems complex to try to make that
4623 @c distinction here.
4624 @emph{Warning:} If you use the @code{step} command while control is
4625 within a function that was compiled without debugging information,
4626 execution proceeds until control reaches a function that does have
4627 debugging information. Likewise, it will not step into a function which
4628 is compiled without debugging information. To step through functions
4629 without debugging information, use the @code{stepi} command, described
4630 below.
4631 @end quotation
4632
4633 The @code{step} command only stops at the first instruction of a source
4634 line. This prevents the multiple stops that could otherwise occur in
4635 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4636 to stop if a function that has debugging information is called within
4637 the line. In other words, @code{step} @emph{steps inside} any functions
4638 called within the line.
4639
4640 Also, the @code{step} command only enters a function if there is line
4641 number information for the function. Otherwise it acts like the
4642 @code{next} command. This avoids problems when using @code{cc -gl}
4643 on MIPS machines. Previously, @code{step} entered subroutines if there
4644 was any debugging information about the routine.
4645
4646 @item step @var{count}
4647 Continue running as in @code{step}, but do so @var{count} times. If a
4648 breakpoint is reached, or a signal not related to stepping occurs before
4649 @var{count} steps, stepping stops right away.
4650
4651 @kindex next
4652 @kindex n @r{(@code{next})}
4653 @item next @r{[}@var{count}@r{]}
4654 Continue to the next source line in the current (innermost) stack frame.
4655 This is similar to @code{step}, but function calls that appear within
4656 the line of code are executed without stopping. Execution stops when
4657 control reaches a different line of code at the original stack level
4658 that was executing when you gave the @code{next} command. This command
4659 is abbreviated @code{n}.
4660
4661 An argument @var{count} is a repeat count, as for @code{step}.
4662
4663
4664 @c FIX ME!! Do we delete this, or is there a way it fits in with
4665 @c the following paragraph? --- Vctoria
4666 @c
4667 @c @code{next} within a function that lacks debugging information acts like
4668 @c @code{step}, but any function calls appearing within the code of the
4669 @c function are executed without stopping.
4670
4671 The @code{next} command only stops at the first instruction of a
4672 source line. This prevents multiple stops that could otherwise occur in
4673 @code{switch} statements, @code{for} loops, etc.
4674
4675 @kindex set step-mode
4676 @item set step-mode
4677 @cindex functions without line info, and stepping
4678 @cindex stepping into functions with no line info
4679 @itemx set step-mode on
4680 The @code{set step-mode on} command causes the @code{step} command to
4681 stop at the first instruction of a function which contains no debug line
4682 information rather than stepping over it.
4683
4684 This is useful in cases where you may be interested in inspecting the
4685 machine instructions of a function which has no symbolic info and do not
4686 want @value{GDBN} to automatically skip over this function.
4687
4688 @item set step-mode off
4689 Causes the @code{step} command to step over any functions which contains no
4690 debug information. This is the default.
4691
4692 @item show step-mode
4693 Show whether @value{GDBN} will stop in or step over functions without
4694 source line debug information.
4695
4696 @kindex finish
4697 @kindex fin @r{(@code{finish})}
4698 @item finish
4699 Continue running until just after function in the selected stack frame
4700 returns. Print the returned value (if any). This command can be
4701 abbreviated as @code{fin}.
4702
4703 Contrast this with the @code{return} command (@pxref{Returning,
4704 ,Returning from a Function}).
4705
4706 @kindex until
4707 @kindex u @r{(@code{until})}
4708 @cindex run until specified location
4709 @item until
4710 @itemx u
4711 Continue running until a source line past the current line, in the
4712 current stack frame, is reached. This command is used to avoid single
4713 stepping through a loop more than once. It is like the @code{next}
4714 command, except that when @code{until} encounters a jump, it
4715 automatically continues execution until the program counter is greater
4716 than the address of the jump.
4717
4718 This means that when you reach the end of a loop after single stepping
4719 though it, @code{until} makes your program continue execution until it
4720 exits the loop. In contrast, a @code{next} command at the end of a loop
4721 simply steps back to the beginning of the loop, which forces you to step
4722 through the next iteration.
4723
4724 @code{until} always stops your program if it attempts to exit the current
4725 stack frame.
4726
4727 @code{until} may produce somewhat counterintuitive results if the order
4728 of machine code does not match the order of the source lines. For
4729 example, in the following excerpt from a debugging session, the @code{f}
4730 (@code{frame}) command shows that execution is stopped at line
4731 @code{206}; yet when we use @code{until}, we get to line @code{195}:
4732
4733 @smallexample
4734 (@value{GDBP}) f
4735 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
4736 206 expand_input();
4737 (@value{GDBP}) until
4738 195 for ( ; argc > 0; NEXTARG) @{
4739 @end smallexample
4740
4741 This happened because, for execution efficiency, the compiler had
4742 generated code for the loop closure test at the end, rather than the
4743 start, of the loop---even though the test in a C @code{for}-loop is
4744 written before the body of the loop. The @code{until} command appeared
4745 to step back to the beginning of the loop when it advanced to this
4746 expression; however, it has not really gone to an earlier
4747 statement---not in terms of the actual machine code.
4748
4749 @code{until} with no argument works by means of single
4750 instruction stepping, and hence is slower than @code{until} with an
4751 argument.
4752
4753 @item until @var{location}
4754 @itemx u @var{location}
4755 Continue running your program until either the specified location is
4756 reached, or the current stack frame returns. @var{location} is any of
4757 the forms described in @ref{Specify Location}.
4758 This form of the command uses temporary breakpoints, and
4759 hence is quicker than @code{until} without an argument. The specified
4760 location is actually reached only if it is in the current frame. This
4761 implies that @code{until} can be used to skip over recursive function
4762 invocations. For instance in the code below, if the current location is
4763 line @code{96}, issuing @code{until 99} will execute the program up to
4764 line @code{99} in the same invocation of factorial, i.e., after the inner
4765 invocations have returned.
4766
4767 @smallexample
4768 94 int factorial (int value)
4769 95 @{
4770 96 if (value > 1) @{
4771 97 value *= factorial (value - 1);
4772 98 @}
4773 99 return (value);
4774 100 @}
4775 @end smallexample
4776
4777
4778 @kindex advance @var{location}
4779 @itemx advance @var{location}
4780 Continue running the program up to the given @var{location}. An argument is
4781 required, which should be of one of the forms described in
4782 @ref{Specify Location}.
4783 Execution will also stop upon exit from the current stack
4784 frame. This command is similar to @code{until}, but @code{advance} will
4785 not skip over recursive function calls, and the target location doesn't
4786 have to be in the same frame as the current one.
4787
4788
4789 @kindex stepi
4790 @kindex si @r{(@code{stepi})}
4791 @item stepi
4792 @itemx stepi @var{arg}
4793 @itemx si
4794 Execute one machine instruction, then stop and return to the debugger.
4795
4796 It is often useful to do @samp{display/i $pc} when stepping by machine
4797 instructions. This makes @value{GDBN} automatically display the next
4798 instruction to be executed, each time your program stops. @xref{Auto
4799 Display,, Automatic Display}.
4800
4801 An argument is a repeat count, as in @code{step}.
4802
4803 @need 750
4804 @kindex nexti
4805 @kindex ni @r{(@code{nexti})}
4806 @item nexti
4807 @itemx nexti @var{arg}
4808 @itemx ni
4809 Execute one machine instruction, but if it is a function call,
4810 proceed until the function returns.
4811
4812 An argument is a repeat count, as in @code{next}.
4813 @end table
4814
4815 @node Signals
4816 @section Signals
4817 @cindex signals
4818
4819 A signal is an asynchronous event that can happen in a program. The
4820 operating system defines the possible kinds of signals, and gives each
4821 kind a name and a number. For example, in Unix @code{SIGINT} is the
4822 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
4823 @code{SIGSEGV} is the signal a program gets from referencing a place in
4824 memory far away from all the areas in use; @code{SIGALRM} occurs when
4825 the alarm clock timer goes off (which happens only if your program has
4826 requested an alarm).
4827
4828 @cindex fatal signals
4829 Some signals, including @code{SIGALRM}, are a normal part of the
4830 functioning of your program. Others, such as @code{SIGSEGV}, indicate
4831 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
4832 program has not specified in advance some other way to handle the signal.
4833 @code{SIGINT} does not indicate an error in your program, but it is normally
4834 fatal so it can carry out the purpose of the interrupt: to kill the program.
4835
4836 @value{GDBN} has the ability to detect any occurrence of a signal in your
4837 program. You can tell @value{GDBN} in advance what to do for each kind of
4838 signal.
4839
4840 @cindex handling signals
4841 Normally, @value{GDBN} is set up to let the non-erroneous signals like
4842 @code{SIGALRM} be silently passed to your program
4843 (so as not to interfere with their role in the program's functioning)
4844 but to stop your program immediately whenever an error signal happens.
4845 You can change these settings with the @code{handle} command.
4846
4847 @table @code
4848 @kindex info signals
4849 @kindex info handle
4850 @item info signals
4851 @itemx info handle
4852 Print a table of all the kinds of signals and how @value{GDBN} has been told to
4853 handle each one. You can use this to see the signal numbers of all
4854 the defined types of signals.
4855
4856 @item info signals @var{sig}
4857 Similar, but print information only about the specified signal number.
4858
4859 @code{info handle} is an alias for @code{info signals}.
4860
4861 @kindex handle
4862 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
4863 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
4864 can be the number of a signal or its name (with or without the
4865 @samp{SIG} at the beginning); a list of signal numbers of the form
4866 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
4867 known signals. Optional arguments @var{keywords}, described below,
4868 say what change to make.
4869 @end table
4870
4871 @c @group
4872 The keywords allowed by the @code{handle} command can be abbreviated.
4873 Their full names are:
4874
4875 @table @code
4876 @item nostop
4877 @value{GDBN} should not stop your program when this signal happens. It may
4878 still print a message telling you that the signal has come in.
4879
4880 @item stop
4881 @value{GDBN} should stop your program when this signal happens. This implies
4882 the @code{print} keyword as well.
4883
4884 @item print
4885 @value{GDBN} should print a message when this signal happens.
4886
4887 @item noprint
4888 @value{GDBN} should not mention the occurrence of the signal at all. This
4889 implies the @code{nostop} keyword as well.
4890
4891 @item pass
4892 @itemx noignore
4893 @value{GDBN} should allow your program to see this signal; your program
4894 can handle the signal, or else it may terminate if the signal is fatal
4895 and not handled. @code{pass} and @code{noignore} are synonyms.
4896
4897 @item nopass
4898 @itemx ignore
4899 @value{GDBN} should not allow your program to see this signal.
4900 @code{nopass} and @code{ignore} are synonyms.
4901 @end table
4902 @c @end group
4903
4904 When a signal stops your program, the signal is not visible to the
4905 program until you
4906 continue. Your program sees the signal then, if @code{pass} is in
4907 effect for the signal in question @emph{at that time}. In other words,
4908 after @value{GDBN} reports a signal, you can use the @code{handle}
4909 command with @code{pass} or @code{nopass} to control whether your
4910 program sees that signal when you continue.
4911
4912 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
4913 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
4914 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
4915 erroneous signals.
4916
4917 You can also use the @code{signal} command to prevent your program from
4918 seeing a signal, or cause it to see a signal it normally would not see,
4919 or to give it any signal at any time. For example, if your program stopped
4920 due to some sort of memory reference error, you might store correct
4921 values into the erroneous variables and continue, hoping to see more
4922 execution; but your program would probably terminate immediately as
4923 a result of the fatal signal once it saw the signal. To prevent this,
4924 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
4925 Program a Signal}.
4926
4927 @cindex extra signal information
4928 @anchor{extra signal information}
4929
4930 On some targets, @value{GDBN} can inspect extra signal information
4931 associated with the intercepted signal, before it is actually
4932 delivered to the program being debugged. This information is exported
4933 by the convenience variable @code{$_siginfo}, and consists of data
4934 that is passed by the kernel to the signal handler at the time of the
4935 receipt of a signal. The data type of the information itself is
4936 target dependent. You can see the data type using the @code{ptype
4937 $_siginfo} command. On Unix systems, it typically corresponds to the
4938 standard @code{siginfo_t} type, as defined in the @file{signal.h}
4939 system header.
4940
4941 Here's an example, on a @sc{gnu}/Linux system, printing the stray
4942 referenced address that raised a segmentation fault.
4943
4944 @smallexample
4945 @group
4946 (@value{GDBP}) continue
4947 Program received signal SIGSEGV, Segmentation fault.
4948 0x0000000000400766 in main ()
4949 69 *(int *)p = 0;
4950 (@value{GDBP}) ptype $_siginfo
4951 type = struct @{
4952 int si_signo;
4953 int si_errno;
4954 int si_code;
4955 union @{
4956 int _pad[28];
4957 struct @{...@} _kill;
4958 struct @{...@} _timer;
4959 struct @{...@} _rt;
4960 struct @{...@} _sigchld;
4961 struct @{...@} _sigfault;
4962 struct @{...@} _sigpoll;
4963 @} _sifields;
4964 @}
4965 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
4966 type = struct @{
4967 void *si_addr;
4968 @}
4969 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
4970 $1 = (void *) 0x7ffff7ff7000
4971 @end group
4972 @end smallexample
4973
4974 Depending on target support, @code{$_siginfo} may also be writable.
4975
4976 @node Thread Stops
4977 @section Stopping and Starting Multi-thread Programs
4978
4979 @cindex stopped threads
4980 @cindex threads, stopped
4981
4982 @cindex continuing threads
4983 @cindex threads, continuing
4984
4985 @value{GDBN} supports debugging programs with multiple threads
4986 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
4987 are two modes of controlling execution of your program within the
4988 debugger. In the default mode, referred to as @dfn{all-stop mode},
4989 when any thread in your program stops (for example, at a breakpoint
4990 or while being stepped), all other threads in the program are also stopped by
4991 @value{GDBN}. On some targets, @value{GDBN} also supports
4992 @dfn{non-stop mode}, in which other threads can continue to run freely while
4993 you examine the stopped thread in the debugger.
4994
4995 @menu
4996 * All-Stop Mode:: All threads stop when GDB takes control
4997 * Non-Stop Mode:: Other threads continue to execute
4998 * Background Execution:: Running your program asynchronously
4999 * Thread-Specific Breakpoints:: Controlling breakpoints
5000 * Interrupted System Calls:: GDB may interfere with system calls
5001 * Observer Mode:: GDB does not alter program behavior
5002 @end menu
5003
5004 @node All-Stop Mode
5005 @subsection All-Stop Mode
5006
5007 @cindex all-stop mode
5008
5009 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
5010 @emph{all} threads of execution stop, not just the current thread. This
5011 allows you to examine the overall state of the program, including
5012 switching between threads, without worrying that things may change
5013 underfoot.
5014
5015 Conversely, whenever you restart the program, @emph{all} threads start
5016 executing. @emph{This is true even when single-stepping} with commands
5017 like @code{step} or @code{next}.
5018
5019 In particular, @value{GDBN} cannot single-step all threads in lockstep.
5020 Since thread scheduling is up to your debugging target's operating
5021 system (not controlled by @value{GDBN}), other threads may
5022 execute more than one statement while the current thread completes a
5023 single step. Moreover, in general other threads stop in the middle of a
5024 statement, rather than at a clean statement boundary, when the program
5025 stops.
5026
5027 You might even find your program stopped in another thread after
5028 continuing or even single-stepping. This happens whenever some other
5029 thread runs into a breakpoint, a signal, or an exception before the
5030 first thread completes whatever you requested.
5031
5032 @cindex automatic thread selection
5033 @cindex switching threads automatically
5034 @cindex threads, automatic switching
5035 Whenever @value{GDBN} stops your program, due to a breakpoint or a
5036 signal, it automatically selects the thread where that breakpoint or
5037 signal happened. @value{GDBN} alerts you to the context switch with a
5038 message such as @samp{[Switching to Thread @var{n}]} to identify the
5039 thread.
5040
5041 On some OSes, you can modify @value{GDBN}'s default behavior by
5042 locking the OS scheduler to allow only a single thread to run.
5043
5044 @table @code
5045 @item set scheduler-locking @var{mode}
5046 @cindex scheduler locking mode
5047 @cindex lock scheduler
5048 Set the scheduler locking mode. If it is @code{off}, then there is no
5049 locking and any thread may run at any time. If @code{on}, then only the
5050 current thread may run when the inferior is resumed. The @code{step}
5051 mode optimizes for single-stepping; it prevents other threads
5052 from preempting the current thread while you are stepping, so that
5053 the focus of debugging does not change unexpectedly.
5054 Other threads only rarely (or never) get a chance to run
5055 when you step. They are more likely to run when you @samp{next} over a
5056 function call, and they are completely free to run when you use commands
5057 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
5058 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
5059 the current thread away from the thread that you are debugging.
5060
5061 @item show scheduler-locking
5062 Display the current scheduler locking mode.
5063 @end table
5064
5065 @cindex resume threads of multiple processes simultaneously
5066 By default, when you issue one of the execution commands such as
5067 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
5068 threads of the current inferior to run. For example, if @value{GDBN}
5069 is attached to two inferiors, each with two threads, the
5070 @code{continue} command resumes only the two threads of the current
5071 inferior. This is useful, for example, when you debug a program that
5072 forks and you want to hold the parent stopped (so that, for instance,
5073 it doesn't run to exit), while you debug the child. In other
5074 situations, you may not be interested in inspecting the current state
5075 of any of the processes @value{GDBN} is attached to, and you may want
5076 to resume them all until some breakpoint is hit. In the latter case,
5077 you can instruct @value{GDBN} to allow all threads of all the
5078 inferiors to run with the @w{@code{set schedule-multiple}} command.
5079
5080 @table @code
5081 @kindex set schedule-multiple
5082 @item set schedule-multiple
5083 Set the mode for allowing threads of multiple processes to be resumed
5084 when an execution command is issued. When @code{on}, all threads of
5085 all processes are allowed to run. When @code{off}, only the threads
5086 of the current process are resumed. The default is @code{off}. The
5087 @code{scheduler-locking} mode takes precedence when set to @code{on},
5088 or while you are stepping and set to @code{step}.
5089
5090 @item show schedule-multiple
5091 Display the current mode for resuming the execution of threads of
5092 multiple processes.
5093 @end table
5094
5095 @node Non-Stop Mode
5096 @subsection Non-Stop Mode
5097
5098 @cindex non-stop mode
5099
5100 @c This section is really only a place-holder, and needs to be expanded
5101 @c with more details.
5102
5103 For some multi-threaded targets, @value{GDBN} supports an optional
5104 mode of operation in which you can examine stopped program threads in
5105 the debugger while other threads continue to execute freely. This
5106 minimizes intrusion when debugging live systems, such as programs
5107 where some threads have real-time constraints or must continue to
5108 respond to external events. This is referred to as @dfn{non-stop} mode.
5109
5110 In non-stop mode, when a thread stops to report a debugging event,
5111 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5112 threads as well, in contrast to the all-stop mode behavior. Additionally,
5113 execution commands such as @code{continue} and @code{step} apply by default
5114 only to the current thread in non-stop mode, rather than all threads as
5115 in all-stop mode. This allows you to control threads explicitly in
5116 ways that are not possible in all-stop mode --- for example, stepping
5117 one thread while allowing others to run freely, stepping
5118 one thread while holding all others stopped, or stepping several threads
5119 independently and simultaneously.
5120
5121 To enter non-stop mode, use this sequence of commands before you run
5122 or attach to your program:
5123
5124 @smallexample
5125 # Enable the async interface.
5126 set target-async 1
5127
5128 # If using the CLI, pagination breaks non-stop.
5129 set pagination off
5130
5131 # Finally, turn it on!
5132 set non-stop on
5133 @end smallexample
5134
5135 You can use these commands to manipulate the non-stop mode setting:
5136
5137 @table @code
5138 @kindex set non-stop
5139 @item set non-stop on
5140 Enable selection of non-stop mode.
5141 @item set non-stop off
5142 Disable selection of non-stop mode.
5143 @kindex show non-stop
5144 @item show non-stop
5145 Show the current non-stop enablement setting.
5146 @end table
5147
5148 Note these commands only reflect whether non-stop mode is enabled,
5149 not whether the currently-executing program is being run in non-stop mode.
5150 In particular, the @code{set non-stop} preference is only consulted when
5151 @value{GDBN} starts or connects to the target program, and it is generally
5152 not possible to switch modes once debugging has started. Furthermore,
5153 since not all targets support non-stop mode, even when you have enabled
5154 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
5155 default.
5156
5157 In non-stop mode, all execution commands apply only to the current thread
5158 by default. That is, @code{continue} only continues one thread.
5159 To continue all threads, issue @code{continue -a} or @code{c -a}.
5160
5161 You can use @value{GDBN}'s background execution commands
5162 (@pxref{Background Execution}) to run some threads in the background
5163 while you continue to examine or step others from @value{GDBN}.
5164 The MI execution commands (@pxref{GDB/MI Program Execution}) are
5165 always executed asynchronously in non-stop mode.
5166
5167 Suspending execution is done with the @code{interrupt} command when
5168 running in the background, or @kbd{Ctrl-c} during foreground execution.
5169 In all-stop mode, this stops the whole process;
5170 but in non-stop mode the interrupt applies only to the current thread.
5171 To stop the whole program, use @code{interrupt -a}.
5172
5173 Other execution commands do not currently support the @code{-a} option.
5174
5175 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
5176 that thread current, as it does in all-stop mode. This is because the
5177 thread stop notifications are asynchronous with respect to @value{GDBN}'s
5178 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
5179 changed to a different thread just as you entered a command to operate on the
5180 previously current thread.
5181
5182 @node Background Execution
5183 @subsection Background Execution
5184
5185 @cindex foreground execution
5186 @cindex background execution
5187 @cindex asynchronous execution
5188 @cindex execution, foreground, background and asynchronous
5189
5190 @value{GDBN}'s execution commands have two variants: the normal
5191 foreground (synchronous) behavior, and a background
5192 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
5193 the program to report that some thread has stopped before prompting for
5194 another command. In background execution, @value{GDBN} immediately gives
5195 a command prompt so that you can issue other commands while your program runs.
5196
5197 You need to explicitly enable asynchronous mode before you can use
5198 background execution commands. You can use these commands to
5199 manipulate the asynchronous mode setting:
5200
5201 @table @code
5202 @kindex set target-async
5203 @item set target-async on
5204 Enable asynchronous mode.
5205 @item set target-async off
5206 Disable asynchronous mode.
5207 @kindex show target-async
5208 @item show target-async
5209 Show the current target-async setting.
5210 @end table
5211
5212 If the target doesn't support async mode, @value{GDBN} issues an error
5213 message if you attempt to use the background execution commands.
5214
5215 To specify background execution, add a @code{&} to the command. For example,
5216 the background form of the @code{continue} command is @code{continue&}, or
5217 just @code{c&}. The execution commands that accept background execution
5218 are:
5219
5220 @table @code
5221 @kindex run&
5222 @item run
5223 @xref{Starting, , Starting your Program}.
5224
5225 @item attach
5226 @kindex attach&
5227 @xref{Attach, , Debugging an Already-running Process}.
5228
5229 @item step
5230 @kindex step&
5231 @xref{Continuing and Stepping, step}.
5232
5233 @item stepi
5234 @kindex stepi&
5235 @xref{Continuing and Stepping, stepi}.
5236
5237 @item next
5238 @kindex next&
5239 @xref{Continuing and Stepping, next}.
5240
5241 @item nexti
5242 @kindex nexti&
5243 @xref{Continuing and Stepping, nexti}.
5244
5245 @item continue
5246 @kindex continue&
5247 @xref{Continuing and Stepping, continue}.
5248
5249 @item finish
5250 @kindex finish&
5251 @xref{Continuing and Stepping, finish}.
5252
5253 @item until
5254 @kindex until&
5255 @xref{Continuing and Stepping, until}.
5256
5257 @end table
5258
5259 Background execution is especially useful in conjunction with non-stop
5260 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
5261 However, you can also use these commands in the normal all-stop mode with
5262 the restriction that you cannot issue another execution command until the
5263 previous one finishes. Examples of commands that are valid in all-stop
5264 mode while the program is running include @code{help} and @code{info break}.
5265
5266 You can interrupt your program while it is running in the background by
5267 using the @code{interrupt} command.
5268
5269 @table @code
5270 @kindex interrupt
5271 @item interrupt
5272 @itemx interrupt -a
5273
5274 Suspend execution of the running program. In all-stop mode,
5275 @code{interrupt} stops the whole process, but in non-stop mode, it stops
5276 only the current thread. To stop the whole program in non-stop mode,
5277 use @code{interrupt -a}.
5278 @end table
5279
5280 @node Thread-Specific Breakpoints
5281 @subsection Thread-Specific Breakpoints
5282
5283 When your program has multiple threads (@pxref{Threads,, Debugging
5284 Programs with Multiple Threads}), you can choose whether to set
5285 breakpoints on all threads, or on a particular thread.
5286
5287 @table @code
5288 @cindex breakpoints and threads
5289 @cindex thread breakpoints
5290 @kindex break @dots{} thread @var{threadno}
5291 @item break @var{linespec} thread @var{threadno}
5292 @itemx break @var{linespec} thread @var{threadno} if @dots{}
5293 @var{linespec} specifies source lines; there are several ways of
5294 writing them (@pxref{Specify Location}), but the effect is always to
5295 specify some source line.
5296
5297 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
5298 to specify that you only want @value{GDBN} to stop the program when a
5299 particular thread reaches this breakpoint. @var{threadno} is one of the
5300 numeric thread identifiers assigned by @value{GDBN}, shown in the first
5301 column of the @samp{info threads} display.
5302
5303 If you do not specify @samp{thread @var{threadno}} when you set a
5304 breakpoint, the breakpoint applies to @emph{all} threads of your
5305 program.
5306
5307 You can use the @code{thread} qualifier on conditional breakpoints as
5308 well; in this case, place @samp{thread @var{threadno}} before or
5309 after the breakpoint condition, like this:
5310
5311 @smallexample
5312 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
5313 @end smallexample
5314
5315 @end table
5316
5317 @node Interrupted System Calls
5318 @subsection Interrupted System Calls
5319
5320 @cindex thread breakpoints and system calls
5321 @cindex system calls and thread breakpoints
5322 @cindex premature return from system calls
5323 There is an unfortunate side effect when using @value{GDBN} to debug
5324 multi-threaded programs. If one thread stops for a
5325 breakpoint, or for some other reason, and another thread is blocked in a
5326 system call, then the system call may return prematurely. This is a
5327 consequence of the interaction between multiple threads and the signals
5328 that @value{GDBN} uses to implement breakpoints and other events that
5329 stop execution.
5330
5331 To handle this problem, your program should check the return value of
5332 each system call and react appropriately. This is good programming
5333 style anyways.
5334
5335 For example, do not write code like this:
5336
5337 @smallexample
5338 sleep (10);
5339 @end smallexample
5340
5341 The call to @code{sleep} will return early if a different thread stops
5342 at a breakpoint or for some other reason.
5343
5344 Instead, write this:
5345
5346 @smallexample
5347 int unslept = 10;
5348 while (unslept > 0)
5349 unslept = sleep (unslept);
5350 @end smallexample
5351
5352 A system call is allowed to return early, so the system is still
5353 conforming to its specification. But @value{GDBN} does cause your
5354 multi-threaded program to behave differently than it would without
5355 @value{GDBN}.
5356
5357 Also, @value{GDBN} uses internal breakpoints in the thread library to
5358 monitor certain events such as thread creation and thread destruction.
5359 When such an event happens, a system call in another thread may return
5360 prematurely, even though your program does not appear to stop.
5361
5362 @node Observer Mode
5363 @subsection Observer Mode
5364
5365 If you want to build on non-stop mode and observe program behavior
5366 without any chance of disruption by @value{GDBN}, you can set
5367 variables to disable all of the debugger's attempts to modify state,
5368 whether by writing memory, inserting breakpoints, etc. These operate
5369 at a low level, intercepting operations from all commands.
5370
5371 When all of these are set to @code{off}, then @value{GDBN} is said to
5372 be @dfn{observer mode}. As a convenience, the variable
5373 @code{observer} can be set to disable these, plus enable non-stop
5374 mode.
5375
5376 Note that @value{GDBN} will not prevent you from making nonsensical
5377 combinations of these settings. For instance, if you have enabled
5378 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
5379 then breakpoints that work by writing trap instructions into the code
5380 stream will still not be able to be placed.
5381
5382 @table @code
5383
5384 @kindex observer
5385 @item set observer on
5386 @itemx set observer off
5387 When set to @code{on}, this disables all the permission variables
5388 below (except for @code{insert-fast-tracepoints}), plus enables
5389 non-stop debugging. Setting this to @code{off} switches back to
5390 normal debugging, though remaining in non-stop mode.
5391
5392 @item show observer
5393 Show whether observer mode is on or off.
5394
5395 @kindex may-write-registers
5396 @item set may-write-registers on
5397 @itemx set may-write-registers off
5398 This controls whether @value{GDBN} will attempt to alter the values of
5399 registers, such as with assignment expressions in @code{print}, or the
5400 @code{jump} command. It defaults to @code{on}.
5401
5402 @item show may-write-registers
5403 Show the current permission to write registers.
5404
5405 @kindex may-write-memory
5406 @item set may-write-memory on
5407 @itemx set may-write-memory off
5408 This controls whether @value{GDBN} will attempt to alter the contents
5409 of memory, such as with assignment expressions in @code{print}. It
5410 defaults to @code{on}.
5411
5412 @item show may-write-memory
5413 Show the current permission to write memory.
5414
5415 @kindex may-insert-breakpoints
5416 @item set may-insert-breakpoints on
5417 @itemx set may-insert-breakpoints off
5418 This controls whether @value{GDBN} will attempt to insert breakpoints.
5419 This affects all breakpoints, including internal breakpoints defined
5420 by @value{GDBN}. It defaults to @code{on}.
5421
5422 @item show may-insert-breakpoints
5423 Show the current permission to insert breakpoints.
5424
5425 @kindex may-insert-tracepoints
5426 @item set may-insert-tracepoints on
5427 @itemx set may-insert-tracepoints off
5428 This controls whether @value{GDBN} will attempt to insert (regular)
5429 tracepoints at the beginning of a tracing experiment. It affects only
5430 non-fast tracepoints, fast tracepoints being under the control of
5431 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
5432
5433 @item show may-insert-tracepoints
5434 Show the current permission to insert tracepoints.
5435
5436 @kindex may-insert-fast-tracepoints
5437 @item set may-insert-fast-tracepoints on
5438 @itemx set may-insert-fast-tracepoints off
5439 This controls whether @value{GDBN} will attempt to insert fast
5440 tracepoints at the beginning of a tracing experiment. It affects only
5441 fast tracepoints, regular (non-fast) tracepoints being under the
5442 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
5443
5444 @item show may-insert-fast-tracepoints
5445 Show the current permission to insert fast tracepoints.
5446
5447 @kindex may-interrupt
5448 @item set may-interrupt on
5449 @itemx set may-interrupt off
5450 This controls whether @value{GDBN} will attempt to interrupt or stop
5451 program execution. When this variable is @code{off}, the
5452 @code{interrupt} command will have no effect, nor will
5453 @kbd{Ctrl-c}. It defaults to @code{on}.
5454
5455 @item show may-interrupt
5456 Show the current permission to interrupt or stop the program.
5457
5458 @end table
5459
5460 @node Reverse Execution
5461 @chapter Running programs backward
5462 @cindex reverse execution
5463 @cindex running programs backward
5464
5465 When you are debugging a program, it is not unusual to realize that
5466 you have gone too far, and some event of interest has already happened.
5467 If the target environment supports it, @value{GDBN} can allow you to
5468 ``rewind'' the program by running it backward.
5469
5470 A target environment that supports reverse execution should be able
5471 to ``undo'' the changes in machine state that have taken place as the
5472 program was executing normally. Variables, registers etc.@: should
5473 revert to their previous values. Obviously this requires a great
5474 deal of sophistication on the part of the target environment; not
5475 all target environments can support reverse execution.
5476
5477 When a program is executed in reverse, the instructions that
5478 have most recently been executed are ``un-executed'', in reverse
5479 order. The program counter runs backward, following the previous
5480 thread of execution in reverse. As each instruction is ``un-executed'',
5481 the values of memory and/or registers that were changed by that
5482 instruction are reverted to their previous states. After executing
5483 a piece of source code in reverse, all side effects of that code
5484 should be ``undone'', and all variables should be returned to their
5485 prior values@footnote{
5486 Note that some side effects are easier to undo than others. For instance,
5487 memory and registers are relatively easy, but device I/O is hard. Some
5488 targets may be able undo things like device I/O, and some may not.
5489
5490 The contract between @value{GDBN} and the reverse executing target
5491 requires only that the target do something reasonable when
5492 @value{GDBN} tells it to execute backwards, and then report the
5493 results back to @value{GDBN}. Whatever the target reports back to
5494 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
5495 assumes that the memory and registers that the target reports are in a
5496 consistant state, but @value{GDBN} accepts whatever it is given.
5497 }.
5498
5499 If you are debugging in a target environment that supports
5500 reverse execution, @value{GDBN} provides the following commands.
5501
5502 @table @code
5503 @kindex reverse-continue
5504 @kindex rc @r{(@code{reverse-continue})}
5505 @item reverse-continue @r{[}@var{ignore-count}@r{]}
5506 @itemx rc @r{[}@var{ignore-count}@r{]}
5507 Beginning at the point where your program last stopped, start executing
5508 in reverse. Reverse execution will stop for breakpoints and synchronous
5509 exceptions (signals), just like normal execution. Behavior of
5510 asynchronous signals depends on the target environment.
5511
5512 @kindex reverse-step
5513 @kindex rs @r{(@code{step})}
5514 @item reverse-step @r{[}@var{count}@r{]}
5515 Run the program backward until control reaches the start of a
5516 different source line; then stop it, and return control to @value{GDBN}.
5517
5518 Like the @code{step} command, @code{reverse-step} will only stop
5519 at the beginning of a source line. It ``un-executes'' the previously
5520 executed source line. If the previous source line included calls to
5521 debuggable functions, @code{reverse-step} will step (backward) into
5522 the called function, stopping at the beginning of the @emph{last}
5523 statement in the called function (typically a return statement).
5524
5525 Also, as with the @code{step} command, if non-debuggable functions are
5526 called, @code{reverse-step} will run thru them backward without stopping.
5527
5528 @kindex reverse-stepi
5529 @kindex rsi @r{(@code{reverse-stepi})}
5530 @item reverse-stepi @r{[}@var{count}@r{]}
5531 Reverse-execute one machine instruction. Note that the instruction
5532 to be reverse-executed is @emph{not} the one pointed to by the program
5533 counter, but the instruction executed prior to that one. For instance,
5534 if the last instruction was a jump, @code{reverse-stepi} will take you
5535 back from the destination of the jump to the jump instruction itself.
5536
5537 @kindex reverse-next
5538 @kindex rn @r{(@code{reverse-next})}
5539 @item reverse-next @r{[}@var{count}@r{]}
5540 Run backward to the beginning of the previous line executed in
5541 the current (innermost) stack frame. If the line contains function
5542 calls, they will be ``un-executed'' without stopping. Starting from
5543 the first line of a function, @code{reverse-next} will take you back
5544 to the caller of that function, @emph{before} the function was called,
5545 just as the normal @code{next} command would take you from the last
5546 line of a function back to its return to its caller
5547 @footnote{Unless the code is too heavily optimized.}.
5548
5549 @kindex reverse-nexti
5550 @kindex rni @r{(@code{reverse-nexti})}
5551 @item reverse-nexti @r{[}@var{count}@r{]}
5552 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
5553 in reverse, except that called functions are ``un-executed'' atomically.
5554 That is, if the previously executed instruction was a return from
5555 another function, @code{reverse-nexti} will continue to execute
5556 in reverse until the call to that function (from the current stack
5557 frame) is reached.
5558
5559 @kindex reverse-finish
5560 @item reverse-finish
5561 Just as the @code{finish} command takes you to the point where the
5562 current function returns, @code{reverse-finish} takes you to the point
5563 where it was called. Instead of ending up at the end of the current
5564 function invocation, you end up at the beginning.
5565
5566 @kindex set exec-direction
5567 @item set exec-direction
5568 Set the direction of target execution.
5569 @itemx set exec-direction reverse
5570 @cindex execute forward or backward in time
5571 @value{GDBN} will perform all execution commands in reverse, until the
5572 exec-direction mode is changed to ``forward''. Affected commands include
5573 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
5574 command cannot be used in reverse mode.
5575 @item set exec-direction forward
5576 @value{GDBN} will perform all execution commands in the normal fashion.
5577 This is the default.
5578 @end table
5579
5580
5581 @node Process Record and Replay
5582 @chapter Recording Inferior's Execution and Replaying It
5583 @cindex process record and replay
5584 @cindex recording inferior's execution and replaying it
5585
5586 On some platforms, @value{GDBN} provides a special @dfn{process record
5587 and replay} target that can record a log of the process execution, and
5588 replay it later with both forward and reverse execution commands.
5589
5590 @cindex replay mode
5591 When this target is in use, if the execution log includes the record
5592 for the next instruction, @value{GDBN} will debug in @dfn{replay
5593 mode}. In the replay mode, the inferior does not really execute code
5594 instructions. Instead, all the events that normally happen during
5595 code execution are taken from the execution log. While code is not
5596 really executed in replay mode, the values of registers (including the
5597 program counter register) and the memory of the inferior are still
5598 changed as they normally would. Their contents are taken from the
5599 execution log.
5600
5601 @cindex record mode
5602 If the record for the next instruction is not in the execution log,
5603 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
5604 inferior executes normally, and @value{GDBN} records the execution log
5605 for future replay.
5606
5607 The process record and replay target supports reverse execution
5608 (@pxref{Reverse Execution}), even if the platform on which the
5609 inferior runs does not. However, the reverse execution is limited in
5610 this case by the range of the instructions recorded in the execution
5611 log. In other words, reverse execution on platforms that don't
5612 support it directly can only be done in the replay mode.
5613
5614 When debugging in the reverse direction, @value{GDBN} will work in
5615 replay mode as long as the execution log includes the record for the
5616 previous instruction; otherwise, it will work in record mode, if the
5617 platform supports reverse execution, or stop if not.
5618
5619 For architecture environments that support process record and replay,
5620 @value{GDBN} provides the following commands:
5621
5622 @table @code
5623 @kindex target record
5624 @kindex record
5625 @kindex rec
5626 @item target record
5627 This command starts the process record and replay target. The process
5628 record and replay target can only debug a process that is already
5629 running. Therefore, you need first to start the process with the
5630 @kbd{run} or @kbd{start} commands, and then start the recording with
5631 the @kbd{target record} command.
5632
5633 Both @code{record} and @code{rec} are aliases of @code{target record}.
5634
5635 @cindex displaced stepping, and process record and replay
5636 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
5637 will be automatically disabled when process record and replay target
5638 is started. That's because the process record and replay target
5639 doesn't support displaced stepping.
5640
5641 @cindex non-stop mode, and process record and replay
5642 @cindex asynchronous execution, and process record and replay
5643 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
5644 the asynchronous execution mode (@pxref{Background Execution}), the
5645 process record and replay target cannot be started because it doesn't
5646 support these two modes.
5647
5648 @kindex record stop
5649 @kindex rec s
5650 @item record stop
5651 Stop the process record and replay target. When process record and
5652 replay target stops, the entire execution log will be deleted and the
5653 inferior will either be terminated, or will remain in its final state.
5654
5655 When you stop the process record and replay target in record mode (at
5656 the end of the execution log), the inferior will be stopped at the
5657 next instruction that would have been recorded. In other words, if
5658 you record for a while and then stop recording, the inferior process
5659 will be left in the same state as if the recording never happened.
5660
5661 On the other hand, if the process record and replay target is stopped
5662 while in replay mode (that is, not at the end of the execution log,
5663 but at some earlier point), the inferior process will become ``live''
5664 at that earlier state, and it will then be possible to continue the
5665 usual ``live'' debugging of the process from that state.
5666
5667 When the inferior process exits, or @value{GDBN} detaches from it,
5668 process record and replay target will automatically stop itself.
5669
5670 @kindex record save
5671 @item record save @var{filename}
5672 Save the execution log to a file @file{@var{filename}}.
5673 Default filename is @file{gdb_record.@var{process_id}}, where
5674 @var{process_id} is the process ID of the inferior.
5675
5676 @kindex record restore
5677 @item record restore @var{filename}
5678 Restore the execution log from a file @file{@var{filename}}.
5679 File must have been created with @code{record save}.
5680
5681 @kindex set record insn-number-max
5682 @item set record insn-number-max @var{limit}
5683 Set the limit of instructions to be recorded. Default value is 200000.
5684
5685 If @var{limit} is a positive number, then @value{GDBN} will start
5686 deleting instructions from the log once the number of the record
5687 instructions becomes greater than @var{limit}. For every new recorded
5688 instruction, @value{GDBN} will delete the earliest recorded
5689 instruction to keep the number of recorded instructions at the limit.
5690 (Since deleting recorded instructions loses information, @value{GDBN}
5691 lets you control what happens when the limit is reached, by means of
5692 the @code{stop-at-limit} option, described below.)
5693
5694 If @var{limit} is zero, @value{GDBN} will never delete recorded
5695 instructions from the execution log. The number of recorded
5696 instructions is unlimited in this case.
5697
5698 @kindex show record insn-number-max
5699 @item show record insn-number-max
5700 Show the limit of instructions to be recorded.
5701
5702 @kindex set record stop-at-limit
5703 @item set record stop-at-limit
5704 Control the behavior when the number of recorded instructions reaches
5705 the limit. If ON (the default), @value{GDBN} will stop when the limit
5706 is reached for the first time and ask you whether you want to stop the
5707 inferior or continue running it and recording the execution log. If
5708 you decide to continue recording, each new recorded instruction will
5709 cause the oldest one to be deleted.
5710
5711 If this option is OFF, @value{GDBN} will automatically delete the
5712 oldest record to make room for each new one, without asking.
5713
5714 @kindex show record stop-at-limit
5715 @item show record stop-at-limit
5716 Show the current setting of @code{stop-at-limit}.
5717
5718 @kindex set record memory-query
5719 @item set record memory-query
5720 Control the behavior when @value{GDBN} is unable to record memory
5721 changes caused by an instruction. If ON, @value{GDBN} will query
5722 whether to stop the inferior in that case.
5723
5724 If this option is OFF (the default), @value{GDBN} will automatically
5725 ignore the effect of such instructions on memory. Later, when
5726 @value{GDBN} replays this execution log, it will mark the log of this
5727 instruction as not accessible, and it will not affect the replay
5728 results.
5729
5730 @kindex show record memory-query
5731 @item show record memory-query
5732 Show the current setting of @code{memory-query}.
5733
5734 @kindex info record
5735 @item info record
5736 Show various statistics about the state of process record and its
5737 in-memory execution log buffer, including:
5738
5739 @itemize @bullet
5740 @item
5741 Whether in record mode or replay mode.
5742 @item
5743 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
5744 @item
5745 Highest recorded instruction number.
5746 @item
5747 Current instruction about to be replayed (if in replay mode).
5748 @item
5749 Number of instructions contained in the execution log.
5750 @item
5751 Maximum number of instructions that may be contained in the execution log.
5752 @end itemize
5753
5754 @kindex record delete
5755 @kindex rec del
5756 @item record delete
5757 When record target runs in replay mode (``in the past''), delete the
5758 subsequent execution log and begin to record a new execution log starting
5759 from the current address. This means you will abandon the previously
5760 recorded ``future'' and begin recording a new ``future''.
5761 @end table
5762
5763
5764 @node Stack
5765 @chapter Examining the Stack
5766
5767 When your program has stopped, the first thing you need to know is where it
5768 stopped and how it got there.
5769
5770 @cindex call stack
5771 Each time your program performs a function call, information about the call
5772 is generated.
5773 That information includes the location of the call in your program,
5774 the arguments of the call,
5775 and the local variables of the function being called.
5776 The information is saved in a block of data called a @dfn{stack frame}.
5777 The stack frames are allocated in a region of memory called the @dfn{call
5778 stack}.
5779
5780 When your program stops, the @value{GDBN} commands for examining the
5781 stack allow you to see all of this information.
5782
5783 @cindex selected frame
5784 One of the stack frames is @dfn{selected} by @value{GDBN} and many
5785 @value{GDBN} commands refer implicitly to the selected frame. In
5786 particular, whenever you ask @value{GDBN} for the value of a variable in
5787 your program, the value is found in the selected frame. There are
5788 special @value{GDBN} commands to select whichever frame you are
5789 interested in. @xref{Selection, ,Selecting a Frame}.
5790
5791 When your program stops, @value{GDBN} automatically selects the
5792 currently executing frame and describes it briefly, similar to the
5793 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
5794
5795 @menu
5796 * Frames:: Stack frames
5797 * Backtrace:: Backtraces
5798 * Selection:: Selecting a frame
5799 * Frame Info:: Information on a frame
5800
5801 @end menu
5802
5803 @node Frames
5804 @section Stack Frames
5805
5806 @cindex frame, definition
5807 @cindex stack frame
5808 The call stack is divided up into contiguous pieces called @dfn{stack
5809 frames}, or @dfn{frames} for short; each frame is the data associated
5810 with one call to one function. The frame contains the arguments given
5811 to the function, the function's local variables, and the address at
5812 which the function is executing.
5813
5814 @cindex initial frame
5815 @cindex outermost frame
5816 @cindex innermost frame
5817 When your program is started, the stack has only one frame, that of the
5818 function @code{main}. This is called the @dfn{initial} frame or the
5819 @dfn{outermost} frame. Each time a function is called, a new frame is
5820 made. Each time a function returns, the frame for that function invocation
5821 is eliminated. If a function is recursive, there can be many frames for
5822 the same function. The frame for the function in which execution is
5823 actually occurring is called the @dfn{innermost} frame. This is the most
5824 recently created of all the stack frames that still exist.
5825
5826 @cindex frame pointer
5827 Inside your program, stack frames are identified by their addresses. A
5828 stack frame consists of many bytes, each of which has its own address; each
5829 kind of computer has a convention for choosing one byte whose
5830 address serves as the address of the frame. Usually this address is kept
5831 in a register called the @dfn{frame pointer register}
5832 (@pxref{Registers, $fp}) while execution is going on in that frame.
5833
5834 @cindex frame number
5835 @value{GDBN} assigns numbers to all existing stack frames, starting with
5836 zero for the innermost frame, one for the frame that called it,
5837 and so on upward. These numbers do not really exist in your program;
5838 they are assigned by @value{GDBN} to give you a way of designating stack
5839 frames in @value{GDBN} commands.
5840
5841 @c The -fomit-frame-pointer below perennially causes hbox overflow
5842 @c underflow problems.
5843 @cindex frameless execution
5844 Some compilers provide a way to compile functions so that they operate
5845 without stack frames. (For example, the @value{NGCC} option
5846 @smallexample
5847 @samp{-fomit-frame-pointer}
5848 @end smallexample
5849 generates functions without a frame.)
5850 This is occasionally done with heavily used library functions to save
5851 the frame setup time. @value{GDBN} has limited facilities for dealing
5852 with these function invocations. If the innermost function invocation
5853 has no stack frame, @value{GDBN} nevertheless regards it as though
5854 it had a separate frame, which is numbered zero as usual, allowing
5855 correct tracing of the function call chain. However, @value{GDBN} has
5856 no provision for frameless functions elsewhere in the stack.
5857
5858 @table @code
5859 @kindex frame@r{, command}
5860 @cindex current stack frame
5861 @item frame @var{args}
5862 The @code{frame} command allows you to move from one stack frame to another,
5863 and to print the stack frame you select. @var{args} may be either the
5864 address of the frame or the stack frame number. Without an argument,
5865 @code{frame} prints the current stack frame.
5866
5867 @kindex select-frame
5868 @cindex selecting frame silently
5869 @item select-frame
5870 The @code{select-frame} command allows you to move from one stack frame
5871 to another without printing the frame. This is the silent version of
5872 @code{frame}.
5873 @end table
5874
5875 @node Backtrace
5876 @section Backtraces
5877
5878 @cindex traceback
5879 @cindex call stack traces
5880 A backtrace is a summary of how your program got where it is. It shows one
5881 line per frame, for many frames, starting with the currently executing
5882 frame (frame zero), followed by its caller (frame one), and on up the
5883 stack.
5884
5885 @table @code
5886 @kindex backtrace
5887 @kindex bt @r{(@code{backtrace})}
5888 @item backtrace
5889 @itemx bt
5890 Print a backtrace of the entire stack: one line per frame for all
5891 frames in the stack.
5892
5893 You can stop the backtrace at any time by typing the system interrupt
5894 character, normally @kbd{Ctrl-c}.
5895
5896 @item backtrace @var{n}
5897 @itemx bt @var{n}
5898 Similar, but print only the innermost @var{n} frames.
5899
5900 @item backtrace -@var{n}
5901 @itemx bt -@var{n}
5902 Similar, but print only the outermost @var{n} frames.
5903
5904 @item backtrace full
5905 @itemx bt full
5906 @itemx bt full @var{n}
5907 @itemx bt full -@var{n}
5908 Print the values of the local variables also. @var{n} specifies the
5909 number of frames to print, as described above.
5910 @end table
5911
5912 @kindex where
5913 @kindex info stack
5914 The names @code{where} and @code{info stack} (abbreviated @code{info s})
5915 are additional aliases for @code{backtrace}.
5916
5917 @cindex multiple threads, backtrace
5918 In a multi-threaded program, @value{GDBN} by default shows the
5919 backtrace only for the current thread. To display the backtrace for
5920 several or all of the threads, use the command @code{thread apply}
5921 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
5922 apply all backtrace}, @value{GDBN} will display the backtrace for all
5923 the threads; this is handy when you debug a core dump of a
5924 multi-threaded program.
5925
5926 Each line in the backtrace shows the frame number and the function name.
5927 The program counter value is also shown---unless you use @code{set
5928 print address off}. The backtrace also shows the source file name and
5929 line number, as well as the arguments to the function. The program
5930 counter value is omitted if it is at the beginning of the code for that
5931 line number.
5932
5933 Here is an example of a backtrace. It was made with the command
5934 @samp{bt 3}, so it shows the innermost three frames.
5935
5936 @smallexample
5937 @group
5938 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5939 at builtin.c:993
5940 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
5941 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
5942 at macro.c:71
5943 (More stack frames follow...)
5944 @end group
5945 @end smallexample
5946
5947 @noindent
5948 The display for frame zero does not begin with a program counter
5949 value, indicating that your program has stopped at the beginning of the
5950 code for line @code{993} of @code{builtin.c}.
5951
5952 @noindent
5953 The value of parameter @code{data} in frame 1 has been replaced by
5954 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
5955 only if it is a scalar (integer, pointer, enumeration, etc). See command
5956 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
5957 on how to configure the way function parameter values are printed.
5958
5959 @cindex value optimized out, in backtrace
5960 @cindex function call arguments, optimized out
5961 If your program was compiled with optimizations, some compilers will
5962 optimize away arguments passed to functions if those arguments are
5963 never used after the call. Such optimizations generate code that
5964 passes arguments through registers, but doesn't store those arguments
5965 in the stack frame. @value{GDBN} has no way of displaying such
5966 arguments in stack frames other than the innermost one. Here's what
5967 such a backtrace might look like:
5968
5969 @smallexample
5970 @group
5971 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5972 at builtin.c:993
5973 #1 0x6e38 in expand_macro (sym=<value optimized out>) at macro.c:242
5974 #2 0x6840 in expand_token (obs=0x0, t=<value optimized out>, td=0xf7fffb08)
5975 at macro.c:71
5976 (More stack frames follow...)
5977 @end group
5978 @end smallexample
5979
5980 @noindent
5981 The values of arguments that were not saved in their stack frames are
5982 shown as @samp{<value optimized out>}.
5983
5984 If you need to display the values of such optimized-out arguments,
5985 either deduce that from other variables whose values depend on the one
5986 you are interested in, or recompile without optimizations.
5987
5988 @cindex backtrace beyond @code{main} function
5989 @cindex program entry point
5990 @cindex startup code, and backtrace
5991 Most programs have a standard user entry point---a place where system
5992 libraries and startup code transition into user code. For C this is
5993 @code{main}@footnote{
5994 Note that embedded programs (the so-called ``free-standing''
5995 environment) are not required to have a @code{main} function as the
5996 entry point. They could even have multiple entry points.}.
5997 When @value{GDBN} finds the entry function in a backtrace
5998 it will terminate the backtrace, to avoid tracing into highly
5999 system-specific (and generally uninteresting) code.
6000
6001 If you need to examine the startup code, or limit the number of levels
6002 in a backtrace, you can change this behavior:
6003
6004 @table @code
6005 @item set backtrace past-main
6006 @itemx set backtrace past-main on
6007 @kindex set backtrace
6008 Backtraces will continue past the user entry point.
6009
6010 @item set backtrace past-main off
6011 Backtraces will stop when they encounter the user entry point. This is the
6012 default.
6013
6014 @item show backtrace past-main
6015 @kindex show backtrace
6016 Display the current user entry point backtrace policy.
6017
6018 @item set backtrace past-entry
6019 @itemx set backtrace past-entry on
6020 Backtraces will continue past the internal entry point of an application.
6021 This entry point is encoded by the linker when the application is built,
6022 and is likely before the user entry point @code{main} (or equivalent) is called.
6023
6024 @item set backtrace past-entry off
6025 Backtraces will stop when they encounter the internal entry point of an
6026 application. This is the default.
6027
6028 @item show backtrace past-entry
6029 Display the current internal entry point backtrace policy.
6030
6031 @item set backtrace limit @var{n}
6032 @itemx set backtrace limit 0
6033 @cindex backtrace limit
6034 Limit the backtrace to @var{n} levels. A value of zero means
6035 unlimited.
6036
6037 @item show backtrace limit
6038 Display the current limit on backtrace levels.
6039 @end table
6040
6041 @node Selection
6042 @section Selecting a Frame
6043
6044 Most commands for examining the stack and other data in your program work on
6045 whichever stack frame is selected at the moment. Here are the commands for
6046 selecting a stack frame; all of them finish by printing a brief description
6047 of the stack frame just selected.
6048
6049 @table @code
6050 @kindex frame@r{, selecting}
6051 @kindex f @r{(@code{frame})}
6052 @item frame @var{n}
6053 @itemx f @var{n}
6054 Select frame number @var{n}. Recall that frame zero is the innermost
6055 (currently executing) frame, frame one is the frame that called the
6056 innermost one, and so on. The highest-numbered frame is the one for
6057 @code{main}.
6058
6059 @item frame @var{addr}
6060 @itemx f @var{addr}
6061 Select the frame at address @var{addr}. This is useful mainly if the
6062 chaining of stack frames has been damaged by a bug, making it
6063 impossible for @value{GDBN} to assign numbers properly to all frames. In
6064 addition, this can be useful when your program has multiple stacks and
6065 switches between them.
6066
6067 On the SPARC architecture, @code{frame} needs two addresses to
6068 select an arbitrary frame: a frame pointer and a stack pointer.
6069
6070 On the MIPS and Alpha architecture, it needs two addresses: a stack
6071 pointer and a program counter.
6072
6073 On the 29k architecture, it needs three addresses: a register stack
6074 pointer, a program counter, and a memory stack pointer.
6075
6076 @kindex up
6077 @item up @var{n}
6078 Move @var{n} frames up the stack. For positive numbers @var{n}, this
6079 advances toward the outermost frame, to higher frame numbers, to frames
6080 that have existed longer. @var{n} defaults to one.
6081
6082 @kindex down
6083 @kindex do @r{(@code{down})}
6084 @item down @var{n}
6085 Move @var{n} frames down the stack. For positive numbers @var{n}, this
6086 advances toward the innermost frame, to lower frame numbers, to frames
6087 that were created more recently. @var{n} defaults to one. You may
6088 abbreviate @code{down} as @code{do}.
6089 @end table
6090
6091 All of these commands end by printing two lines of output describing the
6092 frame. The first line shows the frame number, the function name, the
6093 arguments, and the source file and line number of execution in that
6094 frame. The second line shows the text of that source line.
6095
6096 @need 1000
6097 For example:
6098
6099 @smallexample
6100 @group
6101 (@value{GDBP}) up
6102 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
6103 at env.c:10
6104 10 read_input_file (argv[i]);
6105 @end group
6106 @end smallexample
6107
6108 After such a printout, the @code{list} command with no arguments
6109 prints ten lines centered on the point of execution in the frame.
6110 You can also edit the program at the point of execution with your favorite
6111 editing program by typing @code{edit}.
6112 @xref{List, ,Printing Source Lines},
6113 for details.
6114
6115 @table @code
6116 @kindex down-silently
6117 @kindex up-silently
6118 @item up-silently @var{n}
6119 @itemx down-silently @var{n}
6120 These two commands are variants of @code{up} and @code{down},
6121 respectively; they differ in that they do their work silently, without
6122 causing display of the new frame. They are intended primarily for use
6123 in @value{GDBN} command scripts, where the output might be unnecessary and
6124 distracting.
6125 @end table
6126
6127 @node Frame Info
6128 @section Information About a Frame
6129
6130 There are several other commands to print information about the selected
6131 stack frame.
6132
6133 @table @code
6134 @item frame
6135 @itemx f
6136 When used without any argument, this command does not change which
6137 frame is selected, but prints a brief description of the currently
6138 selected stack frame. It can be abbreviated @code{f}. With an
6139 argument, this command is used to select a stack frame.
6140 @xref{Selection, ,Selecting a Frame}.
6141
6142 @kindex info frame
6143 @kindex info f @r{(@code{info frame})}
6144 @item info frame
6145 @itemx info f
6146 This command prints a verbose description of the selected stack frame,
6147 including:
6148
6149 @itemize @bullet
6150 @item
6151 the address of the frame
6152 @item
6153 the address of the next frame down (called by this frame)
6154 @item
6155 the address of the next frame up (caller of this frame)
6156 @item
6157 the language in which the source code corresponding to this frame is written
6158 @item
6159 the address of the frame's arguments
6160 @item
6161 the address of the frame's local variables
6162 @item
6163 the program counter saved in it (the address of execution in the caller frame)
6164 @item
6165 which registers were saved in the frame
6166 @end itemize
6167
6168 @noindent The verbose description is useful when
6169 something has gone wrong that has made the stack format fail to fit
6170 the usual conventions.
6171
6172 @item info frame @var{addr}
6173 @itemx info f @var{addr}
6174 Print a verbose description of the frame at address @var{addr}, without
6175 selecting that frame. The selected frame remains unchanged by this
6176 command. This requires the same kind of address (more than one for some
6177 architectures) that you specify in the @code{frame} command.
6178 @xref{Selection, ,Selecting a Frame}.
6179
6180 @kindex info args
6181 @item info args
6182 Print the arguments of the selected frame, each on a separate line.
6183
6184 @item info locals
6185 @kindex info locals
6186 Print the local variables of the selected frame, each on a separate
6187 line. These are all variables (declared either static or automatic)
6188 accessible at the point of execution of the selected frame.
6189
6190 @kindex info catch
6191 @cindex catch exceptions, list active handlers
6192 @cindex exception handlers, how to list
6193 @item info catch
6194 Print a list of all the exception handlers that are active in the
6195 current stack frame at the current point of execution. To see other
6196 exception handlers, visit the associated frame (using the @code{up},
6197 @code{down}, or @code{frame} commands); then type @code{info catch}.
6198 @xref{Set Catchpoints, , Setting Catchpoints}.
6199
6200 @end table
6201
6202
6203 @node Source
6204 @chapter Examining Source Files
6205
6206 @value{GDBN} can print parts of your program's source, since the debugging
6207 information recorded in the program tells @value{GDBN} what source files were
6208 used to build it. When your program stops, @value{GDBN} spontaneously prints
6209 the line where it stopped. Likewise, when you select a stack frame
6210 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
6211 execution in that frame has stopped. You can print other portions of
6212 source files by explicit command.
6213
6214 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
6215 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
6216 @value{GDBN} under @sc{gnu} Emacs}.
6217
6218 @menu
6219 * List:: Printing source lines
6220 * Specify Location:: How to specify code locations
6221 * Edit:: Editing source files
6222 * Search:: Searching source files
6223 * Source Path:: Specifying source directories
6224 * Machine Code:: Source and machine code
6225 @end menu
6226
6227 @node List
6228 @section Printing Source Lines
6229
6230 @kindex list
6231 @kindex l @r{(@code{list})}
6232 To print lines from a source file, use the @code{list} command
6233 (abbreviated @code{l}). By default, ten lines are printed.
6234 There are several ways to specify what part of the file you want to
6235 print; see @ref{Specify Location}, for the full list.
6236
6237 Here are the forms of the @code{list} command most commonly used:
6238
6239 @table @code
6240 @item list @var{linenum}
6241 Print lines centered around line number @var{linenum} in the
6242 current source file.
6243
6244 @item list @var{function}
6245 Print lines centered around the beginning of function
6246 @var{function}.
6247
6248 @item list
6249 Print more lines. If the last lines printed were printed with a
6250 @code{list} command, this prints lines following the last lines
6251 printed; however, if the last line printed was a solitary line printed
6252 as part of displaying a stack frame (@pxref{Stack, ,Examining the
6253 Stack}), this prints lines centered around that line.
6254
6255 @item list -
6256 Print lines just before the lines last printed.
6257 @end table
6258
6259 @cindex @code{list}, how many lines to display
6260 By default, @value{GDBN} prints ten source lines with any of these forms of
6261 the @code{list} command. You can change this using @code{set listsize}:
6262
6263 @table @code
6264 @kindex set listsize
6265 @item set listsize @var{count}
6266 Make the @code{list} command display @var{count} source lines (unless
6267 the @code{list} argument explicitly specifies some other number).
6268
6269 @kindex show listsize
6270 @item show listsize
6271 Display the number of lines that @code{list} prints.
6272 @end table
6273
6274 Repeating a @code{list} command with @key{RET} discards the argument,
6275 so it is equivalent to typing just @code{list}. This is more useful
6276 than listing the same lines again. An exception is made for an
6277 argument of @samp{-}; that argument is preserved in repetition so that
6278 each repetition moves up in the source file.
6279
6280 In general, the @code{list} command expects you to supply zero, one or two
6281 @dfn{linespecs}. Linespecs specify source lines; there are several ways
6282 of writing them (@pxref{Specify Location}), but the effect is always
6283 to specify some source line.
6284
6285 Here is a complete description of the possible arguments for @code{list}:
6286
6287 @table @code
6288 @item list @var{linespec}
6289 Print lines centered around the line specified by @var{linespec}.
6290
6291 @item list @var{first},@var{last}
6292 Print lines from @var{first} to @var{last}. Both arguments are
6293 linespecs. When a @code{list} command has two linespecs, and the
6294 source file of the second linespec is omitted, this refers to
6295 the same source file as the first linespec.
6296
6297 @item list ,@var{last}
6298 Print lines ending with @var{last}.
6299
6300 @item list @var{first},
6301 Print lines starting with @var{first}.
6302
6303 @item list +
6304 Print lines just after the lines last printed.
6305
6306 @item list -
6307 Print lines just before the lines last printed.
6308
6309 @item list
6310 As described in the preceding table.
6311 @end table
6312
6313 @node Specify Location
6314 @section Specifying a Location
6315 @cindex specifying location
6316 @cindex linespec
6317
6318 Several @value{GDBN} commands accept arguments that specify a location
6319 of your program's code. Since @value{GDBN} is a source-level
6320 debugger, a location usually specifies some line in the source code;
6321 for that reason, locations are also known as @dfn{linespecs}.
6322
6323 Here are all the different ways of specifying a code location that
6324 @value{GDBN} understands:
6325
6326 @table @code
6327 @item @var{linenum}
6328 Specifies the line number @var{linenum} of the current source file.
6329
6330 @item -@var{offset}
6331 @itemx +@var{offset}
6332 Specifies the line @var{offset} lines before or after the @dfn{current
6333 line}. For the @code{list} command, the current line is the last one
6334 printed; for the breakpoint commands, this is the line at which
6335 execution stopped in the currently selected @dfn{stack frame}
6336 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
6337 used as the second of the two linespecs in a @code{list} command,
6338 this specifies the line @var{offset} lines up or down from the first
6339 linespec.
6340
6341 @item @var{filename}:@var{linenum}
6342 Specifies the line @var{linenum} in the source file @var{filename}.
6343
6344 @item @var{function}
6345 Specifies the line that begins the body of the function @var{function}.
6346 For example, in C, this is the line with the open brace.
6347
6348 @item @var{filename}:@var{function}
6349 Specifies the line that begins the body of the function @var{function}
6350 in the file @var{filename}. You only need the file name with a
6351 function name to avoid ambiguity when there are identically named
6352 functions in different source files.
6353
6354 @item @var{label}
6355 Specifies the line at which the label named @var{label} appears.
6356 @value{GDBN} searches for the label in the function corresponding to
6357 the currently selected stack frame. If there is no current selected
6358 stack frame (for instance, if the inferior is not running), then
6359 @value{GDBN} will not search for a label.
6360
6361 @item *@var{address}
6362 Specifies the program address @var{address}. For line-oriented
6363 commands, such as @code{list} and @code{edit}, this specifies a source
6364 line that contains @var{address}. For @code{break} and other
6365 breakpoint oriented commands, this can be used to set breakpoints in
6366 parts of your program which do not have debugging information or
6367 source files.
6368
6369 Here @var{address} may be any expression valid in the current working
6370 language (@pxref{Languages, working language}) that specifies a code
6371 address. In addition, as a convenience, @value{GDBN} extends the
6372 semantics of expressions used in locations to cover the situations
6373 that frequently happen during debugging. Here are the various forms
6374 of @var{address}:
6375
6376 @table @code
6377 @item @var{expression}
6378 Any expression valid in the current working language.
6379
6380 @item @var{funcaddr}
6381 An address of a function or procedure derived from its name. In C,
6382 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
6383 simply the function's name @var{function} (and actually a special case
6384 of a valid expression). In Pascal and Modula-2, this is
6385 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
6386 (although the Pascal form also works).
6387
6388 This form specifies the address of the function's first instruction,
6389 before the stack frame and arguments have been set up.
6390
6391 @item '@var{filename}'::@var{funcaddr}
6392 Like @var{funcaddr} above, but also specifies the name of the source
6393 file explicitly. This is useful if the name of the function does not
6394 specify the function unambiguously, e.g., if there are several
6395 functions with identical names in different source files.
6396 @end table
6397
6398 @end table
6399
6400
6401 @node Edit
6402 @section Editing Source Files
6403 @cindex editing source files
6404
6405 @kindex edit
6406 @kindex e @r{(@code{edit})}
6407 To edit the lines in a source file, use the @code{edit} command.
6408 The editing program of your choice
6409 is invoked with the current line set to
6410 the active line in the program.
6411 Alternatively, there are several ways to specify what part of the file you
6412 want to print if you want to see other parts of the program:
6413
6414 @table @code
6415 @item edit @var{location}
6416 Edit the source file specified by @code{location}. Editing starts at
6417 that @var{location}, e.g., at the specified source line of the
6418 specified file. @xref{Specify Location}, for all the possible forms
6419 of the @var{location} argument; here are the forms of the @code{edit}
6420 command most commonly used:
6421
6422 @table @code
6423 @item edit @var{number}
6424 Edit the current source file with @var{number} as the active line number.
6425
6426 @item edit @var{function}
6427 Edit the file containing @var{function} at the beginning of its definition.
6428 @end table
6429
6430 @end table
6431
6432 @subsection Choosing your Editor
6433 You can customize @value{GDBN} to use any editor you want
6434 @footnote{
6435 The only restriction is that your editor (say @code{ex}), recognizes the
6436 following command-line syntax:
6437 @smallexample
6438 ex +@var{number} file
6439 @end smallexample
6440 The optional numeric value +@var{number} specifies the number of the line in
6441 the file where to start editing.}.
6442 By default, it is @file{@value{EDITOR}}, but you can change this
6443 by setting the environment variable @code{EDITOR} before using
6444 @value{GDBN}. For example, to configure @value{GDBN} to use the
6445 @code{vi} editor, you could use these commands with the @code{sh} shell:
6446 @smallexample
6447 EDITOR=/usr/bin/vi
6448 export EDITOR
6449 gdb @dots{}
6450 @end smallexample
6451 or in the @code{csh} shell,
6452 @smallexample
6453 setenv EDITOR /usr/bin/vi
6454 gdb @dots{}
6455 @end smallexample
6456
6457 @node Search
6458 @section Searching Source Files
6459 @cindex searching source files
6460
6461 There are two commands for searching through the current source file for a
6462 regular expression.
6463
6464 @table @code
6465 @kindex search
6466 @kindex forward-search
6467 @item forward-search @var{regexp}
6468 @itemx search @var{regexp}
6469 The command @samp{forward-search @var{regexp}} checks each line,
6470 starting with the one following the last line listed, for a match for
6471 @var{regexp}. It lists the line that is found. You can use the
6472 synonym @samp{search @var{regexp}} or abbreviate the command name as
6473 @code{fo}.
6474
6475 @kindex reverse-search
6476 @item reverse-search @var{regexp}
6477 The command @samp{reverse-search @var{regexp}} checks each line, starting
6478 with the one before the last line listed and going backward, for a match
6479 for @var{regexp}. It lists the line that is found. You can abbreviate
6480 this command as @code{rev}.
6481 @end table
6482
6483 @node Source Path
6484 @section Specifying Source Directories
6485
6486 @cindex source path
6487 @cindex directories for source files
6488 Executable programs sometimes do not record the directories of the source
6489 files from which they were compiled, just the names. Even when they do,
6490 the directories could be moved between the compilation and your debugging
6491 session. @value{GDBN} has a list of directories to search for source files;
6492 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
6493 it tries all the directories in the list, in the order they are present
6494 in the list, until it finds a file with the desired name.
6495
6496 For example, suppose an executable references the file
6497 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
6498 @file{/mnt/cross}. The file is first looked up literally; if this
6499 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
6500 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
6501 message is printed. @value{GDBN} does not look up the parts of the
6502 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
6503 Likewise, the subdirectories of the source path are not searched: if
6504 the source path is @file{/mnt/cross}, and the binary refers to
6505 @file{foo.c}, @value{GDBN} would not find it under
6506 @file{/mnt/cross/usr/src/foo-1.0/lib}.
6507
6508 Plain file names, relative file names with leading directories, file
6509 names containing dots, etc.@: are all treated as described above; for
6510 instance, if the source path is @file{/mnt/cross}, and the source file
6511 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
6512 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
6513 that---@file{/mnt/cross/foo.c}.
6514
6515 Note that the executable search path is @emph{not} used to locate the
6516 source files.
6517
6518 Whenever you reset or rearrange the source path, @value{GDBN} clears out
6519 any information it has cached about where source files are found and where
6520 each line is in the file.
6521
6522 @kindex directory
6523 @kindex dir
6524 When you start @value{GDBN}, its source path includes only @samp{cdir}
6525 and @samp{cwd}, in that order.
6526 To add other directories, use the @code{directory} command.
6527
6528 The search path is used to find both program source files and @value{GDBN}
6529 script files (read using the @samp{-command} option and @samp{source} command).
6530
6531 In addition to the source path, @value{GDBN} provides a set of commands
6532 that manage a list of source path substitution rules. A @dfn{substitution
6533 rule} specifies how to rewrite source directories stored in the program's
6534 debug information in case the sources were moved to a different
6535 directory between compilation and debugging. A rule is made of
6536 two strings, the first specifying what needs to be rewritten in
6537 the path, and the second specifying how it should be rewritten.
6538 In @ref{set substitute-path}, we name these two parts @var{from} and
6539 @var{to} respectively. @value{GDBN} does a simple string replacement
6540 of @var{from} with @var{to} at the start of the directory part of the
6541 source file name, and uses that result instead of the original file
6542 name to look up the sources.
6543
6544 Using the previous example, suppose the @file{foo-1.0} tree has been
6545 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
6546 @value{GDBN} to replace @file{/usr/src} in all source path names with
6547 @file{/mnt/cross}. The first lookup will then be
6548 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
6549 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
6550 substitution rule, use the @code{set substitute-path} command
6551 (@pxref{set substitute-path}).
6552
6553 To avoid unexpected substitution results, a rule is applied only if the
6554 @var{from} part of the directory name ends at a directory separator.
6555 For instance, a rule substituting @file{/usr/source} into
6556 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
6557 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
6558 is applied only at the beginning of the directory name, this rule will
6559 not be applied to @file{/root/usr/source/baz.c} either.
6560
6561 In many cases, you can achieve the same result using the @code{directory}
6562 command. However, @code{set substitute-path} can be more efficient in
6563 the case where the sources are organized in a complex tree with multiple
6564 subdirectories. With the @code{directory} command, you need to add each
6565 subdirectory of your project. If you moved the entire tree while
6566 preserving its internal organization, then @code{set substitute-path}
6567 allows you to direct the debugger to all the sources with one single
6568 command.
6569
6570 @code{set substitute-path} is also more than just a shortcut command.
6571 The source path is only used if the file at the original location no
6572 longer exists. On the other hand, @code{set substitute-path} modifies
6573 the debugger behavior to look at the rewritten location instead. So, if
6574 for any reason a source file that is not relevant to your executable is
6575 located at the original location, a substitution rule is the only
6576 method available to point @value{GDBN} at the new location.
6577
6578 @cindex @samp{--with-relocated-sources}
6579 @cindex default source path substitution
6580 You can configure a default source path substitution rule by
6581 configuring @value{GDBN} with the
6582 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
6583 should be the name of a directory under @value{GDBN}'s configured
6584 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
6585 directory names in debug information under @var{dir} will be adjusted
6586 automatically if the installed @value{GDBN} is moved to a new
6587 location. This is useful if @value{GDBN}, libraries or executables
6588 with debug information and corresponding source code are being moved
6589 together.
6590
6591 @table @code
6592 @item directory @var{dirname} @dots{}
6593 @item dir @var{dirname} @dots{}
6594 Add directory @var{dirname} to the front of the source path. Several
6595 directory names may be given to this command, separated by @samp{:}
6596 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
6597 part of absolute file names) or
6598 whitespace. You may specify a directory that is already in the source
6599 path; this moves it forward, so @value{GDBN} searches it sooner.
6600
6601 @kindex cdir
6602 @kindex cwd
6603 @vindex $cdir@r{, convenience variable}
6604 @vindex $cwd@r{, convenience variable}
6605 @cindex compilation directory
6606 @cindex current directory
6607 @cindex working directory
6608 @cindex directory, current
6609 @cindex directory, compilation
6610 You can use the string @samp{$cdir} to refer to the compilation
6611 directory (if one is recorded), and @samp{$cwd} to refer to the current
6612 working directory. @samp{$cwd} is not the same as @samp{.}---the former
6613 tracks the current working directory as it changes during your @value{GDBN}
6614 session, while the latter is immediately expanded to the current
6615 directory at the time you add an entry to the source path.
6616
6617 @item directory
6618 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
6619
6620 @c RET-repeat for @code{directory} is explicitly disabled, but since
6621 @c repeating it would be a no-op we do not say that. (thanks to RMS)
6622
6623 @item show directories
6624 @kindex show directories
6625 Print the source path: show which directories it contains.
6626
6627 @anchor{set substitute-path}
6628 @item set substitute-path @var{from} @var{to}
6629 @kindex set substitute-path
6630 Define a source path substitution rule, and add it at the end of the
6631 current list of existing substitution rules. If a rule with the same
6632 @var{from} was already defined, then the old rule is also deleted.
6633
6634 For example, if the file @file{/foo/bar/baz.c} was moved to
6635 @file{/mnt/cross/baz.c}, then the command
6636
6637 @smallexample
6638 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
6639 @end smallexample
6640
6641 @noindent
6642 will tell @value{GDBN} to replace @samp{/usr/src} with
6643 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
6644 @file{baz.c} even though it was moved.
6645
6646 In the case when more than one substitution rule have been defined,
6647 the rules are evaluated one by one in the order where they have been
6648 defined. The first one matching, if any, is selected to perform
6649 the substitution.
6650
6651 For instance, if we had entered the following commands:
6652
6653 @smallexample
6654 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
6655 (@value{GDBP}) set substitute-path /usr/src /mnt/src
6656 @end smallexample
6657
6658 @noindent
6659 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
6660 @file{/mnt/include/defs.h} by using the first rule. However, it would
6661 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
6662 @file{/mnt/src/lib/foo.c}.
6663
6664
6665 @item unset substitute-path [path]
6666 @kindex unset substitute-path
6667 If a path is specified, search the current list of substitution rules
6668 for a rule that would rewrite that path. Delete that rule if found.
6669 A warning is emitted by the debugger if no rule could be found.
6670
6671 If no path is specified, then all substitution rules are deleted.
6672
6673 @item show substitute-path [path]
6674 @kindex show substitute-path
6675 If a path is specified, then print the source path substitution rule
6676 which would rewrite that path, if any.
6677
6678 If no path is specified, then print all existing source path substitution
6679 rules.
6680
6681 @end table
6682
6683 If your source path is cluttered with directories that are no longer of
6684 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
6685 versions of source. You can correct the situation as follows:
6686
6687 @enumerate
6688 @item
6689 Use @code{directory} with no argument to reset the source path to its default value.
6690
6691 @item
6692 Use @code{directory} with suitable arguments to reinstall the
6693 directories you want in the source path. You can add all the
6694 directories in one command.
6695 @end enumerate
6696
6697 @node Machine Code
6698 @section Source and Machine Code
6699 @cindex source line and its code address
6700
6701 You can use the command @code{info line} to map source lines to program
6702 addresses (and vice versa), and the command @code{disassemble} to display
6703 a range of addresses as machine instructions. You can use the command
6704 @code{set disassemble-next-line} to set whether to disassemble next
6705 source line when execution stops. When run under @sc{gnu} Emacs
6706 mode, the @code{info line} command causes the arrow to point to the
6707 line specified. Also, @code{info line} prints addresses in symbolic form as
6708 well as hex.
6709
6710 @table @code
6711 @kindex info line
6712 @item info line @var{linespec}
6713 Print the starting and ending addresses of the compiled code for
6714 source line @var{linespec}. You can specify source lines in any of
6715 the ways documented in @ref{Specify Location}.
6716 @end table
6717
6718 For example, we can use @code{info line} to discover the location of
6719 the object code for the first line of function
6720 @code{m4_changequote}:
6721
6722 @c FIXME: I think this example should also show the addresses in
6723 @c symbolic form, as they usually would be displayed.
6724 @smallexample
6725 (@value{GDBP}) info line m4_changequote
6726 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
6727 @end smallexample
6728
6729 @noindent
6730 @cindex code address and its source line
6731 We can also inquire (using @code{*@var{addr}} as the form for
6732 @var{linespec}) what source line covers a particular address:
6733 @smallexample
6734 (@value{GDBP}) info line *0x63ff
6735 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
6736 @end smallexample
6737
6738 @cindex @code{$_} and @code{info line}
6739 @cindex @code{x} command, default address
6740 @kindex x@r{(examine), and} info line
6741 After @code{info line}, the default address for the @code{x} command
6742 is changed to the starting address of the line, so that @samp{x/i} is
6743 sufficient to begin examining the machine code (@pxref{Memory,
6744 ,Examining Memory}). Also, this address is saved as the value of the
6745 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
6746 Variables}).
6747
6748 @table @code
6749 @kindex disassemble
6750 @cindex assembly instructions
6751 @cindex instructions, assembly
6752 @cindex machine instructions
6753 @cindex listing machine instructions
6754 @item disassemble
6755 @itemx disassemble /m
6756 @itemx disassemble /r
6757 This specialized command dumps a range of memory as machine
6758 instructions. It can also print mixed source+disassembly by specifying
6759 the @code{/m} modifier and print the raw instructions in hex as well as
6760 in symbolic form by specifying the @code{/r}.
6761 The default memory range is the function surrounding the
6762 program counter of the selected frame. A single argument to this
6763 command is a program counter value; @value{GDBN} dumps the function
6764 surrounding this value. When two arguments are given, they should
6765 be separated by a comma, possibly surrounded by whitespace. The
6766 arguments specify a range of addresses to dump, in one of two forms:
6767
6768 @table @code
6769 @item @var{start},@var{end}
6770 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
6771 @item @var{start},+@var{length}
6772 the addresses from @var{start} (inclusive) to
6773 @code{@var{start}+@var{length}} (exclusive).
6774 @end table
6775
6776 @noindent
6777 When 2 arguments are specified, the name of the function is also
6778 printed (since there could be several functions in the given range).
6779
6780 The argument(s) can be any expression yielding a numeric value, such as
6781 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
6782
6783 If the range of memory being disassembled contains current program counter,
6784 the instruction at that location is shown with a @code{=>} marker.
6785 @end table
6786
6787 The following example shows the disassembly of a range of addresses of
6788 HP PA-RISC 2.0 code:
6789
6790 @smallexample
6791 (@value{GDBP}) disas 0x32c4, 0x32e4
6792 Dump of assembler code from 0x32c4 to 0x32e4:
6793 0x32c4 <main+204>: addil 0,dp
6794 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
6795 0x32cc <main+212>: ldil 0x3000,r31
6796 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
6797 0x32d4 <main+220>: ldo 0(r31),rp
6798 0x32d8 <main+224>: addil -0x800,dp
6799 0x32dc <main+228>: ldo 0x588(r1),r26
6800 0x32e0 <main+232>: ldil 0x3000,r31
6801 End of assembler dump.
6802 @end smallexample
6803
6804 Here is an example showing mixed source+assembly for Intel x86, when the
6805 program is stopped just after function prologue:
6806
6807 @smallexample
6808 (@value{GDBP}) disas /m main
6809 Dump of assembler code for function main:
6810 5 @{
6811 0x08048330 <+0>: push %ebp
6812 0x08048331 <+1>: mov %esp,%ebp
6813 0x08048333 <+3>: sub $0x8,%esp
6814 0x08048336 <+6>: and $0xfffffff0,%esp
6815 0x08048339 <+9>: sub $0x10,%esp
6816
6817 6 printf ("Hello.\n");
6818 => 0x0804833c <+12>: movl $0x8048440,(%esp)
6819 0x08048343 <+19>: call 0x8048284 <puts@@plt>
6820
6821 7 return 0;
6822 8 @}
6823 0x08048348 <+24>: mov $0x0,%eax
6824 0x0804834d <+29>: leave
6825 0x0804834e <+30>: ret
6826
6827 End of assembler dump.
6828 @end smallexample
6829
6830 Here is another example showing raw instructions in hex for AMD x86-64,
6831
6832 @smallexample
6833 (gdb) disas /r 0x400281,+10
6834 Dump of assembler code from 0x400281 to 0x40028b:
6835 0x0000000000400281: 38 36 cmp %dh,(%rsi)
6836 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
6837 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
6838 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
6839 End of assembler dump.
6840 @end smallexample
6841
6842 Some architectures have more than one commonly-used set of instruction
6843 mnemonics or other syntax.
6844
6845 For programs that were dynamically linked and use shared libraries,
6846 instructions that call functions or branch to locations in the shared
6847 libraries might show a seemingly bogus location---it's actually a
6848 location of the relocation table. On some architectures, @value{GDBN}
6849 might be able to resolve these to actual function names.
6850
6851 @table @code
6852 @kindex set disassembly-flavor
6853 @cindex Intel disassembly flavor
6854 @cindex AT&T disassembly flavor
6855 @item set disassembly-flavor @var{instruction-set}
6856 Select the instruction set to use when disassembling the
6857 program via the @code{disassemble} or @code{x/i} commands.
6858
6859 Currently this command is only defined for the Intel x86 family. You
6860 can set @var{instruction-set} to either @code{intel} or @code{att}.
6861 The default is @code{att}, the AT&T flavor used by default by Unix
6862 assemblers for x86-based targets.
6863
6864 @kindex show disassembly-flavor
6865 @item show disassembly-flavor
6866 Show the current setting of the disassembly flavor.
6867 @end table
6868
6869 @table @code
6870 @kindex set disassemble-next-line
6871 @kindex show disassemble-next-line
6872 @item set disassemble-next-line
6873 @itemx show disassemble-next-line
6874 Control whether or not @value{GDBN} will disassemble the next source
6875 line or instruction when execution stops. If ON, @value{GDBN} will
6876 display disassembly of the next source line when execution of the
6877 program being debugged stops. This is @emph{in addition} to
6878 displaying the source line itself, which @value{GDBN} always does if
6879 possible. If the next source line cannot be displayed for some reason
6880 (e.g., if @value{GDBN} cannot find the source file, or there's no line
6881 info in the debug info), @value{GDBN} will display disassembly of the
6882 next @emph{instruction} instead of showing the next source line. If
6883 AUTO, @value{GDBN} will display disassembly of next instruction only
6884 if the source line cannot be displayed. This setting causes
6885 @value{GDBN} to display some feedback when you step through a function
6886 with no line info or whose source file is unavailable. The default is
6887 OFF, which means never display the disassembly of the next line or
6888 instruction.
6889 @end table
6890
6891
6892 @node Data
6893 @chapter Examining Data
6894
6895 @cindex printing data
6896 @cindex examining data
6897 @kindex print
6898 @kindex inspect
6899 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
6900 @c document because it is nonstandard... Under Epoch it displays in a
6901 @c different window or something like that.
6902 The usual way to examine data in your program is with the @code{print}
6903 command (abbreviated @code{p}), or its synonym @code{inspect}. It
6904 evaluates and prints the value of an expression of the language your
6905 program is written in (@pxref{Languages, ,Using @value{GDBN} with
6906 Different Languages}). It may also print the expression using a
6907 Python-based pretty-printer (@pxref{Pretty Printing}).
6908
6909 @table @code
6910 @item print @var{expr}
6911 @itemx print /@var{f} @var{expr}
6912 @var{expr} is an expression (in the source language). By default the
6913 value of @var{expr} is printed in a format appropriate to its data type;
6914 you can choose a different format by specifying @samp{/@var{f}}, where
6915 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
6916 Formats}.
6917
6918 @item print
6919 @itemx print /@var{f}
6920 @cindex reprint the last value
6921 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
6922 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
6923 conveniently inspect the same value in an alternative format.
6924 @end table
6925
6926 A more low-level way of examining data is with the @code{x} command.
6927 It examines data in memory at a specified address and prints it in a
6928 specified format. @xref{Memory, ,Examining Memory}.
6929
6930 If you are interested in information about types, or about how the
6931 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
6932 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
6933 Table}.
6934
6935 @menu
6936 * Expressions:: Expressions
6937 * Ambiguous Expressions:: Ambiguous Expressions
6938 * Variables:: Program variables
6939 * Arrays:: Artificial arrays
6940 * Output Formats:: Output formats
6941 * Memory:: Examining memory
6942 * Auto Display:: Automatic display
6943 * Print Settings:: Print settings
6944 * Pretty Printing:: Python pretty printing
6945 * Value History:: Value history
6946 * Convenience Vars:: Convenience variables
6947 * Registers:: Registers
6948 * Floating Point Hardware:: Floating point hardware
6949 * Vector Unit:: Vector Unit
6950 * OS Information:: Auxiliary data provided by operating system
6951 * Memory Region Attributes:: Memory region attributes
6952 * Dump/Restore Files:: Copy between memory and a file
6953 * Core File Generation:: Cause a program dump its core
6954 * Character Sets:: Debugging programs that use a different
6955 character set than GDB does
6956 * Caching Remote Data:: Data caching for remote targets
6957 * Searching Memory:: Searching memory for a sequence of bytes
6958 @end menu
6959
6960 @node Expressions
6961 @section Expressions
6962
6963 @cindex expressions
6964 @code{print} and many other @value{GDBN} commands accept an expression and
6965 compute its value. Any kind of constant, variable or operator defined
6966 by the programming language you are using is valid in an expression in
6967 @value{GDBN}. This includes conditional expressions, function calls,
6968 casts, and string constants. It also includes preprocessor macros, if
6969 you compiled your program to include this information; see
6970 @ref{Compilation}.
6971
6972 @cindex arrays in expressions
6973 @value{GDBN} supports array constants in expressions input by
6974 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
6975 you can use the command @code{print @{1, 2, 3@}} to create an array
6976 of three integers. If you pass an array to a function or assign it
6977 to a program variable, @value{GDBN} copies the array to memory that
6978 is @code{malloc}ed in the target program.
6979
6980 Because C is so widespread, most of the expressions shown in examples in
6981 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
6982 Languages}, for information on how to use expressions in other
6983 languages.
6984
6985 In this section, we discuss operators that you can use in @value{GDBN}
6986 expressions regardless of your programming language.
6987
6988 @cindex casts, in expressions
6989 Casts are supported in all languages, not just in C, because it is so
6990 useful to cast a number into a pointer in order to examine a structure
6991 at that address in memory.
6992 @c FIXME: casts supported---Mod2 true?
6993
6994 @value{GDBN} supports these operators, in addition to those common
6995 to programming languages:
6996
6997 @table @code
6998 @item @@
6999 @samp{@@} is a binary operator for treating parts of memory as arrays.
7000 @xref{Arrays, ,Artificial Arrays}, for more information.
7001
7002 @item ::
7003 @samp{::} allows you to specify a variable in terms of the file or
7004 function where it is defined. @xref{Variables, ,Program Variables}.
7005
7006 @cindex @{@var{type}@}
7007 @cindex type casting memory
7008 @cindex memory, viewing as typed object
7009 @cindex casts, to view memory
7010 @item @{@var{type}@} @var{addr}
7011 Refers to an object of type @var{type} stored at address @var{addr} in
7012 memory. @var{addr} may be any expression whose value is an integer or
7013 pointer (but parentheses are required around binary operators, just as in
7014 a cast). This construct is allowed regardless of what kind of data is
7015 normally supposed to reside at @var{addr}.
7016 @end table
7017
7018 @node Ambiguous Expressions
7019 @section Ambiguous Expressions
7020 @cindex ambiguous expressions
7021
7022 Expressions can sometimes contain some ambiguous elements. For instance,
7023 some programming languages (notably Ada, C@t{++} and Objective-C) permit
7024 a single function name to be defined several times, for application in
7025 different contexts. This is called @dfn{overloading}. Another example
7026 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
7027 templates and is typically instantiated several times, resulting in
7028 the same function name being defined in different contexts.
7029
7030 In some cases and depending on the language, it is possible to adjust
7031 the expression to remove the ambiguity. For instance in C@t{++}, you
7032 can specify the signature of the function you want to break on, as in
7033 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
7034 qualified name of your function often makes the expression unambiguous
7035 as well.
7036
7037 When an ambiguity that needs to be resolved is detected, the debugger
7038 has the capability to display a menu of numbered choices for each
7039 possibility, and then waits for the selection with the prompt @samp{>}.
7040 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
7041 aborts the current command. If the command in which the expression was
7042 used allows more than one choice to be selected, the next option in the
7043 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
7044 choices.
7045
7046 For example, the following session excerpt shows an attempt to set a
7047 breakpoint at the overloaded symbol @code{String::after}.
7048 We choose three particular definitions of that function name:
7049
7050 @c FIXME! This is likely to change to show arg type lists, at least
7051 @smallexample
7052 @group
7053 (@value{GDBP}) b String::after
7054 [0] cancel
7055 [1] all
7056 [2] file:String.cc; line number:867
7057 [3] file:String.cc; line number:860
7058 [4] file:String.cc; line number:875
7059 [5] file:String.cc; line number:853
7060 [6] file:String.cc; line number:846
7061 [7] file:String.cc; line number:735
7062 > 2 4 6
7063 Breakpoint 1 at 0xb26c: file String.cc, line 867.
7064 Breakpoint 2 at 0xb344: file String.cc, line 875.
7065 Breakpoint 3 at 0xafcc: file String.cc, line 846.
7066 Multiple breakpoints were set.
7067 Use the "delete" command to delete unwanted
7068 breakpoints.
7069 (@value{GDBP})
7070 @end group
7071 @end smallexample
7072
7073 @table @code
7074 @kindex set multiple-symbols
7075 @item set multiple-symbols @var{mode}
7076 @cindex multiple-symbols menu
7077
7078 This option allows you to adjust the debugger behavior when an expression
7079 is ambiguous.
7080
7081 By default, @var{mode} is set to @code{all}. If the command with which
7082 the expression is used allows more than one choice, then @value{GDBN}
7083 automatically selects all possible choices. For instance, inserting
7084 a breakpoint on a function using an ambiguous name results in a breakpoint
7085 inserted on each possible match. However, if a unique choice must be made,
7086 then @value{GDBN} uses the menu to help you disambiguate the expression.
7087 For instance, printing the address of an overloaded function will result
7088 in the use of the menu.
7089
7090 When @var{mode} is set to @code{ask}, the debugger always uses the menu
7091 when an ambiguity is detected.
7092
7093 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
7094 an error due to the ambiguity and the command is aborted.
7095
7096 @kindex show multiple-symbols
7097 @item show multiple-symbols
7098 Show the current value of the @code{multiple-symbols} setting.
7099 @end table
7100
7101 @node Variables
7102 @section Program Variables
7103
7104 The most common kind of expression to use is the name of a variable
7105 in your program.
7106
7107 Variables in expressions are understood in the selected stack frame
7108 (@pxref{Selection, ,Selecting a Frame}); they must be either:
7109
7110 @itemize @bullet
7111 @item
7112 global (or file-static)
7113 @end itemize
7114
7115 @noindent or
7116
7117 @itemize @bullet
7118 @item
7119 visible according to the scope rules of the
7120 programming language from the point of execution in that frame
7121 @end itemize
7122
7123 @noindent This means that in the function
7124
7125 @smallexample
7126 foo (a)
7127 int a;
7128 @{
7129 bar (a);
7130 @{
7131 int b = test ();
7132 bar (b);
7133 @}
7134 @}
7135 @end smallexample
7136
7137 @noindent
7138 you can examine and use the variable @code{a} whenever your program is
7139 executing within the function @code{foo}, but you can only use or
7140 examine the variable @code{b} while your program is executing inside
7141 the block where @code{b} is declared.
7142
7143 @cindex variable name conflict
7144 There is an exception: you can refer to a variable or function whose
7145 scope is a single source file even if the current execution point is not
7146 in this file. But it is possible to have more than one such variable or
7147 function with the same name (in different source files). If that
7148 happens, referring to that name has unpredictable effects. If you wish,
7149 you can specify a static variable in a particular function or file,
7150 using the colon-colon (@code{::}) notation:
7151
7152 @cindex colon-colon, context for variables/functions
7153 @ifnotinfo
7154 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
7155 @cindex @code{::}, context for variables/functions
7156 @end ifnotinfo
7157 @smallexample
7158 @var{file}::@var{variable}
7159 @var{function}::@var{variable}
7160 @end smallexample
7161
7162 @noindent
7163 Here @var{file} or @var{function} is the name of the context for the
7164 static @var{variable}. In the case of file names, you can use quotes to
7165 make sure @value{GDBN} parses the file name as a single word---for example,
7166 to print a global value of @code{x} defined in @file{f2.c}:
7167
7168 @smallexample
7169 (@value{GDBP}) p 'f2.c'::x
7170 @end smallexample
7171
7172 @cindex C@t{++} scope resolution
7173 This use of @samp{::} is very rarely in conflict with the very similar
7174 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
7175 scope resolution operator in @value{GDBN} expressions.
7176 @c FIXME: Um, so what happens in one of those rare cases where it's in
7177 @c conflict?? --mew
7178
7179 @cindex wrong values
7180 @cindex variable values, wrong
7181 @cindex function entry/exit, wrong values of variables
7182 @cindex optimized code, wrong values of variables
7183 @quotation
7184 @emph{Warning:} Occasionally, a local variable may appear to have the
7185 wrong value at certain points in a function---just after entry to a new
7186 scope, and just before exit.
7187 @end quotation
7188 You may see this problem when you are stepping by machine instructions.
7189 This is because, on most machines, it takes more than one instruction to
7190 set up a stack frame (including local variable definitions); if you are
7191 stepping by machine instructions, variables may appear to have the wrong
7192 values until the stack frame is completely built. On exit, it usually
7193 also takes more than one machine instruction to destroy a stack frame;
7194 after you begin stepping through that group of instructions, local
7195 variable definitions may be gone.
7196
7197 This may also happen when the compiler does significant optimizations.
7198 To be sure of always seeing accurate values, turn off all optimization
7199 when compiling.
7200
7201 @cindex ``No symbol "foo" in current context''
7202 Another possible effect of compiler optimizations is to optimize
7203 unused variables out of existence, or assign variables to registers (as
7204 opposed to memory addresses). Depending on the support for such cases
7205 offered by the debug info format used by the compiler, @value{GDBN}
7206 might not be able to display values for such local variables. If that
7207 happens, @value{GDBN} will print a message like this:
7208
7209 @smallexample
7210 No symbol "foo" in current context.
7211 @end smallexample
7212
7213 To solve such problems, either recompile without optimizations, or use a
7214 different debug info format, if the compiler supports several such
7215 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler,
7216 usually supports the @option{-gstabs+} option. @option{-gstabs+}
7217 produces debug info in a format that is superior to formats such as
7218 COFF. You may be able to use DWARF 2 (@option{-gdwarf-2}), which is also
7219 an effective form for debug info. @xref{Debugging Options,,Options
7220 for Debugging Your Program or GCC, gcc.info, Using the @sc{gnu}
7221 Compiler Collection (GCC)}.
7222 @xref{C, ,C and C@t{++}}, for more information about debug info formats
7223 that are best suited to C@t{++} programs.
7224
7225 If you ask to print an object whose contents are unknown to
7226 @value{GDBN}, e.g., because its data type is not completely specified
7227 by the debug information, @value{GDBN} will say @samp{<incomplete
7228 type>}. @xref{Symbols, incomplete type}, for more about this.
7229
7230 Strings are identified as arrays of @code{char} values without specified
7231 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
7232 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
7233 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
7234 defines literal string type @code{"char"} as @code{char} without a sign.
7235 For program code
7236
7237 @smallexample
7238 char var0[] = "A";
7239 signed char var1[] = "A";
7240 @end smallexample
7241
7242 You get during debugging
7243 @smallexample
7244 (gdb) print var0
7245 $1 = "A"
7246 (gdb) print var1
7247 $2 = @{65 'A', 0 '\0'@}
7248 @end smallexample
7249
7250 @node Arrays
7251 @section Artificial Arrays
7252
7253 @cindex artificial array
7254 @cindex arrays
7255 @kindex @@@r{, referencing memory as an array}
7256 It is often useful to print out several successive objects of the
7257 same type in memory; a section of an array, or an array of
7258 dynamically determined size for which only a pointer exists in the
7259 program.
7260
7261 You can do this by referring to a contiguous span of memory as an
7262 @dfn{artificial array}, using the binary operator @samp{@@}. The left
7263 operand of @samp{@@} should be the first element of the desired array
7264 and be an individual object. The right operand should be the desired length
7265 of the array. The result is an array value whose elements are all of
7266 the type of the left argument. The first element is actually the left
7267 argument; the second element comes from bytes of memory immediately
7268 following those that hold the first element, and so on. Here is an
7269 example. If a program says
7270
7271 @smallexample
7272 int *array = (int *) malloc (len * sizeof (int));
7273 @end smallexample
7274
7275 @noindent
7276 you can print the contents of @code{array} with
7277
7278 @smallexample
7279 p *array@@len
7280 @end smallexample
7281
7282 The left operand of @samp{@@} must reside in memory. Array values made
7283 with @samp{@@} in this way behave just like other arrays in terms of
7284 subscripting, and are coerced to pointers when used in expressions.
7285 Artificial arrays most often appear in expressions via the value history
7286 (@pxref{Value History, ,Value History}), after printing one out.
7287
7288 Another way to create an artificial array is to use a cast.
7289 This re-interprets a value as if it were an array.
7290 The value need not be in memory:
7291 @smallexample
7292 (@value{GDBP}) p/x (short[2])0x12345678
7293 $1 = @{0x1234, 0x5678@}
7294 @end smallexample
7295
7296 As a convenience, if you leave the array length out (as in
7297 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
7298 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
7299 @smallexample
7300 (@value{GDBP}) p/x (short[])0x12345678
7301 $2 = @{0x1234, 0x5678@}
7302 @end smallexample
7303
7304 Sometimes the artificial array mechanism is not quite enough; in
7305 moderately complex data structures, the elements of interest may not
7306 actually be adjacent---for example, if you are interested in the values
7307 of pointers in an array. One useful work-around in this situation is
7308 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
7309 Variables}) as a counter in an expression that prints the first
7310 interesting value, and then repeat that expression via @key{RET}. For
7311 instance, suppose you have an array @code{dtab} of pointers to
7312 structures, and you are interested in the values of a field @code{fv}
7313 in each structure. Here is an example of what you might type:
7314
7315 @smallexample
7316 set $i = 0
7317 p dtab[$i++]->fv
7318 @key{RET}
7319 @key{RET}
7320 @dots{}
7321 @end smallexample
7322
7323 @node Output Formats
7324 @section Output Formats
7325
7326 @cindex formatted output
7327 @cindex output formats
7328 By default, @value{GDBN} prints a value according to its data type. Sometimes
7329 this is not what you want. For example, you might want to print a number
7330 in hex, or a pointer in decimal. Or you might want to view data in memory
7331 at a certain address as a character string or as an instruction. To do
7332 these things, specify an @dfn{output format} when you print a value.
7333
7334 The simplest use of output formats is to say how to print a value
7335 already computed. This is done by starting the arguments of the
7336 @code{print} command with a slash and a format letter. The format
7337 letters supported are:
7338
7339 @table @code
7340 @item x
7341 Regard the bits of the value as an integer, and print the integer in
7342 hexadecimal.
7343
7344 @item d
7345 Print as integer in signed decimal.
7346
7347 @item u
7348 Print as integer in unsigned decimal.
7349
7350 @item o
7351 Print as integer in octal.
7352
7353 @item t
7354 Print as integer in binary. The letter @samp{t} stands for ``two''.
7355 @footnote{@samp{b} cannot be used because these format letters are also
7356 used with the @code{x} command, where @samp{b} stands for ``byte'';
7357 see @ref{Memory,,Examining Memory}.}
7358
7359 @item a
7360 @cindex unknown address, locating
7361 @cindex locate address
7362 Print as an address, both absolute in hexadecimal and as an offset from
7363 the nearest preceding symbol. You can use this format used to discover
7364 where (in what function) an unknown address is located:
7365
7366 @smallexample
7367 (@value{GDBP}) p/a 0x54320
7368 $3 = 0x54320 <_initialize_vx+396>
7369 @end smallexample
7370
7371 @noindent
7372 The command @code{info symbol 0x54320} yields similar results.
7373 @xref{Symbols, info symbol}.
7374
7375 @item c
7376 Regard as an integer and print it as a character constant. This
7377 prints both the numerical value and its character representation. The
7378 character representation is replaced with the octal escape @samp{\nnn}
7379 for characters outside the 7-bit @sc{ascii} range.
7380
7381 Without this format, @value{GDBN} displays @code{char},
7382 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
7383 constants. Single-byte members of vectors are displayed as integer
7384 data.
7385
7386 @item f
7387 Regard the bits of the value as a floating point number and print
7388 using typical floating point syntax.
7389
7390 @item s
7391 @cindex printing strings
7392 @cindex printing byte arrays
7393 Regard as a string, if possible. With this format, pointers to single-byte
7394 data are displayed as null-terminated strings and arrays of single-byte data
7395 are displayed as fixed-length strings. Other values are displayed in their
7396 natural types.
7397
7398 Without this format, @value{GDBN} displays pointers to and arrays of
7399 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
7400 strings. Single-byte members of a vector are displayed as an integer
7401 array.
7402
7403 @item r
7404 @cindex raw printing
7405 Print using the @samp{raw} formatting. By default, @value{GDBN} will
7406 use a Python-based pretty-printer, if one is available (@pxref{Pretty
7407 Printing}). This typically results in a higher-level display of the
7408 value's contents. The @samp{r} format bypasses any Python
7409 pretty-printer which might exist.
7410 @end table
7411
7412 For example, to print the program counter in hex (@pxref{Registers}), type
7413
7414 @smallexample
7415 p/x $pc
7416 @end smallexample
7417
7418 @noindent
7419 Note that no space is required before the slash; this is because command
7420 names in @value{GDBN} cannot contain a slash.
7421
7422 To reprint the last value in the value history with a different format,
7423 you can use the @code{print} command with just a format and no
7424 expression. For example, @samp{p/x} reprints the last value in hex.
7425
7426 @node Memory
7427 @section Examining Memory
7428
7429 You can use the command @code{x} (for ``examine'') to examine memory in
7430 any of several formats, independently of your program's data types.
7431
7432 @cindex examining memory
7433 @table @code
7434 @kindex x @r{(examine memory)}
7435 @item x/@var{nfu} @var{addr}
7436 @itemx x @var{addr}
7437 @itemx x
7438 Use the @code{x} command to examine memory.
7439 @end table
7440
7441 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
7442 much memory to display and how to format it; @var{addr} is an
7443 expression giving the address where you want to start displaying memory.
7444 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
7445 Several commands set convenient defaults for @var{addr}.
7446
7447 @table @r
7448 @item @var{n}, the repeat count
7449 The repeat count is a decimal integer; the default is 1. It specifies
7450 how much memory (counting by units @var{u}) to display.
7451 @c This really is **decimal**; unaffected by 'set radix' as of GDB
7452 @c 4.1.2.
7453
7454 @item @var{f}, the display format
7455 The display format is one of the formats used by @code{print}
7456 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
7457 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
7458 The default is @samp{x} (hexadecimal) initially. The default changes
7459 each time you use either @code{x} or @code{print}.
7460
7461 @item @var{u}, the unit size
7462 The unit size is any of
7463
7464 @table @code
7465 @item b
7466 Bytes.
7467 @item h
7468 Halfwords (two bytes).
7469 @item w
7470 Words (four bytes). This is the initial default.
7471 @item g
7472 Giant words (eight bytes).
7473 @end table
7474
7475 Each time you specify a unit size with @code{x}, that size becomes the
7476 default unit the next time you use @code{x}. For the @samp{i} format,
7477 the unit size is ignored and is normally not written. For the @samp{s} format,
7478 the unit size defaults to @samp{b}, unless it is explicitly given.
7479 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
7480 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
7481 Note that the results depend on the programming language of the
7482 current compilation unit. If the language is C, the @samp{s}
7483 modifier will use the UTF-16 encoding while @samp{w} will use
7484 UTF-32. The encoding is set by the programming language and cannot
7485 be altered.
7486
7487 @item @var{addr}, starting display address
7488 @var{addr} is the address where you want @value{GDBN} to begin displaying
7489 memory. The expression need not have a pointer value (though it may);
7490 it is always interpreted as an integer address of a byte of memory.
7491 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
7492 @var{addr} is usually just after the last address examined---but several
7493 other commands also set the default address: @code{info breakpoints} (to
7494 the address of the last breakpoint listed), @code{info line} (to the
7495 starting address of a line), and @code{print} (if you use it to display
7496 a value from memory).
7497 @end table
7498
7499 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
7500 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
7501 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
7502 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
7503 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
7504
7505 Since the letters indicating unit sizes are all distinct from the
7506 letters specifying output formats, you do not have to remember whether
7507 unit size or format comes first; either order works. The output
7508 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
7509 (However, the count @var{n} must come first; @samp{wx4} does not work.)
7510
7511 Even though the unit size @var{u} is ignored for the formats @samp{s}
7512 and @samp{i}, you might still want to use a count @var{n}; for example,
7513 @samp{3i} specifies that you want to see three machine instructions,
7514 including any operands. For convenience, especially when used with
7515 the @code{display} command, the @samp{i} format also prints branch delay
7516 slot instructions, if any, beyond the count specified, which immediately
7517 follow the last instruction that is within the count. The command
7518 @code{disassemble} gives an alternative way of inspecting machine
7519 instructions; see @ref{Machine Code,,Source and Machine Code}.
7520
7521 All the defaults for the arguments to @code{x} are designed to make it
7522 easy to continue scanning memory with minimal specifications each time
7523 you use @code{x}. For example, after you have inspected three machine
7524 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
7525 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
7526 the repeat count @var{n} is used again; the other arguments default as
7527 for successive uses of @code{x}.
7528
7529 When examining machine instructions, the instruction at current program
7530 counter is shown with a @code{=>} marker. For example:
7531
7532 @smallexample
7533 (@value{GDBP}) x/5i $pc-6
7534 0x804837f <main+11>: mov %esp,%ebp
7535 0x8048381 <main+13>: push %ecx
7536 0x8048382 <main+14>: sub $0x4,%esp
7537 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
7538 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
7539 @end smallexample
7540
7541 @cindex @code{$_}, @code{$__}, and value history
7542 The addresses and contents printed by the @code{x} command are not saved
7543 in the value history because there is often too much of them and they
7544 would get in the way. Instead, @value{GDBN} makes these values available for
7545 subsequent use in expressions as values of the convenience variables
7546 @code{$_} and @code{$__}. After an @code{x} command, the last address
7547 examined is available for use in expressions in the convenience variable
7548 @code{$_}. The contents of that address, as examined, are available in
7549 the convenience variable @code{$__}.
7550
7551 If the @code{x} command has a repeat count, the address and contents saved
7552 are from the last memory unit printed; this is not the same as the last
7553 address printed if several units were printed on the last line of output.
7554
7555 @cindex remote memory comparison
7556 @cindex verify remote memory image
7557 When you are debugging a program running on a remote target machine
7558 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
7559 remote machine's memory against the executable file you downloaded to
7560 the target. The @code{compare-sections} command is provided for such
7561 situations.
7562
7563 @table @code
7564 @kindex compare-sections
7565 @item compare-sections @r{[}@var{section-name}@r{]}
7566 Compare the data of a loadable section @var{section-name} in the
7567 executable file of the program being debugged with the same section in
7568 the remote machine's memory, and report any mismatches. With no
7569 arguments, compares all loadable sections. This command's
7570 availability depends on the target's support for the @code{"qCRC"}
7571 remote request.
7572 @end table
7573
7574 @node Auto Display
7575 @section Automatic Display
7576 @cindex automatic display
7577 @cindex display of expressions
7578
7579 If you find that you want to print the value of an expression frequently
7580 (to see how it changes), you might want to add it to the @dfn{automatic
7581 display list} so that @value{GDBN} prints its value each time your program stops.
7582 Each expression added to the list is given a number to identify it;
7583 to remove an expression from the list, you specify that number.
7584 The automatic display looks like this:
7585
7586 @smallexample
7587 2: foo = 38
7588 3: bar[5] = (struct hack *) 0x3804
7589 @end smallexample
7590
7591 @noindent
7592 This display shows item numbers, expressions and their current values. As with
7593 displays you request manually using @code{x} or @code{print}, you can
7594 specify the output format you prefer; in fact, @code{display} decides
7595 whether to use @code{print} or @code{x} depending your format
7596 specification---it uses @code{x} if you specify either the @samp{i}
7597 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
7598
7599 @table @code
7600 @kindex display
7601 @item display @var{expr}
7602 Add the expression @var{expr} to the list of expressions to display
7603 each time your program stops. @xref{Expressions, ,Expressions}.
7604
7605 @code{display} does not repeat if you press @key{RET} again after using it.
7606
7607 @item display/@var{fmt} @var{expr}
7608 For @var{fmt} specifying only a display format and not a size or
7609 count, add the expression @var{expr} to the auto-display list but
7610 arrange to display it each time in the specified format @var{fmt}.
7611 @xref{Output Formats,,Output Formats}.
7612
7613 @item display/@var{fmt} @var{addr}
7614 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
7615 number of units, add the expression @var{addr} as a memory address to
7616 be examined each time your program stops. Examining means in effect
7617 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
7618 @end table
7619
7620 For example, @samp{display/i $pc} can be helpful, to see the machine
7621 instruction about to be executed each time execution stops (@samp{$pc}
7622 is a common name for the program counter; @pxref{Registers, ,Registers}).
7623
7624 @table @code
7625 @kindex delete display
7626 @kindex undisplay
7627 @item undisplay @var{dnums}@dots{}
7628 @itemx delete display @var{dnums}@dots{}
7629 Remove item numbers @var{dnums} from the list of expressions to display.
7630
7631 @code{undisplay} does not repeat if you press @key{RET} after using it.
7632 (Otherwise you would just get the error @samp{No display number @dots{}}.)
7633
7634 @kindex disable display
7635 @item disable display @var{dnums}@dots{}
7636 Disable the display of item numbers @var{dnums}. A disabled display
7637 item is not printed automatically, but is not forgotten. It may be
7638 enabled again later.
7639
7640 @kindex enable display
7641 @item enable display @var{dnums}@dots{}
7642 Enable display of item numbers @var{dnums}. It becomes effective once
7643 again in auto display of its expression, until you specify otherwise.
7644
7645 @item display
7646 Display the current values of the expressions on the list, just as is
7647 done when your program stops.
7648
7649 @kindex info display
7650 @item info display
7651 Print the list of expressions previously set up to display
7652 automatically, each one with its item number, but without showing the
7653 values. This includes disabled expressions, which are marked as such.
7654 It also includes expressions which would not be displayed right now
7655 because they refer to automatic variables not currently available.
7656 @end table
7657
7658 @cindex display disabled out of scope
7659 If a display expression refers to local variables, then it does not make
7660 sense outside the lexical context for which it was set up. Such an
7661 expression is disabled when execution enters a context where one of its
7662 variables is not defined. For example, if you give the command
7663 @code{display last_char} while inside a function with an argument
7664 @code{last_char}, @value{GDBN} displays this argument while your program
7665 continues to stop inside that function. When it stops elsewhere---where
7666 there is no variable @code{last_char}---the display is disabled
7667 automatically. The next time your program stops where @code{last_char}
7668 is meaningful, you can enable the display expression once again.
7669
7670 @node Print Settings
7671 @section Print Settings
7672
7673 @cindex format options
7674 @cindex print settings
7675 @value{GDBN} provides the following ways to control how arrays, structures,
7676 and symbols are printed.
7677
7678 @noindent
7679 These settings are useful for debugging programs in any language:
7680
7681 @table @code
7682 @kindex set print
7683 @item set print address
7684 @itemx set print address on
7685 @cindex print/don't print memory addresses
7686 @value{GDBN} prints memory addresses showing the location of stack
7687 traces, structure values, pointer values, breakpoints, and so forth,
7688 even when it also displays the contents of those addresses. The default
7689 is @code{on}. For example, this is what a stack frame display looks like with
7690 @code{set print address on}:
7691
7692 @smallexample
7693 @group
7694 (@value{GDBP}) f
7695 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
7696 at input.c:530
7697 530 if (lquote != def_lquote)
7698 @end group
7699 @end smallexample
7700
7701 @item set print address off
7702 Do not print addresses when displaying their contents. For example,
7703 this is the same stack frame displayed with @code{set print address off}:
7704
7705 @smallexample
7706 @group
7707 (@value{GDBP}) set print addr off
7708 (@value{GDBP}) f
7709 #0 set_quotes (lq="<<", rq=">>") at input.c:530
7710 530 if (lquote != def_lquote)
7711 @end group
7712 @end smallexample
7713
7714 You can use @samp{set print address off} to eliminate all machine
7715 dependent displays from the @value{GDBN} interface. For example, with
7716 @code{print address off}, you should get the same text for backtraces on
7717 all machines---whether or not they involve pointer arguments.
7718
7719 @kindex show print
7720 @item show print address
7721 Show whether or not addresses are to be printed.
7722 @end table
7723
7724 When @value{GDBN} prints a symbolic address, it normally prints the
7725 closest earlier symbol plus an offset. If that symbol does not uniquely
7726 identify the address (for example, it is a name whose scope is a single
7727 source file), you may need to clarify. One way to do this is with
7728 @code{info line}, for example @samp{info line *0x4537}. Alternately,
7729 you can set @value{GDBN} to print the source file and line number when
7730 it prints a symbolic address:
7731
7732 @table @code
7733 @item set print symbol-filename on
7734 @cindex source file and line of a symbol
7735 @cindex symbol, source file and line
7736 Tell @value{GDBN} to print the source file name and line number of a
7737 symbol in the symbolic form of an address.
7738
7739 @item set print symbol-filename off
7740 Do not print source file name and line number of a symbol. This is the
7741 default.
7742
7743 @item show print symbol-filename
7744 Show whether or not @value{GDBN} will print the source file name and
7745 line number of a symbol in the symbolic form of an address.
7746 @end table
7747
7748 Another situation where it is helpful to show symbol filenames and line
7749 numbers is when disassembling code; @value{GDBN} shows you the line
7750 number and source file that corresponds to each instruction.
7751
7752 Also, you may wish to see the symbolic form only if the address being
7753 printed is reasonably close to the closest earlier symbol:
7754
7755 @table @code
7756 @item set print max-symbolic-offset @var{max-offset}
7757 @cindex maximum value for offset of closest symbol
7758 Tell @value{GDBN} to only display the symbolic form of an address if the
7759 offset between the closest earlier symbol and the address is less than
7760 @var{max-offset}. The default is 0, which tells @value{GDBN}
7761 to always print the symbolic form of an address if any symbol precedes it.
7762
7763 @item show print max-symbolic-offset
7764 Ask how large the maximum offset is that @value{GDBN} prints in a
7765 symbolic address.
7766 @end table
7767
7768 @cindex wild pointer, interpreting
7769 @cindex pointer, finding referent
7770 If you have a pointer and you are not sure where it points, try
7771 @samp{set print symbol-filename on}. Then you can determine the name
7772 and source file location of the variable where it points, using
7773 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
7774 For example, here @value{GDBN} shows that a variable @code{ptt} points
7775 at another variable @code{t}, defined in @file{hi2.c}:
7776
7777 @smallexample
7778 (@value{GDBP}) set print symbol-filename on
7779 (@value{GDBP}) p/a ptt
7780 $4 = 0xe008 <t in hi2.c>
7781 @end smallexample
7782
7783 @quotation
7784 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
7785 does not show the symbol name and filename of the referent, even with
7786 the appropriate @code{set print} options turned on.
7787 @end quotation
7788
7789 Other settings control how different kinds of objects are printed:
7790
7791 @table @code
7792 @item set print array
7793 @itemx set print array on
7794 @cindex pretty print arrays
7795 Pretty print arrays. This format is more convenient to read,
7796 but uses more space. The default is off.
7797
7798 @item set print array off
7799 Return to compressed format for arrays.
7800
7801 @item show print array
7802 Show whether compressed or pretty format is selected for displaying
7803 arrays.
7804
7805 @cindex print array indexes
7806 @item set print array-indexes
7807 @itemx set print array-indexes on
7808 Print the index of each element when displaying arrays. May be more
7809 convenient to locate a given element in the array or quickly find the
7810 index of a given element in that printed array. The default is off.
7811
7812 @item set print array-indexes off
7813 Stop printing element indexes when displaying arrays.
7814
7815 @item show print array-indexes
7816 Show whether the index of each element is printed when displaying
7817 arrays.
7818
7819 @item set print elements @var{number-of-elements}
7820 @cindex number of array elements to print
7821 @cindex limit on number of printed array elements
7822 Set a limit on how many elements of an array @value{GDBN} will print.
7823 If @value{GDBN} is printing a large array, it stops printing after it has
7824 printed the number of elements set by the @code{set print elements} command.
7825 This limit also applies to the display of strings.
7826 When @value{GDBN} starts, this limit is set to 200.
7827 Setting @var{number-of-elements} to zero means that the printing is unlimited.
7828
7829 @item show print elements
7830 Display the number of elements of a large array that @value{GDBN} will print.
7831 If the number is 0, then the printing is unlimited.
7832
7833 @item set print frame-arguments @var{value}
7834 @kindex set print frame-arguments
7835 @cindex printing frame argument values
7836 @cindex print all frame argument values
7837 @cindex print frame argument values for scalars only
7838 @cindex do not print frame argument values
7839 This command allows to control how the values of arguments are printed
7840 when the debugger prints a frame (@pxref{Frames}). The possible
7841 values are:
7842
7843 @table @code
7844 @item all
7845 The values of all arguments are printed.
7846
7847 @item scalars
7848 Print the value of an argument only if it is a scalar. The value of more
7849 complex arguments such as arrays, structures, unions, etc, is replaced
7850 by @code{@dots{}}. This is the default. Here is an example where
7851 only scalar arguments are shown:
7852
7853 @smallexample
7854 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
7855 at frame-args.c:23
7856 @end smallexample
7857
7858 @item none
7859 None of the argument values are printed. Instead, the value of each argument
7860 is replaced by @code{@dots{}}. In this case, the example above now becomes:
7861
7862 @smallexample
7863 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
7864 at frame-args.c:23
7865 @end smallexample
7866 @end table
7867
7868 By default, only scalar arguments are printed. This command can be used
7869 to configure the debugger to print the value of all arguments, regardless
7870 of their type. However, it is often advantageous to not print the value
7871 of more complex parameters. For instance, it reduces the amount of
7872 information printed in each frame, making the backtrace more readable.
7873 Also, it improves performance when displaying Ada frames, because
7874 the computation of large arguments can sometimes be CPU-intensive,
7875 especially in large applications. Setting @code{print frame-arguments}
7876 to @code{scalars} (the default) or @code{none} avoids this computation,
7877 thus speeding up the display of each Ada frame.
7878
7879 @item show print frame-arguments
7880 Show how the value of arguments should be displayed when printing a frame.
7881
7882 @item set print repeats
7883 @cindex repeated array elements
7884 Set the threshold for suppressing display of repeated array
7885 elements. When the number of consecutive identical elements of an
7886 array exceeds the threshold, @value{GDBN} prints the string
7887 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
7888 identical repetitions, instead of displaying the identical elements
7889 themselves. Setting the threshold to zero will cause all elements to
7890 be individually printed. The default threshold is 10.
7891
7892 @item show print repeats
7893 Display the current threshold for printing repeated identical
7894 elements.
7895
7896 @item set print null-stop
7897 @cindex @sc{null} elements in arrays
7898 Cause @value{GDBN} to stop printing the characters of an array when the first
7899 @sc{null} is encountered. This is useful when large arrays actually
7900 contain only short strings.
7901 The default is off.
7902
7903 @item show print null-stop
7904 Show whether @value{GDBN} stops printing an array on the first
7905 @sc{null} character.
7906
7907 @item set print pretty on
7908 @cindex print structures in indented form
7909 @cindex indentation in structure display
7910 Cause @value{GDBN} to print structures in an indented format with one member
7911 per line, like this:
7912
7913 @smallexample
7914 @group
7915 $1 = @{
7916 next = 0x0,
7917 flags = @{
7918 sweet = 1,
7919 sour = 1
7920 @},
7921 meat = 0x54 "Pork"
7922 @}
7923 @end group
7924 @end smallexample
7925
7926 @item set print pretty off
7927 Cause @value{GDBN} to print structures in a compact format, like this:
7928
7929 @smallexample
7930 @group
7931 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
7932 meat = 0x54 "Pork"@}
7933 @end group
7934 @end smallexample
7935
7936 @noindent
7937 This is the default format.
7938
7939 @item show print pretty
7940 Show which format @value{GDBN} is using to print structures.
7941
7942 @item set print sevenbit-strings on
7943 @cindex eight-bit characters in strings
7944 @cindex octal escapes in strings
7945 Print using only seven-bit characters; if this option is set,
7946 @value{GDBN} displays any eight-bit characters (in strings or
7947 character values) using the notation @code{\}@var{nnn}. This setting is
7948 best if you are working in English (@sc{ascii}) and you use the
7949 high-order bit of characters as a marker or ``meta'' bit.
7950
7951 @item set print sevenbit-strings off
7952 Print full eight-bit characters. This allows the use of more
7953 international character sets, and is the default.
7954
7955 @item show print sevenbit-strings
7956 Show whether or not @value{GDBN} is printing only seven-bit characters.
7957
7958 @item set print union on
7959 @cindex unions in structures, printing
7960 Tell @value{GDBN} to print unions which are contained in structures
7961 and other unions. This is the default setting.
7962
7963 @item set print union off
7964 Tell @value{GDBN} not to print unions which are contained in
7965 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
7966 instead.
7967
7968 @item show print union
7969 Ask @value{GDBN} whether or not it will print unions which are contained in
7970 structures and other unions.
7971
7972 For example, given the declarations
7973
7974 @smallexample
7975 typedef enum @{Tree, Bug@} Species;
7976 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
7977 typedef enum @{Caterpillar, Cocoon, Butterfly@}
7978 Bug_forms;
7979
7980 struct thing @{
7981 Species it;
7982 union @{
7983 Tree_forms tree;
7984 Bug_forms bug;
7985 @} form;
7986 @};
7987
7988 struct thing foo = @{Tree, @{Acorn@}@};
7989 @end smallexample
7990
7991 @noindent
7992 with @code{set print union on} in effect @samp{p foo} would print
7993
7994 @smallexample
7995 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
7996 @end smallexample
7997
7998 @noindent
7999 and with @code{set print union off} in effect it would print
8000
8001 @smallexample
8002 $1 = @{it = Tree, form = @{...@}@}
8003 @end smallexample
8004
8005 @noindent
8006 @code{set print union} affects programs written in C-like languages
8007 and in Pascal.
8008 @end table
8009
8010 @need 1000
8011 @noindent
8012 These settings are of interest when debugging C@t{++} programs:
8013
8014 @table @code
8015 @cindex demangling C@t{++} names
8016 @item set print demangle
8017 @itemx set print demangle on
8018 Print C@t{++} names in their source form rather than in the encoded
8019 (``mangled'') form passed to the assembler and linker for type-safe
8020 linkage. The default is on.
8021
8022 @item show print demangle
8023 Show whether C@t{++} names are printed in mangled or demangled form.
8024
8025 @item set print asm-demangle
8026 @itemx set print asm-demangle on
8027 Print C@t{++} names in their source form rather than their mangled form, even
8028 in assembler code printouts such as instruction disassemblies.
8029 The default is off.
8030
8031 @item show print asm-demangle
8032 Show whether C@t{++} names in assembly listings are printed in mangled
8033 or demangled form.
8034
8035 @cindex C@t{++} symbol decoding style
8036 @cindex symbol decoding style, C@t{++}
8037 @kindex set demangle-style
8038 @item set demangle-style @var{style}
8039 Choose among several encoding schemes used by different compilers to
8040 represent C@t{++} names. The choices for @var{style} are currently:
8041
8042 @table @code
8043 @item auto
8044 Allow @value{GDBN} to choose a decoding style by inspecting your program.
8045
8046 @item gnu
8047 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
8048 This is the default.
8049
8050 @item hp
8051 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
8052
8053 @item lucid
8054 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
8055
8056 @item arm
8057 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
8058 @strong{Warning:} this setting alone is not sufficient to allow
8059 debugging @code{cfront}-generated executables. @value{GDBN} would
8060 require further enhancement to permit that.
8061
8062 @end table
8063 If you omit @var{style}, you will see a list of possible formats.
8064
8065 @item show demangle-style
8066 Display the encoding style currently in use for decoding C@t{++} symbols.
8067
8068 @item set print object
8069 @itemx set print object on
8070 @cindex derived type of an object, printing
8071 @cindex display derived types
8072 When displaying a pointer to an object, identify the @emph{actual}
8073 (derived) type of the object rather than the @emph{declared} type, using
8074 the virtual function table.
8075
8076 @item set print object off
8077 Display only the declared type of objects, without reference to the
8078 virtual function table. This is the default setting.
8079
8080 @item show print object
8081 Show whether actual, or declared, object types are displayed.
8082
8083 @item set print static-members
8084 @itemx set print static-members on
8085 @cindex static members of C@t{++} objects
8086 Print static members when displaying a C@t{++} object. The default is on.
8087
8088 @item set print static-members off
8089 Do not print static members when displaying a C@t{++} object.
8090
8091 @item show print static-members
8092 Show whether C@t{++} static members are printed or not.
8093
8094 @item set print pascal_static-members
8095 @itemx set print pascal_static-members on
8096 @cindex static members of Pascal objects
8097 @cindex Pascal objects, static members display
8098 Print static members when displaying a Pascal object. The default is on.
8099
8100 @item set print pascal_static-members off
8101 Do not print static members when displaying a Pascal object.
8102
8103 @item show print pascal_static-members
8104 Show whether Pascal static members are printed or not.
8105
8106 @c These don't work with HP ANSI C++ yet.
8107 @item set print vtbl
8108 @itemx set print vtbl on
8109 @cindex pretty print C@t{++} virtual function tables
8110 @cindex virtual functions (C@t{++}) display
8111 @cindex VTBL display
8112 Pretty print C@t{++} virtual function tables. The default is off.
8113 (The @code{vtbl} commands do not work on programs compiled with the HP
8114 ANSI C@t{++} compiler (@code{aCC}).)
8115
8116 @item set print vtbl off
8117 Do not pretty print C@t{++} virtual function tables.
8118
8119 @item show print vtbl
8120 Show whether C@t{++} virtual function tables are pretty printed, or not.
8121 @end table
8122
8123 @node Pretty Printing
8124 @section Pretty Printing
8125
8126 @value{GDBN} provides a mechanism to allow pretty-printing of values using
8127 Python code. It greatly simplifies the display of complex objects. This
8128 mechanism works for both MI and the CLI.
8129
8130 For example, here is how a C@t{++} @code{std::string} looks without a
8131 pretty-printer:
8132
8133 @smallexample
8134 (@value{GDBP}) print s
8135 $1 = @{
8136 static npos = 4294967295,
8137 _M_dataplus = @{
8138 <std::allocator<char>> = @{
8139 <__gnu_cxx::new_allocator<char>> = @{
8140 <No data fields>@}, <No data fields>
8141 @},
8142 members of std::basic_string<char, std::char_traits<char>,
8143 std::allocator<char> >::_Alloc_hider:
8144 _M_p = 0x804a014 "abcd"
8145 @}
8146 @}
8147 @end smallexample
8148
8149 With a pretty-printer for @code{std::string} only the contents are printed:
8150
8151 @smallexample
8152 (@value{GDBP}) print s
8153 $2 = "abcd"
8154 @end smallexample
8155
8156 For implementing pretty printers for new types you should read the Python API
8157 details (@pxref{Pretty Printing API}).
8158
8159 @node Value History
8160 @section Value History
8161
8162 @cindex value history
8163 @cindex history of values printed by @value{GDBN}
8164 Values printed by the @code{print} command are saved in the @value{GDBN}
8165 @dfn{value history}. This allows you to refer to them in other expressions.
8166 Values are kept until the symbol table is re-read or discarded
8167 (for example with the @code{file} or @code{symbol-file} commands).
8168 When the symbol table changes, the value history is discarded,
8169 since the values may contain pointers back to the types defined in the
8170 symbol table.
8171
8172 @cindex @code{$}
8173 @cindex @code{$$}
8174 @cindex history number
8175 The values printed are given @dfn{history numbers} by which you can
8176 refer to them. These are successive integers starting with one.
8177 @code{print} shows you the history number assigned to a value by
8178 printing @samp{$@var{num} = } before the value; here @var{num} is the
8179 history number.
8180
8181 To refer to any previous value, use @samp{$} followed by the value's
8182 history number. The way @code{print} labels its output is designed to
8183 remind you of this. Just @code{$} refers to the most recent value in
8184 the history, and @code{$$} refers to the value before that.
8185 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
8186 is the value just prior to @code{$$}, @code{$$1} is equivalent to
8187 @code{$$}, and @code{$$0} is equivalent to @code{$}.
8188
8189 For example, suppose you have just printed a pointer to a structure and
8190 want to see the contents of the structure. It suffices to type
8191
8192 @smallexample
8193 p *$
8194 @end smallexample
8195
8196 If you have a chain of structures where the component @code{next} points
8197 to the next one, you can print the contents of the next one with this:
8198
8199 @smallexample
8200 p *$.next
8201 @end smallexample
8202
8203 @noindent
8204 You can print successive links in the chain by repeating this
8205 command---which you can do by just typing @key{RET}.
8206
8207 Note that the history records values, not expressions. If the value of
8208 @code{x} is 4 and you type these commands:
8209
8210 @smallexample
8211 print x
8212 set x=5
8213 @end smallexample
8214
8215 @noindent
8216 then the value recorded in the value history by the @code{print} command
8217 remains 4 even though the value of @code{x} has changed.
8218
8219 @table @code
8220 @kindex show values
8221 @item show values
8222 Print the last ten values in the value history, with their item numbers.
8223 This is like @samp{p@ $$9} repeated ten times, except that @code{show
8224 values} does not change the history.
8225
8226 @item show values @var{n}
8227 Print ten history values centered on history item number @var{n}.
8228
8229 @item show values +
8230 Print ten history values just after the values last printed. If no more
8231 values are available, @code{show values +} produces no display.
8232 @end table
8233
8234 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
8235 same effect as @samp{show values +}.
8236
8237 @node Convenience Vars
8238 @section Convenience Variables
8239
8240 @cindex convenience variables
8241 @cindex user-defined variables
8242 @value{GDBN} provides @dfn{convenience variables} that you can use within
8243 @value{GDBN} to hold on to a value and refer to it later. These variables
8244 exist entirely within @value{GDBN}; they are not part of your program, and
8245 setting a convenience variable has no direct effect on further execution
8246 of your program. That is why you can use them freely.
8247
8248 Convenience variables are prefixed with @samp{$}. Any name preceded by
8249 @samp{$} can be used for a convenience variable, unless it is one of
8250 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
8251 (Value history references, in contrast, are @emph{numbers} preceded
8252 by @samp{$}. @xref{Value History, ,Value History}.)
8253
8254 You can save a value in a convenience variable with an assignment
8255 expression, just as you would set a variable in your program.
8256 For example:
8257
8258 @smallexample
8259 set $foo = *object_ptr
8260 @end smallexample
8261
8262 @noindent
8263 would save in @code{$foo} the value contained in the object pointed to by
8264 @code{object_ptr}.
8265
8266 Using a convenience variable for the first time creates it, but its
8267 value is @code{void} until you assign a new value. You can alter the
8268 value with another assignment at any time.
8269
8270 Convenience variables have no fixed types. You can assign a convenience
8271 variable any type of value, including structures and arrays, even if
8272 that variable already has a value of a different type. The convenience
8273 variable, when used as an expression, has the type of its current value.
8274
8275 @table @code
8276 @kindex show convenience
8277 @cindex show all user variables
8278 @item show convenience
8279 Print a list of convenience variables used so far, and their values.
8280 Abbreviated @code{show conv}.
8281
8282 @kindex init-if-undefined
8283 @cindex convenience variables, initializing
8284 @item init-if-undefined $@var{variable} = @var{expression}
8285 Set a convenience variable if it has not already been set. This is useful
8286 for user-defined commands that keep some state. It is similar, in concept,
8287 to using local static variables with initializers in C (except that
8288 convenience variables are global). It can also be used to allow users to
8289 override default values used in a command script.
8290
8291 If the variable is already defined then the expression is not evaluated so
8292 any side-effects do not occur.
8293 @end table
8294
8295 One of the ways to use a convenience variable is as a counter to be
8296 incremented or a pointer to be advanced. For example, to print
8297 a field from successive elements of an array of structures:
8298
8299 @smallexample
8300 set $i = 0
8301 print bar[$i++]->contents
8302 @end smallexample
8303
8304 @noindent
8305 Repeat that command by typing @key{RET}.
8306
8307 Some convenience variables are created automatically by @value{GDBN} and given
8308 values likely to be useful.
8309
8310 @table @code
8311 @vindex $_@r{, convenience variable}
8312 @item $_
8313 The variable @code{$_} is automatically set by the @code{x} command to
8314 the last address examined (@pxref{Memory, ,Examining Memory}). Other
8315 commands which provide a default address for @code{x} to examine also
8316 set @code{$_} to that address; these commands include @code{info line}
8317 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
8318 except when set by the @code{x} command, in which case it is a pointer
8319 to the type of @code{$__}.
8320
8321 @vindex $__@r{, convenience variable}
8322 @item $__
8323 The variable @code{$__} is automatically set by the @code{x} command
8324 to the value found in the last address examined. Its type is chosen
8325 to match the format in which the data was printed.
8326
8327 @item $_exitcode
8328 @vindex $_exitcode@r{, convenience variable}
8329 The variable @code{$_exitcode} is automatically set to the exit code when
8330 the program being debugged terminates.
8331
8332 @item $_sdata
8333 @vindex $_sdata@r{, inspect, convenience variable}
8334 The variable @code{$_sdata} contains extra collected static tracepoint
8335 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
8336 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
8337 if extra static tracepoint data has not been collected.
8338
8339 @item $_siginfo
8340 @vindex $_siginfo@r{, convenience variable}
8341 The variable @code{$_siginfo} contains extra signal information
8342 (@pxref{extra signal information}). Note that @code{$_siginfo}
8343 could be empty, if the application has not yet received any signals.
8344 For example, it will be empty before you execute the @code{run} command.
8345
8346 @item $_tlb
8347 @vindex $_tlb@r{, convenience variable}
8348 The variable @code{$_tlb} is automatically set when debugging
8349 applications running on MS-Windows in native mode or connected to
8350 gdbserver that supports the @code{qGetTIBAddr} request.
8351 @xref{General Query Packets}.
8352 This variable contains the address of the thread information block.
8353
8354 @end table
8355
8356 On HP-UX systems, if you refer to a function or variable name that
8357 begins with a dollar sign, @value{GDBN} searches for a user or system
8358 name first, before it searches for a convenience variable.
8359
8360 @cindex convenience functions
8361 @value{GDBN} also supplies some @dfn{convenience functions}. These
8362 have a syntax similar to convenience variables. A convenience
8363 function can be used in an expression just like an ordinary function;
8364 however, a convenience function is implemented internally to
8365 @value{GDBN}.
8366
8367 @table @code
8368 @item help function
8369 @kindex help function
8370 @cindex show all convenience functions
8371 Print a list of all convenience functions.
8372 @end table
8373
8374 @node Registers
8375 @section Registers
8376
8377 @cindex registers
8378 You can refer to machine register contents, in expressions, as variables
8379 with names starting with @samp{$}. The names of registers are different
8380 for each machine; use @code{info registers} to see the names used on
8381 your machine.
8382
8383 @table @code
8384 @kindex info registers
8385 @item info registers
8386 Print the names and values of all registers except floating-point
8387 and vector registers (in the selected stack frame).
8388
8389 @kindex info all-registers
8390 @cindex floating point registers
8391 @item info all-registers
8392 Print the names and values of all registers, including floating-point
8393 and vector registers (in the selected stack frame).
8394
8395 @item info registers @var{regname} @dots{}
8396 Print the @dfn{relativized} value of each specified register @var{regname}.
8397 As discussed in detail below, register values are normally relative to
8398 the selected stack frame. @var{regname} may be any register name valid on
8399 the machine you are using, with or without the initial @samp{$}.
8400 @end table
8401
8402 @cindex stack pointer register
8403 @cindex program counter register
8404 @cindex process status register
8405 @cindex frame pointer register
8406 @cindex standard registers
8407 @value{GDBN} has four ``standard'' register names that are available (in
8408 expressions) on most machines---whenever they do not conflict with an
8409 architecture's canonical mnemonics for registers. The register names
8410 @code{$pc} and @code{$sp} are used for the program counter register and
8411 the stack pointer. @code{$fp} is used for a register that contains a
8412 pointer to the current stack frame, and @code{$ps} is used for a
8413 register that contains the processor status. For example,
8414 you could print the program counter in hex with
8415
8416 @smallexample
8417 p/x $pc
8418 @end smallexample
8419
8420 @noindent
8421 or print the instruction to be executed next with
8422
8423 @smallexample
8424 x/i $pc
8425 @end smallexample
8426
8427 @noindent
8428 or add four to the stack pointer@footnote{This is a way of removing
8429 one word from the stack, on machines where stacks grow downward in
8430 memory (most machines, nowadays). This assumes that the innermost
8431 stack frame is selected; setting @code{$sp} is not allowed when other
8432 stack frames are selected. To pop entire frames off the stack,
8433 regardless of machine architecture, use @code{return};
8434 see @ref{Returning, ,Returning from a Function}.} with
8435
8436 @smallexample
8437 set $sp += 4
8438 @end smallexample
8439
8440 Whenever possible, these four standard register names are available on
8441 your machine even though the machine has different canonical mnemonics,
8442 so long as there is no conflict. The @code{info registers} command
8443 shows the canonical names. For example, on the SPARC, @code{info
8444 registers} displays the processor status register as @code{$psr} but you
8445 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
8446 is an alias for the @sc{eflags} register.
8447
8448 @value{GDBN} always considers the contents of an ordinary register as an
8449 integer when the register is examined in this way. Some machines have
8450 special registers which can hold nothing but floating point; these
8451 registers are considered to have floating point values. There is no way
8452 to refer to the contents of an ordinary register as floating point value
8453 (although you can @emph{print} it as a floating point value with
8454 @samp{print/f $@var{regname}}).
8455
8456 Some registers have distinct ``raw'' and ``virtual'' data formats. This
8457 means that the data format in which the register contents are saved by
8458 the operating system is not the same one that your program normally
8459 sees. For example, the registers of the 68881 floating point
8460 coprocessor are always saved in ``extended'' (raw) format, but all C
8461 programs expect to work with ``double'' (virtual) format. In such
8462 cases, @value{GDBN} normally works with the virtual format only (the format
8463 that makes sense for your program), but the @code{info registers} command
8464 prints the data in both formats.
8465
8466 @cindex SSE registers (x86)
8467 @cindex MMX registers (x86)
8468 Some machines have special registers whose contents can be interpreted
8469 in several different ways. For example, modern x86-based machines
8470 have SSE and MMX registers that can hold several values packed
8471 together in several different formats. @value{GDBN} refers to such
8472 registers in @code{struct} notation:
8473
8474 @smallexample
8475 (@value{GDBP}) print $xmm1
8476 $1 = @{
8477 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
8478 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
8479 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
8480 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
8481 v4_int32 = @{0, 20657912, 11, 13@},
8482 v2_int64 = @{88725056443645952, 55834574859@},
8483 uint128 = 0x0000000d0000000b013b36f800000000
8484 @}
8485 @end smallexample
8486
8487 @noindent
8488 To set values of such registers, you need to tell @value{GDBN} which
8489 view of the register you wish to change, as if you were assigning
8490 value to a @code{struct} member:
8491
8492 @smallexample
8493 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
8494 @end smallexample
8495
8496 Normally, register values are relative to the selected stack frame
8497 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
8498 value that the register would contain if all stack frames farther in
8499 were exited and their saved registers restored. In order to see the
8500 true contents of hardware registers, you must select the innermost
8501 frame (with @samp{frame 0}).
8502
8503 However, @value{GDBN} must deduce where registers are saved, from the machine
8504 code generated by your compiler. If some registers are not saved, or if
8505 @value{GDBN} is unable to locate the saved registers, the selected stack
8506 frame makes no difference.
8507
8508 @node Floating Point Hardware
8509 @section Floating Point Hardware
8510 @cindex floating point
8511
8512 Depending on the configuration, @value{GDBN} may be able to give
8513 you more information about the status of the floating point hardware.
8514
8515 @table @code
8516 @kindex info float
8517 @item info float
8518 Display hardware-dependent information about the floating
8519 point unit. The exact contents and layout vary depending on the
8520 floating point chip. Currently, @samp{info float} is supported on
8521 the ARM and x86 machines.
8522 @end table
8523
8524 @node Vector Unit
8525 @section Vector Unit
8526 @cindex vector unit
8527
8528 Depending on the configuration, @value{GDBN} may be able to give you
8529 more information about the status of the vector unit.
8530
8531 @table @code
8532 @kindex info vector
8533 @item info vector
8534 Display information about the vector unit. The exact contents and
8535 layout vary depending on the hardware.
8536 @end table
8537
8538 @node OS Information
8539 @section Operating System Auxiliary Information
8540 @cindex OS information
8541
8542 @value{GDBN} provides interfaces to useful OS facilities that can help
8543 you debug your program.
8544
8545 @cindex @code{ptrace} system call
8546 @cindex @code{struct user} contents
8547 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
8548 machines), it interfaces with the inferior via the @code{ptrace}
8549 system call. The operating system creates a special sata structure,
8550 called @code{struct user}, for this interface. You can use the
8551 command @code{info udot} to display the contents of this data
8552 structure.
8553
8554 @table @code
8555 @item info udot
8556 @kindex info udot
8557 Display the contents of the @code{struct user} maintained by the OS
8558 kernel for the program being debugged. @value{GDBN} displays the
8559 contents of @code{struct user} as a list of hex numbers, similar to
8560 the @code{examine} command.
8561 @end table
8562
8563 @cindex auxiliary vector
8564 @cindex vector, auxiliary
8565 Some operating systems supply an @dfn{auxiliary vector} to programs at
8566 startup. This is akin to the arguments and environment that you
8567 specify for a program, but contains a system-dependent variety of
8568 binary values that tell system libraries important details about the
8569 hardware, operating system, and process. Each value's purpose is
8570 identified by an integer tag; the meanings are well-known but system-specific.
8571 Depending on the configuration and operating system facilities,
8572 @value{GDBN} may be able to show you this information. For remote
8573 targets, this functionality may further depend on the remote stub's
8574 support of the @samp{qXfer:auxv:read} packet, see
8575 @ref{qXfer auxiliary vector read}.
8576
8577 @table @code
8578 @kindex info auxv
8579 @item info auxv
8580 Display the auxiliary vector of the inferior, which can be either a
8581 live process or a core dump file. @value{GDBN} prints each tag value
8582 numerically, and also shows names and text descriptions for recognized
8583 tags. Some values in the vector are numbers, some bit masks, and some
8584 pointers to strings or other data. @value{GDBN} displays each value in the
8585 most appropriate form for a recognized tag, and in hexadecimal for
8586 an unrecognized tag.
8587 @end table
8588
8589 On some targets, @value{GDBN} can access operating-system-specific information
8590 and display it to user, without interpretation. For remote targets,
8591 this functionality depends on the remote stub's support of the
8592 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
8593
8594 @table @code
8595 @kindex info os
8596 @item info os
8597 List the types of OS information available for the target. If the
8598 target does not return a list of possible types, this command will
8599 report an error.
8600
8601 @kindex info os processes
8602 @item info os processes
8603 Display the list of processes on the target. For each process,
8604 @value{GDBN} prints the process identifier, the name of the user, and
8605 the command corresponding to the process.
8606 @end table
8607
8608 @node Memory Region Attributes
8609 @section Memory Region Attributes
8610 @cindex memory region attributes
8611
8612 @dfn{Memory region attributes} allow you to describe special handling
8613 required by regions of your target's memory. @value{GDBN} uses
8614 attributes to determine whether to allow certain types of memory
8615 accesses; whether to use specific width accesses; and whether to cache
8616 target memory. By default the description of memory regions is
8617 fetched from the target (if the current target supports this), but the
8618 user can override the fetched regions.
8619
8620 Defined memory regions can be individually enabled and disabled. When a
8621 memory region is disabled, @value{GDBN} uses the default attributes when
8622 accessing memory in that region. Similarly, if no memory regions have
8623 been defined, @value{GDBN} uses the default attributes when accessing
8624 all memory.
8625
8626 When a memory region is defined, it is given a number to identify it;
8627 to enable, disable, or remove a memory region, you specify that number.
8628
8629 @table @code
8630 @kindex mem
8631 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
8632 Define a memory region bounded by @var{lower} and @var{upper} with
8633 attributes @var{attributes}@dots{}, and add it to the list of regions
8634 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
8635 case: it is treated as the target's maximum memory address.
8636 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
8637
8638 @item mem auto
8639 Discard any user changes to the memory regions and use target-supplied
8640 regions, if available, or no regions if the target does not support.
8641
8642 @kindex delete mem
8643 @item delete mem @var{nums}@dots{}
8644 Remove memory regions @var{nums}@dots{} from the list of regions
8645 monitored by @value{GDBN}.
8646
8647 @kindex disable mem
8648 @item disable mem @var{nums}@dots{}
8649 Disable monitoring of memory regions @var{nums}@dots{}.
8650 A disabled memory region is not forgotten.
8651 It may be enabled again later.
8652
8653 @kindex enable mem
8654 @item enable mem @var{nums}@dots{}
8655 Enable monitoring of memory regions @var{nums}@dots{}.
8656
8657 @kindex info mem
8658 @item info mem
8659 Print a table of all defined memory regions, with the following columns
8660 for each region:
8661
8662 @table @emph
8663 @item Memory Region Number
8664 @item Enabled or Disabled.
8665 Enabled memory regions are marked with @samp{y}.
8666 Disabled memory regions are marked with @samp{n}.
8667
8668 @item Lo Address
8669 The address defining the inclusive lower bound of the memory region.
8670
8671 @item Hi Address
8672 The address defining the exclusive upper bound of the memory region.
8673
8674 @item Attributes
8675 The list of attributes set for this memory region.
8676 @end table
8677 @end table
8678
8679
8680 @subsection Attributes
8681
8682 @subsubsection Memory Access Mode
8683 The access mode attributes set whether @value{GDBN} may make read or
8684 write accesses to a memory region.
8685
8686 While these attributes prevent @value{GDBN} from performing invalid
8687 memory accesses, they do nothing to prevent the target system, I/O DMA,
8688 etc.@: from accessing memory.
8689
8690 @table @code
8691 @item ro
8692 Memory is read only.
8693 @item wo
8694 Memory is write only.
8695 @item rw
8696 Memory is read/write. This is the default.
8697 @end table
8698
8699 @subsubsection Memory Access Size
8700 The access size attribute tells @value{GDBN} to use specific sized
8701 accesses in the memory region. Often memory mapped device registers
8702 require specific sized accesses. If no access size attribute is
8703 specified, @value{GDBN} may use accesses of any size.
8704
8705 @table @code
8706 @item 8
8707 Use 8 bit memory accesses.
8708 @item 16
8709 Use 16 bit memory accesses.
8710 @item 32
8711 Use 32 bit memory accesses.
8712 @item 64
8713 Use 64 bit memory accesses.
8714 @end table
8715
8716 @c @subsubsection Hardware/Software Breakpoints
8717 @c The hardware/software breakpoint attributes set whether @value{GDBN}
8718 @c will use hardware or software breakpoints for the internal breakpoints
8719 @c used by the step, next, finish, until, etc. commands.
8720 @c
8721 @c @table @code
8722 @c @item hwbreak
8723 @c Always use hardware breakpoints
8724 @c @item swbreak (default)
8725 @c @end table
8726
8727 @subsubsection Data Cache
8728 The data cache attributes set whether @value{GDBN} will cache target
8729 memory. While this generally improves performance by reducing debug
8730 protocol overhead, it can lead to incorrect results because @value{GDBN}
8731 does not know about volatile variables or memory mapped device
8732 registers.
8733
8734 @table @code
8735 @item cache
8736 Enable @value{GDBN} to cache target memory.
8737 @item nocache
8738 Disable @value{GDBN} from caching target memory. This is the default.
8739 @end table
8740
8741 @subsection Memory Access Checking
8742 @value{GDBN} can be instructed to refuse accesses to memory that is
8743 not explicitly described. This can be useful if accessing such
8744 regions has undesired effects for a specific target, or to provide
8745 better error checking. The following commands control this behaviour.
8746
8747 @table @code
8748 @kindex set mem inaccessible-by-default
8749 @item set mem inaccessible-by-default [on|off]
8750 If @code{on} is specified, make @value{GDBN} treat memory not
8751 explicitly described by the memory ranges as non-existent and refuse accesses
8752 to such memory. The checks are only performed if there's at least one
8753 memory range defined. If @code{off} is specified, make @value{GDBN}
8754 treat the memory not explicitly described by the memory ranges as RAM.
8755 The default value is @code{on}.
8756 @kindex show mem inaccessible-by-default
8757 @item show mem inaccessible-by-default
8758 Show the current handling of accesses to unknown memory.
8759 @end table
8760
8761
8762 @c @subsubsection Memory Write Verification
8763 @c The memory write verification attributes set whether @value{GDBN}
8764 @c will re-reads data after each write to verify the write was successful.
8765 @c
8766 @c @table @code
8767 @c @item verify
8768 @c @item noverify (default)
8769 @c @end table
8770
8771 @node Dump/Restore Files
8772 @section Copy Between Memory and a File
8773 @cindex dump/restore files
8774 @cindex append data to a file
8775 @cindex dump data to a file
8776 @cindex restore data from a file
8777
8778 You can use the commands @code{dump}, @code{append}, and
8779 @code{restore} to copy data between target memory and a file. The
8780 @code{dump} and @code{append} commands write data to a file, and the
8781 @code{restore} command reads data from a file back into the inferior's
8782 memory. Files may be in binary, Motorola S-record, Intel hex, or
8783 Tektronix Hex format; however, @value{GDBN} can only append to binary
8784 files.
8785
8786 @table @code
8787
8788 @kindex dump
8789 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8790 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
8791 Dump the contents of memory from @var{start_addr} to @var{end_addr},
8792 or the value of @var{expr}, to @var{filename} in the given format.
8793
8794 The @var{format} parameter may be any one of:
8795 @table @code
8796 @item binary
8797 Raw binary form.
8798 @item ihex
8799 Intel hex format.
8800 @item srec
8801 Motorola S-record format.
8802 @item tekhex
8803 Tektronix Hex format.
8804 @end table
8805
8806 @value{GDBN} uses the same definitions of these formats as the
8807 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
8808 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
8809 form.
8810
8811 @kindex append
8812 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8813 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
8814 Append the contents of memory from @var{start_addr} to @var{end_addr},
8815 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
8816 (@value{GDBN} can only append data to files in raw binary form.)
8817
8818 @kindex restore
8819 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
8820 Restore the contents of file @var{filename} into memory. The
8821 @code{restore} command can automatically recognize any known @sc{bfd}
8822 file format, except for raw binary. To restore a raw binary file you
8823 must specify the optional keyword @code{binary} after the filename.
8824
8825 If @var{bias} is non-zero, its value will be added to the addresses
8826 contained in the file. Binary files always start at address zero, so
8827 they will be restored at address @var{bias}. Other bfd files have
8828 a built-in location; they will be restored at offset @var{bias}
8829 from that location.
8830
8831 If @var{start} and/or @var{end} are non-zero, then only data between
8832 file offset @var{start} and file offset @var{end} will be restored.
8833 These offsets are relative to the addresses in the file, before
8834 the @var{bias} argument is applied.
8835
8836 @end table
8837
8838 @node Core File Generation
8839 @section How to Produce a Core File from Your Program
8840 @cindex dump core from inferior
8841
8842 A @dfn{core file} or @dfn{core dump} is a file that records the memory
8843 image of a running process and its process status (register values
8844 etc.). Its primary use is post-mortem debugging of a program that
8845 crashed while it ran outside a debugger. A program that crashes
8846 automatically produces a core file, unless this feature is disabled by
8847 the user. @xref{Files}, for information on invoking @value{GDBN} in
8848 the post-mortem debugging mode.
8849
8850 Occasionally, you may wish to produce a core file of the program you
8851 are debugging in order to preserve a snapshot of its state.
8852 @value{GDBN} has a special command for that.
8853
8854 @table @code
8855 @kindex gcore
8856 @kindex generate-core-file
8857 @item generate-core-file [@var{file}]
8858 @itemx gcore [@var{file}]
8859 Produce a core dump of the inferior process. The optional argument
8860 @var{file} specifies the file name where to put the core dump. If not
8861 specified, the file name defaults to @file{core.@var{pid}}, where
8862 @var{pid} is the inferior process ID.
8863
8864 Note that this command is implemented only for some systems (as of
8865 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
8866 @end table
8867
8868 @node Character Sets
8869 @section Character Sets
8870 @cindex character sets
8871 @cindex charset
8872 @cindex translating between character sets
8873 @cindex host character set
8874 @cindex target character set
8875
8876 If the program you are debugging uses a different character set to
8877 represent characters and strings than the one @value{GDBN} uses itself,
8878 @value{GDBN} can automatically translate between the character sets for
8879 you. The character set @value{GDBN} uses we call the @dfn{host
8880 character set}; the one the inferior program uses we call the
8881 @dfn{target character set}.
8882
8883 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
8884 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
8885 remote protocol (@pxref{Remote Debugging}) to debug a program
8886 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
8887 then the host character set is Latin-1, and the target character set is
8888 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
8889 target-charset EBCDIC-US}, then @value{GDBN} translates between
8890 @sc{ebcdic} and Latin 1 as you print character or string values, or use
8891 character and string literals in expressions.
8892
8893 @value{GDBN} has no way to automatically recognize which character set
8894 the inferior program uses; you must tell it, using the @code{set
8895 target-charset} command, described below.
8896
8897 Here are the commands for controlling @value{GDBN}'s character set
8898 support:
8899
8900 @table @code
8901 @item set target-charset @var{charset}
8902 @kindex set target-charset
8903 Set the current target character set to @var{charset}. To display the
8904 list of supported target character sets, type
8905 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
8906
8907 @item set host-charset @var{charset}
8908 @kindex set host-charset
8909 Set the current host character set to @var{charset}.
8910
8911 By default, @value{GDBN} uses a host character set appropriate to the
8912 system it is running on; you can override that default using the
8913 @code{set host-charset} command. On some systems, @value{GDBN} cannot
8914 automatically determine the appropriate host character set. In this
8915 case, @value{GDBN} uses @samp{UTF-8}.
8916
8917 @value{GDBN} can only use certain character sets as its host character
8918 set. If you type @kbd{@w{set target-charset @key{TAB}@key{TAB}}},
8919 @value{GDBN} will list the host character sets it supports.
8920
8921 @item set charset @var{charset}
8922 @kindex set charset
8923 Set the current host and target character sets to @var{charset}. As
8924 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
8925 @value{GDBN} will list the names of the character sets that can be used
8926 for both host and target.
8927
8928 @item show charset
8929 @kindex show charset
8930 Show the names of the current host and target character sets.
8931
8932 @item show host-charset
8933 @kindex show host-charset
8934 Show the name of the current host character set.
8935
8936 @item show target-charset
8937 @kindex show target-charset
8938 Show the name of the current target character set.
8939
8940 @item set target-wide-charset @var{charset}
8941 @kindex set target-wide-charset
8942 Set the current target's wide character set to @var{charset}. This is
8943 the character set used by the target's @code{wchar_t} type. To
8944 display the list of supported wide character sets, type
8945 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
8946
8947 @item show target-wide-charset
8948 @kindex show target-wide-charset
8949 Show the name of the current target's wide character set.
8950 @end table
8951
8952 Here is an example of @value{GDBN}'s character set support in action.
8953 Assume that the following source code has been placed in the file
8954 @file{charset-test.c}:
8955
8956 @smallexample
8957 #include <stdio.h>
8958
8959 char ascii_hello[]
8960 = @{72, 101, 108, 108, 111, 44, 32, 119,
8961 111, 114, 108, 100, 33, 10, 0@};
8962 char ibm1047_hello[]
8963 = @{200, 133, 147, 147, 150, 107, 64, 166,
8964 150, 153, 147, 132, 90, 37, 0@};
8965
8966 main ()
8967 @{
8968 printf ("Hello, world!\n");
8969 @}
8970 @end smallexample
8971
8972 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
8973 containing the string @samp{Hello, world!} followed by a newline,
8974 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
8975
8976 We compile the program, and invoke the debugger on it:
8977
8978 @smallexample
8979 $ gcc -g charset-test.c -o charset-test
8980 $ gdb -nw charset-test
8981 GNU gdb 2001-12-19-cvs
8982 Copyright 2001 Free Software Foundation, Inc.
8983 @dots{}
8984 (@value{GDBP})
8985 @end smallexample
8986
8987 We can use the @code{show charset} command to see what character sets
8988 @value{GDBN} is currently using to interpret and display characters and
8989 strings:
8990
8991 @smallexample
8992 (@value{GDBP}) show charset
8993 The current host and target character set is `ISO-8859-1'.
8994 (@value{GDBP})
8995 @end smallexample
8996
8997 For the sake of printing this manual, let's use @sc{ascii} as our
8998 initial character set:
8999 @smallexample
9000 (@value{GDBP}) set charset ASCII
9001 (@value{GDBP}) show charset
9002 The current host and target character set is `ASCII'.
9003 (@value{GDBP})
9004 @end smallexample
9005
9006 Let's assume that @sc{ascii} is indeed the correct character set for our
9007 host system --- in other words, let's assume that if @value{GDBN} prints
9008 characters using the @sc{ascii} character set, our terminal will display
9009 them properly. Since our current target character set is also
9010 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
9011
9012 @smallexample
9013 (@value{GDBP}) print ascii_hello
9014 $1 = 0x401698 "Hello, world!\n"
9015 (@value{GDBP}) print ascii_hello[0]
9016 $2 = 72 'H'
9017 (@value{GDBP})
9018 @end smallexample
9019
9020 @value{GDBN} uses the target character set for character and string
9021 literals you use in expressions:
9022
9023 @smallexample
9024 (@value{GDBP}) print '+'
9025 $3 = 43 '+'
9026 (@value{GDBP})
9027 @end smallexample
9028
9029 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
9030 character.
9031
9032 @value{GDBN} relies on the user to tell it which character set the
9033 target program uses. If we print @code{ibm1047_hello} while our target
9034 character set is still @sc{ascii}, we get jibberish:
9035
9036 @smallexample
9037 (@value{GDBP}) print ibm1047_hello
9038 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
9039 (@value{GDBP}) print ibm1047_hello[0]
9040 $5 = 200 '\310'
9041 (@value{GDBP})
9042 @end smallexample
9043
9044 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
9045 @value{GDBN} tells us the character sets it supports:
9046
9047 @smallexample
9048 (@value{GDBP}) set target-charset
9049 ASCII EBCDIC-US IBM1047 ISO-8859-1
9050 (@value{GDBP}) set target-charset
9051 @end smallexample
9052
9053 We can select @sc{ibm1047} as our target character set, and examine the
9054 program's strings again. Now the @sc{ascii} string is wrong, but
9055 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
9056 target character set, @sc{ibm1047}, to the host character set,
9057 @sc{ascii}, and they display correctly:
9058
9059 @smallexample
9060 (@value{GDBP}) set target-charset IBM1047
9061 (@value{GDBP}) show charset
9062 The current host character set is `ASCII'.
9063 The current target character set is `IBM1047'.
9064 (@value{GDBP}) print ascii_hello
9065 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
9066 (@value{GDBP}) print ascii_hello[0]
9067 $7 = 72 '\110'
9068 (@value{GDBP}) print ibm1047_hello
9069 $8 = 0x4016a8 "Hello, world!\n"
9070 (@value{GDBP}) print ibm1047_hello[0]
9071 $9 = 200 'H'
9072 (@value{GDBP})
9073 @end smallexample
9074
9075 As above, @value{GDBN} uses the target character set for character and
9076 string literals you use in expressions:
9077
9078 @smallexample
9079 (@value{GDBP}) print '+'
9080 $10 = 78 '+'
9081 (@value{GDBP})
9082 @end smallexample
9083
9084 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
9085 character.
9086
9087 @node Caching Remote Data
9088 @section Caching Data of Remote Targets
9089 @cindex caching data of remote targets
9090
9091 @value{GDBN} caches data exchanged between the debugger and a
9092 remote target (@pxref{Remote Debugging}). Such caching generally improves
9093 performance, because it reduces the overhead of the remote protocol by
9094 bundling memory reads and writes into large chunks. Unfortunately, simply
9095 caching everything would lead to incorrect results, since @value{GDBN}
9096 does not necessarily know anything about volatile values, memory-mapped I/O
9097 addresses, etc. Furthermore, in non-stop mode (@pxref{Non-Stop Mode})
9098 memory can be changed @emph{while} a gdb command is executing.
9099 Therefore, by default, @value{GDBN} only caches data
9100 known to be on the stack@footnote{In non-stop mode, it is moderately
9101 rare for a running thread to modify the stack of a stopped thread
9102 in a way that would interfere with a backtrace, and caching of
9103 stack reads provides a significant speed up of remote backtraces.}.
9104 Other regions of memory can be explicitly marked as
9105 cacheable; see @pxref{Memory Region Attributes}.
9106
9107 @table @code
9108 @kindex set remotecache
9109 @item set remotecache on
9110 @itemx set remotecache off
9111 This option no longer does anything; it exists for compatibility
9112 with old scripts.
9113
9114 @kindex show remotecache
9115 @item show remotecache
9116 Show the current state of the obsolete remotecache flag.
9117
9118 @kindex set stack-cache
9119 @item set stack-cache on
9120 @itemx set stack-cache off
9121 Enable or disable caching of stack accesses. When @code{ON}, use
9122 caching. By default, this option is @code{ON}.
9123
9124 @kindex show stack-cache
9125 @item show stack-cache
9126 Show the current state of data caching for memory accesses.
9127
9128 @kindex info dcache
9129 @item info dcache @r{[}line@r{]}
9130 Print the information about the data cache performance. The
9131 information displayed includes the dcache width and depth, and for
9132 each cache line, its number, address, and how many times it was
9133 referenced. This command is useful for debugging the data cache
9134 operation.
9135
9136 If a line number is specified, the contents of that line will be
9137 printed in hex.
9138 @end table
9139
9140 @node Searching Memory
9141 @section Search Memory
9142 @cindex searching memory
9143
9144 Memory can be searched for a particular sequence of bytes with the
9145 @code{find} command.
9146
9147 @table @code
9148 @kindex find
9149 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
9150 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
9151 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
9152 etc. The search begins at address @var{start_addr} and continues for either
9153 @var{len} bytes or through to @var{end_addr} inclusive.
9154 @end table
9155
9156 @var{s} and @var{n} are optional parameters.
9157 They may be specified in either order, apart or together.
9158
9159 @table @r
9160 @item @var{s}, search query size
9161 The size of each search query value.
9162
9163 @table @code
9164 @item b
9165 bytes
9166 @item h
9167 halfwords (two bytes)
9168 @item w
9169 words (four bytes)
9170 @item g
9171 giant words (eight bytes)
9172 @end table
9173
9174 All values are interpreted in the current language.
9175 This means, for example, that if the current source language is C/C@t{++}
9176 then searching for the string ``hello'' includes the trailing '\0'.
9177
9178 If the value size is not specified, it is taken from the
9179 value's type in the current language.
9180 This is useful when one wants to specify the search
9181 pattern as a mixture of types.
9182 Note that this means, for example, that in the case of C-like languages
9183 a search for an untyped 0x42 will search for @samp{(int) 0x42}
9184 which is typically four bytes.
9185
9186 @item @var{n}, maximum number of finds
9187 The maximum number of matches to print. The default is to print all finds.
9188 @end table
9189
9190 You can use strings as search values. Quote them with double-quotes
9191 (@code{"}).
9192 The string value is copied into the search pattern byte by byte,
9193 regardless of the endianness of the target and the size specification.
9194
9195 The address of each match found is printed as well as a count of the
9196 number of matches found.
9197
9198 The address of the last value found is stored in convenience variable
9199 @samp{$_}.
9200 A count of the number of matches is stored in @samp{$numfound}.
9201
9202 For example, if stopped at the @code{printf} in this function:
9203
9204 @smallexample
9205 void
9206 hello ()
9207 @{
9208 static char hello[] = "hello-hello";
9209 static struct @{ char c; short s; int i; @}
9210 __attribute__ ((packed)) mixed
9211 = @{ 'c', 0x1234, 0x87654321 @};
9212 printf ("%s\n", hello);
9213 @}
9214 @end smallexample
9215
9216 @noindent
9217 you get during debugging:
9218
9219 @smallexample
9220 (gdb) find &hello[0], +sizeof(hello), "hello"
9221 0x804956d <hello.1620+6>
9222 1 pattern found
9223 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
9224 0x8049567 <hello.1620>
9225 0x804956d <hello.1620+6>
9226 2 patterns found
9227 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
9228 0x8049567 <hello.1620>
9229 1 pattern found
9230 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
9231 0x8049560 <mixed.1625>
9232 1 pattern found
9233 (gdb) print $numfound
9234 $1 = 1
9235 (gdb) print $_
9236 $2 = (void *) 0x8049560
9237 @end smallexample
9238
9239 @node Optimized Code
9240 @chapter Debugging Optimized Code
9241 @cindex optimized code, debugging
9242 @cindex debugging optimized code
9243
9244 Almost all compilers support optimization. With optimization
9245 disabled, the compiler generates assembly code that corresponds
9246 directly to your source code, in a simplistic way. As the compiler
9247 applies more powerful optimizations, the generated assembly code
9248 diverges from your original source code. With help from debugging
9249 information generated by the compiler, @value{GDBN} can map from
9250 the running program back to constructs from your original source.
9251
9252 @value{GDBN} is more accurate with optimization disabled. If you
9253 can recompile without optimization, it is easier to follow the
9254 progress of your program during debugging. But, there are many cases
9255 where you may need to debug an optimized version.
9256
9257 When you debug a program compiled with @samp{-g -O}, remember that the
9258 optimizer has rearranged your code; the debugger shows you what is
9259 really there. Do not be too surprised when the execution path does not
9260 exactly match your source file! An extreme example: if you define a
9261 variable, but never use it, @value{GDBN} never sees that
9262 variable---because the compiler optimizes it out of existence.
9263
9264 Some things do not work as well with @samp{-g -O} as with just
9265 @samp{-g}, particularly on machines with instruction scheduling. If in
9266 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
9267 please report it to us as a bug (including a test case!).
9268 @xref{Variables}, for more information about debugging optimized code.
9269
9270 @menu
9271 * Inline Functions:: How @value{GDBN} presents inlining
9272 @end menu
9273
9274 @node Inline Functions
9275 @section Inline Functions
9276 @cindex inline functions, debugging
9277
9278 @dfn{Inlining} is an optimization that inserts a copy of the function
9279 body directly at each call site, instead of jumping to a shared
9280 routine. @value{GDBN} displays inlined functions just like
9281 non-inlined functions. They appear in backtraces. You can view their
9282 arguments and local variables, step into them with @code{step}, skip
9283 them with @code{next}, and escape from them with @code{finish}.
9284 You can check whether a function was inlined by using the
9285 @code{info frame} command.
9286
9287 For @value{GDBN} to support inlined functions, the compiler must
9288 record information about inlining in the debug information ---
9289 @value{NGCC} using the @sc{dwarf 2} format does this, and several
9290 other compilers do also. @value{GDBN} only supports inlined functions
9291 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
9292 do not emit two required attributes (@samp{DW_AT_call_file} and
9293 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
9294 function calls with earlier versions of @value{NGCC}. It instead
9295 displays the arguments and local variables of inlined functions as
9296 local variables in the caller.
9297
9298 The body of an inlined function is directly included at its call site;
9299 unlike a non-inlined function, there are no instructions devoted to
9300 the call. @value{GDBN} still pretends that the call site and the
9301 start of the inlined function are different instructions. Stepping to
9302 the call site shows the call site, and then stepping again shows
9303 the first line of the inlined function, even though no additional
9304 instructions are executed.
9305
9306 This makes source-level debugging much clearer; you can see both the
9307 context of the call and then the effect of the call. Only stepping by
9308 a single instruction using @code{stepi} or @code{nexti} does not do
9309 this; single instruction steps always show the inlined body.
9310
9311 There are some ways that @value{GDBN} does not pretend that inlined
9312 function calls are the same as normal calls:
9313
9314 @itemize @bullet
9315 @item
9316 You cannot set breakpoints on inlined functions. @value{GDBN}
9317 either reports that there is no symbol with that name, or else sets the
9318 breakpoint only on non-inlined copies of the function. This limitation
9319 will be removed in a future version of @value{GDBN}; until then,
9320 set a breakpoint by line number on the first line of the inlined
9321 function instead.
9322
9323 @item
9324 Setting breakpoints at the call site of an inlined function may not
9325 work, because the call site does not contain any code. @value{GDBN}
9326 may incorrectly move the breakpoint to the next line of the enclosing
9327 function, after the call. This limitation will be removed in a future
9328 version of @value{GDBN}; until then, set a breakpoint on an earlier line
9329 or inside the inlined function instead.
9330
9331 @item
9332 @value{GDBN} cannot locate the return value of inlined calls after
9333 using the @code{finish} command. This is a limitation of compiler-generated
9334 debugging information; after @code{finish}, you can step to the next line
9335 and print a variable where your program stored the return value.
9336
9337 @end itemize
9338
9339
9340 @node Macros
9341 @chapter C Preprocessor Macros
9342
9343 Some languages, such as C and C@t{++}, provide a way to define and invoke
9344 ``preprocessor macros'' which expand into strings of tokens.
9345 @value{GDBN} can evaluate expressions containing macro invocations, show
9346 the result of macro expansion, and show a macro's definition, including
9347 where it was defined.
9348
9349 You may need to compile your program specially to provide @value{GDBN}
9350 with information about preprocessor macros. Most compilers do not
9351 include macros in their debugging information, even when you compile
9352 with the @option{-g} flag. @xref{Compilation}.
9353
9354 A program may define a macro at one point, remove that definition later,
9355 and then provide a different definition after that. Thus, at different
9356 points in the program, a macro may have different definitions, or have
9357 no definition at all. If there is a current stack frame, @value{GDBN}
9358 uses the macros in scope at that frame's source code line. Otherwise,
9359 @value{GDBN} uses the macros in scope at the current listing location;
9360 see @ref{List}.
9361
9362 Whenever @value{GDBN} evaluates an expression, it always expands any
9363 macro invocations present in the expression. @value{GDBN} also provides
9364 the following commands for working with macros explicitly.
9365
9366 @table @code
9367
9368 @kindex macro expand
9369 @cindex macro expansion, showing the results of preprocessor
9370 @cindex preprocessor macro expansion, showing the results of
9371 @cindex expanding preprocessor macros
9372 @item macro expand @var{expression}
9373 @itemx macro exp @var{expression}
9374 Show the results of expanding all preprocessor macro invocations in
9375 @var{expression}. Since @value{GDBN} simply expands macros, but does
9376 not parse the result, @var{expression} need not be a valid expression;
9377 it can be any string of tokens.
9378
9379 @kindex macro exp1
9380 @item macro expand-once @var{expression}
9381 @itemx macro exp1 @var{expression}
9382 @cindex expand macro once
9383 @i{(This command is not yet implemented.)} Show the results of
9384 expanding those preprocessor macro invocations that appear explicitly in
9385 @var{expression}. Macro invocations appearing in that expansion are
9386 left unchanged. This command allows you to see the effect of a
9387 particular macro more clearly, without being confused by further
9388 expansions. Since @value{GDBN} simply expands macros, but does not
9389 parse the result, @var{expression} need not be a valid expression; it
9390 can be any string of tokens.
9391
9392 @kindex info macro
9393 @cindex macro definition, showing
9394 @cindex definition, showing a macro's
9395 @item info macro @var{macro}
9396 Show the definition of the macro named @var{macro}, and describe the
9397 source location or compiler command-line where that definition was established.
9398
9399 @kindex macro define
9400 @cindex user-defined macros
9401 @cindex defining macros interactively
9402 @cindex macros, user-defined
9403 @item macro define @var{macro} @var{replacement-list}
9404 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
9405 Introduce a definition for a preprocessor macro named @var{macro},
9406 invocations of which are replaced by the tokens given in
9407 @var{replacement-list}. The first form of this command defines an
9408 ``object-like'' macro, which takes no arguments; the second form
9409 defines a ``function-like'' macro, which takes the arguments given in
9410 @var{arglist}.
9411
9412 A definition introduced by this command is in scope in every
9413 expression evaluated in @value{GDBN}, until it is removed with the
9414 @code{macro undef} command, described below. The definition overrides
9415 all definitions for @var{macro} present in the program being debugged,
9416 as well as any previous user-supplied definition.
9417
9418 @kindex macro undef
9419 @item macro undef @var{macro}
9420 Remove any user-supplied definition for the macro named @var{macro}.
9421 This command only affects definitions provided with the @code{macro
9422 define} command, described above; it cannot remove definitions present
9423 in the program being debugged.
9424
9425 @kindex macro list
9426 @item macro list
9427 List all the macros defined using the @code{macro define} command.
9428 @end table
9429
9430 @cindex macros, example of debugging with
9431 Here is a transcript showing the above commands in action. First, we
9432 show our source files:
9433
9434 @smallexample
9435 $ cat sample.c
9436 #include <stdio.h>
9437 #include "sample.h"
9438
9439 #define M 42
9440 #define ADD(x) (M + x)
9441
9442 main ()
9443 @{
9444 #define N 28
9445 printf ("Hello, world!\n");
9446 #undef N
9447 printf ("We're so creative.\n");
9448 #define N 1729
9449 printf ("Goodbye, world!\n");
9450 @}
9451 $ cat sample.h
9452 #define Q <
9453 $
9454 @end smallexample
9455
9456 Now, we compile the program using the @sc{gnu} C compiler, @value{NGCC}.
9457 We pass the @option{-gdwarf-2} and @option{-g3} flags to ensure the
9458 compiler includes information about preprocessor macros in the debugging
9459 information.
9460
9461 @smallexample
9462 $ gcc -gdwarf-2 -g3 sample.c -o sample
9463 $
9464 @end smallexample
9465
9466 Now, we start @value{GDBN} on our sample program:
9467
9468 @smallexample
9469 $ gdb -nw sample
9470 GNU gdb 2002-05-06-cvs
9471 Copyright 2002 Free Software Foundation, Inc.
9472 GDB is free software, @dots{}
9473 (@value{GDBP})
9474 @end smallexample
9475
9476 We can expand macros and examine their definitions, even when the
9477 program is not running. @value{GDBN} uses the current listing position
9478 to decide which macro definitions are in scope:
9479
9480 @smallexample
9481 (@value{GDBP}) list main
9482 3
9483 4 #define M 42
9484 5 #define ADD(x) (M + x)
9485 6
9486 7 main ()
9487 8 @{
9488 9 #define N 28
9489 10 printf ("Hello, world!\n");
9490 11 #undef N
9491 12 printf ("We're so creative.\n");
9492 (@value{GDBP}) info macro ADD
9493 Defined at /home/jimb/gdb/macros/play/sample.c:5
9494 #define ADD(x) (M + x)
9495 (@value{GDBP}) info macro Q
9496 Defined at /home/jimb/gdb/macros/play/sample.h:1
9497 included at /home/jimb/gdb/macros/play/sample.c:2
9498 #define Q <
9499 (@value{GDBP}) macro expand ADD(1)
9500 expands to: (42 + 1)
9501 (@value{GDBP}) macro expand-once ADD(1)
9502 expands to: once (M + 1)
9503 (@value{GDBP})
9504 @end smallexample
9505
9506 In the example above, note that @code{macro expand-once} expands only
9507 the macro invocation explicit in the original text --- the invocation of
9508 @code{ADD} --- but does not expand the invocation of the macro @code{M},
9509 which was introduced by @code{ADD}.
9510
9511 Once the program is running, @value{GDBN} uses the macro definitions in
9512 force at the source line of the current stack frame:
9513
9514 @smallexample
9515 (@value{GDBP}) break main
9516 Breakpoint 1 at 0x8048370: file sample.c, line 10.
9517 (@value{GDBP}) run
9518 Starting program: /home/jimb/gdb/macros/play/sample
9519
9520 Breakpoint 1, main () at sample.c:10
9521 10 printf ("Hello, world!\n");
9522 (@value{GDBP})
9523 @end smallexample
9524
9525 At line 10, the definition of the macro @code{N} at line 9 is in force:
9526
9527 @smallexample
9528 (@value{GDBP}) info macro N
9529 Defined at /home/jimb/gdb/macros/play/sample.c:9
9530 #define N 28
9531 (@value{GDBP}) macro expand N Q M
9532 expands to: 28 < 42
9533 (@value{GDBP}) print N Q M
9534 $1 = 1
9535 (@value{GDBP})
9536 @end smallexample
9537
9538 As we step over directives that remove @code{N}'s definition, and then
9539 give it a new definition, @value{GDBN} finds the definition (or lack
9540 thereof) in force at each point:
9541
9542 @smallexample
9543 (@value{GDBP}) next
9544 Hello, world!
9545 12 printf ("We're so creative.\n");
9546 (@value{GDBP}) info macro N
9547 The symbol `N' has no definition as a C/C++ preprocessor macro
9548 at /home/jimb/gdb/macros/play/sample.c:12
9549 (@value{GDBP}) next
9550 We're so creative.
9551 14 printf ("Goodbye, world!\n");
9552 (@value{GDBP}) info macro N
9553 Defined at /home/jimb/gdb/macros/play/sample.c:13
9554 #define N 1729
9555 (@value{GDBP}) macro expand N Q M
9556 expands to: 1729 < 42
9557 (@value{GDBP}) print N Q M
9558 $2 = 0
9559 (@value{GDBP})
9560 @end smallexample
9561
9562 In addition to source files, macros can be defined on the compilation command
9563 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
9564 such a way, @value{GDBN} displays the location of their definition as line zero
9565 of the source file submitted to the compiler.
9566
9567 @smallexample
9568 (@value{GDBP}) info macro __STDC__
9569 Defined at /home/jimb/gdb/macros/play/sample.c:0
9570 -D__STDC__=1
9571 (@value{GDBP})
9572 @end smallexample
9573
9574
9575 @node Tracepoints
9576 @chapter Tracepoints
9577 @c This chapter is based on the documentation written by Michael
9578 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
9579
9580 @cindex tracepoints
9581 In some applications, it is not feasible for the debugger to interrupt
9582 the program's execution long enough for the developer to learn
9583 anything helpful about its behavior. If the program's correctness
9584 depends on its real-time behavior, delays introduced by a debugger
9585 might cause the program to change its behavior drastically, or perhaps
9586 fail, even when the code itself is correct. It is useful to be able
9587 to observe the program's behavior without interrupting it.
9588
9589 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
9590 specify locations in the program, called @dfn{tracepoints}, and
9591 arbitrary expressions to evaluate when those tracepoints are reached.
9592 Later, using the @code{tfind} command, you can examine the values
9593 those expressions had when the program hit the tracepoints. The
9594 expressions may also denote objects in memory---structures or arrays,
9595 for example---whose values @value{GDBN} should record; while visiting
9596 a particular tracepoint, you may inspect those objects as if they were
9597 in memory at that moment. However, because @value{GDBN} records these
9598 values without interacting with you, it can do so quickly and
9599 unobtrusively, hopefully not disturbing the program's behavior.
9600
9601 The tracepoint facility is currently available only for remote
9602 targets. @xref{Targets}. In addition, your remote target must know
9603 how to collect trace data. This functionality is implemented in the
9604 remote stub; however, none of the stubs distributed with @value{GDBN}
9605 support tracepoints as of this writing. The format of the remote
9606 packets used to implement tracepoints are described in @ref{Tracepoint
9607 Packets}.
9608
9609 It is also possible to get trace data from a file, in a manner reminiscent
9610 of corefiles; you specify the filename, and use @code{tfind} to search
9611 through the file. @xref{Trace Files}, for more details.
9612
9613 This chapter describes the tracepoint commands and features.
9614
9615 @menu
9616 * Set Tracepoints::
9617 * Analyze Collected Data::
9618 * Tracepoint Variables::
9619 * Trace Files::
9620 @end menu
9621
9622 @node Set Tracepoints
9623 @section Commands to Set Tracepoints
9624
9625 Before running such a @dfn{trace experiment}, an arbitrary number of
9626 tracepoints can be set. A tracepoint is actually a special type of
9627 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
9628 standard breakpoint commands. For instance, as with breakpoints,
9629 tracepoint numbers are successive integers starting from one, and many
9630 of the commands associated with tracepoints take the tracepoint number
9631 as their argument, to identify which tracepoint to work on.
9632
9633 For each tracepoint, you can specify, in advance, some arbitrary set
9634 of data that you want the target to collect in the trace buffer when
9635 it hits that tracepoint. The collected data can include registers,
9636 local variables, or global data. Later, you can use @value{GDBN}
9637 commands to examine the values these data had at the time the
9638 tracepoint was hit.
9639
9640 Tracepoints do not support every breakpoint feature. Ignore counts on
9641 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
9642 commands when they are hit. Tracepoints may not be thread-specific
9643 either.
9644
9645 @cindex fast tracepoints
9646 Some targets may support @dfn{fast tracepoints}, which are inserted in
9647 a different way (such as with a jump instead of a trap), that is
9648 faster but possibly restricted in where they may be installed.
9649
9650 @cindex static tracepoints
9651 @cindex markers, static tracepoints
9652 @cindex probing markers, static tracepoints
9653 Regular and fast tracepoints are dynamic tracing facilities, meaning
9654 that they can be used to insert tracepoints at (almost) any location
9655 in the target. Some targets may also support controlling @dfn{static
9656 tracepoints} from @value{GDBN}. With static tracing, a set of
9657 instrumentation points, also known as @dfn{markers}, are embedded in
9658 the target program, and can be activated or deactivated by name or
9659 address. These are usually placed at locations which facilitate
9660 investigating what the target is actually doing. @value{GDBN}'s
9661 support for static tracing includes being able to list instrumentation
9662 points, and attach them with @value{GDBN} defined high level
9663 tracepoints that expose the whole range of convenience of
9664 @value{GDBN}'s tracepoints support. Namelly, support for collecting
9665 registers values and values of global or local (to the instrumentation
9666 point) variables; tracepoint conditions and trace state variables.
9667 The act of installing a @value{GDBN} static tracepoint on an
9668 instrumentation point, or marker, is referred to as @dfn{probing} a
9669 static tracepoint marker.
9670
9671 @code{gdbserver} supports tracepoints on some target systems.
9672 @xref{Server,,Tracepoints support in @code{gdbserver}}.
9673
9674 This section describes commands to set tracepoints and associated
9675 conditions and actions.
9676
9677 @menu
9678 * Create and Delete Tracepoints::
9679 * Enable and Disable Tracepoints::
9680 * Tracepoint Passcounts::
9681 * Tracepoint Conditions::
9682 * Trace State Variables::
9683 * Tracepoint Actions::
9684 * Listing Tracepoints::
9685 * Listing Static Tracepoint Markers::
9686 * Starting and Stopping Trace Experiments::
9687 * Tracepoint Restrictions::
9688 @end menu
9689
9690 @node Create and Delete Tracepoints
9691 @subsection Create and Delete Tracepoints
9692
9693 @table @code
9694 @cindex set tracepoint
9695 @kindex trace
9696 @item trace @var{location}
9697 The @code{trace} command is very similar to the @code{break} command.
9698 Its argument @var{location} can be a source line, a function name, or
9699 an address in the target program. @xref{Specify Location}. The
9700 @code{trace} command defines a tracepoint, which is a point in the
9701 target program where the debugger will briefly stop, collect some
9702 data, and then allow the program to continue. Setting a tracepoint or
9703 changing its actions doesn't take effect until the next @code{tstart}
9704 command, and once a trace experiment is running, further changes will
9705 not have any effect until the next trace experiment starts.
9706
9707 Here are some examples of using the @code{trace} command:
9708
9709 @smallexample
9710 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
9711
9712 (@value{GDBP}) @b{trace +2} // 2 lines forward
9713
9714 (@value{GDBP}) @b{trace my_function} // first source line of function
9715
9716 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
9717
9718 (@value{GDBP}) @b{trace *0x2117c4} // an address
9719 @end smallexample
9720
9721 @noindent
9722 You can abbreviate @code{trace} as @code{tr}.
9723
9724 @item trace @var{location} if @var{cond}
9725 Set a tracepoint with condition @var{cond}; evaluate the expression
9726 @var{cond} each time the tracepoint is reached, and collect data only
9727 if the value is nonzero---that is, if @var{cond} evaluates as true.
9728 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
9729 information on tracepoint conditions.
9730
9731 @item ftrace @var{location} [ if @var{cond} ]
9732 @cindex set fast tracepoint
9733 @cindex fast tracepoints, setting
9734 @kindex ftrace
9735 The @code{ftrace} command sets a fast tracepoint. For targets that
9736 support them, fast tracepoints will use a more efficient but possibly
9737 less general technique to trigger data collection, such as a jump
9738 instruction instead of a trap, or some sort of hardware support. It
9739 may not be possible to create a fast tracepoint at the desired
9740 location, in which case the command will exit with an explanatory
9741 message.
9742
9743 @value{GDBN} handles arguments to @code{ftrace} exactly as for
9744 @code{trace}.
9745
9746 @item strace @var{location} [ if @var{cond} ]
9747 @cindex set static tracepoint
9748 @cindex static tracepoints, setting
9749 @cindex probe static tracepoint marker
9750 @kindex strace
9751 The @code{strace} command sets a static tracepoint. For targets that
9752 support it, setting a static tracepoint probes a static
9753 instrumentation point, or marker, found at @var{location}. It may not
9754 be possible to set a static tracepoint at the desired location, in
9755 which case the command will exit with an explanatory message.
9756
9757 @value{GDBN} handles arguments to @code{strace} exactly as for
9758 @code{trace}, with the addition that the user can also specify
9759 @code{-m @var{marker}} as @var{location}. This probes the marker
9760 identified by the @var{marker} string identifier. This identifier
9761 depends on the static tracepoint backend library your program is
9762 using. You can find all the marker identifiers in the @samp{ID} field
9763 of the @code{info static-tracepoint-markers} command output.
9764 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
9765 Markers}. For example, in the following small program using the UST
9766 tracing engine:
9767
9768 @smallexample
9769 main ()
9770 @{
9771 trace_mark(ust, bar33, "str %s", "FOOBAZ");
9772 @}
9773 @end smallexample
9774
9775 @noindent
9776 the marker id is composed of joining the first two arguments to the
9777 @code{trace_mark} call with a slash, which translates to:
9778
9779 @smallexample
9780 (@value{GDBP}) info static-tracepoint-markers
9781 Cnt Enb ID Address What
9782 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
9783 Data: "str %s"
9784 [etc...]
9785 @end smallexample
9786
9787 @noindent
9788 so you may probe the marker above with:
9789
9790 @smallexample
9791 (@value{GDBP}) strace -m ust/bar33
9792 @end smallexample
9793
9794 Static tracepoints accept an extra collect action --- @code{collect
9795 $_sdata}. This collects arbitrary user data passed in the probe point
9796 call to the tracing library. In the UST example above, you'll see
9797 that the third argument to @code{trace_mark} is a printf-like format
9798 string. The user data is then the result of running that formating
9799 string against the following arguments. Note that @code{info
9800 static-tracepoint-markers} command output lists that format string in
9801 the @samp{Data:} field.
9802
9803 You can inspect this data when analyzing the trace buffer, by printing
9804 the $_sdata variable like any other variable available to
9805 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
9806
9807 @vindex $tpnum
9808 @cindex last tracepoint number
9809 @cindex recent tracepoint number
9810 @cindex tracepoint number
9811 The convenience variable @code{$tpnum} records the tracepoint number
9812 of the most recently set tracepoint.
9813
9814 @kindex delete tracepoint
9815 @cindex tracepoint deletion
9816 @item delete tracepoint @r{[}@var{num}@r{]}
9817 Permanently delete one or more tracepoints. With no argument, the
9818 default is to delete all tracepoints. Note that the regular
9819 @code{delete} command can remove tracepoints also.
9820
9821 Examples:
9822
9823 @smallexample
9824 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
9825
9826 (@value{GDBP}) @b{delete trace} // remove all tracepoints
9827 @end smallexample
9828
9829 @noindent
9830 You can abbreviate this command as @code{del tr}.
9831 @end table
9832
9833 @node Enable and Disable Tracepoints
9834 @subsection Enable and Disable Tracepoints
9835
9836 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
9837
9838 @table @code
9839 @kindex disable tracepoint
9840 @item disable tracepoint @r{[}@var{num}@r{]}
9841 Disable tracepoint @var{num}, or all tracepoints if no argument
9842 @var{num} is given. A disabled tracepoint will have no effect during
9843 the next trace experiment, but it is not forgotten. You can re-enable
9844 a disabled tracepoint using the @code{enable tracepoint} command.
9845
9846 @kindex enable tracepoint
9847 @item enable tracepoint @r{[}@var{num}@r{]}
9848 Enable tracepoint @var{num}, or all tracepoints. The enabled
9849 tracepoints will become effective the next time a trace experiment is
9850 run.
9851 @end table
9852
9853 @node Tracepoint Passcounts
9854 @subsection Tracepoint Passcounts
9855
9856 @table @code
9857 @kindex passcount
9858 @cindex tracepoint pass count
9859 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
9860 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
9861 automatically stop a trace experiment. If a tracepoint's passcount is
9862 @var{n}, then the trace experiment will be automatically stopped on
9863 the @var{n}'th time that tracepoint is hit. If the tracepoint number
9864 @var{num} is not specified, the @code{passcount} command sets the
9865 passcount of the most recently defined tracepoint. If no passcount is
9866 given, the trace experiment will run until stopped explicitly by the
9867 user.
9868
9869 Examples:
9870
9871 @smallexample
9872 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
9873 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
9874
9875 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
9876 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
9877 (@value{GDBP}) @b{trace foo}
9878 (@value{GDBP}) @b{pass 3}
9879 (@value{GDBP}) @b{trace bar}
9880 (@value{GDBP}) @b{pass 2}
9881 (@value{GDBP}) @b{trace baz}
9882 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
9883 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
9884 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
9885 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
9886 @end smallexample
9887 @end table
9888
9889 @node Tracepoint Conditions
9890 @subsection Tracepoint Conditions
9891 @cindex conditional tracepoints
9892 @cindex tracepoint conditions
9893
9894 The simplest sort of tracepoint collects data every time your program
9895 reaches a specified place. You can also specify a @dfn{condition} for
9896 a tracepoint. A condition is just a Boolean expression in your
9897 programming language (@pxref{Expressions, ,Expressions}). A
9898 tracepoint with a condition evaluates the expression each time your
9899 program reaches it, and data collection happens only if the condition
9900 is true.
9901
9902 Tracepoint conditions can be specified when a tracepoint is set, by
9903 using @samp{if} in the arguments to the @code{trace} command.
9904 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
9905 also be set or changed at any time with the @code{condition} command,
9906 just as with breakpoints.
9907
9908 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
9909 the conditional expression itself. Instead, @value{GDBN} encodes the
9910 expression into an agent expression (@pxref{Agent Expressions}
9911 suitable for execution on the target, independently of @value{GDBN}.
9912 Global variables become raw memory locations, locals become stack
9913 accesses, and so forth.
9914
9915 For instance, suppose you have a function that is usually called
9916 frequently, but should not be called after an error has occurred. You
9917 could use the following tracepoint command to collect data about calls
9918 of that function that happen while the error code is propagating
9919 through the program; an unconditional tracepoint could end up
9920 collecting thousands of useless trace frames that you would have to
9921 search through.
9922
9923 @smallexample
9924 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
9925 @end smallexample
9926
9927 @node Trace State Variables
9928 @subsection Trace State Variables
9929 @cindex trace state variables
9930
9931 A @dfn{trace state variable} is a special type of variable that is
9932 created and managed by target-side code. The syntax is the same as
9933 that for GDB's convenience variables (a string prefixed with ``$''),
9934 but they are stored on the target. They must be created explicitly,
9935 using a @code{tvariable} command. They are always 64-bit signed
9936 integers.
9937
9938 Trace state variables are remembered by @value{GDBN}, and downloaded
9939 to the target along with tracepoint information when the trace
9940 experiment starts. There are no intrinsic limits on the number of
9941 trace state variables, beyond memory limitations of the target.
9942
9943 @cindex convenience variables, and trace state variables
9944 Although trace state variables are managed by the target, you can use
9945 them in print commands and expressions as if they were convenience
9946 variables; @value{GDBN} will get the current value from the target
9947 while the trace experiment is running. Trace state variables share
9948 the same namespace as other ``$'' variables, which means that you
9949 cannot have trace state variables with names like @code{$23} or
9950 @code{$pc}, nor can you have a trace state variable and a convenience
9951 variable with the same name.
9952
9953 @table @code
9954
9955 @item tvariable $@var{name} [ = @var{expression} ]
9956 @kindex tvariable
9957 The @code{tvariable} command creates a new trace state variable named
9958 @code{$@var{name}}, and optionally gives it an initial value of
9959 @var{expression}. @var{expression} is evaluated when this command is
9960 entered; the result will be converted to an integer if possible,
9961 otherwise @value{GDBN} will report an error. A subsequent
9962 @code{tvariable} command specifying the same name does not create a
9963 variable, but instead assigns the supplied initial value to the
9964 existing variable of that name, overwriting any previous initial
9965 value. The default initial value is 0.
9966
9967 @item info tvariables
9968 @kindex info tvariables
9969 List all the trace state variables along with their initial values.
9970 Their current values may also be displayed, if the trace experiment is
9971 currently running.
9972
9973 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
9974 @kindex delete tvariable
9975 Delete the given trace state variables, or all of them if no arguments
9976 are specified.
9977
9978 @end table
9979
9980 @node Tracepoint Actions
9981 @subsection Tracepoint Action Lists
9982
9983 @table @code
9984 @kindex actions
9985 @cindex tracepoint actions
9986 @item actions @r{[}@var{num}@r{]}
9987 This command will prompt for a list of actions to be taken when the
9988 tracepoint is hit. If the tracepoint number @var{num} is not
9989 specified, this command sets the actions for the one that was most
9990 recently defined (so that you can define a tracepoint and then say
9991 @code{actions} without bothering about its number). You specify the
9992 actions themselves on the following lines, one action at a time, and
9993 terminate the actions list with a line containing just @code{end}. So
9994 far, the only defined actions are @code{collect}, @code{teval}, and
9995 @code{while-stepping}.
9996
9997 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
9998 Commands, ,Breakpoint Command Lists}), except that only the defined
9999 actions are allowed; any other @value{GDBN} command is rejected.
10000
10001 @cindex remove actions from a tracepoint
10002 To remove all actions from a tracepoint, type @samp{actions @var{num}}
10003 and follow it immediately with @samp{end}.
10004
10005 @smallexample
10006 (@value{GDBP}) @b{collect @var{data}} // collect some data
10007
10008 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
10009
10010 (@value{GDBP}) @b{end} // signals the end of actions.
10011 @end smallexample
10012
10013 In the following example, the action list begins with @code{collect}
10014 commands indicating the things to be collected when the tracepoint is
10015 hit. Then, in order to single-step and collect additional data
10016 following the tracepoint, a @code{while-stepping} command is used,
10017 followed by the list of things to be collected after each step in a
10018 sequence of single steps. The @code{while-stepping} command is
10019 terminated by its own separate @code{end} command. Lastly, the action
10020 list is terminated by an @code{end} command.
10021
10022 @smallexample
10023 (@value{GDBP}) @b{trace foo}
10024 (@value{GDBP}) @b{actions}
10025 Enter actions for tracepoint 1, one per line:
10026 > collect bar,baz
10027 > collect $regs
10028 > while-stepping 12
10029 > collect $pc, arr[i]
10030 > end
10031 end
10032 @end smallexample
10033
10034 @kindex collect @r{(tracepoints)}
10035 @item collect @var{expr1}, @var{expr2}, @dots{}
10036 Collect values of the given expressions when the tracepoint is hit.
10037 This command accepts a comma-separated list of any valid expressions.
10038 In addition to global, static, or local variables, the following
10039 special arguments are supported:
10040
10041 @table @code
10042 @item $regs
10043 Collect all registers.
10044
10045 @item $args
10046 Collect all function arguments.
10047
10048 @item $locals
10049 Collect all local variables.
10050
10051 @item $_sdata
10052 @vindex $_sdata@r{, collect}
10053 Collect static tracepoint marker specific data. Only available for
10054 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
10055 Lists}. On the UST static tracepoints library backend, an
10056 instrumentation point resembles a @code{printf} function call. The
10057 tracing library is able to collect user specified data formatted to a
10058 character string using the format provided by the programmer that
10059 instrumented the program. Other backends have similar mechanisms.
10060 Here's an example of a UST marker call:
10061
10062 @smallexample
10063 const char master_name[] = "$your_name";
10064 trace_mark(channel1, marker1, "hello %s", master_name)
10065 @end smallexample
10066
10067 In this case, collecting @code{$_sdata} collects the string
10068 @samp{hello $yourname}. When analyzing the trace buffer, you can
10069 inspect @samp{$_sdata} like any other variable available to
10070 @value{GDBN}.
10071 @end table
10072
10073 You can give several consecutive @code{collect} commands, each one
10074 with a single argument, or one @code{collect} command with several
10075 arguments separated by commas; the effect is the same.
10076
10077 The command @code{info scope} (@pxref{Symbols, info scope}) is
10078 particularly useful for figuring out what data to collect.
10079
10080 @kindex teval @r{(tracepoints)}
10081 @item teval @var{expr1}, @var{expr2}, @dots{}
10082 Evaluate the given expressions when the tracepoint is hit. This
10083 command accepts a comma-separated list of expressions. The results
10084 are discarded, so this is mainly useful for assigning values to trace
10085 state variables (@pxref{Trace State Variables}) without adding those
10086 values to the trace buffer, as would be the case if the @code{collect}
10087 action were used.
10088
10089 @kindex while-stepping @r{(tracepoints)}
10090 @item while-stepping @var{n}
10091 Perform @var{n} single-step instruction traces after the tracepoint,
10092 collecting new data after each step. The @code{while-stepping}
10093 command is followed by the list of what to collect while stepping
10094 (followed by its own @code{end} command):
10095
10096 @smallexample
10097 > while-stepping 12
10098 > collect $regs, myglobal
10099 > end
10100 >
10101 @end smallexample
10102
10103 @noindent
10104 Note that @code{$pc} is not automatically collected by
10105 @code{while-stepping}; you need to explicitly collect that register if
10106 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
10107 @code{stepping}.
10108
10109 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
10110 @kindex set default-collect
10111 @cindex default collection action
10112 This variable is a list of expressions to collect at each tracepoint
10113 hit. It is effectively an additional @code{collect} action prepended
10114 to every tracepoint action list. The expressions are parsed
10115 individually for each tracepoint, so for instance a variable named
10116 @code{xyz} may be interpreted as a global for one tracepoint, and a
10117 local for another, as appropriate to the tracepoint's location.
10118
10119 @item show default-collect
10120 @kindex show default-collect
10121 Show the list of expressions that are collected by default at each
10122 tracepoint hit.
10123
10124 @end table
10125
10126 @node Listing Tracepoints
10127 @subsection Listing Tracepoints
10128
10129 @table @code
10130 @kindex info tracepoints
10131 @kindex info tp
10132 @cindex information about tracepoints
10133 @item info tracepoints @r{[}@var{num}@r{]}
10134 Display information about the tracepoint @var{num}. If you don't
10135 specify a tracepoint number, displays information about all the
10136 tracepoints defined so far. The format is similar to that used for
10137 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
10138 command, simply restricting itself to tracepoints.
10139
10140 A tracepoint's listing may include additional information specific to
10141 tracing:
10142
10143 @itemize @bullet
10144 @item
10145 its passcount as given by the @code{passcount @var{n}} command
10146 @end itemize
10147
10148 @smallexample
10149 (@value{GDBP}) @b{info trace}
10150 Num Type Disp Enb Address What
10151 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
10152 while-stepping 20
10153 collect globfoo, $regs
10154 end
10155 collect globfoo2
10156 end
10157 pass count 1200
10158 (@value{GDBP})
10159 @end smallexample
10160
10161 @noindent
10162 This command can be abbreviated @code{info tp}.
10163 @end table
10164
10165 @node Listing Static Tracepoint Markers
10166 @subsection Listing Static Tracepoint Markers
10167
10168 @table @code
10169 @kindex info static-tracepoint-markers
10170 @cindex information about static tracepoint markers
10171 @item info static-tracepoint-markers
10172 Display information about all static tracepoint markers defined in the
10173 program.
10174
10175 For each marker, the following columns are printed:
10176
10177 @table @emph
10178 @item Count
10179 An incrementing counter, output to help readability. This is not a
10180 stable identifier.
10181 @item ID
10182 The marker ID, as reported by the target.
10183 @item Enabled or Disabled
10184 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
10185 that are not enabled.
10186 @item Address
10187 Where the marker is in your program, as a memory address.
10188 @item What
10189 Where the marker is in the source for your program, as a file and line
10190 number. If the debug information included in the program does not
10191 allow @value{GDBN} to locate the source of the marker, this column
10192 will be left blank.
10193 @end table
10194
10195 @noindent
10196 In addition, the following information may be printed for each marker:
10197
10198 @table @emph
10199 @item Data
10200 User data passed to the tracing library by the marker call. In the
10201 UST backend, this is the format string passed as argument to the
10202 marker call.
10203 @item Static tracepoints probing the marker
10204 The list of static tracepoints attached to the marker.
10205 @end table
10206
10207 @smallexample
10208 (@value{GDBP}) info static-tracepoint-markers
10209 Cnt ID Enb Address What
10210 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
10211 Data: number1 %d number2 %d
10212 Probed by static tracepoints: #2
10213 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
10214 Data: str %s
10215 (@value{GDBP})
10216 @end smallexample
10217 @end table
10218
10219 @node Starting and Stopping Trace Experiments
10220 @subsection Starting and Stopping Trace Experiments
10221
10222 @table @code
10223 @kindex tstart
10224 @cindex start a new trace experiment
10225 @cindex collected data discarded
10226 @item tstart
10227 This command takes no arguments. It starts the trace experiment, and
10228 begins collecting data. This has the side effect of discarding all
10229 the data collected in the trace buffer during the previous trace
10230 experiment.
10231
10232 @kindex tstop
10233 @cindex stop a running trace experiment
10234 @item tstop
10235 This command takes no arguments. It ends the trace experiment, and
10236 stops collecting data.
10237
10238 @strong{Note}: a trace experiment and data collection may stop
10239 automatically if any tracepoint's passcount is reached
10240 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
10241
10242 @kindex tstatus
10243 @cindex status of trace data collection
10244 @cindex trace experiment, status of
10245 @item tstatus
10246 This command displays the status of the current trace data
10247 collection.
10248 @end table
10249
10250 Here is an example of the commands we described so far:
10251
10252 @smallexample
10253 (@value{GDBP}) @b{trace gdb_c_test}
10254 (@value{GDBP}) @b{actions}
10255 Enter actions for tracepoint #1, one per line.
10256 > collect $regs,$locals,$args
10257 > while-stepping 11
10258 > collect $regs
10259 > end
10260 > end
10261 (@value{GDBP}) @b{tstart}
10262 [time passes @dots{}]
10263 (@value{GDBP}) @b{tstop}
10264 @end smallexample
10265
10266 @cindex disconnected tracing
10267 You can choose to continue running the trace experiment even if
10268 @value{GDBN} disconnects from the target, voluntarily or
10269 involuntarily. For commands such as @code{detach}, the debugger will
10270 ask what you want to do with the trace. But for unexpected
10271 terminations (@value{GDBN} crash, network outage), it would be
10272 unfortunate to lose hard-won trace data, so the variable
10273 @code{disconnected-tracing} lets you decide whether the trace should
10274 continue running without @value{GDBN}.
10275
10276 @table @code
10277 @item set disconnected-tracing on
10278 @itemx set disconnected-tracing off
10279 @kindex set disconnected-tracing
10280 Choose whether a tracing run should continue to run if @value{GDBN}
10281 has disconnected from the target. Note that @code{detach} or
10282 @code{quit} will ask you directly what to do about a running trace no
10283 matter what this variable's setting, so the variable is mainly useful
10284 for handling unexpected situations, such as loss of the network.
10285
10286 @item show disconnected-tracing
10287 @kindex show disconnected-tracing
10288 Show the current choice for disconnected tracing.
10289
10290 @end table
10291
10292 When you reconnect to the target, the trace experiment may or may not
10293 still be running; it might have filled the trace buffer in the
10294 meantime, or stopped for one of the other reasons. If it is running,
10295 it will continue after reconnection.
10296
10297 Upon reconnection, the target will upload information about the
10298 tracepoints in effect. @value{GDBN} will then compare that
10299 information to the set of tracepoints currently defined, and attempt
10300 to match them up, allowing for the possibility that the numbers may
10301 have changed due to creation and deletion in the meantime. If one of
10302 the target's tracepoints does not match any in @value{GDBN}, the
10303 debugger will create a new tracepoint, so that you have a number with
10304 which to specify that tracepoint. This matching-up process is
10305 necessarily heuristic, and it may result in useless tracepoints being
10306 created; you may simply delete them if they are of no use.
10307
10308 @cindex circular trace buffer
10309 If your target agent supports a @dfn{circular trace buffer}, then you
10310 can run a trace experiment indefinitely without filling the trace
10311 buffer; when space runs out, the agent deletes already-collected trace
10312 frames, oldest first, until there is enough room to continue
10313 collecting. This is especially useful if your tracepoints are being
10314 hit too often, and your trace gets terminated prematurely because the
10315 buffer is full. To ask for a circular trace buffer, simply set
10316 @samp{circular_trace_buffer} to on. You can set this at any time,
10317 including during tracing; if the agent can do it, it will change
10318 buffer handling on the fly, otherwise it will not take effect until
10319 the next run.
10320
10321 @table @code
10322 @item set circular-trace-buffer on
10323 @itemx set circular-trace-buffer off
10324 @kindex set circular-trace-buffer
10325 Choose whether a tracing run should use a linear or circular buffer
10326 for trace data. A linear buffer will not lose any trace data, but may
10327 fill up prematurely, while a circular buffer will discard old trace
10328 data, but it will have always room for the latest tracepoint hits.
10329
10330 @item show circular-trace-buffer
10331 @kindex show circular-trace-buffer
10332 Show the current choice for the trace buffer. Note that this may not
10333 match the agent's current buffer handling, nor is it guaranteed to
10334 match the setting that might have been in effect during a past run,
10335 for instance if you are looking at frames from a trace file.
10336
10337 @end table
10338
10339 @node Tracepoint Restrictions
10340 @subsection Tracepoint Restrictions
10341
10342 @cindex tracepoint restrictions
10343 There are a number of restrictions on the use of tracepoints. As
10344 described above, tracepoint data gathering occurs on the target
10345 without interaction from @value{GDBN}. Thus the full capabilities of
10346 the debugger are not available during data gathering, and then at data
10347 examination time, you will be limited by only having what was
10348 collected. The following items describe some common problems, but it
10349 is not exhaustive, and you may run into additional difficulties not
10350 mentioned here.
10351
10352 @itemize @bullet
10353
10354 @item
10355 Tracepoint expressions are intended to gather objects (lvalues). Thus
10356 the full flexibility of GDB's expression evaluator is not available.
10357 You cannot call functions, cast objects to aggregate types, access
10358 convenience variables or modify values (except by assignment to trace
10359 state variables). Some language features may implicitly call
10360 functions (for instance Objective-C fields with accessors), and therefore
10361 cannot be collected either.
10362
10363 @item
10364 Collection of local variables, either individually or in bulk with
10365 @code{$locals} or @code{$args}, during @code{while-stepping} may
10366 behave erratically. The stepping action may enter a new scope (for
10367 instance by stepping into a function), or the location of the variable
10368 may change (for instance it is loaded into a register). The
10369 tracepoint data recorded uses the location information for the
10370 variables that is correct for the tracepoint location. When the
10371 tracepoint is created, it is not possible, in general, to determine
10372 where the steps of a @code{while-stepping} sequence will advance the
10373 program---particularly if a conditional branch is stepped.
10374
10375 @item
10376 Collection of an incompletely-initialized or partially-destroyed object
10377 may result in something that @value{GDBN} cannot display, or displays
10378 in a misleading way.
10379
10380 @item
10381 When @value{GDBN} displays a pointer to character it automatically
10382 dereferences the pointer to also display characters of the string
10383 being pointed to. However, collecting the pointer during tracing does
10384 not automatically collect the string. You need to explicitly
10385 dereference the pointer and provide size information if you want to
10386 collect not only the pointer, but the memory pointed to. For example,
10387 @code{*ptr@@50} can be used to collect the 50 element array pointed to
10388 by @code{ptr}.
10389
10390 @item
10391 It is not possible to collect a complete stack backtrace at a
10392 tracepoint. Instead, you may collect the registers and a few hundred
10393 bytes from the stack pointer with something like @code{*$esp@@300}
10394 (adjust to use the name of the actual stack pointer register on your
10395 target architecture, and the amount of stack you wish to capture).
10396 Then the @code{backtrace} command will show a partial backtrace when
10397 using a trace frame. The number of stack frames that can be examined
10398 depends on the sizes of the frames in the collected stack. Note that
10399 if you ask for a block so large that it goes past the bottom of the
10400 stack, the target agent may report an error trying to read from an
10401 invalid address.
10402
10403 @item
10404 If you do not collect registers at a tracepoint, @value{GDBN} can
10405 infer that the value of @code{$pc} must be the same as the address of
10406 the tracepoint and use that when you are looking at a trace frame
10407 for that tracepoint. However, this cannot work if the tracepoint has
10408 multiple locations (for instance if it was set in a function that was
10409 inlined), or if it has a @code{while-stepping} loop. In those cases
10410 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
10411 it to zero.
10412
10413 @end itemize
10414
10415 @node Analyze Collected Data
10416 @section Using the Collected Data
10417
10418 After the tracepoint experiment ends, you use @value{GDBN} commands
10419 for examining the trace data. The basic idea is that each tracepoint
10420 collects a trace @dfn{snapshot} every time it is hit and another
10421 snapshot every time it single-steps. All these snapshots are
10422 consecutively numbered from zero and go into a buffer, and you can
10423 examine them later. The way you examine them is to @dfn{focus} on a
10424 specific trace snapshot. When the remote stub is focused on a trace
10425 snapshot, it will respond to all @value{GDBN} requests for memory and
10426 registers by reading from the buffer which belongs to that snapshot,
10427 rather than from @emph{real} memory or registers of the program being
10428 debugged. This means that @strong{all} @value{GDBN} commands
10429 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
10430 behave as if we were currently debugging the program state as it was
10431 when the tracepoint occurred. Any requests for data that are not in
10432 the buffer will fail.
10433
10434 @menu
10435 * tfind:: How to select a trace snapshot
10436 * tdump:: How to display all data for a snapshot
10437 * save tracepoints:: How to save tracepoints for a future run
10438 @end menu
10439
10440 @node tfind
10441 @subsection @code{tfind @var{n}}
10442
10443 @kindex tfind
10444 @cindex select trace snapshot
10445 @cindex find trace snapshot
10446 The basic command for selecting a trace snapshot from the buffer is
10447 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
10448 counting from zero. If no argument @var{n} is given, the next
10449 snapshot is selected.
10450
10451 Here are the various forms of using the @code{tfind} command.
10452
10453 @table @code
10454 @item tfind start
10455 Find the first snapshot in the buffer. This is a synonym for
10456 @code{tfind 0} (since 0 is the number of the first snapshot).
10457
10458 @item tfind none
10459 Stop debugging trace snapshots, resume @emph{live} debugging.
10460
10461 @item tfind end
10462 Same as @samp{tfind none}.
10463
10464 @item tfind
10465 No argument means find the next trace snapshot.
10466
10467 @item tfind -
10468 Find the previous trace snapshot before the current one. This permits
10469 retracing earlier steps.
10470
10471 @item tfind tracepoint @var{num}
10472 Find the next snapshot associated with tracepoint @var{num}. Search
10473 proceeds forward from the last examined trace snapshot. If no
10474 argument @var{num} is given, it means find the next snapshot collected
10475 for the same tracepoint as the current snapshot.
10476
10477 @item tfind pc @var{addr}
10478 Find the next snapshot associated with the value @var{addr} of the
10479 program counter. Search proceeds forward from the last examined trace
10480 snapshot. If no argument @var{addr} is given, it means find the next
10481 snapshot with the same value of PC as the current snapshot.
10482
10483 @item tfind outside @var{addr1}, @var{addr2}
10484 Find the next snapshot whose PC is outside the given range of
10485 addresses (exclusive).
10486
10487 @item tfind range @var{addr1}, @var{addr2}
10488 Find the next snapshot whose PC is between @var{addr1} and
10489 @var{addr2} (inclusive).
10490
10491 @item tfind line @r{[}@var{file}:@r{]}@var{n}
10492 Find the next snapshot associated with the source line @var{n}. If
10493 the optional argument @var{file} is given, refer to line @var{n} in
10494 that source file. Search proceeds forward from the last examined
10495 trace snapshot. If no argument @var{n} is given, it means find the
10496 next line other than the one currently being examined; thus saying
10497 @code{tfind line} repeatedly can appear to have the same effect as
10498 stepping from line to line in a @emph{live} debugging session.
10499 @end table
10500
10501 The default arguments for the @code{tfind} commands are specifically
10502 designed to make it easy to scan through the trace buffer. For
10503 instance, @code{tfind} with no argument selects the next trace
10504 snapshot, and @code{tfind -} with no argument selects the previous
10505 trace snapshot. So, by giving one @code{tfind} command, and then
10506 simply hitting @key{RET} repeatedly you can examine all the trace
10507 snapshots in order. Or, by saying @code{tfind -} and then hitting
10508 @key{RET} repeatedly you can examine the snapshots in reverse order.
10509 The @code{tfind line} command with no argument selects the snapshot
10510 for the next source line executed. The @code{tfind pc} command with
10511 no argument selects the next snapshot with the same program counter
10512 (PC) as the current frame. The @code{tfind tracepoint} command with
10513 no argument selects the next trace snapshot collected by the same
10514 tracepoint as the current one.
10515
10516 In addition to letting you scan through the trace buffer manually,
10517 these commands make it easy to construct @value{GDBN} scripts that
10518 scan through the trace buffer and print out whatever collected data
10519 you are interested in. Thus, if we want to examine the PC, FP, and SP
10520 registers from each trace frame in the buffer, we can say this:
10521
10522 @smallexample
10523 (@value{GDBP}) @b{tfind start}
10524 (@value{GDBP}) @b{while ($trace_frame != -1)}
10525 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
10526 $trace_frame, $pc, $sp, $fp
10527 > tfind
10528 > end
10529
10530 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
10531 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
10532 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
10533 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
10534 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
10535 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
10536 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
10537 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
10538 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
10539 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
10540 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
10541 @end smallexample
10542
10543 Or, if we want to examine the variable @code{X} at each source line in
10544 the buffer:
10545
10546 @smallexample
10547 (@value{GDBP}) @b{tfind start}
10548 (@value{GDBP}) @b{while ($trace_frame != -1)}
10549 > printf "Frame %d, X == %d\n", $trace_frame, X
10550 > tfind line
10551 > end
10552
10553 Frame 0, X = 1
10554 Frame 7, X = 2
10555 Frame 13, X = 255
10556 @end smallexample
10557
10558 @node tdump
10559 @subsection @code{tdump}
10560 @kindex tdump
10561 @cindex dump all data collected at tracepoint
10562 @cindex tracepoint data, display
10563
10564 This command takes no arguments. It prints all the data collected at
10565 the current trace snapshot.
10566
10567 @smallexample
10568 (@value{GDBP}) @b{trace 444}
10569 (@value{GDBP}) @b{actions}
10570 Enter actions for tracepoint #2, one per line:
10571 > collect $regs, $locals, $args, gdb_long_test
10572 > end
10573
10574 (@value{GDBP}) @b{tstart}
10575
10576 (@value{GDBP}) @b{tfind line 444}
10577 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
10578 at gdb_test.c:444
10579 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
10580
10581 (@value{GDBP}) @b{tdump}
10582 Data collected at tracepoint 2, trace frame 1:
10583 d0 0xc4aa0085 -995491707
10584 d1 0x18 24
10585 d2 0x80 128
10586 d3 0x33 51
10587 d4 0x71aea3d 119204413
10588 d5 0x22 34
10589 d6 0xe0 224
10590 d7 0x380035 3670069
10591 a0 0x19e24a 1696330
10592 a1 0x3000668 50333288
10593 a2 0x100 256
10594 a3 0x322000 3284992
10595 a4 0x3000698 50333336
10596 a5 0x1ad3cc 1758156
10597 fp 0x30bf3c 0x30bf3c
10598 sp 0x30bf34 0x30bf34
10599 ps 0x0 0
10600 pc 0x20b2c8 0x20b2c8
10601 fpcontrol 0x0 0
10602 fpstatus 0x0 0
10603 fpiaddr 0x0 0
10604 p = 0x20e5b4 "gdb-test"
10605 p1 = (void *) 0x11
10606 p2 = (void *) 0x22
10607 p3 = (void *) 0x33
10608 p4 = (void *) 0x44
10609 p5 = (void *) 0x55
10610 p6 = (void *) 0x66
10611 gdb_long_test = 17 '\021'
10612
10613 (@value{GDBP})
10614 @end smallexample
10615
10616 @code{tdump} works by scanning the tracepoint's current collection
10617 actions and printing the value of each expression listed. So
10618 @code{tdump} can fail, if after a run, you change the tracepoint's
10619 actions to mention variables that were not collected during the run.
10620
10621 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
10622 uses the collected value of @code{$pc} to distinguish between trace
10623 frames that were collected at the tracepoint hit, and frames that were
10624 collected while stepping. This allows it to correctly choose whether
10625 to display the basic list of collections, or the collections from the
10626 body of the while-stepping loop. However, if @code{$pc} was not collected,
10627 then @code{tdump} will always attempt to dump using the basic collection
10628 list, and may fail if a while-stepping frame does not include all the
10629 same data that is collected at the tracepoint hit.
10630 @c This is getting pretty arcane, example would be good.
10631
10632 @node save tracepoints
10633 @subsection @code{save tracepoints @var{filename}}
10634 @kindex save tracepoints
10635 @kindex save-tracepoints
10636 @cindex save tracepoints for future sessions
10637
10638 This command saves all current tracepoint definitions together with
10639 their actions and passcounts, into a file @file{@var{filename}}
10640 suitable for use in a later debugging session. To read the saved
10641 tracepoint definitions, use the @code{source} command (@pxref{Command
10642 Files}). The @w{@code{save-tracepoints}} command is a deprecated
10643 alias for @w{@code{save tracepoints}}
10644
10645 @node Tracepoint Variables
10646 @section Convenience Variables for Tracepoints
10647 @cindex tracepoint variables
10648 @cindex convenience variables for tracepoints
10649
10650 @table @code
10651 @vindex $trace_frame
10652 @item (int) $trace_frame
10653 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
10654 snapshot is selected.
10655
10656 @vindex $tracepoint
10657 @item (int) $tracepoint
10658 The tracepoint for the current trace snapshot.
10659
10660 @vindex $trace_line
10661 @item (int) $trace_line
10662 The line number for the current trace snapshot.
10663
10664 @vindex $trace_file
10665 @item (char []) $trace_file
10666 The source file for the current trace snapshot.
10667
10668 @vindex $trace_func
10669 @item (char []) $trace_func
10670 The name of the function containing @code{$tracepoint}.
10671 @end table
10672
10673 Note: @code{$trace_file} is not suitable for use in @code{printf},
10674 use @code{output} instead.
10675
10676 Here's a simple example of using these convenience variables for
10677 stepping through all the trace snapshots and printing some of their
10678 data. Note that these are not the same as trace state variables,
10679 which are managed by the target.
10680
10681 @smallexample
10682 (@value{GDBP}) @b{tfind start}
10683
10684 (@value{GDBP}) @b{while $trace_frame != -1}
10685 > output $trace_file
10686 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
10687 > tfind
10688 > end
10689 @end smallexample
10690
10691 @node Trace Files
10692 @section Using Trace Files
10693 @cindex trace files
10694
10695 In some situations, the target running a trace experiment may no
10696 longer be available; perhaps it crashed, or the hardware was needed
10697 for a different activity. To handle these cases, you can arrange to
10698 dump the trace data into a file, and later use that file as a source
10699 of trace data, via the @code{target tfile} command.
10700
10701 @table @code
10702
10703 @kindex tsave
10704 @item tsave [ -r ] @var{filename}
10705 Save the trace data to @var{filename}. By default, this command
10706 assumes that @var{filename} refers to the host filesystem, so if
10707 necessary @value{GDBN} will copy raw trace data up from the target and
10708 then save it. If the target supports it, you can also supply the
10709 optional argument @code{-r} (``remote'') to direct the target to save
10710 the data directly into @var{filename} in its own filesystem, which may be
10711 more efficient if the trace buffer is very large. (Note, however, that
10712 @code{target tfile} can only read from files accessible to the host.)
10713
10714 @kindex target tfile
10715 @kindex tfile
10716 @item target tfile @var{filename}
10717 Use the file named @var{filename} as a source of trace data. Commands
10718 that examine data work as they do with a live target, but it is not
10719 possible to run any new trace experiments. @code{tstatus} will report
10720 the state of the trace run at the moment the data was saved, as well
10721 as the current trace frame you are examining. @var{filename} must be
10722 on a filesystem accessible to the host.
10723
10724 @end table
10725
10726 @node Overlays
10727 @chapter Debugging Programs That Use Overlays
10728 @cindex overlays
10729
10730 If your program is too large to fit completely in your target system's
10731 memory, you can sometimes use @dfn{overlays} to work around this
10732 problem. @value{GDBN} provides some support for debugging programs that
10733 use overlays.
10734
10735 @menu
10736 * How Overlays Work:: A general explanation of overlays.
10737 * Overlay Commands:: Managing overlays in @value{GDBN}.
10738 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
10739 mapped by asking the inferior.
10740 * Overlay Sample Program:: A sample program using overlays.
10741 @end menu
10742
10743 @node How Overlays Work
10744 @section How Overlays Work
10745 @cindex mapped overlays
10746 @cindex unmapped overlays
10747 @cindex load address, overlay's
10748 @cindex mapped address
10749 @cindex overlay area
10750
10751 Suppose you have a computer whose instruction address space is only 64
10752 kilobytes long, but which has much more memory which can be accessed by
10753 other means: special instructions, segment registers, or memory
10754 management hardware, for example. Suppose further that you want to
10755 adapt a program which is larger than 64 kilobytes to run on this system.
10756
10757 One solution is to identify modules of your program which are relatively
10758 independent, and need not call each other directly; call these modules
10759 @dfn{overlays}. Separate the overlays from the main program, and place
10760 their machine code in the larger memory. Place your main program in
10761 instruction memory, but leave at least enough space there to hold the
10762 largest overlay as well.
10763
10764 Now, to call a function located in an overlay, you must first copy that
10765 overlay's machine code from the large memory into the space set aside
10766 for it in the instruction memory, and then jump to its entry point
10767 there.
10768
10769 @c NB: In the below the mapped area's size is greater or equal to the
10770 @c size of all overlays. This is intentional to remind the developer
10771 @c that overlays don't necessarily need to be the same size.
10772
10773 @smallexample
10774 @group
10775 Data Instruction Larger
10776 Address Space Address Space Address Space
10777 +-----------+ +-----------+ +-----------+
10778 | | | | | |
10779 +-----------+ +-----------+ +-----------+<-- overlay 1
10780 | program | | main | .----| overlay 1 | load address
10781 | variables | | program | | +-----------+
10782 | and heap | | | | | |
10783 +-----------+ | | | +-----------+<-- overlay 2
10784 | | +-----------+ | | | load address
10785 +-----------+ | | | .-| overlay 2 |
10786 | | | | | |
10787 mapped --->+-----------+ | | +-----------+
10788 address | | | | | |
10789 | overlay | <-' | | |
10790 | area | <---' +-----------+<-- overlay 3
10791 | | <---. | | load address
10792 +-----------+ `--| overlay 3 |
10793 | | | |
10794 +-----------+ | |
10795 +-----------+
10796 | |
10797 +-----------+
10798
10799 @anchor{A code overlay}A code overlay
10800 @end group
10801 @end smallexample
10802
10803 The diagram (@pxref{A code overlay}) shows a system with separate data
10804 and instruction address spaces. To map an overlay, the program copies
10805 its code from the larger address space to the instruction address space.
10806 Since the overlays shown here all use the same mapped address, only one
10807 may be mapped at a time. For a system with a single address space for
10808 data and instructions, the diagram would be similar, except that the
10809 program variables and heap would share an address space with the main
10810 program and the overlay area.
10811
10812 An overlay loaded into instruction memory and ready for use is called a
10813 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
10814 instruction memory. An overlay not present (or only partially present)
10815 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
10816 is its address in the larger memory. The mapped address is also called
10817 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
10818 called the @dfn{load memory address}, or @dfn{LMA}.
10819
10820 Unfortunately, overlays are not a completely transparent way to adapt a
10821 program to limited instruction memory. They introduce a new set of
10822 global constraints you must keep in mind as you design your program:
10823
10824 @itemize @bullet
10825
10826 @item
10827 Before calling or returning to a function in an overlay, your program
10828 must make sure that overlay is actually mapped. Otherwise, the call or
10829 return will transfer control to the right address, but in the wrong
10830 overlay, and your program will probably crash.
10831
10832 @item
10833 If the process of mapping an overlay is expensive on your system, you
10834 will need to choose your overlays carefully to minimize their effect on
10835 your program's performance.
10836
10837 @item
10838 The executable file you load onto your system must contain each
10839 overlay's instructions, appearing at the overlay's load address, not its
10840 mapped address. However, each overlay's instructions must be relocated
10841 and its symbols defined as if the overlay were at its mapped address.
10842 You can use GNU linker scripts to specify different load and relocation
10843 addresses for pieces of your program; see @ref{Overlay Description,,,
10844 ld.info, Using ld: the GNU linker}.
10845
10846 @item
10847 The procedure for loading executable files onto your system must be able
10848 to load their contents into the larger address space as well as the
10849 instruction and data spaces.
10850
10851 @end itemize
10852
10853 The overlay system described above is rather simple, and could be
10854 improved in many ways:
10855
10856 @itemize @bullet
10857
10858 @item
10859 If your system has suitable bank switch registers or memory management
10860 hardware, you could use those facilities to make an overlay's load area
10861 contents simply appear at their mapped address in instruction space.
10862 This would probably be faster than copying the overlay to its mapped
10863 area in the usual way.
10864
10865 @item
10866 If your overlays are small enough, you could set aside more than one
10867 overlay area, and have more than one overlay mapped at a time.
10868
10869 @item
10870 You can use overlays to manage data, as well as instructions. In
10871 general, data overlays are even less transparent to your design than
10872 code overlays: whereas code overlays only require care when you call or
10873 return to functions, data overlays require care every time you access
10874 the data. Also, if you change the contents of a data overlay, you
10875 must copy its contents back out to its load address before you can copy a
10876 different data overlay into the same mapped area.
10877
10878 @end itemize
10879
10880
10881 @node Overlay Commands
10882 @section Overlay Commands
10883
10884 To use @value{GDBN}'s overlay support, each overlay in your program must
10885 correspond to a separate section of the executable file. The section's
10886 virtual memory address and load memory address must be the overlay's
10887 mapped and load addresses. Identifying overlays with sections allows
10888 @value{GDBN} to determine the appropriate address of a function or
10889 variable, depending on whether the overlay is mapped or not.
10890
10891 @value{GDBN}'s overlay commands all start with the word @code{overlay};
10892 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
10893
10894 @table @code
10895 @item overlay off
10896 @kindex overlay
10897 Disable @value{GDBN}'s overlay support. When overlay support is
10898 disabled, @value{GDBN} assumes that all functions and variables are
10899 always present at their mapped addresses. By default, @value{GDBN}'s
10900 overlay support is disabled.
10901
10902 @item overlay manual
10903 @cindex manual overlay debugging
10904 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
10905 relies on you to tell it which overlays are mapped, and which are not,
10906 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
10907 commands described below.
10908
10909 @item overlay map-overlay @var{overlay}
10910 @itemx overlay map @var{overlay}
10911 @cindex map an overlay
10912 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
10913 be the name of the object file section containing the overlay. When an
10914 overlay is mapped, @value{GDBN} assumes it can find the overlay's
10915 functions and variables at their mapped addresses. @value{GDBN} assumes
10916 that any other overlays whose mapped ranges overlap that of
10917 @var{overlay} are now unmapped.
10918
10919 @item overlay unmap-overlay @var{overlay}
10920 @itemx overlay unmap @var{overlay}
10921 @cindex unmap an overlay
10922 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
10923 must be the name of the object file section containing the overlay.
10924 When an overlay is unmapped, @value{GDBN} assumes it can find the
10925 overlay's functions and variables at their load addresses.
10926
10927 @item overlay auto
10928 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
10929 consults a data structure the overlay manager maintains in the inferior
10930 to see which overlays are mapped. For details, see @ref{Automatic
10931 Overlay Debugging}.
10932
10933 @item overlay load-target
10934 @itemx overlay load
10935 @cindex reloading the overlay table
10936 Re-read the overlay table from the inferior. Normally, @value{GDBN}
10937 re-reads the table @value{GDBN} automatically each time the inferior
10938 stops, so this command should only be necessary if you have changed the
10939 overlay mapping yourself using @value{GDBN}. This command is only
10940 useful when using automatic overlay debugging.
10941
10942 @item overlay list-overlays
10943 @itemx overlay list
10944 @cindex listing mapped overlays
10945 Display a list of the overlays currently mapped, along with their mapped
10946 addresses, load addresses, and sizes.
10947
10948 @end table
10949
10950 Normally, when @value{GDBN} prints a code address, it includes the name
10951 of the function the address falls in:
10952
10953 @smallexample
10954 (@value{GDBP}) print main
10955 $3 = @{int ()@} 0x11a0 <main>
10956 @end smallexample
10957 @noindent
10958 When overlay debugging is enabled, @value{GDBN} recognizes code in
10959 unmapped overlays, and prints the names of unmapped functions with
10960 asterisks around them. For example, if @code{foo} is a function in an
10961 unmapped overlay, @value{GDBN} prints it this way:
10962
10963 @smallexample
10964 (@value{GDBP}) overlay list
10965 No sections are mapped.
10966 (@value{GDBP}) print foo
10967 $5 = @{int (int)@} 0x100000 <*foo*>
10968 @end smallexample
10969 @noindent
10970 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
10971 name normally:
10972
10973 @smallexample
10974 (@value{GDBP}) overlay list
10975 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
10976 mapped at 0x1016 - 0x104a
10977 (@value{GDBP}) print foo
10978 $6 = @{int (int)@} 0x1016 <foo>
10979 @end smallexample
10980
10981 When overlay debugging is enabled, @value{GDBN} can find the correct
10982 address for functions and variables in an overlay, whether or not the
10983 overlay is mapped. This allows most @value{GDBN} commands, like
10984 @code{break} and @code{disassemble}, to work normally, even on unmapped
10985 code. However, @value{GDBN}'s breakpoint support has some limitations:
10986
10987 @itemize @bullet
10988 @item
10989 @cindex breakpoints in overlays
10990 @cindex overlays, setting breakpoints in
10991 You can set breakpoints in functions in unmapped overlays, as long as
10992 @value{GDBN} can write to the overlay at its load address.
10993 @item
10994 @value{GDBN} can not set hardware or simulator-based breakpoints in
10995 unmapped overlays. However, if you set a breakpoint at the end of your
10996 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
10997 you are using manual overlay management), @value{GDBN} will re-set its
10998 breakpoints properly.
10999 @end itemize
11000
11001
11002 @node Automatic Overlay Debugging
11003 @section Automatic Overlay Debugging
11004 @cindex automatic overlay debugging
11005
11006 @value{GDBN} can automatically track which overlays are mapped and which
11007 are not, given some simple co-operation from the overlay manager in the
11008 inferior. If you enable automatic overlay debugging with the
11009 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
11010 looks in the inferior's memory for certain variables describing the
11011 current state of the overlays.
11012
11013 Here are the variables your overlay manager must define to support
11014 @value{GDBN}'s automatic overlay debugging:
11015
11016 @table @asis
11017
11018 @item @code{_ovly_table}:
11019 This variable must be an array of the following structures:
11020
11021 @smallexample
11022 struct
11023 @{
11024 /* The overlay's mapped address. */
11025 unsigned long vma;
11026
11027 /* The size of the overlay, in bytes. */
11028 unsigned long size;
11029
11030 /* The overlay's load address. */
11031 unsigned long lma;
11032
11033 /* Non-zero if the overlay is currently mapped;
11034 zero otherwise. */
11035 unsigned long mapped;
11036 @}
11037 @end smallexample
11038
11039 @item @code{_novlys}:
11040 This variable must be a four-byte signed integer, holding the total
11041 number of elements in @code{_ovly_table}.
11042
11043 @end table
11044
11045 To decide whether a particular overlay is mapped or not, @value{GDBN}
11046 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
11047 @code{lma} members equal the VMA and LMA of the overlay's section in the
11048 executable file. When @value{GDBN} finds a matching entry, it consults
11049 the entry's @code{mapped} member to determine whether the overlay is
11050 currently mapped.
11051
11052 In addition, your overlay manager may define a function called
11053 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
11054 will silently set a breakpoint there. If the overlay manager then
11055 calls this function whenever it has changed the overlay table, this
11056 will enable @value{GDBN} to accurately keep track of which overlays
11057 are in program memory, and update any breakpoints that may be set
11058 in overlays. This will allow breakpoints to work even if the
11059 overlays are kept in ROM or other non-writable memory while they
11060 are not being executed.
11061
11062 @node Overlay Sample Program
11063 @section Overlay Sample Program
11064 @cindex overlay example program
11065
11066 When linking a program which uses overlays, you must place the overlays
11067 at their load addresses, while relocating them to run at their mapped
11068 addresses. To do this, you must write a linker script (@pxref{Overlay
11069 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
11070 since linker scripts are specific to a particular host system, target
11071 architecture, and target memory layout, this manual cannot provide
11072 portable sample code demonstrating @value{GDBN}'s overlay support.
11073
11074 However, the @value{GDBN} source distribution does contain an overlaid
11075 program, with linker scripts for a few systems, as part of its test
11076 suite. The program consists of the following files from
11077 @file{gdb/testsuite/gdb.base}:
11078
11079 @table @file
11080 @item overlays.c
11081 The main program file.
11082 @item ovlymgr.c
11083 A simple overlay manager, used by @file{overlays.c}.
11084 @item foo.c
11085 @itemx bar.c
11086 @itemx baz.c
11087 @itemx grbx.c
11088 Overlay modules, loaded and used by @file{overlays.c}.
11089 @item d10v.ld
11090 @itemx m32r.ld
11091 Linker scripts for linking the test program on the @code{d10v-elf}
11092 and @code{m32r-elf} targets.
11093 @end table
11094
11095 You can build the test program using the @code{d10v-elf} GCC
11096 cross-compiler like this:
11097
11098 @smallexample
11099 $ d10v-elf-gcc -g -c overlays.c
11100 $ d10v-elf-gcc -g -c ovlymgr.c
11101 $ d10v-elf-gcc -g -c foo.c
11102 $ d10v-elf-gcc -g -c bar.c
11103 $ d10v-elf-gcc -g -c baz.c
11104 $ d10v-elf-gcc -g -c grbx.c
11105 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
11106 baz.o grbx.o -Wl,-Td10v.ld -o overlays
11107 @end smallexample
11108
11109 The build process is identical for any other architecture, except that
11110 you must substitute the appropriate compiler and linker script for the
11111 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
11112
11113
11114 @node Languages
11115 @chapter Using @value{GDBN} with Different Languages
11116 @cindex languages
11117
11118 Although programming languages generally have common aspects, they are
11119 rarely expressed in the same manner. For instance, in ANSI C,
11120 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
11121 Modula-2, it is accomplished by @code{p^}. Values can also be
11122 represented (and displayed) differently. Hex numbers in C appear as
11123 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
11124
11125 @cindex working language
11126 Language-specific information is built into @value{GDBN} for some languages,
11127 allowing you to express operations like the above in your program's
11128 native language, and allowing @value{GDBN} to output values in a manner
11129 consistent with the syntax of your program's native language. The
11130 language you use to build expressions is called the @dfn{working
11131 language}.
11132
11133 @menu
11134 * Setting:: Switching between source languages
11135 * Show:: Displaying the language
11136 * Checks:: Type and range checks
11137 * Supported Languages:: Supported languages
11138 * Unsupported Languages:: Unsupported languages
11139 @end menu
11140
11141 @node Setting
11142 @section Switching Between Source Languages
11143
11144 There are two ways to control the working language---either have @value{GDBN}
11145 set it automatically, or select it manually yourself. You can use the
11146 @code{set language} command for either purpose. On startup, @value{GDBN}
11147 defaults to setting the language automatically. The working language is
11148 used to determine how expressions you type are interpreted, how values
11149 are printed, etc.
11150
11151 In addition to the working language, every source file that
11152 @value{GDBN} knows about has its own working language. For some object
11153 file formats, the compiler might indicate which language a particular
11154 source file is in. However, most of the time @value{GDBN} infers the
11155 language from the name of the file. The language of a source file
11156 controls whether C@t{++} names are demangled---this way @code{backtrace} can
11157 show each frame appropriately for its own language. There is no way to
11158 set the language of a source file from within @value{GDBN}, but you can
11159 set the language associated with a filename extension. @xref{Show, ,
11160 Displaying the Language}.
11161
11162 This is most commonly a problem when you use a program, such
11163 as @code{cfront} or @code{f2c}, that generates C but is written in
11164 another language. In that case, make the
11165 program use @code{#line} directives in its C output; that way
11166 @value{GDBN} will know the correct language of the source code of the original
11167 program, and will display that source code, not the generated C code.
11168
11169 @menu
11170 * Filenames:: Filename extensions and languages.
11171 * Manually:: Setting the working language manually
11172 * Automatically:: Having @value{GDBN} infer the source language
11173 @end menu
11174
11175 @node Filenames
11176 @subsection List of Filename Extensions and Languages
11177
11178 If a source file name ends in one of the following extensions, then
11179 @value{GDBN} infers that its language is the one indicated.
11180
11181 @table @file
11182 @item .ada
11183 @itemx .ads
11184 @itemx .adb
11185 @itemx .a
11186 Ada source file.
11187
11188 @item .c
11189 C source file
11190
11191 @item .C
11192 @itemx .cc
11193 @itemx .cp
11194 @itemx .cpp
11195 @itemx .cxx
11196 @itemx .c++
11197 C@t{++} source file
11198
11199 @item .d
11200 D source file
11201
11202 @item .m
11203 Objective-C source file
11204
11205 @item .f
11206 @itemx .F
11207 Fortran source file
11208
11209 @item .mod
11210 Modula-2 source file
11211
11212 @item .s
11213 @itemx .S
11214 Assembler source file. This actually behaves almost like C, but
11215 @value{GDBN} does not skip over function prologues when stepping.
11216 @end table
11217
11218 In addition, you may set the language associated with a filename
11219 extension. @xref{Show, , Displaying the Language}.
11220
11221 @node Manually
11222 @subsection Setting the Working Language
11223
11224 If you allow @value{GDBN} to set the language automatically,
11225 expressions are interpreted the same way in your debugging session and
11226 your program.
11227
11228 @kindex set language
11229 If you wish, you may set the language manually. To do this, issue the
11230 command @samp{set language @var{lang}}, where @var{lang} is the name of
11231 a language, such as
11232 @code{c} or @code{modula-2}.
11233 For a list of the supported languages, type @samp{set language}.
11234
11235 Setting the language manually prevents @value{GDBN} from updating the working
11236 language automatically. This can lead to confusion if you try
11237 to debug a program when the working language is not the same as the
11238 source language, when an expression is acceptable to both
11239 languages---but means different things. For instance, if the current
11240 source file were written in C, and @value{GDBN} was parsing Modula-2, a
11241 command such as:
11242
11243 @smallexample
11244 print a = b + c
11245 @end smallexample
11246
11247 @noindent
11248 might not have the effect you intended. In C, this means to add
11249 @code{b} and @code{c} and place the result in @code{a}. The result
11250 printed would be the value of @code{a}. In Modula-2, this means to compare
11251 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
11252
11253 @node Automatically
11254 @subsection Having @value{GDBN} Infer the Source Language
11255
11256 To have @value{GDBN} set the working language automatically, use
11257 @samp{set language local} or @samp{set language auto}. @value{GDBN}
11258 then infers the working language. That is, when your program stops in a
11259 frame (usually by encountering a breakpoint), @value{GDBN} sets the
11260 working language to the language recorded for the function in that
11261 frame. If the language for a frame is unknown (that is, if the function
11262 or block corresponding to the frame was defined in a source file that
11263 does not have a recognized extension), the current working language is
11264 not changed, and @value{GDBN} issues a warning.
11265
11266 This may not seem necessary for most programs, which are written
11267 entirely in one source language. However, program modules and libraries
11268 written in one source language can be used by a main program written in
11269 a different source language. Using @samp{set language auto} in this
11270 case frees you from having to set the working language manually.
11271
11272 @node Show
11273 @section Displaying the Language
11274
11275 The following commands help you find out which language is the
11276 working language, and also what language source files were written in.
11277
11278 @table @code
11279 @item show language
11280 @kindex show language
11281 Display the current working language. This is the
11282 language you can use with commands such as @code{print} to
11283 build and compute expressions that may involve variables in your program.
11284
11285 @item info frame
11286 @kindex info frame@r{, show the source language}
11287 Display the source language for this frame. This language becomes the
11288 working language if you use an identifier from this frame.
11289 @xref{Frame Info, ,Information about a Frame}, to identify the other
11290 information listed here.
11291
11292 @item info source
11293 @kindex info source@r{, show the source language}
11294 Display the source language of this source file.
11295 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
11296 information listed here.
11297 @end table
11298
11299 In unusual circumstances, you may have source files with extensions
11300 not in the standard list. You can then set the extension associated
11301 with a language explicitly:
11302
11303 @table @code
11304 @item set extension-language @var{ext} @var{language}
11305 @kindex set extension-language
11306 Tell @value{GDBN} that source files with extension @var{ext} are to be
11307 assumed as written in the source language @var{language}.
11308
11309 @item info extensions
11310 @kindex info extensions
11311 List all the filename extensions and the associated languages.
11312 @end table
11313
11314 @node Checks
11315 @section Type and Range Checking
11316
11317 @quotation
11318 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
11319 checking are included, but they do not yet have any effect. This
11320 section documents the intended facilities.
11321 @end quotation
11322 @c FIXME remove warning when type/range code added
11323
11324 Some languages are designed to guard you against making seemingly common
11325 errors through a series of compile- and run-time checks. These include
11326 checking the type of arguments to functions and operators, and making
11327 sure mathematical overflows are caught at run time. Checks such as
11328 these help to ensure a program's correctness once it has been compiled
11329 by eliminating type mismatches, and providing active checks for range
11330 errors when your program is running.
11331
11332 @value{GDBN} can check for conditions like the above if you wish.
11333 Although @value{GDBN} does not check the statements in your program,
11334 it can check expressions entered directly into @value{GDBN} for
11335 evaluation via the @code{print} command, for example. As with the
11336 working language, @value{GDBN} can also decide whether or not to check
11337 automatically based on your program's source language.
11338 @xref{Supported Languages, ,Supported Languages}, for the default
11339 settings of supported languages.
11340
11341 @menu
11342 * Type Checking:: An overview of type checking
11343 * Range Checking:: An overview of range checking
11344 @end menu
11345
11346 @cindex type checking
11347 @cindex checks, type
11348 @node Type Checking
11349 @subsection An Overview of Type Checking
11350
11351 Some languages, such as Modula-2, are strongly typed, meaning that the
11352 arguments to operators and functions have to be of the correct type,
11353 otherwise an error occurs. These checks prevent type mismatch
11354 errors from ever causing any run-time problems. For example,
11355
11356 @smallexample
11357 1 + 2 @result{} 3
11358 @exdent but
11359 @error{} 1 + 2.3
11360 @end smallexample
11361
11362 The second example fails because the @code{CARDINAL} 1 is not
11363 type-compatible with the @code{REAL} 2.3.
11364
11365 For the expressions you use in @value{GDBN} commands, you can tell the
11366 @value{GDBN} type checker to skip checking;
11367 to treat any mismatches as errors and abandon the expression;
11368 or to only issue warnings when type mismatches occur,
11369 but evaluate the expression anyway. When you choose the last of
11370 these, @value{GDBN} evaluates expressions like the second example above, but
11371 also issues a warning.
11372
11373 Even if you turn type checking off, there may be other reasons
11374 related to type that prevent @value{GDBN} from evaluating an expression.
11375 For instance, @value{GDBN} does not know how to add an @code{int} and
11376 a @code{struct foo}. These particular type errors have nothing to do
11377 with the language in use, and usually arise from expressions, such as
11378 the one described above, which make little sense to evaluate anyway.
11379
11380 Each language defines to what degree it is strict about type. For
11381 instance, both Modula-2 and C require the arguments to arithmetical
11382 operators to be numbers. In C, enumerated types and pointers can be
11383 represented as numbers, so that they are valid arguments to mathematical
11384 operators. @xref{Supported Languages, ,Supported Languages}, for further
11385 details on specific languages.
11386
11387 @value{GDBN} provides some additional commands for controlling the type checker:
11388
11389 @kindex set check type
11390 @kindex show check type
11391 @table @code
11392 @item set check type auto
11393 Set type checking on or off based on the current working language.
11394 @xref{Supported Languages, ,Supported Languages}, for the default settings for
11395 each language.
11396
11397 @item set check type on
11398 @itemx set check type off
11399 Set type checking on or off, overriding the default setting for the
11400 current working language. Issue a warning if the setting does not
11401 match the language default. If any type mismatches occur in
11402 evaluating an expression while type checking is on, @value{GDBN} prints a
11403 message and aborts evaluation of the expression.
11404
11405 @item set check type warn
11406 Cause the type checker to issue warnings, but to always attempt to
11407 evaluate the expression. Evaluating the expression may still
11408 be impossible for other reasons. For example, @value{GDBN} cannot add
11409 numbers and structures.
11410
11411 @item show type
11412 Show the current setting of the type checker, and whether or not @value{GDBN}
11413 is setting it automatically.
11414 @end table
11415
11416 @cindex range checking
11417 @cindex checks, range
11418 @node Range Checking
11419 @subsection An Overview of Range Checking
11420
11421 In some languages (such as Modula-2), it is an error to exceed the
11422 bounds of a type; this is enforced with run-time checks. Such range
11423 checking is meant to ensure program correctness by making sure
11424 computations do not overflow, or indices on an array element access do
11425 not exceed the bounds of the array.
11426
11427 For expressions you use in @value{GDBN} commands, you can tell
11428 @value{GDBN} to treat range errors in one of three ways: ignore them,
11429 always treat them as errors and abandon the expression, or issue
11430 warnings but evaluate the expression anyway.
11431
11432 A range error can result from numerical overflow, from exceeding an
11433 array index bound, or when you type a constant that is not a member
11434 of any type. Some languages, however, do not treat overflows as an
11435 error. In many implementations of C, mathematical overflow causes the
11436 result to ``wrap around'' to lower values---for example, if @var{m} is
11437 the largest integer value, and @var{s} is the smallest, then
11438
11439 @smallexample
11440 @var{m} + 1 @result{} @var{s}
11441 @end smallexample
11442
11443 This, too, is specific to individual languages, and in some cases
11444 specific to individual compilers or machines. @xref{Supported Languages, ,
11445 Supported Languages}, for further details on specific languages.
11446
11447 @value{GDBN} provides some additional commands for controlling the range checker:
11448
11449 @kindex set check range
11450 @kindex show check range
11451 @table @code
11452 @item set check range auto
11453 Set range checking on or off based on the current working language.
11454 @xref{Supported Languages, ,Supported Languages}, for the default settings for
11455 each language.
11456
11457 @item set check range on
11458 @itemx set check range off
11459 Set range checking on or off, overriding the default setting for the
11460 current working language. A warning is issued if the setting does not
11461 match the language default. If a range error occurs and range checking is on,
11462 then a message is printed and evaluation of the expression is aborted.
11463
11464 @item set check range warn
11465 Output messages when the @value{GDBN} range checker detects a range error,
11466 but attempt to evaluate the expression anyway. Evaluating the
11467 expression may still be impossible for other reasons, such as accessing
11468 memory that the process does not own (a typical example from many Unix
11469 systems).
11470
11471 @item show range
11472 Show the current setting of the range checker, and whether or not it is
11473 being set automatically by @value{GDBN}.
11474 @end table
11475
11476 @node Supported Languages
11477 @section Supported Languages
11478
11479 @value{GDBN} supports C, C@t{++}, D, Objective-C, Fortran, Java, Pascal,
11480 assembly, Modula-2, and Ada.
11481 @c This is false ...
11482 Some @value{GDBN} features may be used in expressions regardless of the
11483 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
11484 and the @samp{@{type@}addr} construct (@pxref{Expressions,
11485 ,Expressions}) can be used with the constructs of any supported
11486 language.
11487
11488 The following sections detail to what degree each source language is
11489 supported by @value{GDBN}. These sections are not meant to be language
11490 tutorials or references, but serve only as a reference guide to what the
11491 @value{GDBN} expression parser accepts, and what input and output
11492 formats should look like for different languages. There are many good
11493 books written on each of these languages; please look to these for a
11494 language reference or tutorial.
11495
11496 @menu
11497 * C:: C and C@t{++}
11498 * D:: D
11499 * Objective-C:: Objective-C
11500 * Fortran:: Fortran
11501 * Pascal:: Pascal
11502 * Modula-2:: Modula-2
11503 * Ada:: Ada
11504 @end menu
11505
11506 @node C
11507 @subsection C and C@t{++}
11508
11509 @cindex C and C@t{++}
11510 @cindex expressions in C or C@t{++}
11511
11512 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
11513 to both languages. Whenever this is the case, we discuss those languages
11514 together.
11515
11516 @cindex C@t{++}
11517 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
11518 @cindex @sc{gnu} C@t{++}
11519 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
11520 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
11521 effectively, you must compile your C@t{++} programs with a supported
11522 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
11523 compiler (@code{aCC}).
11524
11525 For best results when using @sc{gnu} C@t{++}, use the DWARF 2 debugging
11526 format; if it doesn't work on your system, try the stabs+ debugging
11527 format. You can select those formats explicitly with the @code{g++}
11528 command-line options @option{-gdwarf-2} and @option{-gstabs+}.
11529 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
11530 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}.
11531
11532 @menu
11533 * C Operators:: C and C@t{++} operators
11534 * C Constants:: C and C@t{++} constants
11535 * C Plus Plus Expressions:: C@t{++} expressions
11536 * C Defaults:: Default settings for C and C@t{++}
11537 * C Checks:: C and C@t{++} type and range checks
11538 * Debugging C:: @value{GDBN} and C
11539 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
11540 * Decimal Floating Point:: Numbers in Decimal Floating Point format
11541 @end menu
11542
11543 @node C Operators
11544 @subsubsection C and C@t{++} Operators
11545
11546 @cindex C and C@t{++} operators
11547
11548 Operators must be defined on values of specific types. For instance,
11549 @code{+} is defined on numbers, but not on structures. Operators are
11550 often defined on groups of types.
11551
11552 For the purposes of C and C@t{++}, the following definitions hold:
11553
11554 @itemize @bullet
11555
11556 @item
11557 @emph{Integral types} include @code{int} with any of its storage-class
11558 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
11559
11560 @item
11561 @emph{Floating-point types} include @code{float}, @code{double}, and
11562 @code{long double} (if supported by the target platform).
11563
11564 @item
11565 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
11566
11567 @item
11568 @emph{Scalar types} include all of the above.
11569
11570 @end itemize
11571
11572 @noindent
11573 The following operators are supported. They are listed here
11574 in order of increasing precedence:
11575
11576 @table @code
11577 @item ,
11578 The comma or sequencing operator. Expressions in a comma-separated list
11579 are evaluated from left to right, with the result of the entire
11580 expression being the last expression evaluated.
11581
11582 @item =
11583 Assignment. The value of an assignment expression is the value
11584 assigned. Defined on scalar types.
11585
11586 @item @var{op}=
11587 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
11588 and translated to @w{@code{@var{a} = @var{a op b}}}.
11589 @w{@code{@var{op}=}} and @code{=} have the same precedence.
11590 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
11591 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
11592
11593 @item ?:
11594 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
11595 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
11596 integral type.
11597
11598 @item ||
11599 Logical @sc{or}. Defined on integral types.
11600
11601 @item &&
11602 Logical @sc{and}. Defined on integral types.
11603
11604 @item |
11605 Bitwise @sc{or}. Defined on integral types.
11606
11607 @item ^
11608 Bitwise exclusive-@sc{or}. Defined on integral types.
11609
11610 @item &
11611 Bitwise @sc{and}. Defined on integral types.
11612
11613 @item ==@r{, }!=
11614 Equality and inequality. Defined on scalar types. The value of these
11615 expressions is 0 for false and non-zero for true.
11616
11617 @item <@r{, }>@r{, }<=@r{, }>=
11618 Less than, greater than, less than or equal, greater than or equal.
11619 Defined on scalar types. The value of these expressions is 0 for false
11620 and non-zero for true.
11621
11622 @item <<@r{, }>>
11623 left shift, and right shift. Defined on integral types.
11624
11625 @item @@
11626 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
11627
11628 @item +@r{, }-
11629 Addition and subtraction. Defined on integral types, floating-point types and
11630 pointer types.
11631
11632 @item *@r{, }/@r{, }%
11633 Multiplication, division, and modulus. Multiplication and division are
11634 defined on integral and floating-point types. Modulus is defined on
11635 integral types.
11636
11637 @item ++@r{, }--
11638 Increment and decrement. When appearing before a variable, the
11639 operation is performed before the variable is used in an expression;
11640 when appearing after it, the variable's value is used before the
11641 operation takes place.
11642
11643 @item *
11644 Pointer dereferencing. Defined on pointer types. Same precedence as
11645 @code{++}.
11646
11647 @item &
11648 Address operator. Defined on variables. Same precedence as @code{++}.
11649
11650 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
11651 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
11652 to examine the address
11653 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
11654 stored.
11655
11656 @item -
11657 Negative. Defined on integral and floating-point types. Same
11658 precedence as @code{++}.
11659
11660 @item !
11661 Logical negation. Defined on integral types. Same precedence as
11662 @code{++}.
11663
11664 @item ~
11665 Bitwise complement operator. Defined on integral types. Same precedence as
11666 @code{++}.
11667
11668
11669 @item .@r{, }->
11670 Structure member, and pointer-to-structure member. For convenience,
11671 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
11672 pointer based on the stored type information.
11673 Defined on @code{struct} and @code{union} data.
11674
11675 @item .*@r{, }->*
11676 Dereferences of pointers to members.
11677
11678 @item []
11679 Array indexing. @code{@var{a}[@var{i}]} is defined as
11680 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
11681
11682 @item ()
11683 Function parameter list. Same precedence as @code{->}.
11684
11685 @item ::
11686 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
11687 and @code{class} types.
11688
11689 @item ::
11690 Doubled colons also represent the @value{GDBN} scope operator
11691 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
11692 above.
11693 @end table
11694
11695 If an operator is redefined in the user code, @value{GDBN} usually
11696 attempts to invoke the redefined version instead of using the operator's
11697 predefined meaning.
11698
11699 @node C Constants
11700 @subsubsection C and C@t{++} Constants
11701
11702 @cindex C and C@t{++} constants
11703
11704 @value{GDBN} allows you to express the constants of C and C@t{++} in the
11705 following ways:
11706
11707 @itemize @bullet
11708 @item
11709 Integer constants are a sequence of digits. Octal constants are
11710 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
11711 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
11712 @samp{l}, specifying that the constant should be treated as a
11713 @code{long} value.
11714
11715 @item
11716 Floating point constants are a sequence of digits, followed by a decimal
11717 point, followed by a sequence of digits, and optionally followed by an
11718 exponent. An exponent is of the form:
11719 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
11720 sequence of digits. The @samp{+} is optional for positive exponents.
11721 A floating-point constant may also end with a letter @samp{f} or
11722 @samp{F}, specifying that the constant should be treated as being of
11723 the @code{float} (as opposed to the default @code{double}) type; or with
11724 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
11725 constant.
11726
11727 @item
11728 Enumerated constants consist of enumerated identifiers, or their
11729 integral equivalents.
11730
11731 @item
11732 Character constants are a single character surrounded by single quotes
11733 (@code{'}), or a number---the ordinal value of the corresponding character
11734 (usually its @sc{ascii} value). Within quotes, the single character may
11735 be represented by a letter or by @dfn{escape sequences}, which are of
11736 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
11737 of the character's ordinal value; or of the form @samp{\@var{x}}, where
11738 @samp{@var{x}} is a predefined special character---for example,
11739 @samp{\n} for newline.
11740
11741 @item
11742 String constants are a sequence of character constants surrounded by
11743 double quotes (@code{"}). Any valid character constant (as described
11744 above) may appear. Double quotes within the string must be preceded by
11745 a backslash, so for instance @samp{"a\"b'c"} is a string of five
11746 characters.
11747
11748 @item
11749 Pointer constants are an integral value. You can also write pointers
11750 to constants using the C operator @samp{&}.
11751
11752 @item
11753 Array constants are comma-separated lists surrounded by braces @samp{@{}
11754 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
11755 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
11756 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
11757 @end itemize
11758
11759 @node C Plus Plus Expressions
11760 @subsubsection C@t{++} Expressions
11761
11762 @cindex expressions in C@t{++}
11763 @value{GDBN} expression handling can interpret most C@t{++} expressions.
11764
11765 @cindex debugging C@t{++} programs
11766 @cindex C@t{++} compilers
11767 @cindex debug formats and C@t{++}
11768 @cindex @value{NGCC} and C@t{++}
11769 @quotation
11770 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
11771 proper compiler and the proper debug format. Currently, @value{GDBN}
11772 works best when debugging C@t{++} code that is compiled with
11773 @value{NGCC} 2.95.3 or with @value{NGCC} 3.1 or newer, using the options
11774 @option{-gdwarf-2} or @option{-gstabs+}. DWARF 2 is preferred over
11775 stabs+. Most configurations of @value{NGCC} emit either DWARF 2 or
11776 stabs+ as their default debug format, so you usually don't need to
11777 specify a debug format explicitly. Other compilers and/or debug formats
11778 are likely to work badly or not at all when using @value{GDBN} to debug
11779 C@t{++} code.
11780 @end quotation
11781
11782 @enumerate
11783
11784 @cindex member functions
11785 @item
11786 Member function calls are allowed; you can use expressions like
11787
11788 @smallexample
11789 count = aml->GetOriginal(x, y)
11790 @end smallexample
11791
11792 @vindex this@r{, inside C@t{++} member functions}
11793 @cindex namespace in C@t{++}
11794 @item
11795 While a member function is active (in the selected stack frame), your
11796 expressions have the same namespace available as the member function;
11797 that is, @value{GDBN} allows implicit references to the class instance
11798 pointer @code{this} following the same rules as C@t{++}.
11799
11800 @cindex call overloaded functions
11801 @cindex overloaded functions, calling
11802 @cindex type conversions in C@t{++}
11803 @item
11804 You can call overloaded functions; @value{GDBN} resolves the function
11805 call to the right definition, with some restrictions. @value{GDBN} does not
11806 perform overload resolution involving user-defined type conversions,
11807 calls to constructors, or instantiations of templates that do not exist
11808 in the program. It also cannot handle ellipsis argument lists or
11809 default arguments.
11810
11811 It does perform integral conversions and promotions, floating-point
11812 promotions, arithmetic conversions, pointer conversions, conversions of
11813 class objects to base classes, and standard conversions such as those of
11814 functions or arrays to pointers; it requires an exact match on the
11815 number of function arguments.
11816
11817 Overload resolution is always performed, unless you have specified
11818 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
11819 ,@value{GDBN} Features for C@t{++}}.
11820
11821 You must specify @code{set overload-resolution off} in order to use an
11822 explicit function signature to call an overloaded function, as in
11823 @smallexample
11824 p 'foo(char,int)'('x', 13)
11825 @end smallexample
11826
11827 The @value{GDBN} command-completion facility can simplify this;
11828 see @ref{Completion, ,Command Completion}.
11829
11830 @cindex reference declarations
11831 @item
11832 @value{GDBN} understands variables declared as C@t{++} references; you can use
11833 them in expressions just as you do in C@t{++} source---they are automatically
11834 dereferenced.
11835
11836 In the parameter list shown when @value{GDBN} displays a frame, the values of
11837 reference variables are not displayed (unlike other variables); this
11838 avoids clutter, since references are often used for large structures.
11839 The @emph{address} of a reference variable is always shown, unless
11840 you have specified @samp{set print address off}.
11841
11842 @item
11843 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
11844 expressions can use it just as expressions in your program do. Since
11845 one scope may be defined in another, you can use @code{::} repeatedly if
11846 necessary, for example in an expression like
11847 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
11848 resolving name scope by reference to source files, in both C and C@t{++}
11849 debugging (@pxref{Variables, ,Program Variables}).
11850 @end enumerate
11851
11852 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
11853 calling virtual functions correctly, printing out virtual bases of
11854 objects, calling functions in a base subobject, casting objects, and
11855 invoking user-defined operators.
11856
11857 @node C Defaults
11858 @subsubsection C and C@t{++} Defaults
11859
11860 @cindex C and C@t{++} defaults
11861
11862 If you allow @value{GDBN} to set type and range checking automatically, they
11863 both default to @code{off} whenever the working language changes to
11864 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
11865 selects the working language.
11866
11867 If you allow @value{GDBN} to set the language automatically, it
11868 recognizes source files whose names end with @file{.c}, @file{.C}, or
11869 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
11870 these files, it sets the working language to C or C@t{++}.
11871 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
11872 for further details.
11873
11874 @c Type checking is (a) primarily motivated by Modula-2, and (b)
11875 @c unimplemented. If (b) changes, it might make sense to let this node
11876 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
11877
11878 @node C Checks
11879 @subsubsection C and C@t{++} Type and Range Checks
11880
11881 @cindex C and C@t{++} checks
11882
11883 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
11884 is not used. However, if you turn type checking on, @value{GDBN}
11885 considers two variables type equivalent if:
11886
11887 @itemize @bullet
11888 @item
11889 The two variables are structured and have the same structure, union, or
11890 enumerated tag.
11891
11892 @item
11893 The two variables have the same type name, or types that have been
11894 declared equivalent through @code{typedef}.
11895
11896 @ignore
11897 @c leaving this out because neither J Gilmore nor R Pesch understand it.
11898 @c FIXME--beers?
11899 @item
11900 The two @code{struct}, @code{union}, or @code{enum} variables are
11901 declared in the same declaration. (Note: this may not be true for all C
11902 compilers.)
11903 @end ignore
11904 @end itemize
11905
11906 Range checking, if turned on, is done on mathematical operations. Array
11907 indices are not checked, since they are often used to index a pointer
11908 that is not itself an array.
11909
11910 @node Debugging C
11911 @subsubsection @value{GDBN} and C
11912
11913 The @code{set print union} and @code{show print union} commands apply to
11914 the @code{union} type. When set to @samp{on}, any @code{union} that is
11915 inside a @code{struct} or @code{class} is also printed. Otherwise, it
11916 appears as @samp{@{...@}}.
11917
11918 The @code{@@} operator aids in the debugging of dynamic arrays, formed
11919 with pointers and a memory allocation function. @xref{Expressions,
11920 ,Expressions}.
11921
11922 @node Debugging C Plus Plus
11923 @subsubsection @value{GDBN} Features for C@t{++}
11924
11925 @cindex commands for C@t{++}
11926
11927 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
11928 designed specifically for use with C@t{++}. Here is a summary:
11929
11930 @table @code
11931 @cindex break in overloaded functions
11932 @item @r{breakpoint menus}
11933 When you want a breakpoint in a function whose name is overloaded,
11934 @value{GDBN} has the capability to display a menu of possible breakpoint
11935 locations to help you specify which function definition you want.
11936 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
11937
11938 @cindex overloading in C@t{++}
11939 @item rbreak @var{regex}
11940 Setting breakpoints using regular expressions is helpful for setting
11941 breakpoints on overloaded functions that are not members of any special
11942 classes.
11943 @xref{Set Breaks, ,Setting Breakpoints}.
11944
11945 @cindex C@t{++} exception handling
11946 @item catch throw
11947 @itemx catch catch
11948 Debug C@t{++} exception handling using these commands. @xref{Set
11949 Catchpoints, , Setting Catchpoints}.
11950
11951 @cindex inheritance
11952 @item ptype @var{typename}
11953 Print inheritance relationships as well as other information for type
11954 @var{typename}.
11955 @xref{Symbols, ,Examining the Symbol Table}.
11956
11957 @cindex C@t{++} symbol display
11958 @item set print demangle
11959 @itemx show print demangle
11960 @itemx set print asm-demangle
11961 @itemx show print asm-demangle
11962 Control whether C@t{++} symbols display in their source form, both when
11963 displaying code as C@t{++} source and when displaying disassemblies.
11964 @xref{Print Settings, ,Print Settings}.
11965
11966 @item set print object
11967 @itemx show print object
11968 Choose whether to print derived (actual) or declared types of objects.
11969 @xref{Print Settings, ,Print Settings}.
11970
11971 @item set print vtbl
11972 @itemx show print vtbl
11973 Control the format for printing virtual function tables.
11974 @xref{Print Settings, ,Print Settings}.
11975 (The @code{vtbl} commands do not work on programs compiled with the HP
11976 ANSI C@t{++} compiler (@code{aCC}).)
11977
11978 @kindex set overload-resolution
11979 @cindex overloaded functions, overload resolution
11980 @item set overload-resolution on
11981 Enable overload resolution for C@t{++} expression evaluation. The default
11982 is on. For overloaded functions, @value{GDBN} evaluates the arguments
11983 and searches for a function whose signature matches the argument types,
11984 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
11985 Expressions, ,C@t{++} Expressions}, for details).
11986 If it cannot find a match, it emits a message.
11987
11988 @item set overload-resolution off
11989 Disable overload resolution for C@t{++} expression evaluation. For
11990 overloaded functions that are not class member functions, @value{GDBN}
11991 chooses the first function of the specified name that it finds in the
11992 symbol table, whether or not its arguments are of the correct type. For
11993 overloaded functions that are class member functions, @value{GDBN}
11994 searches for a function whose signature @emph{exactly} matches the
11995 argument types.
11996
11997 @kindex show overload-resolution
11998 @item show overload-resolution
11999 Show the current setting of overload resolution.
12000
12001 @item @r{Overloaded symbol names}
12002 You can specify a particular definition of an overloaded symbol, using
12003 the same notation that is used to declare such symbols in C@t{++}: type
12004 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
12005 also use the @value{GDBN} command-line word completion facilities to list the
12006 available choices, or to finish the type list for you.
12007 @xref{Completion,, Command Completion}, for details on how to do this.
12008 @end table
12009
12010 @node Decimal Floating Point
12011 @subsubsection Decimal Floating Point format
12012 @cindex decimal floating point format
12013
12014 @value{GDBN} can examine, set and perform computations with numbers in
12015 decimal floating point format, which in the C language correspond to the
12016 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
12017 specified by the extension to support decimal floating-point arithmetic.
12018
12019 There are two encodings in use, depending on the architecture: BID (Binary
12020 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
12021 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
12022 target.
12023
12024 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
12025 to manipulate decimal floating point numbers, it is not possible to convert
12026 (using a cast, for example) integers wider than 32-bit to decimal float.
12027
12028 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
12029 point computations, error checking in decimal float operations ignores
12030 underflow, overflow and divide by zero exceptions.
12031
12032 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
12033 to inspect @code{_Decimal128} values stored in floating point registers.
12034 See @ref{PowerPC,,PowerPC} for more details.
12035
12036 @node D
12037 @subsection D
12038
12039 @cindex D
12040 @value{GDBN} can be used to debug programs written in D and compiled with
12041 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
12042 specific feature --- dynamic arrays.
12043
12044 @node Objective-C
12045 @subsection Objective-C
12046
12047 @cindex Objective-C
12048 This section provides information about some commands and command
12049 options that are useful for debugging Objective-C code. See also
12050 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
12051 few more commands specific to Objective-C support.
12052
12053 @menu
12054 * Method Names in Commands::
12055 * The Print Command with Objective-C::
12056 @end menu
12057
12058 @node Method Names in Commands
12059 @subsubsection Method Names in Commands
12060
12061 The following commands have been extended to accept Objective-C method
12062 names as line specifications:
12063
12064 @kindex clear@r{, and Objective-C}
12065 @kindex break@r{, and Objective-C}
12066 @kindex info line@r{, and Objective-C}
12067 @kindex jump@r{, and Objective-C}
12068 @kindex list@r{, and Objective-C}
12069 @itemize
12070 @item @code{clear}
12071 @item @code{break}
12072 @item @code{info line}
12073 @item @code{jump}
12074 @item @code{list}
12075 @end itemize
12076
12077 A fully qualified Objective-C method name is specified as
12078
12079 @smallexample
12080 -[@var{Class} @var{methodName}]
12081 @end smallexample
12082
12083 where the minus sign is used to indicate an instance method and a
12084 plus sign (not shown) is used to indicate a class method. The class
12085 name @var{Class} and method name @var{methodName} are enclosed in
12086 brackets, similar to the way messages are specified in Objective-C
12087 source code. For example, to set a breakpoint at the @code{create}
12088 instance method of class @code{Fruit} in the program currently being
12089 debugged, enter:
12090
12091 @smallexample
12092 break -[Fruit create]
12093 @end smallexample
12094
12095 To list ten program lines around the @code{initialize} class method,
12096 enter:
12097
12098 @smallexample
12099 list +[NSText initialize]
12100 @end smallexample
12101
12102 In the current version of @value{GDBN}, the plus or minus sign is
12103 required. In future versions of @value{GDBN}, the plus or minus
12104 sign will be optional, but you can use it to narrow the search. It
12105 is also possible to specify just a method name:
12106
12107 @smallexample
12108 break create
12109 @end smallexample
12110
12111 You must specify the complete method name, including any colons. If
12112 your program's source files contain more than one @code{create} method,
12113 you'll be presented with a numbered list of classes that implement that
12114 method. Indicate your choice by number, or type @samp{0} to exit if
12115 none apply.
12116
12117 As another example, to clear a breakpoint established at the
12118 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
12119
12120 @smallexample
12121 clear -[NSWindow makeKeyAndOrderFront:]
12122 @end smallexample
12123
12124 @node The Print Command with Objective-C
12125 @subsubsection The Print Command With Objective-C
12126 @cindex Objective-C, print objects
12127 @kindex print-object
12128 @kindex po @r{(@code{print-object})}
12129
12130 The print command has also been extended to accept methods. For example:
12131
12132 @smallexample
12133 print -[@var{object} hash]
12134 @end smallexample
12135
12136 @cindex print an Objective-C object description
12137 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
12138 @noindent
12139 will tell @value{GDBN} to send the @code{hash} message to @var{object}
12140 and print the result. Also, an additional command has been added,
12141 @code{print-object} or @code{po} for short, which is meant to print
12142 the description of an object. However, this command may only work
12143 with certain Objective-C libraries that have a particular hook
12144 function, @code{_NSPrintForDebugger}, defined.
12145
12146 @node Fortran
12147 @subsection Fortran
12148 @cindex Fortran-specific support in @value{GDBN}
12149
12150 @value{GDBN} can be used to debug programs written in Fortran, but it
12151 currently supports only the features of Fortran 77 language.
12152
12153 @cindex trailing underscore, in Fortran symbols
12154 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
12155 among them) append an underscore to the names of variables and
12156 functions. When you debug programs compiled by those compilers, you
12157 will need to refer to variables and functions with a trailing
12158 underscore.
12159
12160 @menu
12161 * Fortran Operators:: Fortran operators and expressions
12162 * Fortran Defaults:: Default settings for Fortran
12163 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
12164 @end menu
12165
12166 @node Fortran Operators
12167 @subsubsection Fortran Operators and Expressions
12168
12169 @cindex Fortran operators and expressions
12170
12171 Operators must be defined on values of specific types. For instance,
12172 @code{+} is defined on numbers, but not on characters or other non-
12173 arithmetic types. Operators are often defined on groups of types.
12174
12175 @table @code
12176 @item **
12177 The exponentiation operator. It raises the first operand to the power
12178 of the second one.
12179
12180 @item :
12181 The range operator. Normally used in the form of array(low:high) to
12182 represent a section of array.
12183
12184 @item %
12185 The access component operator. Normally used to access elements in derived
12186 types. Also suitable for unions. As unions aren't part of regular Fortran,
12187 this can only happen when accessing a register that uses a gdbarch-defined
12188 union type.
12189 @end table
12190
12191 @node Fortran Defaults
12192 @subsubsection Fortran Defaults
12193
12194 @cindex Fortran Defaults
12195
12196 Fortran symbols are usually case-insensitive, so @value{GDBN} by
12197 default uses case-insensitive matches for Fortran symbols. You can
12198 change that with the @samp{set case-insensitive} command, see
12199 @ref{Symbols}, for the details.
12200
12201 @node Special Fortran Commands
12202 @subsubsection Special Fortran Commands
12203
12204 @cindex Special Fortran commands
12205
12206 @value{GDBN} has some commands to support Fortran-specific features,
12207 such as displaying common blocks.
12208
12209 @table @code
12210 @cindex @code{COMMON} blocks, Fortran
12211 @kindex info common
12212 @item info common @r{[}@var{common-name}@r{]}
12213 This command prints the values contained in the Fortran @code{COMMON}
12214 block whose name is @var{common-name}. With no argument, the names of
12215 all @code{COMMON} blocks visible at the current program location are
12216 printed.
12217 @end table
12218
12219 @node Pascal
12220 @subsection Pascal
12221
12222 @cindex Pascal support in @value{GDBN}, limitations
12223 Debugging Pascal programs which use sets, subranges, file variables, or
12224 nested functions does not currently work. @value{GDBN} does not support
12225 entering expressions, printing values, or similar features using Pascal
12226 syntax.
12227
12228 The Pascal-specific command @code{set print pascal_static-members}
12229 controls whether static members of Pascal objects are displayed.
12230 @xref{Print Settings, pascal_static-members}.
12231
12232 @node Modula-2
12233 @subsection Modula-2
12234
12235 @cindex Modula-2, @value{GDBN} support
12236
12237 The extensions made to @value{GDBN} to support Modula-2 only support
12238 output from the @sc{gnu} Modula-2 compiler (which is currently being
12239 developed). Other Modula-2 compilers are not currently supported, and
12240 attempting to debug executables produced by them is most likely
12241 to give an error as @value{GDBN} reads in the executable's symbol
12242 table.
12243
12244 @cindex expressions in Modula-2
12245 @menu
12246 * M2 Operators:: Built-in operators
12247 * Built-In Func/Proc:: Built-in functions and procedures
12248 * M2 Constants:: Modula-2 constants
12249 * M2 Types:: Modula-2 types
12250 * M2 Defaults:: Default settings for Modula-2
12251 * Deviations:: Deviations from standard Modula-2
12252 * M2 Checks:: Modula-2 type and range checks
12253 * M2 Scope:: The scope operators @code{::} and @code{.}
12254 * GDB/M2:: @value{GDBN} and Modula-2
12255 @end menu
12256
12257 @node M2 Operators
12258 @subsubsection Operators
12259 @cindex Modula-2 operators
12260
12261 Operators must be defined on values of specific types. For instance,
12262 @code{+} is defined on numbers, but not on structures. Operators are
12263 often defined on groups of types. For the purposes of Modula-2, the
12264 following definitions hold:
12265
12266 @itemize @bullet
12267
12268 @item
12269 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
12270 their subranges.
12271
12272 @item
12273 @emph{Character types} consist of @code{CHAR} and its subranges.
12274
12275 @item
12276 @emph{Floating-point types} consist of @code{REAL}.
12277
12278 @item
12279 @emph{Pointer types} consist of anything declared as @code{POINTER TO
12280 @var{type}}.
12281
12282 @item
12283 @emph{Scalar types} consist of all of the above.
12284
12285 @item
12286 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
12287
12288 @item
12289 @emph{Boolean types} consist of @code{BOOLEAN}.
12290 @end itemize
12291
12292 @noindent
12293 The following operators are supported, and appear in order of
12294 increasing precedence:
12295
12296 @table @code
12297 @item ,
12298 Function argument or array index separator.
12299
12300 @item :=
12301 Assignment. The value of @var{var} @code{:=} @var{value} is
12302 @var{value}.
12303
12304 @item <@r{, }>
12305 Less than, greater than on integral, floating-point, or enumerated
12306 types.
12307
12308 @item <=@r{, }>=
12309 Less than or equal to, greater than or equal to
12310 on integral, floating-point and enumerated types, or set inclusion on
12311 set types. Same precedence as @code{<}.
12312
12313 @item =@r{, }<>@r{, }#
12314 Equality and two ways of expressing inequality, valid on scalar types.
12315 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
12316 available for inequality, since @code{#} conflicts with the script
12317 comment character.
12318
12319 @item IN
12320 Set membership. Defined on set types and the types of their members.
12321 Same precedence as @code{<}.
12322
12323 @item OR
12324 Boolean disjunction. Defined on boolean types.
12325
12326 @item AND@r{, }&
12327 Boolean conjunction. Defined on boolean types.
12328
12329 @item @@
12330 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
12331
12332 @item +@r{, }-
12333 Addition and subtraction on integral and floating-point types, or union
12334 and difference on set types.
12335
12336 @item *
12337 Multiplication on integral and floating-point types, or set intersection
12338 on set types.
12339
12340 @item /
12341 Division on floating-point types, or symmetric set difference on set
12342 types. Same precedence as @code{*}.
12343
12344 @item DIV@r{, }MOD
12345 Integer division and remainder. Defined on integral types. Same
12346 precedence as @code{*}.
12347
12348 @item -
12349 Negative. Defined on @code{INTEGER} and @code{REAL} data.
12350
12351 @item ^
12352 Pointer dereferencing. Defined on pointer types.
12353
12354 @item NOT
12355 Boolean negation. Defined on boolean types. Same precedence as
12356 @code{^}.
12357
12358 @item .
12359 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
12360 precedence as @code{^}.
12361
12362 @item []
12363 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
12364
12365 @item ()
12366 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
12367 as @code{^}.
12368
12369 @item ::@r{, }.
12370 @value{GDBN} and Modula-2 scope operators.
12371 @end table
12372
12373 @quotation
12374 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
12375 treats the use of the operator @code{IN}, or the use of operators
12376 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
12377 @code{<=}, and @code{>=} on sets as an error.
12378 @end quotation
12379
12380
12381 @node Built-In Func/Proc
12382 @subsubsection Built-in Functions and Procedures
12383 @cindex Modula-2 built-ins
12384
12385 Modula-2 also makes available several built-in procedures and functions.
12386 In describing these, the following metavariables are used:
12387
12388 @table @var
12389
12390 @item a
12391 represents an @code{ARRAY} variable.
12392
12393 @item c
12394 represents a @code{CHAR} constant or variable.
12395
12396 @item i
12397 represents a variable or constant of integral type.
12398
12399 @item m
12400 represents an identifier that belongs to a set. Generally used in the
12401 same function with the metavariable @var{s}. The type of @var{s} should
12402 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
12403
12404 @item n
12405 represents a variable or constant of integral or floating-point type.
12406
12407 @item r
12408 represents a variable or constant of floating-point type.
12409
12410 @item t
12411 represents a type.
12412
12413 @item v
12414 represents a variable.
12415
12416 @item x
12417 represents a variable or constant of one of many types. See the
12418 explanation of the function for details.
12419 @end table
12420
12421 All Modula-2 built-in procedures also return a result, described below.
12422
12423 @table @code
12424 @item ABS(@var{n})
12425 Returns the absolute value of @var{n}.
12426
12427 @item CAP(@var{c})
12428 If @var{c} is a lower case letter, it returns its upper case
12429 equivalent, otherwise it returns its argument.
12430
12431 @item CHR(@var{i})
12432 Returns the character whose ordinal value is @var{i}.
12433
12434 @item DEC(@var{v})
12435 Decrements the value in the variable @var{v} by one. Returns the new value.
12436
12437 @item DEC(@var{v},@var{i})
12438 Decrements the value in the variable @var{v} by @var{i}. Returns the
12439 new value.
12440
12441 @item EXCL(@var{m},@var{s})
12442 Removes the element @var{m} from the set @var{s}. Returns the new
12443 set.
12444
12445 @item FLOAT(@var{i})
12446 Returns the floating point equivalent of the integer @var{i}.
12447
12448 @item HIGH(@var{a})
12449 Returns the index of the last member of @var{a}.
12450
12451 @item INC(@var{v})
12452 Increments the value in the variable @var{v} by one. Returns the new value.
12453
12454 @item INC(@var{v},@var{i})
12455 Increments the value in the variable @var{v} by @var{i}. Returns the
12456 new value.
12457
12458 @item INCL(@var{m},@var{s})
12459 Adds the element @var{m} to the set @var{s} if it is not already
12460 there. Returns the new set.
12461
12462 @item MAX(@var{t})
12463 Returns the maximum value of the type @var{t}.
12464
12465 @item MIN(@var{t})
12466 Returns the minimum value of the type @var{t}.
12467
12468 @item ODD(@var{i})
12469 Returns boolean TRUE if @var{i} is an odd number.
12470
12471 @item ORD(@var{x})
12472 Returns the ordinal value of its argument. For example, the ordinal
12473 value of a character is its @sc{ascii} value (on machines supporting the
12474 @sc{ascii} character set). @var{x} must be of an ordered type, which include
12475 integral, character and enumerated types.
12476
12477 @item SIZE(@var{x})
12478 Returns the size of its argument. @var{x} can be a variable or a type.
12479
12480 @item TRUNC(@var{r})
12481 Returns the integral part of @var{r}.
12482
12483 @item TSIZE(@var{x})
12484 Returns the size of its argument. @var{x} can be a variable or a type.
12485
12486 @item VAL(@var{t},@var{i})
12487 Returns the member of the type @var{t} whose ordinal value is @var{i}.
12488 @end table
12489
12490 @quotation
12491 @emph{Warning:} Sets and their operations are not yet supported, so
12492 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
12493 an error.
12494 @end quotation
12495
12496 @cindex Modula-2 constants
12497 @node M2 Constants
12498 @subsubsection Constants
12499
12500 @value{GDBN} allows you to express the constants of Modula-2 in the following
12501 ways:
12502
12503 @itemize @bullet
12504
12505 @item
12506 Integer constants are simply a sequence of digits. When used in an
12507 expression, a constant is interpreted to be type-compatible with the
12508 rest of the expression. Hexadecimal integers are specified by a
12509 trailing @samp{H}, and octal integers by a trailing @samp{B}.
12510
12511 @item
12512 Floating point constants appear as a sequence of digits, followed by a
12513 decimal point and another sequence of digits. An optional exponent can
12514 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
12515 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
12516 digits of the floating point constant must be valid decimal (base 10)
12517 digits.
12518
12519 @item
12520 Character constants consist of a single character enclosed by a pair of
12521 like quotes, either single (@code{'}) or double (@code{"}). They may
12522 also be expressed by their ordinal value (their @sc{ascii} value, usually)
12523 followed by a @samp{C}.
12524
12525 @item
12526 String constants consist of a sequence of characters enclosed by a
12527 pair of like quotes, either single (@code{'}) or double (@code{"}).
12528 Escape sequences in the style of C are also allowed. @xref{C
12529 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
12530 sequences.
12531
12532 @item
12533 Enumerated constants consist of an enumerated identifier.
12534
12535 @item
12536 Boolean constants consist of the identifiers @code{TRUE} and
12537 @code{FALSE}.
12538
12539 @item
12540 Pointer constants consist of integral values only.
12541
12542 @item
12543 Set constants are not yet supported.
12544 @end itemize
12545
12546 @node M2 Types
12547 @subsubsection Modula-2 Types
12548 @cindex Modula-2 types
12549
12550 Currently @value{GDBN} can print the following data types in Modula-2
12551 syntax: array types, record types, set types, pointer types, procedure
12552 types, enumerated types, subrange types and base types. You can also
12553 print the contents of variables declared using these type.
12554 This section gives a number of simple source code examples together with
12555 sample @value{GDBN} sessions.
12556
12557 The first example contains the following section of code:
12558
12559 @smallexample
12560 VAR
12561 s: SET OF CHAR ;
12562 r: [20..40] ;
12563 @end smallexample
12564
12565 @noindent
12566 and you can request @value{GDBN} to interrogate the type and value of
12567 @code{r} and @code{s}.
12568
12569 @smallexample
12570 (@value{GDBP}) print s
12571 @{'A'..'C', 'Z'@}
12572 (@value{GDBP}) ptype s
12573 SET OF CHAR
12574 (@value{GDBP}) print r
12575 21
12576 (@value{GDBP}) ptype r
12577 [20..40]
12578 @end smallexample
12579
12580 @noindent
12581 Likewise if your source code declares @code{s} as:
12582
12583 @smallexample
12584 VAR
12585 s: SET ['A'..'Z'] ;
12586 @end smallexample
12587
12588 @noindent
12589 then you may query the type of @code{s} by:
12590
12591 @smallexample
12592 (@value{GDBP}) ptype s
12593 type = SET ['A'..'Z']
12594 @end smallexample
12595
12596 @noindent
12597 Note that at present you cannot interactively manipulate set
12598 expressions using the debugger.
12599
12600 The following example shows how you might declare an array in Modula-2
12601 and how you can interact with @value{GDBN} to print its type and contents:
12602
12603 @smallexample
12604 VAR
12605 s: ARRAY [-10..10] OF CHAR ;
12606 @end smallexample
12607
12608 @smallexample
12609 (@value{GDBP}) ptype s
12610 ARRAY [-10..10] OF CHAR
12611 @end smallexample
12612
12613 Note that the array handling is not yet complete and although the type
12614 is printed correctly, expression handling still assumes that all
12615 arrays have a lower bound of zero and not @code{-10} as in the example
12616 above.
12617
12618 Here are some more type related Modula-2 examples:
12619
12620 @smallexample
12621 TYPE
12622 colour = (blue, red, yellow, green) ;
12623 t = [blue..yellow] ;
12624 VAR
12625 s: t ;
12626 BEGIN
12627 s := blue ;
12628 @end smallexample
12629
12630 @noindent
12631 The @value{GDBN} interaction shows how you can query the data type
12632 and value of a variable.
12633
12634 @smallexample
12635 (@value{GDBP}) print s
12636 $1 = blue
12637 (@value{GDBP}) ptype t
12638 type = [blue..yellow]
12639 @end smallexample
12640
12641 @noindent
12642 In this example a Modula-2 array is declared and its contents
12643 displayed. Observe that the contents are written in the same way as
12644 their @code{C} counterparts.
12645
12646 @smallexample
12647 VAR
12648 s: ARRAY [1..5] OF CARDINAL ;
12649 BEGIN
12650 s[1] := 1 ;
12651 @end smallexample
12652
12653 @smallexample
12654 (@value{GDBP}) print s
12655 $1 = @{1, 0, 0, 0, 0@}
12656 (@value{GDBP}) ptype s
12657 type = ARRAY [1..5] OF CARDINAL
12658 @end smallexample
12659
12660 The Modula-2 language interface to @value{GDBN} also understands
12661 pointer types as shown in this example:
12662
12663 @smallexample
12664 VAR
12665 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
12666 BEGIN
12667 NEW(s) ;
12668 s^[1] := 1 ;
12669 @end smallexample
12670
12671 @noindent
12672 and you can request that @value{GDBN} describes the type of @code{s}.
12673
12674 @smallexample
12675 (@value{GDBP}) ptype s
12676 type = POINTER TO ARRAY [1..5] OF CARDINAL
12677 @end smallexample
12678
12679 @value{GDBN} handles compound types as we can see in this example.
12680 Here we combine array types, record types, pointer types and subrange
12681 types:
12682
12683 @smallexample
12684 TYPE
12685 foo = RECORD
12686 f1: CARDINAL ;
12687 f2: CHAR ;
12688 f3: myarray ;
12689 END ;
12690
12691 myarray = ARRAY myrange OF CARDINAL ;
12692 myrange = [-2..2] ;
12693 VAR
12694 s: POINTER TO ARRAY myrange OF foo ;
12695 @end smallexample
12696
12697 @noindent
12698 and you can ask @value{GDBN} to describe the type of @code{s} as shown
12699 below.
12700
12701 @smallexample
12702 (@value{GDBP}) ptype s
12703 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
12704 f1 : CARDINAL;
12705 f2 : CHAR;
12706 f3 : ARRAY [-2..2] OF CARDINAL;
12707 END
12708 @end smallexample
12709
12710 @node M2 Defaults
12711 @subsubsection Modula-2 Defaults
12712 @cindex Modula-2 defaults
12713
12714 If type and range checking are set automatically by @value{GDBN}, they
12715 both default to @code{on} whenever the working language changes to
12716 Modula-2. This happens regardless of whether you or @value{GDBN}
12717 selected the working language.
12718
12719 If you allow @value{GDBN} to set the language automatically, then entering
12720 code compiled from a file whose name ends with @file{.mod} sets the
12721 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
12722 Infer the Source Language}, for further details.
12723
12724 @node Deviations
12725 @subsubsection Deviations from Standard Modula-2
12726 @cindex Modula-2, deviations from
12727
12728 A few changes have been made to make Modula-2 programs easier to debug.
12729 This is done primarily via loosening its type strictness:
12730
12731 @itemize @bullet
12732 @item
12733 Unlike in standard Modula-2, pointer constants can be formed by
12734 integers. This allows you to modify pointer variables during
12735 debugging. (In standard Modula-2, the actual address contained in a
12736 pointer variable is hidden from you; it can only be modified
12737 through direct assignment to another pointer variable or expression that
12738 returned a pointer.)
12739
12740 @item
12741 C escape sequences can be used in strings and characters to represent
12742 non-printable characters. @value{GDBN} prints out strings with these
12743 escape sequences embedded. Single non-printable characters are
12744 printed using the @samp{CHR(@var{nnn})} format.
12745
12746 @item
12747 The assignment operator (@code{:=}) returns the value of its right-hand
12748 argument.
12749
12750 @item
12751 All built-in procedures both modify @emph{and} return their argument.
12752 @end itemize
12753
12754 @node M2 Checks
12755 @subsubsection Modula-2 Type and Range Checks
12756 @cindex Modula-2 checks
12757
12758 @quotation
12759 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
12760 range checking.
12761 @end quotation
12762 @c FIXME remove warning when type/range checks added
12763
12764 @value{GDBN} considers two Modula-2 variables type equivalent if:
12765
12766 @itemize @bullet
12767 @item
12768 They are of types that have been declared equivalent via a @code{TYPE
12769 @var{t1} = @var{t2}} statement
12770
12771 @item
12772 They have been declared on the same line. (Note: This is true of the
12773 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
12774 @end itemize
12775
12776 As long as type checking is enabled, any attempt to combine variables
12777 whose types are not equivalent is an error.
12778
12779 Range checking is done on all mathematical operations, assignment, array
12780 index bounds, and all built-in functions and procedures.
12781
12782 @node M2 Scope
12783 @subsubsection The Scope Operators @code{::} and @code{.}
12784 @cindex scope
12785 @cindex @code{.}, Modula-2 scope operator
12786 @cindex colon, doubled as scope operator
12787 @ifinfo
12788 @vindex colon-colon@r{, in Modula-2}
12789 @c Info cannot handle :: but TeX can.
12790 @end ifinfo
12791 @ifnotinfo
12792 @vindex ::@r{, in Modula-2}
12793 @end ifnotinfo
12794
12795 There are a few subtle differences between the Modula-2 scope operator
12796 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
12797 similar syntax:
12798
12799 @smallexample
12800
12801 @var{module} . @var{id}
12802 @var{scope} :: @var{id}
12803 @end smallexample
12804
12805 @noindent
12806 where @var{scope} is the name of a module or a procedure,
12807 @var{module} the name of a module, and @var{id} is any declared
12808 identifier within your program, except another module.
12809
12810 Using the @code{::} operator makes @value{GDBN} search the scope
12811 specified by @var{scope} for the identifier @var{id}. If it is not
12812 found in the specified scope, then @value{GDBN} searches all scopes
12813 enclosing the one specified by @var{scope}.
12814
12815 Using the @code{.} operator makes @value{GDBN} search the current scope for
12816 the identifier specified by @var{id} that was imported from the
12817 definition module specified by @var{module}. With this operator, it is
12818 an error if the identifier @var{id} was not imported from definition
12819 module @var{module}, or if @var{id} is not an identifier in
12820 @var{module}.
12821
12822 @node GDB/M2
12823 @subsubsection @value{GDBN} and Modula-2
12824
12825 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
12826 Five subcommands of @code{set print} and @code{show print} apply
12827 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
12828 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
12829 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
12830 analogue in Modula-2.
12831
12832 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
12833 with any language, is not useful with Modula-2. Its
12834 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
12835 created in Modula-2 as they can in C or C@t{++}. However, because an
12836 address can be specified by an integral constant, the construct
12837 @samp{@{@var{type}@}@var{adrexp}} is still useful.
12838
12839 @cindex @code{#} in Modula-2
12840 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
12841 interpreted as the beginning of a comment. Use @code{<>} instead.
12842
12843 @node Ada
12844 @subsection Ada
12845 @cindex Ada
12846
12847 The extensions made to @value{GDBN} for Ada only support
12848 output from the @sc{gnu} Ada (GNAT) compiler.
12849 Other Ada compilers are not currently supported, and
12850 attempting to debug executables produced by them is most likely
12851 to be difficult.
12852
12853
12854 @cindex expressions in Ada
12855 @menu
12856 * Ada Mode Intro:: General remarks on the Ada syntax
12857 and semantics supported by Ada mode
12858 in @value{GDBN}.
12859 * Omissions from Ada:: Restrictions on the Ada expression syntax.
12860 * Additions to Ada:: Extensions of the Ada expression syntax.
12861 * Stopping Before Main Program:: Debugging the program during elaboration.
12862 * Ada Tasks:: Listing and setting breakpoints in tasks.
12863 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
12864 * Ada Glitches:: Known peculiarities of Ada mode.
12865 @end menu
12866
12867 @node Ada Mode Intro
12868 @subsubsection Introduction
12869 @cindex Ada mode, general
12870
12871 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
12872 syntax, with some extensions.
12873 The philosophy behind the design of this subset is
12874
12875 @itemize @bullet
12876 @item
12877 That @value{GDBN} should provide basic literals and access to operations for
12878 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
12879 leaving more sophisticated computations to subprograms written into the
12880 program (which therefore may be called from @value{GDBN}).
12881
12882 @item
12883 That type safety and strict adherence to Ada language restrictions
12884 are not particularly important to the @value{GDBN} user.
12885
12886 @item
12887 That brevity is important to the @value{GDBN} user.
12888 @end itemize
12889
12890 Thus, for brevity, the debugger acts as if all names declared in
12891 user-written packages are directly visible, even if they are not visible
12892 according to Ada rules, thus making it unnecessary to fully qualify most
12893 names with their packages, regardless of context. Where this causes
12894 ambiguity, @value{GDBN} asks the user's intent.
12895
12896 The debugger will start in Ada mode if it detects an Ada main program.
12897 As for other languages, it will enter Ada mode when stopped in a program that
12898 was translated from an Ada source file.
12899
12900 While in Ada mode, you may use `@t{--}' for comments. This is useful
12901 mostly for documenting command files. The standard @value{GDBN} comment
12902 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
12903 middle (to allow based literals).
12904
12905 The debugger supports limited overloading. Given a subprogram call in which
12906 the function symbol has multiple definitions, it will use the number of
12907 actual parameters and some information about their types to attempt to narrow
12908 the set of definitions. It also makes very limited use of context, preferring
12909 procedures to functions in the context of the @code{call} command, and
12910 functions to procedures elsewhere.
12911
12912 @node Omissions from Ada
12913 @subsubsection Omissions from Ada
12914 @cindex Ada, omissions from
12915
12916 Here are the notable omissions from the subset:
12917
12918 @itemize @bullet
12919 @item
12920 Only a subset of the attributes are supported:
12921
12922 @itemize @minus
12923 @item
12924 @t{'First}, @t{'Last}, and @t{'Length}
12925 on array objects (not on types and subtypes).
12926
12927 @item
12928 @t{'Min} and @t{'Max}.
12929
12930 @item
12931 @t{'Pos} and @t{'Val}.
12932
12933 @item
12934 @t{'Tag}.
12935
12936 @item
12937 @t{'Range} on array objects (not subtypes), but only as the right
12938 operand of the membership (@code{in}) operator.
12939
12940 @item
12941 @t{'Access}, @t{'Unchecked_Access}, and
12942 @t{'Unrestricted_Access} (a GNAT extension).
12943
12944 @item
12945 @t{'Address}.
12946 @end itemize
12947
12948 @item
12949 The names in
12950 @code{Characters.Latin_1} are not available and
12951 concatenation is not implemented. Thus, escape characters in strings are
12952 not currently available.
12953
12954 @item
12955 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
12956 equality of representations. They will generally work correctly
12957 for strings and arrays whose elements have integer or enumeration types.
12958 They may not work correctly for arrays whose element
12959 types have user-defined equality, for arrays of real values
12960 (in particular, IEEE-conformant floating point, because of negative
12961 zeroes and NaNs), and for arrays whose elements contain unused bits with
12962 indeterminate values.
12963
12964 @item
12965 The other component-by-component array operations (@code{and}, @code{or},
12966 @code{xor}, @code{not}, and relational tests other than equality)
12967 are not implemented.
12968
12969 @item
12970 @cindex array aggregates (Ada)
12971 @cindex record aggregates (Ada)
12972 @cindex aggregates (Ada)
12973 There is limited support for array and record aggregates. They are
12974 permitted only on the right sides of assignments, as in these examples:
12975
12976 @smallexample
12977 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
12978 (@value{GDBP}) set An_Array := (1, others => 0)
12979 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
12980 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
12981 (@value{GDBP}) set A_Record := (1, "Peter", True);
12982 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
12983 @end smallexample
12984
12985 Changing a
12986 discriminant's value by assigning an aggregate has an
12987 undefined effect if that discriminant is used within the record.
12988 However, you can first modify discriminants by directly assigning to
12989 them (which normally would not be allowed in Ada), and then performing an
12990 aggregate assignment. For example, given a variable @code{A_Rec}
12991 declared to have a type such as:
12992
12993 @smallexample
12994 type Rec (Len : Small_Integer := 0) is record
12995 Id : Integer;
12996 Vals : IntArray (1 .. Len);
12997 end record;
12998 @end smallexample
12999
13000 you can assign a value with a different size of @code{Vals} with two
13001 assignments:
13002
13003 @smallexample
13004 (@value{GDBP}) set A_Rec.Len := 4
13005 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
13006 @end smallexample
13007
13008 As this example also illustrates, @value{GDBN} is very loose about the usual
13009 rules concerning aggregates. You may leave out some of the
13010 components of an array or record aggregate (such as the @code{Len}
13011 component in the assignment to @code{A_Rec} above); they will retain their
13012 original values upon assignment. You may freely use dynamic values as
13013 indices in component associations. You may even use overlapping or
13014 redundant component associations, although which component values are
13015 assigned in such cases is not defined.
13016
13017 @item
13018 Calls to dispatching subprograms are not implemented.
13019
13020 @item
13021 The overloading algorithm is much more limited (i.e., less selective)
13022 than that of real Ada. It makes only limited use of the context in
13023 which a subexpression appears to resolve its meaning, and it is much
13024 looser in its rules for allowing type matches. As a result, some
13025 function calls will be ambiguous, and the user will be asked to choose
13026 the proper resolution.
13027
13028 @item
13029 The @code{new} operator is not implemented.
13030
13031 @item
13032 Entry calls are not implemented.
13033
13034 @item
13035 Aside from printing, arithmetic operations on the native VAX floating-point
13036 formats are not supported.
13037
13038 @item
13039 It is not possible to slice a packed array.
13040
13041 @item
13042 The names @code{True} and @code{False}, when not part of a qualified name,
13043 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
13044 context.
13045 Should your program
13046 redefine these names in a package or procedure (at best a dubious practice),
13047 you will have to use fully qualified names to access their new definitions.
13048 @end itemize
13049
13050 @node Additions to Ada
13051 @subsubsection Additions to Ada
13052 @cindex Ada, deviations from
13053
13054 As it does for other languages, @value{GDBN} makes certain generic
13055 extensions to Ada (@pxref{Expressions}):
13056
13057 @itemize @bullet
13058 @item
13059 If the expression @var{E} is a variable residing in memory (typically
13060 a local variable or array element) and @var{N} is a positive integer,
13061 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
13062 @var{N}-1 adjacent variables following it in memory as an array. In
13063 Ada, this operator is generally not necessary, since its prime use is
13064 in displaying parts of an array, and slicing will usually do this in
13065 Ada. However, there are occasional uses when debugging programs in
13066 which certain debugging information has been optimized away.
13067
13068 @item
13069 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
13070 appears in function or file @var{B}.'' When @var{B} is a file name,
13071 you must typically surround it in single quotes.
13072
13073 @item
13074 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
13075 @var{type} that appears at address @var{addr}.''
13076
13077 @item
13078 A name starting with @samp{$} is a convenience variable
13079 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
13080 @end itemize
13081
13082 In addition, @value{GDBN} provides a few other shortcuts and outright
13083 additions specific to Ada:
13084
13085 @itemize @bullet
13086 @item
13087 The assignment statement is allowed as an expression, returning
13088 its right-hand operand as its value. Thus, you may enter
13089
13090 @smallexample
13091 (@value{GDBP}) set x := y + 3
13092 (@value{GDBP}) print A(tmp := y + 1)
13093 @end smallexample
13094
13095 @item
13096 The semicolon is allowed as an ``operator,'' returning as its value
13097 the value of its right-hand operand.
13098 This allows, for example,
13099 complex conditional breaks:
13100
13101 @smallexample
13102 (@value{GDBP}) break f
13103 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
13104 @end smallexample
13105
13106 @item
13107 Rather than use catenation and symbolic character names to introduce special
13108 characters into strings, one may instead use a special bracket notation,
13109 which is also used to print strings. A sequence of characters of the form
13110 @samp{["@var{XX}"]} within a string or character literal denotes the
13111 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
13112 sequence of characters @samp{["""]} also denotes a single quotation mark
13113 in strings. For example,
13114 @smallexample
13115 "One line.["0a"]Next line.["0a"]"
13116 @end smallexample
13117 @noindent
13118 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
13119 after each period.
13120
13121 @item
13122 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
13123 @t{'Max} is optional (and is ignored in any case). For example, it is valid
13124 to write
13125
13126 @smallexample
13127 (@value{GDBP}) print 'max(x, y)
13128 @end smallexample
13129
13130 @item
13131 When printing arrays, @value{GDBN} uses positional notation when the
13132 array has a lower bound of 1, and uses a modified named notation otherwise.
13133 For example, a one-dimensional array of three integers with a lower bound
13134 of 3 might print as
13135
13136 @smallexample
13137 (3 => 10, 17, 1)
13138 @end smallexample
13139
13140 @noindent
13141 That is, in contrast to valid Ada, only the first component has a @code{=>}
13142 clause.
13143
13144 @item
13145 You may abbreviate attributes in expressions with any unique,
13146 multi-character subsequence of
13147 their names (an exact match gets preference).
13148 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
13149 in place of @t{a'length}.
13150
13151 @item
13152 @cindex quoting Ada internal identifiers
13153 Since Ada is case-insensitive, the debugger normally maps identifiers you type
13154 to lower case. The GNAT compiler uses upper-case characters for
13155 some of its internal identifiers, which are normally of no interest to users.
13156 For the rare occasions when you actually have to look at them,
13157 enclose them in angle brackets to avoid the lower-case mapping.
13158 For example,
13159 @smallexample
13160 (@value{GDBP}) print <JMPBUF_SAVE>[0]
13161 @end smallexample
13162
13163 @item
13164 Printing an object of class-wide type or dereferencing an
13165 access-to-class-wide value will display all the components of the object's
13166 specific type (as indicated by its run-time tag). Likewise, component
13167 selection on such a value will operate on the specific type of the
13168 object.
13169
13170 @end itemize
13171
13172 @node Stopping Before Main Program
13173 @subsubsection Stopping at the Very Beginning
13174
13175 @cindex breakpointing Ada elaboration code
13176 It is sometimes necessary to debug the program during elaboration, and
13177 before reaching the main procedure.
13178 As defined in the Ada Reference
13179 Manual, the elaboration code is invoked from a procedure called
13180 @code{adainit}. To run your program up to the beginning of
13181 elaboration, simply use the following two commands:
13182 @code{tbreak adainit} and @code{run}.
13183
13184 @node Ada Tasks
13185 @subsubsection Extensions for Ada Tasks
13186 @cindex Ada, tasking
13187
13188 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
13189 @value{GDBN} provides the following task-related commands:
13190
13191 @table @code
13192 @kindex info tasks
13193 @item info tasks
13194 This command shows a list of current Ada tasks, as in the following example:
13195
13196
13197 @smallexample
13198 @iftex
13199 @leftskip=0.5cm
13200 @end iftex
13201 (@value{GDBP}) info tasks
13202 ID TID P-ID Pri State Name
13203 1 8088000 0 15 Child Activation Wait main_task
13204 2 80a4000 1 15 Accept Statement b
13205 3 809a800 1 15 Child Activation Wait a
13206 * 4 80ae800 3 15 Runnable c
13207
13208 @end smallexample
13209
13210 @noindent
13211 In this listing, the asterisk before the last task indicates it to be the
13212 task currently being inspected.
13213
13214 @table @asis
13215 @item ID
13216 Represents @value{GDBN}'s internal task number.
13217
13218 @item TID
13219 The Ada task ID.
13220
13221 @item P-ID
13222 The parent's task ID (@value{GDBN}'s internal task number).
13223
13224 @item Pri
13225 The base priority of the task.
13226
13227 @item State
13228 Current state of the task.
13229
13230 @table @code
13231 @item Unactivated
13232 The task has been created but has not been activated. It cannot be
13233 executing.
13234
13235 @item Runnable
13236 The task is not blocked for any reason known to Ada. (It may be waiting
13237 for a mutex, though.) It is conceptually "executing" in normal mode.
13238
13239 @item Terminated
13240 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
13241 that were waiting on terminate alternatives have been awakened and have
13242 terminated themselves.
13243
13244 @item Child Activation Wait
13245 The task is waiting for created tasks to complete activation.
13246
13247 @item Accept Statement
13248 The task is waiting on an accept or selective wait statement.
13249
13250 @item Waiting on entry call
13251 The task is waiting on an entry call.
13252
13253 @item Async Select Wait
13254 The task is waiting to start the abortable part of an asynchronous
13255 select statement.
13256
13257 @item Delay Sleep
13258 The task is waiting on a select statement with only a delay
13259 alternative open.
13260
13261 @item Child Termination Wait
13262 The task is sleeping having completed a master within itself, and is
13263 waiting for the tasks dependent on that master to become terminated or
13264 waiting on a terminate Phase.
13265
13266 @item Wait Child in Term Alt
13267 The task is sleeping waiting for tasks on terminate alternatives to
13268 finish terminating.
13269
13270 @item Accepting RV with @var{taskno}
13271 The task is accepting a rendez-vous with the task @var{taskno}.
13272 @end table
13273
13274 @item Name
13275 Name of the task in the program.
13276
13277 @end table
13278
13279 @kindex info task @var{taskno}
13280 @item info task @var{taskno}
13281 This command shows detailled informations on the specified task, as in
13282 the following example:
13283 @smallexample
13284 @iftex
13285 @leftskip=0.5cm
13286 @end iftex
13287 (@value{GDBP}) info tasks
13288 ID TID P-ID Pri State Name
13289 1 8077880 0 15 Child Activation Wait main_task
13290 * 2 807c468 1 15 Runnable task_1
13291 (@value{GDBP}) info task 2
13292 Ada Task: 0x807c468
13293 Name: task_1
13294 Thread: 0x807f378
13295 Parent: 1 (main_task)
13296 Base Priority: 15
13297 State: Runnable
13298 @end smallexample
13299
13300 @item task
13301 @kindex task@r{ (Ada)}
13302 @cindex current Ada task ID
13303 This command prints the ID of the current task.
13304
13305 @smallexample
13306 @iftex
13307 @leftskip=0.5cm
13308 @end iftex
13309 (@value{GDBP}) info tasks
13310 ID TID P-ID Pri State Name
13311 1 8077870 0 15 Child Activation Wait main_task
13312 * 2 807c458 1 15 Runnable t
13313 (@value{GDBP}) task
13314 [Current task is 2]
13315 @end smallexample
13316
13317 @item task @var{taskno}
13318 @cindex Ada task switching
13319 This command is like the @code{thread @var{threadno}}
13320 command (@pxref{Threads}). It switches the context of debugging
13321 from the current task to the given task.
13322
13323 @smallexample
13324 @iftex
13325 @leftskip=0.5cm
13326 @end iftex
13327 (@value{GDBP}) info tasks
13328 ID TID P-ID Pri State Name
13329 1 8077870 0 15 Child Activation Wait main_task
13330 * 2 807c458 1 15 Runnable t
13331 (@value{GDBP}) task 1
13332 [Switching to task 1]
13333 #0 0x8067726 in pthread_cond_wait ()
13334 (@value{GDBP}) bt
13335 #0 0x8067726 in pthread_cond_wait ()
13336 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
13337 #2 0x805cb63 in system.task_primitives.operations.sleep ()
13338 #3 0x806153e in system.tasking.stages.activate_tasks ()
13339 #4 0x804aacc in un () at un.adb:5
13340 @end smallexample
13341
13342 @item break @var{linespec} task @var{taskno}
13343 @itemx break @var{linespec} task @var{taskno} if @dots{}
13344 @cindex breakpoints and tasks, in Ada
13345 @cindex task breakpoints, in Ada
13346 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
13347 These commands are like the @code{break @dots{} thread @dots{}}
13348 command (@pxref{Thread Stops}).
13349 @var{linespec} specifies source lines, as described
13350 in @ref{Specify Location}.
13351
13352 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
13353 to specify that you only want @value{GDBN} to stop the program when a
13354 particular Ada task reaches this breakpoint. @var{taskno} is one of the
13355 numeric task identifiers assigned by @value{GDBN}, shown in the first
13356 column of the @samp{info tasks} display.
13357
13358 If you do not specify @samp{task @var{taskno}} when you set a
13359 breakpoint, the breakpoint applies to @emph{all} tasks of your
13360 program.
13361
13362 You can use the @code{task} qualifier on conditional breakpoints as
13363 well; in this case, place @samp{task @var{taskno}} before the
13364 breakpoint condition (before the @code{if}).
13365
13366 For example,
13367
13368 @smallexample
13369 @iftex
13370 @leftskip=0.5cm
13371 @end iftex
13372 (@value{GDBP}) info tasks
13373 ID TID P-ID Pri State Name
13374 1 140022020 0 15 Child Activation Wait main_task
13375 2 140045060 1 15 Accept/Select Wait t2
13376 3 140044840 1 15 Runnable t1
13377 * 4 140056040 1 15 Runnable t3
13378 (@value{GDBP}) b 15 task 2
13379 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
13380 (@value{GDBP}) cont
13381 Continuing.
13382 task # 1 running
13383 task # 2 running
13384
13385 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
13386 15 flush;
13387 (@value{GDBP}) info tasks
13388 ID TID P-ID Pri State Name
13389 1 140022020 0 15 Child Activation Wait main_task
13390 * 2 140045060 1 15 Runnable t2
13391 3 140044840 1 15 Runnable t1
13392 4 140056040 1 15 Delay Sleep t3
13393 @end smallexample
13394 @end table
13395
13396 @node Ada Tasks and Core Files
13397 @subsubsection Tasking Support when Debugging Core Files
13398 @cindex Ada tasking and core file debugging
13399
13400 When inspecting a core file, as opposed to debugging a live program,
13401 tasking support may be limited or even unavailable, depending on
13402 the platform being used.
13403 For instance, on x86-linux, the list of tasks is available, but task
13404 switching is not supported. On Tru64, however, task switching will work
13405 as usual.
13406
13407 On certain platforms, including Tru64, the debugger needs to perform some
13408 memory writes in order to provide Ada tasking support. When inspecting
13409 a core file, this means that the core file must be opened with read-write
13410 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
13411 Under these circumstances, you should make a backup copy of the core
13412 file before inspecting it with @value{GDBN}.
13413
13414 @node Ada Glitches
13415 @subsubsection Known Peculiarities of Ada Mode
13416 @cindex Ada, problems
13417
13418 Besides the omissions listed previously (@pxref{Omissions from Ada}),
13419 we know of several problems with and limitations of Ada mode in
13420 @value{GDBN},
13421 some of which will be fixed with planned future releases of the debugger
13422 and the GNU Ada compiler.
13423
13424 @itemize @bullet
13425 @item
13426 Currently, the debugger
13427 has insufficient information to determine whether certain pointers represent
13428 pointers to objects or the objects themselves.
13429 Thus, the user may have to tack an extra @code{.all} after an expression
13430 to get it printed properly.
13431
13432 @item
13433 Static constants that the compiler chooses not to materialize as objects in
13434 storage are invisible to the debugger.
13435
13436 @item
13437 Named parameter associations in function argument lists are ignored (the
13438 argument lists are treated as positional).
13439
13440 @item
13441 Many useful library packages are currently invisible to the debugger.
13442
13443 @item
13444 Fixed-point arithmetic, conversions, input, and output is carried out using
13445 floating-point arithmetic, and may give results that only approximate those on
13446 the host machine.
13447
13448 @item
13449 The GNAT compiler never generates the prefix @code{Standard} for any of
13450 the standard symbols defined by the Ada language. @value{GDBN} knows about
13451 this: it will strip the prefix from names when you use it, and will never
13452 look for a name you have so qualified among local symbols, nor match against
13453 symbols in other packages or subprograms. If you have
13454 defined entities anywhere in your program other than parameters and
13455 local variables whose simple names match names in @code{Standard},
13456 GNAT's lack of qualification here can cause confusion. When this happens,
13457 you can usually resolve the confusion
13458 by qualifying the problematic names with package
13459 @code{Standard} explicitly.
13460 @end itemize
13461
13462 Older versions of the compiler sometimes generate erroneous debugging
13463 information, resulting in the debugger incorrectly printing the value
13464 of affected entities. In some cases, the debugger is able to work
13465 around an issue automatically. In other cases, the debugger is able
13466 to work around the issue, but the work-around has to be specifically
13467 enabled.
13468
13469 @kindex set ada trust-PAD-over-XVS
13470 @kindex show ada trust-PAD-over-XVS
13471 @table @code
13472
13473 @item set ada trust-PAD-over-XVS on
13474 Configure GDB to strictly follow the GNAT encoding when computing the
13475 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
13476 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
13477 a complete description of the encoding used by the GNAT compiler).
13478 This is the default.
13479
13480 @item set ada trust-PAD-over-XVS off
13481 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
13482 sometimes prints the wrong value for certain entities, changing @code{ada
13483 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
13484 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
13485 @code{off}, but this incurs a slight performance penalty, so it is
13486 recommended to leave this setting to @code{on} unless necessary.
13487
13488 @end table
13489
13490 @node Unsupported Languages
13491 @section Unsupported Languages
13492
13493 @cindex unsupported languages
13494 @cindex minimal language
13495 In addition to the other fully-supported programming languages,
13496 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
13497 It does not represent a real programming language, but provides a set
13498 of capabilities close to what the C or assembly languages provide.
13499 This should allow most simple operations to be performed while debugging
13500 an application that uses a language currently not supported by @value{GDBN}.
13501
13502 If the language is set to @code{auto}, @value{GDBN} will automatically
13503 select this language if the current frame corresponds to an unsupported
13504 language.
13505
13506 @node Symbols
13507 @chapter Examining the Symbol Table
13508
13509 The commands described in this chapter allow you to inquire about the
13510 symbols (names of variables, functions and types) defined in your
13511 program. This information is inherent in the text of your program and
13512 does not change as your program executes. @value{GDBN} finds it in your
13513 program's symbol table, in the file indicated when you started @value{GDBN}
13514 (@pxref{File Options, ,Choosing Files}), or by one of the
13515 file-management commands (@pxref{Files, ,Commands to Specify Files}).
13516
13517 @cindex symbol names
13518 @cindex names of symbols
13519 @cindex quoting names
13520 Occasionally, you may need to refer to symbols that contain unusual
13521 characters, which @value{GDBN} ordinarily treats as word delimiters. The
13522 most frequent case is in referring to static variables in other
13523 source files (@pxref{Variables,,Program Variables}). File names
13524 are recorded in object files as debugging symbols, but @value{GDBN} would
13525 ordinarily parse a typical file name, like @file{foo.c}, as the three words
13526 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
13527 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
13528
13529 @smallexample
13530 p 'foo.c'::x
13531 @end smallexample
13532
13533 @noindent
13534 looks up the value of @code{x} in the scope of the file @file{foo.c}.
13535
13536 @table @code
13537 @cindex case-insensitive symbol names
13538 @cindex case sensitivity in symbol names
13539 @kindex set case-sensitive
13540 @item set case-sensitive on
13541 @itemx set case-sensitive off
13542 @itemx set case-sensitive auto
13543 Normally, when @value{GDBN} looks up symbols, it matches their names
13544 with case sensitivity determined by the current source language.
13545 Occasionally, you may wish to control that. The command @code{set
13546 case-sensitive} lets you do that by specifying @code{on} for
13547 case-sensitive matches or @code{off} for case-insensitive ones. If
13548 you specify @code{auto}, case sensitivity is reset to the default
13549 suitable for the source language. The default is case-sensitive
13550 matches for all languages except for Fortran, for which the default is
13551 case-insensitive matches.
13552
13553 @kindex show case-sensitive
13554 @item show case-sensitive
13555 This command shows the current setting of case sensitivity for symbols
13556 lookups.
13557
13558 @kindex info address
13559 @cindex address of a symbol
13560 @item info address @var{symbol}
13561 Describe where the data for @var{symbol} is stored. For a register
13562 variable, this says which register it is kept in. For a non-register
13563 local variable, this prints the stack-frame offset at which the variable
13564 is always stored.
13565
13566 Note the contrast with @samp{print &@var{symbol}}, which does not work
13567 at all for a register variable, and for a stack local variable prints
13568 the exact address of the current instantiation of the variable.
13569
13570 @kindex info symbol
13571 @cindex symbol from address
13572 @cindex closest symbol and offset for an address
13573 @item info symbol @var{addr}
13574 Print the name of a symbol which is stored at the address @var{addr}.
13575 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
13576 nearest symbol and an offset from it:
13577
13578 @smallexample
13579 (@value{GDBP}) info symbol 0x54320
13580 _initialize_vx + 396 in section .text
13581 @end smallexample
13582
13583 @noindent
13584 This is the opposite of the @code{info address} command. You can use
13585 it to find out the name of a variable or a function given its address.
13586
13587 For dynamically linked executables, the name of executable or shared
13588 library containing the symbol is also printed:
13589
13590 @smallexample
13591 (@value{GDBP}) info symbol 0x400225
13592 _start + 5 in section .text of /tmp/a.out
13593 (@value{GDBP}) info symbol 0x2aaaac2811cf
13594 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
13595 @end smallexample
13596
13597 @kindex whatis
13598 @item whatis [@var{arg}]
13599 Print the data type of @var{arg}, which can be either an expression or
13600 a data type. With no argument, print the data type of @code{$}, the
13601 last value in the value history. If @var{arg} is an expression, it is
13602 not actually evaluated, and any side-effecting operations (such as
13603 assignments or function calls) inside it do not take place. If
13604 @var{arg} is a type name, it may be the name of a type or typedef, or
13605 for C code it may have the form @samp{class @var{class-name}},
13606 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
13607 @samp{enum @var{enum-tag}}.
13608 @xref{Expressions, ,Expressions}.
13609
13610 @kindex ptype
13611 @item ptype [@var{arg}]
13612 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
13613 detailed description of the type, instead of just the name of the type.
13614 @xref{Expressions, ,Expressions}.
13615
13616 For example, for this variable declaration:
13617
13618 @smallexample
13619 struct complex @{double real; double imag;@} v;
13620 @end smallexample
13621
13622 @noindent
13623 the two commands give this output:
13624
13625 @smallexample
13626 @group
13627 (@value{GDBP}) whatis v
13628 type = struct complex
13629 (@value{GDBP}) ptype v
13630 type = struct complex @{
13631 double real;
13632 double imag;
13633 @}
13634 @end group
13635 @end smallexample
13636
13637 @noindent
13638 As with @code{whatis}, using @code{ptype} without an argument refers to
13639 the type of @code{$}, the last value in the value history.
13640
13641 @cindex incomplete type
13642 Sometimes, programs use opaque data types or incomplete specifications
13643 of complex data structure. If the debug information included in the
13644 program does not allow @value{GDBN} to display a full declaration of
13645 the data type, it will say @samp{<incomplete type>}. For example,
13646 given these declarations:
13647
13648 @smallexample
13649 struct foo;
13650 struct foo *fooptr;
13651 @end smallexample
13652
13653 @noindent
13654 but no definition for @code{struct foo} itself, @value{GDBN} will say:
13655
13656 @smallexample
13657 (@value{GDBP}) ptype foo
13658 $1 = <incomplete type>
13659 @end smallexample
13660
13661 @noindent
13662 ``Incomplete type'' is C terminology for data types that are not
13663 completely specified.
13664
13665 @kindex info types
13666 @item info types @var{regexp}
13667 @itemx info types
13668 Print a brief description of all types whose names match the regular
13669 expression @var{regexp} (or all types in your program, if you supply
13670 no argument). Each complete typename is matched as though it were a
13671 complete line; thus, @samp{i type value} gives information on all
13672 types in your program whose names include the string @code{value}, but
13673 @samp{i type ^value$} gives information only on types whose complete
13674 name is @code{value}.
13675
13676 This command differs from @code{ptype} in two ways: first, like
13677 @code{whatis}, it does not print a detailed description; second, it
13678 lists all source files where a type is defined.
13679
13680 @kindex info scope
13681 @cindex local variables
13682 @item info scope @var{location}
13683 List all the variables local to a particular scope. This command
13684 accepts a @var{location} argument---a function name, a source line, or
13685 an address preceded by a @samp{*}, and prints all the variables local
13686 to the scope defined by that location. (@xref{Specify Location}, for
13687 details about supported forms of @var{location}.) For example:
13688
13689 @smallexample
13690 (@value{GDBP}) @b{info scope command_line_handler}
13691 Scope for command_line_handler:
13692 Symbol rl is an argument at stack/frame offset 8, length 4.
13693 Symbol linebuffer is in static storage at address 0x150a18, length 4.
13694 Symbol linelength is in static storage at address 0x150a1c, length 4.
13695 Symbol p is a local variable in register $esi, length 4.
13696 Symbol p1 is a local variable in register $ebx, length 4.
13697 Symbol nline is a local variable in register $edx, length 4.
13698 Symbol repeat is a local variable at frame offset -8, length 4.
13699 @end smallexample
13700
13701 @noindent
13702 This command is especially useful for determining what data to collect
13703 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
13704 collect}.
13705
13706 @kindex info source
13707 @item info source
13708 Show information about the current source file---that is, the source file for
13709 the function containing the current point of execution:
13710 @itemize @bullet
13711 @item
13712 the name of the source file, and the directory containing it,
13713 @item
13714 the directory it was compiled in,
13715 @item
13716 its length, in lines,
13717 @item
13718 which programming language it is written in,
13719 @item
13720 whether the executable includes debugging information for that file, and
13721 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
13722 @item
13723 whether the debugging information includes information about
13724 preprocessor macros.
13725 @end itemize
13726
13727
13728 @kindex info sources
13729 @item info sources
13730 Print the names of all source files in your program for which there is
13731 debugging information, organized into two lists: files whose symbols
13732 have already been read, and files whose symbols will be read when needed.
13733
13734 @kindex info functions
13735 @item info functions
13736 Print the names and data types of all defined functions.
13737
13738 @item info functions @var{regexp}
13739 Print the names and data types of all defined functions
13740 whose names contain a match for regular expression @var{regexp}.
13741 Thus, @samp{info fun step} finds all functions whose names
13742 include @code{step}; @samp{info fun ^step} finds those whose names
13743 start with @code{step}. If a function name contains characters
13744 that conflict with the regular expression language (e.g.@:
13745 @samp{operator*()}), they may be quoted with a backslash.
13746
13747 @kindex info variables
13748 @item info variables
13749 Print the names and data types of all variables that are defined
13750 outside of functions (i.e.@: excluding local variables).
13751
13752 @item info variables @var{regexp}
13753 Print the names and data types of all variables (except for local
13754 variables) whose names contain a match for regular expression
13755 @var{regexp}.
13756
13757 @kindex info classes
13758 @cindex Objective-C, classes and selectors
13759 @item info classes
13760 @itemx info classes @var{regexp}
13761 Display all Objective-C classes in your program, or
13762 (with the @var{regexp} argument) all those matching a particular regular
13763 expression.
13764
13765 @kindex info selectors
13766 @item info selectors
13767 @itemx info selectors @var{regexp}
13768 Display all Objective-C selectors in your program, or
13769 (with the @var{regexp} argument) all those matching a particular regular
13770 expression.
13771
13772 @ignore
13773 This was never implemented.
13774 @kindex info methods
13775 @item info methods
13776 @itemx info methods @var{regexp}
13777 The @code{info methods} command permits the user to examine all defined
13778 methods within C@t{++} program, or (with the @var{regexp} argument) a
13779 specific set of methods found in the various C@t{++} classes. Many
13780 C@t{++} classes provide a large number of methods. Thus, the output
13781 from the @code{ptype} command can be overwhelming and hard to use. The
13782 @code{info-methods} command filters the methods, printing only those
13783 which match the regular-expression @var{regexp}.
13784 @end ignore
13785
13786 @cindex reloading symbols
13787 Some systems allow individual object files that make up your program to
13788 be replaced without stopping and restarting your program. For example,
13789 in VxWorks you can simply recompile a defective object file and keep on
13790 running. If you are running on one of these systems, you can allow
13791 @value{GDBN} to reload the symbols for automatically relinked modules:
13792
13793 @table @code
13794 @kindex set symbol-reloading
13795 @item set symbol-reloading on
13796 Replace symbol definitions for the corresponding source file when an
13797 object file with a particular name is seen again.
13798
13799 @item set symbol-reloading off
13800 Do not replace symbol definitions when encountering object files of the
13801 same name more than once. This is the default state; if you are not
13802 running on a system that permits automatic relinking of modules, you
13803 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
13804 may discard symbols when linking large programs, that may contain
13805 several modules (from different directories or libraries) with the same
13806 name.
13807
13808 @kindex show symbol-reloading
13809 @item show symbol-reloading
13810 Show the current @code{on} or @code{off} setting.
13811 @end table
13812
13813 @cindex opaque data types
13814 @kindex set opaque-type-resolution
13815 @item set opaque-type-resolution on
13816 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
13817 declared as a pointer to a @code{struct}, @code{class}, or
13818 @code{union}---for example, @code{struct MyType *}---that is used in one
13819 source file although the full declaration of @code{struct MyType} is in
13820 another source file. The default is on.
13821
13822 A change in the setting of this subcommand will not take effect until
13823 the next time symbols for a file are loaded.
13824
13825 @item set opaque-type-resolution off
13826 Tell @value{GDBN} not to resolve opaque types. In this case, the type
13827 is printed as follows:
13828 @smallexample
13829 @{<no data fields>@}
13830 @end smallexample
13831
13832 @kindex show opaque-type-resolution
13833 @item show opaque-type-resolution
13834 Show whether opaque types are resolved or not.
13835
13836 @kindex maint print symbols
13837 @cindex symbol dump
13838 @kindex maint print psymbols
13839 @cindex partial symbol dump
13840 @item maint print symbols @var{filename}
13841 @itemx maint print psymbols @var{filename}
13842 @itemx maint print msymbols @var{filename}
13843 Write a dump of debugging symbol data into the file @var{filename}.
13844 These commands are used to debug the @value{GDBN} symbol-reading code. Only
13845 symbols with debugging data are included. If you use @samp{maint print
13846 symbols}, @value{GDBN} includes all the symbols for which it has already
13847 collected full details: that is, @var{filename} reflects symbols for
13848 only those files whose symbols @value{GDBN} has read. You can use the
13849 command @code{info sources} to find out which files these are. If you
13850 use @samp{maint print psymbols} instead, the dump shows information about
13851 symbols that @value{GDBN} only knows partially---that is, symbols defined in
13852 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
13853 @samp{maint print msymbols} dumps just the minimal symbol information
13854 required for each object file from which @value{GDBN} has read some symbols.
13855 @xref{Files, ,Commands to Specify Files}, for a discussion of how
13856 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
13857
13858 @kindex maint info symtabs
13859 @kindex maint info psymtabs
13860 @cindex listing @value{GDBN}'s internal symbol tables
13861 @cindex symbol tables, listing @value{GDBN}'s internal
13862 @cindex full symbol tables, listing @value{GDBN}'s internal
13863 @cindex partial symbol tables, listing @value{GDBN}'s internal
13864 @item maint info symtabs @r{[} @var{regexp} @r{]}
13865 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
13866
13867 List the @code{struct symtab} or @code{struct partial_symtab}
13868 structures whose names match @var{regexp}. If @var{regexp} is not
13869 given, list them all. The output includes expressions which you can
13870 copy into a @value{GDBN} debugging this one to examine a particular
13871 structure in more detail. For example:
13872
13873 @smallexample
13874 (@value{GDBP}) maint info psymtabs dwarf2read
13875 @{ objfile /home/gnu/build/gdb/gdb
13876 ((struct objfile *) 0x82e69d0)
13877 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
13878 ((struct partial_symtab *) 0x8474b10)
13879 readin no
13880 fullname (null)
13881 text addresses 0x814d3c8 -- 0x8158074
13882 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
13883 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
13884 dependencies (none)
13885 @}
13886 @}
13887 (@value{GDBP}) maint info symtabs
13888 (@value{GDBP})
13889 @end smallexample
13890 @noindent
13891 We see that there is one partial symbol table whose filename contains
13892 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
13893 and we see that @value{GDBN} has not read in any symtabs yet at all.
13894 If we set a breakpoint on a function, that will cause @value{GDBN} to
13895 read the symtab for the compilation unit containing that function:
13896
13897 @smallexample
13898 (@value{GDBP}) break dwarf2_psymtab_to_symtab
13899 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
13900 line 1574.
13901 (@value{GDBP}) maint info symtabs
13902 @{ objfile /home/gnu/build/gdb/gdb
13903 ((struct objfile *) 0x82e69d0)
13904 @{ symtab /home/gnu/src/gdb/dwarf2read.c
13905 ((struct symtab *) 0x86c1f38)
13906 dirname (null)
13907 fullname (null)
13908 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
13909 linetable ((struct linetable *) 0x8370fa0)
13910 debugformat DWARF 2
13911 @}
13912 @}
13913 (@value{GDBP})
13914 @end smallexample
13915 @end table
13916
13917
13918 @node Altering
13919 @chapter Altering Execution
13920
13921 Once you think you have found an error in your program, you might want to
13922 find out for certain whether correcting the apparent error would lead to
13923 correct results in the rest of the run. You can find the answer by
13924 experiment, using the @value{GDBN} features for altering execution of the
13925 program.
13926
13927 For example, you can store new values into variables or memory
13928 locations, give your program a signal, restart it at a different
13929 address, or even return prematurely from a function.
13930
13931 @menu
13932 * Assignment:: Assignment to variables
13933 * Jumping:: Continuing at a different address
13934 * Signaling:: Giving your program a signal
13935 * Returning:: Returning from a function
13936 * Calling:: Calling your program's functions
13937 * Patching:: Patching your program
13938 @end menu
13939
13940 @node Assignment
13941 @section Assignment to Variables
13942
13943 @cindex assignment
13944 @cindex setting variables
13945 To alter the value of a variable, evaluate an assignment expression.
13946 @xref{Expressions, ,Expressions}. For example,
13947
13948 @smallexample
13949 print x=4
13950 @end smallexample
13951
13952 @noindent
13953 stores the value 4 into the variable @code{x}, and then prints the
13954 value of the assignment expression (which is 4).
13955 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
13956 information on operators in supported languages.
13957
13958 @kindex set variable
13959 @cindex variables, setting
13960 If you are not interested in seeing the value of the assignment, use the
13961 @code{set} command instead of the @code{print} command. @code{set} is
13962 really the same as @code{print} except that the expression's value is
13963 not printed and is not put in the value history (@pxref{Value History,
13964 ,Value History}). The expression is evaluated only for its effects.
13965
13966 If the beginning of the argument string of the @code{set} command
13967 appears identical to a @code{set} subcommand, use the @code{set
13968 variable} command instead of just @code{set}. This command is identical
13969 to @code{set} except for its lack of subcommands. For example, if your
13970 program has a variable @code{width}, you get an error if you try to set
13971 a new value with just @samp{set width=13}, because @value{GDBN} has the
13972 command @code{set width}:
13973
13974 @smallexample
13975 (@value{GDBP}) whatis width
13976 type = double
13977 (@value{GDBP}) p width
13978 $4 = 13
13979 (@value{GDBP}) set width=47
13980 Invalid syntax in expression.
13981 @end smallexample
13982
13983 @noindent
13984 The invalid expression, of course, is @samp{=47}. In
13985 order to actually set the program's variable @code{width}, use
13986
13987 @smallexample
13988 (@value{GDBP}) set var width=47
13989 @end smallexample
13990
13991 Because the @code{set} command has many subcommands that can conflict
13992 with the names of program variables, it is a good idea to use the
13993 @code{set variable} command instead of just @code{set}. For example, if
13994 your program has a variable @code{g}, you run into problems if you try
13995 to set a new value with just @samp{set g=4}, because @value{GDBN} has
13996 the command @code{set gnutarget}, abbreviated @code{set g}:
13997
13998 @smallexample
13999 @group
14000 (@value{GDBP}) whatis g
14001 type = double
14002 (@value{GDBP}) p g
14003 $1 = 1
14004 (@value{GDBP}) set g=4
14005 (@value{GDBP}) p g
14006 $2 = 1
14007 (@value{GDBP}) r
14008 The program being debugged has been started already.
14009 Start it from the beginning? (y or n) y
14010 Starting program: /home/smith/cc_progs/a.out
14011 "/home/smith/cc_progs/a.out": can't open to read symbols:
14012 Invalid bfd target.
14013 (@value{GDBP}) show g
14014 The current BFD target is "=4".
14015 @end group
14016 @end smallexample
14017
14018 @noindent
14019 The program variable @code{g} did not change, and you silently set the
14020 @code{gnutarget} to an invalid value. In order to set the variable
14021 @code{g}, use
14022
14023 @smallexample
14024 (@value{GDBP}) set var g=4
14025 @end smallexample
14026
14027 @value{GDBN} allows more implicit conversions in assignments than C; you can
14028 freely store an integer value into a pointer variable or vice versa,
14029 and you can convert any structure to any other structure that is the
14030 same length or shorter.
14031 @comment FIXME: how do structs align/pad in these conversions?
14032 @comment /doc@cygnus.com 18dec1990
14033
14034 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
14035 construct to generate a value of specified type at a specified address
14036 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
14037 to memory location @code{0x83040} as an integer (which implies a certain size
14038 and representation in memory), and
14039
14040 @smallexample
14041 set @{int@}0x83040 = 4
14042 @end smallexample
14043
14044 @noindent
14045 stores the value 4 into that memory location.
14046
14047 @node Jumping
14048 @section Continuing at a Different Address
14049
14050 Ordinarily, when you continue your program, you do so at the place where
14051 it stopped, with the @code{continue} command. You can instead continue at
14052 an address of your own choosing, with the following commands:
14053
14054 @table @code
14055 @kindex jump
14056 @item jump @var{linespec}
14057 @itemx jump @var{location}
14058 Resume execution at line @var{linespec} or at address given by
14059 @var{location}. Execution stops again immediately if there is a
14060 breakpoint there. @xref{Specify Location}, for a description of the
14061 different forms of @var{linespec} and @var{location}. It is common
14062 practice to use the @code{tbreak} command in conjunction with
14063 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
14064
14065 The @code{jump} command does not change the current stack frame, or
14066 the stack pointer, or the contents of any memory location or any
14067 register other than the program counter. If line @var{linespec} is in
14068 a different function from the one currently executing, the results may
14069 be bizarre if the two functions expect different patterns of arguments or
14070 of local variables. For this reason, the @code{jump} command requests
14071 confirmation if the specified line is not in the function currently
14072 executing. However, even bizarre results are predictable if you are
14073 well acquainted with the machine-language code of your program.
14074 @end table
14075
14076 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
14077 On many systems, you can get much the same effect as the @code{jump}
14078 command by storing a new value into the register @code{$pc}. The
14079 difference is that this does not start your program running; it only
14080 changes the address of where it @emph{will} run when you continue. For
14081 example,
14082
14083 @smallexample
14084 set $pc = 0x485
14085 @end smallexample
14086
14087 @noindent
14088 makes the next @code{continue} command or stepping command execute at
14089 address @code{0x485}, rather than at the address where your program stopped.
14090 @xref{Continuing and Stepping, ,Continuing and Stepping}.
14091
14092 The most common occasion to use the @code{jump} command is to back
14093 up---perhaps with more breakpoints set---over a portion of a program
14094 that has already executed, in order to examine its execution in more
14095 detail.
14096
14097 @c @group
14098 @node Signaling
14099 @section Giving your Program a Signal
14100 @cindex deliver a signal to a program
14101
14102 @table @code
14103 @kindex signal
14104 @item signal @var{signal}
14105 Resume execution where your program stopped, but immediately give it the
14106 signal @var{signal}. @var{signal} can be the name or the number of a
14107 signal. For example, on many systems @code{signal 2} and @code{signal
14108 SIGINT} are both ways of sending an interrupt signal.
14109
14110 Alternatively, if @var{signal} is zero, continue execution without
14111 giving a signal. This is useful when your program stopped on account of
14112 a signal and would ordinary see the signal when resumed with the
14113 @code{continue} command; @samp{signal 0} causes it to resume without a
14114 signal.
14115
14116 @code{signal} does not repeat when you press @key{RET} a second time
14117 after executing the command.
14118 @end table
14119 @c @end group
14120
14121 Invoking the @code{signal} command is not the same as invoking the
14122 @code{kill} utility from the shell. Sending a signal with @code{kill}
14123 causes @value{GDBN} to decide what to do with the signal depending on
14124 the signal handling tables (@pxref{Signals}). The @code{signal} command
14125 passes the signal directly to your program.
14126
14127
14128 @node Returning
14129 @section Returning from a Function
14130
14131 @table @code
14132 @cindex returning from a function
14133 @kindex return
14134 @item return
14135 @itemx return @var{expression}
14136 You can cancel execution of a function call with the @code{return}
14137 command. If you give an
14138 @var{expression} argument, its value is used as the function's return
14139 value.
14140 @end table
14141
14142 When you use @code{return}, @value{GDBN} discards the selected stack frame
14143 (and all frames within it). You can think of this as making the
14144 discarded frame return prematurely. If you wish to specify a value to
14145 be returned, give that value as the argument to @code{return}.
14146
14147 This pops the selected stack frame (@pxref{Selection, ,Selecting a
14148 Frame}), and any other frames inside of it, leaving its caller as the
14149 innermost remaining frame. That frame becomes selected. The
14150 specified value is stored in the registers used for returning values
14151 of functions.
14152
14153 The @code{return} command does not resume execution; it leaves the
14154 program stopped in the state that would exist if the function had just
14155 returned. In contrast, the @code{finish} command (@pxref{Continuing
14156 and Stepping, ,Continuing and Stepping}) resumes execution until the
14157 selected stack frame returns naturally.
14158
14159 @value{GDBN} needs to know how the @var{expression} argument should be set for
14160 the inferior. The concrete registers assignment depends on the OS ABI and the
14161 type being returned by the selected stack frame. For example it is common for
14162 OS ABI to return floating point values in FPU registers while integer values in
14163 CPU registers. Still some ABIs return even floating point values in CPU
14164 registers. Larger integer widths (such as @code{long long int}) also have
14165 specific placement rules. @value{GDBN} already knows the OS ABI from its
14166 current target so it needs to find out also the type being returned to make the
14167 assignment into the right register(s).
14168
14169 Normally, the selected stack frame has debug info. @value{GDBN} will always
14170 use the debug info instead of the implicit type of @var{expression} when the
14171 debug info is available. For example, if you type @kbd{return -1}, and the
14172 function in the current stack frame is declared to return a @code{long long
14173 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
14174 into a @code{long long int}:
14175
14176 @smallexample
14177 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
14178 29 return 31;
14179 (@value{GDBP}) return -1
14180 Make func return now? (y or n) y
14181 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
14182 43 printf ("result=%lld\n", func ());
14183 (@value{GDBP})
14184 @end smallexample
14185
14186 However, if the selected stack frame does not have a debug info, e.g., if the
14187 function was compiled without debug info, @value{GDBN} has to find out the type
14188 to return from user. Specifying a different type by mistake may set the value
14189 in different inferior registers than the caller code expects. For example,
14190 typing @kbd{return -1} with its implicit type @code{int} would set only a part
14191 of a @code{long long int} result for a debug info less function (on 32-bit
14192 architectures). Therefore the user is required to specify the return type by
14193 an appropriate cast explicitly:
14194
14195 @smallexample
14196 Breakpoint 2, 0x0040050b in func ()
14197 (@value{GDBP}) return -1
14198 Return value type not available for selected stack frame.
14199 Please use an explicit cast of the value to return.
14200 (@value{GDBP}) return (long long int) -1
14201 Make selected stack frame return now? (y or n) y
14202 #0 0x00400526 in main ()
14203 (@value{GDBP})
14204 @end smallexample
14205
14206 @node Calling
14207 @section Calling Program Functions
14208
14209 @table @code
14210 @cindex calling functions
14211 @cindex inferior functions, calling
14212 @item print @var{expr}
14213 Evaluate the expression @var{expr} and display the resulting value.
14214 @var{expr} may include calls to functions in the program being
14215 debugged.
14216
14217 @kindex call
14218 @item call @var{expr}
14219 Evaluate the expression @var{expr} without displaying @code{void}
14220 returned values.
14221
14222 You can use this variant of the @code{print} command if you want to
14223 execute a function from your program that does not return anything
14224 (a.k.a.@: @dfn{a void function}), but without cluttering the output
14225 with @code{void} returned values that @value{GDBN} will otherwise
14226 print. If the result is not void, it is printed and saved in the
14227 value history.
14228 @end table
14229
14230 It is possible for the function you call via the @code{print} or
14231 @code{call} command to generate a signal (e.g., if there's a bug in
14232 the function, or if you passed it incorrect arguments). What happens
14233 in that case is controlled by the @code{set unwindonsignal} command.
14234
14235 Similarly, with a C@t{++} program it is possible for the function you
14236 call via the @code{print} or @code{call} command to generate an
14237 exception that is not handled due to the constraints of the dummy
14238 frame. In this case, any exception that is raised in the frame, but has
14239 an out-of-frame exception handler will not be found. GDB builds a
14240 dummy-frame for the inferior function call, and the unwinder cannot
14241 seek for exception handlers outside of this dummy-frame. What happens
14242 in that case is controlled by the
14243 @code{set unwind-on-terminating-exception} command.
14244
14245 @table @code
14246 @item set unwindonsignal
14247 @kindex set unwindonsignal
14248 @cindex unwind stack in called functions
14249 @cindex call dummy stack unwinding
14250 Set unwinding of the stack if a signal is received while in a function
14251 that @value{GDBN} called in the program being debugged. If set to on,
14252 @value{GDBN} unwinds the stack it created for the call and restores
14253 the context to what it was before the call. If set to off (the
14254 default), @value{GDBN} stops in the frame where the signal was
14255 received.
14256
14257 @item show unwindonsignal
14258 @kindex show unwindonsignal
14259 Show the current setting of stack unwinding in the functions called by
14260 @value{GDBN}.
14261
14262 @item set unwind-on-terminating-exception
14263 @kindex set unwind-on-terminating-exception
14264 @cindex unwind stack in called functions with unhandled exceptions
14265 @cindex call dummy stack unwinding on unhandled exception.
14266 Set unwinding of the stack if a C@t{++} exception is raised, but left
14267 unhandled while in a function that @value{GDBN} called in the program being
14268 debugged. If set to on (the default), @value{GDBN} unwinds the stack
14269 it created for the call and restores the context to what it was before
14270 the call. If set to off, @value{GDBN} the exception is delivered to
14271 the default C@t{++} exception handler and the inferior terminated.
14272
14273 @item show unwind-on-terminating-exception
14274 @kindex show unwind-on-terminating-exception
14275 Show the current setting of stack unwinding in the functions called by
14276 @value{GDBN}.
14277
14278 @end table
14279
14280 @cindex weak alias functions
14281 Sometimes, a function you wish to call is actually a @dfn{weak alias}
14282 for another function. In such case, @value{GDBN} might not pick up
14283 the type information, including the types of the function arguments,
14284 which causes @value{GDBN} to call the inferior function incorrectly.
14285 As a result, the called function will function erroneously and may
14286 even crash. A solution to that is to use the name of the aliased
14287 function instead.
14288
14289 @node Patching
14290 @section Patching Programs
14291
14292 @cindex patching binaries
14293 @cindex writing into executables
14294 @cindex writing into corefiles
14295
14296 By default, @value{GDBN} opens the file containing your program's
14297 executable code (or the corefile) read-only. This prevents accidental
14298 alterations to machine code; but it also prevents you from intentionally
14299 patching your program's binary.
14300
14301 If you'd like to be able to patch the binary, you can specify that
14302 explicitly with the @code{set write} command. For example, you might
14303 want to turn on internal debugging flags, or even to make emergency
14304 repairs.
14305
14306 @table @code
14307 @kindex set write
14308 @item set write on
14309 @itemx set write off
14310 If you specify @samp{set write on}, @value{GDBN} opens executable and
14311 core files for both reading and writing; if you specify @kbd{set write
14312 off} (the default), @value{GDBN} opens them read-only.
14313
14314 If you have already loaded a file, you must load it again (using the
14315 @code{exec-file} or @code{core-file} command) after changing @code{set
14316 write}, for your new setting to take effect.
14317
14318 @item show write
14319 @kindex show write
14320 Display whether executable files and core files are opened for writing
14321 as well as reading.
14322 @end table
14323
14324 @node GDB Files
14325 @chapter @value{GDBN} Files
14326
14327 @value{GDBN} needs to know the file name of the program to be debugged,
14328 both in order to read its symbol table and in order to start your
14329 program. To debug a core dump of a previous run, you must also tell
14330 @value{GDBN} the name of the core dump file.
14331
14332 @menu
14333 * Files:: Commands to specify files
14334 * Separate Debug Files:: Debugging information in separate files
14335 * Index Files:: Index files speed up GDB
14336 * Symbol Errors:: Errors reading symbol files
14337 * Data Files:: GDB data files
14338 @end menu
14339
14340 @node Files
14341 @section Commands to Specify Files
14342
14343 @cindex symbol table
14344 @cindex core dump file
14345
14346 You may want to specify executable and core dump file names. The usual
14347 way to do this is at start-up time, using the arguments to
14348 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
14349 Out of @value{GDBN}}).
14350
14351 Occasionally it is necessary to change to a different file during a
14352 @value{GDBN} session. Or you may run @value{GDBN} and forget to
14353 specify a file you want to use. Or you are debugging a remote target
14354 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
14355 Program}). In these situations the @value{GDBN} commands to specify
14356 new files are useful.
14357
14358 @table @code
14359 @cindex executable file
14360 @kindex file
14361 @item file @var{filename}
14362 Use @var{filename} as the program to be debugged. It is read for its
14363 symbols and for the contents of pure memory. It is also the program
14364 executed when you use the @code{run} command. If you do not specify a
14365 directory and the file is not found in the @value{GDBN} working directory,
14366 @value{GDBN} uses the environment variable @code{PATH} as a list of
14367 directories to search, just as the shell does when looking for a program
14368 to run. You can change the value of this variable, for both @value{GDBN}
14369 and your program, using the @code{path} command.
14370
14371 @cindex unlinked object files
14372 @cindex patching object files
14373 You can load unlinked object @file{.o} files into @value{GDBN} using
14374 the @code{file} command. You will not be able to ``run'' an object
14375 file, but you can disassemble functions and inspect variables. Also,
14376 if the underlying BFD functionality supports it, you could use
14377 @kbd{gdb -write} to patch object files using this technique. Note
14378 that @value{GDBN} can neither interpret nor modify relocations in this
14379 case, so branches and some initialized variables will appear to go to
14380 the wrong place. But this feature is still handy from time to time.
14381
14382 @item file
14383 @code{file} with no argument makes @value{GDBN} discard any information it
14384 has on both executable file and the symbol table.
14385
14386 @kindex exec-file
14387 @item exec-file @r{[} @var{filename} @r{]}
14388 Specify that the program to be run (but not the symbol table) is found
14389 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
14390 if necessary to locate your program. Omitting @var{filename} means to
14391 discard information on the executable file.
14392
14393 @kindex symbol-file
14394 @item symbol-file @r{[} @var{filename} @r{]}
14395 Read symbol table information from file @var{filename}. @code{PATH} is
14396 searched when necessary. Use the @code{file} command to get both symbol
14397 table and program to run from the same file.
14398
14399 @code{symbol-file} with no argument clears out @value{GDBN} information on your
14400 program's symbol table.
14401
14402 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
14403 some breakpoints and auto-display expressions. This is because they may
14404 contain pointers to the internal data recording symbols and data types,
14405 which are part of the old symbol table data being discarded inside
14406 @value{GDBN}.
14407
14408 @code{symbol-file} does not repeat if you press @key{RET} again after
14409 executing it once.
14410
14411 When @value{GDBN} is configured for a particular environment, it
14412 understands debugging information in whatever format is the standard
14413 generated for that environment; you may use either a @sc{gnu} compiler, or
14414 other compilers that adhere to the local conventions.
14415 Best results are usually obtained from @sc{gnu} compilers; for example,
14416 using @code{@value{NGCC}} you can generate debugging information for
14417 optimized code.
14418
14419 For most kinds of object files, with the exception of old SVR3 systems
14420 using COFF, the @code{symbol-file} command does not normally read the
14421 symbol table in full right away. Instead, it scans the symbol table
14422 quickly to find which source files and which symbols are present. The
14423 details are read later, one source file at a time, as they are needed.
14424
14425 The purpose of this two-stage reading strategy is to make @value{GDBN}
14426 start up faster. For the most part, it is invisible except for
14427 occasional pauses while the symbol table details for a particular source
14428 file are being read. (The @code{set verbose} command can turn these
14429 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
14430 Warnings and Messages}.)
14431
14432 We have not implemented the two-stage strategy for COFF yet. When the
14433 symbol table is stored in COFF format, @code{symbol-file} reads the
14434 symbol table data in full right away. Note that ``stabs-in-COFF''
14435 still does the two-stage strategy, since the debug info is actually
14436 in stabs format.
14437
14438 @kindex readnow
14439 @cindex reading symbols immediately
14440 @cindex symbols, reading immediately
14441 @item symbol-file @r{[} -readnow @r{]} @var{filename}
14442 @itemx file @r{[} -readnow @r{]} @var{filename}
14443 You can override the @value{GDBN} two-stage strategy for reading symbol
14444 tables by using the @samp{-readnow} option with any of the commands that
14445 load symbol table information, if you want to be sure @value{GDBN} has the
14446 entire symbol table available.
14447
14448 @c FIXME: for now no mention of directories, since this seems to be in
14449 @c flux. 13mar1992 status is that in theory GDB would look either in
14450 @c current dir or in same dir as myprog; but issues like competing
14451 @c GDB's, or clutter in system dirs, mean that in practice right now
14452 @c only current dir is used. FFish says maybe a special GDB hierarchy
14453 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
14454 @c files.
14455
14456 @kindex core-file
14457 @item core-file @r{[}@var{filename}@r{]}
14458 @itemx core
14459 Specify the whereabouts of a core dump file to be used as the ``contents
14460 of memory''. Traditionally, core files contain only some parts of the
14461 address space of the process that generated them; @value{GDBN} can access the
14462 executable file itself for other parts.
14463
14464 @code{core-file} with no argument specifies that no core file is
14465 to be used.
14466
14467 Note that the core file is ignored when your program is actually running
14468 under @value{GDBN}. So, if you have been running your program and you
14469 wish to debug a core file instead, you must kill the subprocess in which
14470 the program is running. To do this, use the @code{kill} command
14471 (@pxref{Kill Process, ,Killing the Child Process}).
14472
14473 @kindex add-symbol-file
14474 @cindex dynamic linking
14475 @item add-symbol-file @var{filename} @var{address}
14476 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
14477 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
14478 The @code{add-symbol-file} command reads additional symbol table
14479 information from the file @var{filename}. You would use this command
14480 when @var{filename} has been dynamically loaded (by some other means)
14481 into the program that is running. @var{address} should be the memory
14482 address at which the file has been loaded; @value{GDBN} cannot figure
14483 this out for itself. You can additionally specify an arbitrary number
14484 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
14485 section name and base address for that section. You can specify any
14486 @var{address} as an expression.
14487
14488 The symbol table of the file @var{filename} is added to the symbol table
14489 originally read with the @code{symbol-file} command. You can use the
14490 @code{add-symbol-file} command any number of times; the new symbol data
14491 thus read keeps adding to the old. To discard all old symbol data
14492 instead, use the @code{symbol-file} command without any arguments.
14493
14494 @cindex relocatable object files, reading symbols from
14495 @cindex object files, relocatable, reading symbols from
14496 @cindex reading symbols from relocatable object files
14497 @cindex symbols, reading from relocatable object files
14498 @cindex @file{.o} files, reading symbols from
14499 Although @var{filename} is typically a shared library file, an
14500 executable file, or some other object file which has been fully
14501 relocated for loading into a process, you can also load symbolic
14502 information from relocatable @file{.o} files, as long as:
14503
14504 @itemize @bullet
14505 @item
14506 the file's symbolic information refers only to linker symbols defined in
14507 that file, not to symbols defined by other object files,
14508 @item
14509 every section the file's symbolic information refers to has actually
14510 been loaded into the inferior, as it appears in the file, and
14511 @item
14512 you can determine the address at which every section was loaded, and
14513 provide these to the @code{add-symbol-file} command.
14514 @end itemize
14515
14516 @noindent
14517 Some embedded operating systems, like Sun Chorus and VxWorks, can load
14518 relocatable files into an already running program; such systems
14519 typically make the requirements above easy to meet. However, it's
14520 important to recognize that many native systems use complex link
14521 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
14522 assembly, for example) that make the requirements difficult to meet. In
14523 general, one cannot assume that using @code{add-symbol-file} to read a
14524 relocatable object file's symbolic information will have the same effect
14525 as linking the relocatable object file into the program in the normal
14526 way.
14527
14528 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
14529
14530 @kindex add-symbol-file-from-memory
14531 @cindex @code{syscall DSO}
14532 @cindex load symbols from memory
14533 @item add-symbol-file-from-memory @var{address}
14534 Load symbols from the given @var{address} in a dynamically loaded
14535 object file whose image is mapped directly into the inferior's memory.
14536 For example, the Linux kernel maps a @code{syscall DSO} into each
14537 process's address space; this DSO provides kernel-specific code for
14538 some system calls. The argument can be any expression whose
14539 evaluation yields the address of the file's shared object file header.
14540 For this command to work, you must have used @code{symbol-file} or
14541 @code{exec-file} commands in advance.
14542
14543 @kindex add-shared-symbol-files
14544 @kindex assf
14545 @item add-shared-symbol-files @var{library-file}
14546 @itemx assf @var{library-file}
14547 The @code{add-shared-symbol-files} command can currently be used only
14548 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
14549 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
14550 @value{GDBN} automatically looks for shared libraries, however if
14551 @value{GDBN} does not find yours, you can invoke
14552 @code{add-shared-symbol-files}. It takes one argument: the shared
14553 library's file name. @code{assf} is a shorthand alias for
14554 @code{add-shared-symbol-files}.
14555
14556 @kindex section
14557 @item section @var{section} @var{addr}
14558 The @code{section} command changes the base address of the named
14559 @var{section} of the exec file to @var{addr}. This can be used if the
14560 exec file does not contain section addresses, (such as in the
14561 @code{a.out} format), or when the addresses specified in the file
14562 itself are wrong. Each section must be changed separately. The
14563 @code{info files} command, described below, lists all the sections and
14564 their addresses.
14565
14566 @kindex info files
14567 @kindex info target
14568 @item info files
14569 @itemx info target
14570 @code{info files} and @code{info target} are synonymous; both print the
14571 current target (@pxref{Targets, ,Specifying a Debugging Target}),
14572 including the names of the executable and core dump files currently in
14573 use by @value{GDBN}, and the files from which symbols were loaded. The
14574 command @code{help target} lists all possible targets rather than
14575 current ones.
14576
14577 @kindex maint info sections
14578 @item maint info sections
14579 Another command that can give you extra information about program sections
14580 is @code{maint info sections}. In addition to the section information
14581 displayed by @code{info files}, this command displays the flags and file
14582 offset of each section in the executable and core dump files. In addition,
14583 @code{maint info sections} provides the following command options (which
14584 may be arbitrarily combined):
14585
14586 @table @code
14587 @item ALLOBJ
14588 Display sections for all loaded object files, including shared libraries.
14589 @item @var{sections}
14590 Display info only for named @var{sections}.
14591 @item @var{section-flags}
14592 Display info only for sections for which @var{section-flags} are true.
14593 The section flags that @value{GDBN} currently knows about are:
14594 @table @code
14595 @item ALLOC
14596 Section will have space allocated in the process when loaded.
14597 Set for all sections except those containing debug information.
14598 @item LOAD
14599 Section will be loaded from the file into the child process memory.
14600 Set for pre-initialized code and data, clear for @code{.bss} sections.
14601 @item RELOC
14602 Section needs to be relocated before loading.
14603 @item READONLY
14604 Section cannot be modified by the child process.
14605 @item CODE
14606 Section contains executable code only.
14607 @item DATA
14608 Section contains data only (no executable code).
14609 @item ROM
14610 Section will reside in ROM.
14611 @item CONSTRUCTOR
14612 Section contains data for constructor/destructor lists.
14613 @item HAS_CONTENTS
14614 Section is not empty.
14615 @item NEVER_LOAD
14616 An instruction to the linker to not output the section.
14617 @item COFF_SHARED_LIBRARY
14618 A notification to the linker that the section contains
14619 COFF shared library information.
14620 @item IS_COMMON
14621 Section contains common symbols.
14622 @end table
14623 @end table
14624 @kindex set trust-readonly-sections
14625 @cindex read-only sections
14626 @item set trust-readonly-sections on
14627 Tell @value{GDBN} that readonly sections in your object file
14628 really are read-only (i.e.@: that their contents will not change).
14629 In that case, @value{GDBN} can fetch values from these sections
14630 out of the object file, rather than from the target program.
14631 For some targets (notably embedded ones), this can be a significant
14632 enhancement to debugging performance.
14633
14634 The default is off.
14635
14636 @item set trust-readonly-sections off
14637 Tell @value{GDBN} not to trust readonly sections. This means that
14638 the contents of the section might change while the program is running,
14639 and must therefore be fetched from the target when needed.
14640
14641 @item show trust-readonly-sections
14642 Show the current setting of trusting readonly sections.
14643 @end table
14644
14645 All file-specifying commands allow both absolute and relative file names
14646 as arguments. @value{GDBN} always converts the file name to an absolute file
14647 name and remembers it that way.
14648
14649 @cindex shared libraries
14650 @anchor{Shared Libraries}
14651 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
14652 and IBM RS/6000 AIX shared libraries.
14653
14654 On MS-Windows @value{GDBN} must be linked with the Expat library to support
14655 shared libraries. @xref{Expat}.
14656
14657 @value{GDBN} automatically loads symbol definitions from shared libraries
14658 when you use the @code{run} command, or when you examine a core file.
14659 (Before you issue the @code{run} command, @value{GDBN} does not understand
14660 references to a function in a shared library, however---unless you are
14661 debugging a core file).
14662
14663 On HP-UX, if the program loads a library explicitly, @value{GDBN}
14664 automatically loads the symbols at the time of the @code{shl_load} call.
14665
14666 @c FIXME: some @value{GDBN} release may permit some refs to undef
14667 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
14668 @c FIXME...lib; check this from time to time when updating manual
14669
14670 There are times, however, when you may wish to not automatically load
14671 symbol definitions from shared libraries, such as when they are
14672 particularly large or there are many of them.
14673
14674 To control the automatic loading of shared library symbols, use the
14675 commands:
14676
14677 @table @code
14678 @kindex set auto-solib-add
14679 @item set auto-solib-add @var{mode}
14680 If @var{mode} is @code{on}, symbols from all shared object libraries
14681 will be loaded automatically when the inferior begins execution, you
14682 attach to an independently started inferior, or when the dynamic linker
14683 informs @value{GDBN} that a new library has been loaded. If @var{mode}
14684 is @code{off}, symbols must be loaded manually, using the
14685 @code{sharedlibrary} command. The default value is @code{on}.
14686
14687 @cindex memory used for symbol tables
14688 If your program uses lots of shared libraries with debug info that
14689 takes large amounts of memory, you can decrease the @value{GDBN}
14690 memory footprint by preventing it from automatically loading the
14691 symbols from shared libraries. To that end, type @kbd{set
14692 auto-solib-add off} before running the inferior, then load each
14693 library whose debug symbols you do need with @kbd{sharedlibrary
14694 @var{regexp}}, where @var{regexp} is a regular expression that matches
14695 the libraries whose symbols you want to be loaded.
14696
14697 @kindex show auto-solib-add
14698 @item show auto-solib-add
14699 Display the current autoloading mode.
14700 @end table
14701
14702 @cindex load shared library
14703 To explicitly load shared library symbols, use the @code{sharedlibrary}
14704 command:
14705
14706 @table @code
14707 @kindex info sharedlibrary
14708 @kindex info share
14709 @item info share @var{regex}
14710 @itemx info sharedlibrary @var{regex}
14711 Print the names of the shared libraries which are currently loaded
14712 that match @var{regex}. If @var{regex} is omitted then print
14713 all shared libraries that are loaded.
14714
14715 @kindex sharedlibrary
14716 @kindex share
14717 @item sharedlibrary @var{regex}
14718 @itemx share @var{regex}
14719 Load shared object library symbols for files matching a
14720 Unix regular expression.
14721 As with files loaded automatically, it only loads shared libraries
14722 required by your program for a core file or after typing @code{run}. If
14723 @var{regex} is omitted all shared libraries required by your program are
14724 loaded.
14725
14726 @item nosharedlibrary
14727 @kindex nosharedlibrary
14728 @cindex unload symbols from shared libraries
14729 Unload all shared object library symbols. This discards all symbols
14730 that have been loaded from all shared libraries. Symbols from shared
14731 libraries that were loaded by explicit user requests are not
14732 discarded.
14733 @end table
14734
14735 Sometimes you may wish that @value{GDBN} stops and gives you control
14736 when any of shared library events happen. Use the @code{set
14737 stop-on-solib-events} command for this:
14738
14739 @table @code
14740 @item set stop-on-solib-events
14741 @kindex set stop-on-solib-events
14742 This command controls whether @value{GDBN} should give you control
14743 when the dynamic linker notifies it about some shared library event.
14744 The most common event of interest is loading or unloading of a new
14745 shared library.
14746
14747 @item show stop-on-solib-events
14748 @kindex show stop-on-solib-events
14749 Show whether @value{GDBN} stops and gives you control when shared
14750 library events happen.
14751 @end table
14752
14753 Shared libraries are also supported in many cross or remote debugging
14754 configurations. @value{GDBN} needs to have access to the target's libraries;
14755 this can be accomplished either by providing copies of the libraries
14756 on the host system, or by asking @value{GDBN} to automatically retrieve the
14757 libraries from the target. If copies of the target libraries are
14758 provided, they need to be the same as the target libraries, although the
14759 copies on the target can be stripped as long as the copies on the host are
14760 not.
14761
14762 @cindex where to look for shared libraries
14763 For remote debugging, you need to tell @value{GDBN} where the target
14764 libraries are, so that it can load the correct copies---otherwise, it
14765 may try to load the host's libraries. @value{GDBN} has two variables
14766 to specify the search directories for target libraries.
14767
14768 @table @code
14769 @cindex prefix for shared library file names
14770 @cindex system root, alternate
14771 @kindex set solib-absolute-prefix
14772 @kindex set sysroot
14773 @item set sysroot @var{path}
14774 Use @var{path} as the system root for the program being debugged. Any
14775 absolute shared library paths will be prefixed with @var{path}; many
14776 runtime loaders store the absolute paths to the shared library in the
14777 target program's memory. If you use @code{set sysroot} to find shared
14778 libraries, they need to be laid out in the same way that they are on
14779 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
14780 under @var{path}.
14781
14782 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
14783 retrieve the target libraries from the remote system. This is only
14784 supported when using a remote target that supports the @code{remote get}
14785 command (@pxref{File Transfer,,Sending files to a remote system}).
14786 The part of @var{path} following the initial @file{remote:}
14787 (if present) is used as system root prefix on the remote file system.
14788 @footnote{If you want to specify a local system root using a directory
14789 that happens to be named @file{remote:}, you need to use some equivalent
14790 variant of the name like @file{./remote:}.}
14791
14792 For targets with an MS-DOS based filesystem, such as MS-Windows and
14793 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
14794 absolute file name with @var{path}. But first, on Unix hosts,
14795 @value{GDBN} converts all backslash directory separators into forward
14796 slashes, because the backslash is not a directory separator on Unix:
14797
14798 @smallexample
14799 c:\foo\bar.dll @result{} c:/foo/bar.dll
14800 @end smallexample
14801
14802 Then, @value{GDBN} attempts prefixing the target file name with
14803 @var{path}, and looks for the resulting file name in the host file
14804 system:
14805
14806 @smallexample
14807 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
14808 @end smallexample
14809
14810 If that does not find the shared library, @value{GDBN} tries removing
14811 the @samp{:} character from the drive spec, both for convenience, and,
14812 for the case of the host file system not supporting file names with
14813 colons:
14814
14815 @smallexample
14816 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
14817 @end smallexample
14818
14819 This makes it possible to have a system root that mirrors a target
14820 with more than one drive. E.g., you may want to setup your local
14821 copies of the target system shared libraries like so (note @samp{c} vs
14822 @samp{z}):
14823
14824 @smallexample
14825 @file{/path/to/sysroot/c/sys/bin/foo.dll}
14826 @file{/path/to/sysroot/c/sys/bin/bar.dll}
14827 @file{/path/to/sysroot/z/sys/bin/bar.dll}
14828 @end smallexample
14829
14830 @noindent
14831 and point the system root at @file{/path/to/sysroot}, so that
14832 @value{GDBN} can find the correct copies of both
14833 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
14834
14835 If that still does not find the shared library, @value{GDBN} tries
14836 removing the whole drive spec from the target file name:
14837
14838 @smallexample
14839 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
14840 @end smallexample
14841
14842 This last lookup makes it possible to not care about the drive name,
14843 if you don't want or need to.
14844
14845 The @code{set solib-absolute-prefix} command is an alias for @code{set
14846 sysroot}.
14847
14848 @cindex default system root
14849 @cindex @samp{--with-sysroot}
14850 You can set the default system root by using the configure-time
14851 @samp{--with-sysroot} option. If the system root is inside
14852 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
14853 @samp{--exec-prefix}), then the default system root will be updated
14854 automatically if the installed @value{GDBN} is moved to a new
14855 location.
14856
14857 @kindex show sysroot
14858 @item show sysroot
14859 Display the current shared library prefix.
14860
14861 @kindex set solib-search-path
14862 @item set solib-search-path @var{path}
14863 If this variable is set, @var{path} is a colon-separated list of
14864 directories to search for shared libraries. @samp{solib-search-path}
14865 is used after @samp{sysroot} fails to locate the library, or if the
14866 path to the library is relative instead of absolute. If you want to
14867 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
14868 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
14869 finding your host's libraries. @samp{sysroot} is preferred; setting
14870 it to a nonexistent directory may interfere with automatic loading
14871 of shared library symbols.
14872
14873 @kindex show solib-search-path
14874 @item show solib-search-path
14875 Display the current shared library search path.
14876
14877 @cindex DOS file-name semantics of file names.
14878 @kindex set target-file-system-kind (unix|dos-based|auto)
14879 @kindex show target-file-system-kind
14880 @item set target-file-system-kind @var{kind}
14881 Set assumed file system kind for target reported file names.
14882
14883 Shared library file names as reported by the target system may not
14884 make sense as is on the system @value{GDBN} is running on. For
14885 example, when remote debugging a target that has MS-DOS based file
14886 system semantics, from a Unix host, the target may be reporting to
14887 @value{GDBN} a list of loaded shared libraries with file names such as
14888 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
14889 drive letters, so the @samp{c:\} prefix is not normally understood as
14890 indicating an absolute file name, and neither is the backslash
14891 normally considered a directory separator character. In that case,
14892 the native file system would interpret this whole absolute file name
14893 as a relative file name with no directory components. This would make
14894 it impossible to point @value{GDBN} at a copy of the remote target's
14895 shared libraries on the host using @code{set sysroot}, and impractical
14896 with @code{set solib-search-path}. Setting
14897 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
14898 to interpret such file names similarly to how the target would, and to
14899 map them to file names valid on @value{GDBN}'s native file system
14900 semantics. The value of @var{kind} can be @code{"auto"}, in addition
14901 to one of the supported file system kinds. In that case, @value{GDBN}
14902 tries to determine the appropriate file system variant based on the
14903 current target's operating system (@pxref{ABI, ,Configuring the
14904 Current ABI}). The supported file system settings are:
14905
14906 @table @code
14907 @item unix
14908 Instruct @value{GDBN} to assume the target file system is of Unix
14909 kind. Only file names starting the forward slash (@samp{/}) character
14910 are considered absolute, and the directory separator character is also
14911 the forward slash.
14912
14913 @item dos-based
14914 Instruct @value{GDBN} to assume the target file system is DOS based.
14915 File names starting with either a forward slash, or a drive letter
14916 followed by a colon (e.g., @samp{c:}), are considered absolute, and
14917 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
14918 considered directory separators.
14919
14920 @item auto
14921 Instruct @value{GDBN} to use the file system kind associated with the
14922 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
14923 This is the default.
14924 @end table
14925 @end table
14926
14927
14928 @node Separate Debug Files
14929 @section Debugging Information in Separate Files
14930 @cindex separate debugging information files
14931 @cindex debugging information in separate files
14932 @cindex @file{.debug} subdirectories
14933 @cindex debugging information directory, global
14934 @cindex global debugging information directory
14935 @cindex build ID, and separate debugging files
14936 @cindex @file{.build-id} directory
14937
14938 @value{GDBN} allows you to put a program's debugging information in a
14939 file separate from the executable itself, in a way that allows
14940 @value{GDBN} to find and load the debugging information automatically.
14941 Since debugging information can be very large---sometimes larger
14942 than the executable code itself---some systems distribute debugging
14943 information for their executables in separate files, which users can
14944 install only when they need to debug a problem.
14945
14946 @value{GDBN} supports two ways of specifying the separate debug info
14947 file:
14948
14949 @itemize @bullet
14950 @item
14951 The executable contains a @dfn{debug link} that specifies the name of
14952 the separate debug info file. The separate debug file's name is
14953 usually @file{@var{executable}.debug}, where @var{executable} is the
14954 name of the corresponding executable file without leading directories
14955 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
14956 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
14957 checksum for the debug file, which @value{GDBN} uses to validate that
14958 the executable and the debug file came from the same build.
14959
14960 @item
14961 The executable contains a @dfn{build ID}, a unique bit string that is
14962 also present in the corresponding debug info file. (This is supported
14963 only on some operating systems, notably those which use the ELF format
14964 for binary files and the @sc{gnu} Binutils.) For more details about
14965 this feature, see the description of the @option{--build-id}
14966 command-line option in @ref{Options, , Command Line Options, ld.info,
14967 The GNU Linker}. The debug info file's name is not specified
14968 explicitly by the build ID, but can be computed from the build ID, see
14969 below.
14970 @end itemize
14971
14972 Depending on the way the debug info file is specified, @value{GDBN}
14973 uses two different methods of looking for the debug file:
14974
14975 @itemize @bullet
14976 @item
14977 For the ``debug link'' method, @value{GDBN} looks up the named file in
14978 the directory of the executable file, then in a subdirectory of that
14979 directory named @file{.debug}, and finally under the global debug
14980 directory, in a subdirectory whose name is identical to the leading
14981 directories of the executable's absolute file name.
14982
14983 @item
14984 For the ``build ID'' method, @value{GDBN} looks in the
14985 @file{.build-id} subdirectory of the global debug directory for a file
14986 named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
14987 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
14988 are the rest of the bit string. (Real build ID strings are 32 or more
14989 hex characters, not 10.)
14990 @end itemize
14991
14992 So, for example, suppose you ask @value{GDBN} to debug
14993 @file{/usr/bin/ls}, which has a debug link that specifies the
14994 file @file{ls.debug}, and a build ID whose value in hex is
14995 @code{abcdef1234}. If the global debug directory is
14996 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
14997 debug information files, in the indicated order:
14998
14999 @itemize @minus
15000 @item
15001 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
15002 @item
15003 @file{/usr/bin/ls.debug}
15004 @item
15005 @file{/usr/bin/.debug/ls.debug}
15006 @item
15007 @file{/usr/lib/debug/usr/bin/ls.debug}.
15008 @end itemize
15009
15010 You can set the global debugging info directory's name, and view the
15011 name @value{GDBN} is currently using.
15012
15013 @table @code
15014
15015 @kindex set debug-file-directory
15016 @item set debug-file-directory @var{directories}
15017 Set the directories which @value{GDBN} searches for separate debugging
15018 information files to @var{directory}. Multiple directory components can be set
15019 concatenating them by a directory separator.
15020
15021 @kindex show debug-file-directory
15022 @item show debug-file-directory
15023 Show the directories @value{GDBN} searches for separate debugging
15024 information files.
15025
15026 @end table
15027
15028 @cindex @code{.gnu_debuglink} sections
15029 @cindex debug link sections
15030 A debug link is a special section of the executable file named
15031 @code{.gnu_debuglink}. The section must contain:
15032
15033 @itemize
15034 @item
15035 A filename, with any leading directory components removed, followed by
15036 a zero byte,
15037 @item
15038 zero to three bytes of padding, as needed to reach the next four-byte
15039 boundary within the section, and
15040 @item
15041 a four-byte CRC checksum, stored in the same endianness used for the
15042 executable file itself. The checksum is computed on the debugging
15043 information file's full contents by the function given below, passing
15044 zero as the @var{crc} argument.
15045 @end itemize
15046
15047 Any executable file format can carry a debug link, as long as it can
15048 contain a section named @code{.gnu_debuglink} with the contents
15049 described above.
15050
15051 @cindex @code{.note.gnu.build-id} sections
15052 @cindex build ID sections
15053 The build ID is a special section in the executable file (and in other
15054 ELF binary files that @value{GDBN} may consider). This section is
15055 often named @code{.note.gnu.build-id}, but that name is not mandatory.
15056 It contains unique identification for the built files---the ID remains
15057 the same across multiple builds of the same build tree. The default
15058 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
15059 content for the build ID string. The same section with an identical
15060 value is present in the original built binary with symbols, in its
15061 stripped variant, and in the separate debugging information file.
15062
15063 The debugging information file itself should be an ordinary
15064 executable, containing a full set of linker symbols, sections, and
15065 debugging information. The sections of the debugging information file
15066 should have the same names, addresses, and sizes as the original file,
15067 but they need not contain any data---much like a @code{.bss} section
15068 in an ordinary executable.
15069
15070 The @sc{gnu} binary utilities (Binutils) package includes the
15071 @samp{objcopy} utility that can produce
15072 the separated executable / debugging information file pairs using the
15073 following commands:
15074
15075 @smallexample
15076 @kbd{objcopy --only-keep-debug foo foo.debug}
15077 @kbd{strip -g foo}
15078 @end smallexample
15079
15080 @noindent
15081 These commands remove the debugging
15082 information from the executable file @file{foo} and place it in the file
15083 @file{foo.debug}. You can use the first, second or both methods to link the
15084 two files:
15085
15086 @itemize @bullet
15087 @item
15088 The debug link method needs the following additional command to also leave
15089 behind a debug link in @file{foo}:
15090
15091 @smallexample
15092 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
15093 @end smallexample
15094
15095 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
15096 a version of the @code{strip} command such that the command @kbd{strip foo -f
15097 foo.debug} has the same functionality as the two @code{objcopy} commands and
15098 the @code{ln -s} command above, together.
15099
15100 @item
15101 Build ID gets embedded into the main executable using @code{ld --build-id} or
15102 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
15103 compatibility fixes for debug files separation are present in @sc{gnu} binary
15104 utilities (Binutils) package since version 2.18.
15105 @end itemize
15106
15107 @noindent
15108
15109 @cindex CRC algorithm definition
15110 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
15111 IEEE 802.3 using the polynomial:
15112
15113 @c TexInfo requires naked braces for multi-digit exponents for Tex
15114 @c output, but this causes HTML output to barf. HTML has to be set using
15115 @c raw commands. So we end up having to specify this equation in 2
15116 @c different ways!
15117 @ifhtml
15118 @display
15119 @html
15120 <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>
15121 + <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
15122 @end html
15123 @end display
15124 @end ifhtml
15125 @ifnothtml
15126 @display
15127 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
15128 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
15129 @end display
15130 @end ifnothtml
15131
15132 The function is computed byte at a time, taking the least
15133 significant bit of each byte first. The initial pattern
15134 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
15135 the final result is inverted to ensure trailing zeros also affect the
15136 CRC.
15137
15138 @emph{Note:} This is the same CRC polynomial as used in handling the
15139 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{Remote Protocol,
15140 , @value{GDBN} Remote Serial Protocol}). However in the
15141 case of the Remote Serial Protocol, the CRC is computed @emph{most}
15142 significant bit first, and the result is not inverted, so trailing
15143 zeros have no effect on the CRC value.
15144
15145 To complete the description, we show below the code of the function
15146 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
15147 initially supplied @code{crc} argument means that an initial call to
15148 this function passing in zero will start computing the CRC using
15149 @code{0xffffffff}.
15150
15151 @kindex gnu_debuglink_crc32
15152 @smallexample
15153 unsigned long
15154 gnu_debuglink_crc32 (unsigned long crc,
15155 unsigned char *buf, size_t len)
15156 @{
15157 static const unsigned long crc32_table[256] =
15158 @{
15159 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
15160 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
15161 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
15162 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
15163 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
15164 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
15165 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
15166 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
15167 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
15168 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
15169 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
15170 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
15171 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
15172 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
15173 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
15174 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
15175 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
15176 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
15177 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
15178 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
15179 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
15180 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
15181 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
15182 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
15183 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
15184 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
15185 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
15186 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
15187 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
15188 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
15189 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
15190 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
15191 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
15192 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
15193 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
15194 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
15195 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
15196 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
15197 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
15198 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
15199 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
15200 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
15201 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
15202 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
15203 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
15204 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
15205 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
15206 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
15207 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
15208 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
15209 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
15210 0x2d02ef8d
15211 @};
15212 unsigned char *end;
15213
15214 crc = ~crc & 0xffffffff;
15215 for (end = buf + len; buf < end; ++buf)
15216 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
15217 return ~crc & 0xffffffff;
15218 @}
15219 @end smallexample
15220
15221 @noindent
15222 This computation does not apply to the ``build ID'' method.
15223
15224
15225 @node Index Files
15226 @section Index Files Speed Up @value{GDBN}
15227 @cindex index files
15228 @cindex @samp{.gdb_index} section
15229
15230 When @value{GDBN} finds a symbol file, it scans the symbols in the
15231 file in order to construct an internal symbol table. This lets most
15232 @value{GDBN} operations work quickly---at the cost of a delay early
15233 on. For large programs, this delay can be quite lengthy, so
15234 @value{GDBN} provides a way to build an index, which speeds up
15235 startup.
15236
15237 The index is stored as a section in the symbol file. @value{GDBN} can
15238 write the index to a file, then you can put it into the symbol file
15239 using @command{objcopy}.
15240
15241 To create an index file, use the @code{save gdb-index} command:
15242
15243 @table @code
15244 @item save gdb-index @var{directory}
15245 @kindex save gdb-index
15246 Create an index file for each symbol file currently known by
15247 @value{GDBN}. Each file is named after its corresponding symbol file,
15248 with @samp{.gdb-index} appended, and is written into the given
15249 @var{directory}.
15250 @end table
15251
15252 Once you have created an index file you can merge it into your symbol
15253 file, here named @file{symfile}, using @command{objcopy}:
15254
15255 @smallexample
15256 $ objcopy --add-section .gdb_index=symfile.gdb-index \
15257 --set-section-flags .gdb_index=readonly symfile symfile
15258 @end smallexample
15259
15260 There are currently some limitation on indices. They only work when
15261 for DWARF debugging information, not stabs. And, they do not
15262 currently work for programs using Ada.
15263
15264 @node Symbol Errors
15265 @section Errors Reading Symbol Files
15266
15267 While reading a symbol file, @value{GDBN} occasionally encounters problems,
15268 such as symbol types it does not recognize, or known bugs in compiler
15269 output. By default, @value{GDBN} does not notify you of such problems, since
15270 they are relatively common and primarily of interest to people
15271 debugging compilers. If you are interested in seeing information
15272 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
15273 only one message about each such type of problem, no matter how many
15274 times the problem occurs; or you can ask @value{GDBN} to print more messages,
15275 to see how many times the problems occur, with the @code{set
15276 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
15277 Messages}).
15278
15279 The messages currently printed, and their meanings, include:
15280
15281 @table @code
15282 @item inner block not inside outer block in @var{symbol}
15283
15284 The symbol information shows where symbol scopes begin and end
15285 (such as at the start of a function or a block of statements). This
15286 error indicates that an inner scope block is not fully contained
15287 in its outer scope blocks.
15288
15289 @value{GDBN} circumvents the problem by treating the inner block as if it had
15290 the same scope as the outer block. In the error message, @var{symbol}
15291 may be shown as ``@code{(don't know)}'' if the outer block is not a
15292 function.
15293
15294 @item block at @var{address} out of order
15295
15296 The symbol information for symbol scope blocks should occur in
15297 order of increasing addresses. This error indicates that it does not
15298 do so.
15299
15300 @value{GDBN} does not circumvent this problem, and has trouble
15301 locating symbols in the source file whose symbols it is reading. (You
15302 can often determine what source file is affected by specifying
15303 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
15304 Messages}.)
15305
15306 @item bad block start address patched
15307
15308 The symbol information for a symbol scope block has a start address
15309 smaller than the address of the preceding source line. This is known
15310 to occur in the SunOS 4.1.1 (and earlier) C compiler.
15311
15312 @value{GDBN} circumvents the problem by treating the symbol scope block as
15313 starting on the previous source line.
15314
15315 @item bad string table offset in symbol @var{n}
15316
15317 @cindex foo
15318 Symbol number @var{n} contains a pointer into the string table which is
15319 larger than the size of the string table.
15320
15321 @value{GDBN} circumvents the problem by considering the symbol to have the
15322 name @code{foo}, which may cause other problems if many symbols end up
15323 with this name.
15324
15325 @item unknown symbol type @code{0x@var{nn}}
15326
15327 The symbol information contains new data types that @value{GDBN} does
15328 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
15329 uncomprehended information, in hexadecimal.
15330
15331 @value{GDBN} circumvents the error by ignoring this symbol information.
15332 This usually allows you to debug your program, though certain symbols
15333 are not accessible. If you encounter such a problem and feel like
15334 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
15335 on @code{complain}, then go up to the function @code{read_dbx_symtab}
15336 and examine @code{*bufp} to see the symbol.
15337
15338 @item stub type has NULL name
15339
15340 @value{GDBN} could not find the full definition for a struct or class.
15341
15342 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
15343 The symbol information for a C@t{++} member function is missing some
15344 information that recent versions of the compiler should have output for
15345 it.
15346
15347 @item info mismatch between compiler and debugger
15348
15349 @value{GDBN} could not parse a type specification output by the compiler.
15350
15351 @end table
15352
15353 @node Data Files
15354 @section GDB Data Files
15355
15356 @cindex prefix for data files
15357 @value{GDBN} will sometimes read an auxiliary data file. These files
15358 are kept in a directory known as the @dfn{data directory}.
15359
15360 You can set the data directory's name, and view the name @value{GDBN}
15361 is currently using.
15362
15363 @table @code
15364 @kindex set data-directory
15365 @item set data-directory @var{directory}
15366 Set the directory which @value{GDBN} searches for auxiliary data files
15367 to @var{directory}.
15368
15369 @kindex show data-directory
15370 @item show data-directory
15371 Show the directory @value{GDBN} searches for auxiliary data files.
15372 @end table
15373
15374 @cindex default data directory
15375 @cindex @samp{--with-gdb-datadir}
15376 You can set the default data directory by using the configure-time
15377 @samp{--with-gdb-datadir} option. If the data directory is inside
15378 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
15379 @samp{--exec-prefix}), then the default data directory will be updated
15380 automatically if the installed @value{GDBN} is moved to a new
15381 location.
15382
15383 @node Targets
15384 @chapter Specifying a Debugging Target
15385
15386 @cindex debugging target
15387 A @dfn{target} is the execution environment occupied by your program.
15388
15389 Often, @value{GDBN} runs in the same host environment as your program;
15390 in that case, the debugging target is specified as a side effect when
15391 you use the @code{file} or @code{core} commands. When you need more
15392 flexibility---for example, running @value{GDBN} on a physically separate
15393 host, or controlling a standalone system over a serial port or a
15394 realtime system over a TCP/IP connection---you can use the @code{target}
15395 command to specify one of the target types configured for @value{GDBN}
15396 (@pxref{Target Commands, ,Commands for Managing Targets}).
15397
15398 @cindex target architecture
15399 It is possible to build @value{GDBN} for several different @dfn{target
15400 architectures}. When @value{GDBN} is built like that, you can choose
15401 one of the available architectures with the @kbd{set architecture}
15402 command.
15403
15404 @table @code
15405 @kindex set architecture
15406 @kindex show architecture
15407 @item set architecture @var{arch}
15408 This command sets the current target architecture to @var{arch}. The
15409 value of @var{arch} can be @code{"auto"}, in addition to one of the
15410 supported architectures.
15411
15412 @item show architecture
15413 Show the current target architecture.
15414
15415 @item set processor
15416 @itemx processor
15417 @kindex set processor
15418 @kindex show processor
15419 These are alias commands for, respectively, @code{set architecture}
15420 and @code{show architecture}.
15421 @end table
15422
15423 @menu
15424 * Active Targets:: Active targets
15425 * Target Commands:: Commands for managing targets
15426 * Byte Order:: Choosing target byte order
15427 @end menu
15428
15429 @node Active Targets
15430 @section Active Targets
15431
15432 @cindex stacking targets
15433 @cindex active targets
15434 @cindex multiple targets
15435
15436 There are multiple classes of targets such as: processes, executable files or
15437 recording sessions. Core files belong to the process class, making core file
15438 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
15439 on multiple active targets, one in each class. This allows you to (for
15440 example) start a process and inspect its activity, while still having access to
15441 the executable file after the process finishes. Or if you start process
15442 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
15443 presented a virtual layer of the recording target, while the process target
15444 remains stopped at the chronologically last point of the process execution.
15445
15446 Use the @code{core-file} and @code{exec-file} commands to select a new core
15447 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
15448 specify as a target a process that is already running, use the @code{attach}
15449 command (@pxref{Attach, ,Debugging an Already-running Process}).
15450
15451 @node Target Commands
15452 @section Commands for Managing Targets
15453
15454 @table @code
15455 @item target @var{type} @var{parameters}
15456 Connects the @value{GDBN} host environment to a target machine or
15457 process. A target is typically a protocol for talking to debugging
15458 facilities. You use the argument @var{type} to specify the type or
15459 protocol of the target machine.
15460
15461 Further @var{parameters} are interpreted by the target protocol, but
15462 typically include things like device names or host names to connect
15463 with, process numbers, and baud rates.
15464
15465 The @code{target} command does not repeat if you press @key{RET} again
15466 after executing the command.
15467
15468 @kindex help target
15469 @item help target
15470 Displays the names of all targets available. To display targets
15471 currently selected, use either @code{info target} or @code{info files}
15472 (@pxref{Files, ,Commands to Specify Files}).
15473
15474 @item help target @var{name}
15475 Describe a particular target, including any parameters necessary to
15476 select it.
15477
15478 @kindex set gnutarget
15479 @item set gnutarget @var{args}
15480 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
15481 knows whether it is reading an @dfn{executable},
15482 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
15483 with the @code{set gnutarget} command. Unlike most @code{target} commands,
15484 with @code{gnutarget} the @code{target} refers to a program, not a machine.
15485
15486 @quotation
15487 @emph{Warning:} To specify a file format with @code{set gnutarget},
15488 you must know the actual BFD name.
15489 @end quotation
15490
15491 @noindent
15492 @xref{Files, , Commands to Specify Files}.
15493
15494 @kindex show gnutarget
15495 @item show gnutarget
15496 Use the @code{show gnutarget} command to display what file format
15497 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
15498 @value{GDBN} will determine the file format for each file automatically,
15499 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
15500 @end table
15501
15502 @cindex common targets
15503 Here are some common targets (available, or not, depending on the GDB
15504 configuration):
15505
15506 @table @code
15507 @kindex target
15508 @item target exec @var{program}
15509 @cindex executable file target
15510 An executable file. @samp{target exec @var{program}} is the same as
15511 @samp{exec-file @var{program}}.
15512
15513 @item target core @var{filename}
15514 @cindex core dump file target
15515 A core dump file. @samp{target core @var{filename}} is the same as
15516 @samp{core-file @var{filename}}.
15517
15518 @item target remote @var{medium}
15519 @cindex remote target
15520 A remote system connected to @value{GDBN} via a serial line or network
15521 connection. This command tells @value{GDBN} to use its own remote
15522 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
15523
15524 For example, if you have a board connected to @file{/dev/ttya} on the
15525 machine running @value{GDBN}, you could say:
15526
15527 @smallexample
15528 target remote /dev/ttya
15529 @end smallexample
15530
15531 @code{target remote} supports the @code{load} command. This is only
15532 useful if you have some other way of getting the stub to the target
15533 system, and you can put it somewhere in memory where it won't get
15534 clobbered by the download.
15535
15536 @item target sim @r{[}@var{simargs}@r{]} @dots{}
15537 @cindex built-in simulator target
15538 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
15539 In general,
15540 @smallexample
15541 target sim
15542 load
15543 run
15544 @end smallexample
15545 @noindent
15546 works; however, you cannot assume that a specific memory map, device
15547 drivers, or even basic I/O is available, although some simulators do
15548 provide these. For info about any processor-specific simulator details,
15549 see the appropriate section in @ref{Embedded Processors, ,Embedded
15550 Processors}.
15551
15552 @end table
15553
15554 Some configurations may include these targets as well:
15555
15556 @table @code
15557
15558 @item target nrom @var{dev}
15559 @cindex NetROM ROM emulator target
15560 NetROM ROM emulator. This target only supports downloading.
15561
15562 @end table
15563
15564 Different targets are available on different configurations of @value{GDBN};
15565 your configuration may have more or fewer targets.
15566
15567 Many remote targets require you to download the executable's code once
15568 you've successfully established a connection. You may wish to control
15569 various aspects of this process.
15570
15571 @table @code
15572
15573 @item set hash
15574 @kindex set hash@r{, for remote monitors}
15575 @cindex hash mark while downloading
15576 This command controls whether a hash mark @samp{#} is displayed while
15577 downloading a file to the remote monitor. If on, a hash mark is
15578 displayed after each S-record is successfully downloaded to the
15579 monitor.
15580
15581 @item show hash
15582 @kindex show hash@r{, for remote monitors}
15583 Show the current status of displaying the hash mark.
15584
15585 @item set debug monitor
15586 @kindex set debug monitor
15587 @cindex display remote monitor communications
15588 Enable or disable display of communications messages between
15589 @value{GDBN} and the remote monitor.
15590
15591 @item show debug monitor
15592 @kindex show debug monitor
15593 Show the current status of displaying communications between
15594 @value{GDBN} and the remote monitor.
15595 @end table
15596
15597 @table @code
15598
15599 @kindex load @var{filename}
15600 @item load @var{filename}
15601 @anchor{load}
15602 Depending on what remote debugging facilities are configured into
15603 @value{GDBN}, the @code{load} command may be available. Where it exists, it
15604 is meant to make @var{filename} (an executable) available for debugging
15605 on the remote system---by downloading, or dynamic linking, for example.
15606 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
15607 the @code{add-symbol-file} command.
15608
15609 If your @value{GDBN} does not have a @code{load} command, attempting to
15610 execute it gets the error message ``@code{You can't do that when your
15611 target is @dots{}}''
15612
15613 The file is loaded at whatever address is specified in the executable.
15614 For some object file formats, you can specify the load address when you
15615 link the program; for other formats, like a.out, the object file format
15616 specifies a fixed address.
15617 @c FIXME! This would be a good place for an xref to the GNU linker doc.
15618
15619 Depending on the remote side capabilities, @value{GDBN} may be able to
15620 load programs into flash memory.
15621
15622 @code{load} does not repeat if you press @key{RET} again after using it.
15623 @end table
15624
15625 @node Byte Order
15626 @section Choosing Target Byte Order
15627
15628 @cindex choosing target byte order
15629 @cindex target byte order
15630
15631 Some types of processors, such as the MIPS, PowerPC, and Renesas SH,
15632 offer the ability to run either big-endian or little-endian byte
15633 orders. Usually the executable or symbol will include a bit to
15634 designate the endian-ness, and you will not need to worry about
15635 which to use. However, you may still find it useful to adjust
15636 @value{GDBN}'s idea of processor endian-ness manually.
15637
15638 @table @code
15639 @kindex set endian
15640 @item set endian big
15641 Instruct @value{GDBN} to assume the target is big-endian.
15642
15643 @item set endian little
15644 Instruct @value{GDBN} to assume the target is little-endian.
15645
15646 @item set endian auto
15647 Instruct @value{GDBN} to use the byte order associated with the
15648 executable.
15649
15650 @item show endian
15651 Display @value{GDBN}'s current idea of the target byte order.
15652
15653 @end table
15654
15655 Note that these commands merely adjust interpretation of symbolic
15656 data on the host, and that they have absolutely no effect on the
15657 target system.
15658
15659
15660 @node Remote Debugging
15661 @chapter Debugging Remote Programs
15662 @cindex remote debugging
15663
15664 If you are trying to debug a program running on a machine that cannot run
15665 @value{GDBN} in the usual way, it is often useful to use remote debugging.
15666 For example, you might use remote debugging on an operating system kernel,
15667 or on a small system which does not have a general purpose operating system
15668 powerful enough to run a full-featured debugger.
15669
15670 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
15671 to make this work with particular debugging targets. In addition,
15672 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
15673 but not specific to any particular target system) which you can use if you
15674 write the remote stubs---the code that runs on the remote system to
15675 communicate with @value{GDBN}.
15676
15677 Other remote targets may be available in your
15678 configuration of @value{GDBN}; use @code{help target} to list them.
15679
15680 @menu
15681 * Connecting:: Connecting to a remote target
15682 * File Transfer:: Sending files to a remote system
15683 * Server:: Using the gdbserver program
15684 * Remote Configuration:: Remote configuration
15685 * Remote Stub:: Implementing a remote stub
15686 @end menu
15687
15688 @node Connecting
15689 @section Connecting to a Remote Target
15690
15691 On the @value{GDBN} host machine, you will need an unstripped copy of
15692 your program, since @value{GDBN} needs symbol and debugging information.
15693 Start up @value{GDBN} as usual, using the name of the local copy of your
15694 program as the first argument.
15695
15696 @cindex @code{target remote}
15697 @value{GDBN} can communicate with the target over a serial line, or
15698 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
15699 each case, @value{GDBN} uses the same protocol for debugging your
15700 program; only the medium carrying the debugging packets varies. The
15701 @code{target remote} command establishes a connection to the target.
15702 Its arguments indicate which medium to use:
15703
15704 @table @code
15705
15706 @item target remote @var{serial-device}
15707 @cindex serial line, @code{target remote}
15708 Use @var{serial-device} to communicate with the target. For example,
15709 to use a serial line connected to the device named @file{/dev/ttyb}:
15710
15711 @smallexample
15712 target remote /dev/ttyb
15713 @end smallexample
15714
15715 If you're using a serial line, you may want to give @value{GDBN} the
15716 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
15717 (@pxref{Remote Configuration, set remotebaud}) before the
15718 @code{target} command.
15719
15720 @item target remote @code{@var{host}:@var{port}}
15721 @itemx target remote @code{tcp:@var{host}:@var{port}}
15722 @cindex @acronym{TCP} port, @code{target remote}
15723 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
15724 The @var{host} may be either a host name or a numeric @acronym{IP}
15725 address; @var{port} must be a decimal number. The @var{host} could be
15726 the target machine itself, if it is directly connected to the net, or
15727 it might be a terminal server which in turn has a serial line to the
15728 target.
15729
15730 For example, to connect to port 2828 on a terminal server named
15731 @code{manyfarms}:
15732
15733 @smallexample
15734 target remote manyfarms:2828
15735 @end smallexample
15736
15737 If your remote target is actually running on the same machine as your
15738 debugger session (e.g.@: a simulator for your target running on the
15739 same host), you can omit the hostname. For example, to connect to
15740 port 1234 on your local machine:
15741
15742 @smallexample
15743 target remote :1234
15744 @end smallexample
15745 @noindent
15746
15747 Note that the colon is still required here.
15748
15749 @item target remote @code{udp:@var{host}:@var{port}}
15750 @cindex @acronym{UDP} port, @code{target remote}
15751 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
15752 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
15753
15754 @smallexample
15755 target remote udp:manyfarms:2828
15756 @end smallexample
15757
15758 When using a @acronym{UDP} connection for remote debugging, you should
15759 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
15760 can silently drop packets on busy or unreliable networks, which will
15761 cause havoc with your debugging session.
15762
15763 @item target remote | @var{command}
15764 @cindex pipe, @code{target remote} to
15765 Run @var{command} in the background and communicate with it using a
15766 pipe. The @var{command} is a shell command, to be parsed and expanded
15767 by the system's command shell, @code{/bin/sh}; it should expect remote
15768 protocol packets on its standard input, and send replies on its
15769 standard output. You could use this to run a stand-alone simulator
15770 that speaks the remote debugging protocol, to make net connections
15771 using programs like @code{ssh}, or for other similar tricks.
15772
15773 If @var{command} closes its standard output (perhaps by exiting),
15774 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
15775 program has already exited, this will have no effect.)
15776
15777 @end table
15778
15779 Once the connection has been established, you can use all the usual
15780 commands to examine and change data. The remote program is already
15781 running; you can use @kbd{step} and @kbd{continue}, and you do not
15782 need to use @kbd{run}.
15783
15784 @cindex interrupting remote programs
15785 @cindex remote programs, interrupting
15786 Whenever @value{GDBN} is waiting for the remote program, if you type the
15787 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
15788 program. This may or may not succeed, depending in part on the hardware
15789 and the serial drivers the remote system uses. If you type the
15790 interrupt character once again, @value{GDBN} displays this prompt:
15791
15792 @smallexample
15793 Interrupted while waiting for the program.
15794 Give up (and stop debugging it)? (y or n)
15795 @end smallexample
15796
15797 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
15798 (If you decide you want to try again later, you can use @samp{target
15799 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
15800 goes back to waiting.
15801
15802 @table @code
15803 @kindex detach (remote)
15804 @item detach
15805 When you have finished debugging the remote program, you can use the
15806 @code{detach} command to release it from @value{GDBN} control.
15807 Detaching from the target normally resumes its execution, but the results
15808 will depend on your particular remote stub. After the @code{detach}
15809 command, @value{GDBN} is free to connect to another target.
15810
15811 @kindex disconnect
15812 @item disconnect
15813 The @code{disconnect} command behaves like @code{detach}, except that
15814 the target is generally not resumed. It will wait for @value{GDBN}
15815 (this instance or another one) to connect and continue debugging. After
15816 the @code{disconnect} command, @value{GDBN} is again free to connect to
15817 another target.
15818
15819 @cindex send command to remote monitor
15820 @cindex extend @value{GDBN} for remote targets
15821 @cindex add new commands for external monitor
15822 @kindex monitor
15823 @item monitor @var{cmd}
15824 This command allows you to send arbitrary commands directly to the
15825 remote monitor. Since @value{GDBN} doesn't care about the commands it
15826 sends like this, this command is the way to extend @value{GDBN}---you
15827 can add new commands that only the external monitor will understand
15828 and implement.
15829 @end table
15830
15831 @node File Transfer
15832 @section Sending files to a remote system
15833 @cindex remote target, file transfer
15834 @cindex file transfer
15835 @cindex sending files to remote systems
15836
15837 Some remote targets offer the ability to transfer files over the same
15838 connection used to communicate with @value{GDBN}. This is convenient
15839 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
15840 running @code{gdbserver} over a network interface. For other targets,
15841 e.g.@: embedded devices with only a single serial port, this may be
15842 the only way to upload or download files.
15843
15844 Not all remote targets support these commands.
15845
15846 @table @code
15847 @kindex remote put
15848 @item remote put @var{hostfile} @var{targetfile}
15849 Copy file @var{hostfile} from the host system (the machine running
15850 @value{GDBN}) to @var{targetfile} on the target system.
15851
15852 @kindex remote get
15853 @item remote get @var{targetfile} @var{hostfile}
15854 Copy file @var{targetfile} from the target system to @var{hostfile}
15855 on the host system.
15856
15857 @kindex remote delete
15858 @item remote delete @var{targetfile}
15859 Delete @var{targetfile} from the target system.
15860
15861 @end table
15862
15863 @node Server
15864 @section Using the @code{gdbserver} Program
15865
15866 @kindex gdbserver
15867 @cindex remote connection without stubs
15868 @code{gdbserver} is a control program for Unix-like systems, which
15869 allows you to connect your program with a remote @value{GDBN} via
15870 @code{target remote}---but without linking in the usual debugging stub.
15871
15872 @code{gdbserver} is not a complete replacement for the debugging stubs,
15873 because it requires essentially the same operating-system facilities
15874 that @value{GDBN} itself does. In fact, a system that can run
15875 @code{gdbserver} to connect to a remote @value{GDBN} could also run
15876 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
15877 because it is a much smaller program than @value{GDBN} itself. It is
15878 also easier to port than all of @value{GDBN}, so you may be able to get
15879 started more quickly on a new system by using @code{gdbserver}.
15880 Finally, if you develop code for real-time systems, you may find that
15881 the tradeoffs involved in real-time operation make it more convenient to
15882 do as much development work as possible on another system, for example
15883 by cross-compiling. You can use @code{gdbserver} to make a similar
15884 choice for debugging.
15885
15886 @value{GDBN} and @code{gdbserver} communicate via either a serial line
15887 or a TCP connection, using the standard @value{GDBN} remote serial
15888 protocol.
15889
15890 @quotation
15891 @emph{Warning:} @code{gdbserver} does not have any built-in security.
15892 Do not run @code{gdbserver} connected to any public network; a
15893 @value{GDBN} connection to @code{gdbserver} provides access to the
15894 target system with the same privileges as the user running
15895 @code{gdbserver}.
15896 @end quotation
15897
15898 @subsection Running @code{gdbserver}
15899 @cindex arguments, to @code{gdbserver}
15900
15901 Run @code{gdbserver} on the target system. You need a copy of the
15902 program you want to debug, including any libraries it requires.
15903 @code{gdbserver} does not need your program's symbol table, so you can
15904 strip the program if necessary to save space. @value{GDBN} on the host
15905 system does all the symbol handling.
15906
15907 To use the server, you must tell it how to communicate with @value{GDBN};
15908 the name of your program; and the arguments for your program. The usual
15909 syntax is:
15910
15911 @smallexample
15912 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
15913 @end smallexample
15914
15915 @var{comm} is either a device name (to use a serial line) or a TCP
15916 hostname and portnumber. For example, to debug Emacs with the argument
15917 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
15918 @file{/dev/com1}:
15919
15920 @smallexample
15921 target> gdbserver /dev/com1 emacs foo.txt
15922 @end smallexample
15923
15924 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
15925 with it.
15926
15927 To use a TCP connection instead of a serial line:
15928
15929 @smallexample
15930 target> gdbserver host:2345 emacs foo.txt
15931 @end smallexample
15932
15933 The only difference from the previous example is the first argument,
15934 specifying that you are communicating with the host @value{GDBN} via
15935 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
15936 expect a TCP connection from machine @samp{host} to local TCP port 2345.
15937 (Currently, the @samp{host} part is ignored.) You can choose any number
15938 you want for the port number as long as it does not conflict with any
15939 TCP ports already in use on the target system (for example, @code{23} is
15940 reserved for @code{telnet}).@footnote{If you choose a port number that
15941 conflicts with another service, @code{gdbserver} prints an error message
15942 and exits.} You must use the same port number with the host @value{GDBN}
15943 @code{target remote} command.
15944
15945 @subsubsection Attaching to a Running Program
15946
15947 On some targets, @code{gdbserver} can also attach to running programs.
15948 This is accomplished via the @code{--attach} argument. The syntax is:
15949
15950 @smallexample
15951 target> gdbserver --attach @var{comm} @var{pid}
15952 @end smallexample
15953
15954 @var{pid} is the process ID of a currently running process. It isn't necessary
15955 to point @code{gdbserver} at a binary for the running process.
15956
15957 @pindex pidof
15958 @cindex attach to a program by name
15959 You can debug processes by name instead of process ID if your target has the
15960 @code{pidof} utility:
15961
15962 @smallexample
15963 target> gdbserver --attach @var{comm} `pidof @var{program}`
15964 @end smallexample
15965
15966 In case more than one copy of @var{program} is running, or @var{program}
15967 has multiple threads, most versions of @code{pidof} support the
15968 @code{-s} option to only return the first process ID.
15969
15970 @subsubsection Multi-Process Mode for @code{gdbserver}
15971 @cindex gdbserver, multiple processes
15972 @cindex multiple processes with gdbserver
15973
15974 When you connect to @code{gdbserver} using @code{target remote},
15975 @code{gdbserver} debugs the specified program only once. When the
15976 program exits, or you detach from it, @value{GDBN} closes the connection
15977 and @code{gdbserver} exits.
15978
15979 If you connect using @kbd{target extended-remote}, @code{gdbserver}
15980 enters multi-process mode. When the debugged program exits, or you
15981 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
15982 though no program is running. The @code{run} and @code{attach}
15983 commands instruct @code{gdbserver} to run or attach to a new program.
15984 The @code{run} command uses @code{set remote exec-file} (@pxref{set
15985 remote exec-file}) to select the program to run. Command line
15986 arguments are supported, except for wildcard expansion and I/O
15987 redirection (@pxref{Arguments}).
15988
15989 To start @code{gdbserver} without supplying an initial command to run
15990 or process ID to attach, use the @option{--multi} command line option.
15991 Then you can connect using @kbd{target extended-remote} and start
15992 the program you want to debug.
15993
15994 @code{gdbserver} does not automatically exit in multi-process mode.
15995 You can terminate it by using @code{monitor exit}
15996 (@pxref{Monitor Commands for gdbserver}).
15997
15998 @subsubsection Other Command-Line Arguments for @code{gdbserver}
15999
16000 The @option{--debug} option tells @code{gdbserver} to display extra
16001 status information about the debugging process. The
16002 @option{--remote-debug} option tells @code{gdbserver} to display
16003 remote protocol debug output. These options are intended for
16004 @code{gdbserver} development and for bug reports to the developers.
16005
16006 The @option{--wrapper} option specifies a wrapper to launch programs
16007 for debugging. The option should be followed by the name of the
16008 wrapper, then any command-line arguments to pass to the wrapper, then
16009 @kbd{--} indicating the end of the wrapper arguments.
16010
16011 @code{gdbserver} runs the specified wrapper program with a combined
16012 command line including the wrapper arguments, then the name of the
16013 program to debug, then any arguments to the program. The wrapper
16014 runs until it executes your program, and then @value{GDBN} gains control.
16015
16016 You can use any program that eventually calls @code{execve} with
16017 its arguments as a wrapper. Several standard Unix utilities do
16018 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
16019 with @code{exec "$@@"} will also work.
16020
16021 For example, you can use @code{env} to pass an environment variable to
16022 the debugged program, without setting the variable in @code{gdbserver}'s
16023 environment:
16024
16025 @smallexample
16026 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
16027 @end smallexample
16028
16029 @subsection Connecting to @code{gdbserver}
16030
16031 Run @value{GDBN} on the host system.
16032
16033 First make sure you have the necessary symbol files. Load symbols for
16034 your application using the @code{file} command before you connect. Use
16035 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
16036 was compiled with the correct sysroot using @code{--with-sysroot}).
16037
16038 The symbol file and target libraries must exactly match the executable
16039 and libraries on the target, with one exception: the files on the host
16040 system should not be stripped, even if the files on the target system
16041 are. Mismatched or missing files will lead to confusing results
16042 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
16043 files may also prevent @code{gdbserver} from debugging multi-threaded
16044 programs.
16045
16046 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
16047 For TCP connections, you must start up @code{gdbserver} prior to using
16048 the @code{target remote} command. Otherwise you may get an error whose
16049 text depends on the host system, but which usually looks something like
16050 @samp{Connection refused}. Don't use the @code{load}
16051 command in @value{GDBN} when using @code{gdbserver}, since the program is
16052 already on the target.
16053
16054 @subsection Monitor Commands for @code{gdbserver}
16055 @cindex monitor commands, for @code{gdbserver}
16056 @anchor{Monitor Commands for gdbserver}
16057
16058 During a @value{GDBN} session using @code{gdbserver}, you can use the
16059 @code{monitor} command to send special requests to @code{gdbserver}.
16060 Here are the available commands.
16061
16062 @table @code
16063 @item monitor help
16064 List the available monitor commands.
16065
16066 @item monitor set debug 0
16067 @itemx monitor set debug 1
16068 Disable or enable general debugging messages.
16069
16070 @item monitor set remote-debug 0
16071 @itemx monitor set remote-debug 1
16072 Disable or enable specific debugging messages associated with the remote
16073 protocol (@pxref{Remote Protocol}).
16074
16075 @item monitor set libthread-db-search-path [PATH]
16076 @cindex gdbserver, search path for @code{libthread_db}
16077 When this command is issued, @var{path} is a colon-separated list of
16078 directories to search for @code{libthread_db} (@pxref{Threads,,set
16079 libthread-db-search-path}). If you omit @var{path},
16080 @samp{libthread-db-search-path} will be reset to an empty list.
16081
16082 @item monitor exit
16083 Tell gdbserver to exit immediately. This command should be followed by
16084 @code{disconnect} to close the debugging session. @code{gdbserver} will
16085 detach from any attached processes and kill any processes it created.
16086 Use @code{monitor exit} to terminate @code{gdbserver} at the end
16087 of a multi-process mode debug session.
16088
16089 @end table
16090
16091 @subsection Tracepoints support in @code{gdbserver}
16092 @cindex tracepoints support in @code{gdbserver}
16093
16094 On some targets, @code{gdbserver} supports tracepoints, fast
16095 tracepoints and static tracepoints.
16096
16097 For fast or static tracepoints to work, a special library called the
16098 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
16099 This library is built and distributed as an integral part of
16100 @code{gdbserver}. In addition, support for static tracepoints
16101 requires building the in-process agent library with static tracepoints
16102 support. At present, the UST (LTTng Userspace Tracer,
16103 @url{http://lttng.org/ust}) tracing engine is supported. This support
16104 is automatically available if UST development headers are found in the
16105 standard include path when @code{gdbserver} is built, or if
16106 @code{gdbserver} was explicitly configured using @option{--with-ust}
16107 to point at such headers. You can explicitly disable the support
16108 using @option{--with-ust=no}.
16109
16110 There are several ways to load the in-process agent in your program:
16111
16112 @table @code
16113 @item Specifying it as dependency at link time
16114
16115 You can link your program dynamically with the in-process agent
16116 library. On most systems, this is accomplished by adding
16117 @code{-linproctrace} to the link command.
16118
16119 @item Using the system's preloading mechanisms
16120
16121 You can force loading the in-process agent at startup time by using
16122 your system's support for preloading shared libraries. Many Unixes
16123 support the concept of preloading user defined libraries. In most
16124 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
16125 in the environment. See also the description of @code{gdbserver}'s
16126 @option{--wrapper} command line option.
16127
16128 @item Using @value{GDBN} to force loading the agent at run time
16129
16130 On some systems, you can force the inferior to load a shared library,
16131 by calling a dynamic loader function in the inferior that takes care
16132 of dynamically looking up and loading a shared library. On most Unix
16133 systems, the function is @code{dlopen}. You'll use the @code{call}
16134 command for that. For example:
16135
16136 @smallexample
16137 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
16138 @end smallexample
16139
16140 Note that on most Unix systems, for the @code{dlopen} function to be
16141 available, the program needs to be linked with @code{-ldl}.
16142 @end table
16143
16144 On systems that have a userspace dynamic loader, like most Unix
16145 systems, when you connect to @code{gdbserver} using @code{target
16146 remote}, you'll find that the program is stopped at the dynamic
16147 loader's entry point, and no shared library has been loaded in the
16148 program's address space yet, including the in-process agent. In that
16149 case, before being able to use any of the fast or static tracepoints
16150 features, you need to let the loader run and load the shared
16151 libraries. The simplest way to do that is to run the program to the
16152 main procedure. E.g., if debugging a C or C@t{++} program, start
16153 @code{gdbserver} like so:
16154
16155 @smallexample
16156 $ gdbserver :9999 myprogram
16157 @end smallexample
16158
16159 Start GDB and connect to @code{gdbserver} like so, and run to main:
16160
16161 @smallexample
16162 $ gdb myprogram
16163 (@value{GDBP}) target remote myhost:9999
16164 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
16165 (@value{GDBP}) b main
16166 (@value{GDBP}) continue
16167 @end smallexample
16168
16169 The in-process tracing agent library should now be loaded into the
16170 process; you can confirm it with the @code{info sharedlibrary}
16171 command, which will list @file{libinproctrace.so} as loaded in the
16172 process. You are now ready to install fast tracepoints, list static
16173 tracepoint markers, probe static tracepoints markers, and start
16174 tracing.
16175
16176 @node Remote Configuration
16177 @section Remote Configuration
16178
16179 @kindex set remote
16180 @kindex show remote
16181 This section documents the configuration options available when
16182 debugging remote programs. For the options related to the File I/O
16183 extensions of the remote protocol, see @ref{system,
16184 system-call-allowed}.
16185
16186 @table @code
16187 @item set remoteaddresssize @var{bits}
16188 @cindex address size for remote targets
16189 @cindex bits in remote address
16190 Set the maximum size of address in a memory packet to the specified
16191 number of bits. @value{GDBN} will mask off the address bits above
16192 that number, when it passes addresses to the remote target. The
16193 default value is the number of bits in the target's address.
16194
16195 @item show remoteaddresssize
16196 Show the current value of remote address size in bits.
16197
16198 @item set remotebaud @var{n}
16199 @cindex baud rate for remote targets
16200 Set the baud rate for the remote serial I/O to @var{n} baud. The
16201 value is used to set the speed of the serial port used for debugging
16202 remote targets.
16203
16204 @item show remotebaud
16205 Show the current speed of the remote connection.
16206
16207 @item set remotebreak
16208 @cindex interrupt remote programs
16209 @cindex BREAK signal instead of Ctrl-C
16210 @anchor{set remotebreak}
16211 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
16212 when you type @kbd{Ctrl-c} to interrupt the program running
16213 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
16214 character instead. The default is off, since most remote systems
16215 expect to see @samp{Ctrl-C} as the interrupt signal.
16216
16217 @item show remotebreak
16218 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
16219 interrupt the remote program.
16220
16221 @item set remoteflow on
16222 @itemx set remoteflow off
16223 @kindex set remoteflow
16224 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
16225 on the serial port used to communicate to the remote target.
16226
16227 @item show remoteflow
16228 @kindex show remoteflow
16229 Show the current setting of hardware flow control.
16230
16231 @item set remotelogbase @var{base}
16232 Set the base (a.k.a.@: radix) of logging serial protocol
16233 communications to @var{base}. Supported values of @var{base} are:
16234 @code{ascii}, @code{octal}, and @code{hex}. The default is
16235 @code{ascii}.
16236
16237 @item show remotelogbase
16238 Show the current setting of the radix for logging remote serial
16239 protocol.
16240
16241 @item set remotelogfile @var{file}
16242 @cindex record serial communications on file
16243 Record remote serial communications on the named @var{file}. The
16244 default is not to record at all.
16245
16246 @item show remotelogfile.
16247 Show the current setting of the file name on which to record the
16248 serial communications.
16249
16250 @item set remotetimeout @var{num}
16251 @cindex timeout for serial communications
16252 @cindex remote timeout
16253 Set the timeout limit to wait for the remote target to respond to
16254 @var{num} seconds. The default is 2 seconds.
16255
16256 @item show remotetimeout
16257 Show the current number of seconds to wait for the remote target
16258 responses.
16259
16260 @cindex limit hardware breakpoints and watchpoints
16261 @cindex remote target, limit break- and watchpoints
16262 @anchor{set remote hardware-watchpoint-limit}
16263 @anchor{set remote hardware-breakpoint-limit}
16264 @item set remote hardware-watchpoint-limit @var{limit}
16265 @itemx set remote hardware-breakpoint-limit @var{limit}
16266 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
16267 watchpoints. A limit of -1, the default, is treated as unlimited.
16268
16269 @item set remote exec-file @var{filename}
16270 @itemx show remote exec-file
16271 @anchor{set remote exec-file}
16272 @cindex executable file, for remote target
16273 Select the file used for @code{run} with @code{target
16274 extended-remote}. This should be set to a filename valid on the
16275 target system. If it is not set, the target will use a default
16276 filename (e.g.@: the last program run).
16277
16278 @item set remote interrupt-sequence
16279 @cindex interrupt remote programs
16280 @cindex select Ctrl-C, BREAK or BREAK-g
16281 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
16282 @samp{BREAK-g} as the
16283 sequence to the remote target in order to interrupt the execution.
16284 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
16285 is high level of serial line for some certain time.
16286 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
16287 It is @code{BREAK} signal followed by character @code{g}.
16288
16289 @item show interrupt-sequence
16290 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
16291 is sent by @value{GDBN} to interrupt the remote program.
16292 @code{BREAK-g} is BREAK signal followed by @code{g} and
16293 also known as Magic SysRq g.
16294
16295 @item set remote interrupt-on-connect
16296 @cindex send interrupt-sequence on start
16297 Specify whether interrupt-sequence is sent to remote target when
16298 @value{GDBN} connects to it. This is mostly needed when you debug
16299 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
16300 which is known as Magic SysRq g in order to connect @value{GDBN}.
16301
16302 @item show interrupt-on-connect
16303 Show whether interrupt-sequence is sent
16304 to remote target when @value{GDBN} connects to it.
16305
16306 @kindex set tcp
16307 @kindex show tcp
16308 @item set tcp auto-retry on
16309 @cindex auto-retry, for remote TCP target
16310 Enable auto-retry for remote TCP connections. This is useful if the remote
16311 debugging agent is launched in parallel with @value{GDBN}; there is a race
16312 condition because the agent may not become ready to accept the connection
16313 before @value{GDBN} attempts to connect. When auto-retry is
16314 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
16315 to establish the connection using the timeout specified by
16316 @code{set tcp connect-timeout}.
16317
16318 @item set tcp auto-retry off
16319 Do not auto-retry failed TCP connections.
16320
16321 @item show tcp auto-retry
16322 Show the current auto-retry setting.
16323
16324 @item set tcp connect-timeout @var{seconds}
16325 @cindex connection timeout, for remote TCP target
16326 @cindex timeout, for remote target connection
16327 Set the timeout for establishing a TCP connection to the remote target to
16328 @var{seconds}. The timeout affects both polling to retry failed connections
16329 (enabled by @code{set tcp auto-retry on}) and waiting for connections
16330 that are merely slow to complete, and represents an approximate cumulative
16331 value.
16332
16333 @item show tcp connect-timeout
16334 Show the current connection timeout setting.
16335 @end table
16336
16337 @cindex remote packets, enabling and disabling
16338 The @value{GDBN} remote protocol autodetects the packets supported by
16339 your debugging stub. If you need to override the autodetection, you
16340 can use these commands to enable or disable individual packets. Each
16341 packet can be set to @samp{on} (the remote target supports this
16342 packet), @samp{off} (the remote target does not support this packet),
16343 or @samp{auto} (detect remote target support for this packet). They
16344 all default to @samp{auto}. For more information about each packet,
16345 see @ref{Remote Protocol}.
16346
16347 During normal use, you should not have to use any of these commands.
16348 If you do, that may be a bug in your remote debugging stub, or a bug
16349 in @value{GDBN}. You may want to report the problem to the
16350 @value{GDBN} developers.
16351
16352 For each packet @var{name}, the command to enable or disable the
16353 packet is @code{set remote @var{name}-packet}. The available settings
16354 are:
16355
16356 @multitable @columnfractions 0.28 0.32 0.25
16357 @item Command Name
16358 @tab Remote Packet
16359 @tab Related Features
16360
16361 @item @code{fetch-register}
16362 @tab @code{p}
16363 @tab @code{info registers}
16364
16365 @item @code{set-register}
16366 @tab @code{P}
16367 @tab @code{set}
16368
16369 @item @code{binary-download}
16370 @tab @code{X}
16371 @tab @code{load}, @code{set}
16372
16373 @item @code{read-aux-vector}
16374 @tab @code{qXfer:auxv:read}
16375 @tab @code{info auxv}
16376
16377 @item @code{symbol-lookup}
16378 @tab @code{qSymbol}
16379 @tab Detecting multiple threads
16380
16381 @item @code{attach}
16382 @tab @code{vAttach}
16383 @tab @code{attach}
16384
16385 @item @code{verbose-resume}
16386 @tab @code{vCont}
16387 @tab Stepping or resuming multiple threads
16388
16389 @item @code{run}
16390 @tab @code{vRun}
16391 @tab @code{run}
16392
16393 @item @code{software-breakpoint}
16394 @tab @code{Z0}
16395 @tab @code{break}
16396
16397 @item @code{hardware-breakpoint}
16398 @tab @code{Z1}
16399 @tab @code{hbreak}
16400
16401 @item @code{write-watchpoint}
16402 @tab @code{Z2}
16403 @tab @code{watch}
16404
16405 @item @code{read-watchpoint}
16406 @tab @code{Z3}
16407 @tab @code{rwatch}
16408
16409 @item @code{access-watchpoint}
16410 @tab @code{Z4}
16411 @tab @code{awatch}
16412
16413 @item @code{target-features}
16414 @tab @code{qXfer:features:read}
16415 @tab @code{set architecture}
16416
16417 @item @code{library-info}
16418 @tab @code{qXfer:libraries:read}
16419 @tab @code{info sharedlibrary}
16420
16421 @item @code{memory-map}
16422 @tab @code{qXfer:memory-map:read}
16423 @tab @code{info mem}
16424
16425 @item @code{read-sdata-object}
16426 @tab @code{qXfer:sdata:read}
16427 @tab @code{print $_sdata}
16428
16429 @item @code{read-spu-object}
16430 @tab @code{qXfer:spu:read}
16431 @tab @code{info spu}
16432
16433 @item @code{write-spu-object}
16434 @tab @code{qXfer:spu:write}
16435 @tab @code{info spu}
16436
16437 @item @code{read-siginfo-object}
16438 @tab @code{qXfer:siginfo:read}
16439 @tab @code{print $_siginfo}
16440
16441 @item @code{write-siginfo-object}
16442 @tab @code{qXfer:siginfo:write}
16443 @tab @code{set $_siginfo}
16444
16445 @item @code{threads}
16446 @tab @code{qXfer:threads:read}
16447 @tab @code{info threads}
16448
16449 @item @code{get-thread-local-@*storage-address}
16450 @tab @code{qGetTLSAddr}
16451 @tab Displaying @code{__thread} variables
16452
16453 @item @code{get-thread-information-block-address}
16454 @tab @code{qGetTIBAddr}
16455 @tab Display MS-Windows Thread Information Block.
16456
16457 @item @code{search-memory}
16458 @tab @code{qSearch:memory}
16459 @tab @code{find}
16460
16461 @item @code{supported-packets}
16462 @tab @code{qSupported}
16463 @tab Remote communications parameters
16464
16465 @item @code{pass-signals}
16466 @tab @code{QPassSignals}
16467 @tab @code{handle @var{signal}}
16468
16469 @item @code{hostio-close-packet}
16470 @tab @code{vFile:close}
16471 @tab @code{remote get}, @code{remote put}
16472
16473 @item @code{hostio-open-packet}
16474 @tab @code{vFile:open}
16475 @tab @code{remote get}, @code{remote put}
16476
16477 @item @code{hostio-pread-packet}
16478 @tab @code{vFile:pread}
16479 @tab @code{remote get}, @code{remote put}
16480
16481 @item @code{hostio-pwrite-packet}
16482 @tab @code{vFile:pwrite}
16483 @tab @code{remote get}, @code{remote put}
16484
16485 @item @code{hostio-unlink-packet}
16486 @tab @code{vFile:unlink}
16487 @tab @code{remote delete}
16488
16489 @item @code{noack-packet}
16490 @tab @code{QStartNoAckMode}
16491 @tab Packet acknowledgment
16492
16493 @item @code{osdata}
16494 @tab @code{qXfer:osdata:read}
16495 @tab @code{info os}
16496
16497 @item @code{query-attached}
16498 @tab @code{qAttached}
16499 @tab Querying remote process attach state.
16500 @end multitable
16501
16502 @node Remote Stub
16503 @section Implementing a Remote Stub
16504
16505 @cindex debugging stub, example
16506 @cindex remote stub, example
16507 @cindex stub example, remote debugging
16508 The stub files provided with @value{GDBN} implement the target side of the
16509 communication protocol, and the @value{GDBN} side is implemented in the
16510 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
16511 these subroutines to communicate, and ignore the details. (If you're
16512 implementing your own stub file, you can still ignore the details: start
16513 with one of the existing stub files. @file{sparc-stub.c} is the best
16514 organized, and therefore the easiest to read.)
16515
16516 @cindex remote serial debugging, overview
16517 To debug a program running on another machine (the debugging
16518 @dfn{target} machine), you must first arrange for all the usual
16519 prerequisites for the program to run by itself. For example, for a C
16520 program, you need:
16521
16522 @enumerate
16523 @item
16524 A startup routine to set up the C runtime environment; these usually
16525 have a name like @file{crt0}. The startup routine may be supplied by
16526 your hardware supplier, or you may have to write your own.
16527
16528 @item
16529 A C subroutine library to support your program's
16530 subroutine calls, notably managing input and output.
16531
16532 @item
16533 A way of getting your program to the other machine---for example, a
16534 download program. These are often supplied by the hardware
16535 manufacturer, but you may have to write your own from hardware
16536 documentation.
16537 @end enumerate
16538
16539 The next step is to arrange for your program to use a serial port to
16540 communicate with the machine where @value{GDBN} is running (the @dfn{host}
16541 machine). In general terms, the scheme looks like this:
16542
16543 @table @emph
16544 @item On the host,
16545 @value{GDBN} already understands how to use this protocol; when everything
16546 else is set up, you can simply use the @samp{target remote} command
16547 (@pxref{Targets,,Specifying a Debugging Target}).
16548
16549 @item On the target,
16550 you must link with your program a few special-purpose subroutines that
16551 implement the @value{GDBN} remote serial protocol. The file containing these
16552 subroutines is called a @dfn{debugging stub}.
16553
16554 On certain remote targets, you can use an auxiliary program
16555 @code{gdbserver} instead of linking a stub into your program.
16556 @xref{Server,,Using the @code{gdbserver} Program}, for details.
16557 @end table
16558
16559 The debugging stub is specific to the architecture of the remote
16560 machine; for example, use @file{sparc-stub.c} to debug programs on
16561 @sc{sparc} boards.
16562
16563 @cindex remote serial stub list
16564 These working remote stubs are distributed with @value{GDBN}:
16565
16566 @table @code
16567
16568 @item i386-stub.c
16569 @cindex @file{i386-stub.c}
16570 @cindex Intel
16571 @cindex i386
16572 For Intel 386 and compatible architectures.
16573
16574 @item m68k-stub.c
16575 @cindex @file{m68k-stub.c}
16576 @cindex Motorola 680x0
16577 @cindex m680x0
16578 For Motorola 680x0 architectures.
16579
16580 @item sh-stub.c
16581 @cindex @file{sh-stub.c}
16582 @cindex Renesas
16583 @cindex SH
16584 For Renesas SH architectures.
16585
16586 @item sparc-stub.c
16587 @cindex @file{sparc-stub.c}
16588 @cindex Sparc
16589 For @sc{sparc} architectures.
16590
16591 @item sparcl-stub.c
16592 @cindex @file{sparcl-stub.c}
16593 @cindex Fujitsu
16594 @cindex SparcLite
16595 For Fujitsu @sc{sparclite} architectures.
16596
16597 @end table
16598
16599 The @file{README} file in the @value{GDBN} distribution may list other
16600 recently added stubs.
16601
16602 @menu
16603 * Stub Contents:: What the stub can do for you
16604 * Bootstrapping:: What you must do for the stub
16605 * Debug Session:: Putting it all together
16606 @end menu
16607
16608 @node Stub Contents
16609 @subsection What the Stub Can Do for You
16610
16611 @cindex remote serial stub
16612 The debugging stub for your architecture supplies these three
16613 subroutines:
16614
16615 @table @code
16616 @item set_debug_traps
16617 @findex set_debug_traps
16618 @cindex remote serial stub, initialization
16619 This routine arranges for @code{handle_exception} to run when your
16620 program stops. You must call this subroutine explicitly near the
16621 beginning of your program.
16622
16623 @item handle_exception
16624 @findex handle_exception
16625 @cindex remote serial stub, main routine
16626 This is the central workhorse, but your program never calls it
16627 explicitly---the setup code arranges for @code{handle_exception} to
16628 run when a trap is triggered.
16629
16630 @code{handle_exception} takes control when your program stops during
16631 execution (for example, on a breakpoint), and mediates communications
16632 with @value{GDBN} on the host machine. This is where the communications
16633 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
16634 representative on the target machine. It begins by sending summary
16635 information on the state of your program, then continues to execute,
16636 retrieving and transmitting any information @value{GDBN} needs, until you
16637 execute a @value{GDBN} command that makes your program resume; at that point,
16638 @code{handle_exception} returns control to your own code on the target
16639 machine.
16640
16641 @item breakpoint
16642 @cindex @code{breakpoint} subroutine, remote
16643 Use this auxiliary subroutine to make your program contain a
16644 breakpoint. Depending on the particular situation, this may be the only
16645 way for @value{GDBN} to get control. For instance, if your target
16646 machine has some sort of interrupt button, you won't need to call this;
16647 pressing the interrupt button transfers control to
16648 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
16649 simply receiving characters on the serial port may also trigger a trap;
16650 again, in that situation, you don't need to call @code{breakpoint} from
16651 your own program---simply running @samp{target remote} from the host
16652 @value{GDBN} session gets control.
16653
16654 Call @code{breakpoint} if none of these is true, or if you simply want
16655 to make certain your program stops at a predetermined point for the
16656 start of your debugging session.
16657 @end table
16658
16659 @node Bootstrapping
16660 @subsection What You Must Do for the Stub
16661
16662 @cindex remote stub, support routines
16663 The debugging stubs that come with @value{GDBN} are set up for a particular
16664 chip architecture, but they have no information about the rest of your
16665 debugging target machine.
16666
16667 First of all you need to tell the stub how to communicate with the
16668 serial port.
16669
16670 @table @code
16671 @item int getDebugChar()
16672 @findex getDebugChar
16673 Write this subroutine to read a single character from the serial port.
16674 It may be identical to @code{getchar} for your target system; a
16675 different name is used to allow you to distinguish the two if you wish.
16676
16677 @item void putDebugChar(int)
16678 @findex putDebugChar
16679 Write this subroutine to write a single character to the serial port.
16680 It may be identical to @code{putchar} for your target system; a
16681 different name is used to allow you to distinguish the two if you wish.
16682 @end table
16683
16684 @cindex control C, and remote debugging
16685 @cindex interrupting remote targets
16686 If you want @value{GDBN} to be able to stop your program while it is
16687 running, you need to use an interrupt-driven serial driver, and arrange
16688 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
16689 character). That is the character which @value{GDBN} uses to tell the
16690 remote system to stop.
16691
16692 Getting the debugging target to return the proper status to @value{GDBN}
16693 probably requires changes to the standard stub; one quick and dirty way
16694 is to just execute a breakpoint instruction (the ``dirty'' part is that
16695 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
16696
16697 Other routines you need to supply are:
16698
16699 @table @code
16700 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
16701 @findex exceptionHandler
16702 Write this function to install @var{exception_address} in the exception
16703 handling tables. You need to do this because the stub does not have any
16704 way of knowing what the exception handling tables on your target system
16705 are like (for example, the processor's table might be in @sc{rom},
16706 containing entries which point to a table in @sc{ram}).
16707 @var{exception_number} is the exception number which should be changed;
16708 its meaning is architecture-dependent (for example, different numbers
16709 might represent divide by zero, misaligned access, etc). When this
16710 exception occurs, control should be transferred directly to
16711 @var{exception_address}, and the processor state (stack, registers,
16712 and so on) should be just as it is when a processor exception occurs. So if
16713 you want to use a jump instruction to reach @var{exception_address}, it
16714 should be a simple jump, not a jump to subroutine.
16715
16716 For the 386, @var{exception_address} should be installed as an interrupt
16717 gate so that interrupts are masked while the handler runs. The gate
16718 should be at privilege level 0 (the most privileged level). The
16719 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
16720 help from @code{exceptionHandler}.
16721
16722 @item void flush_i_cache()
16723 @findex flush_i_cache
16724 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
16725 instruction cache, if any, on your target machine. If there is no
16726 instruction cache, this subroutine may be a no-op.
16727
16728 On target machines that have instruction caches, @value{GDBN} requires this
16729 function to make certain that the state of your program is stable.
16730 @end table
16731
16732 @noindent
16733 You must also make sure this library routine is available:
16734
16735 @table @code
16736 @item void *memset(void *, int, int)
16737 @findex memset
16738 This is the standard library function @code{memset} that sets an area of
16739 memory to a known value. If you have one of the free versions of
16740 @code{libc.a}, @code{memset} can be found there; otherwise, you must
16741 either obtain it from your hardware manufacturer, or write your own.
16742 @end table
16743
16744 If you do not use the GNU C compiler, you may need other standard
16745 library subroutines as well; this varies from one stub to another,
16746 but in general the stubs are likely to use any of the common library
16747 subroutines which @code{@value{NGCC}} generates as inline code.
16748
16749
16750 @node Debug Session
16751 @subsection Putting it All Together
16752
16753 @cindex remote serial debugging summary
16754 In summary, when your program is ready to debug, you must follow these
16755 steps.
16756
16757 @enumerate
16758 @item
16759 Make sure you have defined the supporting low-level routines
16760 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
16761 @display
16762 @code{getDebugChar}, @code{putDebugChar},
16763 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
16764 @end display
16765
16766 @item
16767 Insert these lines near the top of your program:
16768
16769 @smallexample
16770 set_debug_traps();
16771 breakpoint();
16772 @end smallexample
16773
16774 @item
16775 For the 680x0 stub only, you need to provide a variable called
16776 @code{exceptionHook}. Normally you just use:
16777
16778 @smallexample
16779 void (*exceptionHook)() = 0;
16780 @end smallexample
16781
16782 @noindent
16783 but if before calling @code{set_debug_traps}, you set it to point to a
16784 function in your program, that function is called when
16785 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
16786 error). The function indicated by @code{exceptionHook} is called with
16787 one parameter: an @code{int} which is the exception number.
16788
16789 @item
16790 Compile and link together: your program, the @value{GDBN} debugging stub for
16791 your target architecture, and the supporting subroutines.
16792
16793 @item
16794 Make sure you have a serial connection between your target machine and
16795 the @value{GDBN} host, and identify the serial port on the host.
16796
16797 @item
16798 @c The "remote" target now provides a `load' command, so we should
16799 @c document that. FIXME.
16800 Download your program to your target machine (or get it there by
16801 whatever means the manufacturer provides), and start it.
16802
16803 @item
16804 Start @value{GDBN} on the host, and connect to the target
16805 (@pxref{Connecting,,Connecting to a Remote Target}).
16806
16807 @end enumerate
16808
16809 @node Configurations
16810 @chapter Configuration-Specific Information
16811
16812 While nearly all @value{GDBN} commands are available for all native and
16813 cross versions of the debugger, there are some exceptions. This chapter
16814 describes things that are only available in certain configurations.
16815
16816 There are three major categories of configurations: native
16817 configurations, where the host and target are the same, embedded
16818 operating system configurations, which are usually the same for several
16819 different processor architectures, and bare embedded processors, which
16820 are quite different from each other.
16821
16822 @menu
16823 * Native::
16824 * Embedded OS::
16825 * Embedded Processors::
16826 * Architectures::
16827 @end menu
16828
16829 @node Native
16830 @section Native
16831
16832 This section describes details specific to particular native
16833 configurations.
16834
16835 @menu
16836 * HP-UX:: HP-UX
16837 * BSD libkvm Interface:: Debugging BSD kernel memory images
16838 * SVR4 Process Information:: SVR4 process information
16839 * DJGPP Native:: Features specific to the DJGPP port
16840 * Cygwin Native:: Features specific to the Cygwin port
16841 * Hurd Native:: Features specific to @sc{gnu} Hurd
16842 * Neutrino:: Features specific to QNX Neutrino
16843 * Darwin:: Features specific to Darwin
16844 @end menu
16845
16846 @node HP-UX
16847 @subsection HP-UX
16848
16849 On HP-UX systems, if you refer to a function or variable name that
16850 begins with a dollar sign, @value{GDBN} searches for a user or system
16851 name first, before it searches for a convenience variable.
16852
16853
16854 @node BSD libkvm Interface
16855 @subsection BSD libkvm Interface
16856
16857 @cindex libkvm
16858 @cindex kernel memory image
16859 @cindex kernel crash dump
16860
16861 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
16862 interface that provides a uniform interface for accessing kernel virtual
16863 memory images, including live systems and crash dumps. @value{GDBN}
16864 uses this interface to allow you to debug live kernels and kernel crash
16865 dumps on many native BSD configurations. This is implemented as a
16866 special @code{kvm} debugging target. For debugging a live system, load
16867 the currently running kernel into @value{GDBN} and connect to the
16868 @code{kvm} target:
16869
16870 @smallexample
16871 (@value{GDBP}) @b{target kvm}
16872 @end smallexample
16873
16874 For debugging crash dumps, provide the file name of the crash dump as an
16875 argument:
16876
16877 @smallexample
16878 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
16879 @end smallexample
16880
16881 Once connected to the @code{kvm} target, the following commands are
16882 available:
16883
16884 @table @code
16885 @kindex kvm
16886 @item kvm pcb
16887 Set current context from the @dfn{Process Control Block} (PCB) address.
16888
16889 @item kvm proc
16890 Set current context from proc address. This command isn't available on
16891 modern FreeBSD systems.
16892 @end table
16893
16894 @node SVR4 Process Information
16895 @subsection SVR4 Process Information
16896 @cindex /proc
16897 @cindex examine process image
16898 @cindex process info via @file{/proc}
16899
16900 Many versions of SVR4 and compatible systems provide a facility called
16901 @samp{/proc} that can be used to examine the image of a running
16902 process using file-system subroutines. If @value{GDBN} is configured
16903 for an operating system with this facility, the command @code{info
16904 proc} is available to report information about the process running
16905 your program, or about any process running on your system. @code{info
16906 proc} works only on SVR4 systems that include the @code{procfs} code.
16907 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
16908 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
16909
16910 @table @code
16911 @kindex info proc
16912 @cindex process ID
16913 @item info proc
16914 @itemx info proc @var{process-id}
16915 Summarize available information about any running process. If a
16916 process ID is specified by @var{process-id}, display information about
16917 that process; otherwise display information about the program being
16918 debugged. The summary includes the debugged process ID, the command
16919 line used to invoke it, its current working directory, and its
16920 executable file's absolute file name.
16921
16922 On some systems, @var{process-id} can be of the form
16923 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
16924 within a process. If the optional @var{pid} part is missing, it means
16925 a thread from the process being debugged (the leading @samp{/} still
16926 needs to be present, or else @value{GDBN} will interpret the number as
16927 a process ID rather than a thread ID).
16928
16929 @item info proc mappings
16930 @cindex memory address space mappings
16931 Report the memory address space ranges accessible in the program, with
16932 information on whether the process has read, write, or execute access
16933 rights to each range. On @sc{gnu}/Linux systems, each memory range
16934 includes the object file which is mapped to that range, instead of the
16935 memory access rights to that range.
16936
16937 @item info proc stat
16938 @itemx info proc status
16939 @cindex process detailed status information
16940 These subcommands are specific to @sc{gnu}/Linux systems. They show
16941 the process-related information, including the user ID and group ID;
16942 how many threads are there in the process; its virtual memory usage;
16943 the signals that are pending, blocked, and ignored; its TTY; its
16944 consumption of system and user time; its stack size; its @samp{nice}
16945 value; etc. For more information, see the @samp{proc} man page
16946 (type @kbd{man 5 proc} from your shell prompt).
16947
16948 @item info proc all
16949 Show all the information about the process described under all of the
16950 above @code{info proc} subcommands.
16951
16952 @ignore
16953 @comment These sub-options of 'info proc' were not included when
16954 @comment procfs.c was re-written. Keep their descriptions around
16955 @comment against the day when someone finds the time to put them back in.
16956 @kindex info proc times
16957 @item info proc times
16958 Starting time, user CPU time, and system CPU time for your program and
16959 its children.
16960
16961 @kindex info proc id
16962 @item info proc id
16963 Report on the process IDs related to your program: its own process ID,
16964 the ID of its parent, the process group ID, and the session ID.
16965 @end ignore
16966
16967 @item set procfs-trace
16968 @kindex set procfs-trace
16969 @cindex @code{procfs} API calls
16970 This command enables and disables tracing of @code{procfs} API calls.
16971
16972 @item show procfs-trace
16973 @kindex show procfs-trace
16974 Show the current state of @code{procfs} API call tracing.
16975
16976 @item set procfs-file @var{file}
16977 @kindex set procfs-file
16978 Tell @value{GDBN} to write @code{procfs} API trace to the named
16979 @var{file}. @value{GDBN} appends the trace info to the previous
16980 contents of the file. The default is to display the trace on the
16981 standard output.
16982
16983 @item show procfs-file
16984 @kindex show procfs-file
16985 Show the file to which @code{procfs} API trace is written.
16986
16987 @item proc-trace-entry
16988 @itemx proc-trace-exit
16989 @itemx proc-untrace-entry
16990 @itemx proc-untrace-exit
16991 @kindex proc-trace-entry
16992 @kindex proc-trace-exit
16993 @kindex proc-untrace-entry
16994 @kindex proc-untrace-exit
16995 These commands enable and disable tracing of entries into and exits
16996 from the @code{syscall} interface.
16997
16998 @item info pidlist
16999 @kindex info pidlist
17000 @cindex process list, QNX Neutrino
17001 For QNX Neutrino only, this command displays the list of all the
17002 processes and all the threads within each process.
17003
17004 @item info meminfo
17005 @kindex info meminfo
17006 @cindex mapinfo list, QNX Neutrino
17007 For QNX Neutrino only, this command displays the list of all mapinfos.
17008 @end table
17009
17010 @node DJGPP Native
17011 @subsection Features for Debugging @sc{djgpp} Programs
17012 @cindex @sc{djgpp} debugging
17013 @cindex native @sc{djgpp} debugging
17014 @cindex MS-DOS-specific commands
17015
17016 @cindex DPMI
17017 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
17018 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
17019 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
17020 top of real-mode DOS systems and their emulations.
17021
17022 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
17023 defines a few commands specific to the @sc{djgpp} port. This
17024 subsection describes those commands.
17025
17026 @table @code
17027 @kindex info dos
17028 @item info dos
17029 This is a prefix of @sc{djgpp}-specific commands which print
17030 information about the target system and important OS structures.
17031
17032 @kindex sysinfo
17033 @cindex MS-DOS system info
17034 @cindex free memory information (MS-DOS)
17035 @item info dos sysinfo
17036 This command displays assorted information about the underlying
17037 platform: the CPU type and features, the OS version and flavor, the
17038 DPMI version, and the available conventional and DPMI memory.
17039
17040 @cindex GDT
17041 @cindex LDT
17042 @cindex IDT
17043 @cindex segment descriptor tables
17044 @cindex descriptor tables display
17045 @item info dos gdt
17046 @itemx info dos ldt
17047 @itemx info dos idt
17048 These 3 commands display entries from, respectively, Global, Local,
17049 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
17050 tables are data structures which store a descriptor for each segment
17051 that is currently in use. The segment's selector is an index into a
17052 descriptor table; the table entry for that index holds the
17053 descriptor's base address and limit, and its attributes and access
17054 rights.
17055
17056 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
17057 segment (used for both data and the stack), and a DOS segment (which
17058 allows access to DOS/BIOS data structures and absolute addresses in
17059 conventional memory). However, the DPMI host will usually define
17060 additional segments in order to support the DPMI environment.
17061
17062 @cindex garbled pointers
17063 These commands allow to display entries from the descriptor tables.
17064 Without an argument, all entries from the specified table are
17065 displayed. An argument, which should be an integer expression, means
17066 display a single entry whose index is given by the argument. For
17067 example, here's a convenient way to display information about the
17068 debugged program's data segment:
17069
17070 @smallexample
17071 @exdent @code{(@value{GDBP}) info dos ldt $ds}
17072 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
17073 @end smallexample
17074
17075 @noindent
17076 This comes in handy when you want to see whether a pointer is outside
17077 the data segment's limit (i.e.@: @dfn{garbled}).
17078
17079 @cindex page tables display (MS-DOS)
17080 @item info dos pde
17081 @itemx info dos pte
17082 These two commands display entries from, respectively, the Page
17083 Directory and the Page Tables. Page Directories and Page Tables are
17084 data structures which control how virtual memory addresses are mapped
17085 into physical addresses. A Page Table includes an entry for every
17086 page of memory that is mapped into the program's address space; there
17087 may be several Page Tables, each one holding up to 4096 entries. A
17088 Page Directory has up to 4096 entries, one each for every Page Table
17089 that is currently in use.
17090
17091 Without an argument, @kbd{info dos pde} displays the entire Page
17092 Directory, and @kbd{info dos pte} displays all the entries in all of
17093 the Page Tables. An argument, an integer expression, given to the
17094 @kbd{info dos pde} command means display only that entry from the Page
17095 Directory table. An argument given to the @kbd{info dos pte} command
17096 means display entries from a single Page Table, the one pointed to by
17097 the specified entry in the Page Directory.
17098
17099 @cindex direct memory access (DMA) on MS-DOS
17100 These commands are useful when your program uses @dfn{DMA} (Direct
17101 Memory Access), which needs physical addresses to program the DMA
17102 controller.
17103
17104 These commands are supported only with some DPMI servers.
17105
17106 @cindex physical address from linear address
17107 @item info dos address-pte @var{addr}
17108 This command displays the Page Table entry for a specified linear
17109 address. The argument @var{addr} is a linear address which should
17110 already have the appropriate segment's base address added to it,
17111 because this command accepts addresses which may belong to @emph{any}
17112 segment. For example, here's how to display the Page Table entry for
17113 the page where a variable @code{i} is stored:
17114
17115 @smallexample
17116 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
17117 @exdent @code{Page Table entry for address 0x11a00d30:}
17118 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
17119 @end smallexample
17120
17121 @noindent
17122 This says that @code{i} is stored at offset @code{0xd30} from the page
17123 whose physical base address is @code{0x02698000}, and shows all the
17124 attributes of that page.
17125
17126 Note that you must cast the addresses of variables to a @code{char *},
17127 since otherwise the value of @code{__djgpp_base_address}, the base
17128 address of all variables and functions in a @sc{djgpp} program, will
17129 be added using the rules of C pointer arithmetics: if @code{i} is
17130 declared an @code{int}, @value{GDBN} will add 4 times the value of
17131 @code{__djgpp_base_address} to the address of @code{i}.
17132
17133 Here's another example, it displays the Page Table entry for the
17134 transfer buffer:
17135
17136 @smallexample
17137 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
17138 @exdent @code{Page Table entry for address 0x29110:}
17139 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
17140 @end smallexample
17141
17142 @noindent
17143 (The @code{+ 3} offset is because the transfer buffer's address is the
17144 3rd member of the @code{_go32_info_block} structure.) The output
17145 clearly shows that this DPMI server maps the addresses in conventional
17146 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
17147 linear (@code{0x29110}) addresses are identical.
17148
17149 This command is supported only with some DPMI servers.
17150 @end table
17151
17152 @cindex DOS serial data link, remote debugging
17153 In addition to native debugging, the DJGPP port supports remote
17154 debugging via a serial data link. The following commands are specific
17155 to remote serial debugging in the DJGPP port of @value{GDBN}.
17156
17157 @table @code
17158 @kindex set com1base
17159 @kindex set com1irq
17160 @kindex set com2base
17161 @kindex set com2irq
17162 @kindex set com3base
17163 @kindex set com3irq
17164 @kindex set com4base
17165 @kindex set com4irq
17166 @item set com1base @var{addr}
17167 This command sets the base I/O port address of the @file{COM1} serial
17168 port.
17169
17170 @item set com1irq @var{irq}
17171 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
17172 for the @file{COM1} serial port.
17173
17174 There are similar commands @samp{set com2base}, @samp{set com3irq},
17175 etc.@: for setting the port address and the @code{IRQ} lines for the
17176 other 3 COM ports.
17177
17178 @kindex show com1base
17179 @kindex show com1irq
17180 @kindex show com2base
17181 @kindex show com2irq
17182 @kindex show com3base
17183 @kindex show com3irq
17184 @kindex show com4base
17185 @kindex show com4irq
17186 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
17187 display the current settings of the base address and the @code{IRQ}
17188 lines used by the COM ports.
17189
17190 @item info serial
17191 @kindex info serial
17192 @cindex DOS serial port status
17193 This command prints the status of the 4 DOS serial ports. For each
17194 port, it prints whether it's active or not, its I/O base address and
17195 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
17196 counts of various errors encountered so far.
17197 @end table
17198
17199
17200 @node Cygwin Native
17201 @subsection Features for Debugging MS Windows PE Executables
17202 @cindex MS Windows debugging
17203 @cindex native Cygwin debugging
17204 @cindex Cygwin-specific commands
17205
17206 @value{GDBN} supports native debugging of MS Windows programs, including
17207 DLLs with and without symbolic debugging information.
17208
17209 @cindex Ctrl-BREAK, MS-Windows
17210 @cindex interrupt debuggee on MS-Windows
17211 MS-Windows programs that call @code{SetConsoleMode} to switch off the
17212 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
17213 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
17214 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
17215 sequence, which can be used to interrupt the debuggee even if it
17216 ignores @kbd{C-c}.
17217
17218 There are various additional Cygwin-specific commands, described in
17219 this section. Working with DLLs that have no debugging symbols is
17220 described in @ref{Non-debug DLL Symbols}.
17221
17222 @table @code
17223 @kindex info w32
17224 @item info w32
17225 This is a prefix of MS Windows-specific commands which print
17226 information about the target system and important OS structures.
17227
17228 @item info w32 selector
17229 This command displays information returned by
17230 the Win32 API @code{GetThreadSelectorEntry} function.
17231 It takes an optional argument that is evaluated to
17232 a long value to give the information about this given selector.
17233 Without argument, this command displays information
17234 about the six segment registers.
17235
17236 @item info w32 thread-information-block
17237 This command displays thread specific information stored in the
17238 Thread Information Block (readable on the X86 CPU family using @code{$fs}
17239 selector for 32-bit programs and @code{$gs} for 64-bit programs).
17240
17241 @kindex info dll
17242 @item info dll
17243 This is a Cygwin-specific alias of @code{info shared}.
17244
17245 @kindex dll-symbols
17246 @item dll-symbols
17247 This command loads symbols from a dll similarly to
17248 add-sym command but without the need to specify a base address.
17249
17250 @kindex set cygwin-exceptions
17251 @cindex debugging the Cygwin DLL
17252 @cindex Cygwin DLL, debugging
17253 @item set cygwin-exceptions @var{mode}
17254 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
17255 happen inside the Cygwin DLL. If @var{mode} is @code{off},
17256 @value{GDBN} will delay recognition of exceptions, and may ignore some
17257 exceptions which seem to be caused by internal Cygwin DLL
17258 ``bookkeeping''. This option is meant primarily for debugging the
17259 Cygwin DLL itself; the default value is @code{off} to avoid annoying
17260 @value{GDBN} users with false @code{SIGSEGV} signals.
17261
17262 @kindex show cygwin-exceptions
17263 @item show cygwin-exceptions
17264 Displays whether @value{GDBN} will break on exceptions that happen
17265 inside the Cygwin DLL itself.
17266
17267 @kindex set new-console
17268 @item set new-console @var{mode}
17269 If @var{mode} is @code{on} the debuggee will
17270 be started in a new console on next start.
17271 If @var{mode} is @code{off}, the debuggee will
17272 be started in the same console as the debugger.
17273
17274 @kindex show new-console
17275 @item show new-console
17276 Displays whether a new console is used
17277 when the debuggee is started.
17278
17279 @kindex set new-group
17280 @item set new-group @var{mode}
17281 This boolean value controls whether the debuggee should
17282 start a new group or stay in the same group as the debugger.
17283 This affects the way the Windows OS handles
17284 @samp{Ctrl-C}.
17285
17286 @kindex show new-group
17287 @item show new-group
17288 Displays current value of new-group boolean.
17289
17290 @kindex set debugevents
17291 @item set debugevents
17292 This boolean value adds debug output concerning kernel events related
17293 to the debuggee seen by the debugger. This includes events that
17294 signal thread and process creation and exit, DLL loading and
17295 unloading, console interrupts, and debugging messages produced by the
17296 Windows @code{OutputDebugString} API call.
17297
17298 @kindex set debugexec
17299 @item set debugexec
17300 This boolean value adds debug output concerning execute events
17301 (such as resume thread) seen by the debugger.
17302
17303 @kindex set debugexceptions
17304 @item set debugexceptions
17305 This boolean value adds debug output concerning exceptions in the
17306 debuggee seen by the debugger.
17307
17308 @kindex set debugmemory
17309 @item set debugmemory
17310 This boolean value adds debug output concerning debuggee memory reads
17311 and writes by the debugger.
17312
17313 @kindex set shell
17314 @item set shell
17315 This boolean values specifies whether the debuggee is called
17316 via a shell or directly (default value is on).
17317
17318 @kindex show shell
17319 @item show shell
17320 Displays if the debuggee will be started with a shell.
17321
17322 @end table
17323
17324 @menu
17325 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
17326 @end menu
17327
17328 @node Non-debug DLL Symbols
17329 @subsubsection Support for DLLs without Debugging Symbols
17330 @cindex DLLs with no debugging symbols
17331 @cindex Minimal symbols and DLLs
17332
17333 Very often on windows, some of the DLLs that your program relies on do
17334 not include symbolic debugging information (for example,
17335 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
17336 symbols in a DLL, it relies on the minimal amount of symbolic
17337 information contained in the DLL's export table. This section
17338 describes working with such symbols, known internally to @value{GDBN} as
17339 ``minimal symbols''.
17340
17341 Note that before the debugged program has started execution, no DLLs
17342 will have been loaded. The easiest way around this problem is simply to
17343 start the program --- either by setting a breakpoint or letting the
17344 program run once to completion. It is also possible to force
17345 @value{GDBN} to load a particular DLL before starting the executable ---
17346 see the shared library information in @ref{Files}, or the
17347 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
17348 explicitly loading symbols from a DLL with no debugging information will
17349 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
17350 which may adversely affect symbol lookup performance.
17351
17352 @subsubsection DLL Name Prefixes
17353
17354 In keeping with the naming conventions used by the Microsoft debugging
17355 tools, DLL export symbols are made available with a prefix based on the
17356 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
17357 also entered into the symbol table, so @code{CreateFileA} is often
17358 sufficient. In some cases there will be name clashes within a program
17359 (particularly if the executable itself includes full debugging symbols)
17360 necessitating the use of the fully qualified name when referring to the
17361 contents of the DLL. Use single-quotes around the name to avoid the
17362 exclamation mark (``!'') being interpreted as a language operator.
17363
17364 Note that the internal name of the DLL may be all upper-case, even
17365 though the file name of the DLL is lower-case, or vice-versa. Since
17366 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
17367 some confusion. If in doubt, try the @code{info functions} and
17368 @code{info variables} commands or even @code{maint print msymbols}
17369 (@pxref{Symbols}). Here's an example:
17370
17371 @smallexample
17372 (@value{GDBP}) info function CreateFileA
17373 All functions matching regular expression "CreateFileA":
17374
17375 Non-debugging symbols:
17376 0x77e885f4 CreateFileA
17377 0x77e885f4 KERNEL32!CreateFileA
17378 @end smallexample
17379
17380 @smallexample
17381 (@value{GDBP}) info function !
17382 All functions matching regular expression "!":
17383
17384 Non-debugging symbols:
17385 0x6100114c cygwin1!__assert
17386 0x61004034 cygwin1!_dll_crt0@@0
17387 0x61004240 cygwin1!dll_crt0(per_process *)
17388 [etc...]
17389 @end smallexample
17390
17391 @subsubsection Working with Minimal Symbols
17392
17393 Symbols extracted from a DLL's export table do not contain very much
17394 type information. All that @value{GDBN} can do is guess whether a symbol
17395 refers to a function or variable depending on the linker section that
17396 contains the symbol. Also note that the actual contents of the memory
17397 contained in a DLL are not available unless the program is running. This
17398 means that you cannot examine the contents of a variable or disassemble
17399 a function within a DLL without a running program.
17400
17401 Variables are generally treated as pointers and dereferenced
17402 automatically. For this reason, it is often necessary to prefix a
17403 variable name with the address-of operator (``&'') and provide explicit
17404 type information in the command. Here's an example of the type of
17405 problem:
17406
17407 @smallexample
17408 (@value{GDBP}) print 'cygwin1!__argv'
17409 $1 = 268572168
17410 @end smallexample
17411
17412 @smallexample
17413 (@value{GDBP}) x 'cygwin1!__argv'
17414 0x10021610: "\230y\""
17415 @end smallexample
17416
17417 And two possible solutions:
17418
17419 @smallexample
17420 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
17421 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
17422 @end smallexample
17423
17424 @smallexample
17425 (@value{GDBP}) x/2x &'cygwin1!__argv'
17426 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
17427 (@value{GDBP}) x/x 0x10021608
17428 0x10021608: 0x0022fd98
17429 (@value{GDBP}) x/s 0x0022fd98
17430 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
17431 @end smallexample
17432
17433 Setting a break point within a DLL is possible even before the program
17434 starts execution. However, under these circumstances, @value{GDBN} can't
17435 examine the initial instructions of the function in order to skip the
17436 function's frame set-up code. You can work around this by using ``*&''
17437 to set the breakpoint at a raw memory address:
17438
17439 @smallexample
17440 (@value{GDBP}) break *&'python22!PyOS_Readline'
17441 Breakpoint 1 at 0x1e04eff0
17442 @end smallexample
17443
17444 The author of these extensions is not entirely convinced that setting a
17445 break point within a shared DLL like @file{kernel32.dll} is completely
17446 safe.
17447
17448 @node Hurd Native
17449 @subsection Commands Specific to @sc{gnu} Hurd Systems
17450 @cindex @sc{gnu} Hurd debugging
17451
17452 This subsection describes @value{GDBN} commands specific to the
17453 @sc{gnu} Hurd native debugging.
17454
17455 @table @code
17456 @item set signals
17457 @itemx set sigs
17458 @kindex set signals@r{, Hurd command}
17459 @kindex set sigs@r{, Hurd command}
17460 This command toggles the state of inferior signal interception by
17461 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
17462 affected by this command. @code{sigs} is a shorthand alias for
17463 @code{signals}.
17464
17465 @item show signals
17466 @itemx show sigs
17467 @kindex show signals@r{, Hurd command}
17468 @kindex show sigs@r{, Hurd command}
17469 Show the current state of intercepting inferior's signals.
17470
17471 @item set signal-thread
17472 @itemx set sigthread
17473 @kindex set signal-thread
17474 @kindex set sigthread
17475 This command tells @value{GDBN} which thread is the @code{libc} signal
17476 thread. That thread is run when a signal is delivered to a running
17477 process. @code{set sigthread} is the shorthand alias of @code{set
17478 signal-thread}.
17479
17480 @item show signal-thread
17481 @itemx show sigthread
17482 @kindex show signal-thread
17483 @kindex show sigthread
17484 These two commands show which thread will run when the inferior is
17485 delivered a signal.
17486
17487 @item set stopped
17488 @kindex set stopped@r{, Hurd command}
17489 This commands tells @value{GDBN} that the inferior process is stopped,
17490 as with the @code{SIGSTOP} signal. The stopped process can be
17491 continued by delivering a signal to it.
17492
17493 @item show stopped
17494 @kindex show stopped@r{, Hurd command}
17495 This command shows whether @value{GDBN} thinks the debuggee is
17496 stopped.
17497
17498 @item set exceptions
17499 @kindex set exceptions@r{, Hurd command}
17500 Use this command to turn off trapping of exceptions in the inferior.
17501 When exception trapping is off, neither breakpoints nor
17502 single-stepping will work. To restore the default, set exception
17503 trapping on.
17504
17505 @item show exceptions
17506 @kindex show exceptions@r{, Hurd command}
17507 Show the current state of trapping exceptions in the inferior.
17508
17509 @item set task pause
17510 @kindex set task@r{, Hurd commands}
17511 @cindex task attributes (@sc{gnu} Hurd)
17512 @cindex pause current task (@sc{gnu} Hurd)
17513 This command toggles task suspension when @value{GDBN} has control.
17514 Setting it to on takes effect immediately, and the task is suspended
17515 whenever @value{GDBN} gets control. Setting it to off will take
17516 effect the next time the inferior is continued. If this option is set
17517 to off, you can use @code{set thread default pause on} or @code{set
17518 thread pause on} (see below) to pause individual threads.
17519
17520 @item show task pause
17521 @kindex show task@r{, Hurd commands}
17522 Show the current state of task suspension.
17523
17524 @item set task detach-suspend-count
17525 @cindex task suspend count
17526 @cindex detach from task, @sc{gnu} Hurd
17527 This command sets the suspend count the task will be left with when
17528 @value{GDBN} detaches from it.
17529
17530 @item show task detach-suspend-count
17531 Show the suspend count the task will be left with when detaching.
17532
17533 @item set task exception-port
17534 @itemx set task excp
17535 @cindex task exception port, @sc{gnu} Hurd
17536 This command sets the task exception port to which @value{GDBN} will
17537 forward exceptions. The argument should be the value of the @dfn{send
17538 rights} of the task. @code{set task excp} is a shorthand alias.
17539
17540 @item set noninvasive
17541 @cindex noninvasive task options
17542 This command switches @value{GDBN} to a mode that is the least
17543 invasive as far as interfering with the inferior is concerned. This
17544 is the same as using @code{set task pause}, @code{set exceptions}, and
17545 @code{set signals} to values opposite to the defaults.
17546
17547 @item info send-rights
17548 @itemx info receive-rights
17549 @itemx info port-rights
17550 @itemx info port-sets
17551 @itemx info dead-names
17552 @itemx info ports
17553 @itemx info psets
17554 @cindex send rights, @sc{gnu} Hurd
17555 @cindex receive rights, @sc{gnu} Hurd
17556 @cindex port rights, @sc{gnu} Hurd
17557 @cindex port sets, @sc{gnu} Hurd
17558 @cindex dead names, @sc{gnu} Hurd
17559 These commands display information about, respectively, send rights,
17560 receive rights, port rights, port sets, and dead names of a task.
17561 There are also shorthand aliases: @code{info ports} for @code{info
17562 port-rights} and @code{info psets} for @code{info port-sets}.
17563
17564 @item set thread pause
17565 @kindex set thread@r{, Hurd command}
17566 @cindex thread properties, @sc{gnu} Hurd
17567 @cindex pause current thread (@sc{gnu} Hurd)
17568 This command toggles current thread suspension when @value{GDBN} has
17569 control. Setting it to on takes effect immediately, and the current
17570 thread is suspended whenever @value{GDBN} gets control. Setting it to
17571 off will take effect the next time the inferior is continued.
17572 Normally, this command has no effect, since when @value{GDBN} has
17573 control, the whole task is suspended. However, if you used @code{set
17574 task pause off} (see above), this command comes in handy to suspend
17575 only the current thread.
17576
17577 @item show thread pause
17578 @kindex show thread@r{, Hurd command}
17579 This command shows the state of current thread suspension.
17580
17581 @item set thread run
17582 This command sets whether the current thread is allowed to run.
17583
17584 @item show thread run
17585 Show whether the current thread is allowed to run.
17586
17587 @item set thread detach-suspend-count
17588 @cindex thread suspend count, @sc{gnu} Hurd
17589 @cindex detach from thread, @sc{gnu} Hurd
17590 This command sets the suspend count @value{GDBN} will leave on a
17591 thread when detaching. This number is relative to the suspend count
17592 found by @value{GDBN} when it notices the thread; use @code{set thread
17593 takeover-suspend-count} to force it to an absolute value.
17594
17595 @item show thread detach-suspend-count
17596 Show the suspend count @value{GDBN} will leave on the thread when
17597 detaching.
17598
17599 @item set thread exception-port
17600 @itemx set thread excp
17601 Set the thread exception port to which to forward exceptions. This
17602 overrides the port set by @code{set task exception-port} (see above).
17603 @code{set thread excp} is the shorthand alias.
17604
17605 @item set thread takeover-suspend-count
17606 Normally, @value{GDBN}'s thread suspend counts are relative to the
17607 value @value{GDBN} finds when it notices each thread. This command
17608 changes the suspend counts to be absolute instead.
17609
17610 @item set thread default
17611 @itemx show thread default
17612 @cindex thread default settings, @sc{gnu} Hurd
17613 Each of the above @code{set thread} commands has a @code{set thread
17614 default} counterpart (e.g., @code{set thread default pause}, @code{set
17615 thread default exception-port}, etc.). The @code{thread default}
17616 variety of commands sets the default thread properties for all
17617 threads; you can then change the properties of individual threads with
17618 the non-default commands.
17619 @end table
17620
17621
17622 @node Neutrino
17623 @subsection QNX Neutrino
17624 @cindex QNX Neutrino
17625
17626 @value{GDBN} provides the following commands specific to the QNX
17627 Neutrino target:
17628
17629 @table @code
17630 @item set debug nto-debug
17631 @kindex set debug nto-debug
17632 When set to on, enables debugging messages specific to the QNX
17633 Neutrino support.
17634
17635 @item show debug nto-debug
17636 @kindex show debug nto-debug
17637 Show the current state of QNX Neutrino messages.
17638 @end table
17639
17640 @node Darwin
17641 @subsection Darwin
17642 @cindex Darwin
17643
17644 @value{GDBN} provides the following commands specific to the Darwin target:
17645
17646 @table @code
17647 @item set debug darwin @var{num}
17648 @kindex set debug darwin
17649 When set to a non zero value, enables debugging messages specific to
17650 the Darwin support. Higher values produce more verbose output.
17651
17652 @item show debug darwin
17653 @kindex show debug darwin
17654 Show the current state of Darwin messages.
17655
17656 @item set debug mach-o @var{num}
17657 @kindex set debug mach-o
17658 When set to a non zero value, enables debugging messages while
17659 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
17660 file format used on Darwin for object and executable files.) Higher
17661 values produce more verbose output. This is a command to diagnose
17662 problems internal to @value{GDBN} and should not be needed in normal
17663 usage.
17664
17665 @item show debug mach-o
17666 @kindex show debug mach-o
17667 Show the current state of Mach-O file messages.
17668
17669 @item set mach-exceptions on
17670 @itemx set mach-exceptions off
17671 @kindex set mach-exceptions
17672 On Darwin, faults are first reported as a Mach exception and are then
17673 mapped to a Posix signal. Use this command to turn on trapping of
17674 Mach exceptions in the inferior. This might be sometimes useful to
17675 better understand the cause of a fault. The default is off.
17676
17677 @item show mach-exceptions
17678 @kindex show mach-exceptions
17679 Show the current state of exceptions trapping.
17680 @end table
17681
17682
17683 @node Embedded OS
17684 @section Embedded Operating Systems
17685
17686 This section describes configurations involving the debugging of
17687 embedded operating systems that are available for several different
17688 architectures.
17689
17690 @menu
17691 * VxWorks:: Using @value{GDBN} with VxWorks
17692 @end menu
17693
17694 @value{GDBN} includes the ability to debug programs running on
17695 various real-time operating systems.
17696
17697 @node VxWorks
17698 @subsection Using @value{GDBN} with VxWorks
17699
17700 @cindex VxWorks
17701
17702 @table @code
17703
17704 @kindex target vxworks
17705 @item target vxworks @var{machinename}
17706 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
17707 is the target system's machine name or IP address.
17708
17709 @end table
17710
17711 On VxWorks, @code{load} links @var{filename} dynamically on the
17712 current target system as well as adding its symbols in @value{GDBN}.
17713
17714 @value{GDBN} enables developers to spawn and debug tasks running on networked
17715 VxWorks targets from a Unix host. Already-running tasks spawned from
17716 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
17717 both the Unix host and on the VxWorks target. The program
17718 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
17719 installed with the name @code{vxgdb}, to distinguish it from a
17720 @value{GDBN} for debugging programs on the host itself.)
17721
17722 @table @code
17723 @item VxWorks-timeout @var{args}
17724 @kindex vxworks-timeout
17725 All VxWorks-based targets now support the option @code{vxworks-timeout}.
17726 This option is set by the user, and @var{args} represents the number of
17727 seconds @value{GDBN} waits for responses to rpc's. You might use this if
17728 your VxWorks target is a slow software simulator or is on the far side
17729 of a thin network line.
17730 @end table
17731
17732 The following information on connecting to VxWorks was current when
17733 this manual was produced; newer releases of VxWorks may use revised
17734 procedures.
17735
17736 @findex INCLUDE_RDB
17737 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
17738 to include the remote debugging interface routines in the VxWorks
17739 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
17740 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
17741 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
17742 source debugging task @code{tRdbTask} when VxWorks is booted. For more
17743 information on configuring and remaking VxWorks, see the manufacturer's
17744 manual.
17745 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
17746
17747 Once you have included @file{rdb.a} in your VxWorks system image and set
17748 your Unix execution search path to find @value{GDBN}, you are ready to
17749 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
17750 @code{vxgdb}, depending on your installation).
17751
17752 @value{GDBN} comes up showing the prompt:
17753
17754 @smallexample
17755 (vxgdb)
17756 @end smallexample
17757
17758 @menu
17759 * VxWorks Connection:: Connecting to VxWorks
17760 * VxWorks Download:: VxWorks download
17761 * VxWorks Attach:: Running tasks
17762 @end menu
17763
17764 @node VxWorks Connection
17765 @subsubsection Connecting to VxWorks
17766
17767 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
17768 network. To connect to a target whose host name is ``@code{tt}'', type:
17769
17770 @smallexample
17771 (vxgdb) target vxworks tt
17772 @end smallexample
17773
17774 @need 750
17775 @value{GDBN} displays messages like these:
17776
17777 @smallexample
17778 Attaching remote machine across net...
17779 Connected to tt.
17780 @end smallexample
17781
17782 @need 1000
17783 @value{GDBN} then attempts to read the symbol tables of any object modules
17784 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
17785 these files by searching the directories listed in the command search
17786 path (@pxref{Environment, ,Your Program's Environment}); if it fails
17787 to find an object file, it displays a message such as:
17788
17789 @smallexample
17790 prog.o: No such file or directory.
17791 @end smallexample
17792
17793 When this happens, add the appropriate directory to the search path with
17794 the @value{GDBN} command @code{path}, and execute the @code{target}
17795 command again.
17796
17797 @node VxWorks Download
17798 @subsubsection VxWorks Download
17799
17800 @cindex download to VxWorks
17801 If you have connected to the VxWorks target and you want to debug an
17802 object that has not yet been loaded, you can use the @value{GDBN}
17803 @code{load} command to download a file from Unix to VxWorks
17804 incrementally. The object file given as an argument to the @code{load}
17805 command is actually opened twice: first by the VxWorks target in order
17806 to download the code, then by @value{GDBN} in order to read the symbol
17807 table. This can lead to problems if the current working directories on
17808 the two systems differ. If both systems have NFS mounted the same
17809 filesystems, you can avoid these problems by using absolute paths.
17810 Otherwise, it is simplest to set the working directory on both systems
17811 to the directory in which the object file resides, and then to reference
17812 the file by its name, without any path. For instance, a program
17813 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
17814 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
17815 program, type this on VxWorks:
17816
17817 @smallexample
17818 -> cd "@var{vxpath}/vw/demo/rdb"
17819 @end smallexample
17820
17821 @noindent
17822 Then, in @value{GDBN}, type:
17823
17824 @smallexample
17825 (vxgdb) cd @var{hostpath}/vw/demo/rdb
17826 (vxgdb) load prog.o
17827 @end smallexample
17828
17829 @value{GDBN} displays a response similar to this:
17830
17831 @smallexample
17832 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
17833 @end smallexample
17834
17835 You can also use the @code{load} command to reload an object module
17836 after editing and recompiling the corresponding source file. Note that
17837 this makes @value{GDBN} delete all currently-defined breakpoints,
17838 auto-displays, and convenience variables, and to clear the value
17839 history. (This is necessary in order to preserve the integrity of
17840 debugger's data structures that reference the target system's symbol
17841 table.)
17842
17843 @node VxWorks Attach
17844 @subsubsection Running Tasks
17845
17846 @cindex running VxWorks tasks
17847 You can also attach to an existing task using the @code{attach} command as
17848 follows:
17849
17850 @smallexample
17851 (vxgdb) attach @var{task}
17852 @end smallexample
17853
17854 @noindent
17855 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
17856 or suspended when you attach to it. Running tasks are suspended at
17857 the time of attachment.
17858
17859 @node Embedded Processors
17860 @section Embedded Processors
17861
17862 This section goes into details specific to particular embedded
17863 configurations.
17864
17865 @cindex send command to simulator
17866 Whenever a specific embedded processor has a simulator, @value{GDBN}
17867 allows to send an arbitrary command to the simulator.
17868
17869 @table @code
17870 @item sim @var{command}
17871 @kindex sim@r{, a command}
17872 Send an arbitrary @var{command} string to the simulator. Consult the
17873 documentation for the specific simulator in use for information about
17874 acceptable commands.
17875 @end table
17876
17877
17878 @menu
17879 * ARM:: ARM RDI
17880 * M32R/D:: Renesas M32R/D
17881 * M68K:: Motorola M68K
17882 * MicroBlaze:: Xilinx MicroBlaze
17883 * MIPS Embedded:: MIPS Embedded
17884 * OpenRISC 1000:: OpenRisc 1000
17885 * PA:: HP PA Embedded
17886 * PowerPC Embedded:: PowerPC Embedded
17887 * Sparclet:: Tsqware Sparclet
17888 * Sparclite:: Fujitsu Sparclite
17889 * Z8000:: Zilog Z8000
17890 * AVR:: Atmel AVR
17891 * CRIS:: CRIS
17892 * Super-H:: Renesas Super-H
17893 @end menu
17894
17895 @node ARM
17896 @subsection ARM
17897 @cindex ARM RDI
17898
17899 @table @code
17900 @kindex target rdi
17901 @item target rdi @var{dev}
17902 ARM Angel monitor, via RDI library interface to ADP protocol. You may
17903 use this target to communicate with both boards running the Angel
17904 monitor, or with the EmbeddedICE JTAG debug device.
17905
17906 @kindex target rdp
17907 @item target rdp @var{dev}
17908 ARM Demon monitor.
17909
17910 @end table
17911
17912 @value{GDBN} provides the following ARM-specific commands:
17913
17914 @table @code
17915 @item set arm disassembler
17916 @kindex set arm
17917 This commands selects from a list of disassembly styles. The
17918 @code{"std"} style is the standard style.
17919
17920 @item show arm disassembler
17921 @kindex show arm
17922 Show the current disassembly style.
17923
17924 @item set arm apcs32
17925 @cindex ARM 32-bit mode
17926 This command toggles ARM operation mode between 32-bit and 26-bit.
17927
17928 @item show arm apcs32
17929 Display the current usage of the ARM 32-bit mode.
17930
17931 @item set arm fpu @var{fputype}
17932 This command sets the ARM floating-point unit (FPU) type. The
17933 argument @var{fputype} can be one of these:
17934
17935 @table @code
17936 @item auto
17937 Determine the FPU type by querying the OS ABI.
17938 @item softfpa
17939 Software FPU, with mixed-endian doubles on little-endian ARM
17940 processors.
17941 @item fpa
17942 GCC-compiled FPA co-processor.
17943 @item softvfp
17944 Software FPU with pure-endian doubles.
17945 @item vfp
17946 VFP co-processor.
17947 @end table
17948
17949 @item show arm fpu
17950 Show the current type of the FPU.
17951
17952 @item set arm abi
17953 This command forces @value{GDBN} to use the specified ABI.
17954
17955 @item show arm abi
17956 Show the currently used ABI.
17957
17958 @item set arm fallback-mode (arm|thumb|auto)
17959 @value{GDBN} uses the symbol table, when available, to determine
17960 whether instructions are ARM or Thumb. This command controls
17961 @value{GDBN}'s default behavior when the symbol table is not
17962 available. The default is @samp{auto}, which causes @value{GDBN} to
17963 use the current execution mode (from the @code{T} bit in the @code{CPSR}
17964 register).
17965
17966 @item show arm fallback-mode
17967 Show the current fallback instruction mode.
17968
17969 @item set arm force-mode (arm|thumb|auto)
17970 This command overrides use of the symbol table to determine whether
17971 instructions are ARM or Thumb. The default is @samp{auto}, which
17972 causes @value{GDBN} to use the symbol table and then the setting
17973 of @samp{set arm fallback-mode}.
17974
17975 @item show arm force-mode
17976 Show the current forced instruction mode.
17977
17978 @item set debug arm
17979 Toggle whether to display ARM-specific debugging messages from the ARM
17980 target support subsystem.
17981
17982 @item show debug arm
17983 Show whether ARM-specific debugging messages are enabled.
17984 @end table
17985
17986 The following commands are available when an ARM target is debugged
17987 using the RDI interface:
17988
17989 @table @code
17990 @item rdilogfile @r{[}@var{file}@r{]}
17991 @kindex rdilogfile
17992 @cindex ADP (Angel Debugger Protocol) logging
17993 Set the filename for the ADP (Angel Debugger Protocol) packet log.
17994 With an argument, sets the log file to the specified @var{file}. With
17995 no argument, show the current log file name. The default log file is
17996 @file{rdi.log}.
17997
17998 @item rdilogenable @r{[}@var{arg}@r{]}
17999 @kindex rdilogenable
18000 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
18001 enables logging, with an argument 0 or @code{"no"} disables it. With
18002 no arguments displays the current setting. When logging is enabled,
18003 ADP packets exchanged between @value{GDBN} and the RDI target device
18004 are logged to a file.
18005
18006 @item set rdiromatzero
18007 @kindex set rdiromatzero
18008 @cindex ROM at zero address, RDI
18009 Tell @value{GDBN} whether the target has ROM at address 0. If on,
18010 vector catching is disabled, so that zero address can be used. If off
18011 (the default), vector catching is enabled. For this command to take
18012 effect, it needs to be invoked prior to the @code{target rdi} command.
18013
18014 @item show rdiromatzero
18015 @kindex show rdiromatzero
18016 Show the current setting of ROM at zero address.
18017
18018 @item set rdiheartbeat
18019 @kindex set rdiheartbeat
18020 @cindex RDI heartbeat
18021 Enable or disable RDI heartbeat packets. It is not recommended to
18022 turn on this option, since it confuses ARM and EPI JTAG interface, as
18023 well as the Angel monitor.
18024
18025 @item show rdiheartbeat
18026 @kindex show rdiheartbeat
18027 Show the setting of RDI heartbeat packets.
18028 @end table
18029
18030 @table @code
18031 @item target sim @r{[}@var{simargs}@r{]} @dots{}
18032 The @value{GDBN} ARM simulator accepts the following optional arguments.
18033
18034 @table @code
18035 @item --swi-support=@var{type}
18036 Tell the simulator which SWI interfaces to support.
18037 @var{type} may be a comma separated list of the following values.
18038 The default value is @code{all}.
18039
18040 @table @code
18041 @item none
18042 @item demon
18043 @item angel
18044 @item redboot
18045 @item all
18046 @end table
18047 @end table
18048 @end table
18049
18050 @node M32R/D
18051 @subsection Renesas M32R/D and M32R/SDI
18052
18053 @table @code
18054 @kindex target m32r
18055 @item target m32r @var{dev}
18056 Renesas M32R/D ROM monitor.
18057
18058 @kindex target m32rsdi
18059 @item target m32rsdi @var{dev}
18060 Renesas M32R SDI server, connected via parallel port to the board.
18061 @end table
18062
18063 The following @value{GDBN} commands are specific to the M32R monitor:
18064
18065 @table @code
18066 @item set download-path @var{path}
18067 @kindex set download-path
18068 @cindex find downloadable @sc{srec} files (M32R)
18069 Set the default path for finding downloadable @sc{srec} files.
18070
18071 @item show download-path
18072 @kindex show download-path
18073 Show the default path for downloadable @sc{srec} files.
18074
18075 @item set board-address @var{addr}
18076 @kindex set board-address
18077 @cindex M32-EVA target board address
18078 Set the IP address for the M32R-EVA target board.
18079
18080 @item show board-address
18081 @kindex show board-address
18082 Show the current IP address of the target board.
18083
18084 @item set server-address @var{addr}
18085 @kindex set server-address
18086 @cindex download server address (M32R)
18087 Set the IP address for the download server, which is the @value{GDBN}'s
18088 host machine.
18089
18090 @item show server-address
18091 @kindex show server-address
18092 Display the IP address of the download server.
18093
18094 @item upload @r{[}@var{file}@r{]}
18095 @kindex upload@r{, M32R}
18096 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
18097 upload capability. If no @var{file} argument is given, the current
18098 executable file is uploaded.
18099
18100 @item tload @r{[}@var{file}@r{]}
18101 @kindex tload@r{, M32R}
18102 Test the @code{upload} command.
18103 @end table
18104
18105 The following commands are available for M32R/SDI:
18106
18107 @table @code
18108 @item sdireset
18109 @kindex sdireset
18110 @cindex reset SDI connection, M32R
18111 This command resets the SDI connection.
18112
18113 @item sdistatus
18114 @kindex sdistatus
18115 This command shows the SDI connection status.
18116
18117 @item debug_chaos
18118 @kindex debug_chaos
18119 @cindex M32R/Chaos debugging
18120 Instructs the remote that M32R/Chaos debugging is to be used.
18121
18122 @item use_debug_dma
18123 @kindex use_debug_dma
18124 Instructs the remote to use the DEBUG_DMA method of accessing memory.
18125
18126 @item use_mon_code
18127 @kindex use_mon_code
18128 Instructs the remote to use the MON_CODE method of accessing memory.
18129
18130 @item use_ib_break
18131 @kindex use_ib_break
18132 Instructs the remote to set breakpoints by IB break.
18133
18134 @item use_dbt_break
18135 @kindex use_dbt_break
18136 Instructs the remote to set breakpoints by DBT.
18137 @end table
18138
18139 @node M68K
18140 @subsection M68k
18141
18142 The Motorola m68k configuration includes ColdFire support, and a
18143 target command for the following ROM monitor.
18144
18145 @table @code
18146
18147 @kindex target dbug
18148 @item target dbug @var{dev}
18149 dBUG ROM monitor for Motorola ColdFire.
18150
18151 @end table
18152
18153 @node MicroBlaze
18154 @subsection MicroBlaze
18155 @cindex Xilinx MicroBlaze
18156 @cindex XMD, Xilinx Microprocessor Debugger
18157
18158 The MicroBlaze is a soft-core processor supported on various Xilinx
18159 FPGAs, such as Spartan or Virtex series. Boards with these processors
18160 usually have JTAG ports which connect to a host system running the Xilinx
18161 Embedded Development Kit (EDK) or Software Development Kit (SDK).
18162 This host system is used to download the configuration bitstream to
18163 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
18164 communicates with the target board using the JTAG interface and
18165 presents a @code{gdbserver} interface to the board. By default
18166 @code{xmd} uses port @code{1234}. (While it is possible to change
18167 this default port, it requires the use of undocumented @code{xmd}
18168 commands. Contact Xilinx support if you need to do this.)
18169
18170 Use these GDB commands to connect to the MicroBlaze target processor.
18171
18172 @table @code
18173 @item target remote :1234
18174 Use this command to connect to the target if you are running @value{GDBN}
18175 on the same system as @code{xmd}.
18176
18177 @item target remote @var{xmd-host}:1234
18178 Use this command to connect to the target if it is connected to @code{xmd}
18179 running on a different system named @var{xmd-host}.
18180
18181 @item load
18182 Use this command to download a program to the MicroBlaze target.
18183
18184 @item set debug microblaze @var{n}
18185 Enable MicroBlaze-specific debugging messages if non-zero.
18186
18187 @item show debug microblaze @var{n}
18188 Show MicroBlaze-specific debugging level.
18189 @end table
18190
18191 @node MIPS Embedded
18192 @subsection MIPS Embedded
18193
18194 @cindex MIPS boards
18195 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
18196 MIPS board attached to a serial line. This is available when
18197 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
18198
18199 @need 1000
18200 Use these @value{GDBN} commands to specify the connection to your target board:
18201
18202 @table @code
18203 @item target mips @var{port}
18204 @kindex target mips @var{port}
18205 To run a program on the board, start up @code{@value{GDBP}} with the
18206 name of your program as the argument. To connect to the board, use the
18207 command @samp{target mips @var{port}}, where @var{port} is the name of
18208 the serial port connected to the board. If the program has not already
18209 been downloaded to the board, you may use the @code{load} command to
18210 download it. You can then use all the usual @value{GDBN} commands.
18211
18212 For example, this sequence connects to the target board through a serial
18213 port, and loads and runs a program called @var{prog} through the
18214 debugger:
18215
18216 @smallexample
18217 host$ @value{GDBP} @var{prog}
18218 @value{GDBN} is free software and @dots{}
18219 (@value{GDBP}) target mips /dev/ttyb
18220 (@value{GDBP}) load @var{prog}
18221 (@value{GDBP}) run
18222 @end smallexample
18223
18224 @item target mips @var{hostname}:@var{portnumber}
18225 On some @value{GDBN} host configurations, you can specify a TCP
18226 connection (for instance, to a serial line managed by a terminal
18227 concentrator) instead of a serial port, using the syntax
18228 @samp{@var{hostname}:@var{portnumber}}.
18229
18230 @item target pmon @var{port}
18231 @kindex target pmon @var{port}
18232 PMON ROM monitor.
18233
18234 @item target ddb @var{port}
18235 @kindex target ddb @var{port}
18236 NEC's DDB variant of PMON for Vr4300.
18237
18238 @item target lsi @var{port}
18239 @kindex target lsi @var{port}
18240 LSI variant of PMON.
18241
18242 @kindex target r3900
18243 @item target r3900 @var{dev}
18244 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
18245
18246 @kindex target array
18247 @item target array @var{dev}
18248 Array Tech LSI33K RAID controller board.
18249
18250 @end table
18251
18252
18253 @noindent
18254 @value{GDBN} also supports these special commands for MIPS targets:
18255
18256 @table @code
18257 @item set mipsfpu double
18258 @itemx set mipsfpu single
18259 @itemx set mipsfpu none
18260 @itemx set mipsfpu auto
18261 @itemx show mipsfpu
18262 @kindex set mipsfpu
18263 @kindex show mipsfpu
18264 @cindex MIPS remote floating point
18265 @cindex floating point, MIPS remote
18266 If your target board does not support the MIPS floating point
18267 coprocessor, you should use the command @samp{set mipsfpu none} (if you
18268 need this, you may wish to put the command in your @value{GDBN} init
18269 file). This tells @value{GDBN} how to find the return value of
18270 functions which return floating point values. It also allows
18271 @value{GDBN} to avoid saving the floating point registers when calling
18272 functions on the board. If you are using a floating point coprocessor
18273 with only single precision floating point support, as on the @sc{r4650}
18274 processor, use the command @samp{set mipsfpu single}. The default
18275 double precision floating point coprocessor may be selected using
18276 @samp{set mipsfpu double}.
18277
18278 In previous versions the only choices were double precision or no
18279 floating point, so @samp{set mipsfpu on} will select double precision
18280 and @samp{set mipsfpu off} will select no floating point.
18281
18282 As usual, you can inquire about the @code{mipsfpu} variable with
18283 @samp{show mipsfpu}.
18284
18285 @item set timeout @var{seconds}
18286 @itemx set retransmit-timeout @var{seconds}
18287 @itemx show timeout
18288 @itemx show retransmit-timeout
18289 @cindex @code{timeout}, MIPS protocol
18290 @cindex @code{retransmit-timeout}, MIPS protocol
18291 @kindex set timeout
18292 @kindex show timeout
18293 @kindex set retransmit-timeout
18294 @kindex show retransmit-timeout
18295 You can control the timeout used while waiting for a packet, in the MIPS
18296 remote protocol, with the @code{set timeout @var{seconds}} command. The
18297 default is 5 seconds. Similarly, you can control the timeout used while
18298 waiting for an acknowledgment of a packet with the @code{set
18299 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
18300 You can inspect both values with @code{show timeout} and @code{show
18301 retransmit-timeout}. (These commands are @emph{only} available when
18302 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
18303
18304 The timeout set by @code{set timeout} does not apply when @value{GDBN}
18305 is waiting for your program to stop. In that case, @value{GDBN} waits
18306 forever because it has no way of knowing how long the program is going
18307 to run before stopping.
18308
18309 @item set syn-garbage-limit @var{num}
18310 @kindex set syn-garbage-limit@r{, MIPS remote}
18311 @cindex synchronize with remote MIPS target
18312 Limit the maximum number of characters @value{GDBN} should ignore when
18313 it tries to synchronize with the remote target. The default is 10
18314 characters. Setting the limit to -1 means there's no limit.
18315
18316 @item show syn-garbage-limit
18317 @kindex show syn-garbage-limit@r{, MIPS remote}
18318 Show the current limit on the number of characters to ignore when
18319 trying to synchronize with the remote system.
18320
18321 @item set monitor-prompt @var{prompt}
18322 @kindex set monitor-prompt@r{, MIPS remote}
18323 @cindex remote monitor prompt
18324 Tell @value{GDBN} to expect the specified @var{prompt} string from the
18325 remote monitor. The default depends on the target:
18326 @table @asis
18327 @item pmon target
18328 @samp{PMON}
18329 @item ddb target
18330 @samp{NEC010}
18331 @item lsi target
18332 @samp{PMON>}
18333 @end table
18334
18335 @item show monitor-prompt
18336 @kindex show monitor-prompt@r{, MIPS remote}
18337 Show the current strings @value{GDBN} expects as the prompt from the
18338 remote monitor.
18339
18340 @item set monitor-warnings
18341 @kindex set monitor-warnings@r{, MIPS remote}
18342 Enable or disable monitor warnings about hardware breakpoints. This
18343 has effect only for the @code{lsi} target. When on, @value{GDBN} will
18344 display warning messages whose codes are returned by the @code{lsi}
18345 PMON monitor for breakpoint commands.
18346
18347 @item show monitor-warnings
18348 @kindex show monitor-warnings@r{, MIPS remote}
18349 Show the current setting of printing monitor warnings.
18350
18351 @item pmon @var{command}
18352 @kindex pmon@r{, MIPS remote}
18353 @cindex send PMON command
18354 This command allows sending an arbitrary @var{command} string to the
18355 monitor. The monitor must be in debug mode for this to work.
18356 @end table
18357
18358 @node OpenRISC 1000
18359 @subsection OpenRISC 1000
18360 @cindex OpenRISC 1000
18361
18362 @cindex or1k boards
18363 See OR1k Architecture document (@uref{www.opencores.org}) for more information
18364 about platform and commands.
18365
18366 @table @code
18367
18368 @kindex target jtag
18369 @item target jtag jtag://@var{host}:@var{port}
18370
18371 Connects to remote JTAG server.
18372 JTAG remote server can be either an or1ksim or JTAG server,
18373 connected via parallel port to the board.
18374
18375 Example: @code{target jtag jtag://localhost:9999}
18376
18377 @kindex or1ksim
18378 @item or1ksim @var{command}
18379 If connected to @code{or1ksim} OpenRISC 1000 Architectural
18380 Simulator, proprietary commands can be executed.
18381
18382 @kindex info or1k spr
18383 @item info or1k spr
18384 Displays spr groups.
18385
18386 @item info or1k spr @var{group}
18387 @itemx info or1k spr @var{groupno}
18388 Displays register names in selected group.
18389
18390 @item info or1k spr @var{group} @var{register}
18391 @itemx info or1k spr @var{register}
18392 @itemx info or1k spr @var{groupno} @var{registerno}
18393 @itemx info or1k spr @var{registerno}
18394 Shows information about specified spr register.
18395
18396 @kindex spr
18397 @item spr @var{group} @var{register} @var{value}
18398 @itemx spr @var{register @var{value}}
18399 @itemx spr @var{groupno} @var{registerno @var{value}}
18400 @itemx spr @var{registerno @var{value}}
18401 Writes @var{value} to specified spr register.
18402 @end table
18403
18404 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
18405 It is very similar to @value{GDBN} trace, except it does not interfere with normal
18406 program execution and is thus much faster. Hardware breakpoints/watchpoint
18407 triggers can be set using:
18408 @table @code
18409 @item $LEA/$LDATA
18410 Load effective address/data
18411 @item $SEA/$SDATA
18412 Store effective address/data
18413 @item $AEA/$ADATA
18414 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
18415 @item $FETCH
18416 Fetch data
18417 @end table
18418
18419 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
18420 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
18421
18422 @code{htrace} commands:
18423 @cindex OpenRISC 1000 htrace
18424 @table @code
18425 @kindex hwatch
18426 @item hwatch @var{conditional}
18427 Set hardware watchpoint on combination of Load/Store Effective Address(es)
18428 or Data. For example:
18429
18430 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
18431
18432 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
18433
18434 @kindex htrace
18435 @item htrace info
18436 Display information about current HW trace configuration.
18437
18438 @item htrace trigger @var{conditional}
18439 Set starting criteria for HW trace.
18440
18441 @item htrace qualifier @var{conditional}
18442 Set acquisition qualifier for HW trace.
18443
18444 @item htrace stop @var{conditional}
18445 Set HW trace stopping criteria.
18446
18447 @item htrace record [@var{data}]*
18448 Selects the data to be recorded, when qualifier is met and HW trace was
18449 triggered.
18450
18451 @item htrace enable
18452 @itemx htrace disable
18453 Enables/disables the HW trace.
18454
18455 @item htrace rewind [@var{filename}]
18456 Clears currently recorded trace data.
18457
18458 If filename is specified, new trace file is made and any newly collected data
18459 will be written there.
18460
18461 @item htrace print [@var{start} [@var{len}]]
18462 Prints trace buffer, using current record configuration.
18463
18464 @item htrace mode continuous
18465 Set continuous trace mode.
18466
18467 @item htrace mode suspend
18468 Set suspend trace mode.
18469
18470 @end table
18471
18472 @node PowerPC Embedded
18473 @subsection PowerPC Embedded
18474
18475 @cindex DVC register
18476 @value{GDBN} supports using the DVC (Data Value Compare) register to
18477 implement in hardware simple hardware watchpoint conditions of the form:
18478
18479 @smallexample
18480 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
18481 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
18482 @end smallexample
18483
18484 The DVC register will be automatically used whenever @value{GDBN} detects
18485 such pattern in a condition expression. This feature is available in native
18486 @value{GDBN} running on a Linux kernel version 2.6.34 or newer.
18487
18488 @value{GDBN} provides the following PowerPC-specific commands:
18489
18490 @table @code
18491 @kindex set powerpc
18492 @item set powerpc soft-float
18493 @itemx show powerpc soft-float
18494 Force @value{GDBN} to use (or not use) a software floating point calling
18495 convention. By default, @value{GDBN} selects the calling convention based
18496 on the selected architecture and the provided executable file.
18497
18498 @item set powerpc vector-abi
18499 @itemx show powerpc vector-abi
18500 Force @value{GDBN} to use the specified calling convention for vector
18501 arguments and return values. The valid options are @samp{auto};
18502 @samp{generic}, to avoid vector registers even if they are present;
18503 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
18504 registers. By default, @value{GDBN} selects the calling convention
18505 based on the selected architecture and the provided executable file.
18506
18507 @kindex target dink32
18508 @item target dink32 @var{dev}
18509 DINK32 ROM monitor.
18510
18511 @kindex target ppcbug
18512 @item target ppcbug @var{dev}
18513 @kindex target ppcbug1
18514 @item target ppcbug1 @var{dev}
18515 PPCBUG ROM monitor for PowerPC.
18516
18517 @kindex target sds
18518 @item target sds @var{dev}
18519 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
18520 @end table
18521
18522 @cindex SDS protocol
18523 The following commands specific to the SDS protocol are supported
18524 by @value{GDBN}:
18525
18526 @table @code
18527 @item set sdstimeout @var{nsec}
18528 @kindex set sdstimeout
18529 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
18530 default is 2 seconds.
18531
18532 @item show sdstimeout
18533 @kindex show sdstimeout
18534 Show the current value of the SDS timeout.
18535
18536 @item sds @var{command}
18537 @kindex sds@r{, a command}
18538 Send the specified @var{command} string to the SDS monitor.
18539 @end table
18540
18541
18542 @node PA
18543 @subsection HP PA Embedded
18544
18545 @table @code
18546
18547 @kindex target op50n
18548 @item target op50n @var{dev}
18549 OP50N monitor, running on an OKI HPPA board.
18550
18551 @kindex target w89k
18552 @item target w89k @var{dev}
18553 W89K monitor, running on a Winbond HPPA board.
18554
18555 @end table
18556
18557 @node Sparclet
18558 @subsection Tsqware Sparclet
18559
18560 @cindex Sparclet
18561
18562 @value{GDBN} enables developers to debug tasks running on
18563 Sparclet targets from a Unix host.
18564 @value{GDBN} uses code that runs on
18565 both the Unix host and on the Sparclet target. The program
18566 @code{@value{GDBP}} is installed and executed on the Unix host.
18567
18568 @table @code
18569 @item remotetimeout @var{args}
18570 @kindex remotetimeout
18571 @value{GDBN} supports the option @code{remotetimeout}.
18572 This option is set by the user, and @var{args} represents the number of
18573 seconds @value{GDBN} waits for responses.
18574 @end table
18575
18576 @cindex compiling, on Sparclet
18577 When compiling for debugging, include the options @samp{-g} to get debug
18578 information and @samp{-Ttext} to relocate the program to where you wish to
18579 load it on the target. You may also want to add the options @samp{-n} or
18580 @samp{-N} in order to reduce the size of the sections. Example:
18581
18582 @smallexample
18583 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
18584 @end smallexample
18585
18586 You can use @code{objdump} to verify that the addresses are what you intended:
18587
18588 @smallexample
18589 sparclet-aout-objdump --headers --syms prog
18590 @end smallexample
18591
18592 @cindex running, on Sparclet
18593 Once you have set
18594 your Unix execution search path to find @value{GDBN}, you are ready to
18595 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
18596 (or @code{sparclet-aout-gdb}, depending on your installation).
18597
18598 @value{GDBN} comes up showing the prompt:
18599
18600 @smallexample
18601 (gdbslet)
18602 @end smallexample
18603
18604 @menu
18605 * Sparclet File:: Setting the file to debug
18606 * Sparclet Connection:: Connecting to Sparclet
18607 * Sparclet Download:: Sparclet download
18608 * Sparclet Execution:: Running and debugging
18609 @end menu
18610
18611 @node Sparclet File
18612 @subsubsection Setting File to Debug
18613
18614 The @value{GDBN} command @code{file} lets you choose with program to debug.
18615
18616 @smallexample
18617 (gdbslet) file prog
18618 @end smallexample
18619
18620 @need 1000
18621 @value{GDBN} then attempts to read the symbol table of @file{prog}.
18622 @value{GDBN} locates
18623 the file by searching the directories listed in the command search
18624 path.
18625 If the file was compiled with debug information (option @samp{-g}), source
18626 files will be searched as well.
18627 @value{GDBN} locates
18628 the source files by searching the directories listed in the directory search
18629 path (@pxref{Environment, ,Your Program's Environment}).
18630 If it fails
18631 to find a file, it displays a message such as:
18632
18633 @smallexample
18634 prog: No such file or directory.
18635 @end smallexample
18636
18637 When this happens, add the appropriate directories to the search paths with
18638 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
18639 @code{target} command again.
18640
18641 @node Sparclet Connection
18642 @subsubsection Connecting to Sparclet
18643
18644 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
18645 To connect to a target on serial port ``@code{ttya}'', type:
18646
18647 @smallexample
18648 (gdbslet) target sparclet /dev/ttya
18649 Remote target sparclet connected to /dev/ttya
18650 main () at ../prog.c:3
18651 @end smallexample
18652
18653 @need 750
18654 @value{GDBN} displays messages like these:
18655
18656 @smallexample
18657 Connected to ttya.
18658 @end smallexample
18659
18660 @node Sparclet Download
18661 @subsubsection Sparclet Download
18662
18663 @cindex download to Sparclet
18664 Once connected to the Sparclet target,
18665 you can use the @value{GDBN}
18666 @code{load} command to download the file from the host to the target.
18667 The file name and load offset should be given as arguments to the @code{load}
18668 command.
18669 Since the file format is aout, the program must be loaded to the starting
18670 address. You can use @code{objdump} to find out what this value is. The load
18671 offset is an offset which is added to the VMA (virtual memory address)
18672 of each of the file's sections.
18673 For instance, if the program
18674 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
18675 and bss at 0x12010170, in @value{GDBN}, type:
18676
18677 @smallexample
18678 (gdbslet) load prog 0x12010000
18679 Loading section .text, size 0xdb0 vma 0x12010000
18680 @end smallexample
18681
18682 If the code is loaded at a different address then what the program was linked
18683 to, you may need to use the @code{section} and @code{add-symbol-file} commands
18684 to tell @value{GDBN} where to map the symbol table.
18685
18686 @node Sparclet Execution
18687 @subsubsection Running and Debugging
18688
18689 @cindex running and debugging Sparclet programs
18690 You can now begin debugging the task using @value{GDBN}'s execution control
18691 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
18692 manual for the list of commands.
18693
18694 @smallexample
18695 (gdbslet) b main
18696 Breakpoint 1 at 0x12010000: file prog.c, line 3.
18697 (gdbslet) run
18698 Starting program: prog
18699 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
18700 3 char *symarg = 0;
18701 (gdbslet) step
18702 4 char *execarg = "hello!";
18703 (gdbslet)
18704 @end smallexample
18705
18706 @node Sparclite
18707 @subsection Fujitsu Sparclite
18708
18709 @table @code
18710
18711 @kindex target sparclite
18712 @item target sparclite @var{dev}
18713 Fujitsu sparclite boards, used only for the purpose of loading.
18714 You must use an additional command to debug the program.
18715 For example: target remote @var{dev} using @value{GDBN} standard
18716 remote protocol.
18717
18718 @end table
18719
18720 @node Z8000
18721 @subsection Zilog Z8000
18722
18723 @cindex Z8000
18724 @cindex simulator, Z8000
18725 @cindex Zilog Z8000 simulator
18726
18727 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
18728 a Z8000 simulator.
18729
18730 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
18731 unsegmented variant of the Z8000 architecture) or the Z8001 (the
18732 segmented variant). The simulator recognizes which architecture is
18733 appropriate by inspecting the object code.
18734
18735 @table @code
18736 @item target sim @var{args}
18737 @kindex sim
18738 @kindex target sim@r{, with Z8000}
18739 Debug programs on a simulated CPU. If the simulator supports setup
18740 options, specify them via @var{args}.
18741 @end table
18742
18743 @noindent
18744 After specifying this target, you can debug programs for the simulated
18745 CPU in the same style as programs for your host computer; use the
18746 @code{file} command to load a new program image, the @code{run} command
18747 to run your program, and so on.
18748
18749 As well as making available all the usual machine registers
18750 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
18751 additional items of information as specially named registers:
18752
18753 @table @code
18754
18755 @item cycles
18756 Counts clock-ticks in the simulator.
18757
18758 @item insts
18759 Counts instructions run in the simulator.
18760
18761 @item time
18762 Execution time in 60ths of a second.
18763
18764 @end table
18765
18766 You can refer to these values in @value{GDBN} expressions with the usual
18767 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
18768 conditional breakpoint that suspends only after at least 5000
18769 simulated clock ticks.
18770
18771 @node AVR
18772 @subsection Atmel AVR
18773 @cindex AVR
18774
18775 When configured for debugging the Atmel AVR, @value{GDBN} supports the
18776 following AVR-specific commands:
18777
18778 @table @code
18779 @item info io_registers
18780 @kindex info io_registers@r{, AVR}
18781 @cindex I/O registers (Atmel AVR)
18782 This command displays information about the AVR I/O registers. For
18783 each register, @value{GDBN} prints its number and value.
18784 @end table
18785
18786 @node CRIS
18787 @subsection CRIS
18788 @cindex CRIS
18789
18790 When configured for debugging CRIS, @value{GDBN} provides the
18791 following CRIS-specific commands:
18792
18793 @table @code
18794 @item set cris-version @var{ver}
18795 @cindex CRIS version
18796 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
18797 The CRIS version affects register names and sizes. This command is useful in
18798 case autodetection of the CRIS version fails.
18799
18800 @item show cris-version
18801 Show the current CRIS version.
18802
18803 @item set cris-dwarf2-cfi
18804 @cindex DWARF-2 CFI and CRIS
18805 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
18806 Change to @samp{off} when using @code{gcc-cris} whose version is below
18807 @code{R59}.
18808
18809 @item show cris-dwarf2-cfi
18810 Show the current state of using DWARF-2 CFI.
18811
18812 @item set cris-mode @var{mode}
18813 @cindex CRIS mode
18814 Set the current CRIS mode to @var{mode}. It should only be changed when
18815 debugging in guru mode, in which case it should be set to
18816 @samp{guru} (the default is @samp{normal}).
18817
18818 @item show cris-mode
18819 Show the current CRIS mode.
18820 @end table
18821
18822 @node Super-H
18823 @subsection Renesas Super-H
18824 @cindex Super-H
18825
18826 For the Renesas Super-H processor, @value{GDBN} provides these
18827 commands:
18828
18829 @table @code
18830 @item regs
18831 @kindex regs@r{, Super-H}
18832 Show the values of all Super-H registers.
18833
18834 @item set sh calling-convention @var{convention}
18835 @kindex set sh calling-convention
18836 Set the calling-convention used when calling functions from @value{GDBN}.
18837 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
18838 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
18839 convention. If the DWARF-2 information of the called function specifies
18840 that the function follows the Renesas calling convention, the function
18841 is called using the Renesas calling convention. If the calling convention
18842 is set to @samp{renesas}, the Renesas calling convention is always used,
18843 regardless of the DWARF-2 information. This can be used to override the
18844 default of @samp{gcc} if debug information is missing, or the compiler
18845 does not emit the DWARF-2 calling convention entry for a function.
18846
18847 @item show sh calling-convention
18848 @kindex show sh calling-convention
18849 Show the current calling convention setting.
18850
18851 @end table
18852
18853
18854 @node Architectures
18855 @section Architectures
18856
18857 This section describes characteristics of architectures that affect
18858 all uses of @value{GDBN} with the architecture, both native and cross.
18859
18860 @menu
18861 * i386::
18862 * A29K::
18863 * Alpha::
18864 * MIPS::
18865 * HPPA:: HP PA architecture
18866 * SPU:: Cell Broadband Engine SPU architecture
18867 * PowerPC::
18868 @end menu
18869
18870 @node i386
18871 @subsection x86 Architecture-specific Issues
18872
18873 @table @code
18874 @item set struct-convention @var{mode}
18875 @kindex set struct-convention
18876 @cindex struct return convention
18877 @cindex struct/union returned in registers
18878 Set the convention used by the inferior to return @code{struct}s and
18879 @code{union}s from functions to @var{mode}. Possible values of
18880 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
18881 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
18882 are returned on the stack, while @code{"reg"} means that a
18883 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
18884 be returned in a register.
18885
18886 @item show struct-convention
18887 @kindex show struct-convention
18888 Show the current setting of the convention to return @code{struct}s
18889 from functions.
18890 @end table
18891
18892 @node A29K
18893 @subsection A29K
18894
18895 @table @code
18896
18897 @kindex set rstack_high_address
18898 @cindex AMD 29K register stack
18899 @cindex register stack, AMD29K
18900 @item set rstack_high_address @var{address}
18901 On AMD 29000 family processors, registers are saved in a separate
18902 @dfn{register stack}. There is no way for @value{GDBN} to determine the
18903 extent of this stack. Normally, @value{GDBN} just assumes that the
18904 stack is ``large enough''. This may result in @value{GDBN} referencing
18905 memory locations that do not exist. If necessary, you can get around
18906 this problem by specifying the ending address of the register stack with
18907 the @code{set rstack_high_address} command. The argument should be an
18908 address, which you probably want to precede with @samp{0x} to specify in
18909 hexadecimal.
18910
18911 @kindex show rstack_high_address
18912 @item show rstack_high_address
18913 Display the current limit of the register stack, on AMD 29000 family
18914 processors.
18915
18916 @end table
18917
18918 @node Alpha
18919 @subsection Alpha
18920
18921 See the following section.
18922
18923 @node MIPS
18924 @subsection MIPS
18925
18926 @cindex stack on Alpha
18927 @cindex stack on MIPS
18928 @cindex Alpha stack
18929 @cindex MIPS stack
18930 Alpha- and MIPS-based computers use an unusual stack frame, which
18931 sometimes requires @value{GDBN} to search backward in the object code to
18932 find the beginning of a function.
18933
18934 @cindex response time, MIPS debugging
18935 To improve response time (especially for embedded applications, where
18936 @value{GDBN} may be restricted to a slow serial line for this search)
18937 you may want to limit the size of this search, using one of these
18938 commands:
18939
18940 @table @code
18941 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
18942 @item set heuristic-fence-post @var{limit}
18943 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
18944 search for the beginning of a function. A value of @var{0} (the
18945 default) means there is no limit. However, except for @var{0}, the
18946 larger the limit the more bytes @code{heuristic-fence-post} must search
18947 and therefore the longer it takes to run. You should only need to use
18948 this command when debugging a stripped executable.
18949
18950 @item show heuristic-fence-post
18951 Display the current limit.
18952 @end table
18953
18954 @noindent
18955 These commands are available @emph{only} when @value{GDBN} is configured
18956 for debugging programs on Alpha or MIPS processors.
18957
18958 Several MIPS-specific commands are available when debugging MIPS
18959 programs:
18960
18961 @table @code
18962 @item set mips abi @var{arg}
18963 @kindex set mips abi
18964 @cindex set ABI for MIPS
18965 Tell @value{GDBN} which MIPS ABI is used by the inferior. Possible
18966 values of @var{arg} are:
18967
18968 @table @samp
18969 @item auto
18970 The default ABI associated with the current binary (this is the
18971 default).
18972 @item o32
18973 @item o64
18974 @item n32
18975 @item n64
18976 @item eabi32
18977 @item eabi64
18978 @item auto
18979 @end table
18980
18981 @item show mips abi
18982 @kindex show mips abi
18983 Show the MIPS ABI used by @value{GDBN} to debug the inferior.
18984
18985 @item set mipsfpu
18986 @itemx show mipsfpu
18987 @xref{MIPS Embedded, set mipsfpu}.
18988
18989 @item set mips mask-address @var{arg}
18990 @kindex set mips mask-address
18991 @cindex MIPS addresses, masking
18992 This command determines whether the most-significant 32 bits of 64-bit
18993 MIPS addresses are masked off. The argument @var{arg} can be
18994 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
18995 setting, which lets @value{GDBN} determine the correct value.
18996
18997 @item show mips mask-address
18998 @kindex show mips mask-address
18999 Show whether the upper 32 bits of MIPS addresses are masked off or
19000 not.
19001
19002 @item set remote-mips64-transfers-32bit-regs
19003 @kindex set remote-mips64-transfers-32bit-regs
19004 This command controls compatibility with 64-bit MIPS targets that
19005 transfer data in 32-bit quantities. If you have an old MIPS 64 target
19006 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
19007 and 64 bits for other registers, set this option to @samp{on}.
19008
19009 @item show remote-mips64-transfers-32bit-regs
19010 @kindex show remote-mips64-transfers-32bit-regs
19011 Show the current setting of compatibility with older MIPS 64 targets.
19012
19013 @item set debug mips
19014 @kindex set debug mips
19015 This command turns on and off debugging messages for the MIPS-specific
19016 target code in @value{GDBN}.
19017
19018 @item show debug mips
19019 @kindex show debug mips
19020 Show the current setting of MIPS debugging messages.
19021 @end table
19022
19023
19024 @node HPPA
19025 @subsection HPPA
19026 @cindex HPPA support
19027
19028 When @value{GDBN} is debugging the HP PA architecture, it provides the
19029 following special commands:
19030
19031 @table @code
19032 @item set debug hppa
19033 @kindex set debug hppa
19034 This command determines whether HPPA architecture-specific debugging
19035 messages are to be displayed.
19036
19037 @item show debug hppa
19038 Show whether HPPA debugging messages are displayed.
19039
19040 @item maint print unwind @var{address}
19041 @kindex maint print unwind@r{, HPPA}
19042 This command displays the contents of the unwind table entry at the
19043 given @var{address}.
19044
19045 @end table
19046
19047
19048 @node SPU
19049 @subsection Cell Broadband Engine SPU architecture
19050 @cindex Cell Broadband Engine
19051 @cindex SPU
19052
19053 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
19054 it provides the following special commands:
19055
19056 @table @code
19057 @item info spu event
19058 @kindex info spu
19059 Display SPU event facility status. Shows current event mask
19060 and pending event status.
19061
19062 @item info spu signal
19063 Display SPU signal notification facility status. Shows pending
19064 signal-control word and signal notification mode of both signal
19065 notification channels.
19066
19067 @item info spu mailbox
19068 Display SPU mailbox facility status. Shows all pending entries,
19069 in order of processing, in each of the SPU Write Outbound,
19070 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
19071
19072 @item info spu dma
19073 Display MFC DMA status. Shows all pending commands in the MFC
19074 DMA queue. For each entry, opcode, tag, class IDs, effective
19075 and local store addresses and transfer size are shown.
19076
19077 @item info spu proxydma
19078 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
19079 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
19080 and local store addresses and transfer size are shown.
19081
19082 @end table
19083
19084 When @value{GDBN} is debugging a combined PowerPC/SPU application
19085 on the Cell Broadband Engine, it provides in addition the following
19086 special commands:
19087
19088 @table @code
19089 @item set spu stop-on-load @var{arg}
19090 @kindex set spu
19091 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
19092 will give control to the user when a new SPE thread enters its @code{main}
19093 function. The default is @code{off}.
19094
19095 @item show spu stop-on-load
19096 @kindex show spu
19097 Show whether to stop for new SPE threads.
19098
19099 @item set spu auto-flush-cache @var{arg}
19100 Set whether to automatically flush the software-managed cache. When set to
19101 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
19102 cache to be flushed whenever SPE execution stops. This provides a consistent
19103 view of PowerPC memory that is accessed via the cache. If an application
19104 does not use the software-managed cache, this option has no effect.
19105
19106 @item show spu auto-flush-cache
19107 Show whether to automatically flush the software-managed cache.
19108
19109 @end table
19110
19111 @node PowerPC
19112 @subsection PowerPC
19113 @cindex PowerPC architecture
19114
19115 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
19116 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
19117 numbers stored in the floating point registers. These values must be stored
19118 in two consecutive registers, always starting at an even register like
19119 @code{f0} or @code{f2}.
19120
19121 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
19122 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
19123 @code{f2} and @code{f3} for @code{$dl1} and so on.
19124
19125 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
19126 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
19127
19128
19129 @node Controlling GDB
19130 @chapter Controlling @value{GDBN}
19131
19132 You can alter the way @value{GDBN} interacts with you by using the
19133 @code{set} command. For commands controlling how @value{GDBN} displays
19134 data, see @ref{Print Settings, ,Print Settings}. Other settings are
19135 described here.
19136
19137 @menu
19138 * Prompt:: Prompt
19139 * Editing:: Command editing
19140 * Command History:: Command history
19141 * Screen Size:: Screen size
19142 * Numbers:: Numbers
19143 * ABI:: Configuring the current ABI
19144 * Messages/Warnings:: Optional warnings and messages
19145 * Debugging Output:: Optional messages about internal happenings
19146 * Other Misc Settings:: Other Miscellaneous Settings
19147 @end menu
19148
19149 @node Prompt
19150 @section Prompt
19151
19152 @cindex prompt
19153
19154 @value{GDBN} indicates its readiness to read a command by printing a string
19155 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
19156 can change the prompt string with the @code{set prompt} command. For
19157 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
19158 the prompt in one of the @value{GDBN} sessions so that you can always tell
19159 which one you are talking to.
19160
19161 @emph{Note:} @code{set prompt} does not add a space for you after the
19162 prompt you set. This allows you to set a prompt which ends in a space
19163 or a prompt that does not.
19164
19165 @table @code
19166 @kindex set prompt
19167 @item set prompt @var{newprompt}
19168 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
19169
19170 @kindex show prompt
19171 @item show prompt
19172 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
19173 @end table
19174
19175 @node Editing
19176 @section Command Editing
19177 @cindex readline
19178 @cindex command line editing
19179
19180 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
19181 @sc{gnu} library provides consistent behavior for programs which provide a
19182 command line interface to the user. Advantages are @sc{gnu} Emacs-style
19183 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
19184 substitution, and a storage and recall of command history across
19185 debugging sessions.
19186
19187 You may control the behavior of command line editing in @value{GDBN} with the
19188 command @code{set}.
19189
19190 @table @code
19191 @kindex set editing
19192 @cindex editing
19193 @item set editing
19194 @itemx set editing on
19195 Enable command line editing (enabled by default).
19196
19197 @item set editing off
19198 Disable command line editing.
19199
19200 @kindex show editing
19201 @item show editing
19202 Show whether command line editing is enabled.
19203 @end table
19204
19205 @xref{Command Line Editing}, for more details about the Readline
19206 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
19207 encouraged to read that chapter.
19208
19209 @node Command History
19210 @section Command History
19211 @cindex command history
19212
19213 @value{GDBN} can keep track of the commands you type during your
19214 debugging sessions, so that you can be certain of precisely what
19215 happened. Use these commands to manage the @value{GDBN} command
19216 history facility.
19217
19218 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
19219 package, to provide the history facility. @xref{Using History
19220 Interactively}, for the detailed description of the History library.
19221
19222 To issue a command to @value{GDBN} without affecting certain aspects of
19223 the state which is seen by users, prefix it with @samp{server }
19224 (@pxref{Server Prefix}). This
19225 means that this command will not affect the command history, nor will it
19226 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
19227 pressed on a line by itself.
19228
19229 @cindex @code{server}, command prefix
19230 The server prefix does not affect the recording of values into the value
19231 history; to print a value without recording it into the value history,
19232 use the @code{output} command instead of the @code{print} command.
19233
19234 Here is the description of @value{GDBN} commands related to command
19235 history.
19236
19237 @table @code
19238 @cindex history substitution
19239 @cindex history file
19240 @kindex set history filename
19241 @cindex @env{GDBHISTFILE}, environment variable
19242 @item set history filename @var{fname}
19243 Set the name of the @value{GDBN} command history file to @var{fname}.
19244 This is the file where @value{GDBN} reads an initial command history
19245 list, and where it writes the command history from this session when it
19246 exits. You can access this list through history expansion or through
19247 the history command editing characters listed below. This file defaults
19248 to the value of the environment variable @code{GDBHISTFILE}, or to
19249 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
19250 is not set.
19251
19252 @cindex save command history
19253 @kindex set history save
19254 @item set history save
19255 @itemx set history save on
19256 Record command history in a file, whose name may be specified with the
19257 @code{set history filename} command. By default, this option is disabled.
19258
19259 @item set history save off
19260 Stop recording command history in a file.
19261
19262 @cindex history size
19263 @kindex set history size
19264 @cindex @env{HISTSIZE}, environment variable
19265 @item set history size @var{size}
19266 Set the number of commands which @value{GDBN} keeps in its history list.
19267 This defaults to the value of the environment variable
19268 @code{HISTSIZE}, or to 256 if this variable is not set.
19269 @end table
19270
19271 History expansion assigns special meaning to the character @kbd{!}.
19272 @xref{Event Designators}, for more details.
19273
19274 @cindex history expansion, turn on/off
19275 Since @kbd{!} is also the logical not operator in C, history expansion
19276 is off by default. If you decide to enable history expansion with the
19277 @code{set history expansion on} command, you may sometimes need to
19278 follow @kbd{!} (when it is used as logical not, in an expression) with
19279 a space or a tab to prevent it from being expanded. The readline
19280 history facilities do not attempt substitution on the strings
19281 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
19282
19283 The commands to control history expansion are:
19284
19285 @table @code
19286 @item set history expansion on
19287 @itemx set history expansion
19288 @kindex set history expansion
19289 Enable history expansion. History expansion is off by default.
19290
19291 @item set history expansion off
19292 Disable history expansion.
19293
19294 @c @group
19295 @kindex show history
19296 @item show history
19297 @itemx show history filename
19298 @itemx show history save
19299 @itemx show history size
19300 @itemx show history expansion
19301 These commands display the state of the @value{GDBN} history parameters.
19302 @code{show history} by itself displays all four states.
19303 @c @end group
19304 @end table
19305
19306 @table @code
19307 @kindex show commands
19308 @cindex show last commands
19309 @cindex display command history
19310 @item show commands
19311 Display the last ten commands in the command history.
19312
19313 @item show commands @var{n}
19314 Print ten commands centered on command number @var{n}.
19315
19316 @item show commands +
19317 Print ten commands just after the commands last printed.
19318 @end table
19319
19320 @node Screen Size
19321 @section Screen Size
19322 @cindex size of screen
19323 @cindex pauses in output
19324
19325 Certain commands to @value{GDBN} may produce large amounts of
19326 information output to the screen. To help you read all of it,
19327 @value{GDBN} pauses and asks you for input at the end of each page of
19328 output. Type @key{RET} when you want to continue the output, or @kbd{q}
19329 to discard the remaining output. Also, the screen width setting
19330 determines when to wrap lines of output. Depending on what is being
19331 printed, @value{GDBN} tries to break the line at a readable place,
19332 rather than simply letting it overflow onto the following line.
19333
19334 Normally @value{GDBN} knows the size of the screen from the terminal
19335 driver software. For example, on Unix @value{GDBN} uses the termcap data base
19336 together with the value of the @code{TERM} environment variable and the
19337 @code{stty rows} and @code{stty cols} settings. If this is not correct,
19338 you can override it with the @code{set height} and @code{set
19339 width} commands:
19340
19341 @table @code
19342 @kindex set height
19343 @kindex set width
19344 @kindex show width
19345 @kindex show height
19346 @item set height @var{lpp}
19347 @itemx show height
19348 @itemx set width @var{cpl}
19349 @itemx show width
19350 These @code{set} commands specify a screen height of @var{lpp} lines and
19351 a screen width of @var{cpl} characters. The associated @code{show}
19352 commands display the current settings.
19353
19354 If you specify a height of zero lines, @value{GDBN} does not pause during
19355 output no matter how long the output is. This is useful if output is to a
19356 file or to an editor buffer.
19357
19358 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
19359 from wrapping its output.
19360
19361 @item set pagination on
19362 @itemx set pagination off
19363 @kindex set pagination
19364 Turn the output pagination on or off; the default is on. Turning
19365 pagination off is the alternative to @code{set height 0}. Note that
19366 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
19367 Options, -batch}) also automatically disables pagination.
19368
19369 @item show pagination
19370 @kindex show pagination
19371 Show the current pagination mode.
19372 @end table
19373
19374 @node Numbers
19375 @section Numbers
19376 @cindex number representation
19377 @cindex entering numbers
19378
19379 You can always enter numbers in octal, decimal, or hexadecimal in
19380 @value{GDBN} by the usual conventions: octal numbers begin with
19381 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
19382 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
19383 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
19384 10; likewise, the default display for numbers---when no particular
19385 format is specified---is base 10. You can change the default base for
19386 both input and output with the commands described below.
19387
19388 @table @code
19389 @kindex set input-radix
19390 @item set input-radix @var{base}
19391 Set the default base for numeric input. Supported choices
19392 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
19393 specified either unambiguously or using the current input radix; for
19394 example, any of
19395
19396 @smallexample
19397 set input-radix 012
19398 set input-radix 10.
19399 set input-radix 0xa
19400 @end smallexample
19401
19402 @noindent
19403 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
19404 leaves the input radix unchanged, no matter what it was, since
19405 @samp{10}, being without any leading or trailing signs of its base, is
19406 interpreted in the current radix. Thus, if the current radix is 16,
19407 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
19408 change the radix.
19409
19410 @kindex set output-radix
19411 @item set output-radix @var{base}
19412 Set the default base for numeric display. Supported choices
19413 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
19414 specified either unambiguously or using the current input radix.
19415
19416 @kindex show input-radix
19417 @item show input-radix
19418 Display the current default base for numeric input.
19419
19420 @kindex show output-radix
19421 @item show output-radix
19422 Display the current default base for numeric display.
19423
19424 @item set radix @r{[}@var{base}@r{]}
19425 @itemx show radix
19426 @kindex set radix
19427 @kindex show radix
19428 These commands set and show the default base for both input and output
19429 of numbers. @code{set radix} sets the radix of input and output to
19430 the same base; without an argument, it resets the radix back to its
19431 default value of 10.
19432
19433 @end table
19434
19435 @node ABI
19436 @section Configuring the Current ABI
19437
19438 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
19439 application automatically. However, sometimes you need to override its
19440 conclusions. Use these commands to manage @value{GDBN}'s view of the
19441 current ABI.
19442
19443 @cindex OS ABI
19444 @kindex set osabi
19445 @kindex show osabi
19446
19447 One @value{GDBN} configuration can debug binaries for multiple operating
19448 system targets, either via remote debugging or native emulation.
19449 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
19450 but you can override its conclusion using the @code{set osabi} command.
19451 One example where this is useful is in debugging of binaries which use
19452 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
19453 not have the same identifying marks that the standard C library for your
19454 platform provides.
19455
19456 @table @code
19457 @item show osabi
19458 Show the OS ABI currently in use.
19459
19460 @item set osabi
19461 With no argument, show the list of registered available OS ABI's.
19462
19463 @item set osabi @var{abi}
19464 Set the current OS ABI to @var{abi}.
19465 @end table
19466
19467 @cindex float promotion
19468
19469 Generally, the way that an argument of type @code{float} is passed to a
19470 function depends on whether the function is prototyped. For a prototyped
19471 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
19472 according to the architecture's convention for @code{float}. For unprototyped
19473 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
19474 @code{double} and then passed.
19475
19476 Unfortunately, some forms of debug information do not reliably indicate whether
19477 a function is prototyped. If @value{GDBN} calls a function that is not marked
19478 as prototyped, it consults @kbd{set coerce-float-to-double}.
19479
19480 @table @code
19481 @kindex set coerce-float-to-double
19482 @item set coerce-float-to-double
19483 @itemx set coerce-float-to-double on
19484 Arguments of type @code{float} will be promoted to @code{double} when passed
19485 to an unprototyped function. This is the default setting.
19486
19487 @item set coerce-float-to-double off
19488 Arguments of type @code{float} will be passed directly to unprototyped
19489 functions.
19490
19491 @kindex show coerce-float-to-double
19492 @item show coerce-float-to-double
19493 Show the current setting of promoting @code{float} to @code{double}.
19494 @end table
19495
19496 @kindex set cp-abi
19497 @kindex show cp-abi
19498 @value{GDBN} needs to know the ABI used for your program's C@t{++}
19499 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
19500 used to build your application. @value{GDBN} only fully supports
19501 programs with a single C@t{++} ABI; if your program contains code using
19502 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
19503 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
19504 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
19505 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
19506 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
19507 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
19508 ``auto''.
19509
19510 @table @code
19511 @item show cp-abi
19512 Show the C@t{++} ABI currently in use.
19513
19514 @item set cp-abi
19515 With no argument, show the list of supported C@t{++} ABI's.
19516
19517 @item set cp-abi @var{abi}
19518 @itemx set cp-abi auto
19519 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
19520 @end table
19521
19522 @node Messages/Warnings
19523 @section Optional Warnings and Messages
19524
19525 @cindex verbose operation
19526 @cindex optional warnings
19527 By default, @value{GDBN} is silent about its inner workings. If you are
19528 running on a slow machine, you may want to use the @code{set verbose}
19529 command. This makes @value{GDBN} tell you when it does a lengthy
19530 internal operation, so you will not think it has crashed.
19531
19532 Currently, the messages controlled by @code{set verbose} are those
19533 which announce that the symbol table for a source file is being read;
19534 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
19535
19536 @table @code
19537 @kindex set verbose
19538 @item set verbose on
19539 Enables @value{GDBN} output of certain informational messages.
19540
19541 @item set verbose off
19542 Disables @value{GDBN} output of certain informational messages.
19543
19544 @kindex show verbose
19545 @item show verbose
19546 Displays whether @code{set verbose} is on or off.
19547 @end table
19548
19549 By default, if @value{GDBN} encounters bugs in the symbol table of an
19550 object file, it is silent; but if you are debugging a compiler, you may
19551 find this information useful (@pxref{Symbol Errors, ,Errors Reading
19552 Symbol Files}).
19553
19554 @table @code
19555
19556 @kindex set complaints
19557 @item set complaints @var{limit}
19558 Permits @value{GDBN} to output @var{limit} complaints about each type of
19559 unusual symbols before becoming silent about the problem. Set
19560 @var{limit} to zero to suppress all complaints; set it to a large number
19561 to prevent complaints from being suppressed.
19562
19563 @kindex show complaints
19564 @item show complaints
19565 Displays how many symbol complaints @value{GDBN} is permitted to produce.
19566
19567 @end table
19568
19569 @anchor{confirmation requests}
19570 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
19571 lot of stupid questions to confirm certain commands. For example, if
19572 you try to run a program which is already running:
19573
19574 @smallexample
19575 (@value{GDBP}) run
19576 The program being debugged has been started already.
19577 Start it from the beginning? (y or n)
19578 @end smallexample
19579
19580 If you are willing to unflinchingly face the consequences of your own
19581 commands, you can disable this ``feature'':
19582
19583 @table @code
19584
19585 @kindex set confirm
19586 @cindex flinching
19587 @cindex confirmation
19588 @cindex stupid questions
19589 @item set confirm off
19590 Disables confirmation requests. Note that running @value{GDBN} with
19591 the @option{--batch} option (@pxref{Mode Options, -batch}) also
19592 automatically disables confirmation requests.
19593
19594 @item set confirm on
19595 Enables confirmation requests (the default).
19596
19597 @kindex show confirm
19598 @item show confirm
19599 Displays state of confirmation requests.
19600
19601 @end table
19602
19603 @cindex command tracing
19604 If you need to debug user-defined commands or sourced files you may find it
19605 useful to enable @dfn{command tracing}. In this mode each command will be
19606 printed as it is executed, prefixed with one or more @samp{+} symbols, the
19607 quantity denoting the call depth of each command.
19608
19609 @table @code
19610 @kindex set trace-commands
19611 @cindex command scripts, debugging
19612 @item set trace-commands on
19613 Enable command tracing.
19614 @item set trace-commands off
19615 Disable command tracing.
19616 @item show trace-commands
19617 Display the current state of command tracing.
19618 @end table
19619
19620 @node Debugging Output
19621 @section Optional Messages about Internal Happenings
19622 @cindex optional debugging messages
19623
19624 @value{GDBN} has commands that enable optional debugging messages from
19625 various @value{GDBN} subsystems; normally these commands are of
19626 interest to @value{GDBN} maintainers, or when reporting a bug. This
19627 section documents those commands.
19628
19629 @table @code
19630 @kindex set exec-done-display
19631 @item set exec-done-display
19632 Turns on or off the notification of asynchronous commands'
19633 completion. When on, @value{GDBN} will print a message when an
19634 asynchronous command finishes its execution. The default is off.
19635 @kindex show exec-done-display
19636 @item show exec-done-display
19637 Displays the current setting of asynchronous command completion
19638 notification.
19639 @kindex set debug
19640 @cindex gdbarch debugging info
19641 @cindex architecture debugging info
19642 @item set debug arch
19643 Turns on or off display of gdbarch debugging info. The default is off
19644 @kindex show debug
19645 @item show debug arch
19646 Displays the current state of displaying gdbarch debugging info.
19647 @item set debug aix-thread
19648 @cindex AIX threads
19649 Display debugging messages about inner workings of the AIX thread
19650 module.
19651 @item show debug aix-thread
19652 Show the current state of AIX thread debugging info display.
19653 @item set debug dwarf2-die
19654 @cindex DWARF2 DIEs
19655 Dump DWARF2 DIEs after they are read in.
19656 The value is the number of nesting levels to print.
19657 A value of zero turns off the display.
19658 @item show debug dwarf2-die
19659 Show the current state of DWARF2 DIE debugging.
19660 @item set debug displaced
19661 @cindex displaced stepping debugging info
19662 Turns on or off display of @value{GDBN} debugging info for the
19663 displaced stepping support. The default is off.
19664 @item show debug displaced
19665 Displays the current state of displaying @value{GDBN} debugging info
19666 related to displaced stepping.
19667 @item set debug event
19668 @cindex event debugging info
19669 Turns on or off display of @value{GDBN} event debugging info. The
19670 default is off.
19671 @item show debug event
19672 Displays the current state of displaying @value{GDBN} event debugging
19673 info.
19674 @item set debug expression
19675 @cindex expression debugging info
19676 Turns on or off display of debugging info about @value{GDBN}
19677 expression parsing. The default is off.
19678 @item show debug expression
19679 Displays the current state of displaying debugging info about
19680 @value{GDBN} expression parsing.
19681 @item set debug frame
19682 @cindex frame debugging info
19683 Turns on or off display of @value{GDBN} frame debugging info. The
19684 default is off.
19685 @item show debug frame
19686 Displays the current state of displaying @value{GDBN} frame debugging
19687 info.
19688 @item set debug gnu-nat
19689 @cindex @sc{gnu}/Hurd debug messages
19690 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
19691 @item show debug gnu-nat
19692 Show the current state of @sc{gnu}/Hurd debugging messages.
19693 @item set debug infrun
19694 @cindex inferior debugging info
19695 Turns on or off display of @value{GDBN} debugging info for running the inferior.
19696 The default is off. @file{infrun.c} contains GDB's runtime state machine used
19697 for implementing operations such as single-stepping the inferior.
19698 @item show debug infrun
19699 Displays the current state of @value{GDBN} inferior debugging.
19700 @item set debug lin-lwp
19701 @cindex @sc{gnu}/Linux LWP debug messages
19702 @cindex Linux lightweight processes
19703 Turns on or off debugging messages from the Linux LWP debug support.
19704 @item show debug lin-lwp
19705 Show the current state of Linux LWP debugging messages.
19706 @item set debug lin-lwp-async
19707 @cindex @sc{gnu}/Linux LWP async debug messages
19708 @cindex Linux lightweight processes
19709 Turns on or off debugging messages from the Linux LWP async debug support.
19710 @item show debug lin-lwp-async
19711 Show the current state of Linux LWP async debugging messages.
19712 @item set debug observer
19713 @cindex observer debugging info
19714 Turns on or off display of @value{GDBN} observer debugging. This
19715 includes info such as the notification of observable events.
19716 @item show debug observer
19717 Displays the current state of observer debugging.
19718 @item set debug overload
19719 @cindex C@t{++} overload debugging info
19720 Turns on or off display of @value{GDBN} C@t{++} overload debugging
19721 info. This includes info such as ranking of functions, etc. The default
19722 is off.
19723 @item show debug overload
19724 Displays the current state of displaying @value{GDBN} C@t{++} overload
19725 debugging info.
19726 @cindex expression parser, debugging info
19727 @cindex debug expression parser
19728 @item set debug parser
19729 Turns on or off the display of expression parser debugging output.
19730 Internally, this sets the @code{yydebug} variable in the expression
19731 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
19732 details. The default is off.
19733 @item show debug parser
19734 Show the current state of expression parser debugging.
19735 @cindex packets, reporting on stdout
19736 @cindex serial connections, debugging
19737 @cindex debug remote protocol
19738 @cindex remote protocol debugging
19739 @cindex display remote packets
19740 @item set debug remote
19741 Turns on or off display of reports on all packets sent back and forth across
19742 the serial line to the remote machine. The info is printed on the
19743 @value{GDBN} standard output stream. The default is off.
19744 @item show debug remote
19745 Displays the state of display of remote packets.
19746 @item set debug serial
19747 Turns on or off display of @value{GDBN} serial debugging info. The
19748 default is off.
19749 @item show debug serial
19750 Displays the current state of displaying @value{GDBN} serial debugging
19751 info.
19752 @item set debug solib-frv
19753 @cindex FR-V shared-library debugging
19754 Turns on or off debugging messages for FR-V shared-library code.
19755 @item show debug solib-frv
19756 Display the current state of FR-V shared-library code debugging
19757 messages.
19758 @item set debug target
19759 @cindex target debugging info
19760 Turns on or off display of @value{GDBN} target debugging info. This info
19761 includes what is going on at the target level of GDB, as it happens. The
19762 default is 0. Set it to 1 to track events, and to 2 to also track the
19763 value of large memory transfers. Changes to this flag do not take effect
19764 until the next time you connect to a target or use the @code{run} command.
19765 @item show debug target
19766 Displays the current state of displaying @value{GDBN} target debugging
19767 info.
19768 @item set debug timestamp
19769 @cindex timestampping debugging info
19770 Turns on or off display of timestamps with @value{GDBN} debugging info.
19771 When enabled, seconds and microseconds are displayed before each debugging
19772 message.
19773 @item show debug timestamp
19774 Displays the current state of displaying timestamps with @value{GDBN}
19775 debugging info.
19776 @item set debugvarobj
19777 @cindex variable object debugging info
19778 Turns on or off display of @value{GDBN} variable object debugging
19779 info. The default is off.
19780 @item show debugvarobj
19781 Displays the current state of displaying @value{GDBN} variable object
19782 debugging info.
19783 @item set debug xml
19784 @cindex XML parser debugging
19785 Turns on or off debugging messages for built-in XML parsers.
19786 @item show debug xml
19787 Displays the current state of XML debugging messages.
19788 @end table
19789
19790 @node Other Misc Settings
19791 @section Other Miscellaneous Settings
19792 @cindex miscellaneous settings
19793
19794 @table @code
19795 @kindex set interactive-mode
19796 @item set interactive-mode
19797 If @code{on}, forces @value{GDBN} to operate interactively.
19798 If @code{off}, forces @value{GDBN} to operate non-interactively,
19799 If @code{auto} (the default), @value{GDBN} guesses which mode to use,
19800 based on whether the debugger was started in a terminal or not.
19801
19802 In the vast majority of cases, the debugger should be able to guess
19803 correctly which mode should be used. But this setting can be useful
19804 in certain specific cases, such as running a MinGW @value{GDBN}
19805 inside a cygwin window.
19806
19807 @kindex show interactive-mode
19808 @item show interactive-mode
19809 Displays whether the debugger is operating in interactive mode or not.
19810 @end table
19811
19812 @node Extending GDB
19813 @chapter Extending @value{GDBN}
19814 @cindex extending GDB
19815
19816 @value{GDBN} provides two mechanisms for extension. The first is based
19817 on composition of @value{GDBN} commands, and the second is based on the
19818 Python scripting language.
19819
19820 To facilitate the use of these extensions, @value{GDBN} is capable
19821 of evaluating the contents of a file. When doing so, @value{GDBN}
19822 can recognize which scripting language is being used by looking at
19823 the filename extension. Files with an unrecognized filename extension
19824 are always treated as a @value{GDBN} Command Files.
19825 @xref{Command Files,, Command files}.
19826
19827 You can control how @value{GDBN} evaluates these files with the following
19828 setting:
19829
19830 @table @code
19831 @kindex set script-extension
19832 @kindex show script-extension
19833 @item set script-extension off
19834 All scripts are always evaluated as @value{GDBN} Command Files.
19835
19836 @item set script-extension soft
19837 The debugger determines the scripting language based on filename
19838 extension. If this scripting language is supported, @value{GDBN}
19839 evaluates the script using that language. Otherwise, it evaluates
19840 the file as a @value{GDBN} Command File.
19841
19842 @item set script-extension strict
19843 The debugger determines the scripting language based on filename
19844 extension, and evaluates the script using that language. If the
19845 language is not supported, then the evaluation fails.
19846
19847 @item show script-extension
19848 Display the current value of the @code{script-extension} option.
19849
19850 @end table
19851
19852 @menu
19853 * Sequences:: Canned Sequences of Commands
19854 * Python:: Scripting @value{GDBN} using Python
19855 @end menu
19856
19857 @node Sequences
19858 @section Canned Sequences of Commands
19859
19860 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
19861 Command Lists}), @value{GDBN} provides two ways to store sequences of
19862 commands for execution as a unit: user-defined commands and command
19863 files.
19864
19865 @menu
19866 * Define:: How to define your own commands
19867 * Hooks:: Hooks for user-defined commands
19868 * Command Files:: How to write scripts of commands to be stored in a file
19869 * Output:: Commands for controlled output
19870 @end menu
19871
19872 @node Define
19873 @subsection User-defined Commands
19874
19875 @cindex user-defined command
19876 @cindex arguments, to user-defined commands
19877 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
19878 which you assign a new name as a command. This is done with the
19879 @code{define} command. User commands may accept up to 10 arguments
19880 separated by whitespace. Arguments are accessed within the user command
19881 via @code{$arg0@dots{}$arg9}. A trivial example:
19882
19883 @smallexample
19884 define adder
19885 print $arg0 + $arg1 + $arg2
19886 end
19887 @end smallexample
19888
19889 @noindent
19890 To execute the command use:
19891
19892 @smallexample
19893 adder 1 2 3
19894 @end smallexample
19895
19896 @noindent
19897 This defines the command @code{adder}, which prints the sum of
19898 its three arguments. Note the arguments are text substitutions, so they may
19899 reference variables, use complex expressions, or even perform inferior
19900 functions calls.
19901
19902 @cindex argument count in user-defined commands
19903 @cindex how many arguments (user-defined commands)
19904 In addition, @code{$argc} may be used to find out how many arguments have
19905 been passed. This expands to a number in the range 0@dots{}10.
19906
19907 @smallexample
19908 define adder
19909 if $argc == 2
19910 print $arg0 + $arg1
19911 end
19912 if $argc == 3
19913 print $arg0 + $arg1 + $arg2
19914 end
19915 end
19916 @end smallexample
19917
19918 @table @code
19919
19920 @kindex define
19921 @item define @var{commandname}
19922 Define a command named @var{commandname}. If there is already a command
19923 by that name, you are asked to confirm that you want to redefine it.
19924 @var{commandname} may be a bare command name consisting of letters,
19925 numbers, dashes, and underscores. It may also start with any predefined
19926 prefix command. For example, @samp{define target my-target} creates
19927 a user-defined @samp{target my-target} command.
19928
19929 The definition of the command is made up of other @value{GDBN} command lines,
19930 which are given following the @code{define} command. The end of these
19931 commands is marked by a line containing @code{end}.
19932
19933 @kindex document
19934 @kindex end@r{ (user-defined commands)}
19935 @item document @var{commandname}
19936 Document the user-defined command @var{commandname}, so that it can be
19937 accessed by @code{help}. The command @var{commandname} must already be
19938 defined. This command reads lines of documentation just as @code{define}
19939 reads the lines of the command definition, ending with @code{end}.
19940 After the @code{document} command is finished, @code{help} on command
19941 @var{commandname} displays the documentation you have written.
19942
19943 You may use the @code{document} command again to change the
19944 documentation of a command. Redefining the command with @code{define}
19945 does not change the documentation.
19946
19947 @kindex dont-repeat
19948 @cindex don't repeat command
19949 @item dont-repeat
19950 Used inside a user-defined command, this tells @value{GDBN} that this
19951 command should not be repeated when the user hits @key{RET}
19952 (@pxref{Command Syntax, repeat last command}).
19953
19954 @kindex help user-defined
19955 @item help user-defined
19956 List all user-defined commands, with the first line of the documentation
19957 (if any) for each.
19958
19959 @kindex show user
19960 @item show user
19961 @itemx show user @var{commandname}
19962 Display the @value{GDBN} commands used to define @var{commandname} (but
19963 not its documentation). If no @var{commandname} is given, display the
19964 definitions for all user-defined commands.
19965
19966 @cindex infinite recursion in user-defined commands
19967 @kindex show max-user-call-depth
19968 @kindex set max-user-call-depth
19969 @item show max-user-call-depth
19970 @itemx set max-user-call-depth
19971 The value of @code{max-user-call-depth} controls how many recursion
19972 levels are allowed in user-defined commands before @value{GDBN} suspects an
19973 infinite recursion and aborts the command.
19974 @end table
19975
19976 In addition to the above commands, user-defined commands frequently
19977 use control flow commands, described in @ref{Command Files}.
19978
19979 When user-defined commands are executed, the
19980 commands of the definition are not printed. An error in any command
19981 stops execution of the user-defined command.
19982
19983 If used interactively, commands that would ask for confirmation proceed
19984 without asking when used inside a user-defined command. Many @value{GDBN}
19985 commands that normally print messages to say what they are doing omit the
19986 messages when used in a user-defined command.
19987
19988 @node Hooks
19989 @subsection User-defined Command Hooks
19990 @cindex command hooks
19991 @cindex hooks, for commands
19992 @cindex hooks, pre-command
19993
19994 @kindex hook
19995 You may define @dfn{hooks}, which are a special kind of user-defined
19996 command. Whenever you run the command @samp{foo}, if the user-defined
19997 command @samp{hook-foo} exists, it is executed (with no arguments)
19998 before that command.
19999
20000 @cindex hooks, post-command
20001 @kindex hookpost
20002 A hook may also be defined which is run after the command you executed.
20003 Whenever you run the command @samp{foo}, if the user-defined command
20004 @samp{hookpost-foo} exists, it is executed (with no arguments) after
20005 that command. Post-execution hooks may exist simultaneously with
20006 pre-execution hooks, for the same command.
20007
20008 It is valid for a hook to call the command which it hooks. If this
20009 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
20010
20011 @c It would be nice if hookpost could be passed a parameter indicating
20012 @c if the command it hooks executed properly or not. FIXME!
20013
20014 @kindex stop@r{, a pseudo-command}
20015 In addition, a pseudo-command, @samp{stop} exists. Defining
20016 (@samp{hook-stop}) makes the associated commands execute every time
20017 execution stops in your program: before breakpoint commands are run,
20018 displays are printed, or the stack frame is printed.
20019
20020 For example, to ignore @code{SIGALRM} signals while
20021 single-stepping, but treat them normally during normal execution,
20022 you could define:
20023
20024 @smallexample
20025 define hook-stop
20026 handle SIGALRM nopass
20027 end
20028
20029 define hook-run
20030 handle SIGALRM pass
20031 end
20032
20033 define hook-continue
20034 handle SIGALRM pass
20035 end
20036 @end smallexample
20037
20038 As a further example, to hook at the beginning and end of the @code{echo}
20039 command, and to add extra text to the beginning and end of the message,
20040 you could define:
20041
20042 @smallexample
20043 define hook-echo
20044 echo <<<---
20045 end
20046
20047 define hookpost-echo
20048 echo --->>>\n
20049 end
20050
20051 (@value{GDBP}) echo Hello World
20052 <<<---Hello World--->>>
20053 (@value{GDBP})
20054
20055 @end smallexample
20056
20057 You can define a hook for any single-word command in @value{GDBN}, but
20058 not for command aliases; you should define a hook for the basic command
20059 name, e.g.@: @code{backtrace} rather than @code{bt}.
20060 @c FIXME! So how does Joe User discover whether a command is an alias
20061 @c or not?
20062 You can hook a multi-word command by adding @code{hook-} or
20063 @code{hookpost-} to the last word of the command, e.g.@:
20064 @samp{define target hook-remote} to add a hook to @samp{target remote}.
20065
20066 If an error occurs during the execution of your hook, execution of
20067 @value{GDBN} commands stops and @value{GDBN} issues a prompt
20068 (before the command that you actually typed had a chance to run).
20069
20070 If you try to define a hook which does not match any known command, you
20071 get a warning from the @code{define} command.
20072
20073 @node Command Files
20074 @subsection Command Files
20075
20076 @cindex command files
20077 @cindex scripting commands
20078 A command file for @value{GDBN} is a text file made of lines that are
20079 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
20080 also be included. An empty line in a command file does nothing; it
20081 does not mean to repeat the last command, as it would from the
20082 terminal.
20083
20084 You can request the execution of a command file with the @code{source}
20085 command. Note that the @code{source} command is also used to evaluate
20086 scripts that are not Command Files. The exact behavior can be configured
20087 using the @code{script-extension} setting.
20088 @xref{Extending GDB,, Extending GDB}.
20089
20090 @table @code
20091 @kindex source
20092 @cindex execute commands from a file
20093 @item source [-s] [-v] @var{filename}
20094 Execute the command file @var{filename}.
20095 @end table
20096
20097 The lines in a command file are generally executed sequentially,
20098 unless the order of execution is changed by one of the
20099 @emph{flow-control commands} described below. The commands are not
20100 printed as they are executed. An error in any command terminates
20101 execution of the command file and control is returned to the console.
20102
20103 @value{GDBN} first searches for @var{filename} in the current directory.
20104 If the file is not found there, and @var{filename} does not specify a
20105 directory, then @value{GDBN} also looks for the file on the source search path
20106 (specified with the @samp{directory} command);
20107 except that @file{$cdir} is not searched because the compilation directory
20108 is not relevant to scripts.
20109
20110 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
20111 on the search path even if @var{filename} specifies a directory.
20112 The search is done by appending @var{filename} to each element of the
20113 search path. So, for example, if @var{filename} is @file{mylib/myscript}
20114 and the search path contains @file{/home/user} then @value{GDBN} will
20115 look for the script @file{/home/user/mylib/myscript}.
20116 The search is also done if @var{filename} is an absolute path.
20117 For example, if @var{filename} is @file{/tmp/myscript} and
20118 the search path contains @file{/home/user} then @value{GDBN} will
20119 look for the script @file{/home/user/tmp/myscript}.
20120 For DOS-like systems, if @var{filename} contains a drive specification,
20121 it is stripped before concatenation. For example, if @var{filename} is
20122 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
20123 will look for the script @file{c:/tmp/myscript}.
20124
20125 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
20126 each command as it is executed. The option must be given before
20127 @var{filename}, and is interpreted as part of the filename anywhere else.
20128
20129 Commands that would ask for confirmation if used interactively proceed
20130 without asking when used in a command file. Many @value{GDBN} commands that
20131 normally print messages to say what they are doing omit the messages
20132 when called from command files.
20133
20134 @value{GDBN} also accepts command input from standard input. In this
20135 mode, normal output goes to standard output and error output goes to
20136 standard error. Errors in a command file supplied on standard input do
20137 not terminate execution of the command file---execution continues with
20138 the next command.
20139
20140 @smallexample
20141 gdb < cmds > log 2>&1
20142 @end smallexample
20143
20144 (The syntax above will vary depending on the shell used.) This example
20145 will execute commands from the file @file{cmds}. All output and errors
20146 would be directed to @file{log}.
20147
20148 Since commands stored on command files tend to be more general than
20149 commands typed interactively, they frequently need to deal with
20150 complicated situations, such as different or unexpected values of
20151 variables and symbols, changes in how the program being debugged is
20152 built, etc. @value{GDBN} provides a set of flow-control commands to
20153 deal with these complexities. Using these commands, you can write
20154 complex scripts that loop over data structures, execute commands
20155 conditionally, etc.
20156
20157 @table @code
20158 @kindex if
20159 @kindex else
20160 @item if
20161 @itemx else
20162 This command allows to include in your script conditionally executed
20163 commands. The @code{if} command takes a single argument, which is an
20164 expression to evaluate. It is followed by a series of commands that
20165 are executed only if the expression is true (its value is nonzero).
20166 There can then optionally be an @code{else} line, followed by a series
20167 of commands that are only executed if the expression was false. The
20168 end of the list is marked by a line containing @code{end}.
20169
20170 @kindex while
20171 @item while
20172 This command allows to write loops. Its syntax is similar to
20173 @code{if}: the command takes a single argument, which is an expression
20174 to evaluate, and must be followed by the commands to execute, one per
20175 line, terminated by an @code{end}. These commands are called the
20176 @dfn{body} of the loop. The commands in the body of @code{while} are
20177 executed repeatedly as long as the expression evaluates to true.
20178
20179 @kindex loop_break
20180 @item loop_break
20181 This command exits the @code{while} loop in whose body it is included.
20182 Execution of the script continues after that @code{while}s @code{end}
20183 line.
20184
20185 @kindex loop_continue
20186 @item loop_continue
20187 This command skips the execution of the rest of the body of commands
20188 in the @code{while} loop in whose body it is included. Execution
20189 branches to the beginning of the @code{while} loop, where it evaluates
20190 the controlling expression.
20191
20192 @kindex end@r{ (if/else/while commands)}
20193 @item end
20194 Terminate the block of commands that are the body of @code{if},
20195 @code{else}, or @code{while} flow-control commands.
20196 @end table
20197
20198
20199 @node Output
20200 @subsection Commands for Controlled Output
20201
20202 During the execution of a command file or a user-defined command, normal
20203 @value{GDBN} output is suppressed; the only output that appears is what is
20204 explicitly printed by the commands in the definition. This section
20205 describes three commands useful for generating exactly the output you
20206 want.
20207
20208 @table @code
20209 @kindex echo
20210 @item echo @var{text}
20211 @c I do not consider backslash-space a standard C escape sequence
20212 @c because it is not in ANSI.
20213 Print @var{text}. Nonprinting characters can be included in
20214 @var{text} using C escape sequences, such as @samp{\n} to print a
20215 newline. @strong{No newline is printed unless you specify one.}
20216 In addition to the standard C escape sequences, a backslash followed
20217 by a space stands for a space. This is useful for displaying a
20218 string with spaces at the beginning or the end, since leading and
20219 trailing spaces are otherwise trimmed from all arguments.
20220 To print @samp{@w{ }and foo =@w{ }}, use the command
20221 @samp{echo \@w{ }and foo = \@w{ }}.
20222
20223 A backslash at the end of @var{text} can be used, as in C, to continue
20224 the command onto subsequent lines. For example,
20225
20226 @smallexample
20227 echo This is some text\n\
20228 which is continued\n\
20229 onto several lines.\n
20230 @end smallexample
20231
20232 produces the same output as
20233
20234 @smallexample
20235 echo This is some text\n
20236 echo which is continued\n
20237 echo onto several lines.\n
20238 @end smallexample
20239
20240 @kindex output
20241 @item output @var{expression}
20242 Print the value of @var{expression} and nothing but that value: no
20243 newlines, no @samp{$@var{nn} = }. The value is not entered in the
20244 value history either. @xref{Expressions, ,Expressions}, for more information
20245 on expressions.
20246
20247 @item output/@var{fmt} @var{expression}
20248 Print the value of @var{expression} in format @var{fmt}. You can use
20249 the same formats as for @code{print}. @xref{Output Formats,,Output
20250 Formats}, for more information.
20251
20252 @kindex printf
20253 @item printf @var{template}, @var{expressions}@dots{}
20254 Print the values of one or more @var{expressions} under the control of
20255 the string @var{template}. To print several values, make
20256 @var{expressions} be a comma-separated list of individual expressions,
20257 which may be either numbers or pointers. Their values are printed as
20258 specified by @var{template}, exactly as a C program would do by
20259 executing the code below:
20260
20261 @smallexample
20262 printf (@var{template}, @var{expressions}@dots{});
20263 @end smallexample
20264
20265 As in @code{C} @code{printf}, ordinary characters in @var{template}
20266 are printed verbatim, while @dfn{conversion specification} introduced
20267 by the @samp{%} character cause subsequent @var{expressions} to be
20268 evaluated, their values converted and formatted according to type and
20269 style information encoded in the conversion specifications, and then
20270 printed.
20271
20272 For example, you can print two values in hex like this:
20273
20274 @smallexample
20275 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
20276 @end smallexample
20277
20278 @code{printf} supports all the standard @code{C} conversion
20279 specifications, including the flags and modifiers between the @samp{%}
20280 character and the conversion letter, with the following exceptions:
20281
20282 @itemize @bullet
20283 @item
20284 The argument-ordering modifiers, such as @samp{2$}, are not supported.
20285
20286 @item
20287 The modifier @samp{*} is not supported for specifying precision or
20288 width.
20289
20290 @item
20291 The @samp{'} flag (for separation of digits into groups according to
20292 @code{LC_NUMERIC'}) is not supported.
20293
20294 @item
20295 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
20296 supported.
20297
20298 @item
20299 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
20300
20301 @item
20302 The conversion letters @samp{a} and @samp{A} are not supported.
20303 @end itemize
20304
20305 @noindent
20306 Note that the @samp{ll} type modifier is supported only if the
20307 underlying @code{C} implementation used to build @value{GDBN} supports
20308 the @code{long long int} type, and the @samp{L} type modifier is
20309 supported only if @code{long double} type is available.
20310
20311 As in @code{C}, @code{printf} supports simple backslash-escape
20312 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
20313 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
20314 single character. Octal and hexadecimal escape sequences are not
20315 supported.
20316
20317 Additionally, @code{printf} supports conversion specifications for DFP
20318 (@dfn{Decimal Floating Point}) types using the following length modifiers
20319 together with a floating point specifier.
20320 letters:
20321
20322 @itemize @bullet
20323 @item
20324 @samp{H} for printing @code{Decimal32} types.
20325
20326 @item
20327 @samp{D} for printing @code{Decimal64} types.
20328
20329 @item
20330 @samp{DD} for printing @code{Decimal128} types.
20331 @end itemize
20332
20333 If the underlying @code{C} implementation used to build @value{GDBN} has
20334 support for the three length modifiers for DFP types, other modifiers
20335 such as width and precision will also be available for @value{GDBN} to use.
20336
20337 In case there is no such @code{C} support, no additional modifiers will be
20338 available and the value will be printed in the standard way.
20339
20340 Here's an example of printing DFP types using the above conversion letters:
20341 @smallexample
20342 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
20343 @end smallexample
20344
20345 @kindex eval
20346 @item eval @var{template}, @var{expressions}@dots{}
20347 Convert the values of one or more @var{expressions} under the control of
20348 the string @var{template} to a command line, and call it.
20349
20350 @end table
20351
20352 @node Python
20353 @section Scripting @value{GDBN} using Python
20354 @cindex python scripting
20355 @cindex scripting with python
20356
20357 You can script @value{GDBN} using the @uref{http://www.python.org/,
20358 Python programming language}. This feature is available only if
20359 @value{GDBN} was configured using @option{--with-python}.
20360
20361 @cindex python directory
20362 Python scripts used by @value{GDBN} should be installed in
20363 @file{@var{data-directory}/python}, where @var{data-directory} is
20364 the data directory as determined at @value{GDBN} startup (@pxref{Data Files}). This directory, known as the @dfn{python directory},
20365 is automatically added to the Python Search Path in order to allow
20366 the Python interpreter to locate all scripts installed at this location.
20367
20368 @menu
20369 * Python Commands:: Accessing Python from @value{GDBN}.
20370 * Python API:: Accessing @value{GDBN} from Python.
20371 * Auto-loading:: Automatically loading Python code.
20372 @end menu
20373
20374 @node Python Commands
20375 @subsection Python Commands
20376 @cindex python commands
20377 @cindex commands to access python
20378
20379 @value{GDBN} provides one command for accessing the Python interpreter,
20380 and one related setting:
20381
20382 @table @code
20383 @kindex python
20384 @item python @r{[}@var{code}@r{]}
20385 The @code{python} command can be used to evaluate Python code.
20386
20387 If given an argument, the @code{python} command will evaluate the
20388 argument as a Python command. For example:
20389
20390 @smallexample
20391 (@value{GDBP}) python print 23
20392 23
20393 @end smallexample
20394
20395 If you do not provide an argument to @code{python}, it will act as a
20396 multi-line command, like @code{define}. In this case, the Python
20397 script is made up of subsequent command lines, given after the
20398 @code{python} command. This command list is terminated using a line
20399 containing @code{end}. For example:
20400
20401 @smallexample
20402 (@value{GDBP}) python
20403 Type python script
20404 End with a line saying just "end".
20405 >print 23
20406 >end
20407 23
20408 @end smallexample
20409
20410 @kindex maint set python print-stack
20411 @item maint set python print-stack
20412 By default, @value{GDBN} will print a stack trace when an error occurs
20413 in a Python script. This can be controlled using @code{maint set
20414 python print-stack}: if @code{on}, the default, then Python stack
20415 printing is enabled; if @code{off}, then Python stack printing is
20416 disabled.
20417 @end table
20418
20419 It is also possible to execute a Python script from the @value{GDBN}
20420 interpreter:
20421
20422 @table @code
20423 @item source @file{script-name}
20424 The script name must end with @samp{.py} and @value{GDBN} must be configured
20425 to recognize the script language based on filename extension using
20426 the @code{script-extension} setting. @xref{Extending GDB, ,Extending GDB}.
20427
20428 @item python execfile ("script-name")
20429 This method is based on the @code{execfile} Python built-in function,
20430 and thus is always available.
20431 @end table
20432
20433 @node Python API
20434 @subsection Python API
20435 @cindex python api
20436 @cindex programming in python
20437
20438 @cindex python stdout
20439 @cindex python pagination
20440 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
20441 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
20442 A Python program which outputs to one of these streams may have its
20443 output interrupted by the user (@pxref{Screen Size}). In this
20444 situation, a Python @code{KeyboardInterrupt} exception is thrown.
20445
20446 @menu
20447 * Basic Python:: Basic Python Functions.
20448 * Exception Handling::
20449 * Values From Inferior::
20450 * Types In Python:: Python representation of types.
20451 * Pretty Printing API:: Pretty-printing values.
20452 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
20453 * Disabling Pretty-Printers:: Disabling broken printers.
20454 * Inferiors In Python:: Python representation of inferiors (processes)
20455 * Threads In Python:: Accessing inferior threads from Python.
20456 * Commands In Python:: Implementing new commands in Python.
20457 * Parameters In Python:: Adding new @value{GDBN} parameters.
20458 * Functions In Python:: Writing new convenience functions.
20459 * Progspaces In Python:: Program spaces.
20460 * Objfiles In Python:: Object files.
20461 * Frames In Python:: Accessing inferior stack frames from Python.
20462 * Blocks In Python:: Accessing frame blocks from Python.
20463 * Symbols In Python:: Python representation of symbols.
20464 * Symbol Tables In Python:: Python representation of symbol tables.
20465 * Lazy Strings In Python:: Python representation of lazy strings.
20466 * Breakpoints In Python:: Manipulating breakpoints using Python.
20467 @end menu
20468
20469 @node Basic Python
20470 @subsubsection Basic Python
20471
20472 @cindex python functions
20473 @cindex python module
20474 @cindex gdb module
20475 @value{GDBN} introduces a new Python module, named @code{gdb}. All
20476 methods and classes added by @value{GDBN} are placed in this module.
20477 @value{GDBN} automatically @code{import}s the @code{gdb} module for
20478 use in all scripts evaluated by the @code{python} command.
20479
20480 @findex gdb.PYTHONDIR
20481 @defvar PYTHONDIR
20482 A string containing the python directory (@pxref{Python}).
20483 @end defvar
20484
20485 @findex gdb.execute
20486 @defun execute command [from_tty] [to_string]
20487 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
20488 If a GDB exception happens while @var{command} runs, it is
20489 translated as described in @ref{Exception Handling,,Exception Handling}.
20490
20491 @var{from_tty} specifies whether @value{GDBN} ought to consider this
20492 command as having originated from the user invoking it interactively.
20493 It must be a boolean value. If omitted, it defaults to @code{False}.
20494
20495 By default, any output produced by @var{command} is sent to
20496 @value{GDBN}'s standard output. If the @var{to_string} parameter is
20497 @code{True}, then output will be collected by @code{gdb.execute} and
20498 returned as a string. The default is @code{False}, in which case the
20499 return value is @code{None}. If @var{to_string} is @code{True}, the
20500 @value{GDBN} virtual terminal will be temporarily set to unlimited width
20501 and height, and its pagination will be disabled; @pxref{Screen Size}.
20502 @end defun
20503
20504 @findex gdb.breakpoints
20505 @defun breakpoints
20506 Return a sequence holding all of @value{GDBN}'s breakpoints.
20507 @xref{Breakpoints In Python}, for more information.
20508 @end defun
20509
20510 @findex gdb.parameter
20511 @defun parameter parameter
20512 Return the value of a @value{GDBN} parameter. @var{parameter} is a
20513 string naming the parameter to look up; @var{parameter} may contain
20514 spaces if the parameter has a multi-part name. For example,
20515 @samp{print object} is a valid parameter name.
20516
20517 If the named parameter does not exist, this function throws a
20518 @code{RuntimeError}. Otherwise, the parameter's value is converted to
20519 a Python value of the appropriate type, and returned.
20520 @end defun
20521
20522 @findex gdb.history
20523 @defun history number
20524 Return a value from @value{GDBN}'s value history (@pxref{Value
20525 History}). @var{number} indicates which history element to return.
20526 If @var{number} is negative, then @value{GDBN} will take its absolute value
20527 and count backward from the last element (i.e., the most recent element) to
20528 find the value to return. If @var{number} is zero, then @value{GDBN} will
20529 return the most recent element. If the element specified by @var{number}
20530 doesn't exist in the value history, a @code{RuntimeError} exception will be
20531 raised.
20532
20533 If no exception is raised, the return value is always an instance of
20534 @code{gdb.Value} (@pxref{Values From Inferior}).
20535 @end defun
20536
20537 @findex gdb.parse_and_eval
20538 @defun parse_and_eval expression
20539 Parse @var{expression} as an expression in the current language,
20540 evaluate it, and return the result as a @code{gdb.Value}.
20541 @var{expression} must be a string.
20542
20543 This function can be useful when implementing a new command
20544 (@pxref{Commands In Python}), as it provides a way to parse the
20545 command's argument as an expression. It is also useful simply to
20546 compute values, for example, it is the only way to get the value of a
20547 convenience variable (@pxref{Convenience Vars}) as a @code{gdb.Value}.
20548 @end defun
20549
20550 @findex gdb.post_event
20551 @defun post_event event
20552 Put @var{event}, a callable object taking no arguments, into
20553 @value{GDBN}'s internal event queue. This callable will be invoked at
20554 some later point, during @value{GDBN}'s event processing. Events
20555 posted using @code{post_event} will be run in the order in which they
20556 were posted; however, there is no way to know when they will be
20557 processed relative to other events inside @value{GDBN}.
20558
20559 @value{GDBN} is not thread-safe. If your Python program uses multiple
20560 threads, you must be careful to only call @value{GDBN}-specific
20561 functions in the main @value{GDBN} thread. @code{post_event} ensures
20562 this. For example:
20563
20564 @smallexample
20565 (@value{GDBP}) python
20566 >import threading
20567 >
20568 >class Writer():
20569 > def __init__(self, message):
20570 > self.message = message;
20571 > def __call__(self):
20572 > gdb.write(self.message)
20573 >
20574 >class MyThread1 (threading.Thread):
20575 > def run (self):
20576 > gdb.post_event(Writer("Hello "))
20577 >
20578 >class MyThread2 (threading.Thread):
20579 > def run (self):
20580 > gdb.post_event(Writer("World\n"))
20581 >
20582 >MyThread1().start()
20583 >MyThread2().start()
20584 >end
20585 (@value{GDBP}) Hello World
20586 @end smallexample
20587 @end defun
20588
20589 @findex gdb.write
20590 @defun write string
20591 Print a string to @value{GDBN}'s paginated standard output stream.
20592 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
20593 call this function.
20594 @end defun
20595
20596 @findex gdb.flush
20597 @defun flush
20598 Flush @value{GDBN}'s paginated standard output stream. Flushing
20599 @code{sys.stdout} or @code{sys.stderr} will automatically call this
20600 function.
20601 @end defun
20602
20603 @findex gdb.target_charset
20604 @defun target_charset
20605 Return the name of the current target character set (@pxref{Character
20606 Sets}). This differs from @code{gdb.parameter('target-charset')} in
20607 that @samp{auto} is never returned.
20608 @end defun
20609
20610 @findex gdb.target_wide_charset
20611 @defun target_wide_charset
20612 Return the name of the current target wide character set
20613 (@pxref{Character Sets}). This differs from
20614 @code{gdb.parameter('target-wide-charset')} in that @samp{auto} is
20615 never returned.
20616 @end defun
20617
20618 @findex gdb.solib_name
20619 @defun solib_name address
20620 Return the name of the shared library holding the given @var{address}
20621 as a string, or @code{None}.
20622 @end defun
20623
20624 @findex gdb.decode_line
20625 @defun decode_line @r{[}expression@r{]}
20626 Return locations of the line specified by @var{expression}, or of the
20627 current line if no argument was given. This function returns a Python
20628 tuple containing two elements. The first element contains a string
20629 holding any unparsed section of @var{expression} (or @code{None} if
20630 the expression has been fully parsed). The second element contains
20631 either @code{None} or another tuple that contains all the locations
20632 that match the expression represented as @code{gdb.Symtab_and_line}
20633 objects (@pxref{Symbol Tables In Python}). If @var{expression} is
20634 provided, it is decoded the way that @value{GDBN}'s inbuilt
20635 @code{break} or @code{edit} commands do (@pxref{Specify Location}).
20636 @end defun
20637
20638 @node Exception Handling
20639 @subsubsection Exception Handling
20640 @cindex python exceptions
20641 @cindex exceptions, python
20642
20643 When executing the @code{python} command, Python exceptions
20644 uncaught within the Python code are translated to calls to
20645 @value{GDBN} error-reporting mechanism. If the command that called
20646 @code{python} does not handle the error, @value{GDBN} will
20647 terminate it and print an error message containing the Python
20648 exception name, the associated value, and the Python call stack
20649 backtrace at the point where the exception was raised. Example:
20650
20651 @smallexample
20652 (@value{GDBP}) python print foo
20653 Traceback (most recent call last):
20654 File "<string>", line 1, in <module>
20655 NameError: name 'foo' is not defined
20656 @end smallexample
20657
20658 @value{GDBN} errors that happen in @value{GDBN} commands invoked by Python
20659 code are converted to Python @code{RuntimeError} exceptions. User
20660 interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
20661 prompt) is translated to a Python @code{KeyboardInterrupt}
20662 exception. If you catch these exceptions in your Python code, your
20663 exception handler will see @code{RuntimeError} or
20664 @code{KeyboardInterrupt} as the exception type, the @value{GDBN} error
20665 message as its value, and the Python call stack backtrace at the
20666 Python statement closest to where the @value{GDBN} error occured as the
20667 traceback.
20668
20669 @findex gdb.GdbError
20670 When implementing @value{GDBN} commands in Python via @code{gdb.Command},
20671 it is useful to be able to throw an exception that doesn't cause a
20672 traceback to be printed. For example, the user may have invoked the
20673 command incorrectly. Use the @code{gdb.GdbError} exception
20674 to handle this case. Example:
20675
20676 @smallexample
20677 (gdb) python
20678 >class HelloWorld (gdb.Command):
20679 > """Greet the whole world."""
20680 > def __init__ (self):
20681 > super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_OBSCURE)
20682 > def invoke (self, args, from_tty):
20683 > argv = gdb.string_to_argv (args)
20684 > if len (argv) != 0:
20685 > raise gdb.GdbError ("hello-world takes no arguments")
20686 > print "Hello, World!"
20687 >HelloWorld ()
20688 >end
20689 (gdb) hello-world 42
20690 hello-world takes no arguments
20691 @end smallexample
20692
20693 @node Values From Inferior
20694 @subsubsection Values From Inferior
20695 @cindex values from inferior, with Python
20696 @cindex python, working with values from inferior
20697
20698 @cindex @code{gdb.Value}
20699 @value{GDBN} provides values it obtains from the inferior program in
20700 an object of type @code{gdb.Value}. @value{GDBN} uses this object
20701 for its internal bookkeeping of the inferior's values, and for
20702 fetching values when necessary.
20703
20704 Inferior values that are simple scalars can be used directly in
20705 Python expressions that are valid for the value's data type. Here's
20706 an example for an integer or floating-point value @code{some_val}:
20707
20708 @smallexample
20709 bar = some_val + 2
20710 @end smallexample
20711
20712 @noindent
20713 As result of this, @code{bar} will also be a @code{gdb.Value} object
20714 whose values are of the same type as those of @code{some_val}.
20715
20716 Inferior values that are structures or instances of some class can
20717 be accessed using the Python @dfn{dictionary syntax}. For example, if
20718 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
20719 can access its @code{foo} element with:
20720
20721 @smallexample
20722 bar = some_val['foo']
20723 @end smallexample
20724
20725 Again, @code{bar} will also be a @code{gdb.Value} object.
20726
20727 A @code{gdb.Value} that represents a function can be executed via
20728 inferior function call. Any arguments provided to the call must match
20729 the function's prototype, and must be provided in the order specified
20730 by that prototype.
20731
20732 For example, @code{some_val} is a @code{gdb.Value} instance
20733 representing a function that takes two integers as arguments. To
20734 execute this function, call it like so:
20735
20736 @smallexample
20737 result = some_val (10,20)
20738 @end smallexample
20739
20740 Any values returned from a function call will be stored as a
20741 @code{gdb.Value}.
20742
20743 The following attributes are provided:
20744
20745 @table @code
20746 @defivar Value address
20747 If this object is addressable, this read-only attribute holds a
20748 @code{gdb.Value} object representing the address. Otherwise,
20749 this attribute holds @code{None}.
20750 @end defivar
20751
20752 @cindex optimized out value in Python
20753 @defivar Value is_optimized_out
20754 This read-only boolean attribute is true if the compiler optimized out
20755 this value, thus it is not available for fetching from the inferior.
20756 @end defivar
20757
20758 @defivar Value type
20759 The type of this @code{gdb.Value}. The value of this attribute is a
20760 @code{gdb.Type} object.
20761 @end defivar
20762
20763 @defivar Value dynamic_type
20764 The dynamic type of this @code{gdb.Value}. This uses C@t{++} run-time
20765 type information to determine the dynamic type of the value. If this
20766 value is of class type, it will return the class in which the value is
20767 embedded, if any. If this value is of pointer or reference to a class
20768 type, it will compute the dynamic type of the referenced object, and
20769 return a pointer or reference to that type, respectively. In all
20770 other cases, it will return the value's static type.
20771 @end defivar
20772 @end table
20773
20774 The following methods are provided:
20775
20776 @table @code
20777 @defmethod Value cast type
20778 Return a new instance of @code{gdb.Value} that is the result of
20779 casting this instance to the type described by @var{type}, which must
20780 be a @code{gdb.Type} object. If the cast cannot be performed for some
20781 reason, this method throws an exception.
20782 @end defmethod
20783
20784 @defmethod Value dereference
20785 For pointer data types, this method returns a new @code{gdb.Value} object
20786 whose contents is the object pointed to by the pointer. For example, if
20787 @code{foo} is a C pointer to an @code{int}, declared in your C program as
20788
20789 @smallexample
20790 int *foo;
20791 @end smallexample
20792
20793 @noindent
20794 then you can use the corresponding @code{gdb.Value} to access what
20795 @code{foo} points to like this:
20796
20797 @smallexample
20798 bar = foo.dereference ()
20799 @end smallexample
20800
20801 The result @code{bar} will be a @code{gdb.Value} object holding the
20802 value pointed to by @code{foo}.
20803 @end defmethod
20804
20805 @defmethod Value dynamic_cast type
20806 Like @code{Value.cast}, but works as if the C@t{++} @code{dynamic_cast}
20807 operator were used. Consult a C@t{++} reference for details.
20808 @end defmethod
20809
20810 @defmethod Value reinterpret_cast type
20811 Like @code{Value.cast}, but works as if the C@t{++} @code{reinterpret_cast}
20812 operator were used. Consult a C@t{++} reference for details.
20813 @end defmethod
20814
20815 @defmethod Value string @r{[}encoding@r{]} @r{[}errors@r{]} @r{[}length@r{]}
20816 If this @code{gdb.Value} represents a string, then this method
20817 converts the contents to a Python string. Otherwise, this method will
20818 throw an exception.
20819
20820 Strings are recognized in a language-specific way; whether a given
20821 @code{gdb.Value} represents a string is determined by the current
20822 language.
20823
20824 For C-like languages, a value is a string if it is a pointer to or an
20825 array of characters or ints. The string is assumed to be terminated
20826 by a zero of the appropriate width. However if the optional length
20827 argument is given, the string will be converted to that given length,
20828 ignoring any embedded zeros that the string may contain.
20829
20830 If the optional @var{encoding} argument is given, it must be a string
20831 naming the encoding of the string in the @code{gdb.Value}, such as
20832 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
20833 the same encodings as the corresponding argument to Python's
20834 @code{string.decode} method, and the Python codec machinery will be used
20835 to convert the string. If @var{encoding} is not given, or if
20836 @var{encoding} is the empty string, then either the @code{target-charset}
20837 (@pxref{Character Sets}) will be used, or a language-specific encoding
20838 will be used, if the current language is able to supply one.
20839
20840 The optional @var{errors} argument is the same as the corresponding
20841 argument to Python's @code{string.decode} method.
20842
20843 If the optional @var{length} argument is given, the string will be
20844 fetched and converted to the given length.
20845 @end defmethod
20846
20847 @defmethod Value lazy_string @r{[}encoding@r{]} @r{[}length@r{]}
20848 If this @code{gdb.Value} represents a string, then this method
20849 converts the contents to a @code{gdb.LazyString} (@pxref{Lazy Strings
20850 In Python}). Otherwise, this method will throw an exception.
20851
20852 If the optional @var{encoding} argument is given, it must be a string
20853 naming the encoding of the @code{gdb.LazyString}. Some examples are:
20854 @samp{ascii}, @samp{iso-8859-6} or @samp{utf-8}. If the
20855 @var{encoding} argument is an encoding that @value{GDBN} does
20856 recognize, @value{GDBN} will raise an error.
20857
20858 When a lazy string is printed, the @value{GDBN} encoding machinery is
20859 used to convert the string during printing. If the optional
20860 @var{encoding} argument is not provided, or is an empty string,
20861 @value{GDBN} will automatically select the encoding most suitable for
20862 the string type. For further information on encoding in @value{GDBN}
20863 please see @ref{Character Sets}.
20864
20865 If the optional @var{length} argument is given, the string will be
20866 fetched and encoded to the length of characters specified. If
20867 the @var{length} argument is not provided, the string will be fetched
20868 and encoded until a null of appropriate width is found.
20869 @end defmethod
20870 @end table
20871
20872 @node Types In Python
20873 @subsubsection Types In Python
20874 @cindex types in Python
20875 @cindex Python, working with types
20876
20877 @tindex gdb.Type
20878 @value{GDBN} represents types from the inferior using the class
20879 @code{gdb.Type}.
20880
20881 The following type-related functions are available in the @code{gdb}
20882 module:
20883
20884 @findex gdb.lookup_type
20885 @defun lookup_type name [block]
20886 This function looks up a type by name. @var{name} is the name of the
20887 type to look up. It must be a string.
20888
20889 If @var{block} is given, then @var{name} is looked up in that scope.
20890 Otherwise, it is searched for globally.
20891
20892 Ordinarily, this function will return an instance of @code{gdb.Type}.
20893 If the named type cannot be found, it will throw an exception.
20894 @end defun
20895
20896 An instance of @code{Type} has the following attributes:
20897
20898 @table @code
20899 @defivar Type code
20900 The type code for this type. The type code will be one of the
20901 @code{TYPE_CODE_} constants defined below.
20902 @end defivar
20903
20904 @defivar Type sizeof
20905 The size of this type, in target @code{char} units. Usually, a
20906 target's @code{char} type will be an 8-bit byte. However, on some
20907 unusual platforms, this type may have a different size.
20908 @end defivar
20909
20910 @defivar Type tag
20911 The tag name for this type. The tag name is the name after
20912 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
20913 languages have this concept. If this type has no tag name, then
20914 @code{None} is returned.
20915 @end defivar
20916 @end table
20917
20918 The following methods are provided:
20919
20920 @table @code
20921 @defmethod Type fields
20922 For structure and union types, this method returns the fields. Range
20923 types have two fields, the minimum and maximum values. Enum types
20924 have one field per enum constant. Function and method types have one
20925 field per parameter. The base types of C@t{++} classes are also
20926 represented as fields. If the type has no fields, or does not fit
20927 into one of these categories, an empty sequence will be returned.
20928
20929 Each field is an object, with some pre-defined attributes:
20930 @table @code
20931 @item bitpos
20932 This attribute is not available for @code{static} fields (as in
20933 C@t{++} or Java). For non-@code{static} fields, the value is the bit
20934 position of the field.
20935
20936 @item name
20937 The name of the field, or @code{None} for anonymous fields.
20938
20939 @item artificial
20940 This is @code{True} if the field is artificial, usually meaning that
20941 it was provided by the compiler and not the user. This attribute is
20942 always provided, and is @code{False} if the field is not artificial.
20943
20944 @item is_base_class
20945 This is @code{True} if the field represents a base class of a C@t{++}
20946 structure. This attribute is always provided, and is @code{False}
20947 if the field is not a base class of the type that is the argument of
20948 @code{fields}, or if that type was not a C@t{++} class.
20949
20950 @item bitsize
20951 If the field is packed, or is a bitfield, then this will have a
20952 non-zero value, which is the size of the field in bits. Otherwise,
20953 this will be zero; in this case the field's size is given by its type.
20954
20955 @item type
20956 The type of the field. This is usually an instance of @code{Type},
20957 but it can be @code{None} in some situations.
20958 @end table
20959 @end defmethod
20960
20961 @defmethod Type array @var{n1} @r{[}@var{n2}@r{]}
20962 Return a new @code{gdb.Type} object which represents an array of this
20963 type. If one argument is given, it is the inclusive upper bound of
20964 the array; in this case the lower bound is zero. If two arguments are
20965 given, the first argument is the lower bound of the array, and the
20966 second argument is the upper bound of the array. An array's length
20967 must not be negative, but the bounds can be.
20968 @end defmethod
20969
20970 @defmethod Type const
20971 Return a new @code{gdb.Type} object which represents a
20972 @code{const}-qualified variant of this type.
20973 @end defmethod
20974
20975 @defmethod Type volatile
20976 Return a new @code{gdb.Type} object which represents a
20977 @code{volatile}-qualified variant of this type.
20978 @end defmethod
20979
20980 @defmethod Type unqualified
20981 Return a new @code{gdb.Type} object which represents an unqualified
20982 variant of this type. That is, the result is neither @code{const} nor
20983 @code{volatile}.
20984 @end defmethod
20985
20986 @defmethod Type range
20987 Return a Python @code{Tuple} object that contains two elements: the
20988 low bound of the argument type and the high bound of that type. If
20989 the type does not have a range, @value{GDBN} will raise a
20990 @code{RuntimeError} exception.
20991 @end defmethod
20992
20993 @defmethod Type reference
20994 Return a new @code{gdb.Type} object which represents a reference to this
20995 type.
20996 @end defmethod
20997
20998 @defmethod Type pointer
20999 Return a new @code{gdb.Type} object which represents a pointer to this
21000 type.
21001 @end defmethod
21002
21003 @defmethod Type strip_typedefs
21004 Return a new @code{gdb.Type} that represents the real type,
21005 after removing all layers of typedefs.
21006 @end defmethod
21007
21008 @defmethod Type target
21009 Return a new @code{gdb.Type} object which represents the target type
21010 of this type.
21011
21012 For a pointer type, the target type is the type of the pointed-to
21013 object. For an array type (meaning C-like arrays), the target type is
21014 the type of the elements of the array. For a function or method type,
21015 the target type is the type of the return value. For a complex type,
21016 the target type is the type of the elements. For a typedef, the
21017 target type is the aliased type.
21018
21019 If the type does not have a target, this method will throw an
21020 exception.
21021 @end defmethod
21022
21023 @defmethod Type template_argument n [block]
21024 If this @code{gdb.Type} is an instantiation of a template, this will
21025 return a new @code{gdb.Type} which represents the type of the
21026 @var{n}th template argument.
21027
21028 If this @code{gdb.Type} is not a template type, this will throw an
21029 exception. Ordinarily, only C@t{++} code will have template types.
21030
21031 If @var{block} is given, then @var{name} is looked up in that scope.
21032 Otherwise, it is searched for globally.
21033 @end defmethod
21034 @end table
21035
21036
21037 Each type has a code, which indicates what category this type falls
21038 into. The available type categories are represented by constants
21039 defined in the @code{gdb} module:
21040
21041 @table @code
21042 @findex TYPE_CODE_PTR
21043 @findex gdb.TYPE_CODE_PTR
21044 @item TYPE_CODE_PTR
21045 The type is a pointer.
21046
21047 @findex TYPE_CODE_ARRAY
21048 @findex gdb.TYPE_CODE_ARRAY
21049 @item TYPE_CODE_ARRAY
21050 The type is an array.
21051
21052 @findex TYPE_CODE_STRUCT
21053 @findex gdb.TYPE_CODE_STRUCT
21054 @item TYPE_CODE_STRUCT
21055 The type is a structure.
21056
21057 @findex TYPE_CODE_UNION
21058 @findex gdb.TYPE_CODE_UNION
21059 @item TYPE_CODE_UNION
21060 The type is a union.
21061
21062 @findex TYPE_CODE_ENUM
21063 @findex gdb.TYPE_CODE_ENUM
21064 @item TYPE_CODE_ENUM
21065 The type is an enum.
21066
21067 @findex TYPE_CODE_FLAGS
21068 @findex gdb.TYPE_CODE_FLAGS
21069 @item TYPE_CODE_FLAGS
21070 A bit flags type, used for things such as status registers.
21071
21072 @findex TYPE_CODE_FUNC
21073 @findex gdb.TYPE_CODE_FUNC
21074 @item TYPE_CODE_FUNC
21075 The type is a function.
21076
21077 @findex TYPE_CODE_INT
21078 @findex gdb.TYPE_CODE_INT
21079 @item TYPE_CODE_INT
21080 The type is an integer type.
21081
21082 @findex TYPE_CODE_FLT
21083 @findex gdb.TYPE_CODE_FLT
21084 @item TYPE_CODE_FLT
21085 A floating point type.
21086
21087 @findex TYPE_CODE_VOID
21088 @findex gdb.TYPE_CODE_VOID
21089 @item TYPE_CODE_VOID
21090 The special type @code{void}.
21091
21092 @findex TYPE_CODE_SET
21093 @findex gdb.TYPE_CODE_SET
21094 @item TYPE_CODE_SET
21095 A Pascal set type.
21096
21097 @findex TYPE_CODE_RANGE
21098 @findex gdb.TYPE_CODE_RANGE
21099 @item TYPE_CODE_RANGE
21100 A range type, that is, an integer type with bounds.
21101
21102 @findex TYPE_CODE_STRING
21103 @findex gdb.TYPE_CODE_STRING
21104 @item TYPE_CODE_STRING
21105 A string type. Note that this is only used for certain languages with
21106 language-defined string types; C strings are not represented this way.
21107
21108 @findex TYPE_CODE_BITSTRING
21109 @findex gdb.TYPE_CODE_BITSTRING
21110 @item TYPE_CODE_BITSTRING
21111 A string of bits.
21112
21113 @findex TYPE_CODE_ERROR
21114 @findex gdb.TYPE_CODE_ERROR
21115 @item TYPE_CODE_ERROR
21116 An unknown or erroneous type.
21117
21118 @findex TYPE_CODE_METHOD
21119 @findex gdb.TYPE_CODE_METHOD
21120 @item TYPE_CODE_METHOD
21121 A method type, as found in C@t{++} or Java.
21122
21123 @findex TYPE_CODE_METHODPTR
21124 @findex gdb.TYPE_CODE_METHODPTR
21125 @item TYPE_CODE_METHODPTR
21126 A pointer-to-member-function.
21127
21128 @findex TYPE_CODE_MEMBERPTR
21129 @findex gdb.TYPE_CODE_MEMBERPTR
21130 @item TYPE_CODE_MEMBERPTR
21131 A pointer-to-member.
21132
21133 @findex TYPE_CODE_REF
21134 @findex gdb.TYPE_CODE_REF
21135 @item TYPE_CODE_REF
21136 A reference type.
21137
21138 @findex TYPE_CODE_CHAR
21139 @findex gdb.TYPE_CODE_CHAR
21140 @item TYPE_CODE_CHAR
21141 A character type.
21142
21143 @findex TYPE_CODE_BOOL
21144 @findex gdb.TYPE_CODE_BOOL
21145 @item TYPE_CODE_BOOL
21146 A boolean type.
21147
21148 @findex TYPE_CODE_COMPLEX
21149 @findex gdb.TYPE_CODE_COMPLEX
21150 @item TYPE_CODE_COMPLEX
21151 A complex float type.
21152
21153 @findex TYPE_CODE_TYPEDEF
21154 @findex gdb.TYPE_CODE_TYPEDEF
21155 @item TYPE_CODE_TYPEDEF
21156 A typedef to some other type.
21157
21158 @findex TYPE_CODE_NAMESPACE
21159 @findex gdb.TYPE_CODE_NAMESPACE
21160 @item TYPE_CODE_NAMESPACE
21161 A C@t{++} namespace.
21162
21163 @findex TYPE_CODE_DECFLOAT
21164 @findex gdb.TYPE_CODE_DECFLOAT
21165 @item TYPE_CODE_DECFLOAT
21166 A decimal floating point type.
21167
21168 @findex TYPE_CODE_INTERNAL_FUNCTION
21169 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
21170 @item TYPE_CODE_INTERNAL_FUNCTION
21171 A function internal to @value{GDBN}. This is the type used to represent
21172 convenience functions.
21173 @end table
21174
21175 @node Pretty Printing API
21176 @subsubsection Pretty Printing API
21177
21178 An example output is provided (@pxref{Pretty Printing}).
21179
21180 A pretty-printer is just an object that holds a value and implements a
21181 specific interface, defined here.
21182
21183 @defop Operation {pretty printer} children (self)
21184 @value{GDBN} will call this method on a pretty-printer to compute the
21185 children of the pretty-printer's value.
21186
21187 This method must return an object conforming to the Python iterator
21188 protocol. Each item returned by the iterator must be a tuple holding
21189 two elements. The first element is the ``name'' of the child; the
21190 second element is the child's value. The value can be any Python
21191 object which is convertible to a @value{GDBN} value.
21192
21193 This method is optional. If it does not exist, @value{GDBN} will act
21194 as though the value has no children.
21195 @end defop
21196
21197 @defop Operation {pretty printer} display_hint (self)
21198 The CLI may call this method and use its result to change the
21199 formatting of a value. The result will also be supplied to an MI
21200 consumer as a @samp{displayhint} attribute of the variable being
21201 printed.
21202
21203 This method is optional. If it does exist, this method must return a
21204 string.
21205
21206 Some display hints are predefined by @value{GDBN}:
21207
21208 @table @samp
21209 @item array
21210 Indicate that the object being printed is ``array-like''. The CLI
21211 uses this to respect parameters such as @code{set print elements} and
21212 @code{set print array}.
21213
21214 @item map
21215 Indicate that the object being printed is ``map-like'', and that the
21216 children of this value can be assumed to alternate between keys and
21217 values.
21218
21219 @item string
21220 Indicate that the object being printed is ``string-like''. If the
21221 printer's @code{to_string} method returns a Python string of some
21222 kind, then @value{GDBN} will call its internal language-specific
21223 string-printing function to format the string. For the CLI this means
21224 adding quotation marks, possibly escaping some characters, respecting
21225 @code{set print elements}, and the like.
21226 @end table
21227 @end defop
21228
21229 @defop Operation {pretty printer} to_string (self)
21230 @value{GDBN} will call this method to display the string
21231 representation of the value passed to the object's constructor.
21232
21233 When printing from the CLI, if the @code{to_string} method exists,
21234 then @value{GDBN} will prepend its result to the values returned by
21235 @code{children}. Exactly how this formatting is done is dependent on
21236 the display hint, and may change as more hints are added. Also,
21237 depending on the print settings (@pxref{Print Settings}), the CLI may
21238 print just the result of @code{to_string} in a stack trace, omitting
21239 the result of @code{children}.
21240
21241 If this method returns a string, it is printed verbatim.
21242
21243 Otherwise, if this method returns an instance of @code{gdb.Value},
21244 then @value{GDBN} prints this value. This may result in a call to
21245 another pretty-printer.
21246
21247 If instead the method returns a Python value which is convertible to a
21248 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
21249 the resulting value. Again, this may result in a call to another
21250 pretty-printer. Python scalars (integers, floats, and booleans) and
21251 strings are convertible to @code{gdb.Value}; other types are not.
21252
21253 Finally, if this method returns @code{None} then no further operations
21254 are peformed in this method and nothing is printed.
21255
21256 If the result is not one of these types, an exception is raised.
21257 @end defop
21258
21259 @value{GDBN} provides a function which can be used to look up the
21260 default pretty-printer for a @code{gdb.Value}:
21261
21262 @findex gdb.default_visualizer
21263 @defun default_visualizer value
21264 This function takes a @code{gdb.Value} object as an argument. If a
21265 pretty-printer for this value exists, then it is returned. If no such
21266 printer exists, then this returns @code{None}.
21267 @end defun
21268
21269 @node Selecting Pretty-Printers
21270 @subsubsection Selecting Pretty-Printers
21271
21272 The Python list @code{gdb.pretty_printers} contains an array of
21273 functions or callable objects that have been registered via addition
21274 as a pretty-printer.
21275 Each @code{gdb.Progspace} contains a @code{pretty_printers} attribute.
21276 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
21277 attribute.
21278
21279 A function on one of these lists is passed a single @code{gdb.Value}
21280 argument and should return a pretty-printer object conforming to the
21281 interface definition above (@pxref{Pretty Printing API}). If a function
21282 cannot create a pretty-printer for the value, it should return
21283 @code{None}.
21284
21285 @value{GDBN} first checks the @code{pretty_printers} attribute of each
21286 @code{gdb.Objfile} in the current program space and iteratively calls
21287 each enabled function (@pxref{Disabling Pretty-Printers})
21288 in the list for that @code{gdb.Objfile} until it receives
21289 a pretty-printer object.
21290 If no pretty-printer is found in the objfile lists, @value{GDBN} then
21291 searches the pretty-printer list of the current program space,
21292 calling each enabled function until an object is returned.
21293 After these lists have been exhausted, it tries the global
21294 @code{gdb.pretty_printers} list, again calling each enabled function until an
21295 object is returned.
21296
21297 The order in which the objfiles are searched is not specified. For a
21298 given list, functions are always invoked from the head of the list,
21299 and iterated over sequentially until the end of the list, or a printer
21300 object is returned.
21301
21302 Here is an example showing how a @code{std::string} printer might be
21303 written:
21304
21305 @smallexample
21306 class StdStringPrinter:
21307 "Print a std::string"
21308
21309 def __init__ (self, val):
21310 self.val = val
21311
21312 def to_string (self):
21313 return self.val['_M_dataplus']['_M_p']
21314
21315 def display_hint (self):
21316 return 'string'
21317 @end smallexample
21318
21319 And here is an example showing how a lookup function for the printer
21320 example above might be written.
21321
21322 @smallexample
21323 def str_lookup_function (val):
21324
21325 lookup_tag = val.type.tag
21326 regex = re.compile ("^std::basic_string<char,.*>$")
21327 if lookup_tag == None:
21328 return None
21329 if regex.match (lookup_tag):
21330 return StdStringPrinter (val)
21331
21332 return None
21333 @end smallexample
21334
21335 The example lookup function extracts the value's type, and attempts to
21336 match it to a type that it can pretty-print. If it is a type the
21337 printer can pretty-print, it will return a printer object. If not, it
21338 returns @code{None}.
21339
21340 We recommend that you put your core pretty-printers into a Python
21341 package. If your pretty-printers are for use with a library, we
21342 further recommend embedding a version number into the package name.
21343 This practice will enable @value{GDBN} to load multiple versions of
21344 your pretty-printers at the same time, because they will have
21345 different names.
21346
21347 You should write auto-loaded code (@pxref{Auto-loading}) such that it
21348 can be evaluated multiple times without changing its meaning. An
21349 ideal auto-load file will consist solely of @code{import}s of your
21350 printer modules, followed by a call to a register pretty-printers with
21351 the current objfile.
21352
21353 Taken as a whole, this approach will scale nicely to multiple
21354 inferiors, each potentially using a different library version.
21355 Embedding a version number in the Python package name will ensure that
21356 @value{GDBN} is able to load both sets of printers simultaneously.
21357 Then, because the search for pretty-printers is done by objfile, and
21358 because your auto-loaded code took care to register your library's
21359 printers with a specific objfile, @value{GDBN} will find the correct
21360 printers for the specific version of the library used by each
21361 inferior.
21362
21363 To continue the @code{std::string} example (@pxref{Pretty Printing API}),
21364 this code might appear in @code{gdb.libstdcxx.v6}:
21365
21366 @smallexample
21367 def register_printers (objfile):
21368 objfile.pretty_printers.add (str_lookup_function)
21369 @end smallexample
21370
21371 @noindent
21372 And then the corresponding contents of the auto-load file would be:
21373
21374 @smallexample
21375 import gdb.libstdcxx.v6
21376 gdb.libstdcxx.v6.register_printers (gdb.current_objfile ())
21377 @end smallexample
21378
21379 @node Disabling Pretty-Printers
21380 @subsubsection Disabling Pretty-Printers
21381 @cindex disabling pretty-printers
21382
21383 For various reasons a pretty-printer may not work.
21384 For example, the underlying data structure may have changed and
21385 the pretty-printer is out of date.
21386
21387 The consequences of a broken pretty-printer are severe enough that
21388 @value{GDBN} provides support for enabling and disabling individual
21389 printers. For example, if @code{print frame-arguments} is on,
21390 a backtrace can become highly illegible if any argument is printed
21391 with a broken printer.
21392
21393 Pretty-printers are enabled and disabled by attaching an @code{enabled}
21394 attribute to the registered function or callable object. If this attribute
21395 is present and its value is @code{False}, the printer is disabled, otherwise
21396 the printer is enabled.
21397
21398 @node Inferiors In Python
21399 @subsubsection Inferiors In Python
21400 @cindex inferiors in python
21401
21402 @findex gdb.Inferior
21403 Programs which are being run under @value{GDBN} are called inferiors
21404 (@pxref{Inferiors and Programs}). Python scripts can access
21405 information about and manipulate inferiors controlled by @value{GDBN}
21406 via objects of the @code{gdb.Inferior} class.
21407
21408 The following inferior-related functions are available in the @code{gdb}
21409 module:
21410
21411 @defun inferiors
21412 Return a tuple containing all inferior objects.
21413 @end defun
21414
21415 A @code{gdb.Inferior} object has the following attributes:
21416
21417 @table @code
21418 @defivar Inferior num
21419 ID of inferior, as assigned by GDB.
21420 @end defivar
21421
21422 @defivar Inferior pid
21423 Process ID of the inferior, as assigned by the underlying operating
21424 system.
21425 @end defivar
21426
21427 @defivar Inferior was_attached
21428 Boolean signaling whether the inferior was created using `attach', or
21429 started by @value{GDBN} itself.
21430 @end defivar
21431 @end table
21432
21433 A @code{gdb.Inferior} object has the following methods:
21434
21435 @table @code
21436 @defmethod Inferior threads
21437 This method returns a tuple holding all the threads which are valid
21438 when it is called. If there are no valid threads, the method will
21439 return an empty tuple.
21440 @end defmethod
21441
21442 @findex gdb.read_memory
21443 @defmethod Inferior read_memory address length
21444 Read @var{length} bytes of memory from the inferior, starting at
21445 @var{address}. Returns a buffer object, which behaves much like an array
21446 or a string. It can be modified and given to the @code{gdb.write_memory}
21447 function.
21448 @end defmethod
21449
21450 @findex gdb.write_memory
21451 @defmethod Inferior write_memory address buffer @r{[}length@r{]}
21452 Write the contents of @var{buffer} to the inferior, starting at
21453 @var{address}. The @var{buffer} parameter must be a Python object
21454 which supports the buffer protocol, i.e., a string, an array or the
21455 object returned from @code{gdb.read_memory}. If given, @var{length}
21456 determines the number of bytes from @var{buffer} to be written.
21457 @end defmethod
21458
21459 @findex gdb.search_memory
21460 @defmethod Inferior search_memory address length pattern
21461 Search a region of the inferior memory starting at @var{address} with
21462 the given @var{length} using the search pattern supplied in
21463 @var{pattern}. The @var{pattern} parameter must be a Python object
21464 which supports the buffer protocol, i.e., a string, an array or the
21465 object returned from @code{gdb.read_memory}. Returns a Python @code{Long}
21466 containing the address where the pattern was found, or @code{None} if
21467 the pattern could not be found.
21468 @end defmethod
21469 @end table
21470
21471 @node Threads In Python
21472 @subsubsection Threads In Python
21473 @cindex threads in python
21474
21475 @findex gdb.InferiorThread
21476 Python scripts can access information about, and manipulate inferior threads
21477 controlled by @value{GDBN}, via objects of the @code{gdb.InferiorThread} class.
21478
21479 The following thread-related functions are available in the @code{gdb}
21480 module:
21481
21482 @findex gdb.selected_thread
21483 @defun selected_thread
21484 This function returns the thread object for the selected thread. If there
21485 is no selected thread, this will return @code{None}.
21486 @end defun
21487
21488 A @code{gdb.InferiorThread} object has the following attributes:
21489
21490 @table @code
21491 @defivar InferiorThread num
21492 ID of the thread, as assigned by GDB.
21493 @end defivar
21494
21495 @defivar InferiorThread ptid
21496 ID of the thread, as assigned by the operating system. This attribute is a
21497 tuple containing three integers. The first is the Process ID (PID); the second
21498 is the Lightweight Process ID (LWPID), and the third is the Thread ID (TID).
21499 Either the LWPID or TID may be 0, which indicates that the operating system
21500 does not use that identifier.
21501 @end defivar
21502 @end table
21503
21504 A @code{gdb.InferiorThread} object has the following methods:
21505
21506 @table @code
21507 @defmethod InferiorThread switch
21508 This changes @value{GDBN}'s currently selected thread to the one represented
21509 by this object.
21510 @end defmethod
21511
21512 @defmethod InferiorThread is_stopped
21513 Return a Boolean indicating whether the thread is stopped.
21514 @end defmethod
21515
21516 @defmethod InferiorThread is_running
21517 Return a Boolean indicating whether the thread is running.
21518 @end defmethod
21519
21520 @defmethod InferiorThread is_exited
21521 Return a Boolean indicating whether the thread is exited.
21522 @end defmethod
21523 @end table
21524
21525 @node Commands In Python
21526 @subsubsection Commands In Python
21527
21528 @cindex commands in python
21529 @cindex python commands
21530 You can implement new @value{GDBN} CLI commands in Python. A CLI
21531 command is implemented using an instance of the @code{gdb.Command}
21532 class, most commonly using a subclass.
21533
21534 @defmethod Command __init__ name @var{command_class} @r{[}@var{completer_class}@r{]} @r{[}@var{prefix}@r{]}
21535 The object initializer for @code{Command} registers the new command
21536 with @value{GDBN}. This initializer is normally invoked from the
21537 subclass' own @code{__init__} method.
21538
21539 @var{name} is the name of the command. If @var{name} consists of
21540 multiple words, then the initial words are looked for as prefix
21541 commands. In this case, if one of the prefix commands does not exist,
21542 an exception is raised.
21543
21544 There is no support for multi-line commands.
21545
21546 @var{command_class} should be one of the @samp{COMMAND_} constants
21547 defined below. This argument tells @value{GDBN} how to categorize the
21548 new command in the help system.
21549
21550 @var{completer_class} is an optional argument. If given, it should be
21551 one of the @samp{COMPLETE_} constants defined below. This argument
21552 tells @value{GDBN} how to perform completion for this command. If not
21553 given, @value{GDBN} will attempt to complete using the object's
21554 @code{complete} method (see below); if no such method is found, an
21555 error will occur when completion is attempted.
21556
21557 @var{prefix} is an optional argument. If @code{True}, then the new
21558 command is a prefix command; sub-commands of this command may be
21559 registered.
21560
21561 The help text for the new command is taken from the Python
21562 documentation string for the command's class, if there is one. If no
21563 documentation string is provided, the default value ``This command is
21564 not documented.'' is used.
21565 @end defmethod
21566
21567 @cindex don't repeat Python command
21568 @defmethod Command dont_repeat
21569 By default, a @value{GDBN} command is repeated when the user enters a
21570 blank line at the command prompt. A command can suppress this
21571 behavior by invoking the @code{dont_repeat} method. This is similar
21572 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
21573 @end defmethod
21574
21575 @defmethod Command invoke argument from_tty
21576 This method is called by @value{GDBN} when this command is invoked.
21577
21578 @var{argument} is a string. It is the argument to the command, after
21579 leading and trailing whitespace has been stripped.
21580
21581 @var{from_tty} is a boolean argument. When true, this means that the
21582 command was entered by the user at the terminal; when false it means
21583 that the command came from elsewhere.
21584
21585 If this method throws an exception, it is turned into a @value{GDBN}
21586 @code{error} call. Otherwise, the return value is ignored.
21587
21588 @findex gdb.string_to_argv
21589 To break @var{argument} up into an argv-like string use
21590 @code{gdb.string_to_argv}. This function behaves identically to
21591 @value{GDBN}'s internal argument lexer @code{buildargv}.
21592 It is recommended to use this for consistency.
21593 Arguments are separated by spaces and may be quoted.
21594 Example:
21595
21596 @smallexample
21597 print gdb.string_to_argv ("1 2\ \\\"3 '4 \"5' \"6 '7\"")
21598 ['1', '2 "3', '4 "5', "6 '7"]
21599 @end smallexample
21600
21601 @end defmethod
21602
21603 @cindex completion of Python commands
21604 @defmethod Command complete text word
21605 This method is called by @value{GDBN} when the user attempts
21606 completion on this command. All forms of completion are handled by
21607 this method, that is, the @key{TAB} and @key{M-?} key bindings
21608 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
21609 complete}).
21610
21611 The arguments @var{text} and @var{word} are both strings. @var{text}
21612 holds the complete command line up to the cursor's location.
21613 @var{word} holds the last word of the command line; this is computed
21614 using a word-breaking heuristic.
21615
21616 The @code{complete} method can return several values:
21617 @itemize @bullet
21618 @item
21619 If the return value is a sequence, the contents of the sequence are
21620 used as the completions. It is up to @code{complete} to ensure that the
21621 contents actually do complete the word. A zero-length sequence is
21622 allowed, it means that there were no completions available. Only
21623 string elements of the sequence are used; other elements in the
21624 sequence are ignored.
21625
21626 @item
21627 If the return value is one of the @samp{COMPLETE_} constants defined
21628 below, then the corresponding @value{GDBN}-internal completion
21629 function is invoked, and its result is used.
21630
21631 @item
21632 All other results are treated as though there were no available
21633 completions.
21634 @end itemize
21635 @end defmethod
21636
21637 When a new command is registered, it must be declared as a member of
21638 some general class of commands. This is used to classify top-level
21639 commands in the on-line help system; note that prefix commands are not
21640 listed under their own category but rather that of their top-level
21641 command. The available classifications are represented by constants
21642 defined in the @code{gdb} module:
21643
21644 @table @code
21645 @findex COMMAND_NONE
21646 @findex gdb.COMMAND_NONE
21647 @item COMMAND_NONE
21648 The command does not belong to any particular class. A command in
21649 this category will not be displayed in any of the help categories.
21650
21651 @findex COMMAND_RUNNING
21652 @findex gdb.COMMAND_RUNNING
21653 @item COMMAND_RUNNING
21654 The command is related to running the inferior. For example,
21655 @code{start}, @code{step}, and @code{continue} are in this category.
21656 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
21657 commands in this category.
21658
21659 @findex COMMAND_DATA
21660 @findex gdb.COMMAND_DATA
21661 @item COMMAND_DATA
21662 The command is related to data or variables. For example,
21663 @code{call}, @code{find}, and @code{print} are in this category. Type
21664 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
21665 in this category.
21666
21667 @findex COMMAND_STACK
21668 @findex gdb.COMMAND_STACK
21669 @item COMMAND_STACK
21670 The command has to do with manipulation of the stack. For example,
21671 @code{backtrace}, @code{frame}, and @code{return} are in this
21672 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
21673 list of commands in this category.
21674
21675 @findex COMMAND_FILES
21676 @findex gdb.COMMAND_FILES
21677 @item COMMAND_FILES
21678 This class is used for file-related commands. For example,
21679 @code{file}, @code{list} and @code{section} are in this category.
21680 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
21681 commands in this category.
21682
21683 @findex COMMAND_SUPPORT
21684 @findex gdb.COMMAND_SUPPORT
21685 @item COMMAND_SUPPORT
21686 This should be used for ``support facilities'', generally meaning
21687 things that are useful to the user when interacting with @value{GDBN},
21688 but not related to the state of the inferior. For example,
21689 @code{help}, @code{make}, and @code{shell} are in this category. Type
21690 @kbd{help support} at the @value{GDBN} prompt to see a list of
21691 commands in this category.
21692
21693 @findex COMMAND_STATUS
21694 @findex gdb.COMMAND_STATUS
21695 @item COMMAND_STATUS
21696 The command is an @samp{info}-related command, that is, related to the
21697 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
21698 and @code{show} are in this category. Type @kbd{help status} at the
21699 @value{GDBN} prompt to see a list of commands in this category.
21700
21701 @findex COMMAND_BREAKPOINTS
21702 @findex gdb.COMMAND_BREAKPOINTS
21703 @item COMMAND_BREAKPOINTS
21704 The command has to do with breakpoints. For example, @code{break},
21705 @code{clear}, and @code{delete} are in this category. Type @kbd{help
21706 breakpoints} at the @value{GDBN} prompt to see a list of commands in
21707 this category.
21708
21709 @findex COMMAND_TRACEPOINTS
21710 @findex gdb.COMMAND_TRACEPOINTS
21711 @item COMMAND_TRACEPOINTS
21712 The command has to do with tracepoints. For example, @code{trace},
21713 @code{actions}, and @code{tfind} are in this category. Type
21714 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
21715 commands in this category.
21716
21717 @findex COMMAND_OBSCURE
21718 @findex gdb.COMMAND_OBSCURE
21719 @item COMMAND_OBSCURE
21720 The command is only used in unusual circumstances, or is not of
21721 general interest to users. For example, @code{checkpoint},
21722 @code{fork}, and @code{stop} are in this category. Type @kbd{help
21723 obscure} at the @value{GDBN} prompt to see a list of commands in this
21724 category.
21725
21726 @findex COMMAND_MAINTENANCE
21727 @findex gdb.COMMAND_MAINTENANCE
21728 @item COMMAND_MAINTENANCE
21729 The command is only useful to @value{GDBN} maintainers. The
21730 @code{maintenance} and @code{flushregs} commands are in this category.
21731 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
21732 commands in this category.
21733 @end table
21734
21735 A new command can use a predefined completion function, either by
21736 specifying it via an argument at initialization, or by returning it
21737 from the @code{complete} method. These predefined completion
21738 constants are all defined in the @code{gdb} module:
21739
21740 @table @code
21741 @findex COMPLETE_NONE
21742 @findex gdb.COMPLETE_NONE
21743 @item COMPLETE_NONE
21744 This constant means that no completion should be done.
21745
21746 @findex COMPLETE_FILENAME
21747 @findex gdb.COMPLETE_FILENAME
21748 @item COMPLETE_FILENAME
21749 This constant means that filename completion should be performed.
21750
21751 @findex COMPLETE_LOCATION
21752 @findex gdb.COMPLETE_LOCATION
21753 @item COMPLETE_LOCATION
21754 This constant means that location completion should be done.
21755 @xref{Specify Location}.
21756
21757 @findex COMPLETE_COMMAND
21758 @findex gdb.COMPLETE_COMMAND
21759 @item COMPLETE_COMMAND
21760 This constant means that completion should examine @value{GDBN}
21761 command names.
21762
21763 @findex COMPLETE_SYMBOL
21764 @findex gdb.COMPLETE_SYMBOL
21765 @item COMPLETE_SYMBOL
21766 This constant means that completion should be done using symbol names
21767 as the source.
21768 @end table
21769
21770 The following code snippet shows how a trivial CLI command can be
21771 implemented in Python:
21772
21773 @smallexample
21774 class HelloWorld (gdb.Command):
21775 """Greet the whole world."""
21776
21777 def __init__ (self):
21778 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_OBSCURE)
21779
21780 def invoke (self, arg, from_tty):
21781 print "Hello, World!"
21782
21783 HelloWorld ()
21784 @end smallexample
21785
21786 The last line instantiates the class, and is necessary to trigger the
21787 registration of the command with @value{GDBN}. Depending on how the
21788 Python code is read into @value{GDBN}, you may need to import the
21789 @code{gdb} module explicitly.
21790
21791 @node Parameters In Python
21792 @subsubsection Parameters In Python
21793
21794 @cindex parameters in python
21795 @cindex python parameters
21796 @tindex gdb.Parameter
21797 @tindex Parameter
21798 You can implement new @value{GDBN} parameters using Python. A new
21799 parameter is implemented as an instance of the @code{gdb.Parameter}
21800 class.
21801
21802 Parameters are exposed to the user via the @code{set} and
21803 @code{show} commands. @xref{Help}.
21804
21805 There are many parameters that already exist and can be set in
21806 @value{GDBN}. Two examples are: @code{set follow fork} and
21807 @code{set charset}. Setting these parameters influences certain
21808 behavior in @value{GDBN}. Similarly, you can define parameters that
21809 can be used to influence behavior in custom Python scripts and commands.
21810
21811 @defmethod Parameter __init__ name @var{command-class} @var{parameter-class} @r{[}@var{enum-sequence}@r{]}
21812 The object initializer for @code{Parameter} registers the new
21813 parameter with @value{GDBN}. This initializer is normally invoked
21814 from the subclass' own @code{__init__} method.
21815
21816 @var{name} is the name of the new parameter. If @var{name} consists
21817 of multiple words, then the initial words are looked for as prefix
21818 parameters. An example of this can be illustrated with the
21819 @code{set print} set of parameters. If @var{name} is
21820 @code{print foo}, then @code{print} will be searched as the prefix
21821 parameter. In this case the parameter can subsequently be accessed in
21822 @value{GDBN} as @code{set print foo}.
21823
21824 If @var{name} consists of multiple words, and no prefix parameter group
21825 can be found, an exception is raised.
21826
21827 @var{command-class} should be one of the @samp{COMMAND_} constants
21828 (@pxref{Commands In Python}). This argument tells @value{GDBN} how to
21829 categorize the new parameter in the help system.
21830
21831 @var{parameter-class} should be one of the @samp{PARAM_} constants
21832 defined below. This argument tells @value{GDBN} the type of the new
21833 parameter; this information is used for input validation and
21834 completion.
21835
21836 If @var{parameter-class} is @code{PARAM_ENUM}, then
21837 @var{enum-sequence} must be a sequence of strings. These strings
21838 represent the possible values for the parameter.
21839
21840 If @var{parameter-class} is not @code{PARAM_ENUM}, then the presence
21841 of a fourth argument will cause an exception to be thrown.
21842
21843 The help text for the new parameter is taken from the Python
21844 documentation string for the parameter's class, if there is one. If
21845 there is no documentation string, a default value is used.
21846 @end defmethod
21847
21848 @defivar Parameter set_doc
21849 If this attribute exists, and is a string, then its value is used as
21850 the help text for this parameter's @code{set} command. The value is
21851 examined when @code{Parameter.__init__} is invoked; subsequent changes
21852 have no effect.
21853 @end defivar
21854
21855 @defivar Parameter show_doc
21856 If this attribute exists, and is a string, then its value is used as
21857 the help text for this parameter's @code{show} command. The value is
21858 examined when @code{Parameter.__init__} is invoked; subsequent changes
21859 have no effect.
21860 @end defivar
21861
21862 @defivar Parameter value
21863 The @code{value} attribute holds the underlying value of the
21864 parameter. It can be read and assigned to just as any other
21865 attribute. @value{GDBN} does validation when assignments are made.
21866 @end defivar
21867
21868
21869 When a new parameter is defined, its type must be specified. The
21870 available types are represented by constants defined in the @code{gdb}
21871 module:
21872
21873 @table @code
21874 @findex PARAM_BOOLEAN
21875 @findex gdb.PARAM_BOOLEAN
21876 @item PARAM_BOOLEAN
21877 The value is a plain boolean. The Python boolean values, @code{True}
21878 and @code{False} are the only valid values.
21879
21880 @findex PARAM_AUTO_BOOLEAN
21881 @findex gdb.PARAM_AUTO_BOOLEAN
21882 @item PARAM_AUTO_BOOLEAN
21883 The value has three possible states: true, false, and @samp{auto}. In
21884 Python, true and false are represented using boolean constants, and
21885 @samp{auto} is represented using @code{None}.
21886
21887 @findex PARAM_UINTEGER
21888 @findex gdb.PARAM_UINTEGER
21889 @item PARAM_UINTEGER
21890 The value is an unsigned integer. The value of 0 should be
21891 interpreted to mean ``unlimited''.
21892
21893 @findex PARAM_INTEGER
21894 @findex gdb.PARAM_INTEGER
21895 @item PARAM_INTEGER
21896 The value is a signed integer. The value of 0 should be interpreted
21897 to mean ``unlimited''.
21898
21899 @findex PARAM_STRING
21900 @findex gdb.PARAM_STRING
21901 @item PARAM_STRING
21902 The value is a string. When the user modifies the string, any escape
21903 sequences, such as @samp{\t}, @samp{\f}, and octal escapes, are
21904 translated into corresponding characters and encoded into the current
21905 host charset.
21906
21907 @findex PARAM_STRING_NOESCAPE
21908 @findex gdb.PARAM_STRING_NOESCAPE
21909 @item PARAM_STRING_NOESCAPE
21910 The value is a string. When the user modifies the string, escapes are
21911 passed through untranslated.
21912
21913 @findex PARAM_OPTIONAL_FILENAME
21914 @findex gdb.PARAM_OPTIONAL_FILENAME
21915 @item PARAM_OPTIONAL_FILENAME
21916 The value is a either a filename (a string), or @code{None}.
21917
21918 @findex PARAM_FILENAME
21919 @findex gdb.PARAM_FILENAME
21920 @item PARAM_FILENAME
21921 The value is a filename. This is just like
21922 @code{PARAM_STRING_NOESCAPE}, but uses file names for completion.
21923
21924 @findex PARAM_ZINTEGER
21925 @findex gdb.PARAM_ZINTEGER
21926 @item PARAM_ZINTEGER
21927 The value is an integer. This is like @code{PARAM_INTEGER}, except 0
21928 is interpreted as itself.
21929
21930 @findex PARAM_ENUM
21931 @findex gdb.PARAM_ENUM
21932 @item PARAM_ENUM
21933 The value is a string, which must be one of a collection string
21934 constants provided when the parameter is created.
21935 @end table
21936
21937 @node Functions In Python
21938 @subsubsection Writing new convenience functions
21939
21940 @cindex writing convenience functions
21941 @cindex convenience functions in python
21942 @cindex python convenience functions
21943 @tindex gdb.Function
21944 @tindex Function
21945 You can implement new convenience functions (@pxref{Convenience Vars})
21946 in Python. A convenience function is an instance of a subclass of the
21947 class @code{gdb.Function}.
21948
21949 @defmethod Function __init__ name
21950 The initializer for @code{Function} registers the new function with
21951 @value{GDBN}. The argument @var{name} is the name of the function,
21952 a string. The function will be visible to the user as a convenience
21953 variable of type @code{internal function}, whose name is the same as
21954 the given @var{name}.
21955
21956 The documentation for the new function is taken from the documentation
21957 string for the new class.
21958 @end defmethod
21959
21960 @defmethod Function invoke @var{*args}
21961 When a convenience function is evaluated, its arguments are converted
21962 to instances of @code{gdb.Value}, and then the function's
21963 @code{invoke} method is called. Note that @value{GDBN} does not
21964 predetermine the arity of convenience functions. Instead, all
21965 available arguments are passed to @code{invoke}, following the
21966 standard Python calling convention. In particular, a convenience
21967 function can have default values for parameters without ill effect.
21968
21969 The return value of this method is used as its value in the enclosing
21970 expression. If an ordinary Python value is returned, it is converted
21971 to a @code{gdb.Value} following the usual rules.
21972 @end defmethod
21973
21974 The following code snippet shows how a trivial convenience function can
21975 be implemented in Python:
21976
21977 @smallexample
21978 class Greet (gdb.Function):
21979 """Return string to greet someone.
21980 Takes a name as argument."""
21981
21982 def __init__ (self):
21983 super (Greet, self).__init__ ("greet")
21984
21985 def invoke (self, name):
21986 return "Hello, %s!" % name.string ()
21987
21988 Greet ()
21989 @end smallexample
21990
21991 The last line instantiates the class, and is necessary to trigger the
21992 registration of the function with @value{GDBN}. Depending on how the
21993 Python code is read into @value{GDBN}, you may need to import the
21994 @code{gdb} module explicitly.
21995
21996 @node Progspaces In Python
21997 @subsubsection Program Spaces In Python
21998
21999 @cindex progspaces in python
22000 @tindex gdb.Progspace
22001 @tindex Progspace
22002 A program space, or @dfn{progspace}, represents a symbolic view
22003 of an address space.
22004 It consists of all of the objfiles of the program.
22005 @xref{Objfiles In Python}.
22006 @xref{Inferiors and Programs, program spaces}, for more details
22007 about program spaces.
22008
22009 The following progspace-related functions are available in the
22010 @code{gdb} module:
22011
22012 @findex gdb.current_progspace
22013 @defun current_progspace
22014 This function returns the program space of the currently selected inferior.
22015 @xref{Inferiors and Programs}.
22016 @end defun
22017
22018 @findex gdb.progspaces
22019 @defun progspaces
22020 Return a sequence of all the progspaces currently known to @value{GDBN}.
22021 @end defun
22022
22023 Each progspace is represented by an instance of the @code{gdb.Progspace}
22024 class.
22025
22026 @defivar Progspace filename
22027 The file name of the progspace as a string.
22028 @end defivar
22029
22030 @defivar Progspace pretty_printers
22031 The @code{pretty_printers} attribute is a list of functions. It is
22032 used to look up pretty-printers. A @code{Value} is passed to each
22033 function in order; if the function returns @code{None}, then the
22034 search continues. Otherwise, the return value should be an object
22035 which is used to format the value. @xref{Pretty Printing API}, for more
22036 information.
22037 @end defivar
22038
22039 @node Objfiles In Python
22040 @subsubsection Objfiles In Python
22041
22042 @cindex objfiles in python
22043 @tindex gdb.Objfile
22044 @tindex Objfile
22045 @value{GDBN} loads symbols for an inferior from various
22046 symbol-containing files (@pxref{Files}). These include the primary
22047 executable file, any shared libraries used by the inferior, and any
22048 separate debug info files (@pxref{Separate Debug Files}).
22049 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
22050
22051 The following objfile-related functions are available in the
22052 @code{gdb} module:
22053
22054 @findex gdb.current_objfile
22055 @defun current_objfile
22056 When auto-loading a Python script (@pxref{Auto-loading}), @value{GDBN}
22057 sets the ``current objfile'' to the corresponding objfile. This
22058 function returns the current objfile. If there is no current objfile,
22059 this function returns @code{None}.
22060 @end defun
22061
22062 @findex gdb.objfiles
22063 @defun objfiles
22064 Return a sequence of all the objfiles current known to @value{GDBN}.
22065 @xref{Objfiles In Python}.
22066 @end defun
22067
22068 Each objfile is represented by an instance of the @code{gdb.Objfile}
22069 class.
22070
22071 @defivar Objfile filename
22072 The file name of the objfile as a string.
22073 @end defivar
22074
22075 @defivar Objfile pretty_printers
22076 The @code{pretty_printers} attribute is a list of functions. It is
22077 used to look up pretty-printers. A @code{Value} is passed to each
22078 function in order; if the function returns @code{None}, then the
22079 search continues. Otherwise, the return value should be an object
22080 which is used to format the value. @xref{Pretty Printing API}, for more
22081 information.
22082 @end defivar
22083
22084 @node Frames In Python
22085 @subsubsection Accessing inferior stack frames from Python.
22086
22087 @cindex frames in python
22088 When the debugged program stops, @value{GDBN} is able to analyze its call
22089 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
22090 represents a frame in the stack. A @code{gdb.Frame} object is only valid
22091 while its corresponding frame exists in the inferior's stack. If you try
22092 to use an invalid frame object, @value{GDBN} will throw a @code{RuntimeError}
22093 exception.
22094
22095 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
22096 operator, like:
22097
22098 @smallexample
22099 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
22100 True
22101 @end smallexample
22102
22103 The following frame-related functions are available in the @code{gdb} module:
22104
22105 @findex gdb.selected_frame
22106 @defun selected_frame
22107 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
22108 @end defun
22109
22110 @defun frame_stop_reason_string reason
22111 Return a string explaining the reason why @value{GDBN} stopped unwinding
22112 frames, as expressed by the given @var{reason} code (an integer, see the
22113 @code{unwind_stop_reason} method further down in this section).
22114 @end defun
22115
22116 A @code{gdb.Frame} object has the following methods:
22117
22118 @table @code
22119 @defmethod Frame is_valid
22120 Returns true if the @code{gdb.Frame} object is valid, false if not.
22121 A frame object can become invalid if the frame it refers to doesn't
22122 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
22123 an exception if it is invalid at the time the method is called.
22124 @end defmethod
22125
22126 @defmethod Frame name
22127 Returns the function name of the frame, or @code{None} if it can't be
22128 obtained.
22129 @end defmethod
22130
22131 @defmethod Frame type
22132 Returns the type of the frame. The value can be one of
22133 @code{gdb.NORMAL_FRAME}, @code{gdb.DUMMY_FRAME}, @code{gdb.SIGTRAMP_FRAME}
22134 or @code{gdb.SENTINEL_FRAME}.
22135 @end defmethod
22136
22137 @defmethod Frame unwind_stop_reason
22138 Return an integer representing the reason why it's not possible to find
22139 more frames toward the outermost frame. Use
22140 @code{gdb.frame_stop_reason_string} to convert the value returned by this
22141 function to a string.
22142 @end defmethod
22143
22144 @defmethod Frame pc
22145 Returns the frame's resume address.
22146 @end defmethod
22147
22148 @defmethod Frame block
22149 Return the frame's code block. @xref{Blocks In Python}.
22150 @end defmethod
22151
22152 @defmethod Frame function
22153 Return the symbol for the function corresponding to this frame.
22154 @xref{Symbols In Python}.
22155 @end defmethod
22156
22157 @defmethod Frame older
22158 Return the frame that called this frame.
22159 @end defmethod
22160
22161 @defmethod Frame newer
22162 Return the frame called by this frame.
22163 @end defmethod
22164
22165 @defmethod Frame find_sal
22166 Return the frame's symtab and line object.
22167 @xref{Symbol Tables In Python}.
22168 @end defmethod
22169
22170 @defmethod Frame read_var variable @r{[}block@r{]}
22171 Return the value of @var{variable} in this frame. If the optional
22172 argument @var{block} is provided, search for the variable from that
22173 block; otherwise start at the frame's current block (which is
22174 determined by the frame's current program counter). @var{variable}
22175 must be a string or a @code{gdb.Symbol} object. @var{block} must be a
22176 @code{gdb.Block} object.
22177 @end defmethod
22178
22179 @defmethod Frame select
22180 Set this frame to be the selected frame. @xref{Stack, ,Examining the
22181 Stack}.
22182 @end defmethod
22183 @end table
22184
22185 @node Blocks In Python
22186 @subsubsection Accessing frame blocks from Python.
22187
22188 @cindex blocks in python
22189 @tindex gdb.Block
22190
22191 Within each frame, @value{GDBN} maintains information on each block
22192 stored in that frame. These blocks are organized hierarchically, and
22193 are represented individually in Python as a @code{gdb.Block}.
22194 Please see @ref{Frames In Python}, for a more in-depth discussion on
22195 frames. Furthermore, see @ref{Stack, ,Examining the Stack}, for more
22196 detailed technical information on @value{GDBN}'s book-keeping of the
22197 stack.
22198
22199 The following block-related functions are available in the @code{gdb}
22200 module:
22201
22202 @findex gdb.block_for_pc
22203 @defun block_for_pc pc
22204 Return the @code{gdb.Block} containing the given @var{pc} value. If the
22205 block cannot be found for the @var{pc} value specified, the function
22206 will return @code{None}.
22207 @end defun
22208
22209 A @code{gdb.Block} object has the following attributes:
22210
22211 @table @code
22212 @defivar Block start
22213 The start address of the block. This attribute is not writable.
22214 @end defivar
22215
22216 @defivar Block end
22217 The end address of the block. This attribute is not writable.
22218 @end defivar
22219
22220 @defivar Block function
22221 The name of the block represented as a @code{gdb.Symbol}. If the
22222 block is not named, then this attribute holds @code{None}. This
22223 attribute is not writable.
22224 @end defivar
22225
22226 @defivar Block superblock
22227 The block containing this block. If this parent block does not exist,
22228 this attribute holds @code{None}. This attribute is not writable.
22229 @end defivar
22230 @end table
22231
22232 @node Symbols In Python
22233 @subsubsection Python representation of Symbols.
22234
22235 @cindex symbols in python
22236 @tindex gdb.Symbol
22237
22238 @value{GDBN} represents every variable, function and type as an
22239 entry in a symbol table. @xref{Symbols, ,Examining the Symbol Table}.
22240 Similarly, Python represents these symbols in @value{GDBN} with the
22241 @code{gdb.Symbol} object.
22242
22243 The following symbol-related functions are available in the @code{gdb}
22244 module:
22245
22246 @findex gdb.lookup_symbol
22247 @defun lookup_symbol name [block] [domain]
22248 This function searches for a symbol by name. The search scope can be
22249 restricted to the parameters defined in the optional domain and block
22250 arguments.
22251
22252 @var{name} is the name of the symbol. It must be a string. The
22253 optional @var{block} argument restricts the search to symbols visible
22254 in that @var{block}. The @var{block} argument must be a
22255 @code{gdb.Block} object. The optional @var{domain} argument restricts
22256 the search to the domain type. The @var{domain} argument must be a
22257 domain constant defined in the @code{gdb} module and described later
22258 in this chapter.
22259 @end defun
22260
22261 A @code{gdb.Symbol} object has the following attributes:
22262
22263 @table @code
22264 @defivar Symbol symtab
22265 The symbol table in which the symbol appears. This attribute is
22266 represented as a @code{gdb.Symtab} object. @xref{Symbol Tables In
22267 Python}. This attribute is not writable.
22268 @end defivar
22269
22270 @defivar Symbol name
22271 The name of the symbol as a string. This attribute is not writable.
22272 @end defivar
22273
22274 @defivar Symbol linkage_name
22275 The name of the symbol, as used by the linker (i.e., may be mangled).
22276 This attribute is not writable.
22277 @end defivar
22278
22279 @defivar Symbol print_name
22280 The name of the symbol in a form suitable for output. This is either
22281 @code{name} or @code{linkage_name}, depending on whether the user
22282 asked @value{GDBN} to display demangled or mangled names.
22283 @end defivar
22284
22285 @defivar Symbol addr_class
22286 The address class of the symbol. This classifies how to find the value
22287 of a symbol. Each address class is a constant defined in the
22288 @code{gdb} module and described later in this chapter.
22289 @end defivar
22290
22291 @defivar Symbol is_argument
22292 @code{True} if the symbol is an argument of a function.
22293 @end defivar
22294
22295 @defivar Symbol is_constant
22296 @code{True} if the symbol is a constant.
22297 @end defivar
22298
22299 @defivar Symbol is_function
22300 @code{True} if the symbol is a function or a method.
22301 @end defivar
22302
22303 @defivar Symbol is_variable
22304 @code{True} if the symbol is a variable.
22305 @end defivar
22306 @end table
22307
22308 The available domain categories in @code{gdb.Symbol} are represented
22309 as constants in the @code{gdb} module:
22310
22311 @table @code
22312 @findex SYMBOL_UNDEF_DOMAIN
22313 @findex gdb.SYMBOL_UNDEF_DOMAIN
22314 @item SYMBOL_UNDEF_DOMAIN
22315 This is used when a domain has not been discovered or none of the
22316 following domains apply. This usually indicates an error either
22317 in the symbol information or in @value{GDBN}'s handling of symbols.
22318 @findex SYMBOL_VAR_DOMAIN
22319 @findex gdb.SYMBOL_VAR_DOMAIN
22320 @item SYMBOL_VAR_DOMAIN
22321 This domain contains variables, function names, typedef names and enum
22322 type values.
22323 @findex SYMBOL_STRUCT_DOMAIN
22324 @findex gdb.SYMBOL_STRUCT_DOMAIN
22325 @item SYMBOL_STRUCT_DOMAIN
22326 This domain holds struct, union and enum type names.
22327 @findex SYMBOL_LABEL_DOMAIN
22328 @findex gdb.SYMBOL_LABEL_DOMAIN
22329 @item SYMBOL_LABEL_DOMAIN
22330 This domain contains names of labels (for gotos).
22331 @findex SYMBOL_VARIABLES_DOMAIN
22332 @findex gdb.SYMBOL_VARIABLES_DOMAIN
22333 @item SYMBOL_VARIABLES_DOMAIN
22334 This domain holds a subset of the @code{SYMBOLS_VAR_DOMAIN}; it
22335 contains everything minus functions and types.
22336 @findex SYMBOL_FUNCTIONS_DOMAIN
22337 @findex gdb.SYMBOL_FUNCTIONS_DOMAIN
22338 @item SYMBOL_FUNCTION_DOMAIN
22339 This domain contains all functions.
22340 @findex SYMBOL_TYPES_DOMAIN
22341 @findex gdb.SYMBOL_TYPES_DOMAIN
22342 @item SYMBOL_TYPES_DOMAIN
22343 This domain contains all types.
22344 @end table
22345
22346 The available address class categories in @code{gdb.Symbol} are represented
22347 as constants in the @code{gdb} module:
22348
22349 @table @code
22350 @findex SYMBOL_LOC_UNDEF
22351 @findex gdb.SYMBOL_LOC_UNDEF
22352 @item SYMBOL_LOC_UNDEF
22353 If this is returned by address class, it indicates an error either in
22354 the symbol information or in @value{GDBN}'s handling of symbols.
22355 @findex SYMBOL_LOC_CONST
22356 @findex gdb.SYMBOL_LOC_CONST
22357 @item SYMBOL_LOC_CONST
22358 Value is constant int.
22359 @findex SYMBOL_LOC_STATIC
22360 @findex gdb.SYMBOL_LOC_STATIC
22361 @item SYMBOL_LOC_STATIC
22362 Value is at a fixed address.
22363 @findex SYMBOL_LOC_REGISTER
22364 @findex gdb.SYMBOL_LOC_REGISTER
22365 @item SYMBOL_LOC_REGISTER
22366 Value is in a register.
22367 @findex SYMBOL_LOC_ARG
22368 @findex gdb.SYMBOL_LOC_ARG
22369 @item SYMBOL_LOC_ARG
22370 Value is an argument. This value is at the offset stored within the
22371 symbol inside the frame's argument list.
22372 @findex SYMBOL_LOC_REF_ARG
22373 @findex gdb.SYMBOL_LOC_REF_ARG
22374 @item SYMBOL_LOC_REF_ARG
22375 Value address is stored in the frame's argument list. Just like
22376 @code{LOC_ARG} except that the value's address is stored at the
22377 offset, not the value itself.
22378 @findex SYMBOL_LOC_REGPARM_ADDR
22379 @findex gdb.SYMBOL_LOC_REGPARM_ADDR
22380 @item SYMBOL_LOC_REGPARM_ADDR
22381 Value is a specified register. Just like @code{LOC_REGISTER} except
22382 the register holds the address of the argument instead of the argument
22383 itself.
22384 @findex SYMBOL_LOC_LOCAL
22385 @findex gdb.SYMBOL_LOC_LOCAL
22386 @item SYMBOL_LOC_LOCAL
22387 Value is a local variable.
22388 @findex SYMBOL_LOC_TYPEDEF
22389 @findex gdb.SYMBOL_LOC_TYPEDEF
22390 @item SYMBOL_LOC_TYPEDEF
22391 Value not used. Symbols in the domain @code{SYMBOL_STRUCT_DOMAIN} all
22392 have this class.
22393 @findex SYMBOL_LOC_BLOCK
22394 @findex gdb.SYMBOL_LOC_BLOCK
22395 @item SYMBOL_LOC_BLOCK
22396 Value is a block.
22397 @findex SYMBOL_LOC_CONST_BYTES
22398 @findex gdb.SYMBOL_LOC_CONST_BYTES
22399 @item SYMBOL_LOC_CONST_BYTES
22400 Value is a byte-sequence.
22401 @findex SYMBOL_LOC_UNRESOLVED
22402 @findex gdb.SYMBOL_LOC_UNRESOLVED
22403 @item SYMBOL_LOC_UNRESOLVED
22404 Value is at a fixed address, but the address of the variable has to be
22405 determined from the minimal symbol table whenever the variable is
22406 referenced.
22407 @findex SYMBOL_LOC_OPTIMIZED_OUT
22408 @findex gdb.SYMBOL_LOC_OPTIMIZED_OUT
22409 @item SYMBOL_LOC_OPTIMIZED_OUT
22410 The value does not actually exist in the program.
22411 @findex SYMBOL_LOC_COMPUTED
22412 @findex gdb.SYMBOL_LOC_COMPUTED
22413 @item SYMBOL_LOC_COMPUTED
22414 The value's address is a computed location.
22415 @end table
22416
22417 @node Symbol Tables In Python
22418 @subsubsection Symbol table representation in Python.
22419
22420 @cindex symbol tables in python
22421 @tindex gdb.Symtab
22422 @tindex gdb.Symtab_and_line
22423
22424 Access to symbol table data maintained by @value{GDBN} on the inferior
22425 is exposed to Python via two objects: @code{gdb.Symtab_and_line} and
22426 @code{gdb.Symtab}. Symbol table and line data for a frame is returned
22427 from the @code{find_sal} method in @code{gdb.Frame} object.
22428 @xref{Frames In Python}.
22429
22430 For more information on @value{GDBN}'s symbol table management, see
22431 @ref{Symbols, ,Examining the Symbol Table}, for more information.
22432
22433 A @code{gdb.Symtab_and_line} object has the following attributes:
22434
22435 @table @code
22436 @defivar Symtab_and_line symtab
22437 The symbol table object (@code{gdb.Symtab}) for this frame.
22438 This attribute is not writable.
22439 @end defivar
22440
22441 @defivar Symtab_and_line pc
22442 Indicates the current program counter address. This attribute is not
22443 writable.
22444 @end defivar
22445
22446 @defivar Symtab_and_line line
22447 Indicates the current line number for this object. This
22448 attribute is not writable.
22449 @end defivar
22450 @end table
22451
22452 A @code{gdb.Symtab} object has the following attributes:
22453
22454 @table @code
22455 @defivar Symtab filename
22456 The symbol table's source filename. This attribute is not writable.
22457 @end defivar
22458
22459 @defivar Symtab objfile
22460 The symbol table's backing object file. @xref{Objfiles In Python}.
22461 This attribute is not writable.
22462 @end defivar
22463 @end table
22464
22465 The following methods are provided:
22466
22467 @table @code
22468 @defmethod Symtab fullname
22469 Return the symbol table's source absolute file name.
22470 @end defmethod
22471 @end table
22472
22473 @node Breakpoints In Python
22474 @subsubsection Manipulating breakpoints using Python
22475
22476 @cindex breakpoints in python
22477 @tindex gdb.Breakpoint
22478
22479 Python code can manipulate breakpoints via the @code{gdb.Breakpoint}
22480 class.
22481
22482 @defmethod Breakpoint __init__ spec @r{[}type@r{]} @r{[}wp_class@r{]}
22483 Create a new breakpoint. @var{spec} is a string naming the
22484 location of the breakpoint, or an expression that defines a
22485 watchpoint. The contents can be any location recognized by the
22486 @code{break} command, or in the case of a watchpoint, by the @code{watch}
22487 command. The optional @var{type} denotes the breakpoint to create
22488 from the types defined later in this chapter. This argument can be
22489 either: @code{BP_BREAKPOINT} or @code{BP_WATCHPOINT}. @var{type}
22490 defaults to @code{BP_BREAKPOINT}. The optional @var{wp_class}
22491 argument defines the class of watchpoint to create, if @var{type} is
22492 defined as @code{BP_WATCHPOINT}. If a watchpoint class is not
22493 provided, it is assumed to be a @var{WP_WRITE} class.
22494 @end defmethod
22495
22496 The available watchpoint types represented by constants are defined in the
22497 @code{gdb} module:
22498
22499 @table @code
22500 @findex WP_READ
22501 @findex gdb.WP_READ
22502 @item WP_READ
22503 Read only watchpoint.
22504
22505 @findex WP_WRITE
22506 @findex gdb.WP_WRITE
22507 @item WP_WRITE
22508 Write only watchpoint.
22509
22510 @findex WP_ACCESS
22511 @findex gdb.WP_ACCESS
22512 @item WP_ACCESS
22513 Read/Write watchpoint.
22514 @end table
22515
22516 @defmethod Breakpoint is_valid
22517 Return @code{True} if this @code{Breakpoint} object is valid,
22518 @code{False} otherwise. A @code{Breakpoint} object can become invalid
22519 if the user deletes the breakpoint. In this case, the object still
22520 exists, but the underlying breakpoint does not. In the cases of
22521 watchpoint scope, the watchpoint remains valid even if execution of the
22522 inferior leaves the scope of that watchpoint.
22523 @end defmethod
22524
22525 @defivar Breakpoint enabled
22526 This attribute is @code{True} if the breakpoint is enabled, and
22527 @code{False} otherwise. This attribute is writable.
22528 @end defivar
22529
22530 @defivar Breakpoint silent
22531 This attribute is @code{True} if the breakpoint is silent, and
22532 @code{False} otherwise. This attribute is writable.
22533
22534 Note that a breakpoint can also be silent if it has commands and the
22535 first command is @code{silent}. This is not reported by the
22536 @code{silent} attribute.
22537 @end defivar
22538
22539 @defivar Breakpoint thread
22540 If the breakpoint is thread-specific, this attribute holds the thread
22541 id. If the breakpoint is not thread-specific, this attribute is
22542 @code{None}. This attribute is writable.
22543 @end defivar
22544
22545 @defivar Breakpoint task
22546 If the breakpoint is Ada task-specific, this attribute holds the Ada task
22547 id. If the breakpoint is not task-specific (or the underlying
22548 language is not Ada), this attribute is @code{None}. This attribute
22549 is writable.
22550 @end defivar
22551
22552 @defivar Breakpoint ignore_count
22553 This attribute holds the ignore count for the breakpoint, an integer.
22554 This attribute is writable.
22555 @end defivar
22556
22557 @defivar Breakpoint number
22558 This attribute holds the breakpoint's number --- the identifier used by
22559 the user to manipulate the breakpoint. This attribute is not writable.
22560 @end defivar
22561
22562 @defivar Breakpoint type
22563 This attribute holds the breakpoint's type --- the identifier used to
22564 determine the actual breakpoint type or use-case. This attribute is not
22565 writable.
22566 @end defivar
22567
22568 The available types are represented by constants defined in the @code{gdb}
22569 module:
22570
22571 @table @code
22572 @findex BP_BREAKPOINT
22573 @findex gdb.BP_BREAKPOINT
22574 @item BP_BREAKPOINT
22575 Normal code breakpoint.
22576
22577 @findex BP_WATCHPOINT
22578 @findex gdb.BP_WATCHPOINT
22579 @item BP_WATCHPOINT
22580 Watchpoint breakpoint.
22581
22582 @findex BP_HARDWARE_WATCHPOINT
22583 @findex gdb.BP_HARDWARE_WATCHPOINT
22584 @item BP_HARDWARE_WATCHPOINT
22585 Hardware assisted watchpoint.
22586
22587 @findex BP_READ_WATCHPOINT
22588 @findex gdb.BP_READ_WATCHPOINT
22589 @item BP_READ_WATCHPOINT
22590 Hardware assisted read watchpoint.
22591
22592 @findex BP_ACCESS_WATCHPOINT
22593 @findex gdb.BP_ACCESS_WATCHPOINT
22594 @item BP_ACCESS_WATCHPOINT
22595 Hardware assisted access watchpoint.
22596 @end table
22597
22598 @defivar Breakpoint hit_count
22599 This attribute holds the hit count for the breakpoint, an integer.
22600 This attribute is writable, but currently it can only be set to zero.
22601 @end defivar
22602
22603 @defivar Breakpoint location
22604 This attribute holds the location of the breakpoint, as specified by
22605 the user. It is a string. If the breakpoint does not have a location
22606 (that is, it is a watchpoint) the attribute's value is @code{None}. This
22607 attribute is not writable.
22608 @end defivar
22609
22610 @defivar Breakpoint expression
22611 This attribute holds a breakpoint expression, as specified by
22612 the user. It is a string. If the breakpoint does not have an
22613 expression (the breakpoint is not a watchpoint) the attribute's value
22614 is @code{None}. This attribute is not writable.
22615 @end defivar
22616
22617 @defivar Breakpoint condition
22618 This attribute holds the condition of the breakpoint, as specified by
22619 the user. It is a string. If there is no condition, this attribute's
22620 value is @code{None}. This attribute is writable.
22621 @end defivar
22622
22623 @defivar Breakpoint commands
22624 This attribute holds the commands attached to the breakpoint. If
22625 there are commands, this attribute's value is a string holding all the
22626 commands, separated by newlines. If there are no commands, this
22627 attribute is @code{None}. This attribute is not writable.
22628 @end defivar
22629
22630 @node Lazy Strings In Python
22631 @subsubsection Python representation of lazy strings.
22632
22633 @cindex lazy strings in python
22634 @tindex gdb.LazyString
22635
22636 A @dfn{lazy string} is a string whose contents is not retrieved or
22637 encoded until it is needed.
22638
22639 A @code{gdb.LazyString} is represented in @value{GDBN} as an
22640 @code{address} that points to a region of memory, an @code{encoding}
22641 that will be used to encode that region of memory, and a @code{length}
22642 to delimit the region of memory that represents the string. The
22643 difference between a @code{gdb.LazyString} and a string wrapped within
22644 a @code{gdb.Value} is that a @code{gdb.LazyString} will be treated
22645 differently by @value{GDBN} when printing. A @code{gdb.LazyString} is
22646 retrieved and encoded during printing, while a @code{gdb.Value}
22647 wrapping a string is immediately retrieved and encoded on creation.
22648
22649 A @code{gdb.LazyString} object has the following functions:
22650
22651 @defmethod LazyString value
22652 Convert the @code{gdb.LazyString} to a @code{gdb.Value}. This value
22653 will point to the string in memory, but will lose all the delayed
22654 retrieval, encoding and handling that @value{GDBN} applies to a
22655 @code{gdb.LazyString}.
22656 @end defmethod
22657
22658 @defivar LazyString address
22659 This attribute holds the address of the string. This attribute is not
22660 writable.
22661 @end defivar
22662
22663 @defivar LazyString length
22664 This attribute holds the length of the string in characters. If the
22665 length is -1, then the string will be fetched and encoded up to the
22666 first null of appropriate width. This attribute is not writable.
22667 @end defivar
22668
22669 @defivar LazyString encoding
22670 This attribute holds the encoding that will be applied to the string
22671 when the string is printed by @value{GDBN}. If the encoding is not
22672 set, or contains an empty string, then @value{GDBN} will select the
22673 most appropriate encoding when the string is printed. This attribute
22674 is not writable.
22675 @end defivar
22676
22677 @defivar LazyString type
22678 This attribute holds the type that is represented by the lazy string's
22679 type. For a lazy string this will always be a pointer type. To
22680 resolve this to the lazy string's character type, use the type's
22681 @code{target} method. @xref{Types In Python}. This attribute is not
22682 writable.
22683 @end defivar
22684
22685 @node Auto-loading
22686 @subsection Auto-loading
22687 @cindex auto-loading, Python
22688
22689 When a new object file is read (for example, due to the @code{file}
22690 command, or because the inferior has loaded a shared library),
22691 @value{GDBN} will look for Python support scripts in several ways:
22692 @file{@var{objfile}-gdb.py} and @code{.debug_gdb_scripts} section.
22693
22694 @menu
22695 * objfile-gdb.py file:: The @file{@var{objfile}-gdb.py} file
22696 * .debug_gdb_scripts section:: The @code{.debug_gdb_scripts} section
22697 * Which flavor to choose?::
22698 @end menu
22699
22700 The auto-loading feature is useful for supplying application-specific
22701 debugging commands and scripts.
22702
22703 Auto-loading can be enabled or disabled.
22704
22705 @table @code
22706 @kindex maint set python auto-load
22707 @item maint set python auto-load [yes|no]
22708 Enable or disable the Python auto-loading feature.
22709
22710 @kindex maint show python auto-load
22711 @item maint show python auto-load
22712 Show whether Python auto-loading is enabled or disabled.
22713 @end table
22714
22715 When reading an auto-loaded file, @value{GDBN} sets the
22716 @dfn{current objfile}. This is available via the @code{gdb.current_objfile}
22717 function (@pxref{Objfiles In Python}). This can be useful for
22718 registering objfile-specific pretty-printers.
22719
22720 @node objfile-gdb.py file
22721 @subsubsection The @file{@var{objfile}-gdb.py} file
22722 @cindex @file{@var{objfile}-gdb.py}
22723
22724 When a new object file is read, @value{GDBN} looks for
22725 a file named @file{@var{objfile}-gdb.py},
22726 where @var{objfile} is the object file's real name, formed by ensuring
22727 that the file name is absolute, following all symlinks, and resolving
22728 @code{.} and @code{..} components. If this file exists and is
22729 readable, @value{GDBN} will evaluate it as a Python script.
22730
22731 If this file does not exist, and if the parameter
22732 @code{debug-file-directory} is set (@pxref{Separate Debug Files}),
22733 then @value{GDBN} will look for @var{real-name} in all of the
22734 directories mentioned in the value of @code{debug-file-directory}.
22735
22736 Finally, if this file does not exist, then @value{GDBN} will look for
22737 a file named @file{@var{data-directory}/python/auto-load/@var{real-name}}, where
22738 @var{data-directory} is @value{GDBN}'s data directory (available via
22739 @code{show data-directory}, @pxref{Data Files}), and @var{real-name}
22740 is the object file's real name, as described above.
22741
22742 @value{GDBN} does not track which files it has already auto-loaded this way.
22743 @value{GDBN} will load the associated script every time the corresponding
22744 @var{objfile} is opened.
22745 So your @file{-gdb.py} file should be careful to avoid errors if it
22746 is evaluated more than once.
22747
22748 @node .debug_gdb_scripts section
22749 @subsubsection The @code{.debug_gdb_scripts} section
22750 @cindex @code{.debug_gdb_scripts} section
22751
22752 For systems using file formats like ELF and COFF,
22753 when @value{GDBN} loads a new object file
22754 it will look for a special section named @samp{.debug_gdb_scripts}.
22755 If this section exists, its contents is a list of names of scripts to load.
22756
22757 @value{GDBN} will look for each specified script file first in the
22758 current directory and then along the source search path
22759 (@pxref{Source Path, ,Specifying Source Directories}),
22760 except that @file{$cdir} is not searched, since the compilation
22761 directory is not relevant to scripts.
22762
22763 Entries can be placed in section @code{.debug_gdb_scripts} with,
22764 for example, this GCC macro:
22765
22766 @example
22767 /* Note: The "MS" section flags are to remove duplicates. */
22768 #define DEFINE_GDB_SCRIPT(script_name) \
22769 asm("\
22770 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
22771 .byte 1\n\
22772 .asciz \"" script_name "\"\n\
22773 .popsection \n\
22774 ");
22775 @end example
22776
22777 @noindent
22778 Then one can reference the macro in a header or source file like this:
22779
22780 @example
22781 DEFINE_GDB_SCRIPT ("my-app-scripts.py")
22782 @end example
22783
22784 The script name may include directories if desired.
22785
22786 If the macro is put in a header, any application or library
22787 using this header will get a reference to the specified script.
22788
22789 @node Which flavor to choose?
22790 @subsubsection Which flavor to choose?
22791
22792 Given the multiple ways of auto-loading Python scripts, it might not always
22793 be clear which one to choose. This section provides some guidance.
22794
22795 Benefits of the @file{-gdb.py} way:
22796
22797 @itemize @bullet
22798 @item
22799 Can be used with file formats that don't support multiple sections.
22800
22801 @item
22802 Ease of finding scripts for public libraries.
22803
22804 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
22805 in the source search path.
22806 For publicly installed libraries, e.g., @file{libstdc++}, there typically
22807 isn't a source directory in which to find the script.
22808
22809 @item
22810 Doesn't require source code additions.
22811 @end itemize
22812
22813 Benefits of the @code{.debug_gdb_scripts} way:
22814
22815 @itemize @bullet
22816 @item
22817 Works with static linking.
22818
22819 Scripts for libraries done the @file{-gdb.py} way require an objfile to
22820 trigger their loading. When an application is statically linked the only
22821 objfile available is the executable, and it is cumbersome to attach all the
22822 scripts from all the input libraries to the executable's @file{-gdb.py} script.
22823
22824 @item
22825 Works with classes that are entirely inlined.
22826
22827 Some classes can be entirely inlined, and thus there may not be an associated
22828 shared library to attach a @file{-gdb.py} script to.
22829
22830 @item
22831 Scripts needn't be copied out of the source tree.
22832
22833 In some circumstances, apps can be built out of large collections of internal
22834 libraries, and the build infrastructure necessary to install the
22835 @file{-gdb.py} scripts in a place where @value{GDBN} can find them is
22836 cumbersome. It may be easier to specify the scripts in the
22837 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
22838 top of the source tree to the source search path.
22839 @end itemize
22840
22841 @node Interpreters
22842 @chapter Command Interpreters
22843 @cindex command interpreters
22844
22845 @value{GDBN} supports multiple command interpreters, and some command
22846 infrastructure to allow users or user interface writers to switch
22847 between interpreters or run commands in other interpreters.
22848
22849 @value{GDBN} currently supports two command interpreters, the console
22850 interpreter (sometimes called the command-line interpreter or @sc{cli})
22851 and the machine interface interpreter (or @sc{gdb/mi}). This manual
22852 describes both of these interfaces in great detail.
22853
22854 By default, @value{GDBN} will start with the console interpreter.
22855 However, the user may choose to start @value{GDBN} with another
22856 interpreter by specifying the @option{-i} or @option{--interpreter}
22857 startup options. Defined interpreters include:
22858
22859 @table @code
22860 @item console
22861 @cindex console interpreter
22862 The traditional console or command-line interpreter. This is the most often
22863 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
22864 @value{GDBN} will use this interpreter.
22865
22866 @item mi
22867 @cindex mi interpreter
22868 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
22869 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
22870 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
22871 Interface}.
22872
22873 @item mi2
22874 @cindex mi2 interpreter
22875 The current @sc{gdb/mi} interface.
22876
22877 @item mi1
22878 @cindex mi1 interpreter
22879 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
22880
22881 @end table
22882
22883 @cindex invoke another interpreter
22884 The interpreter being used by @value{GDBN} may not be dynamically
22885 switched at runtime. Although possible, this could lead to a very
22886 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
22887 enters the command "interpreter-set console" in a console view,
22888 @value{GDBN} would switch to using the console interpreter, rendering
22889 the IDE inoperable!
22890
22891 @kindex interpreter-exec
22892 Although you may only choose a single interpreter at startup, you may execute
22893 commands in any interpreter from the current interpreter using the appropriate
22894 command. If you are running the console interpreter, simply use the
22895 @code{interpreter-exec} command:
22896
22897 @smallexample
22898 interpreter-exec mi "-data-list-register-names"
22899 @end smallexample
22900
22901 @sc{gdb/mi} has a similar command, although it is only available in versions of
22902 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
22903
22904 @node TUI
22905 @chapter @value{GDBN} Text User Interface
22906 @cindex TUI
22907 @cindex Text User Interface
22908
22909 @menu
22910 * TUI Overview:: TUI overview
22911 * TUI Keys:: TUI key bindings
22912 * TUI Single Key Mode:: TUI single key mode
22913 * TUI Commands:: TUI-specific commands
22914 * TUI Configuration:: TUI configuration variables
22915 @end menu
22916
22917 The @value{GDBN} Text User Interface (TUI) is a terminal
22918 interface which uses the @code{curses} library to show the source
22919 file, the assembly output, the program registers and @value{GDBN}
22920 commands in separate text windows. The TUI mode is supported only
22921 on platforms where a suitable version of the @code{curses} library
22922 is available.
22923
22924 @pindex @value{GDBTUI}
22925 The TUI mode is enabled by default when you invoke @value{GDBN} as
22926 either @samp{@value{GDBTUI}} or @samp{@value{GDBP} -tui}.
22927 You can also switch in and out of TUI mode while @value{GDBN} runs by
22928 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
22929 @xref{TUI Keys, ,TUI Key Bindings}.
22930
22931 @node TUI Overview
22932 @section TUI Overview
22933
22934 In TUI mode, @value{GDBN} can display several text windows:
22935
22936 @table @emph
22937 @item command
22938 This window is the @value{GDBN} command window with the @value{GDBN}
22939 prompt and the @value{GDBN} output. The @value{GDBN} input is still
22940 managed using readline.
22941
22942 @item source
22943 The source window shows the source file of the program. The current
22944 line and active breakpoints are displayed in this window.
22945
22946 @item assembly
22947 The assembly window shows the disassembly output of the program.
22948
22949 @item register
22950 This window shows the processor registers. Registers are highlighted
22951 when their values change.
22952 @end table
22953
22954 The source and assembly windows show the current program position
22955 by highlighting the current line and marking it with a @samp{>} marker.
22956 Breakpoints are indicated with two markers. The first marker
22957 indicates the breakpoint type:
22958
22959 @table @code
22960 @item B
22961 Breakpoint which was hit at least once.
22962
22963 @item b
22964 Breakpoint which was never hit.
22965
22966 @item H
22967 Hardware breakpoint which was hit at least once.
22968
22969 @item h
22970 Hardware breakpoint which was never hit.
22971 @end table
22972
22973 The second marker indicates whether the breakpoint is enabled or not:
22974
22975 @table @code
22976 @item +
22977 Breakpoint is enabled.
22978
22979 @item -
22980 Breakpoint is disabled.
22981 @end table
22982
22983 The source, assembly and register windows are updated when the current
22984 thread changes, when the frame changes, or when the program counter
22985 changes.
22986
22987 These windows are not all visible at the same time. The command
22988 window is always visible. The others can be arranged in several
22989 layouts:
22990
22991 @itemize @bullet
22992 @item
22993 source only,
22994
22995 @item
22996 assembly only,
22997
22998 @item
22999 source and assembly,
23000
23001 @item
23002 source and registers, or
23003
23004 @item
23005 assembly and registers.
23006 @end itemize
23007
23008 A status line above the command window shows the following information:
23009
23010 @table @emph
23011 @item target
23012 Indicates the current @value{GDBN} target.
23013 (@pxref{Targets, ,Specifying a Debugging Target}).
23014
23015 @item process
23016 Gives the current process or thread number.
23017 When no process is being debugged, this field is set to @code{No process}.
23018
23019 @item function
23020 Gives the current function name for the selected frame.
23021 The name is demangled if demangling is turned on (@pxref{Print Settings}).
23022 When there is no symbol corresponding to the current program counter,
23023 the string @code{??} is displayed.
23024
23025 @item line
23026 Indicates the current line number for the selected frame.
23027 When the current line number is not known, the string @code{??} is displayed.
23028
23029 @item pc
23030 Indicates the current program counter address.
23031 @end table
23032
23033 @node TUI Keys
23034 @section TUI Key Bindings
23035 @cindex TUI key bindings
23036
23037 The TUI installs several key bindings in the readline keymaps
23038 (@pxref{Command Line Editing}). The following key bindings
23039 are installed for both TUI mode and the @value{GDBN} standard mode.
23040
23041 @table @kbd
23042 @kindex C-x C-a
23043 @item C-x C-a
23044 @kindex C-x a
23045 @itemx C-x a
23046 @kindex C-x A
23047 @itemx C-x A
23048 Enter or leave the TUI mode. When leaving the TUI mode,
23049 the curses window management stops and @value{GDBN} operates using
23050 its standard mode, writing on the terminal directly. When reentering
23051 the TUI mode, control is given back to the curses windows.
23052 The screen is then refreshed.
23053
23054 @kindex C-x 1
23055 @item C-x 1
23056 Use a TUI layout with only one window. The layout will
23057 either be @samp{source} or @samp{assembly}. When the TUI mode
23058 is not active, it will switch to the TUI mode.
23059
23060 Think of this key binding as the Emacs @kbd{C-x 1} binding.
23061
23062 @kindex C-x 2
23063 @item C-x 2
23064 Use a TUI layout with at least two windows. When the current
23065 layout already has two windows, the next layout with two windows is used.
23066 When a new layout is chosen, one window will always be common to the
23067 previous layout and the new one.
23068
23069 Think of it as the Emacs @kbd{C-x 2} binding.
23070
23071 @kindex C-x o
23072 @item C-x o
23073 Change the active window. The TUI associates several key bindings
23074 (like scrolling and arrow keys) with the active window. This command
23075 gives the focus to the next TUI window.
23076
23077 Think of it as the Emacs @kbd{C-x o} binding.
23078
23079 @kindex C-x s
23080 @item C-x s
23081 Switch in and out of the TUI SingleKey mode that binds single
23082 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
23083 @end table
23084
23085 The following key bindings only work in the TUI mode:
23086
23087 @table @asis
23088 @kindex PgUp
23089 @item @key{PgUp}
23090 Scroll the active window one page up.
23091
23092 @kindex PgDn
23093 @item @key{PgDn}
23094 Scroll the active window one page down.
23095
23096 @kindex Up
23097 @item @key{Up}
23098 Scroll the active window one line up.
23099
23100 @kindex Down
23101 @item @key{Down}
23102 Scroll the active window one line down.
23103
23104 @kindex Left
23105 @item @key{Left}
23106 Scroll the active window one column left.
23107
23108 @kindex Right
23109 @item @key{Right}
23110 Scroll the active window one column right.
23111
23112 @kindex C-L
23113 @item @kbd{C-L}
23114 Refresh the screen.
23115 @end table
23116
23117 Because the arrow keys scroll the active window in the TUI mode, they
23118 are not available for their normal use by readline unless the command
23119 window has the focus. When another window is active, you must use
23120 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
23121 and @kbd{C-f} to control the command window.
23122
23123 @node TUI Single Key Mode
23124 @section TUI Single Key Mode
23125 @cindex TUI single key mode
23126
23127 The TUI also provides a @dfn{SingleKey} mode, which binds several
23128 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
23129 switch into this mode, where the following key bindings are used:
23130
23131 @table @kbd
23132 @kindex c @r{(SingleKey TUI key)}
23133 @item c
23134 continue
23135
23136 @kindex d @r{(SingleKey TUI key)}
23137 @item d
23138 down
23139
23140 @kindex f @r{(SingleKey TUI key)}
23141 @item f
23142 finish
23143
23144 @kindex n @r{(SingleKey TUI key)}
23145 @item n
23146 next
23147
23148 @kindex q @r{(SingleKey TUI key)}
23149 @item q
23150 exit the SingleKey mode.
23151
23152 @kindex r @r{(SingleKey TUI key)}
23153 @item r
23154 run
23155
23156 @kindex s @r{(SingleKey TUI key)}
23157 @item s
23158 step
23159
23160 @kindex u @r{(SingleKey TUI key)}
23161 @item u
23162 up
23163
23164 @kindex v @r{(SingleKey TUI key)}
23165 @item v
23166 info locals
23167
23168 @kindex w @r{(SingleKey TUI key)}
23169 @item w
23170 where
23171 @end table
23172
23173 Other keys temporarily switch to the @value{GDBN} command prompt.
23174 The key that was pressed is inserted in the editing buffer so that
23175 it is possible to type most @value{GDBN} commands without interaction
23176 with the TUI SingleKey mode. Once the command is entered the TUI
23177 SingleKey mode is restored. The only way to permanently leave
23178 this mode is by typing @kbd{q} or @kbd{C-x s}.
23179
23180
23181 @node TUI Commands
23182 @section TUI-specific Commands
23183 @cindex TUI commands
23184
23185 The TUI has specific commands to control the text windows.
23186 These commands are always available, even when @value{GDBN} is not in
23187 the TUI mode. When @value{GDBN} is in the standard mode, most
23188 of these commands will automatically switch to the TUI mode.
23189
23190 Note that if @value{GDBN}'s @code{stdout} is not connected to a
23191 terminal, or @value{GDBN} has been started with the machine interface
23192 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
23193 these commands will fail with an error, because it would not be
23194 possible or desirable to enable curses window management.
23195
23196 @table @code
23197 @item info win
23198 @kindex info win
23199 List and give the size of all displayed windows.
23200
23201 @item layout next
23202 @kindex layout
23203 Display the next layout.
23204
23205 @item layout prev
23206 Display the previous layout.
23207
23208 @item layout src
23209 Display the source window only.
23210
23211 @item layout asm
23212 Display the assembly window only.
23213
23214 @item layout split
23215 Display the source and assembly window.
23216
23217 @item layout regs
23218 Display the register window together with the source or assembly window.
23219
23220 @item focus next
23221 @kindex focus
23222 Make the next window active for scrolling.
23223
23224 @item focus prev
23225 Make the previous window active for scrolling.
23226
23227 @item focus src
23228 Make the source window active for scrolling.
23229
23230 @item focus asm
23231 Make the assembly window active for scrolling.
23232
23233 @item focus regs
23234 Make the register window active for scrolling.
23235
23236 @item focus cmd
23237 Make the command window active for scrolling.
23238
23239 @item refresh
23240 @kindex refresh
23241 Refresh the screen. This is similar to typing @kbd{C-L}.
23242
23243 @item tui reg float
23244 @kindex tui reg
23245 Show the floating point registers in the register window.
23246
23247 @item tui reg general
23248 Show the general registers in the register window.
23249
23250 @item tui reg next
23251 Show the next register group. The list of register groups as well as
23252 their order is target specific. The predefined register groups are the
23253 following: @code{general}, @code{float}, @code{system}, @code{vector},
23254 @code{all}, @code{save}, @code{restore}.
23255
23256 @item tui reg system
23257 Show the system registers in the register window.
23258
23259 @item update
23260 @kindex update
23261 Update the source window and the current execution point.
23262
23263 @item winheight @var{name} +@var{count}
23264 @itemx winheight @var{name} -@var{count}
23265 @kindex winheight
23266 Change the height of the window @var{name} by @var{count}
23267 lines. Positive counts increase the height, while negative counts
23268 decrease it.
23269
23270 @item tabset @var{nchars}
23271 @kindex tabset
23272 Set the width of tab stops to be @var{nchars} characters.
23273 @end table
23274
23275 @node TUI Configuration
23276 @section TUI Configuration Variables
23277 @cindex TUI configuration variables
23278
23279 Several configuration variables control the appearance of TUI windows.
23280
23281 @table @code
23282 @item set tui border-kind @var{kind}
23283 @kindex set tui border-kind
23284 Select the border appearance for the source, assembly and register windows.
23285 The possible values are the following:
23286 @table @code
23287 @item space
23288 Use a space character to draw the border.
23289
23290 @item ascii
23291 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
23292
23293 @item acs
23294 Use the Alternate Character Set to draw the border. The border is
23295 drawn using character line graphics if the terminal supports them.
23296 @end table
23297
23298 @item set tui border-mode @var{mode}
23299 @kindex set tui border-mode
23300 @itemx set tui active-border-mode @var{mode}
23301 @kindex set tui active-border-mode
23302 Select the display attributes for the borders of the inactive windows
23303 or the active window. The @var{mode} can be one of the following:
23304 @table @code
23305 @item normal
23306 Use normal attributes to display the border.
23307
23308 @item standout
23309 Use standout mode.
23310
23311 @item reverse
23312 Use reverse video mode.
23313
23314 @item half
23315 Use half bright mode.
23316
23317 @item half-standout
23318 Use half bright and standout mode.
23319
23320 @item bold
23321 Use extra bright or bold mode.
23322
23323 @item bold-standout
23324 Use extra bright or bold and standout mode.
23325 @end table
23326 @end table
23327
23328 @node Emacs
23329 @chapter Using @value{GDBN} under @sc{gnu} Emacs
23330
23331 @cindex Emacs
23332 @cindex @sc{gnu} Emacs
23333 A special interface allows you to use @sc{gnu} Emacs to view (and
23334 edit) the source files for the program you are debugging with
23335 @value{GDBN}.
23336
23337 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
23338 executable file you want to debug as an argument. This command starts
23339 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
23340 created Emacs buffer.
23341 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
23342
23343 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
23344 things:
23345
23346 @itemize @bullet
23347 @item
23348 All ``terminal'' input and output goes through an Emacs buffer, called
23349 the GUD buffer.
23350
23351 This applies both to @value{GDBN} commands and their output, and to the input
23352 and output done by the program you are debugging.
23353
23354 This is useful because it means that you can copy the text of previous
23355 commands and input them again; you can even use parts of the output
23356 in this way.
23357
23358 All the facilities of Emacs' Shell mode are available for interacting
23359 with your program. In particular, you can send signals the usual
23360 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
23361 stop.
23362
23363 @item
23364 @value{GDBN} displays source code through Emacs.
23365
23366 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
23367 source file for that frame and puts an arrow (@samp{=>}) at the
23368 left margin of the current line. Emacs uses a separate buffer for
23369 source display, and splits the screen to show both your @value{GDBN} session
23370 and the source.
23371
23372 Explicit @value{GDBN} @code{list} or search commands still produce output as
23373 usual, but you probably have no reason to use them from Emacs.
23374 @end itemize
23375
23376 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
23377 a graphical mode, enabled by default, which provides further buffers
23378 that can control the execution and describe the state of your program.
23379 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
23380
23381 If you specify an absolute file name when prompted for the @kbd{M-x
23382 gdb} argument, then Emacs sets your current working directory to where
23383 your program resides. If you only specify the file name, then Emacs
23384 sets your current working directory to to the directory associated
23385 with the previous buffer. In this case, @value{GDBN} may find your
23386 program by searching your environment's @code{PATH} variable, but on
23387 some operating systems it might not find the source. So, although the
23388 @value{GDBN} input and output session proceeds normally, the auxiliary
23389 buffer does not display the current source and line of execution.
23390
23391 The initial working directory of @value{GDBN} is printed on the top
23392 line of the GUD buffer and this serves as a default for the commands
23393 that specify files for @value{GDBN} to operate on. @xref{Files,
23394 ,Commands to Specify Files}.
23395
23396 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
23397 need to call @value{GDBN} by a different name (for example, if you
23398 keep several configurations around, with different names) you can
23399 customize the Emacs variable @code{gud-gdb-command-name} to run the
23400 one you want.
23401
23402 In the GUD buffer, you can use these special Emacs commands in
23403 addition to the standard Shell mode commands:
23404
23405 @table @kbd
23406 @item C-h m
23407 Describe the features of Emacs' GUD Mode.
23408
23409 @item C-c C-s
23410 Execute to another source line, like the @value{GDBN} @code{step} command; also
23411 update the display window to show the current file and location.
23412
23413 @item C-c C-n
23414 Execute to next source line in this function, skipping all function
23415 calls, like the @value{GDBN} @code{next} command. Then update the display window
23416 to show the current file and location.
23417
23418 @item C-c C-i
23419 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
23420 display window accordingly.
23421
23422 @item C-c C-f
23423 Execute until exit from the selected stack frame, like the @value{GDBN}
23424 @code{finish} command.
23425
23426 @item C-c C-r
23427 Continue execution of your program, like the @value{GDBN} @code{continue}
23428 command.
23429
23430 @item C-c <
23431 Go up the number of frames indicated by the numeric argument
23432 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
23433 like the @value{GDBN} @code{up} command.
23434
23435 @item C-c >
23436 Go down the number of frames indicated by the numeric argument, like the
23437 @value{GDBN} @code{down} command.
23438 @end table
23439
23440 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
23441 tells @value{GDBN} to set a breakpoint on the source line point is on.
23442
23443 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
23444 separate frame which shows a backtrace when the GUD buffer is current.
23445 Move point to any frame in the stack and type @key{RET} to make it
23446 become the current frame and display the associated source in the
23447 source buffer. Alternatively, click @kbd{Mouse-2} to make the
23448 selected frame become the current one. In graphical mode, the
23449 speedbar displays watch expressions.
23450
23451 If you accidentally delete the source-display buffer, an easy way to get
23452 it back is to type the command @code{f} in the @value{GDBN} buffer, to
23453 request a frame display; when you run under Emacs, this recreates
23454 the source buffer if necessary to show you the context of the current
23455 frame.
23456
23457 The source files displayed in Emacs are in ordinary Emacs buffers
23458 which are visiting the source files in the usual way. You can edit
23459 the files with these buffers if you wish; but keep in mind that @value{GDBN}
23460 communicates with Emacs in terms of line numbers. If you add or
23461 delete lines from the text, the line numbers that @value{GDBN} knows cease
23462 to correspond properly with the code.
23463
23464 A more detailed description of Emacs' interaction with @value{GDBN} is
23465 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
23466 Emacs Manual}).
23467
23468 @c The following dropped because Epoch is nonstandard. Reactivate
23469 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
23470 @ignore
23471 @kindex Emacs Epoch environment
23472 @kindex Epoch
23473 @kindex inspect
23474
23475 Version 18 of @sc{gnu} Emacs has a built-in window system
23476 called the @code{epoch}
23477 environment. Users of this environment can use a new command,
23478 @code{inspect} which performs identically to @code{print} except that
23479 each value is printed in its own window.
23480 @end ignore
23481
23482
23483 @node GDB/MI
23484 @chapter The @sc{gdb/mi} Interface
23485
23486 @unnumberedsec Function and Purpose
23487
23488 @cindex @sc{gdb/mi}, its purpose
23489 @sc{gdb/mi} is a line based machine oriented text interface to
23490 @value{GDBN} and is activated by specifying using the
23491 @option{--interpreter} command line option (@pxref{Mode Options}). It
23492 is specifically intended to support the development of systems which
23493 use the debugger as just one small component of a larger system.
23494
23495 This chapter is a specification of the @sc{gdb/mi} interface. It is written
23496 in the form of a reference manual.
23497
23498 Note that @sc{gdb/mi} is still under construction, so some of the
23499 features described below are incomplete and subject to change
23500 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
23501
23502 @unnumberedsec Notation and Terminology
23503
23504 @cindex notational conventions, for @sc{gdb/mi}
23505 This chapter uses the following notation:
23506
23507 @itemize @bullet
23508 @item
23509 @code{|} separates two alternatives.
23510
23511 @item
23512 @code{[ @var{something} ]} indicates that @var{something} is optional:
23513 it may or may not be given.
23514
23515 @item
23516 @code{( @var{group} )*} means that @var{group} inside the parentheses
23517 may repeat zero or more times.
23518
23519 @item
23520 @code{( @var{group} )+} means that @var{group} inside the parentheses
23521 may repeat one or more times.
23522
23523 @item
23524 @code{"@var{string}"} means a literal @var{string}.
23525 @end itemize
23526
23527 @ignore
23528 @heading Dependencies
23529 @end ignore
23530
23531 @menu
23532 * GDB/MI General Design::
23533 * GDB/MI Command Syntax::
23534 * GDB/MI Compatibility with CLI::
23535 * GDB/MI Development and Front Ends::
23536 * GDB/MI Output Records::
23537 * GDB/MI Simple Examples::
23538 * GDB/MI Command Description Format::
23539 * GDB/MI Breakpoint Commands::
23540 * GDB/MI Program Context::
23541 * GDB/MI Thread Commands::
23542 * GDB/MI Program Execution::
23543 * GDB/MI Stack Manipulation::
23544 * GDB/MI Variable Objects::
23545 * GDB/MI Data Manipulation::
23546 * GDB/MI Tracepoint Commands::
23547 * GDB/MI Symbol Query::
23548 * GDB/MI File Commands::
23549 @ignore
23550 * GDB/MI Kod Commands::
23551 * GDB/MI Memory Overlay Commands::
23552 * GDB/MI Signal Handling Commands::
23553 @end ignore
23554 * GDB/MI Target Manipulation::
23555 * GDB/MI File Transfer Commands::
23556 * GDB/MI Miscellaneous Commands::
23557 @end menu
23558
23559 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23560 @node GDB/MI General Design
23561 @section @sc{gdb/mi} General Design
23562 @cindex GDB/MI General Design
23563
23564 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
23565 parts---commands sent to @value{GDBN}, responses to those commands
23566 and notifications. Each command results in exactly one response,
23567 indicating either successful completion of the command, or an error.
23568 For the commands that do not resume the target, the response contains the
23569 requested information. For the commands that resume the target, the
23570 response only indicates whether the target was successfully resumed.
23571 Notifications is the mechanism for reporting changes in the state of the
23572 target, or in @value{GDBN} state, that cannot conveniently be associated with
23573 a command and reported as part of that command response.
23574
23575 The important examples of notifications are:
23576 @itemize @bullet
23577
23578 @item
23579 Exec notifications. These are used to report changes in
23580 target state---when a target is resumed, or stopped. It would not
23581 be feasible to include this information in response of resuming
23582 commands, because one resume commands can result in multiple events in
23583 different threads. Also, quite some time may pass before any event
23584 happens in the target, while a frontend needs to know whether the resuming
23585 command itself was successfully executed.
23586
23587 @item
23588 Console output, and status notifications. Console output
23589 notifications are used to report output of CLI commands, as well as
23590 diagnostics for other commands. Status notifications are used to
23591 report the progress of a long-running operation. Naturally, including
23592 this information in command response would mean no output is produced
23593 until the command is finished, which is undesirable.
23594
23595 @item
23596 General notifications. Commands may have various side effects on
23597 the @value{GDBN} or target state beyond their official purpose. For example,
23598 a command may change the selected thread. Although such changes can
23599 be included in command response, using notification allows for more
23600 orthogonal frontend design.
23601
23602 @end itemize
23603
23604 There's no guarantee that whenever an MI command reports an error,
23605 @value{GDBN} or the target are in any specific state, and especially,
23606 the state is not reverted to the state before the MI command was
23607 processed. Therefore, whenever an MI command results in an error,
23608 we recommend that the frontend refreshes all the information shown in
23609 the user interface.
23610
23611
23612 @menu
23613 * Context management::
23614 * Asynchronous and non-stop modes::
23615 * Thread groups::
23616 @end menu
23617
23618 @node Context management
23619 @subsection Context management
23620
23621 In most cases when @value{GDBN} accesses the target, this access is
23622 done in context of a specific thread and frame (@pxref{Frames}).
23623 Often, even when accessing global data, the target requires that a thread
23624 be specified. The CLI interface maintains the selected thread and frame,
23625 and supplies them to target on each command. This is convenient,
23626 because a command line user would not want to specify that information
23627 explicitly on each command, and because user interacts with
23628 @value{GDBN} via a single terminal, so no confusion is possible as
23629 to what thread and frame are the current ones.
23630
23631 In the case of MI, the concept of selected thread and frame is less
23632 useful. First, a frontend can easily remember this information
23633 itself. Second, a graphical frontend can have more than one window,
23634 each one used for debugging a different thread, and the frontend might
23635 want to access additional threads for internal purposes. This
23636 increases the risk that by relying on implicitly selected thread, the
23637 frontend may be operating on a wrong one. Therefore, each MI command
23638 should explicitly specify which thread and frame to operate on. To
23639 make it possible, each MI command accepts the @samp{--thread} and
23640 @samp{--frame} options, the value to each is @value{GDBN} identifier
23641 for thread and frame to operate on.
23642
23643 Usually, each top-level window in a frontend allows the user to select
23644 a thread and a frame, and remembers the user selection for further
23645 operations. However, in some cases @value{GDBN} may suggest that the
23646 current thread be changed. For example, when stopping on a breakpoint
23647 it is reasonable to switch to the thread where breakpoint is hit. For
23648 another example, if the user issues the CLI @samp{thread} command via
23649 the frontend, it is desirable to change the frontend's selected thread to the
23650 one specified by user. @value{GDBN} communicates the suggestion to
23651 change current thread using the @samp{=thread-selected} notification.
23652 No such notification is available for the selected frame at the moment.
23653
23654 Note that historically, MI shares the selected thread with CLI, so
23655 frontends used the @code{-thread-select} to execute commands in the
23656 right context. However, getting this to work right is cumbersome. The
23657 simplest way is for frontend to emit @code{-thread-select} command
23658 before every command. This doubles the number of commands that need
23659 to be sent. The alternative approach is to suppress @code{-thread-select}
23660 if the selected thread in @value{GDBN} is supposed to be identical to the
23661 thread the frontend wants to operate on. However, getting this
23662 optimization right can be tricky. In particular, if the frontend
23663 sends several commands to @value{GDBN}, and one of the commands changes the
23664 selected thread, then the behaviour of subsequent commands will
23665 change. So, a frontend should either wait for response from such
23666 problematic commands, or explicitly add @code{-thread-select} for
23667 all subsequent commands. No frontend is known to do this exactly
23668 right, so it is suggested to just always pass the @samp{--thread} and
23669 @samp{--frame} options.
23670
23671 @node Asynchronous and non-stop modes
23672 @subsection Asynchronous command execution and non-stop mode
23673
23674 On some targets, @value{GDBN} is capable of processing MI commands
23675 even while the target is running. This is called @dfn{asynchronous
23676 command execution} (@pxref{Background Execution}). The frontend may
23677 specify a preferrence for asynchronous execution using the
23678 @code{-gdb-set target-async 1} command, which should be emitted before
23679 either running the executable or attaching to the target. After the
23680 frontend has started the executable or attached to the target, it can
23681 find if asynchronous execution is enabled using the
23682 @code{-list-target-features} command.
23683
23684 Even if @value{GDBN} can accept a command while target is running,
23685 many commands that access the target do not work when the target is
23686 running. Therefore, asynchronous command execution is most useful
23687 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
23688 it is possible to examine the state of one thread, while other threads
23689 are running.
23690
23691 When a given thread is running, MI commands that try to access the
23692 target in the context of that thread may not work, or may work only on
23693 some targets. In particular, commands that try to operate on thread's
23694 stack will not work, on any target. Commands that read memory, or
23695 modify breakpoints, may work or not work, depending on the target. Note
23696 that even commands that operate on global state, such as @code{print},
23697 @code{set}, and breakpoint commands, still access the target in the
23698 context of a specific thread, so frontend should try to find a
23699 stopped thread and perform the operation on that thread (using the
23700 @samp{--thread} option).
23701
23702 Which commands will work in the context of a running thread is
23703 highly target dependent. However, the two commands
23704 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
23705 to find the state of a thread, will always work.
23706
23707 @node Thread groups
23708 @subsection Thread groups
23709 @value{GDBN} may be used to debug several processes at the same time.
23710 On some platfroms, @value{GDBN} may support debugging of several
23711 hardware systems, each one having several cores with several different
23712 processes running on each core. This section describes the MI
23713 mechanism to support such debugging scenarios.
23714
23715 The key observation is that regardless of the structure of the
23716 target, MI can have a global list of threads, because most commands that
23717 accept the @samp{--thread} option do not need to know what process that
23718 thread belongs to. Therefore, it is not necessary to introduce
23719 neither additional @samp{--process} option, nor an notion of the
23720 current process in the MI interface. The only strictly new feature
23721 that is required is the ability to find how the threads are grouped
23722 into processes.
23723
23724 To allow the user to discover such grouping, and to support arbitrary
23725 hierarchy of machines/cores/processes, MI introduces the concept of a
23726 @dfn{thread group}. Thread group is a collection of threads and other
23727 thread groups. A thread group always has a string identifier, a type,
23728 and may have additional attributes specific to the type. A new
23729 command, @code{-list-thread-groups}, returns the list of top-level
23730 thread groups, which correspond to processes that @value{GDBN} is
23731 debugging at the moment. By passing an identifier of a thread group
23732 to the @code{-list-thread-groups} command, it is possible to obtain
23733 the members of specific thread group.
23734
23735 To allow the user to easily discover processes, and other objects, he
23736 wishes to debug, a concept of @dfn{available thread group} is
23737 introduced. Available thread group is an thread group that
23738 @value{GDBN} is not debugging, but that can be attached to, using the
23739 @code{-target-attach} command. The list of available top-level thread
23740 groups can be obtained using @samp{-list-thread-groups --available}.
23741 In general, the content of a thread group may be only retrieved only
23742 after attaching to that thread group.
23743
23744 Thread groups are related to inferiors (@pxref{Inferiors and
23745 Programs}). Each inferior corresponds to a thread group of a special
23746 type @samp{process}, and some additional operations are permitted on
23747 such thread groups.
23748
23749 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23750 @node GDB/MI Command Syntax
23751 @section @sc{gdb/mi} Command Syntax
23752
23753 @menu
23754 * GDB/MI Input Syntax::
23755 * GDB/MI Output Syntax::
23756 @end menu
23757
23758 @node GDB/MI Input Syntax
23759 @subsection @sc{gdb/mi} Input Syntax
23760
23761 @cindex input syntax for @sc{gdb/mi}
23762 @cindex @sc{gdb/mi}, input syntax
23763 @table @code
23764 @item @var{command} @expansion{}
23765 @code{@var{cli-command} | @var{mi-command}}
23766
23767 @item @var{cli-command} @expansion{}
23768 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
23769 @var{cli-command} is any existing @value{GDBN} CLI command.
23770
23771 @item @var{mi-command} @expansion{}
23772 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
23773 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
23774
23775 @item @var{token} @expansion{}
23776 "any sequence of digits"
23777
23778 @item @var{option} @expansion{}
23779 @code{"-" @var{parameter} [ " " @var{parameter} ]}
23780
23781 @item @var{parameter} @expansion{}
23782 @code{@var{non-blank-sequence} | @var{c-string}}
23783
23784 @item @var{operation} @expansion{}
23785 @emph{any of the operations described in this chapter}
23786
23787 @item @var{non-blank-sequence} @expansion{}
23788 @emph{anything, provided it doesn't contain special characters such as
23789 "-", @var{nl}, """ and of course " "}
23790
23791 @item @var{c-string} @expansion{}
23792 @code{""" @var{seven-bit-iso-c-string-content} """}
23793
23794 @item @var{nl} @expansion{}
23795 @code{CR | CR-LF}
23796 @end table
23797
23798 @noindent
23799 Notes:
23800
23801 @itemize @bullet
23802 @item
23803 The CLI commands are still handled by the @sc{mi} interpreter; their
23804 output is described below.
23805
23806 @item
23807 The @code{@var{token}}, when present, is passed back when the command
23808 finishes.
23809
23810 @item
23811 Some @sc{mi} commands accept optional arguments as part of the parameter
23812 list. Each option is identified by a leading @samp{-} (dash) and may be
23813 followed by an optional argument parameter. Options occur first in the
23814 parameter list and can be delimited from normal parameters using
23815 @samp{--} (this is useful when some parameters begin with a dash).
23816 @end itemize
23817
23818 Pragmatics:
23819
23820 @itemize @bullet
23821 @item
23822 We want easy access to the existing CLI syntax (for debugging).
23823
23824 @item
23825 We want it to be easy to spot a @sc{mi} operation.
23826 @end itemize
23827
23828 @node GDB/MI Output Syntax
23829 @subsection @sc{gdb/mi} Output Syntax
23830
23831 @cindex output syntax of @sc{gdb/mi}
23832 @cindex @sc{gdb/mi}, output syntax
23833 The output from @sc{gdb/mi} consists of zero or more out-of-band records
23834 followed, optionally, by a single result record. This result record
23835 is for the most recent command. The sequence of output records is
23836 terminated by @samp{(gdb)}.
23837
23838 If an input command was prefixed with a @code{@var{token}} then the
23839 corresponding output for that command will also be prefixed by that same
23840 @var{token}.
23841
23842 @table @code
23843 @item @var{output} @expansion{}
23844 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
23845
23846 @item @var{result-record} @expansion{}
23847 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
23848
23849 @item @var{out-of-band-record} @expansion{}
23850 @code{@var{async-record} | @var{stream-record}}
23851
23852 @item @var{async-record} @expansion{}
23853 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
23854
23855 @item @var{exec-async-output} @expansion{}
23856 @code{[ @var{token} ] "*" @var{async-output}}
23857
23858 @item @var{status-async-output} @expansion{}
23859 @code{[ @var{token} ] "+" @var{async-output}}
23860
23861 @item @var{notify-async-output} @expansion{}
23862 @code{[ @var{token} ] "=" @var{async-output}}
23863
23864 @item @var{async-output} @expansion{}
23865 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
23866
23867 @item @var{result-class} @expansion{}
23868 @code{"done" | "running" | "connected" | "error" | "exit"}
23869
23870 @item @var{async-class} @expansion{}
23871 @code{"stopped" | @var{others}} (where @var{others} will be added
23872 depending on the needs---this is still in development).
23873
23874 @item @var{result} @expansion{}
23875 @code{ @var{variable} "=" @var{value}}
23876
23877 @item @var{variable} @expansion{}
23878 @code{ @var{string} }
23879
23880 @item @var{value} @expansion{}
23881 @code{ @var{const} | @var{tuple} | @var{list} }
23882
23883 @item @var{const} @expansion{}
23884 @code{@var{c-string}}
23885
23886 @item @var{tuple} @expansion{}
23887 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
23888
23889 @item @var{list} @expansion{}
23890 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
23891 @var{result} ( "," @var{result} )* "]" }
23892
23893 @item @var{stream-record} @expansion{}
23894 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
23895
23896 @item @var{console-stream-output} @expansion{}
23897 @code{"~" @var{c-string}}
23898
23899 @item @var{target-stream-output} @expansion{}
23900 @code{"@@" @var{c-string}}
23901
23902 @item @var{log-stream-output} @expansion{}
23903 @code{"&" @var{c-string}}
23904
23905 @item @var{nl} @expansion{}
23906 @code{CR | CR-LF}
23907
23908 @item @var{token} @expansion{}
23909 @emph{any sequence of digits}.
23910 @end table
23911
23912 @noindent
23913 Notes:
23914
23915 @itemize @bullet
23916 @item
23917 All output sequences end in a single line containing a period.
23918
23919 @item
23920 The @code{@var{token}} is from the corresponding request. Note that
23921 for all async output, while the token is allowed by the grammar and
23922 may be output by future versions of @value{GDBN} for select async
23923 output messages, it is generally omitted. Frontends should treat
23924 all async output as reporting general changes in the state of the
23925 target and there should be no need to associate async output to any
23926 prior command.
23927
23928 @item
23929 @cindex status output in @sc{gdb/mi}
23930 @var{status-async-output} contains on-going status information about the
23931 progress of a slow operation. It can be discarded. All status output is
23932 prefixed by @samp{+}.
23933
23934 @item
23935 @cindex async output in @sc{gdb/mi}
23936 @var{exec-async-output} contains asynchronous state change on the target
23937 (stopped, started, disappeared). All async output is prefixed by
23938 @samp{*}.
23939
23940 @item
23941 @cindex notify output in @sc{gdb/mi}
23942 @var{notify-async-output} contains supplementary information that the
23943 client should handle (e.g., a new breakpoint information). All notify
23944 output is prefixed by @samp{=}.
23945
23946 @item
23947 @cindex console output in @sc{gdb/mi}
23948 @var{console-stream-output} is output that should be displayed as is in the
23949 console. It is the textual response to a CLI command. All the console
23950 output is prefixed by @samp{~}.
23951
23952 @item
23953 @cindex target output in @sc{gdb/mi}
23954 @var{target-stream-output} is the output produced by the target program.
23955 All the target output is prefixed by @samp{@@}.
23956
23957 @item
23958 @cindex log output in @sc{gdb/mi}
23959 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
23960 instance messages that should be displayed as part of an error log. All
23961 the log output is prefixed by @samp{&}.
23962
23963 @item
23964 @cindex list output in @sc{gdb/mi}
23965 New @sc{gdb/mi} commands should only output @var{lists} containing
23966 @var{values}.
23967
23968
23969 @end itemize
23970
23971 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
23972 details about the various output records.
23973
23974 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23975 @node GDB/MI Compatibility with CLI
23976 @section @sc{gdb/mi} Compatibility with CLI
23977
23978 @cindex compatibility, @sc{gdb/mi} and CLI
23979 @cindex @sc{gdb/mi}, compatibility with CLI
23980
23981 For the developers convenience CLI commands can be entered directly,
23982 but there may be some unexpected behaviour. For example, commands
23983 that query the user will behave as if the user replied yes, breakpoint
23984 command lists are not executed and some CLI commands, such as
23985 @code{if}, @code{when} and @code{define}, prompt for further input with
23986 @samp{>}, which is not valid MI output.
23987
23988 This feature may be removed at some stage in the future and it is
23989 recommended that front ends use the @code{-interpreter-exec} command
23990 (@pxref{-interpreter-exec}).
23991
23992 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23993 @node GDB/MI Development and Front Ends
23994 @section @sc{gdb/mi} Development and Front Ends
23995 @cindex @sc{gdb/mi} development
23996
23997 The application which takes the MI output and presents the state of the
23998 program being debugged to the user is called a @dfn{front end}.
23999
24000 Although @sc{gdb/mi} is still incomplete, it is currently being used
24001 by a variety of front ends to @value{GDBN}. This makes it difficult
24002 to introduce new functionality without breaking existing usage. This
24003 section tries to minimize the problems by describing how the protocol
24004 might change.
24005
24006 Some changes in MI need not break a carefully designed front end, and
24007 for these the MI version will remain unchanged. The following is a
24008 list of changes that may occur within one level, so front ends should
24009 parse MI output in a way that can handle them:
24010
24011 @itemize @bullet
24012 @item
24013 New MI commands may be added.
24014
24015 @item
24016 New fields may be added to the output of any MI command.
24017
24018 @item
24019 The range of values for fields with specified values, e.g.,
24020 @code{in_scope} (@pxref{-var-update}) may be extended.
24021
24022 @c The format of field's content e.g type prefix, may change so parse it
24023 @c at your own risk. Yes, in general?
24024
24025 @c The order of fields may change? Shouldn't really matter but it might
24026 @c resolve inconsistencies.
24027 @end itemize
24028
24029 If the changes are likely to break front ends, the MI version level
24030 will be increased by one. This will allow the front end to parse the
24031 output according to the MI version. Apart from mi0, new versions of
24032 @value{GDBN} will not support old versions of MI and it will be the
24033 responsibility of the front end to work with the new one.
24034
24035 @c Starting with mi3, add a new command -mi-version that prints the MI
24036 @c version?
24037
24038 The best way to avoid unexpected changes in MI that might break your front
24039 end is to make your project known to @value{GDBN} developers and
24040 follow development on @email{gdb@@sourceware.org} and
24041 @email{gdb-patches@@sourceware.org}.
24042 @cindex mailing lists
24043
24044 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24045 @node GDB/MI Output Records
24046 @section @sc{gdb/mi} Output Records
24047
24048 @menu
24049 * GDB/MI Result Records::
24050 * GDB/MI Stream Records::
24051 * GDB/MI Async Records::
24052 * GDB/MI Frame Information::
24053 * GDB/MI Thread Information::
24054 @end menu
24055
24056 @node GDB/MI Result Records
24057 @subsection @sc{gdb/mi} Result Records
24058
24059 @cindex result records in @sc{gdb/mi}
24060 @cindex @sc{gdb/mi}, result records
24061 In addition to a number of out-of-band notifications, the response to a
24062 @sc{gdb/mi} command includes one of the following result indications:
24063
24064 @table @code
24065 @findex ^done
24066 @item "^done" [ "," @var{results} ]
24067 The synchronous operation was successful, @code{@var{results}} are the return
24068 values.
24069
24070 @item "^running"
24071 @findex ^running
24072 This result record is equivalent to @samp{^done}. Historically, it
24073 was output instead of @samp{^done} if the command has resumed the
24074 target. This behaviour is maintained for backward compatibility, but
24075 all frontends should treat @samp{^done} and @samp{^running}
24076 identically and rely on the @samp{*running} output record to determine
24077 which threads are resumed.
24078
24079 @item "^connected"
24080 @findex ^connected
24081 @value{GDBN} has connected to a remote target.
24082
24083 @item "^error" "," @var{c-string}
24084 @findex ^error
24085 The operation failed. The @code{@var{c-string}} contains the corresponding
24086 error message.
24087
24088 @item "^exit"
24089 @findex ^exit
24090 @value{GDBN} has terminated.
24091
24092 @end table
24093
24094 @node GDB/MI Stream Records
24095 @subsection @sc{gdb/mi} Stream Records
24096
24097 @cindex @sc{gdb/mi}, stream records
24098 @cindex stream records in @sc{gdb/mi}
24099 @value{GDBN} internally maintains a number of output streams: the console, the
24100 target, and the log. The output intended for each of these streams is
24101 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
24102
24103 Each stream record begins with a unique @dfn{prefix character} which
24104 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
24105 Syntax}). In addition to the prefix, each stream record contains a
24106 @code{@var{string-output}}. This is either raw text (with an implicit new
24107 line) or a quoted C string (which does not contain an implicit newline).
24108
24109 @table @code
24110 @item "~" @var{string-output}
24111 The console output stream contains text that should be displayed in the
24112 CLI console window. It contains the textual responses to CLI commands.
24113
24114 @item "@@" @var{string-output}
24115 The target output stream contains any textual output from the running
24116 target. This is only present when GDB's event loop is truly
24117 asynchronous, which is currently only the case for remote targets.
24118
24119 @item "&" @var{string-output}
24120 The log stream contains debugging messages being produced by @value{GDBN}'s
24121 internals.
24122 @end table
24123
24124 @node GDB/MI Async Records
24125 @subsection @sc{gdb/mi} Async Records
24126
24127 @cindex async records in @sc{gdb/mi}
24128 @cindex @sc{gdb/mi}, async records
24129 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
24130 additional changes that have occurred. Those changes can either be a
24131 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
24132 target activity (e.g., target stopped).
24133
24134 The following is the list of possible async records:
24135
24136 @table @code
24137
24138 @item *running,thread-id="@var{thread}"
24139 The target is now running. The @var{thread} field tells which
24140 specific thread is now running, and can be @samp{all} if all threads
24141 are running. The frontend should assume that no interaction with a
24142 running thread is possible after this notification is produced.
24143 The frontend should not assume that this notification is output
24144 only once for any command. @value{GDBN} may emit this notification
24145 several times, either for different threads, because it cannot resume
24146 all threads together, or even for a single thread, if the thread must
24147 be stepped though some code before letting it run freely.
24148
24149 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
24150 The target has stopped. The @var{reason} field can have one of the
24151 following values:
24152
24153 @table @code
24154 @item breakpoint-hit
24155 A breakpoint was reached.
24156 @item watchpoint-trigger
24157 A watchpoint was triggered.
24158 @item read-watchpoint-trigger
24159 A read watchpoint was triggered.
24160 @item access-watchpoint-trigger
24161 An access watchpoint was triggered.
24162 @item function-finished
24163 An -exec-finish or similar CLI command was accomplished.
24164 @item location-reached
24165 An -exec-until or similar CLI command was accomplished.
24166 @item watchpoint-scope
24167 A watchpoint has gone out of scope.
24168 @item end-stepping-range
24169 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
24170 similar CLI command was accomplished.
24171 @item exited-signalled
24172 The inferior exited because of a signal.
24173 @item exited
24174 The inferior exited.
24175 @item exited-normally
24176 The inferior exited normally.
24177 @item signal-received
24178 A signal was received by the inferior.
24179 @end table
24180
24181 The @var{id} field identifies the thread that directly caused the stop
24182 -- for example by hitting a breakpoint. Depending on whether all-stop
24183 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
24184 stop all threads, or only the thread that directly triggered the stop.
24185 If all threads are stopped, the @var{stopped} field will have the
24186 value of @code{"all"}. Otherwise, the value of the @var{stopped}
24187 field will be a list of thread identifiers. Presently, this list will
24188 always include a single thread, but frontend should be prepared to see
24189 several threads in the list. The @var{core} field reports the
24190 processor core on which the stop event has happened. This field may be absent
24191 if such information is not available.
24192
24193 @item =thread-group-added,id="@var{id}"
24194 @itemx =thread-group-removed,id="@var{id}"
24195 A thread group was either added or removed. The @var{id} field
24196 contains the @value{GDBN} identifier of the thread group. When a thread
24197 group is added, it generally might not be associated with a running
24198 process. When a thread group is removed, its id becomes invalid and
24199 cannot be used in any way.
24200
24201 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
24202 A thread group became associated with a running program,
24203 either because the program was just started or the thread group
24204 was attached to a program. The @var{id} field contains the
24205 @value{GDBN} identifier of the thread group. The @var{pid} field
24206 contains process identifier, specific to the operating system.
24207
24208 @itemx =thread-group-exited,id="@var{id}"
24209 A thread group is no longer associated with a running program,
24210 either because the program has exited, or because it was detached
24211 from. The @var{id} field contains the @value{GDBN} identifier of the
24212 thread group.
24213
24214 @item =thread-created,id="@var{id}",group-id="@var{gid}"
24215 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
24216 A thread either was created, or has exited. The @var{id} field
24217 contains the @value{GDBN} identifier of the thread. The @var{gid}
24218 field identifies the thread group this thread belongs to.
24219
24220 @item =thread-selected,id="@var{id}"
24221 Informs that the selected thread was changed as result of the last
24222 command. This notification is not emitted as result of @code{-thread-select}
24223 command but is emitted whenever an MI command that is not documented
24224 to change the selected thread actually changes it. In particular,
24225 invoking, directly or indirectly (via user-defined command), the CLI
24226 @code{thread} command, will generate this notification.
24227
24228 We suggest that in response to this notification, front ends
24229 highlight the selected thread and cause subsequent commands to apply to
24230 that thread.
24231
24232 @item =library-loaded,...
24233 Reports that a new library file was loaded by the program. This
24234 notification has 4 fields---@var{id}, @var{target-name},
24235 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
24236 opaque identifier of the library. For remote debugging case,
24237 @var{target-name} and @var{host-name} fields give the name of the
24238 library file on the target, and on the host respectively. For native
24239 debugging, both those fields have the same value. The
24240 @var{symbols-loaded} field reports if the debug symbols for this
24241 library are loaded. The @var{thread-group} field, if present,
24242 specifies the id of the thread group in whose context the library was loaded.
24243 If the field is absent, it means the library was loaded in the context
24244 of all present thread groups.
24245
24246 @item =library-unloaded,...
24247 Reports that a library was unloaded by the program. This notification
24248 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
24249 the same meaning as for the @code{=library-loaded} notification.
24250 The @var{thread-group} field, if present, specifies the id of the
24251 thread group in whose context the library was unloaded. If the field is
24252 absent, it means the library was unloaded in the context of all present
24253 thread groups.
24254
24255 @end table
24256
24257 @node GDB/MI Frame Information
24258 @subsection @sc{gdb/mi} Frame Information
24259
24260 Response from many MI commands includes an information about stack
24261 frame. This information is a tuple that may have the following
24262 fields:
24263
24264 @table @code
24265 @item level
24266 The level of the stack frame. The innermost frame has the level of
24267 zero. This field is always present.
24268
24269 @item func
24270 The name of the function corresponding to the frame. This field may
24271 be absent if @value{GDBN} is unable to determine the function name.
24272
24273 @item addr
24274 The code address for the frame. This field is always present.
24275
24276 @item file
24277 The name of the source files that correspond to the frame's code
24278 address. This field may be absent.
24279
24280 @item line
24281 The source line corresponding to the frames' code address. This field
24282 may be absent.
24283
24284 @item from
24285 The name of the binary file (either executable or shared library) the
24286 corresponds to the frame's code address. This field may be absent.
24287
24288 @end table
24289
24290 @node GDB/MI Thread Information
24291 @subsection @sc{gdb/mi} Thread Information
24292
24293 Whenever @value{GDBN} has to report an information about a thread, it
24294 uses a tuple with the following fields:
24295
24296 @table @code
24297 @item id
24298 The numeric id assigned to the thread by @value{GDBN}. This field is
24299 always present.
24300
24301 @item target-id
24302 Target-specific string identifying the thread. This field is always present.
24303
24304 @item details
24305 Additional information about the thread provided by the target.
24306 It is supposed to be human-readable and not interpreted by the
24307 frontend. This field is optional.
24308
24309 @item state
24310 Either @samp{stopped} or @samp{running}, depending on whether the
24311 thread is presently running. This field is always present.
24312
24313 @item core
24314 The value of this field is an integer number of the processor core the
24315 thread was last seen on. This field is optional.
24316 @end table
24317
24318
24319 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24320 @node GDB/MI Simple Examples
24321 @section Simple Examples of @sc{gdb/mi} Interaction
24322 @cindex @sc{gdb/mi}, simple examples
24323
24324 This subsection presents several simple examples of interaction using
24325 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
24326 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
24327 the output received from @sc{gdb/mi}.
24328
24329 Note the line breaks shown in the examples are here only for
24330 readability, they don't appear in the real output.
24331
24332 @subheading Setting a Breakpoint
24333
24334 Setting a breakpoint generates synchronous output which contains detailed
24335 information of the breakpoint.
24336
24337 @smallexample
24338 -> -break-insert main
24339 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
24340 enabled="y",addr="0x08048564",func="main",file="myprog.c",
24341 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
24342 <- (gdb)
24343 @end smallexample
24344
24345 @subheading Program Execution
24346
24347 Program execution generates asynchronous records and MI gives the
24348 reason that execution stopped.
24349
24350 @smallexample
24351 -> -exec-run
24352 <- ^running
24353 <- (gdb)
24354 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
24355 frame=@{addr="0x08048564",func="main",
24356 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
24357 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
24358 <- (gdb)
24359 -> -exec-continue
24360 <- ^running
24361 <- (gdb)
24362 <- *stopped,reason="exited-normally"
24363 <- (gdb)
24364 @end smallexample
24365
24366 @subheading Quitting @value{GDBN}
24367
24368 Quitting @value{GDBN} just prints the result class @samp{^exit}.
24369
24370 @smallexample
24371 -> (gdb)
24372 <- -gdb-exit
24373 <- ^exit
24374 @end smallexample
24375
24376 Please note that @samp{^exit} is printed immediately, but it might
24377 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
24378 performs necessary cleanups, including killing programs being debugged
24379 or disconnecting from debug hardware, so the frontend should wait till
24380 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
24381 fails to exit in reasonable time.
24382
24383 @subheading A Bad Command
24384
24385 Here's what happens if you pass a non-existent command:
24386
24387 @smallexample
24388 -> -rubbish
24389 <- ^error,msg="Undefined MI command: rubbish"
24390 <- (gdb)
24391 @end smallexample
24392
24393
24394 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24395 @node GDB/MI Command Description Format
24396 @section @sc{gdb/mi} Command Description Format
24397
24398 The remaining sections describe blocks of commands. Each block of
24399 commands is laid out in a fashion similar to this section.
24400
24401 @subheading Motivation
24402
24403 The motivation for this collection of commands.
24404
24405 @subheading Introduction
24406
24407 A brief introduction to this collection of commands as a whole.
24408
24409 @subheading Commands
24410
24411 For each command in the block, the following is described:
24412
24413 @subsubheading Synopsis
24414
24415 @smallexample
24416 -command @var{args}@dots{}
24417 @end smallexample
24418
24419 @subsubheading Result
24420
24421 @subsubheading @value{GDBN} Command
24422
24423 The corresponding @value{GDBN} CLI command(s), if any.
24424
24425 @subsubheading Example
24426
24427 Example(s) formatted for readability. Some of the described commands have
24428 not been implemented yet and these are labeled N.A.@: (not available).
24429
24430
24431 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24432 @node GDB/MI Breakpoint Commands
24433 @section @sc{gdb/mi} Breakpoint Commands
24434
24435 @cindex breakpoint commands for @sc{gdb/mi}
24436 @cindex @sc{gdb/mi}, breakpoint commands
24437 This section documents @sc{gdb/mi} commands for manipulating
24438 breakpoints.
24439
24440 @subheading The @code{-break-after} Command
24441 @findex -break-after
24442
24443 @subsubheading Synopsis
24444
24445 @smallexample
24446 -break-after @var{number} @var{count}
24447 @end smallexample
24448
24449 The breakpoint number @var{number} is not in effect until it has been
24450 hit @var{count} times. To see how this is reflected in the output of
24451 the @samp{-break-list} command, see the description of the
24452 @samp{-break-list} command below.
24453
24454 @subsubheading @value{GDBN} Command
24455
24456 The corresponding @value{GDBN} command is @samp{ignore}.
24457
24458 @subsubheading Example
24459
24460 @smallexample
24461 (gdb)
24462 -break-insert main
24463 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
24464 enabled="y",addr="0x000100d0",func="main",file="hello.c",
24465 fullname="/home/foo/hello.c",line="5",times="0"@}
24466 (gdb)
24467 -break-after 1 3
24468 ~
24469 ^done
24470 (gdb)
24471 -break-list
24472 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
24473 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
24474 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
24475 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
24476 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
24477 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
24478 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
24479 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
24480 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
24481 line="5",times="0",ignore="3"@}]@}
24482 (gdb)
24483 @end smallexample
24484
24485 @ignore
24486 @subheading The @code{-break-catch} Command
24487 @findex -break-catch
24488 @end ignore
24489
24490 @subheading The @code{-break-commands} Command
24491 @findex -break-commands
24492
24493 @subsubheading Synopsis
24494
24495 @smallexample
24496 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
24497 @end smallexample
24498
24499 Specifies the CLI commands that should be executed when breakpoint
24500 @var{number} is hit. The parameters @var{command1} to @var{commandN}
24501 are the commands. If no command is specified, any previously-set
24502 commands are cleared. @xref{Break Commands}. Typical use of this
24503 functionality is tracing a program, that is, printing of values of
24504 some variables whenever breakpoint is hit and then continuing.
24505
24506 @subsubheading @value{GDBN} Command
24507
24508 The corresponding @value{GDBN} command is @samp{commands}.
24509
24510 @subsubheading Example
24511
24512 @smallexample
24513 (gdb)
24514 -break-insert main
24515 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
24516 enabled="y",addr="0x000100d0",func="main",file="hello.c",
24517 fullname="/home/foo/hello.c",line="5",times="0"@}
24518 (gdb)
24519 -break-commands 1 "print v" "continue"
24520 ^done
24521 (gdb)
24522 @end smallexample
24523
24524 @subheading The @code{-break-condition} Command
24525 @findex -break-condition
24526
24527 @subsubheading Synopsis
24528
24529 @smallexample
24530 -break-condition @var{number} @var{expr}
24531 @end smallexample
24532
24533 Breakpoint @var{number} will stop the program only if the condition in
24534 @var{expr} is true. The condition becomes part of the
24535 @samp{-break-list} output (see the description of the @samp{-break-list}
24536 command below).
24537
24538 @subsubheading @value{GDBN} Command
24539
24540 The corresponding @value{GDBN} command is @samp{condition}.
24541
24542 @subsubheading Example
24543
24544 @smallexample
24545 (gdb)
24546 -break-condition 1 1
24547 ^done
24548 (gdb)
24549 -break-list
24550 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
24551 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
24552 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
24553 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
24554 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
24555 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
24556 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
24557 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
24558 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
24559 line="5",cond="1",times="0",ignore="3"@}]@}
24560 (gdb)
24561 @end smallexample
24562
24563 @subheading The @code{-break-delete} Command
24564 @findex -break-delete
24565
24566 @subsubheading Synopsis
24567
24568 @smallexample
24569 -break-delete ( @var{breakpoint} )+
24570 @end smallexample
24571
24572 Delete the breakpoint(s) whose number(s) are specified in the argument
24573 list. This is obviously reflected in the breakpoint list.
24574
24575 @subsubheading @value{GDBN} Command
24576
24577 The corresponding @value{GDBN} command is @samp{delete}.
24578
24579 @subsubheading Example
24580
24581 @smallexample
24582 (gdb)
24583 -break-delete 1
24584 ^done
24585 (gdb)
24586 -break-list
24587 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
24588 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
24589 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
24590 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
24591 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
24592 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
24593 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
24594 body=[]@}
24595 (gdb)
24596 @end smallexample
24597
24598 @subheading The @code{-break-disable} Command
24599 @findex -break-disable
24600
24601 @subsubheading Synopsis
24602
24603 @smallexample
24604 -break-disable ( @var{breakpoint} )+
24605 @end smallexample
24606
24607 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
24608 break list is now set to @samp{n} for the named @var{breakpoint}(s).
24609
24610 @subsubheading @value{GDBN} Command
24611
24612 The corresponding @value{GDBN} command is @samp{disable}.
24613
24614 @subsubheading Example
24615
24616 @smallexample
24617 (gdb)
24618 -break-disable 2
24619 ^done
24620 (gdb)
24621 -break-list
24622 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
24623 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
24624 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
24625 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
24626 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
24627 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
24628 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
24629 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
24630 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
24631 line="5",times="0"@}]@}
24632 (gdb)
24633 @end smallexample
24634
24635 @subheading The @code{-break-enable} Command
24636 @findex -break-enable
24637
24638 @subsubheading Synopsis
24639
24640 @smallexample
24641 -break-enable ( @var{breakpoint} )+
24642 @end smallexample
24643
24644 Enable (previously disabled) @var{breakpoint}(s).
24645
24646 @subsubheading @value{GDBN} Command
24647
24648 The corresponding @value{GDBN} command is @samp{enable}.
24649
24650 @subsubheading Example
24651
24652 @smallexample
24653 (gdb)
24654 -break-enable 2
24655 ^done
24656 (gdb)
24657 -break-list
24658 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
24659 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
24660 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
24661 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
24662 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
24663 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
24664 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
24665 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
24666 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
24667 line="5",times="0"@}]@}
24668 (gdb)
24669 @end smallexample
24670
24671 @subheading The @code{-break-info} Command
24672 @findex -break-info
24673
24674 @subsubheading Synopsis
24675
24676 @smallexample
24677 -break-info @var{breakpoint}
24678 @end smallexample
24679
24680 @c REDUNDANT???
24681 Get information about a single breakpoint.
24682
24683 @subsubheading @value{GDBN} Command
24684
24685 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
24686
24687 @subsubheading Example
24688 N.A.
24689
24690 @subheading The @code{-break-insert} Command
24691 @findex -break-insert
24692
24693 @subsubheading Synopsis
24694
24695 @smallexample
24696 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
24697 [ -c @var{condition} ] [ -i @var{ignore-count} ]
24698 [ -p @var{thread} ] [ @var{location} ]
24699 @end smallexample
24700
24701 @noindent
24702 If specified, @var{location}, can be one of:
24703
24704 @itemize @bullet
24705 @item function
24706 @c @item +offset
24707 @c @item -offset
24708 @c @item linenum
24709 @item filename:linenum
24710 @item filename:function
24711 @item *address
24712 @end itemize
24713
24714 The possible optional parameters of this command are:
24715
24716 @table @samp
24717 @item -t
24718 Insert a temporary breakpoint.
24719 @item -h
24720 Insert a hardware breakpoint.
24721 @item -c @var{condition}
24722 Make the breakpoint conditional on @var{condition}.
24723 @item -i @var{ignore-count}
24724 Initialize the @var{ignore-count}.
24725 @item -f
24726 If @var{location} cannot be parsed (for example if it
24727 refers to unknown files or functions), create a pending
24728 breakpoint. Without this flag, @value{GDBN} will report
24729 an error, and won't create a breakpoint, if @var{location}
24730 cannot be parsed.
24731 @item -d
24732 Create a disabled breakpoint.
24733 @item -a
24734 Create a tracepoint. @xref{Tracepoints}. When this parameter
24735 is used together with @samp{-h}, a fast tracepoint is created.
24736 @end table
24737
24738 @subsubheading Result
24739
24740 The result is in the form:
24741
24742 @smallexample
24743 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
24744 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
24745 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
24746 times="@var{times}"@}
24747 @end smallexample
24748
24749 @noindent
24750 where @var{number} is the @value{GDBN} number for this breakpoint,
24751 @var{funcname} is the name of the function where the breakpoint was
24752 inserted, @var{filename} is the name of the source file which contains
24753 this function, @var{lineno} is the source line number within that file
24754 and @var{times} the number of times that the breakpoint has been hit
24755 (always 0 for -break-insert but may be greater for -break-info or -break-list
24756 which use the same output).
24757
24758 Note: this format is open to change.
24759 @c An out-of-band breakpoint instead of part of the result?
24760
24761 @subsubheading @value{GDBN} Command
24762
24763 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
24764 @samp{hbreak}, @samp{thbreak}, and @samp{rbreak}.
24765
24766 @subsubheading Example
24767
24768 @smallexample
24769 (gdb)
24770 -break-insert main
24771 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
24772 fullname="/home/foo/recursive2.c,line="4",times="0"@}
24773 (gdb)
24774 -break-insert -t foo
24775 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
24776 fullname="/home/foo/recursive2.c,line="11",times="0"@}
24777 (gdb)
24778 -break-list
24779 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
24780 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
24781 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
24782 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
24783 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
24784 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
24785 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
24786 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
24787 addr="0x0001072c", func="main",file="recursive2.c",
24788 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
24789 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
24790 addr="0x00010774",func="foo",file="recursive2.c",
24791 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
24792 (gdb)
24793 -break-insert -r foo.*
24794 ~int foo(int, int);
24795 ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
24796 "fullname="/home/foo/recursive2.c",line="11",times="0"@}
24797 (gdb)
24798 @end smallexample
24799
24800 @subheading The @code{-break-list} Command
24801 @findex -break-list
24802
24803 @subsubheading Synopsis
24804
24805 @smallexample
24806 -break-list
24807 @end smallexample
24808
24809 Displays the list of inserted breakpoints, showing the following fields:
24810
24811 @table @samp
24812 @item Number
24813 number of the breakpoint
24814 @item Type
24815 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
24816 @item Disposition
24817 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
24818 or @samp{nokeep}
24819 @item Enabled
24820 is the breakpoint enabled or no: @samp{y} or @samp{n}
24821 @item Address
24822 memory location at which the breakpoint is set
24823 @item What
24824 logical location of the breakpoint, expressed by function name, file
24825 name, line number
24826 @item Times
24827 number of times the breakpoint has been hit
24828 @end table
24829
24830 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
24831 @code{body} field is an empty list.
24832
24833 @subsubheading @value{GDBN} Command
24834
24835 The corresponding @value{GDBN} command is @samp{info break}.
24836
24837 @subsubheading Example
24838
24839 @smallexample
24840 (gdb)
24841 -break-list
24842 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
24843 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
24844 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
24845 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
24846 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
24847 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
24848 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
24849 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
24850 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
24851 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
24852 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
24853 line="13",times="0"@}]@}
24854 (gdb)
24855 @end smallexample
24856
24857 Here's an example of the result when there are no breakpoints:
24858
24859 @smallexample
24860 (gdb)
24861 -break-list
24862 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
24863 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
24864 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
24865 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
24866 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
24867 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
24868 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
24869 body=[]@}
24870 (gdb)
24871 @end smallexample
24872
24873 @subheading The @code{-break-passcount} Command
24874 @findex -break-passcount
24875
24876 @subsubheading Synopsis
24877
24878 @smallexample
24879 -break-passcount @var{tracepoint-number} @var{passcount}
24880 @end smallexample
24881
24882 Set the passcount for tracepoint @var{tracepoint-number} to
24883 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
24884 is not a tracepoint, error is emitted. This corresponds to CLI
24885 command @samp{passcount}.
24886
24887 @subheading The @code{-break-watch} Command
24888 @findex -break-watch
24889
24890 @subsubheading Synopsis
24891
24892 @smallexample
24893 -break-watch [ -a | -r ]
24894 @end smallexample
24895
24896 Create a watchpoint. With the @samp{-a} option it will create an
24897 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
24898 read from or on a write to the memory location. With the @samp{-r}
24899 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
24900 trigger only when the memory location is accessed for reading. Without
24901 either of the options, the watchpoint created is a regular watchpoint,
24902 i.e., it will trigger when the memory location is accessed for writing.
24903 @xref{Set Watchpoints, , Setting Watchpoints}.
24904
24905 Note that @samp{-break-list} will report a single list of watchpoints and
24906 breakpoints inserted.
24907
24908 @subsubheading @value{GDBN} Command
24909
24910 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
24911 @samp{rwatch}.
24912
24913 @subsubheading Example
24914
24915 Setting a watchpoint on a variable in the @code{main} function:
24916
24917 @smallexample
24918 (gdb)
24919 -break-watch x
24920 ^done,wpt=@{number="2",exp="x"@}
24921 (gdb)
24922 -exec-continue
24923 ^running
24924 (gdb)
24925 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
24926 value=@{old="-268439212",new="55"@},
24927 frame=@{func="main",args=[],file="recursive2.c",
24928 fullname="/home/foo/bar/recursive2.c",line="5"@}
24929 (gdb)
24930 @end smallexample
24931
24932 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
24933 the program execution twice: first for the variable changing value, then
24934 for the watchpoint going out of scope.
24935
24936 @smallexample
24937 (gdb)
24938 -break-watch C
24939 ^done,wpt=@{number="5",exp="C"@}
24940 (gdb)
24941 -exec-continue
24942 ^running
24943 (gdb)
24944 *stopped,reason="watchpoint-trigger",
24945 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
24946 frame=@{func="callee4",args=[],
24947 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24948 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
24949 (gdb)
24950 -exec-continue
24951 ^running
24952 (gdb)
24953 *stopped,reason="watchpoint-scope",wpnum="5",
24954 frame=@{func="callee3",args=[@{name="strarg",
24955 value="0x11940 \"A string argument.\""@}],
24956 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24957 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
24958 (gdb)
24959 @end smallexample
24960
24961 Listing breakpoints and watchpoints, at different points in the program
24962 execution. Note that once the watchpoint goes out of scope, it is
24963 deleted.
24964
24965 @smallexample
24966 (gdb)
24967 -break-watch C
24968 ^done,wpt=@{number="2",exp="C"@}
24969 (gdb)
24970 -break-list
24971 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
24972 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
24973 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
24974 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
24975 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
24976 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
24977 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
24978 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
24979 addr="0x00010734",func="callee4",
24980 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24981 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
24982 bkpt=@{number="2",type="watchpoint",disp="keep",
24983 enabled="y",addr="",what="C",times="0"@}]@}
24984 (gdb)
24985 -exec-continue
24986 ^running
24987 (gdb)
24988 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
24989 value=@{old="-276895068",new="3"@},
24990 frame=@{func="callee4",args=[],
24991 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
24992 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
24993 (gdb)
24994 -break-list
24995 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
24996 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
24997 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
24998 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
24999 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25000 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25001 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25002 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
25003 addr="0x00010734",func="callee4",
25004 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25005 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
25006 bkpt=@{number="2",type="watchpoint",disp="keep",
25007 enabled="y",addr="",what="C",times="-5"@}]@}
25008 (gdb)
25009 -exec-continue
25010 ^running
25011 ^done,reason="watchpoint-scope",wpnum="2",
25012 frame=@{func="callee3",args=[@{name="strarg",
25013 value="0x11940 \"A string argument.\""@}],
25014 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25015 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
25016 (gdb)
25017 -break-list
25018 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
25019 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25020 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25021 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25022 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25023 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25024 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25025 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
25026 addr="0x00010734",func="callee4",
25027 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25028 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
25029 times="1"@}]@}
25030 (gdb)
25031 @end smallexample
25032
25033 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25034 @node GDB/MI Program Context
25035 @section @sc{gdb/mi} Program Context
25036
25037 @subheading The @code{-exec-arguments} Command
25038 @findex -exec-arguments
25039
25040
25041 @subsubheading Synopsis
25042
25043 @smallexample
25044 -exec-arguments @var{args}
25045 @end smallexample
25046
25047 Set the inferior program arguments, to be used in the next
25048 @samp{-exec-run}.
25049
25050 @subsubheading @value{GDBN} Command
25051
25052 The corresponding @value{GDBN} command is @samp{set args}.
25053
25054 @subsubheading Example
25055
25056 @smallexample
25057 (gdb)
25058 -exec-arguments -v word
25059 ^done
25060 (gdb)
25061 @end smallexample
25062
25063
25064 @ignore
25065 @subheading The @code{-exec-show-arguments} Command
25066 @findex -exec-show-arguments
25067
25068 @subsubheading Synopsis
25069
25070 @smallexample
25071 -exec-show-arguments
25072 @end smallexample
25073
25074 Print the arguments of the program.
25075
25076 @subsubheading @value{GDBN} Command
25077
25078 The corresponding @value{GDBN} command is @samp{show args}.
25079
25080 @subsubheading Example
25081 N.A.
25082 @end ignore
25083
25084
25085 @subheading The @code{-environment-cd} Command
25086 @findex -environment-cd
25087
25088 @subsubheading Synopsis
25089
25090 @smallexample
25091 -environment-cd @var{pathdir}
25092 @end smallexample
25093
25094 Set @value{GDBN}'s working directory.
25095
25096 @subsubheading @value{GDBN} Command
25097
25098 The corresponding @value{GDBN} command is @samp{cd}.
25099
25100 @subsubheading Example
25101
25102 @smallexample
25103 (gdb)
25104 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
25105 ^done
25106 (gdb)
25107 @end smallexample
25108
25109
25110 @subheading The @code{-environment-directory} Command
25111 @findex -environment-directory
25112
25113 @subsubheading Synopsis
25114
25115 @smallexample
25116 -environment-directory [ -r ] [ @var{pathdir} ]+
25117 @end smallexample
25118
25119 Add directories @var{pathdir} to beginning of search path for source files.
25120 If the @samp{-r} option is used, the search path is reset to the default
25121 search path. If directories @var{pathdir} are supplied in addition to the
25122 @samp{-r} option, the search path is first reset and then addition
25123 occurs as normal.
25124 Multiple directories may be specified, separated by blanks. Specifying
25125 multiple directories in a single command
25126 results in the directories added to the beginning of the
25127 search path in the same order they were presented in the command.
25128 If blanks are needed as
25129 part of a directory name, double-quotes should be used around
25130 the name. In the command output, the path will show up separated
25131 by the system directory-separator character. The directory-separator
25132 character must not be used
25133 in any directory name.
25134 If no directories are specified, the current search path is displayed.
25135
25136 @subsubheading @value{GDBN} Command
25137
25138 The corresponding @value{GDBN} command is @samp{dir}.
25139
25140 @subsubheading Example
25141
25142 @smallexample
25143 (gdb)
25144 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
25145 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
25146 (gdb)
25147 -environment-directory ""
25148 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
25149 (gdb)
25150 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
25151 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
25152 (gdb)
25153 -environment-directory -r
25154 ^done,source-path="$cdir:$cwd"
25155 (gdb)
25156 @end smallexample
25157
25158
25159 @subheading The @code{-environment-path} Command
25160 @findex -environment-path
25161
25162 @subsubheading Synopsis
25163
25164 @smallexample
25165 -environment-path [ -r ] [ @var{pathdir} ]+
25166 @end smallexample
25167
25168 Add directories @var{pathdir} to beginning of search path for object files.
25169 If the @samp{-r} option is used, the search path is reset to the original
25170 search path that existed at gdb start-up. If directories @var{pathdir} are
25171 supplied in addition to the
25172 @samp{-r} option, the search path is first reset and then addition
25173 occurs as normal.
25174 Multiple directories may be specified, separated by blanks. Specifying
25175 multiple directories in a single command
25176 results in the directories added to the beginning of the
25177 search path in the same order they were presented in the command.
25178 If blanks are needed as
25179 part of a directory name, double-quotes should be used around
25180 the name. In the command output, the path will show up separated
25181 by the system directory-separator character. The directory-separator
25182 character must not be used
25183 in any directory name.
25184 If no directories are specified, the current path is displayed.
25185
25186
25187 @subsubheading @value{GDBN} Command
25188
25189 The corresponding @value{GDBN} command is @samp{path}.
25190
25191 @subsubheading Example
25192
25193 @smallexample
25194 (gdb)
25195 -environment-path
25196 ^done,path="/usr/bin"
25197 (gdb)
25198 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
25199 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
25200 (gdb)
25201 -environment-path -r /usr/local/bin
25202 ^done,path="/usr/local/bin:/usr/bin"
25203 (gdb)
25204 @end smallexample
25205
25206
25207 @subheading The @code{-environment-pwd} Command
25208 @findex -environment-pwd
25209
25210 @subsubheading Synopsis
25211
25212 @smallexample
25213 -environment-pwd
25214 @end smallexample
25215
25216 Show the current working directory.
25217
25218 @subsubheading @value{GDBN} Command
25219
25220 The corresponding @value{GDBN} command is @samp{pwd}.
25221
25222 @subsubheading Example
25223
25224 @smallexample
25225 (gdb)
25226 -environment-pwd
25227 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
25228 (gdb)
25229 @end smallexample
25230
25231 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25232 @node GDB/MI Thread Commands
25233 @section @sc{gdb/mi} Thread Commands
25234
25235
25236 @subheading The @code{-thread-info} Command
25237 @findex -thread-info
25238
25239 @subsubheading Synopsis
25240
25241 @smallexample
25242 -thread-info [ @var{thread-id} ]
25243 @end smallexample
25244
25245 Reports information about either a specific thread, if
25246 the @var{thread-id} parameter is present, or about all
25247 threads. When printing information about all threads,
25248 also reports the current thread.
25249
25250 @subsubheading @value{GDBN} Command
25251
25252 The @samp{info thread} command prints the same information
25253 about all threads.
25254
25255 @subsubheading Example
25256
25257 @smallexample
25258 -thread-info
25259 ^done,threads=[
25260 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
25261 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
25262 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
25263 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
25264 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}],
25265 current-thread-id="1"
25266 (gdb)
25267 @end smallexample
25268
25269 The @samp{state} field may have the following values:
25270
25271 @table @code
25272 @item stopped
25273 The thread is stopped. Frame information is available for stopped
25274 threads.
25275
25276 @item running
25277 The thread is running. There's no frame information for running
25278 threads.
25279
25280 @end table
25281
25282 @subheading The @code{-thread-list-ids} Command
25283 @findex -thread-list-ids
25284
25285 @subsubheading Synopsis
25286
25287 @smallexample
25288 -thread-list-ids
25289 @end smallexample
25290
25291 Produces a list of the currently known @value{GDBN} thread ids. At the
25292 end of the list it also prints the total number of such threads.
25293
25294 This command is retained for historical reasons, the
25295 @code{-thread-info} command should be used instead.
25296
25297 @subsubheading @value{GDBN} Command
25298
25299 Part of @samp{info threads} supplies the same information.
25300
25301 @subsubheading Example
25302
25303 @smallexample
25304 (gdb)
25305 -thread-list-ids
25306 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
25307 current-thread-id="1",number-of-threads="3"
25308 (gdb)
25309 @end smallexample
25310
25311
25312 @subheading The @code{-thread-select} Command
25313 @findex -thread-select
25314
25315 @subsubheading Synopsis
25316
25317 @smallexample
25318 -thread-select @var{threadnum}
25319 @end smallexample
25320
25321 Make @var{threadnum} the current thread. It prints the number of the new
25322 current thread, and the topmost frame for that thread.
25323
25324 This command is deprecated in favor of explicitly using the
25325 @samp{--thread} option to each command.
25326
25327 @subsubheading @value{GDBN} Command
25328
25329 The corresponding @value{GDBN} command is @samp{thread}.
25330
25331 @subsubheading Example
25332
25333 @smallexample
25334 (gdb)
25335 -exec-next
25336 ^running
25337 (gdb)
25338 *stopped,reason="end-stepping-range",thread-id="2",line="187",
25339 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
25340 (gdb)
25341 -thread-list-ids
25342 ^done,
25343 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
25344 number-of-threads="3"
25345 (gdb)
25346 -thread-select 3
25347 ^done,new-thread-id="3",
25348 frame=@{level="0",func="vprintf",
25349 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
25350 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
25351 (gdb)
25352 @end smallexample
25353
25354 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25355 @node GDB/MI Program Execution
25356 @section @sc{gdb/mi} Program Execution
25357
25358 These are the asynchronous commands which generate the out-of-band
25359 record @samp{*stopped}. Currently @value{GDBN} only really executes
25360 asynchronously with remote targets and this interaction is mimicked in
25361 other cases.
25362
25363 @subheading The @code{-exec-continue} Command
25364 @findex -exec-continue
25365
25366 @subsubheading Synopsis
25367
25368 @smallexample
25369 -exec-continue [--reverse] [--all|--thread-group N]
25370 @end smallexample
25371
25372 Resumes the execution of the inferior program, which will continue
25373 to execute until it reaches a debugger stop event. If the
25374 @samp{--reverse} option is specified, execution resumes in reverse until
25375 it reaches a stop event. Stop events may include
25376 @itemize @bullet
25377 @item
25378 breakpoints or watchpoints
25379 @item
25380 signals or exceptions
25381 @item
25382 the end of the process (or its beginning under @samp{--reverse})
25383 @item
25384 the end or beginning of a replay log if one is being used.
25385 @end itemize
25386 In all-stop mode (@pxref{All-Stop
25387 Mode}), may resume only one thread, or all threads, depending on the
25388 value of the @samp{scheduler-locking} variable. If @samp{--all} is
25389 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
25390 ignored in all-stop mode. If the @samp{--thread-group} options is
25391 specified, then all threads in that thread group are resumed.
25392
25393 @subsubheading @value{GDBN} Command
25394
25395 The corresponding @value{GDBN} corresponding is @samp{continue}.
25396
25397 @subsubheading Example
25398
25399 @smallexample
25400 -exec-continue
25401 ^running
25402 (gdb)
25403 @@Hello world
25404 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
25405 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
25406 line="13"@}
25407 (gdb)
25408 @end smallexample
25409
25410
25411 @subheading The @code{-exec-finish} Command
25412 @findex -exec-finish
25413
25414 @subsubheading Synopsis
25415
25416 @smallexample
25417 -exec-finish [--reverse]
25418 @end smallexample
25419
25420 Resumes the execution of the inferior program until the current
25421 function is exited. Displays the results returned by the function.
25422 If the @samp{--reverse} option is specified, resumes the reverse
25423 execution of the inferior program until the point where current
25424 function was called.
25425
25426 @subsubheading @value{GDBN} Command
25427
25428 The corresponding @value{GDBN} command is @samp{finish}.
25429
25430 @subsubheading Example
25431
25432 Function returning @code{void}.
25433
25434 @smallexample
25435 -exec-finish
25436 ^running
25437 (gdb)
25438 @@hello from foo
25439 *stopped,reason="function-finished",frame=@{func="main",args=[],
25440 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
25441 (gdb)
25442 @end smallexample
25443
25444 Function returning other than @code{void}. The name of the internal
25445 @value{GDBN} variable storing the result is printed, together with the
25446 value itself.
25447
25448 @smallexample
25449 -exec-finish
25450 ^running
25451 (gdb)
25452 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
25453 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
25454 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
25455 gdb-result-var="$1",return-value="0"
25456 (gdb)
25457 @end smallexample
25458
25459
25460 @subheading The @code{-exec-interrupt} Command
25461 @findex -exec-interrupt
25462
25463 @subsubheading Synopsis
25464
25465 @smallexample
25466 -exec-interrupt [--all|--thread-group N]
25467 @end smallexample
25468
25469 Interrupts the background execution of the target. Note how the token
25470 associated with the stop message is the one for the execution command
25471 that has been interrupted. The token for the interrupt itself only
25472 appears in the @samp{^done} output. If the user is trying to
25473 interrupt a non-running program, an error message will be printed.
25474
25475 Note that when asynchronous execution is enabled, this command is
25476 asynchronous just like other execution commands. That is, first the
25477 @samp{^done} response will be printed, and the target stop will be
25478 reported after that using the @samp{*stopped} notification.
25479
25480 In non-stop mode, only the context thread is interrupted by default.
25481 All threads (in all inferiors) will be interrupted if the
25482 @samp{--all} option is specified. If the @samp{--thread-group}
25483 option is specified, all threads in that group will be interrupted.
25484
25485 @subsubheading @value{GDBN} Command
25486
25487 The corresponding @value{GDBN} command is @samp{interrupt}.
25488
25489 @subsubheading Example
25490
25491 @smallexample
25492 (gdb)
25493 111-exec-continue
25494 111^running
25495
25496 (gdb)
25497 222-exec-interrupt
25498 222^done
25499 (gdb)
25500 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
25501 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
25502 fullname="/home/foo/bar/try.c",line="13"@}
25503 (gdb)
25504
25505 (gdb)
25506 -exec-interrupt
25507 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
25508 (gdb)
25509 @end smallexample
25510
25511 @subheading The @code{-exec-jump} Command
25512 @findex -exec-jump
25513
25514 @subsubheading Synopsis
25515
25516 @smallexample
25517 -exec-jump @var{location}
25518 @end smallexample
25519
25520 Resumes execution of the inferior program at the location specified by
25521 parameter. @xref{Specify Location}, for a description of the
25522 different forms of @var{location}.
25523
25524 @subsubheading @value{GDBN} Command
25525
25526 The corresponding @value{GDBN} command is @samp{jump}.
25527
25528 @subsubheading Example
25529
25530 @smallexample
25531 -exec-jump foo.c:10
25532 *running,thread-id="all"
25533 ^running
25534 @end smallexample
25535
25536
25537 @subheading The @code{-exec-next} Command
25538 @findex -exec-next
25539
25540 @subsubheading Synopsis
25541
25542 @smallexample
25543 -exec-next [--reverse]
25544 @end smallexample
25545
25546 Resumes execution of the inferior program, stopping when the beginning
25547 of the next source line is reached.
25548
25549 If the @samp{--reverse} option is specified, resumes reverse execution
25550 of the inferior program, stopping at the beginning of the previous
25551 source line. If you issue this command on the first line of a
25552 function, it will take you back to the caller of that function, to the
25553 source line where the function was called.
25554
25555
25556 @subsubheading @value{GDBN} Command
25557
25558 The corresponding @value{GDBN} command is @samp{next}.
25559
25560 @subsubheading Example
25561
25562 @smallexample
25563 -exec-next
25564 ^running
25565 (gdb)
25566 *stopped,reason="end-stepping-range",line="8",file="hello.c"
25567 (gdb)
25568 @end smallexample
25569
25570
25571 @subheading The @code{-exec-next-instruction} Command
25572 @findex -exec-next-instruction
25573
25574 @subsubheading Synopsis
25575
25576 @smallexample
25577 -exec-next-instruction [--reverse]
25578 @end smallexample
25579
25580 Executes one machine instruction. If the instruction is a function
25581 call, continues until the function returns. If the program stops at an
25582 instruction in the middle of a source line, the address will be
25583 printed as well.
25584
25585 If the @samp{--reverse} option is specified, resumes reverse execution
25586 of the inferior program, stopping at the previous instruction. If the
25587 previously executed instruction was a return from another function,
25588 it will continue to execute in reverse until the call to that function
25589 (from the current stack frame) is reached.
25590
25591 @subsubheading @value{GDBN} Command
25592
25593 The corresponding @value{GDBN} command is @samp{nexti}.
25594
25595 @subsubheading Example
25596
25597 @smallexample
25598 (gdb)
25599 -exec-next-instruction
25600 ^running
25601
25602 (gdb)
25603 *stopped,reason="end-stepping-range",
25604 addr="0x000100d4",line="5",file="hello.c"
25605 (gdb)
25606 @end smallexample
25607
25608
25609 @subheading The @code{-exec-return} Command
25610 @findex -exec-return
25611
25612 @subsubheading Synopsis
25613
25614 @smallexample
25615 -exec-return
25616 @end smallexample
25617
25618 Makes current function return immediately. Doesn't execute the inferior.
25619 Displays the new current frame.
25620
25621 @subsubheading @value{GDBN} Command
25622
25623 The corresponding @value{GDBN} command is @samp{return}.
25624
25625 @subsubheading Example
25626
25627 @smallexample
25628 (gdb)
25629 200-break-insert callee4
25630 200^done,bkpt=@{number="1",addr="0x00010734",
25631 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
25632 (gdb)
25633 000-exec-run
25634 000^running
25635 (gdb)
25636 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
25637 frame=@{func="callee4",args=[],
25638 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25639 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
25640 (gdb)
25641 205-break-delete
25642 205^done
25643 (gdb)
25644 111-exec-return
25645 111^done,frame=@{level="0",func="callee3",
25646 args=[@{name="strarg",
25647 value="0x11940 \"A string argument.\""@}],
25648 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25649 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
25650 (gdb)
25651 @end smallexample
25652
25653
25654 @subheading The @code{-exec-run} Command
25655 @findex -exec-run
25656
25657 @subsubheading Synopsis
25658
25659 @smallexample
25660 -exec-run [--all | --thread-group N]
25661 @end smallexample
25662
25663 Starts execution of the inferior from the beginning. The inferior
25664 executes until either a breakpoint is encountered or the program
25665 exits. In the latter case the output will include an exit code, if
25666 the program has exited exceptionally.
25667
25668 When no option is specified, the current inferior is started. If the
25669 @samp{--thread-group} option is specified, it should refer to a thread
25670 group of type @samp{process}, and that thread group will be started.
25671 If the @samp{--all} option is specified, then all inferiors will be started.
25672
25673 @subsubheading @value{GDBN} Command
25674
25675 The corresponding @value{GDBN} command is @samp{run}.
25676
25677 @subsubheading Examples
25678
25679 @smallexample
25680 (gdb)
25681 -break-insert main
25682 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
25683 (gdb)
25684 -exec-run
25685 ^running
25686 (gdb)
25687 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
25688 frame=@{func="main",args=[],file="recursive2.c",
25689 fullname="/home/foo/bar/recursive2.c",line="4"@}
25690 (gdb)
25691 @end smallexample
25692
25693 @noindent
25694 Program exited normally:
25695
25696 @smallexample
25697 (gdb)
25698 -exec-run
25699 ^running
25700 (gdb)
25701 x = 55
25702 *stopped,reason="exited-normally"
25703 (gdb)
25704 @end smallexample
25705
25706 @noindent
25707 Program exited exceptionally:
25708
25709 @smallexample
25710 (gdb)
25711 -exec-run
25712 ^running
25713 (gdb)
25714 x = 55
25715 *stopped,reason="exited",exit-code="01"
25716 (gdb)
25717 @end smallexample
25718
25719 Another way the program can terminate is if it receives a signal such as
25720 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
25721
25722 @smallexample
25723 (gdb)
25724 *stopped,reason="exited-signalled",signal-name="SIGINT",
25725 signal-meaning="Interrupt"
25726 @end smallexample
25727
25728
25729 @c @subheading -exec-signal
25730
25731
25732 @subheading The @code{-exec-step} Command
25733 @findex -exec-step
25734
25735 @subsubheading Synopsis
25736
25737 @smallexample
25738 -exec-step [--reverse]
25739 @end smallexample
25740
25741 Resumes execution of the inferior program, stopping when the beginning
25742 of the next source line is reached, if the next source line is not a
25743 function call. If it is, stop at the first instruction of the called
25744 function. If the @samp{--reverse} option is specified, resumes reverse
25745 execution of the inferior program, stopping at the beginning of the
25746 previously executed source line.
25747
25748 @subsubheading @value{GDBN} Command
25749
25750 The corresponding @value{GDBN} command is @samp{step}.
25751
25752 @subsubheading Example
25753
25754 Stepping into a function:
25755
25756 @smallexample
25757 -exec-step
25758 ^running
25759 (gdb)
25760 *stopped,reason="end-stepping-range",
25761 frame=@{func="foo",args=[@{name="a",value="10"@},
25762 @{name="b",value="0"@}],file="recursive2.c",
25763 fullname="/home/foo/bar/recursive2.c",line="11"@}
25764 (gdb)
25765 @end smallexample
25766
25767 Regular stepping:
25768
25769 @smallexample
25770 -exec-step
25771 ^running
25772 (gdb)
25773 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
25774 (gdb)
25775 @end smallexample
25776
25777
25778 @subheading The @code{-exec-step-instruction} Command
25779 @findex -exec-step-instruction
25780
25781 @subsubheading Synopsis
25782
25783 @smallexample
25784 -exec-step-instruction [--reverse]
25785 @end smallexample
25786
25787 Resumes the inferior which executes one machine instruction. If the
25788 @samp{--reverse} option is specified, resumes reverse execution of the
25789 inferior program, stopping at the previously executed instruction.
25790 The output, once @value{GDBN} has stopped, will vary depending on
25791 whether we have stopped in the middle of a source line or not. In the
25792 former case, the address at which the program stopped will be printed
25793 as well.
25794
25795 @subsubheading @value{GDBN} Command
25796
25797 The corresponding @value{GDBN} command is @samp{stepi}.
25798
25799 @subsubheading Example
25800
25801 @smallexample
25802 (gdb)
25803 -exec-step-instruction
25804 ^running
25805
25806 (gdb)
25807 *stopped,reason="end-stepping-range",
25808 frame=@{func="foo",args=[],file="try.c",
25809 fullname="/home/foo/bar/try.c",line="10"@}
25810 (gdb)
25811 -exec-step-instruction
25812 ^running
25813
25814 (gdb)
25815 *stopped,reason="end-stepping-range",
25816 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
25817 fullname="/home/foo/bar/try.c",line="10"@}
25818 (gdb)
25819 @end smallexample
25820
25821
25822 @subheading The @code{-exec-until} Command
25823 @findex -exec-until
25824
25825 @subsubheading Synopsis
25826
25827 @smallexample
25828 -exec-until [ @var{location} ]
25829 @end smallexample
25830
25831 Executes the inferior until the @var{location} specified in the
25832 argument is reached. If there is no argument, the inferior executes
25833 until a source line greater than the current one is reached. The
25834 reason for stopping in this case will be @samp{location-reached}.
25835
25836 @subsubheading @value{GDBN} Command
25837
25838 The corresponding @value{GDBN} command is @samp{until}.
25839
25840 @subsubheading Example
25841
25842 @smallexample
25843 (gdb)
25844 -exec-until recursive2.c:6
25845 ^running
25846 (gdb)
25847 x = 55
25848 *stopped,reason="location-reached",frame=@{func="main",args=[],
25849 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
25850 (gdb)
25851 @end smallexample
25852
25853 @ignore
25854 @subheading -file-clear
25855 Is this going away????
25856 @end ignore
25857
25858 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25859 @node GDB/MI Stack Manipulation
25860 @section @sc{gdb/mi} Stack Manipulation Commands
25861
25862
25863 @subheading The @code{-stack-info-frame} Command
25864 @findex -stack-info-frame
25865
25866 @subsubheading Synopsis
25867
25868 @smallexample
25869 -stack-info-frame
25870 @end smallexample
25871
25872 Get info on the selected frame.
25873
25874 @subsubheading @value{GDBN} Command
25875
25876 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
25877 (without arguments).
25878
25879 @subsubheading Example
25880
25881 @smallexample
25882 (gdb)
25883 -stack-info-frame
25884 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
25885 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25886 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
25887 (gdb)
25888 @end smallexample
25889
25890 @subheading The @code{-stack-info-depth} Command
25891 @findex -stack-info-depth
25892
25893 @subsubheading Synopsis
25894
25895 @smallexample
25896 -stack-info-depth [ @var{max-depth} ]
25897 @end smallexample
25898
25899 Return the depth of the stack. If the integer argument @var{max-depth}
25900 is specified, do not count beyond @var{max-depth} frames.
25901
25902 @subsubheading @value{GDBN} Command
25903
25904 There's no equivalent @value{GDBN} command.
25905
25906 @subsubheading Example
25907
25908 For a stack with frame levels 0 through 11:
25909
25910 @smallexample
25911 (gdb)
25912 -stack-info-depth
25913 ^done,depth="12"
25914 (gdb)
25915 -stack-info-depth 4
25916 ^done,depth="4"
25917 (gdb)
25918 -stack-info-depth 12
25919 ^done,depth="12"
25920 (gdb)
25921 -stack-info-depth 11
25922 ^done,depth="11"
25923 (gdb)
25924 -stack-info-depth 13
25925 ^done,depth="12"
25926 (gdb)
25927 @end smallexample
25928
25929 @subheading The @code{-stack-list-arguments} Command
25930 @findex -stack-list-arguments
25931
25932 @subsubheading Synopsis
25933
25934 @smallexample
25935 -stack-list-arguments @var{print-values}
25936 [ @var{low-frame} @var{high-frame} ]
25937 @end smallexample
25938
25939 Display a list of the arguments for the frames between @var{low-frame}
25940 and @var{high-frame} (inclusive). If @var{low-frame} and
25941 @var{high-frame} are not provided, list the arguments for the whole
25942 call stack. If the two arguments are equal, show the single frame
25943 at the corresponding level. It is an error if @var{low-frame} is
25944 larger than the actual number of frames. On the other hand,
25945 @var{high-frame} may be larger than the actual number of frames, in
25946 which case only existing frames will be returned.
25947
25948 If @var{print-values} is 0 or @code{--no-values}, print only the names of
25949 the variables; if it is 1 or @code{--all-values}, print also their
25950 values; and if it is 2 or @code{--simple-values}, print the name,
25951 type and value for simple data types, and the name and type for arrays,
25952 structures and unions.
25953
25954 Use of this command to obtain arguments in a single frame is
25955 deprecated in favor of the @samp{-stack-list-variables} command.
25956
25957 @subsubheading @value{GDBN} Command
25958
25959 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
25960 @samp{gdb_get_args} command which partially overlaps with the
25961 functionality of @samp{-stack-list-arguments}.
25962
25963 @subsubheading Example
25964
25965 @smallexample
25966 (gdb)
25967 -stack-list-frames
25968 ^done,
25969 stack=[
25970 frame=@{level="0",addr="0x00010734",func="callee4",
25971 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25972 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
25973 frame=@{level="1",addr="0x0001076c",func="callee3",
25974 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25975 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
25976 frame=@{level="2",addr="0x0001078c",func="callee2",
25977 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25978 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
25979 frame=@{level="3",addr="0x000107b4",func="callee1",
25980 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25981 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
25982 frame=@{level="4",addr="0x000107e0",func="main",
25983 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25984 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
25985 (gdb)
25986 -stack-list-arguments 0
25987 ^done,
25988 stack-args=[
25989 frame=@{level="0",args=[]@},
25990 frame=@{level="1",args=[name="strarg"]@},
25991 frame=@{level="2",args=[name="intarg",name="strarg"]@},
25992 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
25993 frame=@{level="4",args=[]@}]
25994 (gdb)
25995 -stack-list-arguments 1
25996 ^done,
25997 stack-args=[
25998 frame=@{level="0",args=[]@},
25999 frame=@{level="1",
26000 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
26001 frame=@{level="2",args=[
26002 @{name="intarg",value="2"@},
26003 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
26004 @{frame=@{level="3",args=[
26005 @{name="intarg",value="2"@},
26006 @{name="strarg",value="0x11940 \"A string argument.\""@},
26007 @{name="fltarg",value="3.5"@}]@},
26008 frame=@{level="4",args=[]@}]
26009 (gdb)
26010 -stack-list-arguments 0 2 2
26011 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
26012 (gdb)
26013 -stack-list-arguments 1 2 2
26014 ^done,stack-args=[frame=@{level="2",
26015 args=[@{name="intarg",value="2"@},
26016 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
26017 (gdb)
26018 @end smallexample
26019
26020 @c @subheading -stack-list-exception-handlers
26021
26022
26023 @subheading The @code{-stack-list-frames} Command
26024 @findex -stack-list-frames
26025
26026 @subsubheading Synopsis
26027
26028 @smallexample
26029 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
26030 @end smallexample
26031
26032 List the frames currently on the stack. For each frame it displays the
26033 following info:
26034
26035 @table @samp
26036 @item @var{level}
26037 The frame number, 0 being the topmost frame, i.e., the innermost function.
26038 @item @var{addr}
26039 The @code{$pc} value for that frame.
26040 @item @var{func}
26041 Function name.
26042 @item @var{file}
26043 File name of the source file where the function lives.
26044 @item @var{line}
26045 Line number corresponding to the @code{$pc}.
26046 @end table
26047
26048 If invoked without arguments, this command prints a backtrace for the
26049 whole stack. If given two integer arguments, it shows the frames whose
26050 levels are between the two arguments (inclusive). If the two arguments
26051 are equal, it shows the single frame at the corresponding level. It is
26052 an error if @var{low-frame} is larger than the actual number of
26053 frames. On the other hand, @var{high-frame} may be larger than the
26054 actual number of frames, in which case only existing frames will be returned.
26055
26056 @subsubheading @value{GDBN} Command
26057
26058 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
26059
26060 @subsubheading Example
26061
26062 Full stack backtrace:
26063
26064 @smallexample
26065 (gdb)
26066 -stack-list-frames
26067 ^done,stack=
26068 [frame=@{level="0",addr="0x0001076c",func="foo",
26069 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
26070 frame=@{level="1",addr="0x000107a4",func="foo",
26071 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26072 frame=@{level="2",addr="0x000107a4",func="foo",
26073 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26074 frame=@{level="3",addr="0x000107a4",func="foo",
26075 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26076 frame=@{level="4",addr="0x000107a4",func="foo",
26077 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26078 frame=@{level="5",addr="0x000107a4",func="foo",
26079 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26080 frame=@{level="6",addr="0x000107a4",func="foo",
26081 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26082 frame=@{level="7",addr="0x000107a4",func="foo",
26083 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26084 frame=@{level="8",addr="0x000107a4",func="foo",
26085 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26086 frame=@{level="9",addr="0x000107a4",func="foo",
26087 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26088 frame=@{level="10",addr="0x000107a4",func="foo",
26089 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26090 frame=@{level="11",addr="0x00010738",func="main",
26091 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
26092 (gdb)
26093 @end smallexample
26094
26095 Show frames between @var{low_frame} and @var{high_frame}:
26096
26097 @smallexample
26098 (gdb)
26099 -stack-list-frames 3 5
26100 ^done,stack=
26101 [frame=@{level="3",addr="0x000107a4",func="foo",
26102 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26103 frame=@{level="4",addr="0x000107a4",func="foo",
26104 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26105 frame=@{level="5",addr="0x000107a4",func="foo",
26106 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
26107 (gdb)
26108 @end smallexample
26109
26110 Show a single frame:
26111
26112 @smallexample
26113 (gdb)
26114 -stack-list-frames 3 3
26115 ^done,stack=
26116 [frame=@{level="3",addr="0x000107a4",func="foo",
26117 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
26118 (gdb)
26119 @end smallexample
26120
26121
26122 @subheading The @code{-stack-list-locals} Command
26123 @findex -stack-list-locals
26124
26125 @subsubheading Synopsis
26126
26127 @smallexample
26128 -stack-list-locals @var{print-values}
26129 @end smallexample
26130
26131 Display the local variable names for the selected frame. If
26132 @var{print-values} is 0 or @code{--no-values}, print only the names of
26133 the variables; if it is 1 or @code{--all-values}, print also their
26134 values; and if it is 2 or @code{--simple-values}, print the name,
26135 type and value for simple data types, and the name and type for arrays,
26136 structures and unions. In this last case, a frontend can immediately
26137 display the value of simple data types and create variable objects for
26138 other data types when the user wishes to explore their values in
26139 more detail.
26140
26141 This command is deprecated in favor of the
26142 @samp{-stack-list-variables} command.
26143
26144 @subsubheading @value{GDBN} Command
26145
26146 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
26147
26148 @subsubheading Example
26149
26150 @smallexample
26151 (gdb)
26152 -stack-list-locals 0
26153 ^done,locals=[name="A",name="B",name="C"]
26154 (gdb)
26155 -stack-list-locals --all-values
26156 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
26157 @{name="C",value="@{1, 2, 3@}"@}]
26158 -stack-list-locals --simple-values
26159 ^done,locals=[@{name="A",type="int",value="1"@},
26160 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
26161 (gdb)
26162 @end smallexample
26163
26164 @subheading The @code{-stack-list-variables} Command
26165 @findex -stack-list-variables
26166
26167 @subsubheading Synopsis
26168
26169 @smallexample
26170 -stack-list-variables @var{print-values}
26171 @end smallexample
26172
26173 Display the names of local variables and function arguments for the selected frame. If
26174 @var{print-values} is 0 or @code{--no-values}, print only the names of
26175 the variables; if it is 1 or @code{--all-values}, print also their
26176 values; and if it is 2 or @code{--simple-values}, print the name,
26177 type and value for simple data types, and the name and type for arrays,
26178 structures and unions.
26179
26180 @subsubheading Example
26181
26182 @smallexample
26183 (gdb)
26184 -stack-list-variables --thread 1 --frame 0 --all-values
26185 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
26186 (gdb)
26187 @end smallexample
26188
26189
26190 @subheading The @code{-stack-select-frame} Command
26191 @findex -stack-select-frame
26192
26193 @subsubheading Synopsis
26194
26195 @smallexample
26196 -stack-select-frame @var{framenum}
26197 @end smallexample
26198
26199 Change the selected frame. Select a different frame @var{framenum} on
26200 the stack.
26201
26202 This command in deprecated in favor of passing the @samp{--frame}
26203 option to every command.
26204
26205 @subsubheading @value{GDBN} Command
26206
26207 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
26208 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
26209
26210 @subsubheading Example
26211
26212 @smallexample
26213 (gdb)
26214 -stack-select-frame 2
26215 ^done
26216 (gdb)
26217 @end smallexample
26218
26219 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26220 @node GDB/MI Variable Objects
26221 @section @sc{gdb/mi} Variable Objects
26222
26223 @ignore
26224
26225 @subheading Motivation for Variable Objects in @sc{gdb/mi}
26226
26227 For the implementation of a variable debugger window (locals, watched
26228 expressions, etc.), we are proposing the adaptation of the existing code
26229 used by @code{Insight}.
26230
26231 The two main reasons for that are:
26232
26233 @enumerate 1
26234 @item
26235 It has been proven in practice (it is already on its second generation).
26236
26237 @item
26238 It will shorten development time (needless to say how important it is
26239 now).
26240 @end enumerate
26241
26242 The original interface was designed to be used by Tcl code, so it was
26243 slightly changed so it could be used through @sc{gdb/mi}. This section
26244 describes the @sc{gdb/mi} operations that will be available and gives some
26245 hints about their use.
26246
26247 @emph{Note}: In addition to the set of operations described here, we
26248 expect the @sc{gui} implementation of a variable window to require, at
26249 least, the following operations:
26250
26251 @itemize @bullet
26252 @item @code{-gdb-show} @code{output-radix}
26253 @item @code{-stack-list-arguments}
26254 @item @code{-stack-list-locals}
26255 @item @code{-stack-select-frame}
26256 @end itemize
26257
26258 @end ignore
26259
26260 @subheading Introduction to Variable Objects
26261
26262 @cindex variable objects in @sc{gdb/mi}
26263
26264 Variable objects are "object-oriented" MI interface for examining and
26265 changing values of expressions. Unlike some other MI interfaces that
26266 work with expressions, variable objects are specifically designed for
26267 simple and efficient presentation in the frontend. A variable object
26268 is identified by string name. When a variable object is created, the
26269 frontend specifies the expression for that variable object. The
26270 expression can be a simple variable, or it can be an arbitrary complex
26271 expression, and can even involve CPU registers. After creating a
26272 variable object, the frontend can invoke other variable object
26273 operations---for example to obtain or change the value of a variable
26274 object, or to change display format.
26275
26276 Variable objects have hierarchical tree structure. Any variable object
26277 that corresponds to a composite type, such as structure in C, has
26278 a number of child variable objects, for example corresponding to each
26279 element of a structure. A child variable object can itself have
26280 children, recursively. Recursion ends when we reach
26281 leaf variable objects, which always have built-in types. Child variable
26282 objects are created only by explicit request, so if a frontend
26283 is not interested in the children of a particular variable object, no
26284 child will be created.
26285
26286 For a leaf variable object it is possible to obtain its value as a
26287 string, or set the value from a string. String value can be also
26288 obtained for a non-leaf variable object, but it's generally a string
26289 that only indicates the type of the object, and does not list its
26290 contents. Assignment to a non-leaf variable object is not allowed.
26291
26292 A frontend does not need to read the values of all variable objects each time
26293 the program stops. Instead, MI provides an update command that lists all
26294 variable objects whose values has changed since the last update
26295 operation. This considerably reduces the amount of data that must
26296 be transferred to the frontend. As noted above, children variable
26297 objects are created on demand, and only leaf variable objects have a
26298 real value. As result, gdb will read target memory only for leaf
26299 variables that frontend has created.
26300
26301 The automatic update is not always desirable. For example, a frontend
26302 might want to keep a value of some expression for future reference,
26303 and never update it. For another example, fetching memory is
26304 relatively slow for embedded targets, so a frontend might want
26305 to disable automatic update for the variables that are either not
26306 visible on the screen, or ``closed''. This is possible using so
26307 called ``frozen variable objects''. Such variable objects are never
26308 implicitly updated.
26309
26310 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
26311 fixed variable object, the expression is parsed when the variable
26312 object is created, including associating identifiers to specific
26313 variables. The meaning of expression never changes. For a floating
26314 variable object the values of variables whose names appear in the
26315 expressions are re-evaluated every time in the context of the current
26316 frame. Consider this example:
26317
26318 @smallexample
26319 void do_work(...)
26320 @{
26321 struct work_state state;
26322
26323 if (...)
26324 do_work(...);
26325 @}
26326 @end smallexample
26327
26328 If a fixed variable object for the @code{state} variable is created in
26329 this function, and we enter the recursive call, the the variable
26330 object will report the value of @code{state} in the top-level
26331 @code{do_work} invocation. On the other hand, a floating variable
26332 object will report the value of @code{state} in the current frame.
26333
26334 If an expression specified when creating a fixed variable object
26335 refers to a local variable, the variable object becomes bound to the
26336 thread and frame in which the variable object is created. When such
26337 variable object is updated, @value{GDBN} makes sure that the
26338 thread/frame combination the variable object is bound to still exists,
26339 and re-evaluates the variable object in context of that thread/frame.
26340
26341 The following is the complete set of @sc{gdb/mi} operations defined to
26342 access this functionality:
26343
26344 @multitable @columnfractions .4 .6
26345 @item @strong{Operation}
26346 @tab @strong{Description}
26347
26348 @item @code{-enable-pretty-printing}
26349 @tab enable Python-based pretty-printing
26350 @item @code{-var-create}
26351 @tab create a variable object
26352 @item @code{-var-delete}
26353 @tab delete the variable object and/or its children
26354 @item @code{-var-set-format}
26355 @tab set the display format of this variable
26356 @item @code{-var-show-format}
26357 @tab show the display format of this variable
26358 @item @code{-var-info-num-children}
26359 @tab tells how many children this object has
26360 @item @code{-var-list-children}
26361 @tab return a list of the object's children
26362 @item @code{-var-info-type}
26363 @tab show the type of this variable object
26364 @item @code{-var-info-expression}
26365 @tab print parent-relative expression that this variable object represents
26366 @item @code{-var-info-path-expression}
26367 @tab print full expression that this variable object represents
26368 @item @code{-var-show-attributes}
26369 @tab is this variable editable? does it exist here?
26370 @item @code{-var-evaluate-expression}
26371 @tab get the value of this variable
26372 @item @code{-var-assign}
26373 @tab set the value of this variable
26374 @item @code{-var-update}
26375 @tab update the variable and its children
26376 @item @code{-var-set-frozen}
26377 @tab set frozeness attribute
26378 @item @code{-var-set-update-range}
26379 @tab set range of children to display on update
26380 @end multitable
26381
26382 In the next subsection we describe each operation in detail and suggest
26383 how it can be used.
26384
26385 @subheading Description And Use of Operations on Variable Objects
26386
26387 @subheading The @code{-enable-pretty-printing} Command
26388 @findex -enable-pretty-printing
26389
26390 @smallexample
26391 -enable-pretty-printing
26392 @end smallexample
26393
26394 @value{GDBN} allows Python-based visualizers to affect the output of the
26395 MI variable object commands. However, because there was no way to
26396 implement this in a fully backward-compatible way, a front end must
26397 request that this functionality be enabled.
26398
26399 Once enabled, this feature cannot be disabled.
26400
26401 Note that if Python support has not been compiled into @value{GDBN},
26402 this command will still succeed (and do nothing).
26403
26404 This feature is currently (as of @value{GDBN} 7.0) experimental, and
26405 may work differently in future versions of @value{GDBN}.
26406
26407 @subheading The @code{-var-create} Command
26408 @findex -var-create
26409
26410 @subsubheading Synopsis
26411
26412 @smallexample
26413 -var-create @{@var{name} | "-"@}
26414 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
26415 @end smallexample
26416
26417 This operation creates a variable object, which allows the monitoring of
26418 a variable, the result of an expression, a memory cell or a CPU
26419 register.
26420
26421 The @var{name} parameter is the string by which the object can be
26422 referenced. It must be unique. If @samp{-} is specified, the varobj
26423 system will generate a string ``varNNNNNN'' automatically. It will be
26424 unique provided that one does not specify @var{name} of that format.
26425 The command fails if a duplicate name is found.
26426
26427 The frame under which the expression should be evaluated can be
26428 specified by @var{frame-addr}. A @samp{*} indicates that the current
26429 frame should be used. A @samp{@@} indicates that a floating variable
26430 object must be created.
26431
26432 @var{expression} is any expression valid on the current language set (must not
26433 begin with a @samp{*}), or one of the following:
26434
26435 @itemize @bullet
26436 @item
26437 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
26438
26439 @item
26440 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
26441
26442 @item
26443 @samp{$@var{regname}} --- a CPU register name
26444 @end itemize
26445
26446 @cindex dynamic varobj
26447 A varobj's contents may be provided by a Python-based pretty-printer. In this
26448 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
26449 have slightly different semantics in some cases. If the
26450 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
26451 will never create a dynamic varobj. This ensures backward
26452 compatibility for existing clients.
26453
26454 @subsubheading Result
26455
26456 This operation returns attributes of the newly-created varobj. These
26457 are:
26458
26459 @table @samp
26460 @item name
26461 The name of the varobj.
26462
26463 @item numchild
26464 The number of children of the varobj. This number is not necessarily
26465 reliable for a dynamic varobj. Instead, you must examine the
26466 @samp{has_more} attribute.
26467
26468 @item value
26469 The varobj's scalar value. For a varobj whose type is some sort of
26470 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
26471 will not be interesting.
26472
26473 @item type
26474 The varobj's type. This is a string representation of the type, as
26475 would be printed by the @value{GDBN} CLI.
26476
26477 @item thread-id
26478 If a variable object is bound to a specific thread, then this is the
26479 thread's identifier.
26480
26481 @item has_more
26482 For a dynamic varobj, this indicates whether there appear to be any
26483 children available. For a non-dynamic varobj, this will be 0.
26484
26485 @item dynamic
26486 This attribute will be present and have the value @samp{1} if the
26487 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
26488 then this attribute will not be present.
26489
26490 @item displayhint
26491 A dynamic varobj can supply a display hint to the front end. The
26492 value comes directly from the Python pretty-printer object's
26493 @code{display_hint} method. @xref{Pretty Printing API}.
26494 @end table
26495
26496 Typical output will look like this:
26497
26498 @smallexample
26499 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
26500 has_more="@var{has_more}"
26501 @end smallexample
26502
26503
26504 @subheading The @code{-var-delete} Command
26505 @findex -var-delete
26506
26507 @subsubheading Synopsis
26508
26509 @smallexample
26510 -var-delete [ -c ] @var{name}
26511 @end smallexample
26512
26513 Deletes a previously created variable object and all of its children.
26514 With the @samp{-c} option, just deletes the children.
26515
26516 Returns an error if the object @var{name} is not found.
26517
26518
26519 @subheading The @code{-var-set-format} Command
26520 @findex -var-set-format
26521
26522 @subsubheading Synopsis
26523
26524 @smallexample
26525 -var-set-format @var{name} @var{format-spec}
26526 @end smallexample
26527
26528 Sets the output format for the value of the object @var{name} to be
26529 @var{format-spec}.
26530
26531 @anchor{-var-set-format}
26532 The syntax for the @var{format-spec} is as follows:
26533
26534 @smallexample
26535 @var{format-spec} @expansion{}
26536 @{binary | decimal | hexadecimal | octal | natural@}
26537 @end smallexample
26538
26539 The natural format is the default format choosen automatically
26540 based on the variable type (like decimal for an @code{int}, hex
26541 for pointers, etc.).
26542
26543 For a variable with children, the format is set only on the
26544 variable itself, and the children are not affected.
26545
26546 @subheading The @code{-var-show-format} Command
26547 @findex -var-show-format
26548
26549 @subsubheading Synopsis
26550
26551 @smallexample
26552 -var-show-format @var{name}
26553 @end smallexample
26554
26555 Returns the format used to display the value of the object @var{name}.
26556
26557 @smallexample
26558 @var{format} @expansion{}
26559 @var{format-spec}
26560 @end smallexample
26561
26562
26563 @subheading The @code{-var-info-num-children} Command
26564 @findex -var-info-num-children
26565
26566 @subsubheading Synopsis
26567
26568 @smallexample
26569 -var-info-num-children @var{name}
26570 @end smallexample
26571
26572 Returns the number of children of a variable object @var{name}:
26573
26574 @smallexample
26575 numchild=@var{n}
26576 @end smallexample
26577
26578 Note that this number is not completely reliable for a dynamic varobj.
26579 It will return the current number of children, but more children may
26580 be available.
26581
26582
26583 @subheading The @code{-var-list-children} Command
26584 @findex -var-list-children
26585
26586 @subsubheading Synopsis
26587
26588 @smallexample
26589 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
26590 @end smallexample
26591 @anchor{-var-list-children}
26592
26593 Return a list of the children of the specified variable object and
26594 create variable objects for them, if they do not already exist. With
26595 a single argument or if @var{print-values} has a value of 0 or
26596 @code{--no-values}, print only the names of the variables; if
26597 @var{print-values} is 1 or @code{--all-values}, also print their
26598 values; and if it is 2 or @code{--simple-values} print the name and
26599 value for simple data types and just the name for arrays, structures
26600 and unions.
26601
26602 @var{from} and @var{to}, if specified, indicate the range of children
26603 to report. If @var{from} or @var{to} is less than zero, the range is
26604 reset and all children will be reported. Otherwise, children starting
26605 at @var{from} (zero-based) and up to and excluding @var{to} will be
26606 reported.
26607
26608 If a child range is requested, it will only affect the current call to
26609 @code{-var-list-children}, but not future calls to @code{-var-update}.
26610 For this, you must instead use @code{-var-set-update-range}. The
26611 intent of this approach is to enable a front end to implement any
26612 update approach it likes; for example, scrolling a view may cause the
26613 front end to request more children with @code{-var-list-children}, and
26614 then the front end could call @code{-var-set-update-range} with a
26615 different range to ensure that future updates are restricted to just
26616 the visible items.
26617
26618 For each child the following results are returned:
26619
26620 @table @var
26621
26622 @item name
26623 Name of the variable object created for this child.
26624
26625 @item exp
26626 The expression to be shown to the user by the front end to designate this child.
26627 For example this may be the name of a structure member.
26628
26629 For a dynamic varobj, this value cannot be used to form an
26630 expression. There is no way to do this at all with a dynamic varobj.
26631
26632 For C/C@t{++} structures there are several pseudo children returned to
26633 designate access qualifiers. For these pseudo children @var{exp} is
26634 @samp{public}, @samp{private}, or @samp{protected}. In this case the
26635 type and value are not present.
26636
26637 A dynamic varobj will not report the access qualifying
26638 pseudo-children, regardless of the language. This information is not
26639 available at all with a dynamic varobj.
26640
26641 @item numchild
26642 Number of children this child has. For a dynamic varobj, this will be
26643 0.
26644
26645 @item type
26646 The type of the child.
26647
26648 @item value
26649 If values were requested, this is the value.
26650
26651 @item thread-id
26652 If this variable object is associated with a thread, this is the thread id.
26653 Otherwise this result is not present.
26654
26655 @item frozen
26656 If the variable object is frozen, this variable will be present with a value of 1.
26657 @end table
26658
26659 The result may have its own attributes:
26660
26661 @table @samp
26662 @item displayhint
26663 A dynamic varobj can supply a display hint to the front end. The
26664 value comes directly from the Python pretty-printer object's
26665 @code{display_hint} method. @xref{Pretty Printing API}.
26666
26667 @item has_more
26668 This is an integer attribute which is nonzero if there are children
26669 remaining after the end of the selected range.
26670 @end table
26671
26672 @subsubheading Example
26673
26674 @smallexample
26675 (gdb)
26676 -var-list-children n
26677 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
26678 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
26679 (gdb)
26680 -var-list-children --all-values n
26681 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
26682 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
26683 @end smallexample
26684
26685
26686 @subheading The @code{-var-info-type} Command
26687 @findex -var-info-type
26688
26689 @subsubheading Synopsis
26690
26691 @smallexample
26692 -var-info-type @var{name}
26693 @end smallexample
26694
26695 Returns the type of the specified variable @var{name}. The type is
26696 returned as a string in the same format as it is output by the
26697 @value{GDBN} CLI:
26698
26699 @smallexample
26700 type=@var{typename}
26701 @end smallexample
26702
26703
26704 @subheading The @code{-var-info-expression} Command
26705 @findex -var-info-expression
26706
26707 @subsubheading Synopsis
26708
26709 @smallexample
26710 -var-info-expression @var{name}
26711 @end smallexample
26712
26713 Returns a string that is suitable for presenting this
26714 variable object in user interface. The string is generally
26715 not valid expression in the current language, and cannot be evaluated.
26716
26717 For example, if @code{a} is an array, and variable object
26718 @code{A} was created for @code{a}, then we'll get this output:
26719
26720 @smallexample
26721 (gdb) -var-info-expression A.1
26722 ^done,lang="C",exp="1"
26723 @end smallexample
26724
26725 @noindent
26726 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
26727
26728 Note that the output of the @code{-var-list-children} command also
26729 includes those expressions, so the @code{-var-info-expression} command
26730 is of limited use.
26731
26732 @subheading The @code{-var-info-path-expression} Command
26733 @findex -var-info-path-expression
26734
26735 @subsubheading Synopsis
26736
26737 @smallexample
26738 -var-info-path-expression @var{name}
26739 @end smallexample
26740
26741 Returns an expression that can be evaluated in the current
26742 context and will yield the same value that a variable object has.
26743 Compare this with the @code{-var-info-expression} command, which
26744 result can be used only for UI presentation. Typical use of
26745 the @code{-var-info-path-expression} command is creating a
26746 watchpoint from a variable object.
26747
26748 This command is currently not valid for children of a dynamic varobj,
26749 and will give an error when invoked on one.
26750
26751 For example, suppose @code{C} is a C@t{++} class, derived from class
26752 @code{Base}, and that the @code{Base} class has a member called
26753 @code{m_size}. Assume a variable @code{c} is has the type of
26754 @code{C} and a variable object @code{C} was created for variable
26755 @code{c}. Then, we'll get this output:
26756 @smallexample
26757 (gdb) -var-info-path-expression C.Base.public.m_size
26758 ^done,path_expr=((Base)c).m_size)
26759 @end smallexample
26760
26761 @subheading The @code{-var-show-attributes} Command
26762 @findex -var-show-attributes
26763
26764 @subsubheading Synopsis
26765
26766 @smallexample
26767 -var-show-attributes @var{name}
26768 @end smallexample
26769
26770 List attributes of the specified variable object @var{name}:
26771
26772 @smallexample
26773 status=@var{attr} [ ( ,@var{attr} )* ]
26774 @end smallexample
26775
26776 @noindent
26777 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
26778
26779 @subheading The @code{-var-evaluate-expression} Command
26780 @findex -var-evaluate-expression
26781
26782 @subsubheading Synopsis
26783
26784 @smallexample
26785 -var-evaluate-expression [-f @var{format-spec}] @var{name}
26786 @end smallexample
26787
26788 Evaluates the expression that is represented by the specified variable
26789 object and returns its value as a string. The format of the string
26790 can be specified with the @samp{-f} option. The possible values of
26791 this option are the same as for @code{-var-set-format}
26792 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
26793 the current display format will be used. The current display format
26794 can be changed using the @code{-var-set-format} command.
26795
26796 @smallexample
26797 value=@var{value}
26798 @end smallexample
26799
26800 Note that one must invoke @code{-var-list-children} for a variable
26801 before the value of a child variable can be evaluated.
26802
26803 @subheading The @code{-var-assign} Command
26804 @findex -var-assign
26805
26806 @subsubheading Synopsis
26807
26808 @smallexample
26809 -var-assign @var{name} @var{expression}
26810 @end smallexample
26811
26812 Assigns the value of @var{expression} to the variable object specified
26813 by @var{name}. The object must be @samp{editable}. If the variable's
26814 value is altered by the assign, the variable will show up in any
26815 subsequent @code{-var-update} list.
26816
26817 @subsubheading Example
26818
26819 @smallexample
26820 (gdb)
26821 -var-assign var1 3
26822 ^done,value="3"
26823 (gdb)
26824 -var-update *
26825 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
26826 (gdb)
26827 @end smallexample
26828
26829 @subheading The @code{-var-update} Command
26830 @findex -var-update
26831
26832 @subsubheading Synopsis
26833
26834 @smallexample
26835 -var-update [@var{print-values}] @{@var{name} | "*"@}
26836 @end smallexample
26837
26838 Reevaluate the expressions corresponding to the variable object
26839 @var{name} and all its direct and indirect children, and return the
26840 list of variable objects whose values have changed; @var{name} must
26841 be a root variable object. Here, ``changed'' means that the result of
26842 @code{-var-evaluate-expression} before and after the
26843 @code{-var-update} is different. If @samp{*} is used as the variable
26844 object names, all existing variable objects are updated, except
26845 for frozen ones (@pxref{-var-set-frozen}). The option
26846 @var{print-values} determines whether both names and values, or just
26847 names are printed. The possible values of this option are the same
26848 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
26849 recommended to use the @samp{--all-values} option, to reduce the
26850 number of MI commands needed on each program stop.
26851
26852 With the @samp{*} parameter, if a variable object is bound to a
26853 currently running thread, it will not be updated, without any
26854 diagnostic.
26855
26856 If @code{-var-set-update-range} was previously used on a varobj, then
26857 only the selected range of children will be reported.
26858
26859 @code{-var-update} reports all the changed varobjs in a tuple named
26860 @samp{changelist}.
26861
26862 Each item in the change list is itself a tuple holding:
26863
26864 @table @samp
26865 @item name
26866 The name of the varobj.
26867
26868 @item value
26869 If values were requested for this update, then this field will be
26870 present and will hold the value of the varobj.
26871
26872 @item in_scope
26873 @anchor{-var-update}
26874 This field is a string which may take one of three values:
26875
26876 @table @code
26877 @item "true"
26878 The variable object's current value is valid.
26879
26880 @item "false"
26881 The variable object does not currently hold a valid value but it may
26882 hold one in the future if its associated expression comes back into
26883 scope.
26884
26885 @item "invalid"
26886 The variable object no longer holds a valid value.
26887 This can occur when the executable file being debugged has changed,
26888 either through recompilation or by using the @value{GDBN} @code{file}
26889 command. The front end should normally choose to delete these variable
26890 objects.
26891 @end table
26892
26893 In the future new values may be added to this list so the front should
26894 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
26895
26896 @item type_changed
26897 This is only present if the varobj is still valid. If the type
26898 changed, then this will be the string @samp{true}; otherwise it will
26899 be @samp{false}.
26900
26901 @item new_type
26902 If the varobj's type changed, then this field will be present and will
26903 hold the new type.
26904
26905 @item new_num_children
26906 For a dynamic varobj, if the number of children changed, or if the
26907 type changed, this will be the new number of children.
26908
26909 The @samp{numchild} field in other varobj responses is generally not
26910 valid for a dynamic varobj -- it will show the number of children that
26911 @value{GDBN} knows about, but because dynamic varobjs lazily
26912 instantiate their children, this will not reflect the number of
26913 children which may be available.
26914
26915 The @samp{new_num_children} attribute only reports changes to the
26916 number of children known by @value{GDBN}. This is the only way to
26917 detect whether an update has removed children (which necessarily can
26918 only happen at the end of the update range).
26919
26920 @item displayhint
26921 The display hint, if any.
26922
26923 @item has_more
26924 This is an integer value, which will be 1 if there are more children
26925 available outside the varobj's update range.
26926
26927 @item dynamic
26928 This attribute will be present and have the value @samp{1} if the
26929 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
26930 then this attribute will not be present.
26931
26932 @item new_children
26933 If new children were added to a dynamic varobj within the selected
26934 update range (as set by @code{-var-set-update-range}), then they will
26935 be listed in this attribute.
26936 @end table
26937
26938 @subsubheading Example
26939
26940 @smallexample
26941 (gdb)
26942 -var-assign var1 3
26943 ^done,value="3"
26944 (gdb)
26945 -var-update --all-values var1
26946 ^done,changelist=[@{name="var1",value="3",in_scope="true",
26947 type_changed="false"@}]
26948 (gdb)
26949 @end smallexample
26950
26951 @subheading The @code{-var-set-frozen} Command
26952 @findex -var-set-frozen
26953 @anchor{-var-set-frozen}
26954
26955 @subsubheading Synopsis
26956
26957 @smallexample
26958 -var-set-frozen @var{name} @var{flag}
26959 @end smallexample
26960
26961 Set the frozenness flag on the variable object @var{name}. The
26962 @var{flag} parameter should be either @samp{1} to make the variable
26963 frozen or @samp{0} to make it unfrozen. If a variable object is
26964 frozen, then neither itself, nor any of its children, are
26965 implicitly updated by @code{-var-update} of
26966 a parent variable or by @code{-var-update *}. Only
26967 @code{-var-update} of the variable itself will update its value and
26968 values of its children. After a variable object is unfrozen, it is
26969 implicitly updated by all subsequent @code{-var-update} operations.
26970 Unfreezing a variable does not update it, only subsequent
26971 @code{-var-update} does.
26972
26973 @subsubheading Example
26974
26975 @smallexample
26976 (gdb)
26977 -var-set-frozen V 1
26978 ^done
26979 (gdb)
26980 @end smallexample
26981
26982 @subheading The @code{-var-set-update-range} command
26983 @findex -var-set-update-range
26984 @anchor{-var-set-update-range}
26985
26986 @subsubheading Synopsis
26987
26988 @smallexample
26989 -var-set-update-range @var{name} @var{from} @var{to}
26990 @end smallexample
26991
26992 Set the range of children to be returned by future invocations of
26993 @code{-var-update}.
26994
26995 @var{from} and @var{to} indicate the range of children to report. If
26996 @var{from} or @var{to} is less than zero, the range is reset and all
26997 children will be reported. Otherwise, children starting at @var{from}
26998 (zero-based) and up to and excluding @var{to} will be reported.
26999
27000 @subsubheading Example
27001
27002 @smallexample
27003 (gdb)
27004 -var-set-update-range V 1 2
27005 ^done
27006 @end smallexample
27007
27008 @subheading The @code{-var-set-visualizer} command
27009 @findex -var-set-visualizer
27010 @anchor{-var-set-visualizer}
27011
27012 @subsubheading Synopsis
27013
27014 @smallexample
27015 -var-set-visualizer @var{name} @var{visualizer}
27016 @end smallexample
27017
27018 Set a visualizer for the variable object @var{name}.
27019
27020 @var{visualizer} is the visualizer to use. The special value
27021 @samp{None} means to disable any visualizer in use.
27022
27023 If not @samp{None}, @var{visualizer} must be a Python expression.
27024 This expression must evaluate to a callable object which accepts a
27025 single argument. @value{GDBN} will call this object with the value of
27026 the varobj @var{name} as an argument (this is done so that the same
27027 Python pretty-printing code can be used for both the CLI and MI).
27028 When called, this object must return an object which conforms to the
27029 pretty-printing interface (@pxref{Pretty Printing API}).
27030
27031 The pre-defined function @code{gdb.default_visualizer} may be used to
27032 select a visualizer by following the built-in process
27033 (@pxref{Selecting Pretty-Printers}). This is done automatically when
27034 a varobj is created, and so ordinarily is not needed.
27035
27036 This feature is only available if Python support is enabled. The MI
27037 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
27038 can be used to check this.
27039
27040 @subsubheading Example
27041
27042 Resetting the visualizer:
27043
27044 @smallexample
27045 (gdb)
27046 -var-set-visualizer V None
27047 ^done
27048 @end smallexample
27049
27050 Reselecting the default (type-based) visualizer:
27051
27052 @smallexample
27053 (gdb)
27054 -var-set-visualizer V gdb.default_visualizer
27055 ^done
27056 @end smallexample
27057
27058 Suppose @code{SomeClass} is a visualizer class. A lambda expression
27059 can be used to instantiate this class for a varobj:
27060
27061 @smallexample
27062 (gdb)
27063 -var-set-visualizer V "lambda val: SomeClass()"
27064 ^done
27065 @end smallexample
27066
27067 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27068 @node GDB/MI Data Manipulation
27069 @section @sc{gdb/mi} Data Manipulation
27070
27071 @cindex data manipulation, in @sc{gdb/mi}
27072 @cindex @sc{gdb/mi}, data manipulation
27073 This section describes the @sc{gdb/mi} commands that manipulate data:
27074 examine memory and registers, evaluate expressions, etc.
27075
27076 @c REMOVED FROM THE INTERFACE.
27077 @c @subheading -data-assign
27078 @c Change the value of a program variable. Plenty of side effects.
27079 @c @subsubheading GDB Command
27080 @c set variable
27081 @c @subsubheading Example
27082 @c N.A.
27083
27084 @subheading The @code{-data-disassemble} Command
27085 @findex -data-disassemble
27086
27087 @subsubheading Synopsis
27088
27089 @smallexample
27090 -data-disassemble
27091 [ -s @var{start-addr} -e @var{end-addr} ]
27092 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
27093 -- @var{mode}
27094 @end smallexample
27095
27096 @noindent
27097 Where:
27098
27099 @table @samp
27100 @item @var{start-addr}
27101 is the beginning address (or @code{$pc})
27102 @item @var{end-addr}
27103 is the end address
27104 @item @var{filename}
27105 is the name of the file to disassemble
27106 @item @var{linenum}
27107 is the line number to disassemble around
27108 @item @var{lines}
27109 is the number of disassembly lines to be produced. If it is -1,
27110 the whole function will be disassembled, in case no @var{end-addr} is
27111 specified. If @var{end-addr} is specified as a non-zero value, and
27112 @var{lines} is lower than the number of disassembly lines between
27113 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
27114 displayed; if @var{lines} is higher than the number of lines between
27115 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
27116 are displayed.
27117 @item @var{mode}
27118 is either 0 (meaning only disassembly) or 1 (meaning mixed source and
27119 disassembly).
27120 @end table
27121
27122 @subsubheading Result
27123
27124 The output for each instruction is composed of four fields:
27125
27126 @itemize @bullet
27127 @item Address
27128 @item Func-name
27129 @item Offset
27130 @item Instruction
27131 @end itemize
27132
27133 Note that whatever included in the instruction field, is not manipulated
27134 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
27135
27136 @subsubheading @value{GDBN} Command
27137
27138 There's no direct mapping from this command to the CLI.
27139
27140 @subsubheading Example
27141
27142 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
27143
27144 @smallexample
27145 (gdb)
27146 -data-disassemble -s $pc -e "$pc + 20" -- 0
27147 ^done,
27148 asm_insns=[
27149 @{address="0x000107c0",func-name="main",offset="4",
27150 inst="mov 2, %o0"@},
27151 @{address="0x000107c4",func-name="main",offset="8",
27152 inst="sethi %hi(0x11800), %o2"@},
27153 @{address="0x000107c8",func-name="main",offset="12",
27154 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
27155 @{address="0x000107cc",func-name="main",offset="16",
27156 inst="sethi %hi(0x11800), %o2"@},
27157 @{address="0x000107d0",func-name="main",offset="20",
27158 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
27159 (gdb)
27160 @end smallexample
27161
27162 Disassemble the whole @code{main} function. Line 32 is part of
27163 @code{main}.
27164
27165 @smallexample
27166 -data-disassemble -f basics.c -l 32 -- 0
27167 ^done,asm_insns=[
27168 @{address="0x000107bc",func-name="main",offset="0",
27169 inst="save %sp, -112, %sp"@},
27170 @{address="0x000107c0",func-name="main",offset="4",
27171 inst="mov 2, %o0"@},
27172 @{address="0x000107c4",func-name="main",offset="8",
27173 inst="sethi %hi(0x11800), %o2"@},
27174 [@dots{}]
27175 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
27176 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
27177 (gdb)
27178 @end smallexample
27179
27180 Disassemble 3 instructions from the start of @code{main}:
27181
27182 @smallexample
27183 (gdb)
27184 -data-disassemble -f basics.c -l 32 -n 3 -- 0
27185 ^done,asm_insns=[
27186 @{address="0x000107bc",func-name="main",offset="0",
27187 inst="save %sp, -112, %sp"@},
27188 @{address="0x000107c0",func-name="main",offset="4",
27189 inst="mov 2, %o0"@},
27190 @{address="0x000107c4",func-name="main",offset="8",
27191 inst="sethi %hi(0x11800), %o2"@}]
27192 (gdb)
27193 @end smallexample
27194
27195 Disassemble 3 instructions from the start of @code{main} in mixed mode:
27196
27197 @smallexample
27198 (gdb)
27199 -data-disassemble -f basics.c -l 32 -n 3 -- 1
27200 ^done,asm_insns=[
27201 src_and_asm_line=@{line="31",
27202 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
27203 testsuite/gdb.mi/basics.c",line_asm_insn=[
27204 @{address="0x000107bc",func-name="main",offset="0",
27205 inst="save %sp, -112, %sp"@}]@},
27206 src_and_asm_line=@{line="32",
27207 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
27208 testsuite/gdb.mi/basics.c",line_asm_insn=[
27209 @{address="0x000107c0",func-name="main",offset="4",
27210 inst="mov 2, %o0"@},
27211 @{address="0x000107c4",func-name="main",offset="8",
27212 inst="sethi %hi(0x11800), %o2"@}]@}]
27213 (gdb)
27214 @end smallexample
27215
27216
27217 @subheading The @code{-data-evaluate-expression} Command
27218 @findex -data-evaluate-expression
27219
27220 @subsubheading Synopsis
27221
27222 @smallexample
27223 -data-evaluate-expression @var{expr}
27224 @end smallexample
27225
27226 Evaluate @var{expr} as an expression. The expression could contain an
27227 inferior function call. The function call will execute synchronously.
27228 If the expression contains spaces, it must be enclosed in double quotes.
27229
27230 @subsubheading @value{GDBN} Command
27231
27232 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
27233 @samp{call}. In @code{gdbtk} only, there's a corresponding
27234 @samp{gdb_eval} command.
27235
27236 @subsubheading Example
27237
27238 In the following example, the numbers that precede the commands are the
27239 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
27240 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
27241 output.
27242
27243 @smallexample
27244 211-data-evaluate-expression A
27245 211^done,value="1"
27246 (gdb)
27247 311-data-evaluate-expression &A
27248 311^done,value="0xefffeb7c"
27249 (gdb)
27250 411-data-evaluate-expression A+3
27251 411^done,value="4"
27252 (gdb)
27253 511-data-evaluate-expression "A + 3"
27254 511^done,value="4"
27255 (gdb)
27256 @end smallexample
27257
27258
27259 @subheading The @code{-data-list-changed-registers} Command
27260 @findex -data-list-changed-registers
27261
27262 @subsubheading Synopsis
27263
27264 @smallexample
27265 -data-list-changed-registers
27266 @end smallexample
27267
27268 Display a list of the registers that have changed.
27269
27270 @subsubheading @value{GDBN} Command
27271
27272 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
27273 has the corresponding command @samp{gdb_changed_register_list}.
27274
27275 @subsubheading Example
27276
27277 On a PPC MBX board:
27278
27279 @smallexample
27280 (gdb)
27281 -exec-continue
27282 ^running
27283
27284 (gdb)
27285 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
27286 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
27287 line="5"@}
27288 (gdb)
27289 -data-list-changed-registers
27290 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
27291 "10","11","13","14","15","16","17","18","19","20","21","22","23",
27292 "24","25","26","27","28","30","31","64","65","66","67","69"]
27293 (gdb)
27294 @end smallexample
27295
27296
27297 @subheading The @code{-data-list-register-names} Command
27298 @findex -data-list-register-names
27299
27300 @subsubheading Synopsis
27301
27302 @smallexample
27303 -data-list-register-names [ ( @var{regno} )+ ]
27304 @end smallexample
27305
27306 Show a list of register names for the current target. If no arguments
27307 are given, it shows a list of the names of all the registers. If
27308 integer numbers are given as arguments, it will print a list of the
27309 names of the registers corresponding to the arguments. To ensure
27310 consistency between a register name and its number, the output list may
27311 include empty register names.
27312
27313 @subsubheading @value{GDBN} Command
27314
27315 @value{GDBN} does not have a command which corresponds to
27316 @samp{-data-list-register-names}. In @code{gdbtk} there is a
27317 corresponding command @samp{gdb_regnames}.
27318
27319 @subsubheading Example
27320
27321 For the PPC MBX board:
27322 @smallexample
27323 (gdb)
27324 -data-list-register-names
27325 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
27326 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
27327 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
27328 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
27329 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
27330 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
27331 "", "pc","ps","cr","lr","ctr","xer"]
27332 (gdb)
27333 -data-list-register-names 1 2 3
27334 ^done,register-names=["r1","r2","r3"]
27335 (gdb)
27336 @end smallexample
27337
27338 @subheading The @code{-data-list-register-values} Command
27339 @findex -data-list-register-values
27340
27341 @subsubheading Synopsis
27342
27343 @smallexample
27344 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
27345 @end smallexample
27346
27347 Display the registers' contents. @var{fmt} is the format according to
27348 which the registers' contents are to be returned, followed by an optional
27349 list of numbers specifying the registers to display. A missing list of
27350 numbers indicates that the contents of all the registers must be returned.
27351
27352 Allowed formats for @var{fmt} are:
27353
27354 @table @code
27355 @item x
27356 Hexadecimal
27357 @item o
27358 Octal
27359 @item t
27360 Binary
27361 @item d
27362 Decimal
27363 @item r
27364 Raw
27365 @item N
27366 Natural
27367 @end table
27368
27369 @subsubheading @value{GDBN} Command
27370
27371 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
27372 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
27373
27374 @subsubheading Example
27375
27376 For a PPC MBX board (note: line breaks are for readability only, they
27377 don't appear in the actual output):
27378
27379 @smallexample
27380 (gdb)
27381 -data-list-register-values r 64 65
27382 ^done,register-values=[@{number="64",value="0xfe00a300"@},
27383 @{number="65",value="0x00029002"@}]
27384 (gdb)
27385 -data-list-register-values x
27386 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
27387 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
27388 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
27389 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
27390 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
27391 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
27392 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
27393 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
27394 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
27395 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
27396 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
27397 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
27398 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
27399 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
27400 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
27401 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
27402 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
27403 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
27404 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
27405 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
27406 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
27407 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
27408 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
27409 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
27410 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
27411 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
27412 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
27413 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
27414 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
27415 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
27416 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
27417 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
27418 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
27419 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
27420 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
27421 @{number="69",value="0x20002b03"@}]
27422 (gdb)
27423 @end smallexample
27424
27425
27426 @subheading The @code{-data-read-memory} Command
27427 @findex -data-read-memory
27428
27429 This command is deprecated, use @code{-data-read-memory-bytes} instead.
27430
27431 @subsubheading Synopsis
27432
27433 @smallexample
27434 -data-read-memory [ -o @var{byte-offset} ]
27435 @var{address} @var{word-format} @var{word-size}
27436 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
27437 @end smallexample
27438
27439 @noindent
27440 where:
27441
27442 @table @samp
27443 @item @var{address}
27444 An expression specifying the address of the first memory word to be
27445 read. Complex expressions containing embedded white space should be
27446 quoted using the C convention.
27447
27448 @item @var{word-format}
27449 The format to be used to print the memory words. The notation is the
27450 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
27451 ,Output Formats}).
27452
27453 @item @var{word-size}
27454 The size of each memory word in bytes.
27455
27456 @item @var{nr-rows}
27457 The number of rows in the output table.
27458
27459 @item @var{nr-cols}
27460 The number of columns in the output table.
27461
27462 @item @var{aschar}
27463 If present, indicates that each row should include an @sc{ascii} dump. The
27464 value of @var{aschar} is used as a padding character when a byte is not a
27465 member of the printable @sc{ascii} character set (printable @sc{ascii}
27466 characters are those whose code is between 32 and 126, inclusively).
27467
27468 @item @var{byte-offset}
27469 An offset to add to the @var{address} before fetching memory.
27470 @end table
27471
27472 This command displays memory contents as a table of @var{nr-rows} by
27473 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
27474 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
27475 (returned as @samp{total-bytes}). Should less than the requested number
27476 of bytes be returned by the target, the missing words are identified
27477 using @samp{N/A}. The number of bytes read from the target is returned
27478 in @samp{nr-bytes} and the starting address used to read memory in
27479 @samp{addr}.
27480
27481 The address of the next/previous row or page is available in
27482 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
27483 @samp{prev-page}.
27484
27485 @subsubheading @value{GDBN} Command
27486
27487 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
27488 @samp{gdb_get_mem} memory read command.
27489
27490 @subsubheading Example
27491
27492 Read six bytes of memory starting at @code{bytes+6} but then offset by
27493 @code{-6} bytes. Format as three rows of two columns. One byte per
27494 word. Display each word in hex.
27495
27496 @smallexample
27497 (gdb)
27498 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
27499 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
27500 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
27501 prev-page="0x0000138a",memory=[
27502 @{addr="0x00001390",data=["0x00","0x01"]@},
27503 @{addr="0x00001392",data=["0x02","0x03"]@},
27504 @{addr="0x00001394",data=["0x04","0x05"]@}]
27505 (gdb)
27506 @end smallexample
27507
27508 Read two bytes of memory starting at address @code{shorts + 64} and
27509 display as a single word formatted in decimal.
27510
27511 @smallexample
27512 (gdb)
27513 5-data-read-memory shorts+64 d 2 1 1
27514 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
27515 next-row="0x00001512",prev-row="0x0000150e",
27516 next-page="0x00001512",prev-page="0x0000150e",memory=[
27517 @{addr="0x00001510",data=["128"]@}]
27518 (gdb)
27519 @end smallexample
27520
27521 Read thirty two bytes of memory starting at @code{bytes+16} and format
27522 as eight rows of four columns. Include a string encoding with @samp{x}
27523 used as the non-printable character.
27524
27525 @smallexample
27526 (gdb)
27527 4-data-read-memory bytes+16 x 1 8 4 x
27528 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
27529 next-row="0x000013c0",prev-row="0x0000139c",
27530 next-page="0x000013c0",prev-page="0x00001380",memory=[
27531 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
27532 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
27533 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
27534 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
27535 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
27536 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
27537 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
27538 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
27539 (gdb)
27540 @end smallexample
27541
27542 @subheading The @code{-data-read-memory-bytes} Command
27543 @findex -data-read-memory-bytes
27544
27545 @subsubheading Synopsis
27546
27547 @smallexample
27548 -data-read-memory-bytes [ -o @var{byte-offset} ]
27549 @var{address} @var{count}
27550 @end smallexample
27551
27552 @noindent
27553 where:
27554
27555 @table @samp
27556 @item @var{address}
27557 An expression specifying the address of the first memory word to be
27558 read. Complex expressions containing embedded white space should be
27559 quoted using the C convention.
27560
27561 @item @var{count}
27562 The number of bytes to read. This should be an integer literal.
27563
27564 @item @var{byte-offset}
27565 The offsets in bytes relative to @var{address} at which to start
27566 reading. This should be an integer literal. This option is provided
27567 so that a frontend is not required to first evaluate address and then
27568 perform address arithmetics itself.
27569
27570 @end table
27571
27572 This command attempts to read all accessible memory regions in the
27573 specified range. First, all regions marked as unreadable in the memory
27574 map (if one is defined) will be skipped. @xref{Memory Region
27575 Attributes}. Second, @value{GDBN} will attempt to read the remaining
27576 regions. For each one, if reading full region results in an errors,
27577 @value{GDBN} will try to read a subset of the region.
27578
27579 In general, every single byte in the region may be readable or not,
27580 and the only way to read every readable byte is to try a read at
27581 every address, which is not practical. Therefore, @value{GDBN} will
27582 attempt to read all accessible bytes at either beginning or the end
27583 of the region, using a binary division scheme. This heuristic works
27584 well for reading accross a memory map boundary. Note that if a region
27585 has a readable range that is neither at the beginning or the end,
27586 @value{GDBN} will not read it.
27587
27588 The result record (@pxref{GDB/MI Result Records}) that is output of
27589 the command includes a field named @samp{memory} whose content is a
27590 list of tuples. Each tuple represent a successfully read memory block
27591 and has the following fields:
27592
27593 @table @code
27594 @item begin
27595 The start address of the memory block, as hexadecimal literal.
27596
27597 @item end
27598 The end address of the memory block, as hexadecimal literal.
27599
27600 @item offset
27601 The offset of the memory block, as hexadecimal literal, relative to
27602 the start address passed to @code{-data-read-memory-bytes}.
27603
27604 @item contents
27605 The contents of the memory block, in hex.
27606
27607 @end table
27608
27609
27610
27611 @subsubheading @value{GDBN} Command
27612
27613 The corresponding @value{GDBN} command is @samp{x}.
27614
27615 @subsubheading Example
27616
27617 @smallexample
27618 (gdb)
27619 -data-read-memory-bytes &a 10
27620 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
27621 end="0xbffff15e",
27622 contents="01000000020000000300"@}]
27623 (gdb)
27624 @end smallexample
27625
27626
27627 @subheading The @code{-data-write-memory-bytes} Command
27628 @findex -data-write-memory-bytes
27629
27630 @subsubheading Synopsis
27631
27632 @smallexample
27633 -data-write-memory-bytes @var{address} @var{contents}
27634 @end smallexample
27635
27636 @noindent
27637 where:
27638
27639 @table @samp
27640 @item @var{address}
27641 An expression specifying the address of the first memory word to be
27642 read. Complex expressions containing embedded white space should be
27643 quoted using the C convention.
27644
27645 @item @var{contents}
27646 The hex-encoded bytes to write.
27647
27648 @end table
27649
27650 @subsubheading @value{GDBN} Command
27651
27652 There's no corresponding @value{GDBN} command.
27653
27654 @subsubheading Example
27655
27656 @smallexample
27657 (gdb)
27658 -data-write-memory-bytes &a "aabbccdd"
27659 ^done
27660 (gdb)
27661 @end smallexample
27662
27663
27664 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27665 @node GDB/MI Tracepoint Commands
27666 @section @sc{gdb/mi} Tracepoint Commands
27667
27668 The commands defined in this section implement MI support for
27669 tracepoints. For detailed introduction, see @ref{Tracepoints}.
27670
27671 @subheading The @code{-trace-find} Command
27672 @findex -trace-find
27673
27674 @subsubheading Synopsis
27675
27676 @smallexample
27677 -trace-find @var{mode} [@var{parameters}@dots{}]
27678 @end smallexample
27679
27680 Find a trace frame using criteria defined by @var{mode} and
27681 @var{parameters}. The following table lists permissible
27682 modes and their parameters. For details of operation, see @ref{tfind}.
27683
27684 @table @samp
27685
27686 @item none
27687 No parameters are required. Stops examining trace frames.
27688
27689 @item frame-number
27690 An integer is required as parameter. Selects tracepoint frame with
27691 that index.
27692
27693 @item tracepoint-number
27694 An integer is required as parameter. Finds next
27695 trace frame that corresponds to tracepoint with the specified number.
27696
27697 @item pc
27698 An address is required as parameter. Finds
27699 next trace frame that corresponds to any tracepoint at the specified
27700 address.
27701
27702 @item pc-inside-range
27703 Two addresses are required as parameters. Finds next trace
27704 frame that corresponds to a tracepoint at an address inside the
27705 specified range. Both bounds are considered to be inside the range.
27706
27707 @item pc-outside-range
27708 Two addresses are required as parameters. Finds
27709 next trace frame that corresponds to a tracepoint at an address outside
27710 the specified range. Both bounds are considered to be inside the range.
27711
27712 @item line
27713 Line specification is required as parameter. @xref{Specify Location}.
27714 Finds next trace frame that corresponds to a tracepoint at
27715 the specified location.
27716
27717 @end table
27718
27719 If @samp{none} was passed as @var{mode}, the response does not
27720 have fields. Otherwise, the response may have the following fields:
27721
27722 @table @samp
27723 @item found
27724 This field has either @samp{0} or @samp{1} as the value, depending
27725 on whether a matching tracepoint was found.
27726
27727 @item traceframe
27728 The index of the found traceframe. This field is present iff
27729 the @samp{found} field has value of @samp{1}.
27730
27731 @item tracepoint
27732 The index of the found tracepoint. This field is present iff
27733 the @samp{found} field has value of @samp{1}.
27734
27735 @item frame
27736 The information about the frame corresponding to the found trace
27737 frame. This field is present only if a trace frame was found.
27738 @xref{GDB/MI Frame Information}, for description of this field.
27739
27740 @end table
27741
27742 @subsubheading @value{GDBN} Command
27743
27744 The corresponding @value{GDBN} command is @samp{tfind}.
27745
27746 @subheading -trace-define-variable
27747 @findex -trace-define-variable
27748
27749 @subsubheading Synopsis
27750
27751 @smallexample
27752 -trace-define-variable @var{name} [ @var{value} ]
27753 @end smallexample
27754
27755 Create trace variable @var{name} if it does not exist. If
27756 @var{value} is specified, sets the initial value of the specified
27757 trace variable to that value. Note that the @var{name} should start
27758 with the @samp{$} character.
27759
27760 @subsubheading @value{GDBN} Command
27761
27762 The corresponding @value{GDBN} command is @samp{tvariable}.
27763
27764 @subheading -trace-list-variables
27765 @findex -trace-list-variables
27766
27767 @subsubheading Synopsis
27768
27769 @smallexample
27770 -trace-list-variables
27771 @end smallexample
27772
27773 Return a table of all defined trace variables. Each element of the
27774 table has the following fields:
27775
27776 @table @samp
27777 @item name
27778 The name of the trace variable. This field is always present.
27779
27780 @item initial
27781 The initial value. This is a 64-bit signed integer. This
27782 field is always present.
27783
27784 @item current
27785 The value the trace variable has at the moment. This is a 64-bit
27786 signed integer. This field is absent iff current value is
27787 not defined, for example if the trace was never run, or is
27788 presently running.
27789
27790 @end table
27791
27792 @subsubheading @value{GDBN} Command
27793
27794 The corresponding @value{GDBN} command is @samp{tvariables}.
27795
27796 @subsubheading Example
27797
27798 @smallexample
27799 (gdb)
27800 -trace-list-variables
27801 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
27802 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
27803 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
27804 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
27805 body=[variable=@{name="$trace_timestamp",initial="0"@}
27806 variable=@{name="$foo",initial="10",current="15"@}]@}
27807 (gdb)
27808 @end smallexample
27809
27810 @subheading -trace-save
27811 @findex -trace-save
27812
27813 @subsubheading Synopsis
27814
27815 @smallexample
27816 -trace-save [-r ] @var{filename}
27817 @end smallexample
27818
27819 Saves the collected trace data to @var{filename}. Without the
27820 @samp{-r} option, the data is downloaded from the target and saved
27821 in a local file. With the @samp{-r} option the target is asked
27822 to perform the save.
27823
27824 @subsubheading @value{GDBN} Command
27825
27826 The corresponding @value{GDBN} command is @samp{tsave}.
27827
27828
27829 @subheading -trace-start
27830 @findex -trace-start
27831
27832 @subsubheading Synopsis
27833
27834 @smallexample
27835 -trace-start
27836 @end smallexample
27837
27838 Starts a tracing experiments. The result of this command does not
27839 have any fields.
27840
27841 @subsubheading @value{GDBN} Command
27842
27843 The corresponding @value{GDBN} command is @samp{tstart}.
27844
27845 @subheading -trace-status
27846 @findex -trace-status
27847
27848 @subsubheading Synopsis
27849
27850 @smallexample
27851 -trace-status
27852 @end smallexample
27853
27854 Obtains the status of a tracing experiment. The result may include
27855 the following fields:
27856
27857 @table @samp
27858
27859 @item supported
27860 May have a value of either @samp{0}, when no tracing operations are
27861 supported, @samp{1}, when all tracing operations are supported, or
27862 @samp{file} when examining trace file. In the latter case, examining
27863 of trace frame is possible but new tracing experiement cannot be
27864 started. This field is always present.
27865
27866 @item running
27867 May have a value of either @samp{0} or @samp{1} depending on whether
27868 tracing experiement is in progress on target. This field is present
27869 if @samp{supported} field is not @samp{0}.
27870
27871 @item stop-reason
27872 Report the reason why the tracing was stopped last time. This field
27873 may be absent iff tracing was never stopped on target yet. The
27874 value of @samp{request} means the tracing was stopped as result of
27875 the @code{-trace-stop} command. The value of @samp{overflow} means
27876 the tracing buffer is full. The value of @samp{disconnection} means
27877 tracing was automatically stopped when @value{GDBN} has disconnected.
27878 The value of @samp{passcount} means tracing was stopped when a
27879 tracepoint was passed a maximal number of times for that tracepoint.
27880 This field is present if @samp{supported} field is not @samp{0}.
27881
27882 @item stopping-tracepoint
27883 The number of tracepoint whose passcount as exceeded. This field is
27884 present iff the @samp{stop-reason} field has the value of
27885 @samp{passcount}.
27886
27887 @item frames
27888 @itemx frames-created
27889 The @samp{frames} field is a count of the total number of trace frames
27890 in the trace buffer, while @samp{frames-created} is the total created
27891 during the run, including ones that were discarded, such as when a
27892 circular trace buffer filled up. Both fields are optional.
27893
27894 @item buffer-size
27895 @itemx buffer-free
27896 These fields tell the current size of the tracing buffer and the
27897 remaining space. These fields are optional.
27898
27899 @item circular
27900 The value of the circular trace buffer flag. @code{1} means that the
27901 trace buffer is circular and old trace frames will be discarded if
27902 necessary to make room, @code{0} means that the trace buffer is linear
27903 and may fill up.
27904
27905 @item disconnected
27906 The value of the disconnected tracing flag. @code{1} means that
27907 tracing will continue after @value{GDBN} disconnects, @code{0} means
27908 that the trace run will stop.
27909
27910 @end table
27911
27912 @subsubheading @value{GDBN} Command
27913
27914 The corresponding @value{GDBN} command is @samp{tstatus}.
27915
27916 @subheading -trace-stop
27917 @findex -trace-stop
27918
27919 @subsubheading Synopsis
27920
27921 @smallexample
27922 -trace-stop
27923 @end smallexample
27924
27925 Stops a tracing experiment. The result of this command has the same
27926 fields as @code{-trace-status}, except that the @samp{supported} and
27927 @samp{running} fields are not output.
27928
27929 @subsubheading @value{GDBN} Command
27930
27931 The corresponding @value{GDBN} command is @samp{tstop}.
27932
27933
27934 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27935 @node GDB/MI Symbol Query
27936 @section @sc{gdb/mi} Symbol Query Commands
27937
27938
27939 @ignore
27940 @subheading The @code{-symbol-info-address} Command
27941 @findex -symbol-info-address
27942
27943 @subsubheading Synopsis
27944
27945 @smallexample
27946 -symbol-info-address @var{symbol}
27947 @end smallexample
27948
27949 Describe where @var{symbol} is stored.
27950
27951 @subsubheading @value{GDBN} Command
27952
27953 The corresponding @value{GDBN} command is @samp{info address}.
27954
27955 @subsubheading Example
27956 N.A.
27957
27958
27959 @subheading The @code{-symbol-info-file} Command
27960 @findex -symbol-info-file
27961
27962 @subsubheading Synopsis
27963
27964 @smallexample
27965 -symbol-info-file
27966 @end smallexample
27967
27968 Show the file for the symbol.
27969
27970 @subsubheading @value{GDBN} Command
27971
27972 There's no equivalent @value{GDBN} command. @code{gdbtk} has
27973 @samp{gdb_find_file}.
27974
27975 @subsubheading Example
27976 N.A.
27977
27978
27979 @subheading The @code{-symbol-info-function} Command
27980 @findex -symbol-info-function
27981
27982 @subsubheading Synopsis
27983
27984 @smallexample
27985 -symbol-info-function
27986 @end smallexample
27987
27988 Show which function the symbol lives in.
27989
27990 @subsubheading @value{GDBN} Command
27991
27992 @samp{gdb_get_function} in @code{gdbtk}.
27993
27994 @subsubheading Example
27995 N.A.
27996
27997
27998 @subheading The @code{-symbol-info-line} Command
27999 @findex -symbol-info-line
28000
28001 @subsubheading Synopsis
28002
28003 @smallexample
28004 -symbol-info-line
28005 @end smallexample
28006
28007 Show the core addresses of the code for a source line.
28008
28009 @subsubheading @value{GDBN} Command
28010
28011 The corresponding @value{GDBN} command is @samp{info line}.
28012 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
28013
28014 @subsubheading Example
28015 N.A.
28016
28017
28018 @subheading The @code{-symbol-info-symbol} Command
28019 @findex -symbol-info-symbol
28020
28021 @subsubheading Synopsis
28022
28023 @smallexample
28024 -symbol-info-symbol @var{addr}
28025 @end smallexample
28026
28027 Describe what symbol is at location @var{addr}.
28028
28029 @subsubheading @value{GDBN} Command
28030
28031 The corresponding @value{GDBN} command is @samp{info symbol}.
28032
28033 @subsubheading Example
28034 N.A.
28035
28036
28037 @subheading The @code{-symbol-list-functions} Command
28038 @findex -symbol-list-functions
28039
28040 @subsubheading Synopsis
28041
28042 @smallexample
28043 -symbol-list-functions
28044 @end smallexample
28045
28046 List the functions in the executable.
28047
28048 @subsubheading @value{GDBN} Command
28049
28050 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
28051 @samp{gdb_search} in @code{gdbtk}.
28052
28053 @subsubheading Example
28054 N.A.
28055 @end ignore
28056
28057
28058 @subheading The @code{-symbol-list-lines} Command
28059 @findex -symbol-list-lines
28060
28061 @subsubheading Synopsis
28062
28063 @smallexample
28064 -symbol-list-lines @var{filename}
28065 @end smallexample
28066
28067 Print the list of lines that contain code and their associated program
28068 addresses for the given source filename. The entries are sorted in
28069 ascending PC order.
28070
28071 @subsubheading @value{GDBN} Command
28072
28073 There is no corresponding @value{GDBN} command.
28074
28075 @subsubheading Example
28076 @smallexample
28077 (gdb)
28078 -symbol-list-lines basics.c
28079 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
28080 (gdb)
28081 @end smallexample
28082
28083
28084 @ignore
28085 @subheading The @code{-symbol-list-types} Command
28086 @findex -symbol-list-types
28087
28088 @subsubheading Synopsis
28089
28090 @smallexample
28091 -symbol-list-types
28092 @end smallexample
28093
28094 List all the type names.
28095
28096 @subsubheading @value{GDBN} Command
28097
28098 The corresponding commands are @samp{info types} in @value{GDBN},
28099 @samp{gdb_search} in @code{gdbtk}.
28100
28101 @subsubheading Example
28102 N.A.
28103
28104
28105 @subheading The @code{-symbol-list-variables} Command
28106 @findex -symbol-list-variables
28107
28108 @subsubheading Synopsis
28109
28110 @smallexample
28111 -symbol-list-variables
28112 @end smallexample
28113
28114 List all the global and static variable names.
28115
28116 @subsubheading @value{GDBN} Command
28117
28118 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
28119
28120 @subsubheading Example
28121 N.A.
28122
28123
28124 @subheading The @code{-symbol-locate} Command
28125 @findex -symbol-locate
28126
28127 @subsubheading Synopsis
28128
28129 @smallexample
28130 -symbol-locate
28131 @end smallexample
28132
28133 @subsubheading @value{GDBN} Command
28134
28135 @samp{gdb_loc} in @code{gdbtk}.
28136
28137 @subsubheading Example
28138 N.A.
28139
28140
28141 @subheading The @code{-symbol-type} Command
28142 @findex -symbol-type
28143
28144 @subsubheading Synopsis
28145
28146 @smallexample
28147 -symbol-type @var{variable}
28148 @end smallexample
28149
28150 Show type of @var{variable}.
28151
28152 @subsubheading @value{GDBN} Command
28153
28154 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
28155 @samp{gdb_obj_variable}.
28156
28157 @subsubheading Example
28158 N.A.
28159 @end ignore
28160
28161
28162 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28163 @node GDB/MI File Commands
28164 @section @sc{gdb/mi} File Commands
28165
28166 This section describes the GDB/MI commands to specify executable file names
28167 and to read in and obtain symbol table information.
28168
28169 @subheading The @code{-file-exec-and-symbols} Command
28170 @findex -file-exec-and-symbols
28171
28172 @subsubheading Synopsis
28173
28174 @smallexample
28175 -file-exec-and-symbols @var{file}
28176 @end smallexample
28177
28178 Specify the executable file to be debugged. This file is the one from
28179 which the symbol table is also read. If no file is specified, the
28180 command clears the executable and symbol information. If breakpoints
28181 are set when using this command with no arguments, @value{GDBN} will produce
28182 error messages. Otherwise, no output is produced, except a completion
28183 notification.
28184
28185 @subsubheading @value{GDBN} Command
28186
28187 The corresponding @value{GDBN} command is @samp{file}.
28188
28189 @subsubheading Example
28190
28191 @smallexample
28192 (gdb)
28193 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
28194 ^done
28195 (gdb)
28196 @end smallexample
28197
28198
28199 @subheading The @code{-file-exec-file} Command
28200 @findex -file-exec-file
28201
28202 @subsubheading Synopsis
28203
28204 @smallexample
28205 -file-exec-file @var{file}
28206 @end smallexample
28207
28208 Specify the executable file to be debugged. Unlike
28209 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
28210 from this file. If used without argument, @value{GDBN} clears the information
28211 about the executable file. No output is produced, except a completion
28212 notification.
28213
28214 @subsubheading @value{GDBN} Command
28215
28216 The corresponding @value{GDBN} command is @samp{exec-file}.
28217
28218 @subsubheading Example
28219
28220 @smallexample
28221 (gdb)
28222 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
28223 ^done
28224 (gdb)
28225 @end smallexample
28226
28227
28228 @ignore
28229 @subheading The @code{-file-list-exec-sections} Command
28230 @findex -file-list-exec-sections
28231
28232 @subsubheading Synopsis
28233
28234 @smallexample
28235 -file-list-exec-sections
28236 @end smallexample
28237
28238 List the sections of the current executable file.
28239
28240 @subsubheading @value{GDBN} Command
28241
28242 The @value{GDBN} command @samp{info file} shows, among the rest, the same
28243 information as this command. @code{gdbtk} has a corresponding command
28244 @samp{gdb_load_info}.
28245
28246 @subsubheading Example
28247 N.A.
28248 @end ignore
28249
28250
28251 @subheading The @code{-file-list-exec-source-file} Command
28252 @findex -file-list-exec-source-file
28253
28254 @subsubheading Synopsis
28255
28256 @smallexample
28257 -file-list-exec-source-file
28258 @end smallexample
28259
28260 List the line number, the current source file, and the absolute path
28261 to the current source file for the current executable. The macro
28262 information field has a value of @samp{1} or @samp{0} depending on
28263 whether or not the file includes preprocessor macro information.
28264
28265 @subsubheading @value{GDBN} Command
28266
28267 The @value{GDBN} equivalent is @samp{info source}
28268
28269 @subsubheading Example
28270
28271 @smallexample
28272 (gdb)
28273 123-file-list-exec-source-file
28274 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
28275 (gdb)
28276 @end smallexample
28277
28278
28279 @subheading The @code{-file-list-exec-source-files} Command
28280 @findex -file-list-exec-source-files
28281
28282 @subsubheading Synopsis
28283
28284 @smallexample
28285 -file-list-exec-source-files
28286 @end smallexample
28287
28288 List the source files for the current executable.
28289
28290 It will always output the filename, but only when @value{GDBN} can find
28291 the absolute file name of a source file, will it output the fullname.
28292
28293 @subsubheading @value{GDBN} Command
28294
28295 The @value{GDBN} equivalent is @samp{info sources}.
28296 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
28297
28298 @subsubheading Example
28299 @smallexample
28300 (gdb)
28301 -file-list-exec-source-files
28302 ^done,files=[
28303 @{file=foo.c,fullname=/home/foo.c@},
28304 @{file=/home/bar.c,fullname=/home/bar.c@},
28305 @{file=gdb_could_not_find_fullpath.c@}]
28306 (gdb)
28307 @end smallexample
28308
28309 @ignore
28310 @subheading The @code{-file-list-shared-libraries} Command
28311 @findex -file-list-shared-libraries
28312
28313 @subsubheading Synopsis
28314
28315 @smallexample
28316 -file-list-shared-libraries
28317 @end smallexample
28318
28319 List the shared libraries in the program.
28320
28321 @subsubheading @value{GDBN} Command
28322
28323 The corresponding @value{GDBN} command is @samp{info shared}.
28324
28325 @subsubheading Example
28326 N.A.
28327
28328
28329 @subheading The @code{-file-list-symbol-files} Command
28330 @findex -file-list-symbol-files
28331
28332 @subsubheading Synopsis
28333
28334 @smallexample
28335 -file-list-symbol-files
28336 @end smallexample
28337
28338 List symbol files.
28339
28340 @subsubheading @value{GDBN} Command
28341
28342 The corresponding @value{GDBN} command is @samp{info file} (part of it).
28343
28344 @subsubheading Example
28345 N.A.
28346 @end ignore
28347
28348
28349 @subheading The @code{-file-symbol-file} Command
28350 @findex -file-symbol-file
28351
28352 @subsubheading Synopsis
28353
28354 @smallexample
28355 -file-symbol-file @var{file}
28356 @end smallexample
28357
28358 Read symbol table info from the specified @var{file} argument. When
28359 used without arguments, clears @value{GDBN}'s symbol table info. No output is
28360 produced, except for a completion notification.
28361
28362 @subsubheading @value{GDBN} Command
28363
28364 The corresponding @value{GDBN} command is @samp{symbol-file}.
28365
28366 @subsubheading Example
28367
28368 @smallexample
28369 (gdb)
28370 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
28371 ^done
28372 (gdb)
28373 @end smallexample
28374
28375 @ignore
28376 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28377 @node GDB/MI Memory Overlay Commands
28378 @section @sc{gdb/mi} Memory Overlay Commands
28379
28380 The memory overlay commands are not implemented.
28381
28382 @c @subheading -overlay-auto
28383
28384 @c @subheading -overlay-list-mapping-state
28385
28386 @c @subheading -overlay-list-overlays
28387
28388 @c @subheading -overlay-map
28389
28390 @c @subheading -overlay-off
28391
28392 @c @subheading -overlay-on
28393
28394 @c @subheading -overlay-unmap
28395
28396 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28397 @node GDB/MI Signal Handling Commands
28398 @section @sc{gdb/mi} Signal Handling Commands
28399
28400 Signal handling commands are not implemented.
28401
28402 @c @subheading -signal-handle
28403
28404 @c @subheading -signal-list-handle-actions
28405
28406 @c @subheading -signal-list-signal-types
28407 @end ignore
28408
28409
28410 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28411 @node GDB/MI Target Manipulation
28412 @section @sc{gdb/mi} Target Manipulation Commands
28413
28414
28415 @subheading The @code{-target-attach} Command
28416 @findex -target-attach
28417
28418 @subsubheading Synopsis
28419
28420 @smallexample
28421 -target-attach @var{pid} | @var{gid} | @var{file}
28422 @end smallexample
28423
28424 Attach to a process @var{pid} or a file @var{file} outside of
28425 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
28426 group, the id previously returned by
28427 @samp{-list-thread-groups --available} must be used.
28428
28429 @subsubheading @value{GDBN} Command
28430
28431 The corresponding @value{GDBN} command is @samp{attach}.
28432
28433 @subsubheading Example
28434 @smallexample
28435 (gdb)
28436 -target-attach 34
28437 =thread-created,id="1"
28438 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
28439 ^done
28440 (gdb)
28441 @end smallexample
28442
28443 @ignore
28444 @subheading The @code{-target-compare-sections} Command
28445 @findex -target-compare-sections
28446
28447 @subsubheading Synopsis
28448
28449 @smallexample
28450 -target-compare-sections [ @var{section} ]
28451 @end smallexample
28452
28453 Compare data of section @var{section} on target to the exec file.
28454 Without the argument, all sections are compared.
28455
28456 @subsubheading @value{GDBN} Command
28457
28458 The @value{GDBN} equivalent is @samp{compare-sections}.
28459
28460 @subsubheading Example
28461 N.A.
28462 @end ignore
28463
28464
28465 @subheading The @code{-target-detach} Command
28466 @findex -target-detach
28467
28468 @subsubheading Synopsis
28469
28470 @smallexample
28471 -target-detach [ @var{pid} | @var{gid} ]
28472 @end smallexample
28473
28474 Detach from the remote target which normally resumes its execution.
28475 If either @var{pid} or @var{gid} is specified, detaches from either
28476 the specified process, or specified thread group. There's no output.
28477
28478 @subsubheading @value{GDBN} Command
28479
28480 The corresponding @value{GDBN} command is @samp{detach}.
28481
28482 @subsubheading Example
28483
28484 @smallexample
28485 (gdb)
28486 -target-detach
28487 ^done
28488 (gdb)
28489 @end smallexample
28490
28491
28492 @subheading The @code{-target-disconnect} Command
28493 @findex -target-disconnect
28494
28495 @subsubheading Synopsis
28496
28497 @smallexample
28498 -target-disconnect
28499 @end smallexample
28500
28501 Disconnect from the remote target. There's no output and the target is
28502 generally not resumed.
28503
28504 @subsubheading @value{GDBN} Command
28505
28506 The corresponding @value{GDBN} command is @samp{disconnect}.
28507
28508 @subsubheading Example
28509
28510 @smallexample
28511 (gdb)
28512 -target-disconnect
28513 ^done
28514 (gdb)
28515 @end smallexample
28516
28517
28518 @subheading The @code{-target-download} Command
28519 @findex -target-download
28520
28521 @subsubheading Synopsis
28522
28523 @smallexample
28524 -target-download
28525 @end smallexample
28526
28527 Loads the executable onto the remote target.
28528 It prints out an update message every half second, which includes the fields:
28529
28530 @table @samp
28531 @item section
28532 The name of the section.
28533 @item section-sent
28534 The size of what has been sent so far for that section.
28535 @item section-size
28536 The size of the section.
28537 @item total-sent
28538 The total size of what was sent so far (the current and the previous sections).
28539 @item total-size
28540 The size of the overall executable to download.
28541 @end table
28542
28543 @noindent
28544 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
28545 @sc{gdb/mi} Output Syntax}).
28546
28547 In addition, it prints the name and size of the sections, as they are
28548 downloaded. These messages include the following fields:
28549
28550 @table @samp
28551 @item section
28552 The name of the section.
28553 @item section-size
28554 The size of the section.
28555 @item total-size
28556 The size of the overall executable to download.
28557 @end table
28558
28559 @noindent
28560 At the end, a summary is printed.
28561
28562 @subsubheading @value{GDBN} Command
28563
28564 The corresponding @value{GDBN} command is @samp{load}.
28565
28566 @subsubheading Example
28567
28568 Note: each status message appears on a single line. Here the messages
28569 have been broken down so that they can fit onto a page.
28570
28571 @smallexample
28572 (gdb)
28573 -target-download
28574 +download,@{section=".text",section-size="6668",total-size="9880"@}
28575 +download,@{section=".text",section-sent="512",section-size="6668",
28576 total-sent="512",total-size="9880"@}
28577 +download,@{section=".text",section-sent="1024",section-size="6668",
28578 total-sent="1024",total-size="9880"@}
28579 +download,@{section=".text",section-sent="1536",section-size="6668",
28580 total-sent="1536",total-size="9880"@}
28581 +download,@{section=".text",section-sent="2048",section-size="6668",
28582 total-sent="2048",total-size="9880"@}
28583 +download,@{section=".text",section-sent="2560",section-size="6668",
28584 total-sent="2560",total-size="9880"@}
28585 +download,@{section=".text",section-sent="3072",section-size="6668",
28586 total-sent="3072",total-size="9880"@}
28587 +download,@{section=".text",section-sent="3584",section-size="6668",
28588 total-sent="3584",total-size="9880"@}
28589 +download,@{section=".text",section-sent="4096",section-size="6668",
28590 total-sent="4096",total-size="9880"@}
28591 +download,@{section=".text",section-sent="4608",section-size="6668",
28592 total-sent="4608",total-size="9880"@}
28593 +download,@{section=".text",section-sent="5120",section-size="6668",
28594 total-sent="5120",total-size="9880"@}
28595 +download,@{section=".text",section-sent="5632",section-size="6668",
28596 total-sent="5632",total-size="9880"@}
28597 +download,@{section=".text",section-sent="6144",section-size="6668",
28598 total-sent="6144",total-size="9880"@}
28599 +download,@{section=".text",section-sent="6656",section-size="6668",
28600 total-sent="6656",total-size="9880"@}
28601 +download,@{section=".init",section-size="28",total-size="9880"@}
28602 +download,@{section=".fini",section-size="28",total-size="9880"@}
28603 +download,@{section=".data",section-size="3156",total-size="9880"@}
28604 +download,@{section=".data",section-sent="512",section-size="3156",
28605 total-sent="7236",total-size="9880"@}
28606 +download,@{section=".data",section-sent="1024",section-size="3156",
28607 total-sent="7748",total-size="9880"@}
28608 +download,@{section=".data",section-sent="1536",section-size="3156",
28609 total-sent="8260",total-size="9880"@}
28610 +download,@{section=".data",section-sent="2048",section-size="3156",
28611 total-sent="8772",total-size="9880"@}
28612 +download,@{section=".data",section-sent="2560",section-size="3156",
28613 total-sent="9284",total-size="9880"@}
28614 +download,@{section=".data",section-sent="3072",section-size="3156",
28615 total-sent="9796",total-size="9880"@}
28616 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
28617 write-rate="429"
28618 (gdb)
28619 @end smallexample
28620
28621
28622 @ignore
28623 @subheading The @code{-target-exec-status} Command
28624 @findex -target-exec-status
28625
28626 @subsubheading Synopsis
28627
28628 @smallexample
28629 -target-exec-status
28630 @end smallexample
28631
28632 Provide information on the state of the target (whether it is running or
28633 not, for instance).
28634
28635 @subsubheading @value{GDBN} Command
28636
28637 There's no equivalent @value{GDBN} command.
28638
28639 @subsubheading Example
28640 N.A.
28641
28642
28643 @subheading The @code{-target-list-available-targets} Command
28644 @findex -target-list-available-targets
28645
28646 @subsubheading Synopsis
28647
28648 @smallexample
28649 -target-list-available-targets
28650 @end smallexample
28651
28652 List the possible targets to connect to.
28653
28654 @subsubheading @value{GDBN} Command
28655
28656 The corresponding @value{GDBN} command is @samp{help target}.
28657
28658 @subsubheading Example
28659 N.A.
28660
28661
28662 @subheading The @code{-target-list-current-targets} Command
28663 @findex -target-list-current-targets
28664
28665 @subsubheading Synopsis
28666
28667 @smallexample
28668 -target-list-current-targets
28669 @end smallexample
28670
28671 Describe the current target.
28672
28673 @subsubheading @value{GDBN} Command
28674
28675 The corresponding information is printed by @samp{info file} (among
28676 other things).
28677
28678 @subsubheading Example
28679 N.A.
28680
28681
28682 @subheading The @code{-target-list-parameters} Command
28683 @findex -target-list-parameters
28684
28685 @subsubheading Synopsis
28686
28687 @smallexample
28688 -target-list-parameters
28689 @end smallexample
28690
28691 @c ????
28692 @end ignore
28693
28694 @subsubheading @value{GDBN} Command
28695
28696 No equivalent.
28697
28698 @subsubheading Example
28699 N.A.
28700
28701
28702 @subheading The @code{-target-select} Command
28703 @findex -target-select
28704
28705 @subsubheading Synopsis
28706
28707 @smallexample
28708 -target-select @var{type} @var{parameters @dots{}}
28709 @end smallexample
28710
28711 Connect @value{GDBN} to the remote target. This command takes two args:
28712
28713 @table @samp
28714 @item @var{type}
28715 The type of target, for instance @samp{remote}, etc.
28716 @item @var{parameters}
28717 Device names, host names and the like. @xref{Target Commands, ,
28718 Commands for Managing Targets}, for more details.
28719 @end table
28720
28721 The output is a connection notification, followed by the address at
28722 which the target program is, in the following form:
28723
28724 @smallexample
28725 ^connected,addr="@var{address}",func="@var{function name}",
28726 args=[@var{arg list}]
28727 @end smallexample
28728
28729 @subsubheading @value{GDBN} Command
28730
28731 The corresponding @value{GDBN} command is @samp{target}.
28732
28733 @subsubheading Example
28734
28735 @smallexample
28736 (gdb)
28737 -target-select remote /dev/ttya
28738 ^connected,addr="0xfe00a300",func="??",args=[]
28739 (gdb)
28740 @end smallexample
28741
28742 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28743 @node GDB/MI File Transfer Commands
28744 @section @sc{gdb/mi} File Transfer Commands
28745
28746
28747 @subheading The @code{-target-file-put} Command
28748 @findex -target-file-put
28749
28750 @subsubheading Synopsis
28751
28752 @smallexample
28753 -target-file-put @var{hostfile} @var{targetfile}
28754 @end smallexample
28755
28756 Copy file @var{hostfile} from the host system (the machine running
28757 @value{GDBN}) to @var{targetfile} on the target system.
28758
28759 @subsubheading @value{GDBN} Command
28760
28761 The corresponding @value{GDBN} command is @samp{remote put}.
28762
28763 @subsubheading Example
28764
28765 @smallexample
28766 (gdb)
28767 -target-file-put localfile remotefile
28768 ^done
28769 (gdb)
28770 @end smallexample
28771
28772
28773 @subheading The @code{-target-file-get} Command
28774 @findex -target-file-get
28775
28776 @subsubheading Synopsis
28777
28778 @smallexample
28779 -target-file-get @var{targetfile} @var{hostfile}
28780 @end smallexample
28781
28782 Copy file @var{targetfile} from the target system to @var{hostfile}
28783 on the host system.
28784
28785 @subsubheading @value{GDBN} Command
28786
28787 The corresponding @value{GDBN} command is @samp{remote get}.
28788
28789 @subsubheading Example
28790
28791 @smallexample
28792 (gdb)
28793 -target-file-get remotefile localfile
28794 ^done
28795 (gdb)
28796 @end smallexample
28797
28798
28799 @subheading The @code{-target-file-delete} Command
28800 @findex -target-file-delete
28801
28802 @subsubheading Synopsis
28803
28804 @smallexample
28805 -target-file-delete @var{targetfile}
28806 @end smallexample
28807
28808 Delete @var{targetfile} from the target system.
28809
28810 @subsubheading @value{GDBN} Command
28811
28812 The corresponding @value{GDBN} command is @samp{remote delete}.
28813
28814 @subsubheading Example
28815
28816 @smallexample
28817 (gdb)
28818 -target-file-delete remotefile
28819 ^done
28820 (gdb)
28821 @end smallexample
28822
28823
28824 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28825 @node GDB/MI Miscellaneous Commands
28826 @section Miscellaneous @sc{gdb/mi} Commands
28827
28828 @c @subheading -gdb-complete
28829
28830 @subheading The @code{-gdb-exit} Command
28831 @findex -gdb-exit
28832
28833 @subsubheading Synopsis
28834
28835 @smallexample
28836 -gdb-exit
28837 @end smallexample
28838
28839 Exit @value{GDBN} immediately.
28840
28841 @subsubheading @value{GDBN} Command
28842
28843 Approximately corresponds to @samp{quit}.
28844
28845 @subsubheading Example
28846
28847 @smallexample
28848 (gdb)
28849 -gdb-exit
28850 ^exit
28851 @end smallexample
28852
28853
28854 @ignore
28855 @subheading The @code{-exec-abort} Command
28856 @findex -exec-abort
28857
28858 @subsubheading Synopsis
28859
28860 @smallexample
28861 -exec-abort
28862 @end smallexample
28863
28864 Kill the inferior running program.
28865
28866 @subsubheading @value{GDBN} Command
28867
28868 The corresponding @value{GDBN} command is @samp{kill}.
28869
28870 @subsubheading Example
28871 N.A.
28872 @end ignore
28873
28874
28875 @subheading The @code{-gdb-set} Command
28876 @findex -gdb-set
28877
28878 @subsubheading Synopsis
28879
28880 @smallexample
28881 -gdb-set
28882 @end smallexample
28883
28884 Set an internal @value{GDBN} variable.
28885 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
28886
28887 @subsubheading @value{GDBN} Command
28888
28889 The corresponding @value{GDBN} command is @samp{set}.
28890
28891 @subsubheading Example
28892
28893 @smallexample
28894 (gdb)
28895 -gdb-set $foo=3
28896 ^done
28897 (gdb)
28898 @end smallexample
28899
28900
28901 @subheading The @code{-gdb-show} Command
28902 @findex -gdb-show
28903
28904 @subsubheading Synopsis
28905
28906 @smallexample
28907 -gdb-show
28908 @end smallexample
28909
28910 Show the current value of a @value{GDBN} variable.
28911
28912 @subsubheading @value{GDBN} Command
28913
28914 The corresponding @value{GDBN} command is @samp{show}.
28915
28916 @subsubheading Example
28917
28918 @smallexample
28919 (gdb)
28920 -gdb-show annotate
28921 ^done,value="0"
28922 (gdb)
28923 @end smallexample
28924
28925 @c @subheading -gdb-source
28926
28927
28928 @subheading The @code{-gdb-version} Command
28929 @findex -gdb-version
28930
28931 @subsubheading Synopsis
28932
28933 @smallexample
28934 -gdb-version
28935 @end smallexample
28936
28937 Show version information for @value{GDBN}. Used mostly in testing.
28938
28939 @subsubheading @value{GDBN} Command
28940
28941 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
28942 default shows this information when you start an interactive session.
28943
28944 @subsubheading Example
28945
28946 @c This example modifies the actual output from GDB to avoid overfull
28947 @c box in TeX.
28948 @smallexample
28949 (gdb)
28950 -gdb-version
28951 ~GNU gdb 5.2.1
28952 ~Copyright 2000 Free Software Foundation, Inc.
28953 ~GDB is free software, covered by the GNU General Public License, and
28954 ~you are welcome to change it and/or distribute copies of it under
28955 ~ certain conditions.
28956 ~Type "show copying" to see the conditions.
28957 ~There is absolutely no warranty for GDB. Type "show warranty" for
28958 ~ details.
28959 ~This GDB was configured as
28960 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
28961 ^done
28962 (gdb)
28963 @end smallexample
28964
28965 @subheading The @code{-list-features} Command
28966 @findex -list-features
28967
28968 Returns a list of particular features of the MI protocol that
28969 this version of gdb implements. A feature can be a command,
28970 or a new field in an output of some command, or even an
28971 important bugfix. While a frontend can sometimes detect presence
28972 of a feature at runtime, it is easier to perform detection at debugger
28973 startup.
28974
28975 The command returns a list of strings, with each string naming an
28976 available feature. Each returned string is just a name, it does not
28977 have any internal structure. The list of possible feature names
28978 is given below.
28979
28980 Example output:
28981
28982 @smallexample
28983 (gdb) -list-features
28984 ^done,result=["feature1","feature2"]
28985 @end smallexample
28986
28987 The current list of features is:
28988
28989 @table @samp
28990 @item frozen-varobjs
28991 Indicates presence of the @code{-var-set-frozen} command, as well
28992 as possible presense of the @code{frozen} field in the output
28993 of @code{-varobj-create}.
28994 @item pending-breakpoints
28995 Indicates presence of the @option{-f} option to the @code{-break-insert} command.
28996 @item python
28997 Indicates presence of Python scripting support, Python-based
28998 pretty-printing commands, and possible presence of the
28999 @samp{display_hint} field in the output of @code{-var-list-children}
29000 @item thread-info
29001 Indicates presence of the @code{-thread-info} command.
29002 @item data-read-memory-bytes
29003 Indicates presense of the @code{-data-read-memory-bytes} and the
29004 @code{-data-write-memory-bytes} commands.
29005
29006 @end table
29007
29008 @subheading The @code{-list-target-features} Command
29009 @findex -list-target-features
29010
29011 Returns a list of particular features that are supported by the
29012 target. Those features affect the permitted MI commands, but
29013 unlike the features reported by the @code{-list-features} command, the
29014 features depend on which target GDB is using at the moment. Whenever
29015 a target can change, due to commands such as @code{-target-select},
29016 @code{-target-attach} or @code{-exec-run}, the list of target features
29017 may change, and the frontend should obtain it again.
29018 Example output:
29019
29020 @smallexample
29021 (gdb) -list-features
29022 ^done,result=["async"]
29023 @end smallexample
29024
29025 The current list of features is:
29026
29027 @table @samp
29028 @item async
29029 Indicates that the target is capable of asynchronous command
29030 execution, which means that @value{GDBN} will accept further commands
29031 while the target is running.
29032
29033 @end table
29034
29035 @subheading The @code{-list-thread-groups} Command
29036 @findex -list-thread-groups
29037
29038 @subheading Synopsis
29039
29040 @smallexample
29041 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
29042 @end smallexample
29043
29044 Lists thread groups (@pxref{Thread groups}). When a single thread
29045 group is passed as the argument, lists the children of that group.
29046 When several thread group are passed, lists information about those
29047 thread groups. Without any parameters, lists information about all
29048 top-level thread groups.
29049
29050 Normally, thread groups that are being debugged are reported.
29051 With the @samp{--available} option, @value{GDBN} reports thread groups
29052 available on the target.
29053
29054 The output of this command may have either a @samp{threads} result or
29055 a @samp{groups} result. The @samp{thread} result has a list of tuples
29056 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
29057 Information}). The @samp{groups} result has a list of tuples as value,
29058 each tuple describing a thread group. If top-level groups are
29059 requested (that is, no parameter is passed), or when several groups
29060 are passed, the output always has a @samp{groups} result. The format
29061 of the @samp{group} result is described below.
29062
29063 To reduce the number of roundtrips it's possible to list thread groups
29064 together with their children, by passing the @samp{--recurse} option
29065 and the recursion depth. Presently, only recursion depth of 1 is
29066 permitted. If this option is present, then every reported thread group
29067 will also include its children, either as @samp{group} or
29068 @samp{threads} field.
29069
29070 In general, any combination of option and parameters is permitted, with
29071 the following caveats:
29072
29073 @itemize @bullet
29074 @item
29075 When a single thread group is passed, the output will typically
29076 be the @samp{threads} result. Because threads may not contain
29077 anything, the @samp{recurse} option will be ignored.
29078
29079 @item
29080 When the @samp{--available} option is passed, limited information may
29081 be available. In particular, the list of threads of a process might
29082 be inaccessible. Further, specifying specific thread groups might
29083 not give any performance advantage over listing all thread groups.
29084 The frontend should assume that @samp{-list-thread-groups --available}
29085 is always an expensive operation and cache the results.
29086
29087 @end itemize
29088
29089 The @samp{groups} result is a list of tuples, where each tuple may
29090 have the following fields:
29091
29092 @table @code
29093 @item id
29094 Identifier of the thread group. This field is always present.
29095 The identifier is an opaque string; frontends should not try to
29096 convert it to an integer, even though it might look like one.
29097
29098 @item type
29099 The type of the thread group. At present, only @samp{process} is a
29100 valid type.
29101
29102 @item pid
29103 The target-specific process identifier. This field is only present
29104 for thread groups of type @samp{process} and only if the process exists.
29105
29106 @item num_children
29107 The number of children this thread group has. This field may be
29108 absent for an available thread group.
29109
29110 @item threads
29111 This field has a list of tuples as value, each tuple describing a
29112 thread. It may be present if the @samp{--recurse} option is
29113 specified, and it's actually possible to obtain the threads.
29114
29115 @item cores
29116 This field is a list of integers, each identifying a core that one
29117 thread of the group is running on. This field may be absent if
29118 such information is not available.
29119
29120 @item executable
29121 The name of the executable file that corresponds to this thread group.
29122 The field is only present for thread groups of type @samp{process},
29123 and only if there is a corresponding executable file.
29124
29125 @end table
29126
29127 @subheading Example
29128
29129 @smallexample
29130 @value{GDBP}
29131 -list-thread-groups
29132 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
29133 -list-thread-groups 17
29134 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
29135 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
29136 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
29137 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
29138 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
29139 -list-thread-groups --available
29140 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
29141 -list-thread-groups --available --recurse 1
29142 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
29143 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
29144 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
29145 -list-thread-groups --available --recurse 1 17 18
29146 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
29147 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
29148 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
29149 @end smallexample
29150
29151
29152 @subheading The @code{-add-inferior} Command
29153 @findex -add-inferior
29154
29155 @subheading Synopsis
29156
29157 @smallexample
29158 -add-inferior
29159 @end smallexample
29160
29161 Creates a new inferior (@pxref{Inferiors and Programs}). The created
29162 inferior is not associated with any executable. Such association may
29163 be established with the @samp{-file-exec-and-symbols} command
29164 (@pxref{GDB/MI File Commands}). The command response has a single
29165 field, @samp{thread-group}, whose value is the identifier of the
29166 thread group corresponding to the new inferior.
29167
29168 @subheading Example
29169
29170 @smallexample
29171 @value{GDBP}
29172 -add-inferior
29173 ^done,thread-group="i3"
29174 @end smallexample
29175
29176 @subheading The @code{-interpreter-exec} Command
29177 @findex -interpreter-exec
29178
29179 @subheading Synopsis
29180
29181 @smallexample
29182 -interpreter-exec @var{interpreter} @var{command}
29183 @end smallexample
29184 @anchor{-interpreter-exec}
29185
29186 Execute the specified @var{command} in the given @var{interpreter}.
29187
29188 @subheading @value{GDBN} Command
29189
29190 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
29191
29192 @subheading Example
29193
29194 @smallexample
29195 (gdb)
29196 -interpreter-exec console "break main"
29197 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
29198 &"During symbol reading, bad structure-type format.\n"
29199 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
29200 ^done
29201 (gdb)
29202 @end smallexample
29203
29204 @subheading The @code{-inferior-tty-set} Command
29205 @findex -inferior-tty-set
29206
29207 @subheading Synopsis
29208
29209 @smallexample
29210 -inferior-tty-set /dev/pts/1
29211 @end smallexample
29212
29213 Set terminal for future runs of the program being debugged.
29214
29215 @subheading @value{GDBN} Command
29216
29217 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
29218
29219 @subheading Example
29220
29221 @smallexample
29222 (gdb)
29223 -inferior-tty-set /dev/pts/1
29224 ^done
29225 (gdb)
29226 @end smallexample
29227
29228 @subheading The @code{-inferior-tty-show} Command
29229 @findex -inferior-tty-show
29230
29231 @subheading Synopsis
29232
29233 @smallexample
29234 -inferior-tty-show
29235 @end smallexample
29236
29237 Show terminal for future runs of program being debugged.
29238
29239 @subheading @value{GDBN} Command
29240
29241 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
29242
29243 @subheading Example
29244
29245 @smallexample
29246 (gdb)
29247 -inferior-tty-set /dev/pts/1
29248 ^done
29249 (gdb)
29250 -inferior-tty-show
29251 ^done,inferior_tty_terminal="/dev/pts/1"
29252 (gdb)
29253 @end smallexample
29254
29255 @subheading The @code{-enable-timings} Command
29256 @findex -enable-timings
29257
29258 @subheading Synopsis
29259
29260 @smallexample
29261 -enable-timings [yes | no]
29262 @end smallexample
29263
29264 Toggle the printing of the wallclock, user and system times for an MI
29265 command as a field in its output. This command is to help frontend
29266 developers optimize the performance of their code. No argument is
29267 equivalent to @samp{yes}.
29268
29269 @subheading @value{GDBN} Command
29270
29271 No equivalent.
29272
29273 @subheading Example
29274
29275 @smallexample
29276 (gdb)
29277 -enable-timings
29278 ^done
29279 (gdb)
29280 -break-insert main
29281 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29282 addr="0x080484ed",func="main",file="myprog.c",
29283 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
29284 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
29285 (gdb)
29286 -enable-timings no
29287 ^done
29288 (gdb)
29289 -exec-run
29290 ^running
29291 (gdb)
29292 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
29293 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
29294 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
29295 fullname="/home/nickrob/myprog.c",line="73"@}
29296 (gdb)
29297 @end smallexample
29298
29299 @node Annotations
29300 @chapter @value{GDBN} Annotations
29301
29302 This chapter describes annotations in @value{GDBN}. Annotations were
29303 designed to interface @value{GDBN} to graphical user interfaces or other
29304 similar programs which want to interact with @value{GDBN} at a
29305 relatively high level.
29306
29307 The annotation mechanism has largely been superseded by @sc{gdb/mi}
29308 (@pxref{GDB/MI}).
29309
29310 @ignore
29311 This is Edition @value{EDITION}, @value{DATE}.
29312 @end ignore
29313
29314 @menu
29315 * Annotations Overview:: What annotations are; the general syntax.
29316 * Server Prefix:: Issuing a command without affecting user state.
29317 * Prompting:: Annotations marking @value{GDBN}'s need for input.
29318 * Errors:: Annotations for error messages.
29319 * Invalidation:: Some annotations describe things now invalid.
29320 * Annotations for Running::
29321 Whether the program is running, how it stopped, etc.
29322 * Source Annotations:: Annotations describing source code.
29323 @end menu
29324
29325 @node Annotations Overview
29326 @section What is an Annotation?
29327 @cindex annotations
29328
29329 Annotations start with a newline character, two @samp{control-z}
29330 characters, and the name of the annotation. If there is no additional
29331 information associated with this annotation, the name of the annotation
29332 is followed immediately by a newline. If there is additional
29333 information, the name of the annotation is followed by a space, the
29334 additional information, and a newline. The additional information
29335 cannot contain newline characters.
29336
29337 Any output not beginning with a newline and two @samp{control-z}
29338 characters denotes literal output from @value{GDBN}. Currently there is
29339 no need for @value{GDBN} to output a newline followed by two
29340 @samp{control-z} characters, but if there was such a need, the
29341 annotations could be extended with an @samp{escape} annotation which
29342 means those three characters as output.
29343
29344 The annotation @var{level}, which is specified using the
29345 @option{--annotate} command line option (@pxref{Mode Options}), controls
29346 how much information @value{GDBN} prints together with its prompt,
29347 values of expressions, source lines, and other types of output. Level 0
29348 is for no annotations, level 1 is for use when @value{GDBN} is run as a
29349 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
29350 for programs that control @value{GDBN}, and level 2 annotations have
29351 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
29352 Interface, annotate, GDB's Obsolete Annotations}).
29353
29354 @table @code
29355 @kindex set annotate
29356 @item set annotate @var{level}
29357 The @value{GDBN} command @code{set annotate} sets the level of
29358 annotations to the specified @var{level}.
29359
29360 @item show annotate
29361 @kindex show annotate
29362 Show the current annotation level.
29363 @end table
29364
29365 This chapter describes level 3 annotations.
29366
29367 A simple example of starting up @value{GDBN} with annotations is:
29368
29369 @smallexample
29370 $ @kbd{gdb --annotate=3}
29371 GNU gdb 6.0
29372 Copyright 2003 Free Software Foundation, Inc.
29373 GDB is free software, covered by the GNU General Public License,
29374 and you are welcome to change it and/or distribute copies of it
29375 under certain conditions.
29376 Type "show copying" to see the conditions.
29377 There is absolutely no warranty for GDB. Type "show warranty"
29378 for details.
29379 This GDB was configured as "i386-pc-linux-gnu"
29380
29381 ^Z^Zpre-prompt
29382 (@value{GDBP})
29383 ^Z^Zprompt
29384 @kbd{quit}
29385
29386 ^Z^Zpost-prompt
29387 $
29388 @end smallexample
29389
29390 Here @samp{quit} is input to @value{GDBN}; the rest is output from
29391 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
29392 denotes a @samp{control-z} character) are annotations; the rest is
29393 output from @value{GDBN}.
29394
29395 @node Server Prefix
29396 @section The Server Prefix
29397 @cindex server prefix
29398
29399 If you prefix a command with @samp{server } then it will not affect
29400 the command history, nor will it affect @value{GDBN}'s notion of which
29401 command to repeat if @key{RET} is pressed on a line by itself. This
29402 means that commands can be run behind a user's back by a front-end in
29403 a transparent manner.
29404
29405 The @code{server } prefix does not affect the recording of values into
29406 the value history; to print a value without recording it into the
29407 value history, use the @code{output} command instead of the
29408 @code{print} command.
29409
29410 Using this prefix also disables confirmation requests
29411 (@pxref{confirmation requests}).
29412
29413 @node Prompting
29414 @section Annotation for @value{GDBN} Input
29415
29416 @cindex annotations for prompts
29417 When @value{GDBN} prompts for input, it annotates this fact so it is possible
29418 to know when to send output, when the output from a given command is
29419 over, etc.
29420
29421 Different kinds of input each have a different @dfn{input type}. Each
29422 input type has three annotations: a @code{pre-} annotation, which
29423 denotes the beginning of any prompt which is being output, a plain
29424 annotation, which denotes the end of the prompt, and then a @code{post-}
29425 annotation which denotes the end of any echo which may (or may not) be
29426 associated with the input. For example, the @code{prompt} input type
29427 features the following annotations:
29428
29429 @smallexample
29430 ^Z^Zpre-prompt
29431 ^Z^Zprompt
29432 ^Z^Zpost-prompt
29433 @end smallexample
29434
29435 The input types are
29436
29437 @table @code
29438 @findex pre-prompt annotation
29439 @findex prompt annotation
29440 @findex post-prompt annotation
29441 @item prompt
29442 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
29443
29444 @findex pre-commands annotation
29445 @findex commands annotation
29446 @findex post-commands annotation
29447 @item commands
29448 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
29449 command. The annotations are repeated for each command which is input.
29450
29451 @findex pre-overload-choice annotation
29452 @findex overload-choice annotation
29453 @findex post-overload-choice annotation
29454 @item overload-choice
29455 When @value{GDBN} wants the user to select between various overloaded functions.
29456
29457 @findex pre-query annotation
29458 @findex query annotation
29459 @findex post-query annotation
29460 @item query
29461 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
29462
29463 @findex pre-prompt-for-continue annotation
29464 @findex prompt-for-continue annotation
29465 @findex post-prompt-for-continue annotation
29466 @item prompt-for-continue
29467 When @value{GDBN} is asking the user to press return to continue. Note: Don't
29468 expect this to work well; instead use @code{set height 0} to disable
29469 prompting. This is because the counting of lines is buggy in the
29470 presence of annotations.
29471 @end table
29472
29473 @node Errors
29474 @section Errors
29475 @cindex annotations for errors, warnings and interrupts
29476
29477 @findex quit annotation
29478 @smallexample
29479 ^Z^Zquit
29480 @end smallexample
29481
29482 This annotation occurs right before @value{GDBN} responds to an interrupt.
29483
29484 @findex error annotation
29485 @smallexample
29486 ^Z^Zerror
29487 @end smallexample
29488
29489 This annotation occurs right before @value{GDBN} responds to an error.
29490
29491 Quit and error annotations indicate that any annotations which @value{GDBN} was
29492 in the middle of may end abruptly. For example, if a
29493 @code{value-history-begin} annotation is followed by a @code{error}, one
29494 cannot expect to receive the matching @code{value-history-end}. One
29495 cannot expect not to receive it either, however; an error annotation
29496 does not necessarily mean that @value{GDBN} is immediately returning all the way
29497 to the top level.
29498
29499 @findex error-begin annotation
29500 A quit or error annotation may be preceded by
29501
29502 @smallexample
29503 ^Z^Zerror-begin
29504 @end smallexample
29505
29506 Any output between that and the quit or error annotation is the error
29507 message.
29508
29509 Warning messages are not yet annotated.
29510 @c If we want to change that, need to fix warning(), type_error(),
29511 @c range_error(), and possibly other places.
29512
29513 @node Invalidation
29514 @section Invalidation Notices
29515
29516 @cindex annotations for invalidation messages
29517 The following annotations say that certain pieces of state may have
29518 changed.
29519
29520 @table @code
29521 @findex frames-invalid annotation
29522 @item ^Z^Zframes-invalid
29523
29524 The frames (for example, output from the @code{backtrace} command) may
29525 have changed.
29526
29527 @findex breakpoints-invalid annotation
29528 @item ^Z^Zbreakpoints-invalid
29529
29530 The breakpoints may have changed. For example, the user just added or
29531 deleted a breakpoint.
29532 @end table
29533
29534 @node Annotations for Running
29535 @section Running the Program
29536 @cindex annotations for running programs
29537
29538 @findex starting annotation
29539 @findex stopping annotation
29540 When the program starts executing due to a @value{GDBN} command such as
29541 @code{step} or @code{continue},
29542
29543 @smallexample
29544 ^Z^Zstarting
29545 @end smallexample
29546
29547 is output. When the program stops,
29548
29549 @smallexample
29550 ^Z^Zstopped
29551 @end smallexample
29552
29553 is output. Before the @code{stopped} annotation, a variety of
29554 annotations describe how the program stopped.
29555
29556 @table @code
29557 @findex exited annotation
29558 @item ^Z^Zexited @var{exit-status}
29559 The program exited, and @var{exit-status} is the exit status (zero for
29560 successful exit, otherwise nonzero).
29561
29562 @findex signalled annotation
29563 @findex signal-name annotation
29564 @findex signal-name-end annotation
29565 @findex signal-string annotation
29566 @findex signal-string-end annotation
29567 @item ^Z^Zsignalled
29568 The program exited with a signal. After the @code{^Z^Zsignalled}, the
29569 annotation continues:
29570
29571 @smallexample
29572 @var{intro-text}
29573 ^Z^Zsignal-name
29574 @var{name}
29575 ^Z^Zsignal-name-end
29576 @var{middle-text}
29577 ^Z^Zsignal-string
29578 @var{string}
29579 ^Z^Zsignal-string-end
29580 @var{end-text}
29581 @end smallexample
29582
29583 @noindent
29584 where @var{name} is the name of the signal, such as @code{SIGILL} or
29585 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
29586 as @code{Illegal Instruction} or @code{Segmentation fault}.
29587 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
29588 user's benefit and have no particular format.
29589
29590 @findex signal annotation
29591 @item ^Z^Zsignal
29592 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
29593 just saying that the program received the signal, not that it was
29594 terminated with it.
29595
29596 @findex breakpoint annotation
29597 @item ^Z^Zbreakpoint @var{number}
29598 The program hit breakpoint number @var{number}.
29599
29600 @findex watchpoint annotation
29601 @item ^Z^Zwatchpoint @var{number}
29602 The program hit watchpoint number @var{number}.
29603 @end table
29604
29605 @node Source Annotations
29606 @section Displaying Source
29607 @cindex annotations for source display
29608
29609 @findex source annotation
29610 The following annotation is used instead of displaying source code:
29611
29612 @smallexample
29613 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
29614 @end smallexample
29615
29616 where @var{filename} is an absolute file name indicating which source
29617 file, @var{line} is the line number within that file (where 1 is the
29618 first line in the file), @var{character} is the character position
29619 within the file (where 0 is the first character in the file) (for most
29620 debug formats this will necessarily point to the beginning of a line),
29621 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
29622 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
29623 @var{addr} is the address in the target program associated with the
29624 source which is being displayed. @var{addr} is in the form @samp{0x}
29625 followed by one or more lowercase hex digits (note that this does not
29626 depend on the language).
29627
29628 @node JIT Interface
29629 @chapter JIT Compilation Interface
29630 @cindex just-in-time compilation
29631 @cindex JIT compilation interface
29632
29633 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
29634 interface. A JIT compiler is a program or library that generates native
29635 executable code at runtime and executes it, usually in order to achieve good
29636 performance while maintaining platform independence.
29637
29638 Programs that use JIT compilation are normally difficult to debug because
29639 portions of their code are generated at runtime, instead of being loaded from
29640 object files, which is where @value{GDBN} normally finds the program's symbols
29641 and debug information. In order to debug programs that use JIT compilation,
29642 @value{GDBN} has an interface that allows the program to register in-memory
29643 symbol files with @value{GDBN} at runtime.
29644
29645 If you are using @value{GDBN} to debug a program that uses this interface, then
29646 it should work transparently so long as you have not stripped the binary. If
29647 you are developing a JIT compiler, then the interface is documented in the rest
29648 of this chapter. At this time, the only known client of this interface is the
29649 LLVM JIT.
29650
29651 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
29652 JIT compiler communicates with @value{GDBN} by writing data into a global
29653 variable and calling a fuction at a well-known symbol. When @value{GDBN}
29654 attaches, it reads a linked list of symbol files from the global variable to
29655 find existing code, and puts a breakpoint in the function so that it can find
29656 out about additional code.
29657
29658 @menu
29659 * Declarations:: Relevant C struct declarations
29660 * Registering Code:: Steps to register code
29661 * Unregistering Code:: Steps to unregister code
29662 @end menu
29663
29664 @node Declarations
29665 @section JIT Declarations
29666
29667 These are the relevant struct declarations that a C program should include to
29668 implement the interface:
29669
29670 @smallexample
29671 typedef enum
29672 @{
29673 JIT_NOACTION = 0,
29674 JIT_REGISTER_FN,
29675 JIT_UNREGISTER_FN
29676 @} jit_actions_t;
29677
29678 struct jit_code_entry
29679 @{
29680 struct jit_code_entry *next_entry;
29681 struct jit_code_entry *prev_entry;
29682 const char *symfile_addr;
29683 uint64_t symfile_size;
29684 @};
29685
29686 struct jit_descriptor
29687 @{
29688 uint32_t version;
29689 /* This type should be jit_actions_t, but we use uint32_t
29690 to be explicit about the bitwidth. */
29691 uint32_t action_flag;
29692 struct jit_code_entry *relevant_entry;
29693 struct jit_code_entry *first_entry;
29694 @};
29695
29696 /* GDB puts a breakpoint in this function. */
29697 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
29698
29699 /* Make sure to specify the version statically, because the
29700 debugger may check the version before we can set it. */
29701 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
29702 @end smallexample
29703
29704 If the JIT is multi-threaded, then it is important that the JIT synchronize any
29705 modifications to this global data properly, which can easily be done by putting
29706 a global mutex around modifications to these structures.
29707
29708 @node Registering Code
29709 @section Registering Code
29710
29711 To register code with @value{GDBN}, the JIT should follow this protocol:
29712
29713 @itemize @bullet
29714 @item
29715 Generate an object file in memory with symbols and other desired debug
29716 information. The file must include the virtual addresses of the sections.
29717
29718 @item
29719 Create a code entry for the file, which gives the start and size of the symbol
29720 file.
29721
29722 @item
29723 Add it to the linked list in the JIT descriptor.
29724
29725 @item
29726 Point the relevant_entry field of the descriptor at the entry.
29727
29728 @item
29729 Set @code{action_flag} to @code{JIT_REGISTER} and call
29730 @code{__jit_debug_register_code}.
29731 @end itemize
29732
29733 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
29734 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
29735 new code. However, the linked list must still be maintained in order to allow
29736 @value{GDBN} to attach to a running process and still find the symbol files.
29737
29738 @node Unregistering Code
29739 @section Unregistering Code
29740
29741 If code is freed, then the JIT should use the following protocol:
29742
29743 @itemize @bullet
29744 @item
29745 Remove the code entry corresponding to the code from the linked list.
29746
29747 @item
29748 Point the @code{relevant_entry} field of the descriptor at the code entry.
29749
29750 @item
29751 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
29752 @code{__jit_debug_register_code}.
29753 @end itemize
29754
29755 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
29756 and the JIT will leak the memory used for the associated symbol files.
29757
29758 @node GDB Bugs
29759 @chapter Reporting Bugs in @value{GDBN}
29760 @cindex bugs in @value{GDBN}
29761 @cindex reporting bugs in @value{GDBN}
29762
29763 Your bug reports play an essential role in making @value{GDBN} reliable.
29764
29765 Reporting a bug may help you by bringing a solution to your problem, or it
29766 may not. But in any case the principal function of a bug report is to help
29767 the entire community by making the next version of @value{GDBN} work better. Bug
29768 reports are your contribution to the maintenance of @value{GDBN}.
29769
29770 In order for a bug report to serve its purpose, you must include the
29771 information that enables us to fix the bug.
29772
29773 @menu
29774 * Bug Criteria:: Have you found a bug?
29775 * Bug Reporting:: How to report bugs
29776 @end menu
29777
29778 @node Bug Criteria
29779 @section Have You Found a Bug?
29780 @cindex bug criteria
29781
29782 If you are not sure whether you have found a bug, here are some guidelines:
29783
29784 @itemize @bullet
29785 @cindex fatal signal
29786 @cindex debugger crash
29787 @cindex crash of debugger
29788 @item
29789 If the debugger gets a fatal signal, for any input whatever, that is a
29790 @value{GDBN} bug. Reliable debuggers never crash.
29791
29792 @cindex error on valid input
29793 @item
29794 If @value{GDBN} produces an error message for valid input, that is a
29795 bug. (Note that if you're cross debugging, the problem may also be
29796 somewhere in the connection to the target.)
29797
29798 @cindex invalid input
29799 @item
29800 If @value{GDBN} does not produce an error message for invalid input,
29801 that is a bug. However, you should note that your idea of
29802 ``invalid input'' might be our idea of ``an extension'' or ``support
29803 for traditional practice''.
29804
29805 @item
29806 If you are an experienced user of debugging tools, your suggestions
29807 for improvement of @value{GDBN} are welcome in any case.
29808 @end itemize
29809
29810 @node Bug Reporting
29811 @section How to Report Bugs
29812 @cindex bug reports
29813 @cindex @value{GDBN} bugs, reporting
29814
29815 A number of companies and individuals offer support for @sc{gnu} products.
29816 If you obtained @value{GDBN} from a support organization, we recommend you
29817 contact that organization first.
29818
29819 You can find contact information for many support companies and
29820 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
29821 distribution.
29822 @c should add a web page ref...
29823
29824 @ifset BUGURL
29825 @ifset BUGURL_DEFAULT
29826 In any event, we also recommend that you submit bug reports for
29827 @value{GDBN}. The preferred method is to submit them directly using
29828 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
29829 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
29830 be used.
29831
29832 @strong{Do not send bug reports to @samp{info-gdb}, or to
29833 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
29834 not want to receive bug reports. Those that do have arranged to receive
29835 @samp{bug-gdb}.
29836
29837 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
29838 serves as a repeater. The mailing list and the newsgroup carry exactly
29839 the same messages. Often people think of posting bug reports to the
29840 newsgroup instead of mailing them. This appears to work, but it has one
29841 problem which can be crucial: a newsgroup posting often lacks a mail
29842 path back to the sender. Thus, if we need to ask for more information,
29843 we may be unable to reach you. For this reason, it is better to send
29844 bug reports to the mailing list.
29845 @end ifset
29846 @ifclear BUGURL_DEFAULT
29847 In any event, we also recommend that you submit bug reports for
29848 @value{GDBN} to @value{BUGURL}.
29849 @end ifclear
29850 @end ifset
29851
29852 The fundamental principle of reporting bugs usefully is this:
29853 @strong{report all the facts}. If you are not sure whether to state a
29854 fact or leave it out, state it!
29855
29856 Often people omit facts because they think they know what causes the
29857 problem and assume that some details do not matter. Thus, you might
29858 assume that the name of the variable you use in an example does not matter.
29859 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
29860 stray memory reference which happens to fetch from the location where that
29861 name is stored in memory; perhaps, if the name were different, the contents
29862 of that location would fool the debugger into doing the right thing despite
29863 the bug. Play it safe and give a specific, complete example. That is the
29864 easiest thing for you to do, and the most helpful.
29865
29866 Keep in mind that the purpose of a bug report is to enable us to fix the
29867 bug. It may be that the bug has been reported previously, but neither
29868 you nor we can know that unless your bug report is complete and
29869 self-contained.
29870
29871 Sometimes people give a few sketchy facts and ask, ``Does this ring a
29872 bell?'' Those bug reports are useless, and we urge everyone to
29873 @emph{refuse to respond to them} except to chide the sender to report
29874 bugs properly.
29875
29876 To enable us to fix the bug, you should include all these things:
29877
29878 @itemize @bullet
29879 @item
29880 The version of @value{GDBN}. @value{GDBN} announces it if you start
29881 with no arguments; you can also print it at any time using @code{show
29882 version}.
29883
29884 Without this, we will not know whether there is any point in looking for
29885 the bug in the current version of @value{GDBN}.
29886
29887 @item
29888 The type of machine you are using, and the operating system name and
29889 version number.
29890
29891 @item
29892 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
29893 ``@value{GCC}--2.8.1''.
29894
29895 @item
29896 What compiler (and its version) was used to compile the program you are
29897 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
29898 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
29899 to get this information; for other compilers, see the documentation for
29900 those compilers.
29901
29902 @item
29903 The command arguments you gave the compiler to compile your example and
29904 observe the bug. For example, did you use @samp{-O}? To guarantee
29905 you will not omit something important, list them all. A copy of the
29906 Makefile (or the output from make) is sufficient.
29907
29908 If we were to try to guess the arguments, we would probably guess wrong
29909 and then we might not encounter the bug.
29910
29911 @item
29912 A complete input script, and all necessary source files, that will
29913 reproduce the bug.
29914
29915 @item
29916 A description of what behavior you observe that you believe is
29917 incorrect. For example, ``It gets a fatal signal.''
29918
29919 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
29920 will certainly notice it. But if the bug is incorrect output, we might
29921 not notice unless it is glaringly wrong. You might as well not give us
29922 a chance to make a mistake.
29923
29924 Even if the problem you experience is a fatal signal, you should still
29925 say so explicitly. Suppose something strange is going on, such as, your
29926 copy of @value{GDBN} is out of synch, or you have encountered a bug in
29927 the C library on your system. (This has happened!) Your copy might
29928 crash and ours would not. If you told us to expect a crash, then when
29929 ours fails to crash, we would know that the bug was not happening for
29930 us. If you had not told us to expect a crash, then we would not be able
29931 to draw any conclusion from our observations.
29932
29933 @pindex script
29934 @cindex recording a session script
29935 To collect all this information, you can use a session recording program
29936 such as @command{script}, which is available on many Unix systems.
29937 Just run your @value{GDBN} session inside @command{script} and then
29938 include the @file{typescript} file with your bug report.
29939
29940 Another way to record a @value{GDBN} session is to run @value{GDBN}
29941 inside Emacs and then save the entire buffer to a file.
29942
29943 @item
29944 If you wish to suggest changes to the @value{GDBN} source, send us context
29945 diffs. If you even discuss something in the @value{GDBN} source, refer to
29946 it by context, not by line number.
29947
29948 The line numbers in our development sources will not match those in your
29949 sources. Your line numbers would convey no useful information to us.
29950
29951 @end itemize
29952
29953 Here are some things that are not necessary:
29954
29955 @itemize @bullet
29956 @item
29957 A description of the envelope of the bug.
29958
29959 Often people who encounter a bug spend a lot of time investigating
29960 which changes to the input file will make the bug go away and which
29961 changes will not affect it.
29962
29963 This is often time consuming and not very useful, because the way we
29964 will find the bug is by running a single example under the debugger
29965 with breakpoints, not by pure deduction from a series of examples.
29966 We recommend that you save your time for something else.
29967
29968 Of course, if you can find a simpler example to report @emph{instead}
29969 of the original one, that is a convenience for us. Errors in the
29970 output will be easier to spot, running under the debugger will take
29971 less time, and so on.
29972
29973 However, simplification is not vital; if you do not want to do this,
29974 report the bug anyway and send us the entire test case you used.
29975
29976 @item
29977 A patch for the bug.
29978
29979 A patch for the bug does help us if it is a good one. But do not omit
29980 the necessary information, such as the test case, on the assumption that
29981 a patch is all we need. We might see problems with your patch and decide
29982 to fix the problem another way, or we might not understand it at all.
29983
29984 Sometimes with a program as complicated as @value{GDBN} it is very hard to
29985 construct an example that will make the program follow a certain path
29986 through the code. If you do not send us the example, we will not be able
29987 to construct one, so we will not be able to verify that the bug is fixed.
29988
29989 And if we cannot understand what bug you are trying to fix, or why your
29990 patch should be an improvement, we will not install it. A test case will
29991 help us to understand.
29992
29993 @item
29994 A guess about what the bug is or what it depends on.
29995
29996 Such guesses are usually wrong. Even we cannot guess right about such
29997 things without first using the debugger to find the facts.
29998 @end itemize
29999
30000 @c The readline documentation is distributed with the readline code
30001 @c and consists of the two following files:
30002 @c rluser.texinfo
30003 @c inc-hist.texinfo
30004 @c Use -I with makeinfo to point to the appropriate directory,
30005 @c environment var TEXINPUTS with TeX.
30006 @include rluser.texi
30007 @include inc-hist.texinfo
30008
30009
30010 @node Formatting Documentation
30011 @appendix Formatting Documentation
30012
30013 @cindex @value{GDBN} reference card
30014 @cindex reference card
30015 The @value{GDBN} 4 release includes an already-formatted reference card, ready
30016 for printing with PostScript or Ghostscript, in the @file{gdb}
30017 subdirectory of the main source directory@footnote{In
30018 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
30019 release.}. If you can use PostScript or Ghostscript with your printer,
30020 you can print the reference card immediately with @file{refcard.ps}.
30021
30022 The release also includes the source for the reference card. You
30023 can format it, using @TeX{}, by typing:
30024
30025 @smallexample
30026 make refcard.dvi
30027 @end smallexample
30028
30029 The @value{GDBN} reference card is designed to print in @dfn{landscape}
30030 mode on US ``letter'' size paper;
30031 that is, on a sheet 11 inches wide by 8.5 inches
30032 high. You will need to specify this form of printing as an option to
30033 your @sc{dvi} output program.
30034
30035 @cindex documentation
30036
30037 All the documentation for @value{GDBN} comes as part of the machine-readable
30038 distribution. The documentation is written in Texinfo format, which is
30039 a documentation system that uses a single source file to produce both
30040 on-line information and a printed manual. You can use one of the Info
30041 formatting commands to create the on-line version of the documentation
30042 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
30043
30044 @value{GDBN} includes an already formatted copy of the on-line Info
30045 version of this manual in the @file{gdb} subdirectory. The main Info
30046 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
30047 subordinate files matching @samp{gdb.info*} in the same directory. If
30048 necessary, you can print out these files, or read them with any editor;
30049 but they are easier to read using the @code{info} subsystem in @sc{gnu}
30050 Emacs or the standalone @code{info} program, available as part of the
30051 @sc{gnu} Texinfo distribution.
30052
30053 If you want to format these Info files yourself, you need one of the
30054 Info formatting programs, such as @code{texinfo-format-buffer} or
30055 @code{makeinfo}.
30056
30057 If you have @code{makeinfo} installed, and are in the top level
30058 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
30059 version @value{GDBVN}), you can make the Info file by typing:
30060
30061 @smallexample
30062 cd gdb
30063 make gdb.info
30064 @end smallexample
30065
30066 If you want to typeset and print copies of this manual, you need @TeX{},
30067 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
30068 Texinfo definitions file.
30069
30070 @TeX{} is a typesetting program; it does not print files directly, but
30071 produces output files called @sc{dvi} files. To print a typeset
30072 document, you need a program to print @sc{dvi} files. If your system
30073 has @TeX{} installed, chances are it has such a program. The precise
30074 command to use depends on your system; @kbd{lpr -d} is common; another
30075 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
30076 require a file name without any extension or a @samp{.dvi} extension.
30077
30078 @TeX{} also requires a macro definitions file called
30079 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
30080 written in Texinfo format. On its own, @TeX{} cannot either read or
30081 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
30082 and is located in the @file{gdb-@var{version-number}/texinfo}
30083 directory.
30084
30085 If you have @TeX{} and a @sc{dvi} printer program installed, you can
30086 typeset and print this manual. First switch to the @file{gdb}
30087 subdirectory of the main source directory (for example, to
30088 @file{gdb-@value{GDBVN}/gdb}) and type:
30089
30090 @smallexample
30091 make gdb.dvi
30092 @end smallexample
30093
30094 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
30095
30096 @node Installing GDB
30097 @appendix Installing @value{GDBN}
30098 @cindex installation
30099
30100 @menu
30101 * Requirements:: Requirements for building @value{GDBN}
30102 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
30103 * Separate Objdir:: Compiling @value{GDBN} in another directory
30104 * Config Names:: Specifying names for hosts and targets
30105 * Configure Options:: Summary of options for configure
30106 * System-wide configuration:: Having a system-wide init file
30107 @end menu
30108
30109 @node Requirements
30110 @section Requirements for Building @value{GDBN}
30111 @cindex building @value{GDBN}, requirements for
30112
30113 Building @value{GDBN} requires various tools and packages to be available.
30114 Other packages will be used only if they are found.
30115
30116 @heading Tools/Packages Necessary for Building @value{GDBN}
30117 @table @asis
30118 @item ISO C90 compiler
30119 @value{GDBN} is written in ISO C90. It should be buildable with any
30120 working C90 compiler, e.g.@: GCC.
30121
30122 @end table
30123
30124 @heading Tools/Packages Optional for Building @value{GDBN}
30125 @table @asis
30126 @item Expat
30127 @anchor{Expat}
30128 @value{GDBN} can use the Expat XML parsing library. This library may be
30129 included with your operating system distribution; if it is not, you
30130 can get the latest version from @url{http://expat.sourceforge.net}.
30131 The @file{configure} script will search for this library in several
30132 standard locations; if it is installed in an unusual path, you can
30133 use the @option{--with-libexpat-prefix} option to specify its location.
30134
30135 Expat is used for:
30136
30137 @itemize @bullet
30138 @item
30139 Remote protocol memory maps (@pxref{Memory Map Format})
30140 @item
30141 Target descriptions (@pxref{Target Descriptions})
30142 @item
30143 Remote shared library lists (@pxref{Library List Format})
30144 @item
30145 MS-Windows shared libraries (@pxref{Shared Libraries})
30146 @end itemize
30147
30148 @item zlib
30149 @cindex compressed debug sections
30150 @value{GDBN} will use the @samp{zlib} library, if available, to read
30151 compressed debug sections. Some linkers, such as GNU gold, are capable
30152 of producing binaries with compressed debug sections. If @value{GDBN}
30153 is compiled with @samp{zlib}, it will be able to read the debug
30154 information in such binaries.
30155
30156 The @samp{zlib} library is likely included with your operating system
30157 distribution; if it is not, you can get the latest version from
30158 @url{http://zlib.net}.
30159
30160 @item iconv
30161 @value{GDBN}'s features related to character sets (@pxref{Character
30162 Sets}) require a functioning @code{iconv} implementation. If you are
30163 on a GNU system, then this is provided by the GNU C Library. Some
30164 other systems also provide a working @code{iconv}.
30165
30166 On systems with @code{iconv}, you can install GNU Libiconv. If you
30167 have previously installed Libiconv, you can use the
30168 @option{--with-libiconv-prefix} option to configure.
30169
30170 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
30171 arrange to build Libiconv if a directory named @file{libiconv} appears
30172 in the top-most source directory. If Libiconv is built this way, and
30173 if the operating system does not provide a suitable @code{iconv}
30174 implementation, then the just-built library will automatically be used
30175 by @value{GDBN}. One easy way to set this up is to download GNU
30176 Libiconv, unpack it, and then rename the directory holding the
30177 Libiconv source code to @samp{libiconv}.
30178 @end table
30179
30180 @node Running Configure
30181 @section Invoking the @value{GDBN} @file{configure} Script
30182 @cindex configuring @value{GDBN}
30183 @value{GDBN} comes with a @file{configure} script that automates the process
30184 of preparing @value{GDBN} for installation; you can then use @code{make} to
30185 build the @code{gdb} program.
30186 @iftex
30187 @c irrelevant in info file; it's as current as the code it lives with.
30188 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
30189 look at the @file{README} file in the sources; we may have improved the
30190 installation procedures since publishing this manual.}
30191 @end iftex
30192
30193 The @value{GDBN} distribution includes all the source code you need for
30194 @value{GDBN} in a single directory, whose name is usually composed by
30195 appending the version number to @samp{gdb}.
30196
30197 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
30198 @file{gdb-@value{GDBVN}} directory. That directory contains:
30199
30200 @table @code
30201 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
30202 script for configuring @value{GDBN} and all its supporting libraries
30203
30204 @item gdb-@value{GDBVN}/gdb
30205 the source specific to @value{GDBN} itself
30206
30207 @item gdb-@value{GDBVN}/bfd
30208 source for the Binary File Descriptor library
30209
30210 @item gdb-@value{GDBVN}/include
30211 @sc{gnu} include files
30212
30213 @item gdb-@value{GDBVN}/libiberty
30214 source for the @samp{-liberty} free software library
30215
30216 @item gdb-@value{GDBVN}/opcodes
30217 source for the library of opcode tables and disassemblers
30218
30219 @item gdb-@value{GDBVN}/readline
30220 source for the @sc{gnu} command-line interface
30221
30222 @item gdb-@value{GDBVN}/glob
30223 source for the @sc{gnu} filename pattern-matching subroutine
30224
30225 @item gdb-@value{GDBVN}/mmalloc
30226 source for the @sc{gnu} memory-mapped malloc package
30227 @end table
30228
30229 The simplest way to configure and build @value{GDBN} is to run @file{configure}
30230 from the @file{gdb-@var{version-number}} source directory, which in
30231 this example is the @file{gdb-@value{GDBVN}} directory.
30232
30233 First switch to the @file{gdb-@var{version-number}} source directory
30234 if you are not already in it; then run @file{configure}. Pass the
30235 identifier for the platform on which @value{GDBN} will run as an
30236 argument.
30237
30238 For example:
30239
30240 @smallexample
30241 cd gdb-@value{GDBVN}
30242 ./configure @var{host}
30243 make
30244 @end smallexample
30245
30246 @noindent
30247 where @var{host} is an identifier such as @samp{sun4} or
30248 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
30249 (You can often leave off @var{host}; @file{configure} tries to guess the
30250 correct value by examining your system.)
30251
30252 Running @samp{configure @var{host}} and then running @code{make} builds the
30253 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
30254 libraries, then @code{gdb} itself. The configured source files, and the
30255 binaries, are left in the corresponding source directories.
30256
30257 @need 750
30258 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
30259 system does not recognize this automatically when you run a different
30260 shell, you may need to run @code{sh} on it explicitly:
30261
30262 @smallexample
30263 sh configure @var{host}
30264 @end smallexample
30265
30266 If you run @file{configure} from a directory that contains source
30267 directories for multiple libraries or programs, such as the
30268 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
30269 @file{configure}
30270 creates configuration files for every directory level underneath (unless
30271 you tell it not to, with the @samp{--norecursion} option).
30272
30273 You should run the @file{configure} script from the top directory in the
30274 source tree, the @file{gdb-@var{version-number}} directory. If you run
30275 @file{configure} from one of the subdirectories, you will configure only
30276 that subdirectory. That is usually not what you want. In particular,
30277 if you run the first @file{configure} from the @file{gdb} subdirectory
30278 of the @file{gdb-@var{version-number}} directory, you will omit the
30279 configuration of @file{bfd}, @file{readline}, and other sibling
30280 directories of the @file{gdb} subdirectory. This leads to build errors
30281 about missing include files such as @file{bfd/bfd.h}.
30282
30283 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
30284 However, you should make sure that the shell on your path (named by
30285 the @samp{SHELL} environment variable) is publicly readable. Remember
30286 that @value{GDBN} uses the shell to start your program---some systems refuse to
30287 let @value{GDBN} debug child processes whose programs are not readable.
30288
30289 @node Separate Objdir
30290 @section Compiling @value{GDBN} in Another Directory
30291
30292 If you want to run @value{GDBN} versions for several host or target machines,
30293 you need a different @code{gdb} compiled for each combination of
30294 host and target. @file{configure} is designed to make this easy by
30295 allowing you to generate each configuration in a separate subdirectory,
30296 rather than in the source directory. If your @code{make} program
30297 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
30298 @code{make} in each of these directories builds the @code{gdb}
30299 program specified there.
30300
30301 To build @code{gdb} in a separate directory, run @file{configure}
30302 with the @samp{--srcdir} option to specify where to find the source.
30303 (You also need to specify a path to find @file{configure}
30304 itself from your working directory. If the path to @file{configure}
30305 would be the same as the argument to @samp{--srcdir}, you can leave out
30306 the @samp{--srcdir} option; it is assumed.)
30307
30308 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
30309 separate directory for a Sun 4 like this:
30310
30311 @smallexample
30312 @group
30313 cd gdb-@value{GDBVN}
30314 mkdir ../gdb-sun4
30315 cd ../gdb-sun4
30316 ../gdb-@value{GDBVN}/configure sun4
30317 make
30318 @end group
30319 @end smallexample
30320
30321 When @file{configure} builds a configuration using a remote source
30322 directory, it creates a tree for the binaries with the same structure
30323 (and using the same names) as the tree under the source directory. In
30324 the example, you'd find the Sun 4 library @file{libiberty.a} in the
30325 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
30326 @file{gdb-sun4/gdb}.
30327
30328 Make sure that your path to the @file{configure} script has just one
30329 instance of @file{gdb} in it. If your path to @file{configure} looks
30330 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
30331 one subdirectory of @value{GDBN}, not the whole package. This leads to
30332 build errors about missing include files such as @file{bfd/bfd.h}.
30333
30334 One popular reason to build several @value{GDBN} configurations in separate
30335 directories is to configure @value{GDBN} for cross-compiling (where
30336 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
30337 programs that run on another machine---the @dfn{target}).
30338 You specify a cross-debugging target by
30339 giving the @samp{--target=@var{target}} option to @file{configure}.
30340
30341 When you run @code{make} to build a program or library, you must run
30342 it in a configured directory---whatever directory you were in when you
30343 called @file{configure} (or one of its subdirectories).
30344
30345 The @code{Makefile} that @file{configure} generates in each source
30346 directory also runs recursively. If you type @code{make} in a source
30347 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
30348 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
30349 will build all the required libraries, and then build GDB.
30350
30351 When you have multiple hosts or targets configured in separate
30352 directories, you can run @code{make} on them in parallel (for example,
30353 if they are NFS-mounted on each of the hosts); they will not interfere
30354 with each other.
30355
30356 @node Config Names
30357 @section Specifying Names for Hosts and Targets
30358
30359 The specifications used for hosts and targets in the @file{configure}
30360 script are based on a three-part naming scheme, but some short predefined
30361 aliases are also supported. The full naming scheme encodes three pieces
30362 of information in the following pattern:
30363
30364 @smallexample
30365 @var{architecture}-@var{vendor}-@var{os}
30366 @end smallexample
30367
30368 For example, you can use the alias @code{sun4} as a @var{host} argument,
30369 or as the value for @var{target} in a @code{--target=@var{target}}
30370 option. The equivalent full name is @samp{sparc-sun-sunos4}.
30371
30372 The @file{configure} script accompanying @value{GDBN} does not provide
30373 any query facility to list all supported host and target names or
30374 aliases. @file{configure} calls the Bourne shell script
30375 @code{config.sub} to map abbreviations to full names; you can read the
30376 script, if you wish, or you can use it to test your guesses on
30377 abbreviations---for example:
30378
30379 @smallexample
30380 % sh config.sub i386-linux
30381 i386-pc-linux-gnu
30382 % sh config.sub alpha-linux
30383 alpha-unknown-linux-gnu
30384 % sh config.sub hp9k700
30385 hppa1.1-hp-hpux
30386 % sh config.sub sun4
30387 sparc-sun-sunos4.1.1
30388 % sh config.sub sun3
30389 m68k-sun-sunos4.1.1
30390 % sh config.sub i986v
30391 Invalid configuration `i986v': machine `i986v' not recognized
30392 @end smallexample
30393
30394 @noindent
30395 @code{config.sub} is also distributed in the @value{GDBN} source
30396 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
30397
30398 @node Configure Options
30399 @section @file{configure} Options
30400
30401 Here is a summary of the @file{configure} options and arguments that
30402 are most often useful for building @value{GDBN}. @file{configure} also has
30403 several other options not listed here. @inforef{What Configure
30404 Does,,configure.info}, for a full explanation of @file{configure}.
30405
30406 @smallexample
30407 configure @r{[}--help@r{]}
30408 @r{[}--prefix=@var{dir}@r{]}
30409 @r{[}--exec-prefix=@var{dir}@r{]}
30410 @r{[}--srcdir=@var{dirname}@r{]}
30411 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
30412 @r{[}--target=@var{target}@r{]}
30413 @var{host}
30414 @end smallexample
30415
30416 @noindent
30417 You may introduce options with a single @samp{-} rather than
30418 @samp{--} if you prefer; but you may abbreviate option names if you use
30419 @samp{--}.
30420
30421 @table @code
30422 @item --help
30423 Display a quick summary of how to invoke @file{configure}.
30424
30425 @item --prefix=@var{dir}
30426 Configure the source to install programs and files under directory
30427 @file{@var{dir}}.
30428
30429 @item --exec-prefix=@var{dir}
30430 Configure the source to install programs under directory
30431 @file{@var{dir}}.
30432
30433 @c avoid splitting the warning from the explanation:
30434 @need 2000
30435 @item --srcdir=@var{dirname}
30436 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
30437 @code{make} that implements the @code{VPATH} feature.}@*
30438 Use this option to make configurations in directories separate from the
30439 @value{GDBN} source directories. Among other things, you can use this to
30440 build (or maintain) several configurations simultaneously, in separate
30441 directories. @file{configure} writes configuration-specific files in
30442 the current directory, but arranges for them to use the source in the
30443 directory @var{dirname}. @file{configure} creates directories under
30444 the working directory in parallel to the source directories below
30445 @var{dirname}.
30446
30447 @item --norecursion
30448 Configure only the directory level where @file{configure} is executed; do not
30449 propagate configuration to subdirectories.
30450
30451 @item --target=@var{target}
30452 Configure @value{GDBN} for cross-debugging programs running on the specified
30453 @var{target}. Without this option, @value{GDBN} is configured to debug
30454 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
30455
30456 There is no convenient way to generate a list of all available targets.
30457
30458 @item @var{host} @dots{}
30459 Configure @value{GDBN} to run on the specified @var{host}.
30460
30461 There is no convenient way to generate a list of all available hosts.
30462 @end table
30463
30464 There are many other options available as well, but they are generally
30465 needed for special purposes only.
30466
30467 @node System-wide configuration
30468 @section System-wide configuration and settings
30469 @cindex system-wide init file
30470
30471 @value{GDBN} can be configured to have a system-wide init file;
30472 this file will be read and executed at startup (@pxref{Startup, , What
30473 @value{GDBN} does during startup}).
30474
30475 Here is the corresponding configure option:
30476
30477 @table @code
30478 @item --with-system-gdbinit=@var{file}
30479 Specify that the default location of the system-wide init file is
30480 @var{file}.
30481 @end table
30482
30483 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
30484 it may be subject to relocation. Two possible cases:
30485
30486 @itemize @bullet
30487 @item
30488 If the default location of this init file contains @file{$prefix},
30489 it will be subject to relocation. Suppose that the configure options
30490 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
30491 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
30492 init file is looked for as @file{$install/etc/gdbinit} instead of
30493 @file{$prefix/etc/gdbinit}.
30494
30495 @item
30496 By contrast, if the default location does not contain the prefix,
30497 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
30498 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
30499 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
30500 wherever @value{GDBN} is installed.
30501 @end itemize
30502
30503 @node Maintenance Commands
30504 @appendix Maintenance Commands
30505 @cindex maintenance commands
30506 @cindex internal commands
30507
30508 In addition to commands intended for @value{GDBN} users, @value{GDBN}
30509 includes a number of commands intended for @value{GDBN} developers,
30510 that are not documented elsewhere in this manual. These commands are
30511 provided here for reference. (For commands that turn on debugging
30512 messages, see @ref{Debugging Output}.)
30513
30514 @table @code
30515 @kindex maint agent
30516 @kindex maint agent-eval
30517 @item maint agent @var{expression}
30518 @itemx maint agent-eval @var{expression}
30519 Translate the given @var{expression} into remote agent bytecodes.
30520 This command is useful for debugging the Agent Expression mechanism
30521 (@pxref{Agent Expressions}). The @samp{agent} version produces an
30522 expression useful for data collection, such as by tracepoints, while
30523 @samp{maint agent-eval} produces an expression that evaluates directly
30524 to a result. For instance, a collection expression for @code{globa +
30525 globb} will include bytecodes to record four bytes of memory at each
30526 of the addresses of @code{globa} and @code{globb}, while discarding
30527 the result of the addition, while an evaluation expression will do the
30528 addition and return the sum.
30529
30530 @kindex maint info breakpoints
30531 @item @anchor{maint info breakpoints}maint info breakpoints
30532 Using the same format as @samp{info breakpoints}, display both the
30533 breakpoints you've set explicitly, and those @value{GDBN} is using for
30534 internal purposes. Internal breakpoints are shown with negative
30535 breakpoint numbers. The type column identifies what kind of breakpoint
30536 is shown:
30537
30538 @table @code
30539 @item breakpoint
30540 Normal, explicitly set breakpoint.
30541
30542 @item watchpoint
30543 Normal, explicitly set watchpoint.
30544
30545 @item longjmp
30546 Internal breakpoint, used to handle correctly stepping through
30547 @code{longjmp} calls.
30548
30549 @item longjmp resume
30550 Internal breakpoint at the target of a @code{longjmp}.
30551
30552 @item until
30553 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
30554
30555 @item finish
30556 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
30557
30558 @item shlib events
30559 Shared library events.
30560
30561 @end table
30562
30563 @kindex set displaced-stepping
30564 @kindex show displaced-stepping
30565 @cindex displaced stepping support
30566 @cindex out-of-line single-stepping
30567 @item set displaced-stepping
30568 @itemx show displaced-stepping
30569 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
30570 if the target supports it. Displaced stepping is a way to single-step
30571 over breakpoints without removing them from the inferior, by executing
30572 an out-of-line copy of the instruction that was originally at the
30573 breakpoint location. It is also known as out-of-line single-stepping.
30574
30575 @table @code
30576 @item set displaced-stepping on
30577 If the target architecture supports it, @value{GDBN} will use
30578 displaced stepping to step over breakpoints.
30579
30580 @item set displaced-stepping off
30581 @value{GDBN} will not use displaced stepping to step over breakpoints,
30582 even if such is supported by the target architecture.
30583
30584 @cindex non-stop mode, and @samp{set displaced-stepping}
30585 @item set displaced-stepping auto
30586 This is the default mode. @value{GDBN} will use displaced stepping
30587 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
30588 architecture supports displaced stepping.
30589 @end table
30590
30591 @kindex maint check-symtabs
30592 @item maint check-symtabs
30593 Check the consistency of psymtabs and symtabs.
30594
30595 @kindex maint cplus first_component
30596 @item maint cplus first_component @var{name}
30597 Print the first C@t{++} class/namespace component of @var{name}.
30598
30599 @kindex maint cplus namespace
30600 @item maint cplus namespace
30601 Print the list of possible C@t{++} namespaces.
30602
30603 @kindex maint demangle
30604 @item maint demangle @var{name}
30605 Demangle a C@t{++} or Objective-C mangled @var{name}.
30606
30607 @kindex maint deprecate
30608 @kindex maint undeprecate
30609 @cindex deprecated commands
30610 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
30611 @itemx maint undeprecate @var{command}
30612 Deprecate or undeprecate the named @var{command}. Deprecated commands
30613 cause @value{GDBN} to issue a warning when you use them. The optional
30614 argument @var{replacement} says which newer command should be used in
30615 favor of the deprecated one; if it is given, @value{GDBN} will mention
30616 the replacement as part of the warning.
30617
30618 @kindex maint dump-me
30619 @item maint dump-me
30620 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
30621 Cause a fatal signal in the debugger and force it to dump its core.
30622 This is supported only on systems which support aborting a program
30623 with the @code{SIGQUIT} signal.
30624
30625 @kindex maint internal-error
30626 @kindex maint internal-warning
30627 @item maint internal-error @r{[}@var{message-text}@r{]}
30628 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
30629 Cause @value{GDBN} to call the internal function @code{internal_error}
30630 or @code{internal_warning} and hence behave as though an internal error
30631 or internal warning has been detected. In addition to reporting the
30632 internal problem, these functions give the user the opportunity to
30633 either quit @value{GDBN} or create a core file of the current
30634 @value{GDBN} session.
30635
30636 These commands take an optional parameter @var{message-text} that is
30637 used as the text of the error or warning message.
30638
30639 Here's an example of using @code{internal-error}:
30640
30641 @smallexample
30642 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
30643 @dots{}/maint.c:121: internal-error: testing, 1, 2
30644 A problem internal to GDB has been detected. Further
30645 debugging may prove unreliable.
30646 Quit this debugging session? (y or n) @kbd{n}
30647 Create a core file? (y or n) @kbd{n}
30648 (@value{GDBP})
30649 @end smallexample
30650
30651 @cindex @value{GDBN} internal error
30652 @cindex internal errors, control of @value{GDBN} behavior
30653
30654 @kindex maint set internal-error
30655 @kindex maint show internal-error
30656 @kindex maint set internal-warning
30657 @kindex maint show internal-warning
30658 @item maint set internal-error @var{action} [ask|yes|no]
30659 @itemx maint show internal-error @var{action}
30660 @itemx maint set internal-warning @var{action} [ask|yes|no]
30661 @itemx maint show internal-warning @var{action}
30662 When @value{GDBN} reports an internal problem (error or warning) it
30663 gives the user the opportunity to both quit @value{GDBN} and create a
30664 core file of the current @value{GDBN} session. These commands let you
30665 override the default behaviour for each particular @var{action},
30666 described in the table below.
30667
30668 @table @samp
30669 @item quit
30670 You can specify that @value{GDBN} should always (yes) or never (no)
30671 quit. The default is to ask the user what to do.
30672
30673 @item corefile
30674 You can specify that @value{GDBN} should always (yes) or never (no)
30675 create a core file. The default is to ask the user what to do.
30676 @end table
30677
30678 @kindex maint packet
30679 @item maint packet @var{text}
30680 If @value{GDBN} is talking to an inferior via the serial protocol,
30681 then this command sends the string @var{text} to the inferior, and
30682 displays the response packet. @value{GDBN} supplies the initial
30683 @samp{$} character, the terminating @samp{#} character, and the
30684 checksum.
30685
30686 @kindex maint print architecture
30687 @item maint print architecture @r{[}@var{file}@r{]}
30688 Print the entire architecture configuration. The optional argument
30689 @var{file} names the file where the output goes.
30690
30691 @kindex maint print c-tdesc
30692 @item maint print c-tdesc
30693 Print the current target description (@pxref{Target Descriptions}) as
30694 a C source file. The created source file can be used in @value{GDBN}
30695 when an XML parser is not available to parse the description.
30696
30697 @kindex maint print dummy-frames
30698 @item maint print dummy-frames
30699 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
30700
30701 @smallexample
30702 (@value{GDBP}) @kbd{b add}
30703 @dots{}
30704 (@value{GDBP}) @kbd{print add(2,3)}
30705 Breakpoint 2, add (a=2, b=3) at @dots{}
30706 58 return (a + b);
30707 The program being debugged stopped while in a function called from GDB.
30708 @dots{}
30709 (@value{GDBP}) @kbd{maint print dummy-frames}
30710 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
30711 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
30712 call_lo=0x01014000 call_hi=0x01014001
30713 (@value{GDBP})
30714 @end smallexample
30715
30716 Takes an optional file parameter.
30717
30718 @kindex maint print registers
30719 @kindex maint print raw-registers
30720 @kindex maint print cooked-registers
30721 @kindex maint print register-groups
30722 @item maint print registers @r{[}@var{file}@r{]}
30723 @itemx maint print raw-registers @r{[}@var{file}@r{]}
30724 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
30725 @itemx maint print register-groups @r{[}@var{file}@r{]}
30726 Print @value{GDBN}'s internal register data structures.
30727
30728 The command @code{maint print raw-registers} includes the contents of
30729 the raw register cache; the command @code{maint print cooked-registers}
30730 includes the (cooked) value of all registers, including registers which
30731 aren't available on the target nor visible to user; and the
30732 command @code{maint print register-groups} includes the groups that each
30733 register is a member of. @xref{Registers,, Registers, gdbint,
30734 @value{GDBN} Internals}.
30735
30736 These commands take an optional parameter, a file name to which to
30737 write the information.
30738
30739 @kindex maint print reggroups
30740 @item maint print reggroups @r{[}@var{file}@r{]}
30741 Print @value{GDBN}'s internal register group data structures. The
30742 optional argument @var{file} tells to what file to write the
30743 information.
30744
30745 The register groups info looks like this:
30746
30747 @smallexample
30748 (@value{GDBP}) @kbd{maint print reggroups}
30749 Group Type
30750 general user
30751 float user
30752 all user
30753 vector user
30754 system user
30755 save internal
30756 restore internal
30757 @end smallexample
30758
30759 @kindex flushregs
30760 @item flushregs
30761 This command forces @value{GDBN} to flush its internal register cache.
30762
30763 @kindex maint print objfiles
30764 @cindex info for known object files
30765 @item maint print objfiles
30766 Print a dump of all known object files. For each object file, this
30767 command prints its name, address in memory, and all of its psymtabs
30768 and symtabs.
30769
30770 @kindex maint print section-scripts
30771 @cindex info for known .debug_gdb_scripts-loaded scripts
30772 @item maint print section-scripts [@var{regexp}]
30773 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
30774 If @var{regexp} is specified, only print scripts loaded by object files
30775 matching @var{regexp}.
30776 For each script, this command prints its name as specified in the objfile,
30777 and the full path if known.
30778 @xref{.debug_gdb_scripts section}.
30779
30780 @kindex maint print statistics
30781 @cindex bcache statistics
30782 @item maint print statistics
30783 This command prints, for each object file in the program, various data
30784 about that object file followed by the byte cache (@dfn{bcache})
30785 statistics for the object file. The objfile data includes the number
30786 of minimal, partial, full, and stabs symbols, the number of types
30787 defined by the objfile, the number of as yet unexpanded psym tables,
30788 the number of line tables and string tables, and the amount of memory
30789 used by the various tables. The bcache statistics include the counts,
30790 sizes, and counts of duplicates of all and unique objects, max,
30791 average, and median entry size, total memory used and its overhead and
30792 savings, and various measures of the hash table size and chain
30793 lengths.
30794
30795 @kindex maint print target-stack
30796 @cindex target stack description
30797 @item maint print target-stack
30798 A @dfn{target} is an interface between the debugger and a particular
30799 kind of file or process. Targets can be stacked in @dfn{strata},
30800 so that more than one target can potentially respond to a request.
30801 In particular, memory accesses will walk down the stack of targets
30802 until they find a target that is interested in handling that particular
30803 address.
30804
30805 This command prints a short description of each layer that was pushed on
30806 the @dfn{target stack}, starting from the top layer down to the bottom one.
30807
30808 @kindex maint print type
30809 @cindex type chain of a data type
30810 @item maint print type @var{expr}
30811 Print the type chain for a type specified by @var{expr}. The argument
30812 can be either a type name or a symbol. If it is a symbol, the type of
30813 that symbol is described. The type chain produced by this command is
30814 a recursive definition of the data type as stored in @value{GDBN}'s
30815 data structures, including its flags and contained types.
30816
30817 @kindex maint set dwarf2 always-disassemble
30818 @kindex maint show dwarf2 always-disassemble
30819 @item maint set dwarf2 always-disassemble
30820 @item maint show dwarf2 always-disassemble
30821 Control the behavior of @code{info address} when using DWARF debugging
30822 information.
30823
30824 The default is @code{off}, which means that @value{GDBN} should try to
30825 describe a variable's location in an easily readable format. When
30826 @code{on}, @value{GDBN} will instead display the DWARF location
30827 expression in an assembly-like format. Note that some locations are
30828 too complex for @value{GDBN} to describe simply; in this case you will
30829 always see the disassembly form.
30830
30831 Here is an example of the resulting disassembly:
30832
30833 @smallexample
30834 (gdb) info addr argc
30835 Symbol "argc" is a complex DWARF expression:
30836 1: DW_OP_fbreg 0
30837 @end smallexample
30838
30839 For more information on these expressions, see
30840 @uref{http://www.dwarfstd.org/, the DWARF standard}.
30841
30842 @kindex maint set dwarf2 max-cache-age
30843 @kindex maint show dwarf2 max-cache-age
30844 @item maint set dwarf2 max-cache-age
30845 @itemx maint show dwarf2 max-cache-age
30846 Control the DWARF 2 compilation unit cache.
30847
30848 @cindex DWARF 2 compilation units cache
30849 In object files with inter-compilation-unit references, such as those
30850 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
30851 reader needs to frequently refer to previously read compilation units.
30852 This setting controls how long a compilation unit will remain in the
30853 cache if it is not referenced. A higher limit means that cached
30854 compilation units will be stored in memory longer, and more total
30855 memory will be used. Setting it to zero disables caching, which will
30856 slow down @value{GDBN} startup, but reduce memory consumption.
30857
30858 @kindex maint set profile
30859 @kindex maint show profile
30860 @cindex profiling GDB
30861 @item maint set profile
30862 @itemx maint show profile
30863 Control profiling of @value{GDBN}.
30864
30865 Profiling will be disabled until you use the @samp{maint set profile}
30866 command to enable it. When you enable profiling, the system will begin
30867 collecting timing and execution count data; when you disable profiling or
30868 exit @value{GDBN}, the results will be written to a log file. Remember that
30869 if you use profiling, @value{GDBN} will overwrite the profiling log file
30870 (often called @file{gmon.out}). If you have a record of important profiling
30871 data in a @file{gmon.out} file, be sure to move it to a safe location.
30872
30873 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
30874 compiled with the @samp{-pg} compiler option.
30875
30876 @kindex maint set show-debug-regs
30877 @kindex maint show show-debug-regs
30878 @cindex hardware debug registers
30879 @item maint set show-debug-regs
30880 @itemx maint show show-debug-regs
30881 Control whether to show variables that mirror the hardware debug
30882 registers. Use @code{ON} to enable, @code{OFF} to disable. If
30883 enabled, the debug registers values are shown when @value{GDBN} inserts or
30884 removes a hardware breakpoint or watchpoint, and when the inferior
30885 triggers a hardware-assisted breakpoint or watchpoint.
30886
30887 @kindex maint set show-all-tib
30888 @kindex maint show show-all-tib
30889 @item maint set show-all-tib
30890 @itemx maint show show-all-tib
30891 Control whether to show all non zero areas within a 1k block starting
30892 at thread local base, when using the @samp{info w32 thread-information-block}
30893 command.
30894
30895 @kindex maint space
30896 @cindex memory used by commands
30897 @item maint space
30898 Control whether to display memory usage for each command. If set to a
30899 nonzero value, @value{GDBN} will display how much memory each command
30900 took, following the command's own output. This can also be requested
30901 by invoking @value{GDBN} with the @option{--statistics} command-line
30902 switch (@pxref{Mode Options}).
30903
30904 @kindex maint time
30905 @cindex time of command execution
30906 @item maint time
30907 Control whether to display the execution time for each command. If
30908 set to a nonzero value, @value{GDBN} will display how much time it
30909 took to execute each command, following the command's own output.
30910 The time is not printed for the commands that run the target, since
30911 there's no mechanism currently to compute how much time was spend
30912 by @value{GDBN} and how much time was spend by the program been debugged.
30913 it's not possibly currently
30914 This can also be requested by invoking @value{GDBN} with the
30915 @option{--statistics} command-line switch (@pxref{Mode Options}).
30916
30917 @kindex maint translate-address
30918 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
30919 Find the symbol stored at the location specified by the address
30920 @var{addr} and an optional section name @var{section}. If found,
30921 @value{GDBN} prints the name of the closest symbol and an offset from
30922 the symbol's location to the specified address. This is similar to
30923 the @code{info address} command (@pxref{Symbols}), except that this
30924 command also allows to find symbols in other sections.
30925
30926 If section was not specified, the section in which the symbol was found
30927 is also printed. For dynamically linked executables, the name of
30928 executable or shared library containing the symbol is printed as well.
30929
30930 @end table
30931
30932 The following command is useful for non-interactive invocations of
30933 @value{GDBN}, such as in the test suite.
30934
30935 @table @code
30936 @item set watchdog @var{nsec}
30937 @kindex set watchdog
30938 @cindex watchdog timer
30939 @cindex timeout for commands
30940 Set the maximum number of seconds @value{GDBN} will wait for the
30941 target operation to finish. If this time expires, @value{GDBN}
30942 reports and error and the command is aborted.
30943
30944 @item show watchdog
30945 Show the current setting of the target wait timeout.
30946 @end table
30947
30948 @node Remote Protocol
30949 @appendix @value{GDBN} Remote Serial Protocol
30950
30951 @menu
30952 * Overview::
30953 * Packets::
30954 * Stop Reply Packets::
30955 * General Query Packets::
30956 * Architecture-Specific Protocol Details::
30957 * Tracepoint Packets::
30958 * Host I/O Packets::
30959 * Interrupts::
30960 * Notification Packets::
30961 * Remote Non-Stop::
30962 * Packet Acknowledgment::
30963 * Examples::
30964 * File-I/O Remote Protocol Extension::
30965 * Library List Format::
30966 * Memory Map Format::
30967 * Thread List Format::
30968 @end menu
30969
30970 @node Overview
30971 @section Overview
30972
30973 There may be occasions when you need to know something about the
30974 protocol---for example, if there is only one serial port to your target
30975 machine, you might want your program to do something special if it
30976 recognizes a packet meant for @value{GDBN}.
30977
30978 In the examples below, @samp{->} and @samp{<-} are used to indicate
30979 transmitted and received data, respectively.
30980
30981 @cindex protocol, @value{GDBN} remote serial
30982 @cindex serial protocol, @value{GDBN} remote
30983 @cindex remote serial protocol
30984 All @value{GDBN} commands and responses (other than acknowledgments
30985 and notifications, see @ref{Notification Packets}) are sent as a
30986 @var{packet}. A @var{packet} is introduced with the character
30987 @samp{$}, the actual @var{packet-data}, and the terminating character
30988 @samp{#} followed by a two-digit @var{checksum}:
30989
30990 @smallexample
30991 @code{$}@var{packet-data}@code{#}@var{checksum}
30992 @end smallexample
30993 @noindent
30994
30995 @cindex checksum, for @value{GDBN} remote
30996 @noindent
30997 The two-digit @var{checksum} is computed as the modulo 256 sum of all
30998 characters between the leading @samp{$} and the trailing @samp{#} (an
30999 eight bit unsigned checksum).
31000
31001 Implementors should note that prior to @value{GDBN} 5.0 the protocol
31002 specification also included an optional two-digit @var{sequence-id}:
31003
31004 @smallexample
31005 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
31006 @end smallexample
31007
31008 @cindex sequence-id, for @value{GDBN} remote
31009 @noindent
31010 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
31011 has never output @var{sequence-id}s. Stubs that handle packets added
31012 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
31013
31014 When either the host or the target machine receives a packet, the first
31015 response expected is an acknowledgment: either @samp{+} (to indicate
31016 the package was received correctly) or @samp{-} (to request
31017 retransmission):
31018
31019 @smallexample
31020 -> @code{$}@var{packet-data}@code{#}@var{checksum}
31021 <- @code{+}
31022 @end smallexample
31023 @noindent
31024
31025 The @samp{+}/@samp{-} acknowledgments can be disabled
31026 once a connection is established.
31027 @xref{Packet Acknowledgment}, for details.
31028
31029 The host (@value{GDBN}) sends @var{command}s, and the target (the
31030 debugging stub incorporated in your program) sends a @var{response}. In
31031 the case of step and continue @var{command}s, the response is only sent
31032 when the operation has completed, and the target has again stopped all
31033 threads in all attached processes. This is the default all-stop mode
31034 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
31035 execution mode; see @ref{Remote Non-Stop}, for details.
31036
31037 @var{packet-data} consists of a sequence of characters with the
31038 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
31039 exceptions).
31040
31041 @cindex remote protocol, field separator
31042 Fields within the packet should be separated using @samp{,} @samp{;} or
31043 @samp{:}. Except where otherwise noted all numbers are represented in
31044 @sc{hex} with leading zeros suppressed.
31045
31046 Implementors should note that prior to @value{GDBN} 5.0, the character
31047 @samp{:} could not appear as the third character in a packet (as it
31048 would potentially conflict with the @var{sequence-id}).
31049
31050 @cindex remote protocol, binary data
31051 @anchor{Binary Data}
31052 Binary data in most packets is encoded either as two hexadecimal
31053 digits per byte of binary data. This allowed the traditional remote
31054 protocol to work over connections which were only seven-bit clean.
31055 Some packets designed more recently assume an eight-bit clean
31056 connection, and use a more efficient encoding to send and receive
31057 binary data.
31058
31059 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
31060 as an escape character. Any escaped byte is transmitted as the escape
31061 character followed by the original character XORed with @code{0x20}.
31062 For example, the byte @code{0x7d} would be transmitted as the two
31063 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
31064 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
31065 @samp{@}}) must always be escaped. Responses sent by the stub
31066 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
31067 is not interpreted as the start of a run-length encoded sequence
31068 (described next).
31069
31070 Response @var{data} can be run-length encoded to save space.
31071 Run-length encoding replaces runs of identical characters with one
31072 instance of the repeated character, followed by a @samp{*} and a
31073 repeat count. The repeat count is itself sent encoded, to avoid
31074 binary characters in @var{data}: a value of @var{n} is sent as
31075 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
31076 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
31077 code 32) for a repeat count of 3. (This is because run-length
31078 encoding starts to win for counts 3 or more.) Thus, for example,
31079 @samp{0* } is a run-length encoding of ``0000'': the space character
31080 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
31081 3}} more times.
31082
31083 The printable characters @samp{#} and @samp{$} or with a numeric value
31084 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
31085 seven repeats (@samp{$}) can be expanded using a repeat count of only
31086 five (@samp{"}). For example, @samp{00000000} can be encoded as
31087 @samp{0*"00}.
31088
31089 The error response returned for some packets includes a two character
31090 error number. That number is not well defined.
31091
31092 @cindex empty response, for unsupported packets
31093 For any @var{command} not supported by the stub, an empty response
31094 (@samp{$#00}) should be returned. That way it is possible to extend the
31095 protocol. A newer @value{GDBN} can tell if a packet is supported based
31096 on that response.
31097
31098 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
31099 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
31100 optional.
31101
31102 @node Packets
31103 @section Packets
31104
31105 The following table provides a complete list of all currently defined
31106 @var{command}s and their corresponding response @var{data}.
31107 @xref{File-I/O Remote Protocol Extension}, for details about the File
31108 I/O extension of the remote protocol.
31109
31110 Each packet's description has a template showing the packet's overall
31111 syntax, followed by an explanation of the packet's meaning. We
31112 include spaces in some of the templates for clarity; these are not
31113 part of the packet's syntax. No @value{GDBN} packet uses spaces to
31114 separate its components. For example, a template like @samp{foo
31115 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
31116 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
31117 @var{baz}. @value{GDBN} does not transmit a space character between the
31118 @samp{foo} and the @var{bar}, or between the @var{bar} and the
31119 @var{baz}.
31120
31121 @cindex @var{thread-id}, in remote protocol
31122 @anchor{thread-id syntax}
31123 Several packets and replies include a @var{thread-id} field to identify
31124 a thread. Normally these are positive numbers with a target-specific
31125 interpretation, formatted as big-endian hex strings. A @var{thread-id}
31126 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
31127 pick any thread.
31128
31129 In addition, the remote protocol supports a multiprocess feature in
31130 which the @var{thread-id} syntax is extended to optionally include both
31131 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
31132 The @var{pid} (process) and @var{tid} (thread) components each have the
31133 format described above: a positive number with target-specific
31134 interpretation formatted as a big-endian hex string, literal @samp{-1}
31135 to indicate all processes or threads (respectively), or @samp{0} to
31136 indicate an arbitrary process or thread. Specifying just a process, as
31137 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
31138 error to specify all processes but a specific thread, such as
31139 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
31140 for those packets and replies explicitly documented to include a process
31141 ID, rather than a @var{thread-id}.
31142
31143 The multiprocess @var{thread-id} syntax extensions are only used if both
31144 @value{GDBN} and the stub report support for the @samp{multiprocess}
31145 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
31146 more information.
31147
31148 Note that all packet forms beginning with an upper- or lower-case
31149 letter, other than those described here, are reserved for future use.
31150
31151 Here are the packet descriptions.
31152
31153 @table @samp
31154
31155 @item !
31156 @cindex @samp{!} packet
31157 @anchor{extended mode}
31158 Enable extended mode. In extended mode, the remote server is made
31159 persistent. The @samp{R} packet is used to restart the program being
31160 debugged.
31161
31162 Reply:
31163 @table @samp
31164 @item OK
31165 The remote target both supports and has enabled extended mode.
31166 @end table
31167
31168 @item ?
31169 @cindex @samp{?} packet
31170 Indicate the reason the target halted. The reply is the same as for
31171 step and continue. This packet has a special interpretation when the
31172 target is in non-stop mode; see @ref{Remote Non-Stop}.
31173
31174 Reply:
31175 @xref{Stop Reply Packets}, for the reply specifications.
31176
31177 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
31178 @cindex @samp{A} packet
31179 Initialized @code{argv[]} array passed into program. @var{arglen}
31180 specifies the number of bytes in the hex encoded byte stream
31181 @var{arg}. See @code{gdbserver} for more details.
31182
31183 Reply:
31184 @table @samp
31185 @item OK
31186 The arguments were set.
31187 @item E @var{NN}
31188 An error occurred.
31189 @end table
31190
31191 @item b @var{baud}
31192 @cindex @samp{b} packet
31193 (Don't use this packet; its behavior is not well-defined.)
31194 Change the serial line speed to @var{baud}.
31195
31196 JTC: @emph{When does the transport layer state change? When it's
31197 received, or after the ACK is transmitted. In either case, there are
31198 problems if the command or the acknowledgment packet is dropped.}
31199
31200 Stan: @emph{If people really wanted to add something like this, and get
31201 it working for the first time, they ought to modify ser-unix.c to send
31202 some kind of out-of-band message to a specially-setup stub and have the
31203 switch happen "in between" packets, so that from remote protocol's point
31204 of view, nothing actually happened.}
31205
31206 @item B @var{addr},@var{mode}
31207 @cindex @samp{B} packet
31208 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
31209 breakpoint at @var{addr}.
31210
31211 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
31212 (@pxref{insert breakpoint or watchpoint packet}).
31213
31214 @cindex @samp{bc} packet
31215 @anchor{bc}
31216 @item bc
31217 Backward continue. Execute the target system in reverse. No parameter.
31218 @xref{Reverse Execution}, for more information.
31219
31220 Reply:
31221 @xref{Stop Reply Packets}, for the reply specifications.
31222
31223 @cindex @samp{bs} packet
31224 @anchor{bs}
31225 @item bs
31226 Backward single step. Execute one instruction in reverse. No parameter.
31227 @xref{Reverse Execution}, for more information.
31228
31229 Reply:
31230 @xref{Stop Reply Packets}, for the reply specifications.
31231
31232 @item c @r{[}@var{addr}@r{]}
31233 @cindex @samp{c} packet
31234 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
31235 resume at current address.
31236
31237 Reply:
31238 @xref{Stop Reply Packets}, for the reply specifications.
31239
31240 @item C @var{sig}@r{[};@var{addr}@r{]}
31241 @cindex @samp{C} packet
31242 Continue with signal @var{sig} (hex signal number). If
31243 @samp{;@var{addr}} is omitted, resume at same address.
31244
31245 Reply:
31246 @xref{Stop Reply Packets}, for the reply specifications.
31247
31248 @item d
31249 @cindex @samp{d} packet
31250 Toggle debug flag.
31251
31252 Don't use this packet; instead, define a general set packet
31253 (@pxref{General Query Packets}).
31254
31255 @item D
31256 @itemx D;@var{pid}
31257 @cindex @samp{D} packet
31258 The first form of the packet is used to detach @value{GDBN} from the
31259 remote system. It is sent to the remote target
31260 before @value{GDBN} disconnects via the @code{detach} command.
31261
31262 The second form, including a process ID, is used when multiprocess
31263 protocol extensions are enabled (@pxref{multiprocess extensions}), to
31264 detach only a specific process. The @var{pid} is specified as a
31265 big-endian hex string.
31266
31267 Reply:
31268 @table @samp
31269 @item OK
31270 for success
31271 @item E @var{NN}
31272 for an error
31273 @end table
31274
31275 @item F @var{RC},@var{EE},@var{CF};@var{XX}
31276 @cindex @samp{F} packet
31277 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
31278 This is part of the File-I/O protocol extension. @xref{File-I/O
31279 Remote Protocol Extension}, for the specification.
31280
31281 @item g
31282 @anchor{read registers packet}
31283 @cindex @samp{g} packet
31284 Read general registers.
31285
31286 Reply:
31287 @table @samp
31288 @item @var{XX@dots{}}
31289 Each byte of register data is described by two hex digits. The bytes
31290 with the register are transmitted in target byte order. The size of
31291 each register and their position within the @samp{g} packet are
31292 determined by the @value{GDBN} internal gdbarch functions
31293 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
31294 specification of several standard @samp{g} packets is specified below.
31295 @item E @var{NN}
31296 for an error.
31297 @end table
31298
31299 @item G @var{XX@dots{}}
31300 @cindex @samp{G} packet
31301 Write general registers. @xref{read registers packet}, for a
31302 description of the @var{XX@dots{}} data.
31303
31304 Reply:
31305 @table @samp
31306 @item OK
31307 for success
31308 @item E @var{NN}
31309 for an error
31310 @end table
31311
31312 @item H @var{c} @var{thread-id}
31313 @cindex @samp{H} packet
31314 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
31315 @samp{G}, et.al.). @var{c} depends on the operation to be performed: it
31316 should be @samp{c} for step and continue operations, @samp{g} for other
31317 operations. The thread designator @var{thread-id} has the format and
31318 interpretation described in @ref{thread-id syntax}.
31319
31320 Reply:
31321 @table @samp
31322 @item OK
31323 for success
31324 @item E @var{NN}
31325 for an error
31326 @end table
31327
31328 @c FIXME: JTC:
31329 @c 'H': How restrictive (or permissive) is the thread model. If a
31330 @c thread is selected and stopped, are other threads allowed
31331 @c to continue to execute? As I mentioned above, I think the
31332 @c semantics of each command when a thread is selected must be
31333 @c described. For example:
31334 @c
31335 @c 'g': If the stub supports threads and a specific thread is
31336 @c selected, returns the register block from that thread;
31337 @c otherwise returns current registers.
31338 @c
31339 @c 'G' If the stub supports threads and a specific thread is
31340 @c selected, sets the registers of the register block of
31341 @c that thread; otherwise sets current registers.
31342
31343 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
31344 @anchor{cycle step packet}
31345 @cindex @samp{i} packet
31346 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
31347 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
31348 step starting at that address.
31349
31350 @item I
31351 @cindex @samp{I} packet
31352 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
31353 step packet}.
31354
31355 @item k
31356 @cindex @samp{k} packet
31357 Kill request.
31358
31359 FIXME: @emph{There is no description of how to operate when a specific
31360 thread context has been selected (i.e.@: does 'k' kill only that
31361 thread?)}.
31362
31363 @item m @var{addr},@var{length}
31364 @cindex @samp{m} packet
31365 Read @var{length} bytes of memory starting at address @var{addr}.
31366 Note that @var{addr} may not be aligned to any particular boundary.
31367
31368 The stub need not use any particular size or alignment when gathering
31369 data from memory for the response; even if @var{addr} is word-aligned
31370 and @var{length} is a multiple of the word size, the stub is free to
31371 use byte accesses, or not. For this reason, this packet may not be
31372 suitable for accessing memory-mapped I/O devices.
31373 @cindex alignment of remote memory accesses
31374 @cindex size of remote memory accesses
31375 @cindex memory, alignment and size of remote accesses
31376
31377 Reply:
31378 @table @samp
31379 @item @var{XX@dots{}}
31380 Memory contents; each byte is transmitted as a two-digit hexadecimal
31381 number. The reply may contain fewer bytes than requested if the
31382 server was able to read only part of the region of memory.
31383 @item E @var{NN}
31384 @var{NN} is errno
31385 @end table
31386
31387 @item M @var{addr},@var{length}:@var{XX@dots{}}
31388 @cindex @samp{M} packet
31389 Write @var{length} bytes of memory starting at address @var{addr}.
31390 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
31391 hexadecimal number.
31392
31393 Reply:
31394 @table @samp
31395 @item OK
31396 for success
31397 @item E @var{NN}
31398 for an error (this includes the case where only part of the data was
31399 written).
31400 @end table
31401
31402 @item p @var{n}
31403 @cindex @samp{p} packet
31404 Read the value of register @var{n}; @var{n} is in hex.
31405 @xref{read registers packet}, for a description of how the returned
31406 register value is encoded.
31407
31408 Reply:
31409 @table @samp
31410 @item @var{XX@dots{}}
31411 the register's value
31412 @item E @var{NN}
31413 for an error
31414 @item
31415 Indicating an unrecognized @var{query}.
31416 @end table
31417
31418 @item P @var{n@dots{}}=@var{r@dots{}}
31419 @anchor{write register packet}
31420 @cindex @samp{P} packet
31421 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
31422 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
31423 digits for each byte in the register (target byte order).
31424
31425 Reply:
31426 @table @samp
31427 @item OK
31428 for success
31429 @item E @var{NN}
31430 for an error
31431 @end table
31432
31433 @item q @var{name} @var{params}@dots{}
31434 @itemx Q @var{name} @var{params}@dots{}
31435 @cindex @samp{q} packet
31436 @cindex @samp{Q} packet
31437 General query (@samp{q}) and set (@samp{Q}). These packets are
31438 described fully in @ref{General Query Packets}.
31439
31440 @item r
31441 @cindex @samp{r} packet
31442 Reset the entire system.
31443
31444 Don't use this packet; use the @samp{R} packet instead.
31445
31446 @item R @var{XX}
31447 @cindex @samp{R} packet
31448 Restart the program being debugged. @var{XX}, while needed, is ignored.
31449 This packet is only available in extended mode (@pxref{extended mode}).
31450
31451 The @samp{R} packet has no reply.
31452
31453 @item s @r{[}@var{addr}@r{]}
31454 @cindex @samp{s} packet
31455 Single step. @var{addr} is the address at which to resume. If
31456 @var{addr} is omitted, resume at same address.
31457
31458 Reply:
31459 @xref{Stop Reply Packets}, for the reply specifications.
31460
31461 @item S @var{sig}@r{[};@var{addr}@r{]}
31462 @anchor{step with signal packet}
31463 @cindex @samp{S} packet
31464 Step with signal. This is analogous to the @samp{C} packet, but
31465 requests a single-step, rather than a normal resumption of execution.
31466
31467 Reply:
31468 @xref{Stop Reply Packets}, for the reply specifications.
31469
31470 @item t @var{addr}:@var{PP},@var{MM}
31471 @cindex @samp{t} packet
31472 Search backwards starting at address @var{addr} for a match with pattern
31473 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
31474 @var{addr} must be at least 3 digits.
31475
31476 @item T @var{thread-id}
31477 @cindex @samp{T} packet
31478 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
31479
31480 Reply:
31481 @table @samp
31482 @item OK
31483 thread is still alive
31484 @item E @var{NN}
31485 thread is dead
31486 @end table
31487
31488 @item v
31489 Packets starting with @samp{v} are identified by a multi-letter name,
31490 up to the first @samp{;} or @samp{?} (or the end of the packet).
31491
31492 @item vAttach;@var{pid}
31493 @cindex @samp{vAttach} packet
31494 Attach to a new process with the specified process ID @var{pid}.
31495 The process ID is a
31496 hexadecimal integer identifying the process. In all-stop mode, all
31497 threads in the attached process are stopped; in non-stop mode, it may be
31498 attached without being stopped if that is supported by the target.
31499
31500 @c In non-stop mode, on a successful vAttach, the stub should set the
31501 @c current thread to a thread of the newly-attached process. After
31502 @c attaching, GDB queries for the attached process's thread ID with qC.
31503 @c Also note that, from a user perspective, whether or not the
31504 @c target is stopped on attach in non-stop mode depends on whether you
31505 @c use the foreground or background version of the attach command, not
31506 @c on what vAttach does; GDB does the right thing with respect to either
31507 @c stopping or restarting threads.
31508
31509 This packet is only available in extended mode (@pxref{extended mode}).
31510
31511 Reply:
31512 @table @samp
31513 @item E @var{nn}
31514 for an error
31515 @item @r{Any stop packet}
31516 for success in all-stop mode (@pxref{Stop Reply Packets})
31517 @item OK
31518 for success in non-stop mode (@pxref{Remote Non-Stop})
31519 @end table
31520
31521 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
31522 @cindex @samp{vCont} packet
31523 Resume the inferior, specifying different actions for each thread.
31524 If an action is specified with no @var{thread-id}, then it is applied to any
31525 threads that don't have a specific action specified; if no default action is
31526 specified then other threads should remain stopped in all-stop mode and
31527 in their current state in non-stop mode.
31528 Specifying multiple
31529 default actions is an error; specifying no actions is also an error.
31530 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
31531
31532 Currently supported actions are:
31533
31534 @table @samp
31535 @item c
31536 Continue.
31537 @item C @var{sig}
31538 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
31539 @item s
31540 Step.
31541 @item S @var{sig}
31542 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
31543 @item t
31544 Stop.
31545 @end table
31546
31547 The optional argument @var{addr} normally associated with the
31548 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
31549 not supported in @samp{vCont}.
31550
31551 The @samp{t} action is only relevant in non-stop mode
31552 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
31553 A stop reply should be generated for any affected thread not already stopped.
31554 When a thread is stopped by means of a @samp{t} action,
31555 the corresponding stop reply should indicate that the thread has stopped with
31556 signal @samp{0}, regardless of whether the target uses some other signal
31557 as an implementation detail.
31558
31559 Reply:
31560 @xref{Stop Reply Packets}, for the reply specifications.
31561
31562 @item vCont?
31563 @cindex @samp{vCont?} packet
31564 Request a list of actions supported by the @samp{vCont} packet.
31565
31566 Reply:
31567 @table @samp
31568 @item vCont@r{[};@var{action}@dots{}@r{]}
31569 The @samp{vCont} packet is supported. Each @var{action} is a supported
31570 command in the @samp{vCont} packet.
31571 @item
31572 The @samp{vCont} packet is not supported.
31573 @end table
31574
31575 @item vFile:@var{operation}:@var{parameter}@dots{}
31576 @cindex @samp{vFile} packet
31577 Perform a file operation on the target system. For details,
31578 see @ref{Host I/O Packets}.
31579
31580 @item vFlashErase:@var{addr},@var{length}
31581 @cindex @samp{vFlashErase} packet
31582 Direct the stub to erase @var{length} bytes of flash starting at
31583 @var{addr}. The region may enclose any number of flash blocks, but
31584 its start and end must fall on block boundaries, as indicated by the
31585 flash block size appearing in the memory map (@pxref{Memory Map
31586 Format}). @value{GDBN} groups flash memory programming operations
31587 together, and sends a @samp{vFlashDone} request after each group; the
31588 stub is allowed to delay erase operation until the @samp{vFlashDone}
31589 packet is received.
31590
31591 The stub must support @samp{vCont} if it reports support for
31592 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
31593 this case @samp{vCont} actions can be specified to apply to all threads
31594 in a process by using the @samp{p@var{pid}.-1} form of the
31595 @var{thread-id}.
31596
31597 Reply:
31598 @table @samp
31599 @item OK
31600 for success
31601 @item E @var{NN}
31602 for an error
31603 @end table
31604
31605 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
31606 @cindex @samp{vFlashWrite} packet
31607 Direct the stub to write data to flash address @var{addr}. The data
31608 is passed in binary form using the same encoding as for the @samp{X}
31609 packet (@pxref{Binary Data}). The memory ranges specified by
31610 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
31611 not overlap, and must appear in order of increasing addresses
31612 (although @samp{vFlashErase} packets for higher addresses may already
31613 have been received; the ordering is guaranteed only between
31614 @samp{vFlashWrite} packets). If a packet writes to an address that was
31615 neither erased by a preceding @samp{vFlashErase} packet nor by some other
31616 target-specific method, the results are unpredictable.
31617
31618
31619 Reply:
31620 @table @samp
31621 @item OK
31622 for success
31623 @item E.memtype
31624 for vFlashWrite addressing non-flash memory
31625 @item E @var{NN}
31626 for an error
31627 @end table
31628
31629 @item vFlashDone
31630 @cindex @samp{vFlashDone} packet
31631 Indicate to the stub that flash programming operation is finished.
31632 The stub is permitted to delay or batch the effects of a group of
31633 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
31634 @samp{vFlashDone} packet is received. The contents of the affected
31635 regions of flash memory are unpredictable until the @samp{vFlashDone}
31636 request is completed.
31637
31638 @item vKill;@var{pid}
31639 @cindex @samp{vKill} packet
31640 Kill the process with the specified process ID. @var{pid} is a
31641 hexadecimal integer identifying the process. This packet is used in
31642 preference to @samp{k} when multiprocess protocol extensions are
31643 supported; see @ref{multiprocess extensions}.
31644
31645 Reply:
31646 @table @samp
31647 @item E @var{nn}
31648 for an error
31649 @item OK
31650 for success
31651 @end table
31652
31653 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
31654 @cindex @samp{vRun} packet
31655 Run the program @var{filename}, passing it each @var{argument} on its
31656 command line. The file and arguments are hex-encoded strings. If
31657 @var{filename} is an empty string, the stub may use a default program
31658 (e.g.@: the last program run). The program is created in the stopped
31659 state.
31660
31661 @c FIXME: What about non-stop mode?
31662
31663 This packet is only available in extended mode (@pxref{extended mode}).
31664
31665 Reply:
31666 @table @samp
31667 @item E @var{nn}
31668 for an error
31669 @item @r{Any stop packet}
31670 for success (@pxref{Stop Reply Packets})
31671 @end table
31672
31673 @item vStopped
31674 @anchor{vStopped packet}
31675 @cindex @samp{vStopped} packet
31676
31677 In non-stop mode (@pxref{Remote Non-Stop}), acknowledge a previous stop
31678 reply and prompt for the stub to report another one.
31679
31680 Reply:
31681 @table @samp
31682 @item @r{Any stop packet}
31683 if there is another unreported stop event (@pxref{Stop Reply Packets})
31684 @item OK
31685 if there are no unreported stop events
31686 @end table
31687
31688 @item X @var{addr},@var{length}:@var{XX@dots{}}
31689 @anchor{X packet}
31690 @cindex @samp{X} packet
31691 Write data to memory, where the data is transmitted in binary.
31692 @var{addr} is address, @var{length} is number of bytes,
31693 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
31694
31695 Reply:
31696 @table @samp
31697 @item OK
31698 for success
31699 @item E @var{NN}
31700 for an error
31701 @end table
31702
31703 @item z @var{type},@var{addr},@var{kind}
31704 @itemx Z @var{type},@var{addr},@var{kind}
31705 @anchor{insert breakpoint or watchpoint packet}
31706 @cindex @samp{z} packet
31707 @cindex @samp{Z} packets
31708 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
31709 watchpoint starting at address @var{address} of kind @var{kind}.
31710
31711 Each breakpoint and watchpoint packet @var{type} is documented
31712 separately.
31713
31714 @emph{Implementation notes: A remote target shall return an empty string
31715 for an unrecognized breakpoint or watchpoint packet @var{type}. A
31716 remote target shall support either both or neither of a given
31717 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
31718 avoid potential problems with duplicate packets, the operations should
31719 be implemented in an idempotent way.}
31720
31721 @item z0,@var{addr},@var{kind}
31722 @itemx Z0,@var{addr},@var{kind}
31723 @cindex @samp{z0} packet
31724 @cindex @samp{Z0} packet
31725 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
31726 @var{addr} of type @var{kind}.
31727
31728 A memory breakpoint is implemented by replacing the instruction at
31729 @var{addr} with a software breakpoint or trap instruction. The
31730 @var{kind} is target-specific and typically indicates the size of
31731 the breakpoint in bytes that should be inserted. E.g., the @sc{arm}
31732 and @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
31733 architectures have additional meanings for @var{kind};
31734 see @ref{Architecture-Specific Protocol Details}.
31735
31736 @emph{Implementation note: It is possible for a target to copy or move
31737 code that contains memory breakpoints (e.g., when implementing
31738 overlays). The behavior of this packet, in the presence of such a
31739 target, is not defined.}
31740
31741 Reply:
31742 @table @samp
31743 @item OK
31744 success
31745 @item
31746 not supported
31747 @item E @var{NN}
31748 for an error
31749 @end table
31750
31751 @item z1,@var{addr},@var{kind}
31752 @itemx Z1,@var{addr},@var{kind}
31753 @cindex @samp{z1} packet
31754 @cindex @samp{Z1} packet
31755 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
31756 address @var{addr}.
31757
31758 A hardware breakpoint is implemented using a mechanism that is not
31759 dependant on being able to modify the target's memory. @var{kind}
31760 has the same meaning as in @samp{Z0} packets.
31761
31762 @emph{Implementation note: A hardware breakpoint is not affected by code
31763 movement.}
31764
31765 Reply:
31766 @table @samp
31767 @item OK
31768 success
31769 @item
31770 not supported
31771 @item E @var{NN}
31772 for an error
31773 @end table
31774
31775 @item z2,@var{addr},@var{kind}
31776 @itemx Z2,@var{addr},@var{kind}
31777 @cindex @samp{z2} packet
31778 @cindex @samp{Z2} packet
31779 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
31780 @var{kind} is interpreted as the number of bytes to watch.
31781
31782 Reply:
31783 @table @samp
31784 @item OK
31785 success
31786 @item
31787 not supported
31788 @item E @var{NN}
31789 for an error
31790 @end table
31791
31792 @item z3,@var{addr},@var{kind}
31793 @itemx Z3,@var{addr},@var{kind}
31794 @cindex @samp{z3} packet
31795 @cindex @samp{Z3} packet
31796 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
31797 @var{kind} is interpreted as the number of bytes to watch.
31798
31799 Reply:
31800 @table @samp
31801 @item OK
31802 success
31803 @item
31804 not supported
31805 @item E @var{NN}
31806 for an error
31807 @end table
31808
31809 @item z4,@var{addr},@var{kind}
31810 @itemx Z4,@var{addr},@var{kind}
31811 @cindex @samp{z4} packet
31812 @cindex @samp{Z4} packet
31813 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
31814 @var{kind} is interpreted as the number of bytes to watch.
31815
31816 Reply:
31817 @table @samp
31818 @item OK
31819 success
31820 @item
31821 not supported
31822 @item E @var{NN}
31823 for an error
31824 @end table
31825
31826 @end table
31827
31828 @node Stop Reply Packets
31829 @section Stop Reply Packets
31830 @cindex stop reply packets
31831
31832 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
31833 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
31834 receive any of the below as a reply. Except for @samp{?}
31835 and @samp{vStopped}, that reply is only returned
31836 when the target halts. In the below the exact meaning of @dfn{signal
31837 number} is defined by the header @file{include/gdb/signals.h} in the
31838 @value{GDBN} source code.
31839
31840 As in the description of request packets, we include spaces in the
31841 reply templates for clarity; these are not part of the reply packet's
31842 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
31843 components.
31844
31845 @table @samp
31846
31847 @item S @var{AA}
31848 The program received signal number @var{AA} (a two-digit hexadecimal
31849 number). This is equivalent to a @samp{T} response with no
31850 @var{n}:@var{r} pairs.
31851
31852 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
31853 @cindex @samp{T} packet reply
31854 The program received signal number @var{AA} (a two-digit hexadecimal
31855 number). This is equivalent to an @samp{S} response, except that the
31856 @samp{@var{n}:@var{r}} pairs can carry values of important registers
31857 and other information directly in the stop reply packet, reducing
31858 round-trip latency. Single-step and breakpoint traps are reported
31859 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
31860
31861 @itemize @bullet
31862 @item
31863 If @var{n} is a hexadecimal number, it is a register number, and the
31864 corresponding @var{r} gives that register's value. @var{r} is a
31865 series of bytes in target byte order, with each byte given by a
31866 two-digit hex number.
31867
31868 @item
31869 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
31870 the stopped thread, as specified in @ref{thread-id syntax}.
31871
31872 @item
31873 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
31874 the core on which the stop event was detected.
31875
31876 @item
31877 If @var{n} is a recognized @dfn{stop reason}, it describes a more
31878 specific event that stopped the target. The currently defined stop
31879 reasons are listed below. @var{aa} should be @samp{05}, the trap
31880 signal. At most one stop reason should be present.
31881
31882 @item
31883 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
31884 and go on to the next; this allows us to extend the protocol in the
31885 future.
31886 @end itemize
31887
31888 The currently defined stop reasons are:
31889
31890 @table @samp
31891 @item watch
31892 @itemx rwatch
31893 @itemx awatch
31894 The packet indicates a watchpoint hit, and @var{r} is the data address, in
31895 hex.
31896
31897 @cindex shared library events, remote reply
31898 @item library
31899 The packet indicates that the loaded libraries have changed.
31900 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
31901 list of loaded libraries. @var{r} is ignored.
31902
31903 @cindex replay log events, remote reply
31904 @item replaylog
31905 The packet indicates that the target cannot continue replaying
31906 logged execution events, because it has reached the end (or the
31907 beginning when executing backward) of the log. The value of @var{r}
31908 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
31909 for more information.
31910 @end table
31911
31912 @item W @var{AA}
31913 @itemx W @var{AA} ; process:@var{pid}
31914 The process exited, and @var{AA} is the exit status. This is only
31915 applicable to certain targets.
31916
31917 The second form of the response, including the process ID of the exited
31918 process, can be used only when @value{GDBN} has reported support for
31919 multiprocess protocol extensions; see @ref{multiprocess extensions}.
31920 The @var{pid} is formatted as a big-endian hex string.
31921
31922 @item X @var{AA}
31923 @itemx X @var{AA} ; process:@var{pid}
31924 The process terminated with signal @var{AA}.
31925
31926 The second form of the response, including the process ID of the
31927 terminated process, can be used only when @value{GDBN} has reported
31928 support for multiprocess protocol extensions; see @ref{multiprocess
31929 extensions}. The @var{pid} is formatted as a big-endian hex string.
31930
31931 @item O @var{XX}@dots{}
31932 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
31933 written as the program's console output. This can happen at any time
31934 while the program is running and the debugger should continue to wait
31935 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
31936
31937 @item F @var{call-id},@var{parameter}@dots{}
31938 @var{call-id} is the identifier which says which host system call should
31939 be called. This is just the name of the function. Translation into the
31940 correct system call is only applicable as it's defined in @value{GDBN}.
31941 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
31942 system calls.
31943
31944 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
31945 this very system call.
31946
31947 The target replies with this packet when it expects @value{GDBN} to
31948 call a host system call on behalf of the target. @value{GDBN} replies
31949 with an appropriate @samp{F} packet and keeps up waiting for the next
31950 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
31951 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
31952 Protocol Extension}, for more details.
31953
31954 @end table
31955
31956 @node General Query Packets
31957 @section General Query Packets
31958 @cindex remote query requests
31959
31960 Packets starting with @samp{q} are @dfn{general query packets};
31961 packets starting with @samp{Q} are @dfn{general set packets}. General
31962 query and set packets are a semi-unified form for retrieving and
31963 sending information to and from the stub.
31964
31965 The initial letter of a query or set packet is followed by a name
31966 indicating what sort of thing the packet applies to. For example,
31967 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
31968 definitions with the stub. These packet names follow some
31969 conventions:
31970
31971 @itemize @bullet
31972 @item
31973 The name must not contain commas, colons or semicolons.
31974 @item
31975 Most @value{GDBN} query and set packets have a leading upper case
31976 letter.
31977 @item
31978 The names of custom vendor packets should use a company prefix, in
31979 lower case, followed by a period. For example, packets designed at
31980 the Acme Corporation might begin with @samp{qacme.foo} (for querying
31981 foos) or @samp{Qacme.bar} (for setting bars).
31982 @end itemize
31983
31984 The name of a query or set packet should be separated from any
31985 parameters by a @samp{:}; the parameters themselves should be
31986 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
31987 full packet name, and check for a separator or the end of the packet,
31988 in case two packet names share a common prefix. New packets should not begin
31989 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
31990 packets predate these conventions, and have arguments without any terminator
31991 for the packet name; we suspect they are in widespread use in places that
31992 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
31993 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
31994 packet.}.
31995
31996 Like the descriptions of the other packets, each description here
31997 has a template showing the packet's overall syntax, followed by an
31998 explanation of the packet's meaning. We include spaces in some of the
31999 templates for clarity; these are not part of the packet's syntax. No
32000 @value{GDBN} packet uses spaces to separate its components.
32001
32002 Here are the currently defined query and set packets:
32003
32004 @table @samp
32005
32006 @item QAllow:@var{op}:@var{val}@dots{}
32007 @cindex @samp{QAllow} packet
32008 Specify which operations @value{GDBN} expects to request of the
32009 target, as a semicolon-separated list of operation name and value
32010 pairs. Possible values for @var{op} include @samp{WriteReg},
32011 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
32012 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
32013 indicating that @value{GDBN} will not request the operation, or 1,
32014 indicating that it may. (The target can then use this to set up its
32015 own internals optimally, for instance if the debugger never expects to
32016 insert breakpoints, it may not need to install its own trap handler.)
32017
32018 @item qC
32019 @cindex current thread, remote request
32020 @cindex @samp{qC} packet
32021 Return the current thread ID.
32022
32023 Reply:
32024 @table @samp
32025 @item QC @var{thread-id}
32026 Where @var{thread-id} is a thread ID as documented in
32027 @ref{thread-id syntax}.
32028 @item @r{(anything else)}
32029 Any other reply implies the old thread ID.
32030 @end table
32031
32032 @item qCRC:@var{addr},@var{length}
32033 @cindex CRC of memory block, remote request
32034 @cindex @samp{qCRC} packet
32035 Compute the CRC checksum of a block of memory using CRC-32 defined in
32036 IEEE 802.3. The CRC is computed byte at a time, taking the most
32037 significant bit of each byte first. The initial pattern code
32038 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
32039
32040 @emph{Note:} This is the same CRC used in validating separate debug
32041 files (@pxref{Separate Debug Files, , Debugging Information in Separate
32042 Files}). However the algorithm is slightly different. When validating
32043 separate debug files, the CRC is computed taking the @emph{least}
32044 significant bit of each byte first, and the final result is inverted to
32045 detect trailing zeros.
32046
32047 Reply:
32048 @table @samp
32049 @item E @var{NN}
32050 An error (such as memory fault)
32051 @item C @var{crc32}
32052 The specified memory region's checksum is @var{crc32}.
32053 @end table
32054
32055 @item qfThreadInfo
32056 @itemx qsThreadInfo
32057 @cindex list active threads, remote request
32058 @cindex @samp{qfThreadInfo} packet
32059 @cindex @samp{qsThreadInfo} packet
32060 Obtain a list of all active thread IDs from the target (OS). Since there
32061 may be too many active threads to fit into one reply packet, this query
32062 works iteratively: it may require more than one query/reply sequence to
32063 obtain the entire list of threads. The first query of the sequence will
32064 be the @samp{qfThreadInfo} query; subsequent queries in the
32065 sequence will be the @samp{qsThreadInfo} query.
32066
32067 NOTE: This packet replaces the @samp{qL} query (see below).
32068
32069 Reply:
32070 @table @samp
32071 @item m @var{thread-id}
32072 A single thread ID
32073 @item m @var{thread-id},@var{thread-id}@dots{}
32074 a comma-separated list of thread IDs
32075 @item l
32076 (lower case letter @samp{L}) denotes end of list.
32077 @end table
32078
32079 In response to each query, the target will reply with a list of one or
32080 more thread IDs, separated by commas.
32081 @value{GDBN} will respond to each reply with a request for more thread
32082 ids (using the @samp{qs} form of the query), until the target responds
32083 with @samp{l} (lower-case ell, for @dfn{last}).
32084 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
32085 fields.
32086
32087 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
32088 @cindex get thread-local storage address, remote request
32089 @cindex @samp{qGetTLSAddr} packet
32090 Fetch the address associated with thread local storage specified
32091 by @var{thread-id}, @var{offset}, and @var{lm}.
32092
32093 @var{thread-id} is the thread ID associated with the
32094 thread for which to fetch the TLS address. @xref{thread-id syntax}.
32095
32096 @var{offset} is the (big endian, hex encoded) offset associated with the
32097 thread local variable. (This offset is obtained from the debug
32098 information associated with the variable.)
32099
32100 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
32101 the load module associated with the thread local storage. For example,
32102 a @sc{gnu}/Linux system will pass the link map address of the shared
32103 object associated with the thread local storage under consideration.
32104 Other operating environments may choose to represent the load module
32105 differently, so the precise meaning of this parameter will vary.
32106
32107 Reply:
32108 @table @samp
32109 @item @var{XX}@dots{}
32110 Hex encoded (big endian) bytes representing the address of the thread
32111 local storage requested.
32112
32113 @item E @var{nn}
32114 An error occurred. @var{nn} are hex digits.
32115
32116 @item
32117 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
32118 @end table
32119
32120 @item qGetTIBAddr:@var{thread-id}
32121 @cindex get thread information block address
32122 @cindex @samp{qGetTIBAddr} packet
32123 Fetch address of the Windows OS specific Thread Information Block.
32124
32125 @var{thread-id} is the thread ID associated with the thread.
32126
32127 Reply:
32128 @table @samp
32129 @item @var{XX}@dots{}
32130 Hex encoded (big endian) bytes representing the linear address of the
32131 thread information block.
32132
32133 @item E @var{nn}
32134 An error occured. This means that either the thread was not found, or the
32135 address could not be retrieved.
32136
32137 @item
32138 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
32139 @end table
32140
32141 @item qL @var{startflag} @var{threadcount} @var{nextthread}
32142 Obtain thread information from RTOS. Where: @var{startflag} (one hex
32143 digit) is one to indicate the first query and zero to indicate a
32144 subsequent query; @var{threadcount} (two hex digits) is the maximum
32145 number of threads the response packet can contain; and @var{nextthread}
32146 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
32147 returned in the response as @var{argthread}.
32148
32149 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
32150
32151 Reply:
32152 @table @samp
32153 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
32154 Where: @var{count} (two hex digits) is the number of threads being
32155 returned; @var{done} (one hex digit) is zero to indicate more threads
32156 and one indicates no further threads; @var{argthreadid} (eight hex
32157 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
32158 is a sequence of thread IDs from the target. @var{threadid} (eight hex
32159 digits). See @code{remote.c:parse_threadlist_response()}.
32160 @end table
32161
32162 @item qOffsets
32163 @cindex section offsets, remote request
32164 @cindex @samp{qOffsets} packet
32165 Get section offsets that the target used when relocating the downloaded
32166 image.
32167
32168 Reply:
32169 @table @samp
32170 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
32171 Relocate the @code{Text} section by @var{xxx} from its original address.
32172 Relocate the @code{Data} section by @var{yyy} from its original address.
32173 If the object file format provides segment information (e.g.@: @sc{elf}
32174 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
32175 segments by the supplied offsets.
32176
32177 @emph{Note: while a @code{Bss} offset may be included in the response,
32178 @value{GDBN} ignores this and instead applies the @code{Data} offset
32179 to the @code{Bss} section.}
32180
32181 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
32182 Relocate the first segment of the object file, which conventionally
32183 contains program code, to a starting address of @var{xxx}. If
32184 @samp{DataSeg} is specified, relocate the second segment, which
32185 conventionally contains modifiable data, to a starting address of
32186 @var{yyy}. @value{GDBN} will report an error if the object file
32187 does not contain segment information, or does not contain at least
32188 as many segments as mentioned in the reply. Extra segments are
32189 kept at fixed offsets relative to the last relocated segment.
32190 @end table
32191
32192 @item qP @var{mode} @var{thread-id}
32193 @cindex thread information, remote request
32194 @cindex @samp{qP} packet
32195 Returns information on @var{thread-id}. Where: @var{mode} is a hex
32196 encoded 32 bit mode; @var{thread-id} is a thread ID
32197 (@pxref{thread-id syntax}).
32198
32199 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
32200 (see below).
32201
32202 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
32203
32204 @item QNonStop:1
32205 @item QNonStop:0
32206 @cindex non-stop mode, remote request
32207 @cindex @samp{QNonStop} packet
32208 @anchor{QNonStop}
32209 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
32210 @xref{Remote Non-Stop}, for more information.
32211
32212 Reply:
32213 @table @samp
32214 @item OK
32215 The request succeeded.
32216
32217 @item E @var{nn}
32218 An error occurred. @var{nn} are hex digits.
32219
32220 @item
32221 An empty reply indicates that @samp{QNonStop} is not supported by
32222 the stub.
32223 @end table
32224
32225 This packet is not probed by default; the remote stub must request it,
32226 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
32227 Use of this packet is controlled by the @code{set non-stop} command;
32228 @pxref{Non-Stop Mode}.
32229
32230 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
32231 @cindex pass signals to inferior, remote request
32232 @cindex @samp{QPassSignals} packet
32233 @anchor{QPassSignals}
32234 Each listed @var{signal} should be passed directly to the inferior process.
32235 Signals are numbered identically to continue packets and stop replies
32236 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
32237 strictly greater than the previous item. These signals do not need to stop
32238 the inferior, or be reported to @value{GDBN}. All other signals should be
32239 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
32240 combine; any earlier @samp{QPassSignals} list is completely replaced by the
32241 new list. This packet improves performance when using @samp{handle
32242 @var{signal} nostop noprint pass}.
32243
32244 Reply:
32245 @table @samp
32246 @item OK
32247 The request succeeded.
32248
32249 @item E @var{nn}
32250 An error occurred. @var{nn} are hex digits.
32251
32252 @item
32253 An empty reply indicates that @samp{QPassSignals} is not supported by
32254 the stub.
32255 @end table
32256
32257 Use of this packet is controlled by the @code{set remote pass-signals}
32258 command (@pxref{Remote Configuration, set remote pass-signals}).
32259 This packet is not probed by default; the remote stub must request it,
32260 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
32261
32262 @item qRcmd,@var{command}
32263 @cindex execute remote command, remote request
32264 @cindex @samp{qRcmd} packet
32265 @var{command} (hex encoded) is passed to the local interpreter for
32266 execution. Invalid commands should be reported using the output
32267 string. Before the final result packet, the target may also respond
32268 with a number of intermediate @samp{O@var{output}} console output
32269 packets. @emph{Implementors should note that providing access to a
32270 stubs's interpreter may have security implications}.
32271
32272 Reply:
32273 @table @samp
32274 @item OK
32275 A command response with no output.
32276 @item @var{OUTPUT}
32277 A command response with the hex encoded output string @var{OUTPUT}.
32278 @item E @var{NN}
32279 Indicate a badly formed request.
32280 @item
32281 An empty reply indicates that @samp{qRcmd} is not recognized.
32282 @end table
32283
32284 (Note that the @code{qRcmd} packet's name is separated from the
32285 command by a @samp{,}, not a @samp{:}, contrary to the naming
32286 conventions above. Please don't use this packet as a model for new
32287 packets.)
32288
32289 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
32290 @cindex searching memory, in remote debugging
32291 @cindex @samp{qSearch:memory} packet
32292 @anchor{qSearch memory}
32293 Search @var{length} bytes at @var{address} for @var{search-pattern}.
32294 @var{address} and @var{length} are encoded in hex.
32295 @var{search-pattern} is a sequence of bytes, hex encoded.
32296
32297 Reply:
32298 @table @samp
32299 @item 0
32300 The pattern was not found.
32301 @item 1,address
32302 The pattern was found at @var{address}.
32303 @item E @var{NN}
32304 A badly formed request or an error was encountered while searching memory.
32305 @item
32306 An empty reply indicates that @samp{qSearch:memory} is not recognized.
32307 @end table
32308
32309 @item QStartNoAckMode
32310 @cindex @samp{QStartNoAckMode} packet
32311 @anchor{QStartNoAckMode}
32312 Request that the remote stub disable the normal @samp{+}/@samp{-}
32313 protocol acknowledgments (@pxref{Packet Acknowledgment}).
32314
32315 Reply:
32316 @table @samp
32317 @item OK
32318 The stub has switched to no-acknowledgment mode.
32319 @value{GDBN} acknowledges this reponse,
32320 but neither the stub nor @value{GDBN} shall send or expect further
32321 @samp{+}/@samp{-} acknowledgments in the current connection.
32322 @item
32323 An empty reply indicates that the stub does not support no-acknowledgment mode.
32324 @end table
32325
32326 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
32327 @cindex supported packets, remote query
32328 @cindex features of the remote protocol
32329 @cindex @samp{qSupported} packet
32330 @anchor{qSupported}
32331 Tell the remote stub about features supported by @value{GDBN}, and
32332 query the stub for features it supports. This packet allows
32333 @value{GDBN} and the remote stub to take advantage of each others'
32334 features. @samp{qSupported} also consolidates multiple feature probes
32335 at startup, to improve @value{GDBN} performance---a single larger
32336 packet performs better than multiple smaller probe packets on
32337 high-latency links. Some features may enable behavior which must not
32338 be on by default, e.g.@: because it would confuse older clients or
32339 stubs. Other features may describe packets which could be
32340 automatically probed for, but are not. These features must be
32341 reported before @value{GDBN} will use them. This ``default
32342 unsupported'' behavior is not appropriate for all packets, but it
32343 helps to keep the initial connection time under control with new
32344 versions of @value{GDBN} which support increasing numbers of packets.
32345
32346 Reply:
32347 @table @samp
32348 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
32349 The stub supports or does not support each returned @var{stubfeature},
32350 depending on the form of each @var{stubfeature} (see below for the
32351 possible forms).
32352 @item
32353 An empty reply indicates that @samp{qSupported} is not recognized,
32354 or that no features needed to be reported to @value{GDBN}.
32355 @end table
32356
32357 The allowed forms for each feature (either a @var{gdbfeature} in the
32358 @samp{qSupported} packet, or a @var{stubfeature} in the response)
32359 are:
32360
32361 @table @samp
32362 @item @var{name}=@var{value}
32363 The remote protocol feature @var{name} is supported, and associated
32364 with the specified @var{value}. The format of @var{value} depends
32365 on the feature, but it must not include a semicolon.
32366 @item @var{name}+
32367 The remote protocol feature @var{name} is supported, and does not
32368 need an associated value.
32369 @item @var{name}-
32370 The remote protocol feature @var{name} is not supported.
32371 @item @var{name}?
32372 The remote protocol feature @var{name} may be supported, and
32373 @value{GDBN} should auto-detect support in some other way when it is
32374 needed. This form will not be used for @var{gdbfeature} notifications,
32375 but may be used for @var{stubfeature} responses.
32376 @end table
32377
32378 Whenever the stub receives a @samp{qSupported} request, the
32379 supplied set of @value{GDBN} features should override any previous
32380 request. This allows @value{GDBN} to put the stub in a known
32381 state, even if the stub had previously been communicating with
32382 a different version of @value{GDBN}.
32383
32384 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
32385 are defined:
32386
32387 @table @samp
32388 @item multiprocess
32389 This feature indicates whether @value{GDBN} supports multiprocess
32390 extensions to the remote protocol. @value{GDBN} does not use such
32391 extensions unless the stub also reports that it supports them by
32392 including @samp{multiprocess+} in its @samp{qSupported} reply.
32393 @xref{multiprocess extensions}, for details.
32394
32395 @item xmlRegisters
32396 This feature indicates that @value{GDBN} supports the XML target
32397 description. If the stub sees @samp{xmlRegisters=} with target
32398 specific strings separated by a comma, it will report register
32399 description.
32400
32401 @item qRelocInsn
32402 This feature indicates whether @value{GDBN} supports the
32403 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
32404 instruction reply packet}).
32405 @end table
32406
32407 Stubs should ignore any unknown values for
32408 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
32409 packet supports receiving packets of unlimited length (earlier
32410 versions of @value{GDBN} may reject overly long responses). Additional values
32411 for @var{gdbfeature} may be defined in the future to let the stub take
32412 advantage of new features in @value{GDBN}, e.g.@: incompatible
32413 improvements in the remote protocol---the @samp{multiprocess} feature is
32414 an example of such a feature. The stub's reply should be independent
32415 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
32416 describes all the features it supports, and then the stub replies with
32417 all the features it supports.
32418
32419 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
32420 responses, as long as each response uses one of the standard forms.
32421
32422 Some features are flags. A stub which supports a flag feature
32423 should respond with a @samp{+} form response. Other features
32424 require values, and the stub should respond with an @samp{=}
32425 form response.
32426
32427 Each feature has a default value, which @value{GDBN} will use if
32428 @samp{qSupported} is not available or if the feature is not mentioned
32429 in the @samp{qSupported} response. The default values are fixed; a
32430 stub is free to omit any feature responses that match the defaults.
32431
32432 Not all features can be probed, but for those which can, the probing
32433 mechanism is useful: in some cases, a stub's internal
32434 architecture may not allow the protocol layer to know some information
32435 about the underlying target in advance. This is especially common in
32436 stubs which may be configured for multiple targets.
32437
32438 These are the currently defined stub features and their properties:
32439
32440 @multitable @columnfractions 0.35 0.2 0.12 0.2
32441 @c NOTE: The first row should be @headitem, but we do not yet require
32442 @c a new enough version of Texinfo (4.7) to use @headitem.
32443 @item Feature Name
32444 @tab Value Required
32445 @tab Default
32446 @tab Probe Allowed
32447
32448 @item @samp{PacketSize}
32449 @tab Yes
32450 @tab @samp{-}
32451 @tab No
32452
32453 @item @samp{qXfer:auxv:read}
32454 @tab No
32455 @tab @samp{-}
32456 @tab Yes
32457
32458 @item @samp{qXfer:features:read}
32459 @tab No
32460 @tab @samp{-}
32461 @tab Yes
32462
32463 @item @samp{qXfer:libraries:read}
32464 @tab No
32465 @tab @samp{-}
32466 @tab Yes
32467
32468 @item @samp{qXfer:memory-map:read}
32469 @tab No
32470 @tab @samp{-}
32471 @tab Yes
32472
32473 @item @samp{qXfer:sdata:read}
32474 @tab No
32475 @tab @samp{-}
32476 @tab Yes
32477
32478 @item @samp{qXfer:spu:read}
32479 @tab No
32480 @tab @samp{-}
32481 @tab Yes
32482
32483 @item @samp{qXfer:spu:write}
32484 @tab No
32485 @tab @samp{-}
32486 @tab Yes
32487
32488 @item @samp{qXfer:siginfo:read}
32489 @tab No
32490 @tab @samp{-}
32491 @tab Yes
32492
32493 @item @samp{qXfer:siginfo:write}
32494 @tab No
32495 @tab @samp{-}
32496 @tab Yes
32497
32498 @item @samp{qXfer:threads:read}
32499 @tab No
32500 @tab @samp{-}
32501 @tab Yes
32502
32503
32504 @item @samp{QNonStop}
32505 @tab No
32506 @tab @samp{-}
32507 @tab Yes
32508
32509 @item @samp{QPassSignals}
32510 @tab No
32511 @tab @samp{-}
32512 @tab Yes
32513
32514 @item @samp{QStartNoAckMode}
32515 @tab No
32516 @tab @samp{-}
32517 @tab Yes
32518
32519 @item @samp{multiprocess}
32520 @tab No
32521 @tab @samp{-}
32522 @tab No
32523
32524 @item @samp{ConditionalTracepoints}
32525 @tab No
32526 @tab @samp{-}
32527 @tab No
32528
32529 @item @samp{ReverseContinue}
32530 @tab No
32531 @tab @samp{-}
32532 @tab No
32533
32534 @item @samp{ReverseStep}
32535 @tab No
32536 @tab @samp{-}
32537 @tab No
32538
32539 @item @samp{TracepointSource}
32540 @tab No
32541 @tab @samp{-}
32542 @tab No
32543
32544 @item @samp{QAllow}
32545 @tab No
32546 @tab @samp{-}
32547 @tab No
32548
32549 @end multitable
32550
32551 These are the currently defined stub features, in more detail:
32552
32553 @table @samp
32554 @cindex packet size, remote protocol
32555 @item PacketSize=@var{bytes}
32556 The remote stub can accept packets up to at least @var{bytes} in
32557 length. @value{GDBN} will send packets up to this size for bulk
32558 transfers, and will never send larger packets. This is a limit on the
32559 data characters in the packet, including the frame and checksum.
32560 There is no trailing NUL byte in a remote protocol packet; if the stub
32561 stores packets in a NUL-terminated format, it should allow an extra
32562 byte in its buffer for the NUL. If this stub feature is not supported,
32563 @value{GDBN} guesses based on the size of the @samp{g} packet response.
32564
32565 @item qXfer:auxv:read
32566 The remote stub understands the @samp{qXfer:auxv:read} packet
32567 (@pxref{qXfer auxiliary vector read}).
32568
32569 @item qXfer:features:read
32570 The remote stub understands the @samp{qXfer:features:read} packet
32571 (@pxref{qXfer target description read}).
32572
32573 @item qXfer:libraries:read
32574 The remote stub understands the @samp{qXfer:libraries:read} packet
32575 (@pxref{qXfer library list read}).
32576
32577 @item qXfer:memory-map:read
32578 The remote stub understands the @samp{qXfer:memory-map:read} packet
32579 (@pxref{qXfer memory map read}).
32580
32581 @item qXfer:sdata:read
32582 The remote stub understands the @samp{qXfer:sdata:read} packet
32583 (@pxref{qXfer sdata read}).
32584
32585 @item qXfer:spu:read
32586 The remote stub understands the @samp{qXfer:spu:read} packet
32587 (@pxref{qXfer spu read}).
32588
32589 @item qXfer:spu:write
32590 The remote stub understands the @samp{qXfer:spu:write} packet
32591 (@pxref{qXfer spu write}).
32592
32593 @item qXfer:siginfo:read
32594 The remote stub understands the @samp{qXfer:siginfo:read} packet
32595 (@pxref{qXfer siginfo read}).
32596
32597 @item qXfer:siginfo:write
32598 The remote stub understands the @samp{qXfer:siginfo:write} packet
32599 (@pxref{qXfer siginfo write}).
32600
32601 @item qXfer:threads:read
32602 The remote stub understands the @samp{qXfer:threads:read} packet
32603 (@pxref{qXfer threads read}).
32604
32605 @item QNonStop
32606 The remote stub understands the @samp{QNonStop} packet
32607 (@pxref{QNonStop}).
32608
32609 @item QPassSignals
32610 The remote stub understands the @samp{QPassSignals} packet
32611 (@pxref{QPassSignals}).
32612
32613 @item QStartNoAckMode
32614 The remote stub understands the @samp{QStartNoAckMode} packet and
32615 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
32616
32617 @item multiprocess
32618 @anchor{multiprocess extensions}
32619 @cindex multiprocess extensions, in remote protocol
32620 The remote stub understands the multiprocess extensions to the remote
32621 protocol syntax. The multiprocess extensions affect the syntax of
32622 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
32623 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
32624 replies. Note that reporting this feature indicates support for the
32625 syntactic extensions only, not that the stub necessarily supports
32626 debugging of more than one process at a time. The stub must not use
32627 multiprocess extensions in packet replies unless @value{GDBN} has also
32628 indicated it supports them in its @samp{qSupported} request.
32629
32630 @item qXfer:osdata:read
32631 The remote stub understands the @samp{qXfer:osdata:read} packet
32632 ((@pxref{qXfer osdata read}).
32633
32634 @item ConditionalTracepoints
32635 The remote stub accepts and implements conditional expressions defined
32636 for tracepoints (@pxref{Tracepoint Conditions}).
32637
32638 @item ReverseContinue
32639 The remote stub accepts and implements the reverse continue packet
32640 (@pxref{bc}).
32641
32642 @item ReverseStep
32643 The remote stub accepts and implements the reverse step packet
32644 (@pxref{bs}).
32645
32646 @item TracepointSource
32647 The remote stub understands the @samp{QTDPsrc} packet that supplies
32648 the source form of tracepoint definitions.
32649
32650 @item QAllow
32651 The remote stub understands the @samp{QAllow} packet.
32652
32653 @item StaticTracepoint
32654 @cindex static tracepoints, in remote protocol
32655 The remote stub supports static tracepoints.
32656
32657 @end table
32658
32659 @item qSymbol::
32660 @cindex symbol lookup, remote request
32661 @cindex @samp{qSymbol} packet
32662 Notify the target that @value{GDBN} is prepared to serve symbol lookup
32663 requests. Accept requests from the target for the values of symbols.
32664
32665 Reply:
32666 @table @samp
32667 @item OK
32668 The target does not need to look up any (more) symbols.
32669 @item qSymbol:@var{sym_name}
32670 The target requests the value of symbol @var{sym_name} (hex encoded).
32671 @value{GDBN} may provide the value by using the
32672 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
32673 below.
32674 @end table
32675
32676 @item qSymbol:@var{sym_value}:@var{sym_name}
32677 Set the value of @var{sym_name} to @var{sym_value}.
32678
32679 @var{sym_name} (hex encoded) is the name of a symbol whose value the
32680 target has previously requested.
32681
32682 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
32683 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
32684 will be empty.
32685
32686 Reply:
32687 @table @samp
32688 @item OK
32689 The target does not need to look up any (more) symbols.
32690 @item qSymbol:@var{sym_name}
32691 The target requests the value of a new symbol @var{sym_name} (hex
32692 encoded). @value{GDBN} will continue to supply the values of symbols
32693 (if available), until the target ceases to request them.
32694 @end table
32695
32696 @item qTBuffer
32697 @item QTBuffer
32698 @item QTDisconnected
32699 @itemx QTDP
32700 @itemx QTDPsrc
32701 @itemx QTDV
32702 @itemx qTfP
32703 @itemx qTfV
32704 @itemx QTFrame
32705 @xref{Tracepoint Packets}.
32706
32707 @item qThreadExtraInfo,@var{thread-id}
32708 @cindex thread attributes info, remote request
32709 @cindex @samp{qThreadExtraInfo} packet
32710 Obtain a printable string description of a thread's attributes from
32711 the target OS. @var{thread-id} is a thread ID;
32712 see @ref{thread-id syntax}. This
32713 string may contain anything that the target OS thinks is interesting
32714 for @value{GDBN} to tell the user about the thread. The string is
32715 displayed in @value{GDBN}'s @code{info threads} display. Some
32716 examples of possible thread extra info strings are @samp{Runnable}, or
32717 @samp{Blocked on Mutex}.
32718
32719 Reply:
32720 @table @samp
32721 @item @var{XX}@dots{}
32722 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
32723 comprising the printable string containing the extra information about
32724 the thread's attributes.
32725 @end table
32726
32727 (Note that the @code{qThreadExtraInfo} packet's name is separated from
32728 the command by a @samp{,}, not a @samp{:}, contrary to the naming
32729 conventions above. Please don't use this packet as a model for new
32730 packets.)
32731
32732 @item QTSave
32733 @item qTsP
32734 @item qTsV
32735 @itemx QTStart
32736 @itemx QTStop
32737 @itemx QTinit
32738 @itemx QTro
32739 @itemx qTStatus
32740 @itemx qTV
32741 @itemx qTfSTM
32742 @itemx qTsSTM
32743 @itemx qTSTMat
32744 @xref{Tracepoint Packets}.
32745
32746 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
32747 @cindex read special object, remote request
32748 @cindex @samp{qXfer} packet
32749 @anchor{qXfer read}
32750 Read uninterpreted bytes from the target's special data area
32751 identified by the keyword @var{object}. Request @var{length} bytes
32752 starting at @var{offset} bytes into the data. The content and
32753 encoding of @var{annex} is specific to @var{object}; it can supply
32754 additional details about what data to access.
32755
32756 Here are the specific requests of this form defined so far. All
32757 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
32758 formats, listed below.
32759
32760 @table @samp
32761 @item qXfer:auxv:read::@var{offset},@var{length}
32762 @anchor{qXfer auxiliary vector read}
32763 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
32764 auxiliary vector}. Note @var{annex} must be empty.
32765
32766 This packet is not probed by default; the remote stub must request it,
32767 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
32768
32769 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
32770 @anchor{qXfer target description read}
32771 Access the @dfn{target description}. @xref{Target Descriptions}. The
32772 annex specifies which XML document to access. The main description is
32773 always loaded from the @samp{target.xml} annex.
32774
32775 This packet is not probed by default; the remote stub must request it,
32776 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
32777
32778 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
32779 @anchor{qXfer library list read}
32780 Access the target's list of loaded libraries. @xref{Library List Format}.
32781 The annex part of the generic @samp{qXfer} packet must be empty
32782 (@pxref{qXfer read}).
32783
32784 Targets which maintain a list of libraries in the program's memory do
32785 not need to implement this packet; it is designed for platforms where
32786 the operating system manages the list of loaded libraries.
32787
32788 This packet is not probed by default; the remote stub must request it,
32789 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
32790
32791 @item qXfer:memory-map:read::@var{offset},@var{length}
32792 @anchor{qXfer memory map read}
32793 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
32794 annex part of the generic @samp{qXfer} packet must be empty
32795 (@pxref{qXfer read}).
32796
32797 This packet is not probed by default; the remote stub must request it,
32798 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
32799
32800 @item qXfer:sdata:read::@var{offset},@var{length}
32801 @anchor{qXfer sdata read}
32802
32803 Read contents of the extra collected static tracepoint marker
32804 information. The annex part of the generic @samp{qXfer} packet must
32805 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
32806 Action Lists}.
32807
32808 This packet is not probed by default; the remote stub must request it,
32809 by supplying an appropriate @samp{qSupported} response
32810 (@pxref{qSupported}).
32811
32812 @item qXfer:siginfo:read::@var{offset},@var{length}
32813 @anchor{qXfer siginfo read}
32814 Read contents of the extra signal information on the target
32815 system. The annex part of the generic @samp{qXfer} packet must be
32816 empty (@pxref{qXfer read}).
32817
32818 This packet is not probed by default; the remote stub must request it,
32819 by supplying an appropriate @samp{qSupported} response
32820 (@pxref{qSupported}).
32821
32822 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
32823 @anchor{qXfer spu read}
32824 Read contents of an @code{spufs} file on the target system. The
32825 annex specifies which file to read; it must be of the form
32826 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
32827 in the target process, and @var{name} identifes the @code{spufs} file
32828 in that context to be accessed.
32829
32830 This packet is not probed by default; the remote stub must request it,
32831 by supplying an appropriate @samp{qSupported} response
32832 (@pxref{qSupported}).
32833
32834 @item qXfer:threads:read::@var{offset},@var{length}
32835 @anchor{qXfer threads read}
32836 Access the list of threads on target. @xref{Thread List Format}. The
32837 annex part of the generic @samp{qXfer} packet must be empty
32838 (@pxref{qXfer read}).
32839
32840 This packet is not probed by default; the remote stub must request it,
32841 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
32842
32843 @item qXfer:osdata:read::@var{offset},@var{length}
32844 @anchor{qXfer osdata read}
32845 Access the target's @dfn{operating system information}.
32846 @xref{Operating System Information}.
32847
32848 @end table
32849
32850 Reply:
32851 @table @samp
32852 @item m @var{data}
32853 Data @var{data} (@pxref{Binary Data}) has been read from the
32854 target. There may be more data at a higher address (although
32855 it is permitted to return @samp{m} even for the last valid
32856 block of data, as long as at least one byte of data was read).
32857 @var{data} may have fewer bytes than the @var{length} in the
32858 request.
32859
32860 @item l @var{data}
32861 Data @var{data} (@pxref{Binary Data}) has been read from the target.
32862 There is no more data to be read. @var{data} may have fewer bytes
32863 than the @var{length} in the request.
32864
32865 @item l
32866 The @var{offset} in the request is at the end of the data.
32867 There is no more data to be read.
32868
32869 @item E00
32870 The request was malformed, or @var{annex} was invalid.
32871
32872 @item E @var{nn}
32873 The offset was invalid, or there was an error encountered reading the data.
32874 @var{nn} is a hex-encoded @code{errno} value.
32875
32876 @item
32877 An empty reply indicates the @var{object} string was not recognized by
32878 the stub, or that the object does not support reading.
32879 @end table
32880
32881 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
32882 @cindex write data into object, remote request
32883 @anchor{qXfer write}
32884 Write uninterpreted bytes into the target's special data area
32885 identified by the keyword @var{object}, starting at @var{offset} bytes
32886 into the data. @var{data}@dots{} is the binary-encoded data
32887 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
32888 is specific to @var{object}; it can supply additional details about what data
32889 to access.
32890
32891 Here are the specific requests of this form defined so far. All
32892 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
32893 formats, listed below.
32894
32895 @table @samp
32896 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
32897 @anchor{qXfer siginfo write}
32898 Write @var{data} to the extra signal information on the target system.
32899 The annex part of the generic @samp{qXfer} packet must be
32900 empty (@pxref{qXfer write}).
32901
32902 This packet is not probed by default; the remote stub must request it,
32903 by supplying an appropriate @samp{qSupported} response
32904 (@pxref{qSupported}).
32905
32906 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
32907 @anchor{qXfer spu write}
32908 Write @var{data} to an @code{spufs} file on the target system. The
32909 annex specifies which file to write; it must be of the form
32910 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
32911 in the target process, and @var{name} identifes the @code{spufs} file
32912 in that context to be accessed.
32913
32914 This packet is not probed by default; the remote stub must request it,
32915 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
32916 @end table
32917
32918 Reply:
32919 @table @samp
32920 @item @var{nn}
32921 @var{nn} (hex encoded) is the number of bytes written.
32922 This may be fewer bytes than supplied in the request.
32923
32924 @item E00
32925 The request was malformed, or @var{annex} was invalid.
32926
32927 @item E @var{nn}
32928 The offset was invalid, or there was an error encountered writing the data.
32929 @var{nn} is a hex-encoded @code{errno} value.
32930
32931 @item
32932 An empty reply indicates the @var{object} string was not
32933 recognized by the stub, or that the object does not support writing.
32934 @end table
32935
32936 @item qXfer:@var{object}:@var{operation}:@dots{}
32937 Requests of this form may be added in the future. When a stub does
32938 not recognize the @var{object} keyword, or its support for
32939 @var{object} does not recognize the @var{operation} keyword, the stub
32940 must respond with an empty packet.
32941
32942 @item qAttached:@var{pid}
32943 @cindex query attached, remote request
32944 @cindex @samp{qAttached} packet
32945 Return an indication of whether the remote server attached to an
32946 existing process or created a new process. When the multiprocess
32947 protocol extensions are supported (@pxref{multiprocess extensions}),
32948 @var{pid} is an integer in hexadecimal format identifying the target
32949 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
32950 the query packet will be simplified as @samp{qAttached}.
32951
32952 This query is used, for example, to know whether the remote process
32953 should be detached or killed when a @value{GDBN} session is ended with
32954 the @code{quit} command.
32955
32956 Reply:
32957 @table @samp
32958 @item 1
32959 The remote server attached to an existing process.
32960 @item 0
32961 The remote server created a new process.
32962 @item E @var{NN}
32963 A badly formed request or an error was encountered.
32964 @end table
32965
32966 @end table
32967
32968 @node Architecture-Specific Protocol Details
32969 @section Architecture-Specific Protocol Details
32970
32971 This section describes how the remote protocol is applied to specific
32972 target architectures. Also see @ref{Standard Target Features}, for
32973 details of XML target descriptions for each architecture.
32974
32975 @subsection ARM
32976
32977 @subsubsection Breakpoint Kinds
32978
32979 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
32980
32981 @table @r
32982
32983 @item 2
32984 16-bit Thumb mode breakpoint.
32985
32986 @item 3
32987 32-bit Thumb mode (Thumb-2) breakpoint.
32988
32989 @item 4
32990 32-bit ARM mode breakpoint.
32991
32992 @end table
32993
32994 @subsection MIPS
32995
32996 @subsubsection Register Packet Format
32997
32998 The following @code{g}/@code{G} packets have previously been defined.
32999 In the below, some thirty-two bit registers are transferred as
33000 sixty-four bits. Those registers should be zero/sign extended (which?)
33001 to fill the space allocated. Register bytes are transferred in target
33002 byte order. The two nibbles within a register byte are transferred
33003 most-significant - least-significant.
33004
33005 @table @r
33006
33007 @item MIPS32
33008
33009 All registers are transferred as thirty-two bit quantities in the order:
33010 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
33011 registers; fsr; fir; fp.
33012
33013 @item MIPS64
33014
33015 All registers are transferred as sixty-four bit quantities (including
33016 thirty-two bit registers such as @code{sr}). The ordering is the same
33017 as @code{MIPS32}.
33018
33019 @end table
33020
33021 @node Tracepoint Packets
33022 @section Tracepoint Packets
33023 @cindex tracepoint packets
33024 @cindex packets, tracepoint
33025
33026 Here we describe the packets @value{GDBN} uses to implement
33027 tracepoints (@pxref{Tracepoints}).
33028
33029 @table @samp
33030
33031 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
33032 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
33033 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
33034 the tracepoint is disabled. @var{step} is the tracepoint's step
33035 count, and @var{pass} is its pass count. If an @samp{F} is present,
33036 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
33037 the number of bytes that the target should copy elsewhere to make room
33038 for the tracepoint. If an @samp{X} is present, it introduces a
33039 tracepoint condition, which consists of a hexadecimal length, followed
33040 by a comma and hex-encoded bytes, in a manner similar to action
33041 encodings as described below. If the trailing @samp{-} is present,
33042 further @samp{QTDP} packets will follow to specify this tracepoint's
33043 actions.
33044
33045 Replies:
33046 @table @samp
33047 @item OK
33048 The packet was understood and carried out.
33049 @item qRelocInsn
33050 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
33051 @item
33052 The packet was not recognized.
33053 @end table
33054
33055 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
33056 Define actions to be taken when a tracepoint is hit. @var{n} and
33057 @var{addr} must be the same as in the initial @samp{QTDP} packet for
33058 this tracepoint. This packet may only be sent immediately after
33059 another @samp{QTDP} packet that ended with a @samp{-}. If the
33060 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
33061 specifying more actions for this tracepoint.
33062
33063 In the series of action packets for a given tracepoint, at most one
33064 can have an @samp{S} before its first @var{action}. If such a packet
33065 is sent, it and the following packets define ``while-stepping''
33066 actions. Any prior packets define ordinary actions --- that is, those
33067 taken when the tracepoint is first hit. If no action packet has an
33068 @samp{S}, then all the packets in the series specify ordinary
33069 tracepoint actions.
33070
33071 The @samp{@var{action}@dots{}} portion of the packet is a series of
33072 actions, concatenated without separators. Each action has one of the
33073 following forms:
33074
33075 @table @samp
33076
33077 @item R @var{mask}
33078 Collect the registers whose bits are set in @var{mask}. @var{mask} is
33079 a hexadecimal number whose @var{i}'th bit is set if register number
33080 @var{i} should be collected. (The least significant bit is numbered
33081 zero.) Note that @var{mask} may be any number of digits long; it may
33082 not fit in a 32-bit word.
33083
33084 @item M @var{basereg},@var{offset},@var{len}
33085 Collect @var{len} bytes of memory starting at the address in register
33086 number @var{basereg}, plus @var{offset}. If @var{basereg} is
33087 @samp{-1}, then the range has a fixed address: @var{offset} is the
33088 address of the lowest byte to collect. The @var{basereg},
33089 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
33090 values (the @samp{-1} value for @var{basereg} is a special case).
33091
33092 @item X @var{len},@var{expr}
33093 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
33094 it directs. @var{expr} is an agent expression, as described in
33095 @ref{Agent Expressions}. Each byte of the expression is encoded as a
33096 two-digit hex number in the packet; @var{len} is the number of bytes
33097 in the expression (and thus one-half the number of hex digits in the
33098 packet).
33099
33100 @end table
33101
33102 Any number of actions may be packed together in a single @samp{QTDP}
33103 packet, as long as the packet does not exceed the maximum packet
33104 length (400 bytes, for many stubs). There may be only one @samp{R}
33105 action per tracepoint, and it must precede any @samp{M} or @samp{X}
33106 actions. Any registers referred to by @samp{M} and @samp{X} actions
33107 must be collected by a preceding @samp{R} action. (The
33108 ``while-stepping'' actions are treated as if they were attached to a
33109 separate tracepoint, as far as these restrictions are concerned.)
33110
33111 Replies:
33112 @table @samp
33113 @item OK
33114 The packet was understood and carried out.
33115 @item qRelocInsn
33116 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
33117 @item
33118 The packet was not recognized.
33119 @end table
33120
33121 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
33122 @cindex @samp{QTDPsrc} packet
33123 Specify a source string of tracepoint @var{n} at address @var{addr}.
33124 This is useful to get accurate reproduction of the tracepoints
33125 originally downloaded at the beginning of the trace run. @var{type}
33126 is the name of the tracepoint part, such as @samp{cond} for the
33127 tracepoint's conditional expression (see below for a list of types), while
33128 @var{bytes} is the string, encoded in hexadecimal.
33129
33130 @var{start} is the offset of the @var{bytes} within the overall source
33131 string, while @var{slen} is the total length of the source string.
33132 This is intended for handling source strings that are longer than will
33133 fit in a single packet.
33134 @c Add detailed example when this info is moved into a dedicated
33135 @c tracepoint descriptions section.
33136
33137 The available string types are @samp{at} for the location,
33138 @samp{cond} for the conditional, and @samp{cmd} for an action command.
33139 @value{GDBN} sends a separate packet for each command in the action
33140 list, in the same order in which the commands are stored in the list.
33141
33142 The target does not need to do anything with source strings except
33143 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
33144 query packets.
33145
33146 Although this packet is optional, and @value{GDBN} will only send it
33147 if the target replies with @samp{TracepointSource} @xref{General
33148 Query Packets}, it makes both disconnected tracing and trace files
33149 much easier to use. Otherwise the user must be careful that the
33150 tracepoints in effect while looking at trace frames are identical to
33151 the ones in effect during the trace run; even a small discrepancy
33152 could cause @samp{tdump} not to work, or a particular trace frame not
33153 be found.
33154
33155 @item QTDV:@var{n}:@var{value}
33156 @cindex define trace state variable, remote request
33157 @cindex @samp{QTDV} packet
33158 Create a new trace state variable, number @var{n}, with an initial
33159 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
33160 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
33161 the option of not using this packet for initial values of zero; the
33162 target should simply create the trace state variables as they are
33163 mentioned in expressions.
33164
33165 @item QTFrame:@var{n}
33166 Select the @var{n}'th tracepoint frame from the buffer, and use the
33167 register and memory contents recorded there to answer subsequent
33168 request packets from @value{GDBN}.
33169
33170 A successful reply from the stub indicates that the stub has found the
33171 requested frame. The response is a series of parts, concatenated
33172 without separators, describing the frame we selected. Each part has
33173 one of the following forms:
33174
33175 @table @samp
33176 @item F @var{f}
33177 The selected frame is number @var{n} in the trace frame buffer;
33178 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
33179 was no frame matching the criteria in the request packet.
33180
33181 @item T @var{t}
33182 The selected trace frame records a hit of tracepoint number @var{t};
33183 @var{t} is a hexadecimal number.
33184
33185 @end table
33186
33187 @item QTFrame:pc:@var{addr}
33188 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
33189 currently selected frame whose PC is @var{addr};
33190 @var{addr} is a hexadecimal number.
33191
33192 @item QTFrame:tdp:@var{t}
33193 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
33194 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
33195 is a hexadecimal number.
33196
33197 @item QTFrame:range:@var{start}:@var{end}
33198 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
33199 currently selected frame whose PC is between @var{start} (inclusive)
33200 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
33201 numbers.
33202
33203 @item QTFrame:outside:@var{start}:@var{end}
33204 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
33205 frame @emph{outside} the given range of addresses (exclusive).
33206
33207 @item QTStart
33208 Begin the tracepoint experiment. Begin collecting data from
33209 tracepoint hits in the trace frame buffer. This packet supports the
33210 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
33211 instruction reply packet}).
33212
33213 @item QTStop
33214 End the tracepoint experiment. Stop collecting trace frames.
33215
33216 @item QTinit
33217 Clear the table of tracepoints, and empty the trace frame buffer.
33218
33219 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
33220 Establish the given ranges of memory as ``transparent''. The stub
33221 will answer requests for these ranges from memory's current contents,
33222 if they were not collected as part of the tracepoint hit.
33223
33224 @value{GDBN} uses this to mark read-only regions of memory, like those
33225 containing program code. Since these areas never change, they should
33226 still have the same contents they did when the tracepoint was hit, so
33227 there's no reason for the stub to refuse to provide their contents.
33228
33229 @item QTDisconnected:@var{value}
33230 Set the choice to what to do with the tracing run when @value{GDBN}
33231 disconnects from the target. A @var{value} of 1 directs the target to
33232 continue the tracing run, while 0 tells the target to stop tracing if
33233 @value{GDBN} is no longer in the picture.
33234
33235 @item qTStatus
33236 Ask the stub if there is a trace experiment running right now.
33237
33238 The reply has the form:
33239
33240 @table @samp
33241
33242 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
33243 @var{running} is a single digit @code{1} if the trace is presently
33244 running, or @code{0} if not. It is followed by semicolon-separated
33245 optional fields that an agent may use to report additional status.
33246
33247 @end table
33248
33249 If the trace is not running, the agent may report any of several
33250 explanations as one of the optional fields:
33251
33252 @table @samp
33253
33254 @item tnotrun:0
33255 No trace has been run yet.
33256
33257 @item tstop:0
33258 The trace was stopped by a user-originated stop command.
33259
33260 @item tfull:0
33261 The trace stopped because the trace buffer filled up.
33262
33263 @item tdisconnected:0
33264 The trace stopped because @value{GDBN} disconnected from the target.
33265
33266 @item tpasscount:@var{tpnum}
33267 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
33268
33269 @item terror:@var{text}:@var{tpnum}
33270 The trace stopped because tracepoint @var{tpnum} had an error. The
33271 string @var{text} is available to describe the nature of the error
33272 (for instance, a divide by zero in the condition expression).
33273 @var{text} is hex encoded.
33274
33275 @item tunknown:0
33276 The trace stopped for some other reason.
33277
33278 @end table
33279
33280 Additional optional fields supply statistical and other information.
33281 Although not required, they are extremely useful for users monitoring
33282 the progress of a trace run. If a trace has stopped, and these
33283 numbers are reported, they must reflect the state of the just-stopped
33284 trace.
33285
33286 @table @samp
33287
33288 @item tframes:@var{n}
33289 The number of trace frames in the buffer.
33290
33291 @item tcreated:@var{n}
33292 The total number of trace frames created during the run. This may
33293 be larger than the trace frame count, if the buffer is circular.
33294
33295 @item tsize:@var{n}
33296 The total size of the trace buffer, in bytes.
33297
33298 @item tfree:@var{n}
33299 The number of bytes still unused in the buffer.
33300
33301 @item circular:@var{n}
33302 The value of the circular trace buffer flag. @code{1} means that the
33303 trace buffer is circular and old trace frames will be discarded if
33304 necessary to make room, @code{0} means that the trace buffer is linear
33305 and may fill up.
33306
33307 @item disconn:@var{n}
33308 The value of the disconnected tracing flag. @code{1} means that
33309 tracing will continue after @value{GDBN} disconnects, @code{0} means
33310 that the trace run will stop.
33311
33312 @end table
33313
33314 @item qTV:@var{var}
33315 @cindex trace state variable value, remote request
33316 @cindex @samp{qTV} packet
33317 Ask the stub for the value of the trace state variable number @var{var}.
33318
33319 Replies:
33320 @table @samp
33321 @item V@var{value}
33322 The value of the variable is @var{value}. This will be the current
33323 value of the variable if the user is examining a running target, or a
33324 saved value if the variable was collected in the trace frame that the
33325 user is looking at. Note that multiple requests may result in
33326 different reply values, such as when requesting values while the
33327 program is running.
33328
33329 @item U
33330 The value of the variable is unknown. This would occur, for example,
33331 if the user is examining a trace frame in which the requested variable
33332 was not collected.
33333 @end table
33334
33335 @item qTfP
33336 @itemx qTsP
33337 These packets request data about tracepoints that are being used by
33338 the target. @value{GDBN} sends @code{qTfP} to get the first piece
33339 of data, and multiple @code{qTsP} to get additional pieces. Replies
33340 to these packets generally take the form of the @code{QTDP} packets
33341 that define tracepoints. (FIXME add detailed syntax)
33342
33343 @item qTfV
33344 @itemx qTsV
33345 These packets request data about trace state variables that are on the
33346 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
33347 and multiple @code{qTsV} to get additional variables. Replies to
33348 these packets follow the syntax of the @code{QTDV} packets that define
33349 trace state variables.
33350
33351 @item qTfSTM
33352 @itemx qTsSTM
33353 These packets request data about static tracepoint markers that exist
33354 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
33355 first piece of data, and multiple @code{qTsSTM} to get additional
33356 pieces. Replies to these packets take the following form:
33357
33358 Reply:
33359 @table @samp
33360 @item m @var{address}:@var{id}:@var{extra}
33361 A single marker
33362 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
33363 a comma-separated list of markers
33364 @item l
33365 (lower case letter @samp{L}) denotes end of list.
33366 @item E @var{nn}
33367 An error occurred. @var{nn} are hex digits.
33368 @item
33369 An empty reply indicates that the request is not supported by the
33370 stub.
33371 @end table
33372
33373 @var{address} is encoded in hex.
33374 @var{id} and @var{extra} are strings encoded in hex.
33375
33376 In response to each query, the target will reply with a list of one or
33377 more markers, separated by commas. @value{GDBN} will respond to each
33378 reply with a request for more markers (using the @samp{qs} form of the
33379 query), until the target responds with @samp{l} (lower-case ell, for
33380 @dfn{last}).
33381
33382 @item qTSTMat:@var{address}
33383 This packets requests data about static tracepoint markers in the
33384 target program at @var{address}. Replies to this packet follow the
33385 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
33386 tracepoint markers.
33387
33388 @item QTSave:@var{filename}
33389 This packet directs the target to save trace data to the file name
33390 @var{filename} in the target's filesystem. @var{filename} is encoded
33391 as a hex string; the interpretation of the file name (relative vs
33392 absolute, wild cards, etc) is up to the target.
33393
33394 @item qTBuffer:@var{offset},@var{len}
33395 Return up to @var{len} bytes of the current contents of trace buffer,
33396 starting at @var{offset}. The trace buffer is treated as if it were
33397 a contiguous collection of traceframes, as per the trace file format.
33398 The reply consists as many hex-encoded bytes as the target can deliver
33399 in a packet; it is not an error to return fewer than were asked for.
33400 A reply consisting of just @code{l} indicates that no bytes are
33401 available.
33402
33403 @item QTBuffer:circular:@var{value}
33404 This packet directs the target to use a circular trace buffer if
33405 @var{value} is 1, or a linear buffer if the value is 0.
33406
33407 @end table
33408
33409 @subsection Relocate instruction reply packet
33410 When installing fast tracepoints in memory, the target may need to
33411 relocate the instruction currently at the tracepoint address to a
33412 different address in memory. For most instructions, a simple copy is
33413 enough, but, for example, call instructions that implicitly push the
33414 return address on the stack, and relative branches or other
33415 PC-relative instructions require offset adjustment, so that the effect
33416 of executing the instruction at a different address is the same as if
33417 it had executed in the original location.
33418
33419 In response to several of the tracepoint packets, the target may also
33420 respond with a number of intermediate @samp{qRelocInsn} request
33421 packets before the final result packet, to have @value{GDBN} handle
33422 this relocation operation. If a packet supports this mechanism, its
33423 documentation will explicitly say so. See for example the above
33424 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
33425 format of the request is:
33426
33427 @table @samp
33428 @item qRelocInsn:@var{from};@var{to}
33429
33430 This requests @value{GDBN} to copy instruction at address @var{from}
33431 to address @var{to}, possibly adjusted so that executing the
33432 instruction at @var{to} has the same effect as executing it at
33433 @var{from}. @value{GDBN} writes the adjusted instruction to target
33434 memory starting at @var{to}.
33435 @end table
33436
33437 Replies:
33438 @table @samp
33439 @item qRelocInsn:@var{adjusted_size}
33440 Informs the stub the relocation is complete. @var{adjusted_size} is
33441 the length in bytes of resulting relocated instruction sequence.
33442 @item E @var{NN}
33443 A badly formed request was detected, or an error was encountered while
33444 relocating the instruction.
33445 @end table
33446
33447 @node Host I/O Packets
33448 @section Host I/O Packets
33449 @cindex Host I/O, remote protocol
33450 @cindex file transfer, remote protocol
33451
33452 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
33453 operations on the far side of a remote link. For example, Host I/O is
33454 used to upload and download files to a remote target with its own
33455 filesystem. Host I/O uses the same constant values and data structure
33456 layout as the target-initiated File-I/O protocol. However, the
33457 Host I/O packets are structured differently. The target-initiated
33458 protocol relies on target memory to store parameters and buffers.
33459 Host I/O requests are initiated by @value{GDBN}, and the
33460 target's memory is not involved. @xref{File-I/O Remote Protocol
33461 Extension}, for more details on the target-initiated protocol.
33462
33463 The Host I/O request packets all encode a single operation along with
33464 its arguments. They have this format:
33465
33466 @table @samp
33467
33468 @item vFile:@var{operation}: @var{parameter}@dots{}
33469 @var{operation} is the name of the particular request; the target
33470 should compare the entire packet name up to the second colon when checking
33471 for a supported operation. The format of @var{parameter} depends on
33472 the operation. Numbers are always passed in hexadecimal. Negative
33473 numbers have an explicit minus sign (i.e.@: two's complement is not
33474 used). Strings (e.g.@: filenames) are encoded as a series of
33475 hexadecimal bytes. The last argument to a system call may be a
33476 buffer of escaped binary data (@pxref{Binary Data}).
33477
33478 @end table
33479
33480 The valid responses to Host I/O packets are:
33481
33482 @table @samp
33483
33484 @item F @var{result} [, @var{errno}] [; @var{attachment}]
33485 @var{result} is the integer value returned by this operation, usually
33486 non-negative for success and -1 for errors. If an error has occured,
33487 @var{errno} will be included in the result. @var{errno} will have a
33488 value defined by the File-I/O protocol (@pxref{Errno Values}). For
33489 operations which return data, @var{attachment} supplies the data as a
33490 binary buffer. Binary buffers in response packets are escaped in the
33491 normal way (@pxref{Binary Data}). See the individual packet
33492 documentation for the interpretation of @var{result} and
33493 @var{attachment}.
33494
33495 @item
33496 An empty response indicates that this operation is not recognized.
33497
33498 @end table
33499
33500 These are the supported Host I/O operations:
33501
33502 @table @samp
33503 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
33504 Open a file at @var{pathname} and return a file descriptor for it, or
33505 return -1 if an error occurs. @var{pathname} is a string,
33506 @var{flags} is an integer indicating a mask of open flags
33507 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
33508 of mode bits to use if the file is created (@pxref{mode_t Values}).
33509 @xref{open}, for details of the open flags and mode values.
33510
33511 @item vFile:close: @var{fd}
33512 Close the open file corresponding to @var{fd} and return 0, or
33513 -1 if an error occurs.
33514
33515 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
33516 Read data from the open file corresponding to @var{fd}. Up to
33517 @var{count} bytes will be read from the file, starting at @var{offset}
33518 relative to the start of the file. The target may read fewer bytes;
33519 common reasons include packet size limits and an end-of-file
33520 condition. The number of bytes read is returned. Zero should only be
33521 returned for a successful read at the end of the file, or if
33522 @var{count} was zero.
33523
33524 The data read should be returned as a binary attachment on success.
33525 If zero bytes were read, the response should include an empty binary
33526 attachment (i.e.@: a trailing semicolon). The return value is the
33527 number of target bytes read; the binary attachment may be longer if
33528 some characters were escaped.
33529
33530 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
33531 Write @var{data} (a binary buffer) to the open file corresponding
33532 to @var{fd}. Start the write at @var{offset} from the start of the
33533 file. Unlike many @code{write} system calls, there is no
33534 separate @var{count} argument; the length of @var{data} in the
33535 packet is used. @samp{vFile:write} returns the number of bytes written,
33536 which may be shorter than the length of @var{data}, or -1 if an
33537 error occurred.
33538
33539 @item vFile:unlink: @var{pathname}
33540 Delete the file at @var{pathname} on the target. Return 0,
33541 or -1 if an error occurs. @var{pathname} is a string.
33542
33543 @end table
33544
33545 @node Interrupts
33546 @section Interrupts
33547 @cindex interrupts (remote protocol)
33548
33549 When a program on the remote target is running, @value{GDBN} may
33550 attempt to interrupt it by sending a @samp{Ctrl-C}, @code{BREAK} or
33551 a @code{BREAK} followed by @code{g},
33552 control of which is specified via @value{GDBN}'s @samp{interrupt-sequence}.
33553
33554 The precise meaning of @code{BREAK} is defined by the transport
33555 mechanism and may, in fact, be undefined. @value{GDBN} does not
33556 currently define a @code{BREAK} mechanism for any of the network
33557 interfaces except for TCP, in which case @value{GDBN} sends the
33558 @code{telnet} BREAK sequence.
33559
33560 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
33561 transport mechanisms. It is represented by sending the single byte
33562 @code{0x03} without any of the usual packet overhead described in
33563 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
33564 transmitted as part of a packet, it is considered to be packet data
33565 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
33566 (@pxref{X packet}), used for binary downloads, may include an unescaped
33567 @code{0x03} as part of its packet.
33568
33569 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
33570 When Linux kernel receives this sequence from serial port,
33571 it stops execution and connects to gdb.
33572
33573 Stubs are not required to recognize these interrupt mechanisms and the
33574 precise meaning associated with receipt of the interrupt is
33575 implementation defined. If the target supports debugging of multiple
33576 threads and/or processes, it should attempt to interrupt all
33577 currently-executing threads and processes.
33578 If the stub is successful at interrupting the
33579 running program, it should send one of the stop
33580 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
33581 of successfully stopping the program in all-stop mode, and a stop reply
33582 for each stopped thread in non-stop mode.
33583 Interrupts received while the
33584 program is stopped are discarded.
33585
33586 @node Notification Packets
33587 @section Notification Packets
33588 @cindex notification packets
33589 @cindex packets, notification
33590
33591 The @value{GDBN} remote serial protocol includes @dfn{notifications},
33592 packets that require no acknowledgment. Both the GDB and the stub
33593 may send notifications (although the only notifications defined at
33594 present are sent by the stub). Notifications carry information
33595 without incurring the round-trip latency of an acknowledgment, and so
33596 are useful for low-impact communications where occasional packet loss
33597 is not a problem.
33598
33599 A notification packet has the form @samp{% @var{data} #
33600 @var{checksum}}, where @var{data} is the content of the notification,
33601 and @var{checksum} is a checksum of @var{data}, computed and formatted
33602 as for ordinary @value{GDBN} packets. A notification's @var{data}
33603 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
33604 receiving a notification, the recipient sends no @samp{+} or @samp{-}
33605 to acknowledge the notification's receipt or to report its corruption.
33606
33607 Every notification's @var{data} begins with a name, which contains no
33608 colon characters, followed by a colon character.
33609
33610 Recipients should silently ignore corrupted notifications and
33611 notifications they do not understand. Recipients should restart
33612 timeout periods on receipt of a well-formed notification, whether or
33613 not they understand it.
33614
33615 Senders should only send the notifications described here when this
33616 protocol description specifies that they are permitted. In the
33617 future, we may extend the protocol to permit existing notifications in
33618 new contexts; this rule helps older senders avoid confusing newer
33619 recipients.
33620
33621 (Older versions of @value{GDBN} ignore bytes received until they see
33622 the @samp{$} byte that begins an ordinary packet, so new stubs may
33623 transmit notifications without fear of confusing older clients. There
33624 are no notifications defined for @value{GDBN} to send at the moment, but we
33625 assume that most older stubs would ignore them, as well.)
33626
33627 The following notification packets from the stub to @value{GDBN} are
33628 defined:
33629
33630 @table @samp
33631 @item Stop: @var{reply}
33632 Report an asynchronous stop event in non-stop mode.
33633 The @var{reply} has the form of a stop reply, as
33634 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
33635 for information on how these notifications are acknowledged by
33636 @value{GDBN}.
33637 @end table
33638
33639 @node Remote Non-Stop
33640 @section Remote Protocol Support for Non-Stop Mode
33641
33642 @value{GDBN}'s remote protocol supports non-stop debugging of
33643 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
33644 supports non-stop mode, it should report that to @value{GDBN} by including
33645 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
33646
33647 @value{GDBN} typically sends a @samp{QNonStop} packet only when
33648 establishing a new connection with the stub. Entering non-stop mode
33649 does not alter the state of any currently-running threads, but targets
33650 must stop all threads in any already-attached processes when entering
33651 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
33652 probe the target state after a mode change.
33653
33654 In non-stop mode, when an attached process encounters an event that
33655 would otherwise be reported with a stop reply, it uses the
33656 asynchronous notification mechanism (@pxref{Notification Packets}) to
33657 inform @value{GDBN}. In contrast to all-stop mode, where all threads
33658 in all processes are stopped when a stop reply is sent, in non-stop
33659 mode only the thread reporting the stop event is stopped. That is,
33660 when reporting a @samp{S} or @samp{T} response to indicate completion
33661 of a step operation, hitting a breakpoint, or a fault, only the
33662 affected thread is stopped; any other still-running threads continue
33663 to run. When reporting a @samp{W} or @samp{X} response, all running
33664 threads belonging to other attached processes continue to run.
33665
33666 Only one stop reply notification at a time may be pending; if
33667 additional stop events occur before @value{GDBN} has acknowledged the
33668 previous notification, they must be queued by the stub for later
33669 synchronous transmission in response to @samp{vStopped} packets from
33670 @value{GDBN}. Because the notification mechanism is unreliable,
33671 the stub is permitted to resend a stop reply notification
33672 if it believes @value{GDBN} may not have received it. @value{GDBN}
33673 ignores additional stop reply notifications received before it has
33674 finished processing a previous notification and the stub has completed
33675 sending any queued stop events.
33676
33677 Otherwise, @value{GDBN} must be prepared to receive a stop reply
33678 notification at any time. Specifically, they may appear when
33679 @value{GDBN} is not otherwise reading input from the stub, or when
33680 @value{GDBN} is expecting to read a normal synchronous response or a
33681 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
33682 Notification packets are distinct from any other communication from
33683 the stub so there is no ambiguity.
33684
33685 After receiving a stop reply notification, @value{GDBN} shall
33686 acknowledge it by sending a @samp{vStopped} packet (@pxref{vStopped packet})
33687 as a regular, synchronous request to the stub. Such acknowledgment
33688 is not required to happen immediately, as @value{GDBN} is permitted to
33689 send other, unrelated packets to the stub first, which the stub should
33690 process normally.
33691
33692 Upon receiving a @samp{vStopped} packet, if the stub has other queued
33693 stop events to report to @value{GDBN}, it shall respond by sending a
33694 normal stop reply response. @value{GDBN} shall then send another
33695 @samp{vStopped} packet to solicit further responses; again, it is
33696 permitted to send other, unrelated packets as well which the stub
33697 should process normally.
33698
33699 If the stub receives a @samp{vStopped} packet and there are no
33700 additional stop events to report, the stub shall return an @samp{OK}
33701 response. At this point, if further stop events occur, the stub shall
33702 send a new stop reply notification, @value{GDBN} shall accept the
33703 notification, and the process shall be repeated.
33704
33705 In non-stop mode, the target shall respond to the @samp{?} packet as
33706 follows. First, any incomplete stop reply notification/@samp{vStopped}
33707 sequence in progress is abandoned. The target must begin a new
33708 sequence reporting stop events for all stopped threads, whether or not
33709 it has previously reported those events to @value{GDBN}. The first
33710 stop reply is sent as a synchronous reply to the @samp{?} packet, and
33711 subsequent stop replies are sent as responses to @samp{vStopped} packets
33712 using the mechanism described above. The target must not send
33713 asynchronous stop reply notifications until the sequence is complete.
33714 If all threads are running when the target receives the @samp{?} packet,
33715 or if the target is not attached to any process, it shall respond
33716 @samp{OK}.
33717
33718 @node Packet Acknowledgment
33719 @section Packet Acknowledgment
33720
33721 @cindex acknowledgment, for @value{GDBN} remote
33722 @cindex packet acknowledgment, for @value{GDBN} remote
33723 By default, when either the host or the target machine receives a packet,
33724 the first response expected is an acknowledgment: either @samp{+} (to indicate
33725 the package was received correctly) or @samp{-} (to request retransmission).
33726 This mechanism allows the @value{GDBN} remote protocol to operate over
33727 unreliable transport mechanisms, such as a serial line.
33728
33729 In cases where the transport mechanism is itself reliable (such as a pipe or
33730 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
33731 It may be desirable to disable them in that case to reduce communication
33732 overhead, or for other reasons. This can be accomplished by means of the
33733 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
33734
33735 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
33736 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
33737 and response format still includes the normal checksum, as described in
33738 @ref{Overview}, but the checksum may be ignored by the receiver.
33739
33740 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
33741 no-acknowledgment mode, it should report that to @value{GDBN}
33742 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
33743 @pxref{qSupported}.
33744 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
33745 disabled via the @code{set remote noack-packet off} command
33746 (@pxref{Remote Configuration}),
33747 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
33748 Only then may the stub actually turn off packet acknowledgments.
33749 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
33750 response, which can be safely ignored by the stub.
33751
33752 Note that @code{set remote noack-packet} command only affects negotiation
33753 between @value{GDBN} and the stub when subsequent connections are made;
33754 it does not affect the protocol acknowledgment state for any current
33755 connection.
33756 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
33757 new connection is established,
33758 there is also no protocol request to re-enable the acknowledgments
33759 for the current connection, once disabled.
33760
33761 @node Examples
33762 @section Examples
33763
33764 Example sequence of a target being re-started. Notice how the restart
33765 does not get any direct output:
33766
33767 @smallexample
33768 -> @code{R00}
33769 <- @code{+}
33770 @emph{target restarts}
33771 -> @code{?}
33772 <- @code{+}
33773 <- @code{T001:1234123412341234}
33774 -> @code{+}
33775 @end smallexample
33776
33777 Example sequence of a target being stepped by a single instruction:
33778
33779 @smallexample
33780 -> @code{G1445@dots{}}
33781 <- @code{+}
33782 -> @code{s}
33783 <- @code{+}
33784 @emph{time passes}
33785 <- @code{T001:1234123412341234}
33786 -> @code{+}
33787 -> @code{g}
33788 <- @code{+}
33789 <- @code{1455@dots{}}
33790 -> @code{+}
33791 @end smallexample
33792
33793 @node File-I/O Remote Protocol Extension
33794 @section File-I/O Remote Protocol Extension
33795 @cindex File-I/O remote protocol extension
33796
33797 @menu
33798 * File-I/O Overview::
33799 * Protocol Basics::
33800 * The F Request Packet::
33801 * The F Reply Packet::
33802 * The Ctrl-C Message::
33803 * Console I/O::
33804 * List of Supported Calls::
33805 * Protocol-specific Representation of Datatypes::
33806 * Constants::
33807 * File-I/O Examples::
33808 @end menu
33809
33810 @node File-I/O Overview
33811 @subsection File-I/O Overview
33812 @cindex file-i/o overview
33813
33814 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
33815 target to use the host's file system and console I/O to perform various
33816 system calls. System calls on the target system are translated into a
33817 remote protocol packet to the host system, which then performs the needed
33818 actions and returns a response packet to the target system.
33819 This simulates file system operations even on targets that lack file systems.
33820
33821 The protocol is defined to be independent of both the host and target systems.
33822 It uses its own internal representation of datatypes and values. Both
33823 @value{GDBN} and the target's @value{GDBN} stub are responsible for
33824 translating the system-dependent value representations into the internal
33825 protocol representations when data is transmitted.
33826
33827 The communication is synchronous. A system call is possible only when
33828 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
33829 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
33830 the target is stopped to allow deterministic access to the target's
33831 memory. Therefore File-I/O is not interruptible by target signals. On
33832 the other hand, it is possible to interrupt File-I/O by a user interrupt
33833 (@samp{Ctrl-C}) within @value{GDBN}.
33834
33835 The target's request to perform a host system call does not finish
33836 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
33837 after finishing the system call, the target returns to continuing the
33838 previous activity (continue, step). No additional continue or step
33839 request from @value{GDBN} is required.
33840
33841 @smallexample
33842 (@value{GDBP}) continue
33843 <- target requests 'system call X'
33844 target is stopped, @value{GDBN} executes system call
33845 -> @value{GDBN} returns result
33846 ... target continues, @value{GDBN} returns to wait for the target
33847 <- target hits breakpoint and sends a Txx packet
33848 @end smallexample
33849
33850 The protocol only supports I/O on the console and to regular files on
33851 the host file system. Character or block special devices, pipes,
33852 named pipes, sockets or any other communication method on the host
33853 system are not supported by this protocol.
33854
33855 File I/O is not supported in non-stop mode.
33856
33857 @node Protocol Basics
33858 @subsection Protocol Basics
33859 @cindex protocol basics, file-i/o
33860
33861 The File-I/O protocol uses the @code{F} packet as the request as well
33862 as reply packet. Since a File-I/O system call can only occur when
33863 @value{GDBN} is waiting for a response from the continuing or stepping target,
33864 the File-I/O request is a reply that @value{GDBN} has to expect as a result
33865 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
33866 This @code{F} packet contains all information needed to allow @value{GDBN}
33867 to call the appropriate host system call:
33868
33869 @itemize @bullet
33870 @item
33871 A unique identifier for the requested system call.
33872
33873 @item
33874 All parameters to the system call. Pointers are given as addresses
33875 in the target memory address space. Pointers to strings are given as
33876 pointer/length pair. Numerical values are given as they are.
33877 Numerical control flags are given in a protocol-specific representation.
33878
33879 @end itemize
33880
33881 At this point, @value{GDBN} has to perform the following actions.
33882
33883 @itemize @bullet
33884 @item
33885 If the parameters include pointer values to data needed as input to a
33886 system call, @value{GDBN} requests this data from the target with a
33887 standard @code{m} packet request. This additional communication has to be
33888 expected by the target implementation and is handled as any other @code{m}
33889 packet.
33890
33891 @item
33892 @value{GDBN} translates all value from protocol representation to host
33893 representation as needed. Datatypes are coerced into the host types.
33894
33895 @item
33896 @value{GDBN} calls the system call.
33897
33898 @item
33899 It then coerces datatypes back to protocol representation.
33900
33901 @item
33902 If the system call is expected to return data in buffer space specified
33903 by pointer parameters to the call, the data is transmitted to the
33904 target using a @code{M} or @code{X} packet. This packet has to be expected
33905 by the target implementation and is handled as any other @code{M} or @code{X}
33906 packet.
33907
33908 @end itemize
33909
33910 Eventually @value{GDBN} replies with another @code{F} packet which contains all
33911 necessary information for the target to continue. This at least contains
33912
33913 @itemize @bullet
33914 @item
33915 Return value.
33916
33917 @item
33918 @code{errno}, if has been changed by the system call.
33919
33920 @item
33921 ``Ctrl-C'' flag.
33922
33923 @end itemize
33924
33925 After having done the needed type and value coercion, the target continues
33926 the latest continue or step action.
33927
33928 @node The F Request Packet
33929 @subsection The @code{F} Request Packet
33930 @cindex file-i/o request packet
33931 @cindex @code{F} request packet
33932
33933 The @code{F} request packet has the following format:
33934
33935 @table @samp
33936 @item F@var{call-id},@var{parameter@dots{}}
33937
33938 @var{call-id} is the identifier to indicate the host system call to be called.
33939 This is just the name of the function.
33940
33941 @var{parameter@dots{}} are the parameters to the system call.
33942 Parameters are hexadecimal integer values, either the actual values in case
33943 of scalar datatypes, pointers to target buffer space in case of compound
33944 datatypes and unspecified memory areas, or pointer/length pairs in case
33945 of string parameters. These are appended to the @var{call-id} as a
33946 comma-delimited list. All values are transmitted in ASCII
33947 string representation, pointer/length pairs separated by a slash.
33948
33949 @end table
33950
33951
33952
33953 @node The F Reply Packet
33954 @subsection The @code{F} Reply Packet
33955 @cindex file-i/o reply packet
33956 @cindex @code{F} reply packet
33957
33958 The @code{F} reply packet has the following format:
33959
33960 @table @samp
33961
33962 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
33963
33964 @var{retcode} is the return code of the system call as hexadecimal value.
33965
33966 @var{errno} is the @code{errno} set by the call, in protocol-specific
33967 representation.
33968 This parameter can be omitted if the call was successful.
33969
33970 @var{Ctrl-C flag} is only sent if the user requested a break. In this
33971 case, @var{errno} must be sent as well, even if the call was successful.
33972 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
33973
33974 @smallexample
33975 F0,0,C
33976 @end smallexample
33977
33978 @noindent
33979 or, if the call was interrupted before the host call has been performed:
33980
33981 @smallexample
33982 F-1,4,C
33983 @end smallexample
33984
33985 @noindent
33986 assuming 4 is the protocol-specific representation of @code{EINTR}.
33987
33988 @end table
33989
33990
33991 @node The Ctrl-C Message
33992 @subsection The @samp{Ctrl-C} Message
33993 @cindex ctrl-c message, in file-i/o protocol
33994
33995 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
33996 reply packet (@pxref{The F Reply Packet}),
33997 the target should behave as if it had
33998 gotten a break message. The meaning for the target is ``system call
33999 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
34000 (as with a break message) and return to @value{GDBN} with a @code{T02}
34001 packet.
34002
34003 It's important for the target to know in which
34004 state the system call was interrupted. There are two possible cases:
34005
34006 @itemize @bullet
34007 @item
34008 The system call hasn't been performed on the host yet.
34009
34010 @item
34011 The system call on the host has been finished.
34012
34013 @end itemize
34014
34015 These two states can be distinguished by the target by the value of the
34016 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
34017 call hasn't been performed. This is equivalent to the @code{EINTR} handling
34018 on POSIX systems. In any other case, the target may presume that the
34019 system call has been finished --- successfully or not --- and should behave
34020 as if the break message arrived right after the system call.
34021
34022 @value{GDBN} must behave reliably. If the system call has not been called
34023 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
34024 @code{errno} in the packet. If the system call on the host has been finished
34025 before the user requests a break, the full action must be finished by
34026 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
34027 The @code{F} packet may only be sent when either nothing has happened
34028 or the full action has been completed.
34029
34030 @node Console I/O
34031 @subsection Console I/O
34032 @cindex console i/o as part of file-i/o
34033
34034 By default and if not explicitly closed by the target system, the file
34035 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
34036 on the @value{GDBN} console is handled as any other file output operation
34037 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
34038 by @value{GDBN} so that after the target read request from file descriptor
34039 0 all following typing is buffered until either one of the following
34040 conditions is met:
34041
34042 @itemize @bullet
34043 @item
34044 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
34045 @code{read}
34046 system call is treated as finished.
34047
34048 @item
34049 The user presses @key{RET}. This is treated as end of input with a trailing
34050 newline.
34051
34052 @item
34053 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
34054 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
34055
34056 @end itemize
34057
34058 If the user has typed more characters than fit in the buffer given to
34059 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
34060 either another @code{read(0, @dots{})} is requested by the target, or debugging
34061 is stopped at the user's request.
34062
34063
34064 @node List of Supported Calls
34065 @subsection List of Supported Calls
34066 @cindex list of supported file-i/o calls
34067
34068 @menu
34069 * open::
34070 * close::
34071 * read::
34072 * write::
34073 * lseek::
34074 * rename::
34075 * unlink::
34076 * stat/fstat::
34077 * gettimeofday::
34078 * isatty::
34079 * system::
34080 @end menu
34081
34082 @node open
34083 @unnumberedsubsubsec open
34084 @cindex open, file-i/o system call
34085
34086 @table @asis
34087 @item Synopsis:
34088 @smallexample
34089 int open(const char *pathname, int flags);
34090 int open(const char *pathname, int flags, mode_t mode);
34091 @end smallexample
34092
34093 @item Request:
34094 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
34095
34096 @noindent
34097 @var{flags} is the bitwise @code{OR} of the following values:
34098
34099 @table @code
34100 @item O_CREAT
34101 If the file does not exist it will be created. The host
34102 rules apply as far as file ownership and time stamps
34103 are concerned.
34104
34105 @item O_EXCL
34106 When used with @code{O_CREAT}, if the file already exists it is
34107 an error and open() fails.
34108
34109 @item O_TRUNC
34110 If the file already exists and the open mode allows
34111 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
34112 truncated to zero length.
34113
34114 @item O_APPEND
34115 The file is opened in append mode.
34116
34117 @item O_RDONLY
34118 The file is opened for reading only.
34119
34120 @item O_WRONLY
34121 The file is opened for writing only.
34122
34123 @item O_RDWR
34124 The file is opened for reading and writing.
34125 @end table
34126
34127 @noindent
34128 Other bits are silently ignored.
34129
34130
34131 @noindent
34132 @var{mode} is the bitwise @code{OR} of the following values:
34133
34134 @table @code
34135 @item S_IRUSR
34136 User has read permission.
34137
34138 @item S_IWUSR
34139 User has write permission.
34140
34141 @item S_IRGRP
34142 Group has read permission.
34143
34144 @item S_IWGRP
34145 Group has write permission.
34146
34147 @item S_IROTH
34148 Others have read permission.
34149
34150 @item S_IWOTH
34151 Others have write permission.
34152 @end table
34153
34154 @noindent
34155 Other bits are silently ignored.
34156
34157
34158 @item Return value:
34159 @code{open} returns the new file descriptor or -1 if an error
34160 occurred.
34161
34162 @item Errors:
34163
34164 @table @code
34165 @item EEXIST
34166 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
34167
34168 @item EISDIR
34169 @var{pathname} refers to a directory.
34170
34171 @item EACCES
34172 The requested access is not allowed.
34173
34174 @item ENAMETOOLONG
34175 @var{pathname} was too long.
34176
34177 @item ENOENT
34178 A directory component in @var{pathname} does not exist.
34179
34180 @item ENODEV
34181 @var{pathname} refers to a device, pipe, named pipe or socket.
34182
34183 @item EROFS
34184 @var{pathname} refers to a file on a read-only filesystem and
34185 write access was requested.
34186
34187 @item EFAULT
34188 @var{pathname} is an invalid pointer value.
34189
34190 @item ENOSPC
34191 No space on device to create the file.
34192
34193 @item EMFILE
34194 The process already has the maximum number of files open.
34195
34196 @item ENFILE
34197 The limit on the total number of files open on the system
34198 has been reached.
34199
34200 @item EINTR
34201 The call was interrupted by the user.
34202 @end table
34203
34204 @end table
34205
34206 @node close
34207 @unnumberedsubsubsec close
34208 @cindex close, file-i/o system call
34209
34210 @table @asis
34211 @item Synopsis:
34212 @smallexample
34213 int close(int fd);
34214 @end smallexample
34215
34216 @item Request:
34217 @samp{Fclose,@var{fd}}
34218
34219 @item Return value:
34220 @code{close} returns zero on success, or -1 if an error occurred.
34221
34222 @item Errors:
34223
34224 @table @code
34225 @item EBADF
34226 @var{fd} isn't a valid open file descriptor.
34227
34228 @item EINTR
34229 The call was interrupted by the user.
34230 @end table
34231
34232 @end table
34233
34234 @node read
34235 @unnumberedsubsubsec read
34236 @cindex read, file-i/o system call
34237
34238 @table @asis
34239 @item Synopsis:
34240 @smallexample
34241 int read(int fd, void *buf, unsigned int count);
34242 @end smallexample
34243
34244 @item Request:
34245 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
34246
34247 @item Return value:
34248 On success, the number of bytes read is returned.
34249 Zero indicates end of file. If count is zero, read
34250 returns zero as well. On error, -1 is returned.
34251
34252 @item Errors:
34253
34254 @table @code
34255 @item EBADF
34256 @var{fd} is not a valid file descriptor or is not open for
34257 reading.
34258
34259 @item EFAULT
34260 @var{bufptr} is an invalid pointer value.
34261
34262 @item EINTR
34263 The call was interrupted by the user.
34264 @end table
34265
34266 @end table
34267
34268 @node write
34269 @unnumberedsubsubsec write
34270 @cindex write, file-i/o system call
34271
34272 @table @asis
34273 @item Synopsis:
34274 @smallexample
34275 int write(int fd, const void *buf, unsigned int count);
34276 @end smallexample
34277
34278 @item Request:
34279 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
34280
34281 @item Return value:
34282 On success, the number of bytes written are returned.
34283 Zero indicates nothing was written. On error, -1
34284 is returned.
34285
34286 @item Errors:
34287
34288 @table @code
34289 @item EBADF
34290 @var{fd} is not a valid file descriptor or is not open for
34291 writing.
34292
34293 @item EFAULT
34294 @var{bufptr} is an invalid pointer value.
34295
34296 @item EFBIG
34297 An attempt was made to write a file that exceeds the
34298 host-specific maximum file size allowed.
34299
34300 @item ENOSPC
34301 No space on device to write the data.
34302
34303 @item EINTR
34304 The call was interrupted by the user.
34305 @end table
34306
34307 @end table
34308
34309 @node lseek
34310 @unnumberedsubsubsec lseek
34311 @cindex lseek, file-i/o system call
34312
34313 @table @asis
34314 @item Synopsis:
34315 @smallexample
34316 long lseek (int fd, long offset, int flag);
34317 @end smallexample
34318
34319 @item Request:
34320 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
34321
34322 @var{flag} is one of:
34323
34324 @table @code
34325 @item SEEK_SET
34326 The offset is set to @var{offset} bytes.
34327
34328 @item SEEK_CUR
34329 The offset is set to its current location plus @var{offset}
34330 bytes.
34331
34332 @item SEEK_END
34333 The offset is set to the size of the file plus @var{offset}
34334 bytes.
34335 @end table
34336
34337 @item Return value:
34338 On success, the resulting unsigned offset in bytes from
34339 the beginning of the file is returned. Otherwise, a
34340 value of -1 is returned.
34341
34342 @item Errors:
34343
34344 @table @code
34345 @item EBADF
34346 @var{fd} is not a valid open file descriptor.
34347
34348 @item ESPIPE
34349 @var{fd} is associated with the @value{GDBN} console.
34350
34351 @item EINVAL
34352 @var{flag} is not a proper value.
34353
34354 @item EINTR
34355 The call was interrupted by the user.
34356 @end table
34357
34358 @end table
34359
34360 @node rename
34361 @unnumberedsubsubsec rename
34362 @cindex rename, file-i/o system call
34363
34364 @table @asis
34365 @item Synopsis:
34366 @smallexample
34367 int rename(const char *oldpath, const char *newpath);
34368 @end smallexample
34369
34370 @item Request:
34371 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
34372
34373 @item Return value:
34374 On success, zero is returned. On error, -1 is returned.
34375
34376 @item Errors:
34377
34378 @table @code
34379 @item EISDIR
34380 @var{newpath} is an existing directory, but @var{oldpath} is not a
34381 directory.
34382
34383 @item EEXIST
34384 @var{newpath} is a non-empty directory.
34385
34386 @item EBUSY
34387 @var{oldpath} or @var{newpath} is a directory that is in use by some
34388 process.
34389
34390 @item EINVAL
34391 An attempt was made to make a directory a subdirectory
34392 of itself.
34393
34394 @item ENOTDIR
34395 A component used as a directory in @var{oldpath} or new
34396 path is not a directory. Or @var{oldpath} is a directory
34397 and @var{newpath} exists but is not a directory.
34398
34399 @item EFAULT
34400 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
34401
34402 @item EACCES
34403 No access to the file or the path of the file.
34404
34405 @item ENAMETOOLONG
34406
34407 @var{oldpath} or @var{newpath} was too long.
34408
34409 @item ENOENT
34410 A directory component in @var{oldpath} or @var{newpath} does not exist.
34411
34412 @item EROFS
34413 The file is on a read-only filesystem.
34414
34415 @item ENOSPC
34416 The device containing the file has no room for the new
34417 directory entry.
34418
34419 @item EINTR
34420 The call was interrupted by the user.
34421 @end table
34422
34423 @end table
34424
34425 @node unlink
34426 @unnumberedsubsubsec unlink
34427 @cindex unlink, file-i/o system call
34428
34429 @table @asis
34430 @item Synopsis:
34431 @smallexample
34432 int unlink(const char *pathname);
34433 @end smallexample
34434
34435 @item Request:
34436 @samp{Funlink,@var{pathnameptr}/@var{len}}
34437
34438 @item Return value:
34439 On success, zero is returned. On error, -1 is returned.
34440
34441 @item Errors:
34442
34443 @table @code
34444 @item EACCES
34445 No access to the file or the path of the file.
34446
34447 @item EPERM
34448 The system does not allow unlinking of directories.
34449
34450 @item EBUSY
34451 The file @var{pathname} cannot be unlinked because it's
34452 being used by another process.
34453
34454 @item EFAULT
34455 @var{pathnameptr} is an invalid pointer value.
34456
34457 @item ENAMETOOLONG
34458 @var{pathname} was too long.
34459
34460 @item ENOENT
34461 A directory component in @var{pathname} does not exist.
34462
34463 @item ENOTDIR
34464 A component of the path is not a directory.
34465
34466 @item EROFS
34467 The file is on a read-only filesystem.
34468
34469 @item EINTR
34470 The call was interrupted by the user.
34471 @end table
34472
34473 @end table
34474
34475 @node stat/fstat
34476 @unnumberedsubsubsec stat/fstat
34477 @cindex fstat, file-i/o system call
34478 @cindex stat, file-i/o system call
34479
34480 @table @asis
34481 @item Synopsis:
34482 @smallexample
34483 int stat(const char *pathname, struct stat *buf);
34484 int fstat(int fd, struct stat *buf);
34485 @end smallexample
34486
34487 @item Request:
34488 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
34489 @samp{Ffstat,@var{fd},@var{bufptr}}
34490
34491 @item Return value:
34492 On success, zero is returned. On error, -1 is returned.
34493
34494 @item Errors:
34495
34496 @table @code
34497 @item EBADF
34498 @var{fd} is not a valid open file.
34499
34500 @item ENOENT
34501 A directory component in @var{pathname} does not exist or the
34502 path is an empty string.
34503
34504 @item ENOTDIR
34505 A component of the path is not a directory.
34506
34507 @item EFAULT
34508 @var{pathnameptr} is an invalid pointer value.
34509
34510 @item EACCES
34511 No access to the file or the path of the file.
34512
34513 @item ENAMETOOLONG
34514 @var{pathname} was too long.
34515
34516 @item EINTR
34517 The call was interrupted by the user.
34518 @end table
34519
34520 @end table
34521
34522 @node gettimeofday
34523 @unnumberedsubsubsec gettimeofday
34524 @cindex gettimeofday, file-i/o system call
34525
34526 @table @asis
34527 @item Synopsis:
34528 @smallexample
34529 int gettimeofday(struct timeval *tv, void *tz);
34530 @end smallexample
34531
34532 @item Request:
34533 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
34534
34535 @item Return value:
34536 On success, 0 is returned, -1 otherwise.
34537
34538 @item Errors:
34539
34540 @table @code
34541 @item EINVAL
34542 @var{tz} is a non-NULL pointer.
34543
34544 @item EFAULT
34545 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
34546 @end table
34547
34548 @end table
34549
34550 @node isatty
34551 @unnumberedsubsubsec isatty
34552 @cindex isatty, file-i/o system call
34553
34554 @table @asis
34555 @item Synopsis:
34556 @smallexample
34557 int isatty(int fd);
34558 @end smallexample
34559
34560 @item Request:
34561 @samp{Fisatty,@var{fd}}
34562
34563 @item Return value:
34564 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
34565
34566 @item Errors:
34567
34568 @table @code
34569 @item EINTR
34570 The call was interrupted by the user.
34571 @end table
34572
34573 @end table
34574
34575 Note that the @code{isatty} call is treated as a special case: it returns
34576 1 to the target if the file descriptor is attached
34577 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
34578 would require implementing @code{ioctl} and would be more complex than
34579 needed.
34580
34581
34582 @node system
34583 @unnumberedsubsubsec system
34584 @cindex system, file-i/o system call
34585
34586 @table @asis
34587 @item Synopsis:
34588 @smallexample
34589 int system(const char *command);
34590 @end smallexample
34591
34592 @item Request:
34593 @samp{Fsystem,@var{commandptr}/@var{len}}
34594
34595 @item Return value:
34596 If @var{len} is zero, the return value indicates whether a shell is
34597 available. A zero return value indicates a shell is not available.
34598 For non-zero @var{len}, the value returned is -1 on error and the
34599 return status of the command otherwise. Only the exit status of the
34600 command is returned, which is extracted from the host's @code{system}
34601 return value by calling @code{WEXITSTATUS(retval)}. In case
34602 @file{/bin/sh} could not be executed, 127 is returned.
34603
34604 @item Errors:
34605
34606 @table @code
34607 @item EINTR
34608 The call was interrupted by the user.
34609 @end table
34610
34611 @end table
34612
34613 @value{GDBN} takes over the full task of calling the necessary host calls
34614 to perform the @code{system} call. The return value of @code{system} on
34615 the host is simplified before it's returned
34616 to the target. Any termination signal information from the child process
34617 is discarded, and the return value consists
34618 entirely of the exit status of the called command.
34619
34620 Due to security concerns, the @code{system} call is by default refused
34621 by @value{GDBN}. The user has to allow this call explicitly with the
34622 @code{set remote system-call-allowed 1} command.
34623
34624 @table @code
34625 @item set remote system-call-allowed
34626 @kindex set remote system-call-allowed
34627 Control whether to allow the @code{system} calls in the File I/O
34628 protocol for the remote target. The default is zero (disabled).
34629
34630 @item show remote system-call-allowed
34631 @kindex show remote system-call-allowed
34632 Show whether the @code{system} calls are allowed in the File I/O
34633 protocol.
34634 @end table
34635
34636 @node Protocol-specific Representation of Datatypes
34637 @subsection Protocol-specific Representation of Datatypes
34638 @cindex protocol-specific representation of datatypes, in file-i/o protocol
34639
34640 @menu
34641 * Integral Datatypes::
34642 * Pointer Values::
34643 * Memory Transfer::
34644 * struct stat::
34645 * struct timeval::
34646 @end menu
34647
34648 @node Integral Datatypes
34649 @unnumberedsubsubsec Integral Datatypes
34650 @cindex integral datatypes, in file-i/o protocol
34651
34652 The integral datatypes used in the system calls are @code{int},
34653 @code{unsigned int}, @code{long}, @code{unsigned long},
34654 @code{mode_t}, and @code{time_t}.
34655
34656 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
34657 implemented as 32 bit values in this protocol.
34658
34659 @code{long} and @code{unsigned long} are implemented as 64 bit types.
34660
34661 @xref{Limits}, for corresponding MIN and MAX values (similar to those
34662 in @file{limits.h}) to allow range checking on host and target.
34663
34664 @code{time_t} datatypes are defined as seconds since the Epoch.
34665
34666 All integral datatypes transferred as part of a memory read or write of a
34667 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
34668 byte order.
34669
34670 @node Pointer Values
34671 @unnumberedsubsubsec Pointer Values
34672 @cindex pointer values, in file-i/o protocol
34673
34674 Pointers to target data are transmitted as they are. An exception
34675 is made for pointers to buffers for which the length isn't
34676 transmitted as part of the function call, namely strings. Strings
34677 are transmitted as a pointer/length pair, both as hex values, e.g.@:
34678
34679 @smallexample
34680 @code{1aaf/12}
34681 @end smallexample
34682
34683 @noindent
34684 which is a pointer to data of length 18 bytes at position 0x1aaf.
34685 The length is defined as the full string length in bytes, including
34686 the trailing null byte. For example, the string @code{"hello world"}
34687 at address 0x123456 is transmitted as
34688
34689 @smallexample
34690 @code{123456/d}
34691 @end smallexample
34692
34693 @node Memory Transfer
34694 @unnumberedsubsubsec Memory Transfer
34695 @cindex memory transfer, in file-i/o protocol
34696
34697 Structured data which is transferred using a memory read or write (for
34698 example, a @code{struct stat}) is expected to be in a protocol-specific format
34699 with all scalar multibyte datatypes being big endian. Translation to
34700 this representation needs to be done both by the target before the @code{F}
34701 packet is sent, and by @value{GDBN} before
34702 it transfers memory to the target. Transferred pointers to structured
34703 data should point to the already-coerced data at any time.
34704
34705
34706 @node struct stat
34707 @unnumberedsubsubsec struct stat
34708 @cindex struct stat, in file-i/o protocol
34709
34710 The buffer of type @code{struct stat} used by the target and @value{GDBN}
34711 is defined as follows:
34712
34713 @smallexample
34714 struct stat @{
34715 unsigned int st_dev; /* device */
34716 unsigned int st_ino; /* inode */
34717 mode_t st_mode; /* protection */
34718 unsigned int st_nlink; /* number of hard links */
34719 unsigned int st_uid; /* user ID of owner */
34720 unsigned int st_gid; /* group ID of owner */
34721 unsigned int st_rdev; /* device type (if inode device) */
34722 unsigned long st_size; /* total size, in bytes */
34723 unsigned long st_blksize; /* blocksize for filesystem I/O */
34724 unsigned long st_blocks; /* number of blocks allocated */
34725 time_t st_atime; /* time of last access */
34726 time_t st_mtime; /* time of last modification */
34727 time_t st_ctime; /* time of last change */
34728 @};
34729 @end smallexample
34730
34731 The integral datatypes conform to the definitions given in the
34732 appropriate section (see @ref{Integral Datatypes}, for details) so this
34733 structure is of size 64 bytes.
34734
34735 The values of several fields have a restricted meaning and/or
34736 range of values.
34737
34738 @table @code
34739
34740 @item st_dev
34741 A value of 0 represents a file, 1 the console.
34742
34743 @item st_ino
34744 No valid meaning for the target. Transmitted unchanged.
34745
34746 @item st_mode
34747 Valid mode bits are described in @ref{Constants}. Any other
34748 bits have currently no meaning for the target.
34749
34750 @item st_uid
34751 @itemx st_gid
34752 @itemx st_rdev
34753 No valid meaning for the target. Transmitted unchanged.
34754
34755 @item st_atime
34756 @itemx st_mtime
34757 @itemx st_ctime
34758 These values have a host and file system dependent
34759 accuracy. Especially on Windows hosts, the file system may not
34760 support exact timing values.
34761 @end table
34762
34763 The target gets a @code{struct stat} of the above representation and is
34764 responsible for coercing it to the target representation before
34765 continuing.
34766
34767 Note that due to size differences between the host, target, and protocol
34768 representations of @code{struct stat} members, these members could eventually
34769 get truncated on the target.
34770
34771 @node struct timeval
34772 @unnumberedsubsubsec struct timeval
34773 @cindex struct timeval, in file-i/o protocol
34774
34775 The buffer of type @code{struct timeval} used by the File-I/O protocol
34776 is defined as follows:
34777
34778 @smallexample
34779 struct timeval @{
34780 time_t tv_sec; /* second */
34781 long tv_usec; /* microsecond */
34782 @};
34783 @end smallexample
34784
34785 The integral datatypes conform to the definitions given in the
34786 appropriate section (see @ref{Integral Datatypes}, for details) so this
34787 structure is of size 8 bytes.
34788
34789 @node Constants
34790 @subsection Constants
34791 @cindex constants, in file-i/o protocol
34792
34793 The following values are used for the constants inside of the
34794 protocol. @value{GDBN} and target are responsible for translating these
34795 values before and after the call as needed.
34796
34797 @menu
34798 * Open Flags::
34799 * mode_t Values::
34800 * Errno Values::
34801 * Lseek Flags::
34802 * Limits::
34803 @end menu
34804
34805 @node Open Flags
34806 @unnumberedsubsubsec Open Flags
34807 @cindex open flags, in file-i/o protocol
34808
34809 All values are given in hexadecimal representation.
34810
34811 @smallexample
34812 O_RDONLY 0x0
34813 O_WRONLY 0x1
34814 O_RDWR 0x2
34815 O_APPEND 0x8
34816 O_CREAT 0x200
34817 O_TRUNC 0x400
34818 O_EXCL 0x800
34819 @end smallexample
34820
34821 @node mode_t Values
34822 @unnumberedsubsubsec mode_t Values
34823 @cindex mode_t values, in file-i/o protocol
34824
34825 All values are given in octal representation.
34826
34827 @smallexample
34828 S_IFREG 0100000
34829 S_IFDIR 040000
34830 S_IRUSR 0400
34831 S_IWUSR 0200
34832 S_IXUSR 0100
34833 S_IRGRP 040
34834 S_IWGRP 020
34835 S_IXGRP 010
34836 S_IROTH 04
34837 S_IWOTH 02
34838 S_IXOTH 01
34839 @end smallexample
34840
34841 @node Errno Values
34842 @unnumberedsubsubsec Errno Values
34843 @cindex errno values, in file-i/o protocol
34844
34845 All values are given in decimal representation.
34846
34847 @smallexample
34848 EPERM 1
34849 ENOENT 2
34850 EINTR 4
34851 EBADF 9
34852 EACCES 13
34853 EFAULT 14
34854 EBUSY 16
34855 EEXIST 17
34856 ENODEV 19
34857 ENOTDIR 20
34858 EISDIR 21
34859 EINVAL 22
34860 ENFILE 23
34861 EMFILE 24
34862 EFBIG 27
34863 ENOSPC 28
34864 ESPIPE 29
34865 EROFS 30
34866 ENAMETOOLONG 91
34867 EUNKNOWN 9999
34868 @end smallexample
34869
34870 @code{EUNKNOWN} is used as a fallback error value if a host system returns
34871 any error value not in the list of supported error numbers.
34872
34873 @node Lseek Flags
34874 @unnumberedsubsubsec Lseek Flags
34875 @cindex lseek flags, in file-i/o protocol
34876
34877 @smallexample
34878 SEEK_SET 0
34879 SEEK_CUR 1
34880 SEEK_END 2
34881 @end smallexample
34882
34883 @node Limits
34884 @unnumberedsubsubsec Limits
34885 @cindex limits, in file-i/o protocol
34886
34887 All values are given in decimal representation.
34888
34889 @smallexample
34890 INT_MIN -2147483648
34891 INT_MAX 2147483647
34892 UINT_MAX 4294967295
34893 LONG_MIN -9223372036854775808
34894 LONG_MAX 9223372036854775807
34895 ULONG_MAX 18446744073709551615
34896 @end smallexample
34897
34898 @node File-I/O Examples
34899 @subsection File-I/O Examples
34900 @cindex file-i/o examples
34901
34902 Example sequence of a write call, file descriptor 3, buffer is at target
34903 address 0x1234, 6 bytes should be written:
34904
34905 @smallexample
34906 <- @code{Fwrite,3,1234,6}
34907 @emph{request memory read from target}
34908 -> @code{m1234,6}
34909 <- XXXXXX
34910 @emph{return "6 bytes written"}
34911 -> @code{F6}
34912 @end smallexample
34913
34914 Example sequence of a read call, file descriptor 3, buffer is at target
34915 address 0x1234, 6 bytes should be read:
34916
34917 @smallexample
34918 <- @code{Fread,3,1234,6}
34919 @emph{request memory write to target}
34920 -> @code{X1234,6:XXXXXX}
34921 @emph{return "6 bytes read"}
34922 -> @code{F6}
34923 @end smallexample
34924
34925 Example sequence of a read call, call fails on the host due to invalid
34926 file descriptor (@code{EBADF}):
34927
34928 @smallexample
34929 <- @code{Fread,3,1234,6}
34930 -> @code{F-1,9}
34931 @end smallexample
34932
34933 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
34934 host is called:
34935
34936 @smallexample
34937 <- @code{Fread,3,1234,6}
34938 -> @code{F-1,4,C}
34939 <- @code{T02}
34940 @end smallexample
34941
34942 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
34943 host is called:
34944
34945 @smallexample
34946 <- @code{Fread,3,1234,6}
34947 -> @code{X1234,6:XXXXXX}
34948 <- @code{T02}
34949 @end smallexample
34950
34951 @node Library List Format
34952 @section Library List Format
34953 @cindex library list format, remote protocol
34954
34955 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
34956 same process as your application to manage libraries. In this case,
34957 @value{GDBN} can use the loader's symbol table and normal memory
34958 operations to maintain a list of shared libraries. On other
34959 platforms, the operating system manages loaded libraries.
34960 @value{GDBN} can not retrieve the list of currently loaded libraries
34961 through memory operations, so it uses the @samp{qXfer:libraries:read}
34962 packet (@pxref{qXfer library list read}) instead. The remote stub
34963 queries the target's operating system and reports which libraries
34964 are loaded.
34965
34966 The @samp{qXfer:libraries:read} packet returns an XML document which
34967 lists loaded libraries and their offsets. Each library has an
34968 associated name and one or more segment or section base addresses,
34969 which report where the library was loaded in memory.
34970
34971 For the common case of libraries that are fully linked binaries, the
34972 library should have a list of segments. If the target supports
34973 dynamic linking of a relocatable object file, its library XML element
34974 should instead include a list of allocated sections. The segment or
34975 section bases are start addresses, not relocation offsets; they do not
34976 depend on the library's link-time base addresses.
34977
34978 @value{GDBN} must be linked with the Expat library to support XML
34979 library lists. @xref{Expat}.
34980
34981 A simple memory map, with one loaded library relocated by a single
34982 offset, looks like this:
34983
34984 @smallexample
34985 <library-list>
34986 <library name="/lib/libc.so.6">
34987 <segment address="0x10000000"/>
34988 </library>
34989 </library-list>
34990 @end smallexample
34991
34992 Another simple memory map, with one loaded library with three
34993 allocated sections (.text, .data, .bss), looks like this:
34994
34995 @smallexample
34996 <library-list>
34997 <library name="sharedlib.o">
34998 <section address="0x10000000"/>
34999 <section address="0x20000000"/>
35000 <section address="0x30000000"/>
35001 </library>
35002 </library-list>
35003 @end smallexample
35004
35005 The format of a library list is described by this DTD:
35006
35007 @smallexample
35008 <!-- library-list: Root element with versioning -->
35009 <!ELEMENT library-list (library)*>
35010 <!ATTLIST library-list version CDATA #FIXED "1.0">
35011 <!ELEMENT library (segment*, section*)>
35012 <!ATTLIST library name CDATA #REQUIRED>
35013 <!ELEMENT segment EMPTY>
35014 <!ATTLIST segment address CDATA #REQUIRED>
35015 <!ELEMENT section EMPTY>
35016 <!ATTLIST section address CDATA #REQUIRED>
35017 @end smallexample
35018
35019 In addition, segments and section descriptors cannot be mixed within a
35020 single library element, and you must supply at least one segment or
35021 section for each library.
35022
35023 @node Memory Map Format
35024 @section Memory Map Format
35025 @cindex memory map format
35026
35027 To be able to write into flash memory, @value{GDBN} needs to obtain a
35028 memory map from the target. This section describes the format of the
35029 memory map.
35030
35031 The memory map is obtained using the @samp{qXfer:memory-map:read}
35032 (@pxref{qXfer memory map read}) packet and is an XML document that
35033 lists memory regions.
35034
35035 @value{GDBN} must be linked with the Expat library to support XML
35036 memory maps. @xref{Expat}.
35037
35038 The top-level structure of the document is shown below:
35039
35040 @smallexample
35041 <?xml version="1.0"?>
35042 <!DOCTYPE memory-map
35043 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
35044 "http://sourceware.org/gdb/gdb-memory-map.dtd">
35045 <memory-map>
35046 region...
35047 </memory-map>
35048 @end smallexample
35049
35050 Each region can be either:
35051
35052 @itemize
35053
35054 @item
35055 A region of RAM starting at @var{addr} and extending for @var{length}
35056 bytes from there:
35057
35058 @smallexample
35059 <memory type="ram" start="@var{addr}" length="@var{length}"/>
35060 @end smallexample
35061
35062
35063 @item
35064 A region of read-only memory:
35065
35066 @smallexample
35067 <memory type="rom" start="@var{addr}" length="@var{length}"/>
35068 @end smallexample
35069
35070
35071 @item
35072 A region of flash memory, with erasure blocks @var{blocksize}
35073 bytes in length:
35074
35075 @smallexample
35076 <memory type="flash" start="@var{addr}" length="@var{length}">
35077 <property name="blocksize">@var{blocksize}</property>
35078 </memory>
35079 @end smallexample
35080
35081 @end itemize
35082
35083 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
35084 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
35085 packets to write to addresses in such ranges.
35086
35087 The formal DTD for memory map format is given below:
35088
35089 @smallexample
35090 <!-- ................................................... -->
35091 <!-- Memory Map XML DTD ................................ -->
35092 <!-- File: memory-map.dtd .............................. -->
35093 <!-- .................................... .............. -->
35094 <!-- memory-map.dtd -->
35095 <!-- memory-map: Root element with versioning -->
35096 <!ELEMENT memory-map (memory | property)>
35097 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
35098 <!ELEMENT memory (property)>
35099 <!-- memory: Specifies a memory region,
35100 and its type, or device. -->
35101 <!ATTLIST memory type CDATA #REQUIRED
35102 start CDATA #REQUIRED
35103 length CDATA #REQUIRED
35104 device CDATA #IMPLIED>
35105 <!-- property: Generic attribute tag -->
35106 <!ELEMENT property (#PCDATA | property)*>
35107 <!ATTLIST property name CDATA #REQUIRED>
35108 @end smallexample
35109
35110 @node Thread List Format
35111 @section Thread List Format
35112 @cindex thread list format
35113
35114 To efficiently update the list of threads and their attributes,
35115 @value{GDBN} issues the @samp{qXfer:threads:read} packet
35116 (@pxref{qXfer threads read}) and obtains the XML document with
35117 the following structure:
35118
35119 @smallexample
35120 <?xml version="1.0"?>
35121 <threads>
35122 <thread id="id" core="0">
35123 ... description ...
35124 </thread>
35125 </threads>
35126 @end smallexample
35127
35128 Each @samp{thread} element must have the @samp{id} attribute that
35129 identifies the thread (@pxref{thread-id syntax}). The
35130 @samp{core} attribute, if present, specifies which processor core
35131 the thread was last executing on. The content of the of @samp{thread}
35132 element is interpreted as human-readable auxilliary information.
35133
35134 @include agentexpr.texi
35135
35136 @node Trace File Format
35137 @appendix Trace File Format
35138 @cindex trace file format
35139
35140 The trace file comes in three parts: a header, a textual description
35141 section, and a trace frame section with binary data.
35142
35143 The header has the form @code{\x7fTRACE0\n}. The first byte is
35144 @code{0x7f} so as to indicate that the file contains binary data,
35145 while the @code{0} is a version number that may have different values
35146 in the future.
35147
35148 The description section consists of multiple lines of @sc{ascii} text
35149 separated by newline characters (@code{0xa}). The lines may include a
35150 variety of optional descriptive or context-setting information, such
35151 as tracepoint definitions or register set size. @value{GDBN} will
35152 ignore any line that it does not recognize. An empty line marks the end
35153 of this section.
35154
35155 @c FIXME add some specific types of data
35156
35157 The trace frame section consists of a number of consecutive frames.
35158 Each frame begins with a two-byte tracepoint number, followed by a
35159 four-byte size giving the amount of data in the frame. The data in
35160 the frame consists of a number of blocks, each introduced by a
35161 character indicating its type (at least register, memory, and trace
35162 state variable). The data in this section is raw binary, not a
35163 hexadecimal or other encoding; its endianness matches the target's
35164 endianness.
35165
35166 @c FIXME bi-arch may require endianness/arch info in description section
35167
35168 @table @code
35169 @item R @var{bytes}
35170 Register block. The number and ordering of bytes matches that of a
35171 @code{g} packet in the remote protocol. Note that these are the
35172 actual bytes, in target order and @value{GDBN} register order, not a
35173 hexadecimal encoding.
35174
35175 @item M @var{address} @var{length} @var{bytes}...
35176 Memory block. This is a contiguous block of memory, at the 8-byte
35177 address @var{address}, with a 2-byte length @var{length}, followed by
35178 @var{length} bytes.
35179
35180 @item V @var{number} @var{value}
35181 Trace state variable block. This records the 8-byte signed value
35182 @var{value} of trace state variable numbered @var{number}.
35183
35184 @end table
35185
35186 Future enhancements of the trace file format may include additional types
35187 of blocks.
35188
35189 @node Target Descriptions
35190 @appendix Target Descriptions
35191 @cindex target descriptions
35192
35193 @strong{Warning:} target descriptions are still under active development,
35194 and the contents and format may change between @value{GDBN} releases.
35195 The format is expected to stabilize in the future.
35196
35197 One of the challenges of using @value{GDBN} to debug embedded systems
35198 is that there are so many minor variants of each processor
35199 architecture in use. It is common practice for vendors to start with
35200 a standard processor core --- ARM, PowerPC, or MIPS, for example ---
35201 and then make changes to adapt it to a particular market niche. Some
35202 architectures have hundreds of variants, available from dozens of
35203 vendors. This leads to a number of problems:
35204
35205 @itemize @bullet
35206 @item
35207 With so many different customized processors, it is difficult for
35208 the @value{GDBN} maintainers to keep up with the changes.
35209 @item
35210 Since individual variants may have short lifetimes or limited
35211 audiences, it may not be worthwhile to carry information about every
35212 variant in the @value{GDBN} source tree.
35213 @item
35214 When @value{GDBN} does support the architecture of the embedded system
35215 at hand, the task of finding the correct architecture name to give the
35216 @command{set architecture} command can be error-prone.
35217 @end itemize
35218
35219 To address these problems, the @value{GDBN} remote protocol allows a
35220 target system to not only identify itself to @value{GDBN}, but to
35221 actually describe its own features. This lets @value{GDBN} support
35222 processor variants it has never seen before --- to the extent that the
35223 descriptions are accurate, and that @value{GDBN} understands them.
35224
35225 @value{GDBN} must be linked with the Expat library to support XML
35226 target descriptions. @xref{Expat}.
35227
35228 @menu
35229 * Retrieving Descriptions:: How descriptions are fetched from a target.
35230 * Target Description Format:: The contents of a target description.
35231 * Predefined Target Types:: Standard types available for target
35232 descriptions.
35233 * Standard Target Features:: Features @value{GDBN} knows about.
35234 @end menu
35235
35236 @node Retrieving Descriptions
35237 @section Retrieving Descriptions
35238
35239 Target descriptions can be read from the target automatically, or
35240 specified by the user manually. The default behavior is to read the
35241 description from the target. @value{GDBN} retrieves it via the remote
35242 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
35243 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
35244 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
35245 XML document, of the form described in @ref{Target Description
35246 Format}.
35247
35248 Alternatively, you can specify a file to read for the target description.
35249 If a file is set, the target will not be queried. The commands to
35250 specify a file are:
35251
35252 @table @code
35253 @cindex set tdesc filename
35254 @item set tdesc filename @var{path}
35255 Read the target description from @var{path}.
35256
35257 @cindex unset tdesc filename
35258 @item unset tdesc filename
35259 Do not read the XML target description from a file. @value{GDBN}
35260 will use the description supplied by the current target.
35261
35262 @cindex show tdesc filename
35263 @item show tdesc filename
35264 Show the filename to read for a target description, if any.
35265 @end table
35266
35267
35268 @node Target Description Format
35269 @section Target Description Format
35270 @cindex target descriptions, XML format
35271
35272 A target description annex is an @uref{http://www.w3.org/XML/, XML}
35273 document which complies with the Document Type Definition provided in
35274 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
35275 means you can use generally available tools like @command{xmllint} to
35276 check that your feature descriptions are well-formed and valid.
35277 However, to help people unfamiliar with XML write descriptions for
35278 their targets, we also describe the grammar here.
35279
35280 Target descriptions can identify the architecture of the remote target
35281 and (for some architectures) provide information about custom register
35282 sets. They can also identify the OS ABI of the remote target.
35283 @value{GDBN} can use this information to autoconfigure for your
35284 target, or to warn you if you connect to an unsupported target.
35285
35286 Here is a simple target description:
35287
35288 @smallexample
35289 <target version="1.0">
35290 <architecture>i386:x86-64</architecture>
35291 </target>
35292 @end smallexample
35293
35294 @noindent
35295 This minimal description only says that the target uses
35296 the x86-64 architecture.
35297
35298 A target description has the following overall form, with [ ] marking
35299 optional elements and @dots{} marking repeatable elements. The elements
35300 are explained further below.
35301
35302 @smallexample
35303 <?xml version="1.0"?>
35304 <!DOCTYPE target SYSTEM "gdb-target.dtd">
35305 <target version="1.0">
35306 @r{[}@var{architecture}@r{]}
35307 @r{[}@var{osabi}@r{]}
35308 @r{[}@var{compatible}@r{]}
35309 @r{[}@var{feature}@dots{}@r{]}
35310 </target>
35311 @end smallexample
35312
35313 @noindent
35314 The description is generally insensitive to whitespace and line
35315 breaks, under the usual common-sense rules. The XML version
35316 declaration and document type declaration can generally be omitted
35317 (@value{GDBN} does not require them), but specifying them may be
35318 useful for XML validation tools. The @samp{version} attribute for
35319 @samp{<target>} may also be omitted, but we recommend
35320 including it; if future versions of @value{GDBN} use an incompatible
35321 revision of @file{gdb-target.dtd}, they will detect and report
35322 the version mismatch.
35323
35324 @subsection Inclusion
35325 @cindex target descriptions, inclusion
35326 @cindex XInclude
35327 @ifnotinfo
35328 @cindex <xi:include>
35329 @end ifnotinfo
35330
35331 It can sometimes be valuable to split a target description up into
35332 several different annexes, either for organizational purposes, or to
35333 share files between different possible target descriptions. You can
35334 divide a description into multiple files by replacing any element of
35335 the target description with an inclusion directive of the form:
35336
35337 @smallexample
35338 <xi:include href="@var{document}"/>
35339 @end smallexample
35340
35341 @noindent
35342 When @value{GDBN} encounters an element of this form, it will retrieve
35343 the named XML @var{document}, and replace the inclusion directive with
35344 the contents of that document. If the current description was read
35345 using @samp{qXfer}, then so will be the included document;
35346 @var{document} will be interpreted as the name of an annex. If the
35347 current description was read from a file, @value{GDBN} will look for
35348 @var{document} as a file in the same directory where it found the
35349 original description.
35350
35351 @subsection Architecture
35352 @cindex <architecture>
35353
35354 An @samp{<architecture>} element has this form:
35355
35356 @smallexample
35357 <architecture>@var{arch}</architecture>
35358 @end smallexample
35359
35360 @var{arch} is one of the architectures from the set accepted by
35361 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
35362
35363 @subsection OS ABI
35364 @cindex @code{<osabi>}
35365
35366 This optional field was introduced in @value{GDBN} version 7.0.
35367 Previous versions of @value{GDBN} ignore it.
35368
35369 An @samp{<osabi>} element has this form:
35370
35371 @smallexample
35372 <osabi>@var{abi-name}</osabi>
35373 @end smallexample
35374
35375 @var{abi-name} is an OS ABI name from the same selection accepted by
35376 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
35377
35378 @subsection Compatible Architecture
35379 @cindex @code{<compatible>}
35380
35381 This optional field was introduced in @value{GDBN} version 7.0.
35382 Previous versions of @value{GDBN} ignore it.
35383
35384 A @samp{<compatible>} element has this form:
35385
35386 @smallexample
35387 <compatible>@var{arch}</compatible>
35388 @end smallexample
35389
35390 @var{arch} is one of the architectures from the set accepted by
35391 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
35392
35393 A @samp{<compatible>} element is used to specify that the target
35394 is able to run binaries in some other than the main target architecture
35395 given by the @samp{<architecture>} element. For example, on the
35396 Cell Broadband Engine, the main architecture is @code{powerpc:common}
35397 or @code{powerpc:common64}, but the system is able to run binaries
35398 in the @code{spu} architecture as well. The way to describe this
35399 capability with @samp{<compatible>} is as follows:
35400
35401 @smallexample
35402 <architecture>powerpc:common</architecture>
35403 <compatible>spu</compatible>
35404 @end smallexample
35405
35406 @subsection Features
35407 @cindex <feature>
35408
35409 Each @samp{<feature>} describes some logical portion of the target
35410 system. Features are currently used to describe available CPU
35411 registers and the types of their contents. A @samp{<feature>} element
35412 has this form:
35413
35414 @smallexample
35415 <feature name="@var{name}">
35416 @r{[}@var{type}@dots{}@r{]}
35417 @var{reg}@dots{}
35418 </feature>
35419 @end smallexample
35420
35421 @noindent
35422 Each feature's name should be unique within the description. The name
35423 of a feature does not matter unless @value{GDBN} has some special
35424 knowledge of the contents of that feature; if it does, the feature
35425 should have its standard name. @xref{Standard Target Features}.
35426
35427 @subsection Types
35428
35429 Any register's value is a collection of bits which @value{GDBN} must
35430 interpret. The default interpretation is a two's complement integer,
35431 but other types can be requested by name in the register description.
35432 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
35433 Target Types}), and the description can define additional composite types.
35434
35435 Each type element must have an @samp{id} attribute, which gives
35436 a unique (within the containing @samp{<feature>}) name to the type.
35437 Types must be defined before they are used.
35438
35439 @cindex <vector>
35440 Some targets offer vector registers, which can be treated as arrays
35441 of scalar elements. These types are written as @samp{<vector>} elements,
35442 specifying the array element type, @var{type}, and the number of elements,
35443 @var{count}:
35444
35445 @smallexample
35446 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
35447 @end smallexample
35448
35449 @cindex <union>
35450 If a register's value is usefully viewed in multiple ways, define it
35451 with a union type containing the useful representations. The
35452 @samp{<union>} element contains one or more @samp{<field>} elements,
35453 each of which has a @var{name} and a @var{type}:
35454
35455 @smallexample
35456 <union id="@var{id}">
35457 <field name="@var{name}" type="@var{type}"/>
35458 @dots{}
35459 </union>
35460 @end smallexample
35461
35462 @cindex <struct>
35463 If a register's value is composed from several separate values, define
35464 it with a structure type. There are two forms of the @samp{<struct>}
35465 element; a @samp{<struct>} element must either contain only bitfields
35466 or contain no bitfields. If the structure contains only bitfields,
35467 its total size in bytes must be specified, each bitfield must have an
35468 explicit start and end, and bitfields are automatically assigned an
35469 integer type. The field's @var{start} should be less than or
35470 equal to its @var{end}, and zero represents the least significant bit.
35471
35472 @smallexample
35473 <struct id="@var{id}" size="@var{size}">
35474 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
35475 @dots{}
35476 </struct>
35477 @end smallexample
35478
35479 If the structure contains no bitfields, then each field has an
35480 explicit type, and no implicit padding is added.
35481
35482 @smallexample
35483 <struct id="@var{id}">
35484 <field name="@var{name}" type="@var{type}"/>
35485 @dots{}
35486 </struct>
35487 @end smallexample
35488
35489 @cindex <flags>
35490 If a register's value is a series of single-bit flags, define it with
35491 a flags type. The @samp{<flags>} element has an explicit @var{size}
35492 and contains one or more @samp{<field>} elements. Each field has a
35493 @var{name}, a @var{start}, and an @var{end}. Only single-bit flags
35494 are supported.
35495
35496 @smallexample
35497 <flags id="@var{id}" size="@var{size}">
35498 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
35499 @dots{}
35500 </flags>
35501 @end smallexample
35502
35503 @subsection Registers
35504 @cindex <reg>
35505
35506 Each register is represented as an element with this form:
35507
35508 @smallexample
35509 <reg name="@var{name}"
35510 bitsize="@var{size}"
35511 @r{[}regnum="@var{num}"@r{]}
35512 @r{[}save-restore="@var{save-restore}"@r{]}
35513 @r{[}type="@var{type}"@r{]}
35514 @r{[}group="@var{group}"@r{]}/>
35515 @end smallexample
35516
35517 @noindent
35518 The components are as follows:
35519
35520 @table @var
35521
35522 @item name
35523 The register's name; it must be unique within the target description.
35524
35525 @item bitsize
35526 The register's size, in bits.
35527
35528 @item regnum
35529 The register's number. If omitted, a register's number is one greater
35530 than that of the previous register (either in the current feature or in
35531 a preceeding feature); the first register in the target description
35532 defaults to zero. This register number is used to read or write
35533 the register; e.g.@: it is used in the remote @code{p} and @code{P}
35534 packets, and registers appear in the @code{g} and @code{G} packets
35535 in order of increasing register number.
35536
35537 @item save-restore
35538 Whether the register should be preserved across inferior function
35539 calls; this must be either @code{yes} or @code{no}. The default is
35540 @code{yes}, which is appropriate for most registers except for
35541 some system control registers; this is not related to the target's
35542 ABI.
35543
35544 @item type
35545 The type of the register. @var{type} may be a predefined type, a type
35546 defined in the current feature, or one of the special types @code{int}
35547 and @code{float}. @code{int} is an integer type of the correct size
35548 for @var{bitsize}, and @code{float} is a floating point type (in the
35549 architecture's normal floating point format) of the correct size for
35550 @var{bitsize}. The default is @code{int}.
35551
35552 @item group
35553 The register group to which this register belongs. @var{group} must
35554 be either @code{general}, @code{float}, or @code{vector}. If no
35555 @var{group} is specified, @value{GDBN} will not display the register
35556 in @code{info registers}.
35557
35558 @end table
35559
35560 @node Predefined Target Types
35561 @section Predefined Target Types
35562 @cindex target descriptions, predefined types
35563
35564 Type definitions in the self-description can build up composite types
35565 from basic building blocks, but can not define fundamental types. Instead,
35566 standard identifiers are provided by @value{GDBN} for the fundamental
35567 types. The currently supported types are:
35568
35569 @table @code
35570
35571 @item int8
35572 @itemx int16
35573 @itemx int32
35574 @itemx int64
35575 @itemx int128
35576 Signed integer types holding the specified number of bits.
35577
35578 @item uint8
35579 @itemx uint16
35580 @itemx uint32
35581 @itemx uint64
35582 @itemx uint128
35583 Unsigned integer types holding the specified number of bits.
35584
35585 @item code_ptr
35586 @itemx data_ptr
35587 Pointers to unspecified code and data. The program counter and
35588 any dedicated return address register may be marked as code
35589 pointers; printing a code pointer converts it into a symbolic
35590 address. The stack pointer and any dedicated address registers
35591 may be marked as data pointers.
35592
35593 @item ieee_single
35594 Single precision IEEE floating point.
35595
35596 @item ieee_double
35597 Double precision IEEE floating point.
35598
35599 @item arm_fpa_ext
35600 The 12-byte extended precision format used by ARM FPA registers.
35601
35602 @item i387_ext
35603 The 10-byte extended precision format used by x87 registers.
35604
35605 @item i386_eflags
35606 32bit @sc{eflags} register used by x86.
35607
35608 @item i386_mxcsr
35609 32bit @sc{mxcsr} register used by x86.
35610
35611 @end table
35612
35613 @node Standard Target Features
35614 @section Standard Target Features
35615 @cindex target descriptions, standard features
35616
35617 A target description must contain either no registers or all the
35618 target's registers. If the description contains no registers, then
35619 @value{GDBN} will assume a default register layout, selected based on
35620 the architecture. If the description contains any registers, the
35621 default layout will not be used; the standard registers must be
35622 described in the target description, in such a way that @value{GDBN}
35623 can recognize them.
35624
35625 This is accomplished by giving specific names to feature elements
35626 which contain standard registers. @value{GDBN} will look for features
35627 with those names and verify that they contain the expected registers;
35628 if any known feature is missing required registers, or if any required
35629 feature is missing, @value{GDBN} will reject the target
35630 description. You can add additional registers to any of the
35631 standard features --- @value{GDBN} will display them just as if
35632 they were added to an unrecognized feature.
35633
35634 This section lists the known features and their expected contents.
35635 Sample XML documents for these features are included in the
35636 @value{GDBN} source tree, in the directory @file{gdb/features}.
35637
35638 Names recognized by @value{GDBN} should include the name of the
35639 company or organization which selected the name, and the overall
35640 architecture to which the feature applies; so e.g.@: the feature
35641 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
35642
35643 The names of registers are not case sensitive for the purpose
35644 of recognizing standard features, but @value{GDBN} will only display
35645 registers using the capitalization used in the description.
35646
35647 @menu
35648 * ARM Features::
35649 * i386 Features::
35650 * MIPS Features::
35651 * M68K Features::
35652 * PowerPC Features::
35653 @end menu
35654
35655
35656 @node ARM Features
35657 @subsection ARM Features
35658 @cindex target descriptions, ARM features
35659
35660 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
35661 ARM targets.
35662 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
35663 @samp{lr}, @samp{pc}, and @samp{cpsr}.
35664
35665 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
35666 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
35667 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
35668 and @samp{xpsr}.
35669
35670 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
35671 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
35672
35673 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
35674 it should contain at least registers @samp{wR0} through @samp{wR15} and
35675 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
35676 @samp{wCSSF}, and @samp{wCASF} registers are optional.
35677
35678 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
35679 should contain at least registers @samp{d0} through @samp{d15}. If
35680 they are present, @samp{d16} through @samp{d31} should also be included.
35681 @value{GDBN} will synthesize the single-precision registers from
35682 halves of the double-precision registers.
35683
35684 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
35685 need to contain registers; it instructs @value{GDBN} to display the
35686 VFP double-precision registers as vectors and to synthesize the
35687 quad-precision registers from pairs of double-precision registers.
35688 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
35689 be present and include 32 double-precision registers.
35690
35691 @node i386 Features
35692 @subsection i386 Features
35693 @cindex target descriptions, i386 features
35694
35695 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
35696 targets. It should describe the following registers:
35697
35698 @itemize @minus
35699 @item
35700 @samp{eax} through @samp{edi} plus @samp{eip} for i386
35701 @item
35702 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
35703 @item
35704 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
35705 @samp{fs}, @samp{gs}
35706 @item
35707 @samp{st0} through @samp{st7}
35708 @item
35709 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
35710 @samp{foseg}, @samp{fooff} and @samp{fop}
35711 @end itemize
35712
35713 The register sets may be different, depending on the target.
35714
35715 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
35716 describe registers:
35717
35718 @itemize @minus
35719 @item
35720 @samp{xmm0} through @samp{xmm7} for i386
35721 @item
35722 @samp{xmm0} through @samp{xmm15} for amd64
35723 @item
35724 @samp{mxcsr}
35725 @end itemize
35726
35727 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
35728 @samp{org.gnu.gdb.i386.sse} feature. It should
35729 describe the upper 128 bits of @sc{ymm} registers:
35730
35731 @itemize @minus
35732 @item
35733 @samp{ymm0h} through @samp{ymm7h} for i386
35734 @item
35735 @samp{ymm0h} through @samp{ymm15h} for amd64
35736 @item
35737 @end itemize
35738
35739 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
35740 describe a single register, @samp{orig_eax}.
35741
35742 @node MIPS Features
35743 @subsection MIPS Features
35744 @cindex target descriptions, MIPS features
35745
35746 The @samp{org.gnu.gdb.mips.cpu} feature is required for MIPS targets.
35747 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
35748 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
35749 on the target.
35750
35751 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
35752 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
35753 registers. They may be 32-bit or 64-bit depending on the target.
35754
35755 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
35756 it may be optional in a future version of @value{GDBN}. It should
35757 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
35758 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
35759
35760 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
35761 contain a single register, @samp{restart}, which is used by the
35762 Linux kernel to control restartable syscalls.
35763
35764 @node M68K Features
35765 @subsection M68K Features
35766 @cindex target descriptions, M68K features
35767
35768 @table @code
35769 @item @samp{org.gnu.gdb.m68k.core}
35770 @itemx @samp{org.gnu.gdb.coldfire.core}
35771 @itemx @samp{org.gnu.gdb.fido.core}
35772 One of those features must be always present.
35773 The feature that is present determines which flavor of m68k is
35774 used. The feature that is present should contain registers
35775 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
35776 @samp{sp}, @samp{ps} and @samp{pc}.
35777
35778 @item @samp{org.gnu.gdb.coldfire.fp}
35779 This feature is optional. If present, it should contain registers
35780 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
35781 @samp{fpiaddr}.
35782 @end table
35783
35784 @node PowerPC Features
35785 @subsection PowerPC Features
35786 @cindex target descriptions, PowerPC features
35787
35788 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
35789 targets. It should contain registers @samp{r0} through @samp{r31},
35790 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
35791 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
35792
35793 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
35794 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
35795
35796 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
35797 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
35798 and @samp{vrsave}.
35799
35800 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
35801 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
35802 will combine these registers with the floating point registers
35803 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
35804 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
35805 through @samp{vs63}, the set of vector registers for POWER7.
35806
35807 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
35808 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
35809 @samp{spefscr}. SPE targets should provide 32-bit registers in
35810 @samp{org.gnu.gdb.power.core} and provide the upper halves in
35811 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
35812 these to present registers @samp{ev0} through @samp{ev31} to the
35813 user.
35814
35815 @node Operating System Information
35816 @appendix Operating System Information
35817 @cindex operating system information
35818
35819 @menu
35820 * Process list::
35821 @end menu
35822
35823 Users of @value{GDBN} often wish to obtain information about the state of
35824 the operating system running on the target---for example the list of
35825 processes, or the list of open files. This section describes the
35826 mechanism that makes it possible. This mechanism is similar to the
35827 target features mechanism (@pxref{Target Descriptions}), but focuses
35828 on a different aspect of target.
35829
35830 Operating system information is retrived from the target via the
35831 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
35832 read}). The object name in the request should be @samp{osdata}, and
35833 the @var{annex} identifies the data to be fetched.
35834
35835 @node Process list
35836 @appendixsection Process list
35837 @cindex operating system information, process list
35838
35839 When requesting the process list, the @var{annex} field in the
35840 @samp{qXfer} request should be @samp{processes}. The returned data is
35841 an XML document. The formal syntax of this document is defined in
35842 @file{gdb/features/osdata.dtd}.
35843
35844 An example document is:
35845
35846 @smallexample
35847 <?xml version="1.0"?>
35848 <!DOCTYPE target SYSTEM "osdata.dtd">
35849 <osdata type="processes">
35850 <item>
35851 <column name="pid">1</column>
35852 <column name="user">root</column>
35853 <column name="command">/sbin/init</column>
35854 <column name="cores">1,2,3</column>
35855 </item>
35856 </osdata>
35857 @end smallexample
35858
35859 Each item should include a column whose name is @samp{pid}. The value
35860 of that column should identify the process on the target. The
35861 @samp{user} and @samp{command} columns are optional, and will be
35862 displayed by @value{GDBN}. The @samp{cores} column, if present,
35863 should contain a comma-separated list of cores that this process
35864 is running on. Target may provide additional columns,
35865 which @value{GDBN} currently ignores.
35866
35867 @include gpl.texi
35868
35869 @node GNU Free Documentation License
35870 @appendix GNU Free Documentation License
35871 @include fdl.texi
35872
35873 @node Index
35874 @unnumbered Index
35875
35876 @printindex cp
35877
35878 @tex
35879 % I think something like @colophon should be in texinfo. In the
35880 % meantime:
35881 \long\def\colophon{\hbox to0pt{}\vfill
35882 \centerline{The body of this manual is set in}
35883 \centerline{\fontname\tenrm,}
35884 \centerline{with headings in {\bf\fontname\tenbf}}
35885 \centerline{and examples in {\tt\fontname\tentt}.}
35886 \centerline{{\it\fontname\tenit\/},}
35887 \centerline{{\bf\fontname\tenbf}, and}
35888 \centerline{{\sl\fontname\tensl\/}}
35889 \centerline{are used for emphasis.}\vfill}
35890 \page\colophon
35891 % Blame: doc@cygnus.com, 1991.
35892 @end tex
35893
35894 @bye
This page took 0.942563 seconds and 4 git commands to generate.