1 \input texinfo @c -*- texinfo -*-
2 @setfilename gdbint.info
4 @dircategory Programming & development tools.
6 * Gdb-Internals: (gdbint). The GNU debugger's internals.
10 This file documents the internals of the GNU debugger @value{GDBN}.
11 Copyright 1990,1991,1992,1993,1994,1996,1998,1999,2000,2001,2002,2003
12 Free Software Foundation, Inc.
13 Contributed by Cygnus Solutions. Written by John Gilmore.
14 Second Edition by Stan Shebs.
16 Permission is granted to copy, distribute and/or modify this document
17 under the terms of the GNU Free Documentation License, Version 1.1 or
18 any later version published by the Free Software Foundation; with no
19 Invariant Sections, with the Front-Cover Texts being ``A GNU Manual,''
20 and with the Back-Cover Texts as in (a) below.
22 (a) The FSF's Back-Cover Text is: ``You have freedom to copy and modify
23 this GNU Manual, like GNU software. Copies published by the Free
24 Software Foundation raise funds for GNU development.''
27 @setchapternewpage off
28 @settitle @value{GDBN} Internals
34 @title @value{GDBN} Internals
35 @subtitle{A guide to the internals of the GNU debugger}
37 @author Cygnus Solutions
38 @author Second Edition:
40 @author Cygnus Solutions
43 \def\$#1${{#1}} % Kluge: collect RCS revision info without $...$
44 \xdef\manvers{\$Revision$} % For use in headers, footers too
46 \hfill Cygnus Solutions\par
48 \hfill \TeX{}info \texinfoversion\par
52 @vskip 0pt plus 1filll
53 Copyright @copyright{} 1990,1991,1992,1993,1994,1996,1998,1999,2000,2001,
54 2002, 2003 Free Software Foundation, Inc.
56 Permission is granted to copy, distribute and/or modify this document
57 under the terms of the GNU Free Documentation License, Version 1.1 or
58 any later version published by the Free Software Foundation; with no
59 Invariant Sections, with the Front-Cover Texts being ``A GNU Manual,''
60 and with the Back-Cover Texts as in (a) below.
62 (a) The FSF's Back-Cover Text is: ``You have freedom to copy and modify
63 this GNU Manual, like GNU software. Copies published by the Free
64 Software Foundation raise funds for GNU development.''
70 @c Perhaps this should be the title of the document (but only for info,
71 @c not for TeX). Existing GNU manuals seem inconsistent on this point.
72 @top Scope of this Document
74 This document documents the internals of the GNU debugger, @value{GDBN}. It
75 includes description of @value{GDBN}'s key algorithms and operations, as well
76 as the mechanisms that adapt @value{GDBN} to specific hosts and targets.
82 * Errors and Warnings::
88 * Target Architecture Definition::
89 * Target Vector Definition::
98 * GNU Free Documentation License:: The license for this documentation
104 @chapter Requirements
105 @cindex requirements for @value{GDBN}
107 Before diving into the internals, you should understand the formal
108 requirements and other expectations for @value{GDBN}. Although some
109 of these may seem obvious, there have been proposals for @value{GDBN}
110 that have run counter to these requirements.
112 First of all, @value{GDBN} is a debugger. It's not designed to be a
113 front panel for embedded systems. It's not a text editor. It's not a
114 shell. It's not a programming environment.
116 @value{GDBN} is an interactive tool. Although a batch mode is
117 available, @value{GDBN}'s primary role is to interact with a human
120 @value{GDBN} should be responsive to the user. A programmer hot on
121 the trail of a nasty bug, and operating under a looming deadline, is
122 going to be very impatient of everything, including the response time
123 to debugger commands.
125 @value{GDBN} should be relatively permissive, such as for expressions.
126 While the compiler should be picky (or have the option to be made
127 picky), since source code lives for a long time usually, the
128 programmer doing debugging shouldn't be spending time figuring out to
129 mollify the debugger.
131 @value{GDBN} will be called upon to deal with really large programs.
132 Executable sizes of 50 to 100 megabytes occur regularly, and we've
133 heard reports of programs approaching 1 gigabyte in size.
135 @value{GDBN} should be able to run everywhere. No other debugger is
136 available for even half as many configurations as @value{GDBN}
140 @node Overall Structure
142 @chapter Overall Structure
144 @value{GDBN} consists of three major subsystems: user interface,
145 symbol handling (the @dfn{symbol side}), and target system handling (the
148 The user interface consists of several actual interfaces, plus
151 The symbol side consists of object file readers, debugging info
152 interpreters, symbol table management, source language expression
153 parsing, type and value printing.
155 The target side consists of execution control, stack frame analysis, and
156 physical target manipulation.
158 The target side/symbol side division is not formal, and there are a
159 number of exceptions. For instance, core file support involves symbolic
160 elements (the basic core file reader is in BFD) and target elements (it
161 supplies the contents of memory and the values of registers). Instead,
162 this division is useful for understanding how the minor subsystems
165 @section The Symbol Side
167 The symbolic side of @value{GDBN} can be thought of as ``everything
168 you can do in @value{GDBN} without having a live program running''.
169 For instance, you can look at the types of variables, and evaluate
170 many kinds of expressions.
172 @section The Target Side
174 The target side of @value{GDBN} is the ``bits and bytes manipulator''.
175 Although it may make reference to symbolic info here and there, most
176 of the target side will run with only a stripped executable
177 available---or even no executable at all, in remote debugging cases.
179 Operations such as disassembly, stack frame crawls, and register
180 display, are able to work with no symbolic info at all. In some cases,
181 such as disassembly, @value{GDBN} will use symbolic info to present addresses
182 relative to symbols rather than as raw numbers, but it will work either
185 @section Configurations
189 @dfn{Host} refers to attributes of the system where @value{GDBN} runs.
190 @dfn{Target} refers to the system where the program being debugged
191 executes. In most cases they are the same machine, in which case a
192 third type of @dfn{Native} attributes come into play.
194 Defines and include files needed to build on the host are host support.
195 Examples are tty support, system defined types, host byte order, host
198 Defines and information needed to handle the target format are target
199 dependent. Examples are the stack frame format, instruction set,
200 breakpoint instruction, registers, and how to set up and tear down the stack
203 Information that is only needed when the host and target are the same,
204 is native dependent. One example is Unix child process support; if the
205 host and target are not the same, doing a fork to start the target
206 process is a bad idea. The various macros needed for finding the
207 registers in the @code{upage}, running @code{ptrace}, and such are all
208 in the native-dependent files.
210 Another example of native-dependent code is support for features that
211 are really part of the target environment, but which require
212 @code{#include} files that are only available on the host system. Core
213 file handling and @code{setjmp} handling are two common cases.
215 When you want to make @value{GDBN} work ``native'' on a particular machine, you
216 have to include all three kinds of information.
224 @value{GDBN} uses a number of debugging-specific algorithms. They are
225 often not very complicated, but get lost in the thicket of special
226 cases and real-world issues. This chapter describes the basic
227 algorithms and mentions some of the specific target definitions that
233 @cindex call stack frame
234 A frame is a construct that @value{GDBN} uses to keep track of calling
235 and called functions.
237 @findex create_new_frame
239 @code{FRAME_FP} in the machine description has no meaning to the
240 machine-independent part of @value{GDBN}, except that it is used when
241 setting up a new frame from scratch, as follows:
244 create_new_frame (read_register (FP_REGNUM), read_pc ()));
247 @cindex frame pointer register
248 Other than that, all the meaning imparted to @code{FP_REGNUM} is
249 imparted by the machine-dependent code. So, @code{FP_REGNUM} can have
250 any value that is convenient for the code that creates new frames.
251 (@code{create_new_frame} calls @code{INIT_EXTRA_FRAME_INFO} if it is
252 defined; that is where you should use the @code{FP_REGNUM} value, if
253 your frames are nonstandard.)
256 Given a @value{GDBN} frame, define @code{FRAME_CHAIN} to determine the
257 address of the calling function's frame. This will be used to create a
258 new @value{GDBN} frame struct, and then @code{INIT_EXTRA_FRAME_INFO} and
259 @code{DEPRECATED_INIT_FRAME_PC} will be called for the new frame.
261 @section Breakpoint Handling
264 In general, a breakpoint is a user-designated location in the program
265 where the user wants to regain control if program execution ever reaches
268 There are two main ways to implement breakpoints; either as ``hardware''
269 breakpoints or as ``software'' breakpoints.
271 @cindex hardware breakpoints
272 @cindex program counter
273 Hardware breakpoints are sometimes available as a builtin debugging
274 features with some chips. Typically these work by having dedicated
275 register into which the breakpoint address may be stored. If the PC
276 (shorthand for @dfn{program counter})
277 ever matches a value in a breakpoint registers, the CPU raises an
278 exception and reports it to @value{GDBN}.
280 Another possibility is when an emulator is in use; many emulators
281 include circuitry that watches the address lines coming out from the
282 processor, and force it to stop if the address matches a breakpoint's
285 A third possibility is that the target already has the ability to do
286 breakpoints somehow; for instance, a ROM monitor may do its own
287 software breakpoints. So although these are not literally ``hardware
288 breakpoints'', from @value{GDBN}'s point of view they work the same;
289 @value{GDBN} need not do nothing more than set the breakpoint and wait
290 for something to happen.
292 Since they depend on hardware resources, hardware breakpoints may be
293 limited in number; when the user asks for more, @value{GDBN} will
294 start trying to set software breakpoints. (On some architectures,
295 notably the 32-bit x86 platforms, @value{GDBN} cannot always know
296 whether there's enough hardware resources to insert all the hardware
297 breakpoints and watchpoints. On those platforms, @value{GDBN} prints
298 an error message only when the program being debugged is continued.)
300 @cindex software breakpoints
301 Software breakpoints require @value{GDBN} to do somewhat more work.
302 The basic theory is that @value{GDBN} will replace a program
303 instruction with a trap, illegal divide, or some other instruction
304 that will cause an exception, and then when it's encountered,
305 @value{GDBN} will take the exception and stop the program. When the
306 user says to continue, @value{GDBN} will restore the original
307 instruction, single-step, re-insert the trap, and continue on.
309 Since it literally overwrites the program being tested, the program area
310 must be writable, so this technique won't work on programs in ROM. It
311 can also distort the behavior of programs that examine themselves,
312 although such a situation would be highly unusual.
314 Also, the software breakpoint instruction should be the smallest size of
315 instruction, so it doesn't overwrite an instruction that might be a jump
316 target, and cause disaster when the program jumps into the middle of the
317 breakpoint instruction. (Strictly speaking, the breakpoint must be no
318 larger than the smallest interval between instructions that may be jump
319 targets; perhaps there is an architecture where only even-numbered
320 instructions may jumped to.) Note that it's possible for an instruction
321 set not to have any instructions usable for a software breakpoint,
322 although in practice only the ARC has failed to define such an
326 The basic definition of the software breakpoint is the macro
329 Basic breakpoint object handling is in @file{breakpoint.c}. However,
330 much of the interesting breakpoint action is in @file{infrun.c}.
332 @section Single Stepping
334 @section Signal Handling
336 @section Thread Handling
338 @section Inferior Function Calls
340 @section Longjmp Support
342 @cindex @code{longjmp} debugging
343 @value{GDBN} has support for figuring out that the target is doing a
344 @code{longjmp} and for stopping at the target of the jump, if we are
345 stepping. This is done with a few specialized internal breakpoints,
346 which are visible in the output of the @samp{maint info breakpoint}
349 @findex GET_LONGJMP_TARGET
350 To make this work, you need to define a macro called
351 @code{GET_LONGJMP_TARGET}, which will examine the @code{jmp_buf}
352 structure and extract the longjmp target address. Since @code{jmp_buf}
353 is target specific, you will need to define it in the appropriate
354 @file{tm-@var{target}.h} file. Look in @file{tm-sun4os4.h} and
355 @file{sparc-tdep.c} for examples of how to do this.
360 Watchpoints are a special kind of breakpoints (@pxref{Algorithms,
361 breakpoints}) which break when data is accessed rather than when some
362 instruction is executed. When you have data which changes without
363 your knowing what code does that, watchpoints are the silver bullet to
364 hunt down and kill such bugs.
366 @cindex hardware watchpoints
367 @cindex software watchpoints
368 Watchpoints can be either hardware-assisted or not; the latter type is
369 known as ``software watchpoints.'' @value{GDBN} always uses
370 hardware-assisted watchpoints if they are available, and falls back on
371 software watchpoints otherwise. Typical situations where @value{GDBN}
372 will use software watchpoints are:
376 The watched memory region is too large for the underlying hardware
377 watchpoint support. For example, each x86 debug register can watch up
378 to 4 bytes of memory, so trying to watch data structures whose size is
379 more than 16 bytes will cause @value{GDBN} to use software
383 The value of the expression to be watched depends on data held in
384 registers (as opposed to memory).
387 Too many different watchpoints requested. (On some architectures,
388 this situation is impossible to detect until the debugged program is
389 resumed.) Note that x86 debug registers are used both for hardware
390 breakpoints and for watchpoints, so setting too many hardware
391 breakpoints might cause watchpoint insertion to fail.
394 No hardware-assisted watchpoints provided by the target
398 Software watchpoints are very slow, since @value{GDBN} needs to
399 single-step the program being debugged and test the value of the
400 watched expression(s) after each instruction. The rest of this
401 section is mostly irrelevant for software watchpoints.
403 @value{GDBN} uses several macros and primitives to support hardware
407 @findex TARGET_HAS_HARDWARE_WATCHPOINTS
408 @item TARGET_HAS_HARDWARE_WATCHPOINTS
409 If defined, the target supports hardware watchpoints.
411 @findex TARGET_CAN_USE_HARDWARE_WATCHPOINT
412 @item TARGET_CAN_USE_HARDWARE_WATCHPOINT (@var{type}, @var{count}, @var{other})
413 Return the number of hardware watchpoints of type @var{type} that are
414 possible to be set. The value is positive if @var{count} watchpoints
415 of this type can be set, zero if setting watchpoints of this type is
416 not supported, and negative if @var{count} is more than the maximum
417 number of watchpoints of type @var{type} that can be set. @var{other}
418 is non-zero if other types of watchpoints are currently enabled (there
419 are architectures which cannot set watchpoints of different types at
422 @findex TARGET_REGION_OK_FOR_HW_WATCHPOINT
423 @item TARGET_REGION_OK_FOR_HW_WATCHPOINT (@var{addr}, @var{len})
424 Return non-zero if hardware watchpoints can be used to watch a region
425 whose address is @var{addr} and whose length in bytes is @var{len}.
427 @findex TARGET_REGION_SIZE_OK_FOR_HW_WATCHPOINT
428 @item TARGET_REGION_SIZE_OK_FOR_HW_WATCHPOINT (@var{size})
429 Return non-zero if hardware watchpoints can be used to watch a region
430 whose size is @var{size}. @value{GDBN} only uses this macro as a
431 fall-back, in case @code{TARGET_REGION_OK_FOR_HW_WATCHPOINT} is not
434 @findex TARGET_DISABLE_HW_WATCHPOINTS
435 @item TARGET_DISABLE_HW_WATCHPOINTS (@var{pid})
436 Disables watchpoints in the process identified by @var{pid}. This is
437 used, e.g., on HP-UX which provides operations to disable and enable
438 the page-level memory protection that implements hardware watchpoints
441 @findex TARGET_ENABLE_HW_WATCHPOINTS
442 @item TARGET_ENABLE_HW_WATCHPOINTS (@var{pid})
443 Enables watchpoints in the process identified by @var{pid}. This is
444 used, e.g., on HP-UX which provides operations to disable and enable
445 the page-level memory protection that implements hardware watchpoints
448 @findex target_insert_watchpoint
449 @findex target_remove_watchpoint
450 @item target_insert_watchpoint (@var{addr}, @var{len}, @var{type})
451 @itemx target_remove_watchpoint (@var{addr}, @var{len}, @var{type})
452 Insert or remove a hardware watchpoint starting at @var{addr}, for
453 @var{len} bytes. @var{type} is the watchpoint type, one of the
454 possible values of the enumerated data type @code{target_hw_bp_type},
455 defined by @file{breakpoint.h} as follows:
458 enum target_hw_bp_type
460 hw_write = 0, /* Common (write) HW watchpoint */
461 hw_read = 1, /* Read HW watchpoint */
462 hw_access = 2, /* Access (read or write) HW watchpoint */
463 hw_execute = 3 /* Execute HW breakpoint */
468 These two macros should return 0 for success, non-zero for failure.
470 @cindex insert or remove hardware breakpoint
471 @findex target_remove_hw_breakpoint
472 @findex target_insert_hw_breakpoint
473 @item target_remove_hw_breakpoint (@var{addr}, @var{shadow})
474 @itemx target_insert_hw_breakpoint (@var{addr}, @var{shadow})
475 Insert or remove a hardware-assisted breakpoint at address @var{addr}.
476 Returns zero for success, non-zero for failure. @var{shadow} is the
477 real contents of the byte where the breakpoint has been inserted; it
478 is generally not valid when hardware breakpoints are used, but since
479 no other code touches these values, the implementations of the above
480 two macros can use them for their internal purposes.
482 @findex target_stopped_data_address
483 @item target_stopped_data_address ()
484 If the inferior has some watchpoint that triggered, return the address
485 associated with that watchpoint. Otherwise, return zero.
487 @findex DECR_PC_AFTER_HW_BREAK
488 @item DECR_PC_AFTER_HW_BREAK
489 If defined, @value{GDBN} decrements the program counter by the value
490 of @code{DECR_PC_AFTER_HW_BREAK} after a hardware break-point. This
491 overrides the value of @code{DECR_PC_AFTER_BREAK} when a breakpoint
492 that breaks is a hardware-assisted breakpoint.
494 @findex HAVE_STEPPABLE_WATCHPOINT
495 @item HAVE_STEPPABLE_WATCHPOINT
496 If defined to a non-zero value, it is not necessary to disable a
497 watchpoint to step over it.
499 @findex HAVE_NONSTEPPABLE_WATCHPOINT
500 @item HAVE_NONSTEPPABLE_WATCHPOINT
501 If defined to a non-zero value, @value{GDBN} should disable a
502 watchpoint to step the inferior over it.
504 @findex HAVE_CONTINUABLE_WATCHPOINT
505 @item HAVE_CONTINUABLE_WATCHPOINT
506 If defined to a non-zero value, it is possible to continue the
507 inferior after a watchpoint has been hit.
509 @findex CANNOT_STEP_HW_WATCHPOINTS
510 @item CANNOT_STEP_HW_WATCHPOINTS
511 If this is defined to a non-zero value, @value{GDBN} will remove all
512 watchpoints before stepping the inferior.
514 @findex STOPPED_BY_WATCHPOINT
515 @item STOPPED_BY_WATCHPOINT (@var{wait_status})
516 Return non-zero if stopped by a watchpoint. @var{wait_status} is of
517 the type @code{struct target_waitstatus}, defined by @file{target.h}.
520 @subsection x86 Watchpoints
521 @cindex x86 debug registers
522 @cindex watchpoints, on x86
524 The 32-bit Intel x86 (a.k.a.@: ia32) processors feature special debug
525 registers designed to facilitate debugging. @value{GDBN} provides a
526 generic library of functions that x86-based ports can use to implement
527 support for watchpoints and hardware-assisted breakpoints. This
528 subsection documents the x86 watchpoint facilities in @value{GDBN}.
530 To use the generic x86 watchpoint support, a port should do the
534 @findex I386_USE_GENERIC_WATCHPOINTS
536 Define the macro @code{I386_USE_GENERIC_WATCHPOINTS} somewhere in the
537 target-dependent headers.
540 Include the @file{config/i386/nm-i386.h} header file @emph{after}
541 defining @code{I386_USE_GENERIC_WATCHPOINTS}.
544 Add @file{i386-nat.o} to the value of the Make variable
545 @code{NATDEPFILES} (@pxref{Native Debugging, NATDEPFILES}) or
546 @code{TDEPFILES} (@pxref{Target Architecture Definition, TDEPFILES}).
549 Provide implementations for the @code{I386_DR_LOW_*} macros described
550 below. Typically, each macro should call a target-specific function
551 which does the real work.
554 The x86 watchpoint support works by maintaining mirror images of the
555 debug registers. Values are copied between the mirror images and the
556 real debug registers via a set of macros which each target needs to
560 @findex I386_DR_LOW_SET_CONTROL
561 @item I386_DR_LOW_SET_CONTROL (@var{val})
562 Set the Debug Control (DR7) register to the value @var{val}.
564 @findex I386_DR_LOW_SET_ADDR
565 @item I386_DR_LOW_SET_ADDR (@var{idx}, @var{addr})
566 Put the address @var{addr} into the debug register number @var{idx}.
568 @findex I386_DR_LOW_RESET_ADDR
569 @item I386_DR_LOW_RESET_ADDR (@var{idx})
570 Reset (i.e.@: zero out) the address stored in the debug register
573 @findex I386_DR_LOW_GET_STATUS
574 @item I386_DR_LOW_GET_STATUS
575 Return the value of the Debug Status (DR6) register. This value is
576 used immediately after it is returned by
577 @code{I386_DR_LOW_GET_STATUS}, so as to support per-thread status
581 For each one of the 4 debug registers (whose indices are from 0 to 3)
582 that store addresses, a reference count is maintained by @value{GDBN},
583 to allow sharing of debug registers by several watchpoints. This
584 allows users to define several watchpoints that watch the same
585 expression, but with different conditions and/or commands, without
586 wasting debug registers which are in short supply. @value{GDBN}
587 maintains the reference counts internally, targets don't have to do
588 anything to use this feature.
590 The x86 debug registers can each watch a region that is 1, 2, or 4
591 bytes long. The ia32 architecture requires that each watched region
592 be appropriately aligned: 2-byte region on 2-byte boundary, 4-byte
593 region on 4-byte boundary. However, the x86 watchpoint support in
594 @value{GDBN} can watch unaligned regions and regions larger than 4
595 bytes (up to 16 bytes) by allocating several debug registers to watch
596 a single region. This allocation of several registers per a watched
597 region is also done automatically without target code intervention.
599 The generic x86 watchpoint support provides the following API for the
600 @value{GDBN}'s application code:
603 @findex i386_region_ok_for_watchpoint
604 @item i386_region_ok_for_watchpoint (@var{addr}, @var{len})
605 The macro @code{TARGET_REGION_OK_FOR_HW_WATCHPOINT} is set to call
606 this function. It counts the number of debug registers required to
607 watch a given region, and returns a non-zero value if that number is
608 less than 4, the number of debug registers available to x86
611 @findex i386_stopped_data_address
612 @item i386_stopped_data_address (void)
613 The macros @code{STOPPED_BY_WATCHPOINT} and
614 @code{target_stopped_data_address} are set to call this function. The
615 argument passed to @code{STOPPED_BY_WATCHPOINT} is ignored. This
616 function examines the breakpoint condition bits in the DR6 Debug
617 Status register, as returned by the @code{I386_DR_LOW_GET_STATUS}
618 macro, and returns the address associated with the first bit that is
621 @findex i386_insert_watchpoint
622 @findex i386_remove_watchpoint
623 @item i386_insert_watchpoint (@var{addr}, @var{len}, @var{type})
624 @itemx i386_remove_watchpoint (@var{addr}, @var{len}, @var{type})
625 Insert or remove a watchpoint. The macros
626 @code{target_insert_watchpoint} and @code{target_remove_watchpoint}
627 are set to call these functions. @code{i386_insert_watchpoint} first
628 looks for a debug register which is already set to watch the same
629 region for the same access types; if found, it just increments the
630 reference count of that debug register, thus implementing debug
631 register sharing between watchpoints. If no such register is found,
632 the function looks for a vacant debug register, sets its mirrored
633 value to @var{addr}, sets the mirrored value of DR7 Debug Control
634 register as appropriate for the @var{len} and @var{type} parameters,
635 and then passes the new values of the debug register and DR7 to the
636 inferior by calling @code{I386_DR_LOW_SET_ADDR} and
637 @code{I386_DR_LOW_SET_CONTROL}. If more than one debug register is
638 required to cover the given region, the above process is repeated for
641 @code{i386_remove_watchpoint} does the opposite: it resets the address
642 in the mirrored value of the debug register and its read/write and
643 length bits in the mirrored value of DR7, then passes these new
644 values to the inferior via @code{I386_DR_LOW_RESET_ADDR} and
645 @code{I386_DR_LOW_SET_CONTROL}. If a register is shared by several
646 watchpoints, each time a @code{i386_remove_watchpoint} is called, it
647 decrements the reference count, and only calls
648 @code{I386_DR_LOW_RESET_ADDR} and @code{I386_DR_LOW_SET_CONTROL} when
649 the count goes to zero.
651 @findex i386_insert_hw_breakpoint
652 @findex i386_remove_hw_breakpoint
653 @item i386_insert_hw_breakpoint (@var{addr}, @var{shadow}
654 @itemx i386_remove_hw_breakpoint (@var{addr}, @var{shadow})
655 These functions insert and remove hardware-assisted breakpoints. The
656 macros @code{target_insert_hw_breakpoint} and
657 @code{target_remove_hw_breakpoint} are set to call these functions.
658 These functions work like @code{i386_insert_watchpoint} and
659 @code{i386_remove_watchpoint}, respectively, except that they set up
660 the debug registers to watch instruction execution, and each
661 hardware-assisted breakpoint always requires exactly one debug
664 @findex i386_stopped_by_hwbp
665 @item i386_stopped_by_hwbp (void)
666 This function returns non-zero if the inferior has some watchpoint or
667 hardware breakpoint that triggered. It works like
668 @code{i386_stopped_data_address}, except that it doesn't return the
669 address whose watchpoint triggered.
671 @findex i386_cleanup_dregs
672 @item i386_cleanup_dregs (void)
673 This function clears all the reference counts, addresses, and control
674 bits in the mirror images of the debug registers. It doesn't affect
675 the actual debug registers in the inferior process.
682 x86 processors support setting watchpoints on I/O reads or writes.
683 However, since no target supports this (as of March 2001), and since
684 @code{enum target_hw_bp_type} doesn't even have an enumeration for I/O
685 watchpoints, this feature is not yet available to @value{GDBN} running
689 x86 processors can enable watchpoints locally, for the current task
690 only, or globally, for all the tasks. For each debug register,
691 there's a bit in the DR7 Debug Control register that determines
692 whether the associated address is watched locally or globally. The
693 current implementation of x86 watchpoint support in @value{GDBN}
694 always sets watchpoints to be locally enabled, since global
695 watchpoints might interfere with the underlying OS and are probably
696 unavailable in many platforms.
699 @node Errors and Warnings
700 @chapter Errors and Warnings
705 On encountering a problem, in addition to notifying the user,
706 @value{GDBN} can either abandon the operation or take evasive action
711 An operation is aborted by calling an error function:
713 @emph{prototype of simple error goes here}
718 @section Internal or External
719 @cindex internal error
720 @cindex external error
727 @chapter User Interface
729 @value{GDBN} has several user interfaces. Although the command-line interface
730 is the most common and most familiar, there are others.
732 @section Command Interpreter
734 @cindex command interpreter
736 The command interpreter in @value{GDBN} is fairly simple. It is designed to
737 allow for the set of commands to be augmented dynamically, and also
738 has a recursive subcommand capability, where the first argument to
739 a command may itself direct a lookup on a different command list.
741 For instance, the @samp{set} command just starts a lookup on the
742 @code{setlist} command list, while @samp{set thread} recurses
743 to the @code{set_thread_cmd_list}.
747 To add commands in general, use @code{add_cmd}. @code{add_com} adds to
748 the main command list, and should be used for those commands. The usual
749 place to add commands is in the @code{_initialize_@var{xyz}} routines at
750 the ends of most source files.
752 @findex add_setshow_cmd
753 @findex add_setshow_cmd_full
754 To add paired @samp{set} and @samp{show} commands, use
755 @code{add_setshow_cmd} or @code{add_setshow_cmd_full}. The former is
756 a slightly simpler interface which is useful when you don't need to
757 further modify the new command structures, while the latter returns
758 the new command structures for manipulation.
760 @cindex deprecating commands
761 @findex deprecate_cmd
762 Before removing commands from the command set it is a good idea to
763 deprecate them for some time. Use @code{deprecate_cmd} on commands or
764 aliases to set the deprecated flag. @code{deprecate_cmd} takes a
765 @code{struct cmd_list_element} as it's first argument. You can use the
766 return value from @code{add_com} or @code{add_cmd} to deprecate the
767 command immediately after it is created.
769 The first time a command is used the user will be warned and offered a
770 replacement (if one exists). Note that the replacement string passed to
771 @code{deprecate_cmd} should be the full name of the command, i.e. the
772 entire string the user should type at the command line.
774 @section UI-Independent Output---the @code{ui_out} Functions
775 @c This section is based on the documentation written by Fernando
776 @c Nasser <fnasser@redhat.com>.
778 @cindex @code{ui_out} functions
779 The @code{ui_out} functions present an abstraction level for the
780 @value{GDBN} output code. They hide the specifics of different user
781 interfaces supported by @value{GDBN}, and thus free the programmer
782 from the need to write several versions of the same code, one each for
783 every UI, to produce output.
785 @subsection Overview and Terminology
787 In general, execution of each @value{GDBN} command produces some sort
788 of output, and can even generate an input request.
790 Output can be generated for the following purposes:
794 to display a @emph{result} of an operation;
797 to convey @emph{info} or produce side-effects of a requested
801 to provide a @emph{notification} of an asynchronous event (including
802 progress indication of a prolonged asynchronous operation);
805 to display @emph{error messages} (including warnings);
808 to show @emph{debug data};
811 to @emph{query} or prompt a user for input (a special case).
815 This section mainly concentrates on how to build result output,
816 although some of it also applies to other kinds of output.
818 Generation of output that displays the results of an operation
819 involves one or more of the following:
823 output of the actual data
826 formatting the output as appropriate for console output, to make it
827 easily readable by humans
830 machine oriented formatting--a more terse formatting to allow for easy
831 parsing by programs which read @value{GDBN}'s output
834 annotation, whose purpose is to help legacy GUIs to identify interesting
838 The @code{ui_out} routines take care of the first three aspects.
839 Annotations are provided by separate annotation routines. Note that use
840 of annotations for an interface between a GUI and @value{GDBN} is
843 Output can be in the form of a single item, which we call a @dfn{field};
844 a @dfn{list} consisting of identical fields; a @dfn{tuple} consisting of
845 non-identical fields; or a @dfn{table}, which is a tuple consisting of a
846 header and a body. In a BNF-like form:
849 @item <table> @expansion{}
850 @code{<header> <body>}
851 @item <header> @expansion{}
852 @code{@{ <column> @}}
853 @item <column> @expansion{}
854 @code{<width> <alignment> <title>}
855 @item <body> @expansion{}
860 @subsection General Conventions
862 Most @code{ui_out} routines are of type @code{void}, the exceptions are
863 @code{ui_out_stream_new} (which returns a pointer to the newly created
864 object) and the @code{make_cleanup} routines.
866 The first parameter is always the @code{ui_out} vector object, a pointer
867 to a @code{struct ui_out}.
869 The @var{format} parameter is like in @code{printf} family of functions.
870 When it is present, there must also be a variable list of arguments
871 sufficient used to satisfy the @code{%} specifiers in the supplied
874 When a character string argument is not used in a @code{ui_out} function
875 call, a @code{NULL} pointer has to be supplied instead.
878 @subsection Table, Tuple and List Functions
880 @cindex list output functions
881 @cindex table output functions
882 @cindex tuple output functions
883 This section introduces @code{ui_out} routines for building lists,
884 tuples and tables. The routines to output the actual data items
885 (fields) are presented in the next section.
887 To recap: A @dfn{tuple} is a sequence of @dfn{fields}, each field
888 containing information about an object; a @dfn{list} is a sequence of
889 fields where each field describes an identical object.
891 Use the @dfn{table} functions when your output consists of a list of
892 rows (tuples) and the console output should include a heading. Use this
893 even when you are listing just one object but you still want the header.
895 @cindex nesting level in @code{ui_out} functions
896 Tables can not be nested. Tuples and lists can be nested up to a
897 maximum of five levels.
899 The overall structure of the table output code is something like this:
914 Here is the description of table-, tuple- and list-related @code{ui_out}
917 @deftypefun void ui_out_table_begin (struct ui_out *@var{uiout}, int @var{nbrofcols}, int @var{nr_rows}, const char *@var{tblid})
918 The function @code{ui_out_table_begin} marks the beginning of the output
919 of a table. It should always be called before any other @code{ui_out}
920 function for a given table. @var{nbrofcols} is the number of columns in
921 the table. @var{nr_rows} is the number of rows in the table.
922 @var{tblid} is an optional string identifying the table. The string
923 pointed to by @var{tblid} is copied by the implementation of
924 @code{ui_out_table_begin}, so the application can free the string if it
927 The companion function @code{ui_out_table_end}, described below, marks
928 the end of the table's output.
931 @deftypefun void ui_out_table_header (struct ui_out *@var{uiout}, int @var{width}, enum ui_align @var{alignment}, const char *@var{colhdr})
932 @code{ui_out_table_header} provides the header information for a single
933 table column. You call this function several times, one each for every
934 column of the table, after @code{ui_out_table_begin}, but before
935 @code{ui_out_table_body}.
937 The value of @var{width} gives the column width in characters. The
938 value of @var{alignment} is one of @code{left}, @code{center}, and
939 @code{right}, and it specifies how to align the header: left-justify,
940 center, or right-justify it. @var{colhdr} points to a string that
941 specifies the column header; the implementation copies that string, so
942 column header strings in @code{malloc}ed storage can be freed after the
946 @deftypefun void ui_out_table_body (struct ui_out *@var{uiout})
947 This function delimits the table header from the table body.
950 @deftypefun void ui_out_table_end (struct ui_out *@var{uiout})
951 This function signals the end of a table's output. It should be called
952 after the table body has been produced by the list and field output
955 There should be exactly one call to @code{ui_out_table_end} for each
956 call to @code{ui_out_table_begin}, otherwise the @code{ui_out} functions
957 will signal an internal error.
960 The output of the tuples that represent the table rows must follow the
961 call to @code{ui_out_table_body} and precede the call to
962 @code{ui_out_table_end}. You build a tuple by calling
963 @code{ui_out_tuple_begin} and @code{ui_out_tuple_end}, with suitable
964 calls to functions which actually output fields between them.
966 @deftypefun void ui_out_tuple_begin (struct ui_out *@var{uiout}, const char *@var{id})
967 This function marks the beginning of a tuple output. @var{id} points
968 to an optional string that identifies the tuple; it is copied by the
969 implementation, and so strings in @code{malloc}ed storage can be freed
973 @deftypefun void ui_out_tuple_end (struct ui_out *@var{uiout})
974 This function signals an end of a tuple output. There should be exactly
975 one call to @code{ui_out_tuple_end} for each call to
976 @code{ui_out_tuple_begin}, otherwise an internal @value{GDBN} error will
980 @deftypefun struct cleanup *make_cleanup_ui_out_tuple_begin_end (struct ui_out *@var{uiout}, const char *@var{id})
981 This function first opens the tuple and then establishes a cleanup
982 (@pxref{Coding, Cleanups}) to close the tuple. It provides a convenient
983 and correct implementation of the non-portable@footnote{The function
984 cast is not portable ISO C.} code sequence:
986 struct cleanup *old_cleanup;
987 ui_out_tuple_begin (uiout, "...");
988 old_cleanup = make_cleanup ((void(*)(void *)) ui_out_tuple_end,
993 @deftypefun void ui_out_list_begin (struct ui_out *@var{uiout}, const char *@var{id})
994 This function marks the beginning of a list output. @var{id} points to
995 an optional string that identifies the list; it is copied by the
996 implementation, and so strings in @code{malloc}ed storage can be freed
1000 @deftypefun void ui_out_list_end (struct ui_out *@var{uiout})
1001 This function signals an end of a list output. There should be exactly
1002 one call to @code{ui_out_list_end} for each call to
1003 @code{ui_out_list_begin}, otherwise an internal @value{GDBN} error will
1007 @deftypefun struct cleanup *make_cleanup_ui_out_list_begin_end (struct ui_out *@var{uiout}, const char *@var{id})
1008 Similar to @code{make_cleanup_ui_out_tuple_begin_end}, this function
1009 opens a list and then establishes cleanup (@pxref{Coding, Cleanups})
1010 that will close the list.list.
1013 @subsection Item Output Functions
1015 @cindex item output functions
1016 @cindex field output functions
1018 The functions described below produce output for the actual data
1019 items, or fields, which contain information about the object.
1021 Choose the appropriate function accordingly to your particular needs.
1023 @deftypefun void ui_out_field_fmt (struct ui_out *@var{uiout}, char *@var{fldname}, char *@var{format}, ...)
1024 This is the most general output function. It produces the
1025 representation of the data in the variable-length argument list
1026 according to formatting specifications in @var{format}, a
1027 @code{printf}-like format string. The optional argument @var{fldname}
1028 supplies the name of the field. The data items themselves are
1029 supplied as additional arguments after @var{format}.
1031 This generic function should be used only when it is not possible to
1032 use one of the specialized versions (see below).
1035 @deftypefun void ui_out_field_int (struct ui_out *@var{uiout}, const char *@var{fldname}, int @var{value})
1036 This function outputs a value of an @code{int} variable. It uses the
1037 @code{"%d"} output conversion specification. @var{fldname} specifies
1038 the name of the field.
1041 @deftypefun void ui_out_field_fmt_int (struct ui_out *@var{uiout}, int @var{width}, enum ui_align @var{alignment}, const char *@var{fldname}, int @var{value})
1042 This function outputs a value of an @code{int} variable. It differs from
1043 @code{ui_out_field_int} in that the caller specifies the desired @var{width} and @var{alignment} of the output.
1044 @var{fldname} specifies
1045 the name of the field.
1048 @deftypefun void ui_out_field_core_addr (struct ui_out *@var{uiout}, const char *@var{fldname}, CORE_ADDR @var{address})
1049 This function outputs an address.
1052 @deftypefun void ui_out_field_string (struct ui_out *@var{uiout}, const char *@var{fldname}, const char *@var{string})
1053 This function outputs a string using the @code{"%s"} conversion
1057 Sometimes, there's a need to compose your output piece by piece using
1058 functions that operate on a stream, such as @code{value_print} or
1059 @code{fprintf_symbol_filtered}. These functions accept an argument of
1060 the type @code{struct ui_file *}, a pointer to a @code{ui_file} object
1061 used to store the data stream used for the output. When you use one
1062 of these functions, you need a way to pass their results stored in a
1063 @code{ui_file} object to the @code{ui_out} functions. To this end,
1064 you first create a @code{ui_stream} object by calling
1065 @code{ui_out_stream_new}, pass the @code{stream} member of that
1066 @code{ui_stream} object to @code{value_print} and similar functions,
1067 and finally call @code{ui_out_field_stream} to output the field you
1068 constructed. When the @code{ui_stream} object is no longer needed,
1069 you should destroy it and free its memory by calling
1070 @code{ui_out_stream_delete}.
1072 @deftypefun struct ui_stream *ui_out_stream_new (struct ui_out *@var{uiout})
1073 This function creates a new @code{ui_stream} object which uses the
1074 same output methods as the @code{ui_out} object whose pointer is
1075 passed in @var{uiout}. It returns a pointer to the newly created
1076 @code{ui_stream} object.
1079 @deftypefun void ui_out_stream_delete (struct ui_stream *@var{streambuf})
1080 This functions destroys a @code{ui_stream} object specified by
1084 @deftypefun void ui_out_field_stream (struct ui_out *@var{uiout}, const char *@var{fieldname}, struct ui_stream *@var{streambuf})
1085 This function consumes all the data accumulated in
1086 @code{streambuf->stream} and outputs it like
1087 @code{ui_out_field_string} does. After a call to
1088 @code{ui_out_field_stream}, the accumulated data no longer exists, but
1089 the stream is still valid and may be used for producing more fields.
1092 @strong{Important:} If there is any chance that your code could bail
1093 out before completing output generation and reaching the point where
1094 @code{ui_out_stream_delete} is called, it is necessary to set up a
1095 cleanup, to avoid leaking memory and other resources. Here's a
1096 skeleton code to do that:
1099 struct ui_stream *mybuf = ui_out_stream_new (uiout);
1100 struct cleanup *old = make_cleanup (ui_out_stream_delete, mybuf);
1105 If the function already has the old cleanup chain set (for other kinds
1106 of cleanups), you just have to add your cleanup to it:
1109 mybuf = ui_out_stream_new (uiout);
1110 make_cleanup (ui_out_stream_delete, mybuf);
1113 Note that with cleanups in place, you should not call
1114 @code{ui_out_stream_delete} directly, or you would attempt to free the
1117 @subsection Utility Output Functions
1119 @deftypefun void ui_out_field_skip (struct ui_out *@var{uiout}, const char *@var{fldname})
1120 This function skips a field in a table. Use it if you have to leave
1121 an empty field without disrupting the table alignment. The argument
1122 @var{fldname} specifies a name for the (missing) filed.
1125 @deftypefun void ui_out_text (struct ui_out *@var{uiout}, const char *@var{string})
1126 This function outputs the text in @var{string} in a way that makes it
1127 easy to be read by humans. For example, the console implementation of
1128 this method filters the text through a built-in pager, to prevent it
1129 from scrolling off the visible portion of the screen.
1131 Use this function for printing relatively long chunks of text around
1132 the actual field data: the text it produces is not aligned according
1133 to the table's format. Use @code{ui_out_field_string} to output a
1134 string field, and use @code{ui_out_message}, described below, to
1135 output short messages.
1138 @deftypefun void ui_out_spaces (struct ui_out *@var{uiout}, int @var{nspaces})
1139 This function outputs @var{nspaces} spaces. It is handy to align the
1140 text produced by @code{ui_out_text} with the rest of the table or
1144 @deftypefun void ui_out_message (struct ui_out *@var{uiout}, int @var{verbosity}, const char *@var{format}, ...)
1145 This function produces a formatted message, provided that the current
1146 verbosity level is at least as large as given by @var{verbosity}. The
1147 current verbosity level is specified by the user with the @samp{set
1148 verbositylevel} command.@footnote{As of this writing (April 2001),
1149 setting verbosity level is not yet implemented, and is always returned
1150 as zero. So calling @code{ui_out_message} with a @var{verbosity}
1151 argument more than zero will cause the message to never be printed.}
1154 @deftypefun void ui_out_wrap_hint (struct ui_out *@var{uiout}, char *@var{indent})
1155 This function gives the console output filter (a paging filter) a hint
1156 of where to break lines which are too long. Ignored for all other
1157 output consumers. @var{indent}, if non-@code{NULL}, is the string to
1158 be printed to indent the wrapped text on the next line; it must remain
1159 accessible until the next call to @code{ui_out_wrap_hint}, or until an
1160 explicit newline is produced by one of the other functions. If
1161 @var{indent} is @code{NULL}, the wrapped text will not be indented.
1164 @deftypefun void ui_out_flush (struct ui_out *@var{uiout})
1165 This function flushes whatever output has been accumulated so far, if
1166 the UI buffers output.
1170 @subsection Examples of Use of @code{ui_out} functions
1172 @cindex using @code{ui_out} functions
1173 @cindex @code{ui_out} functions, usage examples
1174 This section gives some practical examples of using the @code{ui_out}
1175 functions to generalize the old console-oriented code in
1176 @value{GDBN}. The examples all come from functions defined on the
1177 @file{breakpoints.c} file.
1179 This example, from the @code{breakpoint_1} function, shows how to
1182 The original code was:
1185 if (!found_a_breakpoint++)
1187 annotate_breakpoints_headers ();
1190 printf_filtered ("Num ");
1192 printf_filtered ("Type ");
1194 printf_filtered ("Disp ");
1196 printf_filtered ("Enb ");
1200 printf_filtered ("Address ");
1203 printf_filtered ("What\n");
1205 annotate_breakpoints_table ();
1209 Here's the new version:
1212 nr_printable_breakpoints = @dots{};
1215 ui_out_table_begin (ui, 6, nr_printable_breakpoints, "BreakpointTable");
1217 ui_out_table_begin (ui, 5, nr_printable_breakpoints, "BreakpointTable");
1219 if (nr_printable_breakpoints > 0)
1220 annotate_breakpoints_headers ();
1221 if (nr_printable_breakpoints > 0)
1223 ui_out_table_header (uiout, 3, ui_left, "number", "Num"); /* 1 */
1224 if (nr_printable_breakpoints > 0)
1226 ui_out_table_header (uiout, 14, ui_left, "type", "Type"); /* 2 */
1227 if (nr_printable_breakpoints > 0)
1229 ui_out_table_header (uiout, 4, ui_left, "disp", "Disp"); /* 3 */
1230 if (nr_printable_breakpoints > 0)
1232 ui_out_table_header (uiout, 3, ui_left, "enabled", "Enb"); /* 4 */
1235 if (nr_printable_breakpoints > 0)
1237 if (TARGET_ADDR_BIT <= 32)
1238 ui_out_table_header (uiout, 10, ui_left, "addr", "Address");/* 5 */
1240 ui_out_table_header (uiout, 18, ui_left, "addr", "Address");/* 5 */
1242 if (nr_printable_breakpoints > 0)
1244 ui_out_table_header (uiout, 40, ui_noalign, "what", "What"); /* 6 */
1245 ui_out_table_body (uiout);
1246 if (nr_printable_breakpoints > 0)
1247 annotate_breakpoints_table ();
1250 This example, from the @code{print_one_breakpoint} function, shows how
1251 to produce the actual data for the table whose structure was defined
1252 in the above example. The original code was:
1257 printf_filtered ("%-3d ", b->number);
1259 if ((int)b->type > (sizeof(bptypes)/sizeof(bptypes[0]))
1260 || ((int) b->type != bptypes[(int) b->type].type))
1261 internal_error ("bptypes table does not describe type #%d.",
1263 printf_filtered ("%-14s ", bptypes[(int)b->type].description);
1265 printf_filtered ("%-4s ", bpdisps[(int)b->disposition]);
1267 printf_filtered ("%-3c ", bpenables[(int)b->enable]);
1271 This is the new version:
1275 ui_out_tuple_begin (uiout, "bkpt");
1277 ui_out_field_int (uiout, "number", b->number);
1279 if (((int) b->type > (sizeof (bptypes) / sizeof (bptypes[0])))
1280 || ((int) b->type != bptypes[(int) b->type].type))
1281 internal_error ("bptypes table does not describe type #%d.",
1283 ui_out_field_string (uiout, "type", bptypes[(int)b->type].description);
1285 ui_out_field_string (uiout, "disp", bpdisps[(int)b->disposition]);
1287 ui_out_field_fmt (uiout, "enabled", "%c", bpenables[(int)b->enable]);
1291 This example, also from @code{print_one_breakpoint}, shows how to
1292 produce a complicated output field using the @code{print_expression}
1293 functions which requires a stream to be passed. It also shows how to
1294 automate stream destruction with cleanups. The original code was:
1298 print_expression (b->exp, gdb_stdout);
1304 struct ui_stream *stb = ui_out_stream_new (uiout);
1305 struct cleanup *old_chain = make_cleanup_ui_out_stream_delete (stb);
1308 print_expression (b->exp, stb->stream);
1309 ui_out_field_stream (uiout, "what", local_stream);
1312 This example, also from @code{print_one_breakpoint}, shows how to use
1313 @code{ui_out_text} and @code{ui_out_field_string}. The original code
1318 if (b->dll_pathname == NULL)
1319 printf_filtered ("<any library> ");
1321 printf_filtered ("library \"%s\" ", b->dll_pathname);
1328 if (b->dll_pathname == NULL)
1330 ui_out_field_string (uiout, "what", "<any library>");
1331 ui_out_spaces (uiout, 1);
1335 ui_out_text (uiout, "library \"");
1336 ui_out_field_string (uiout, "what", b->dll_pathname);
1337 ui_out_text (uiout, "\" ");
1341 The following example from @code{print_one_breakpoint} shows how to
1342 use @code{ui_out_field_int} and @code{ui_out_spaces}. The original
1347 if (b->forked_inferior_pid != 0)
1348 printf_filtered ("process %d ", b->forked_inferior_pid);
1355 if (b->forked_inferior_pid != 0)
1357 ui_out_text (uiout, "process ");
1358 ui_out_field_int (uiout, "what", b->forked_inferior_pid);
1359 ui_out_spaces (uiout, 1);
1363 Here's an example of using @code{ui_out_field_string}. The original
1368 if (b->exec_pathname != NULL)
1369 printf_filtered ("program \"%s\" ", b->exec_pathname);
1376 if (b->exec_pathname != NULL)
1378 ui_out_text (uiout, "program \"");
1379 ui_out_field_string (uiout, "what", b->exec_pathname);
1380 ui_out_text (uiout, "\" ");
1384 Finally, here's an example of printing an address. The original code:
1388 printf_filtered ("%s ",
1389 local_hex_string_custom ((unsigned long) b->address, "08l"));
1396 ui_out_field_core_addr (uiout, "Address", b->address);
1400 @section Console Printing
1409 @cindex @code{libgdb}
1410 @code{libgdb} 1.0 was an abortive project of years ago. The theory was
1411 to provide an API to @value{GDBN}'s functionality.
1414 @cindex @code{libgdb}
1415 @code{libgdb} 2.0 is an ongoing effort to update @value{GDBN} so that is
1416 better able to support graphical and other environments.
1418 Since @code{libgdb} development is on-going, its architecture is still
1419 evolving. The following components have so far been identified:
1423 Observer - @file{gdb-events.h}.
1425 Builder - @file{ui-out.h}
1427 Event Loop - @file{event-loop.h}
1429 Library - @file{gdb.h}
1432 The model that ties these components together is described below.
1434 @section The @code{libgdb} Model
1436 A client of @code{libgdb} interacts with the library in two ways.
1440 As an observer (using @file{gdb-events}) receiving notifications from
1441 @code{libgdb} of any internal state changes (break point changes, run
1444 As a client querying @code{libgdb} (using the @file{ui-out} builder) to
1445 obtain various status values from @value{GDBN}.
1448 Since @code{libgdb} could have multiple clients (e.g. a GUI supporting
1449 the existing @value{GDBN} CLI), those clients must co-operate when
1450 controlling @code{libgdb}. In particular, a client must ensure that
1451 @code{libgdb} is idle (i.e. no other client is using @code{libgdb})
1452 before responding to a @file{gdb-event} by making a query.
1454 @section CLI support
1456 At present @value{GDBN}'s CLI is very much entangled in with the core of
1457 @code{libgdb}. Consequently, a client wishing to include the CLI in
1458 their interface needs to carefully co-ordinate its own and the CLI's
1461 It is suggested that the client set @code{libgdb} up to be bi-modal
1462 (alternate between CLI and client query modes). The notes below sketch
1467 The client registers itself as an observer of @code{libgdb}.
1469 The client create and install @code{cli-out} builder using its own
1470 versions of the @code{ui-file} @code{gdb_stderr}, @code{gdb_stdtarg} and
1471 @code{gdb_stdout} streams.
1473 The client creates a separate custom @code{ui-out} builder that is only
1474 used while making direct queries to @code{libgdb}.
1477 When the client receives input intended for the CLI, it simply passes it
1478 along. Since the @code{cli-out} builder is installed by default, all
1479 the CLI output in response to that command is routed (pronounced rooted)
1480 through to the client controlled @code{gdb_stdout} et.@: al.@: streams.
1481 At the same time, the client is kept abreast of internal changes by
1482 virtue of being a @code{libgdb} observer.
1484 The only restriction on the client is that it must wait until
1485 @code{libgdb} becomes idle before initiating any queries (using the
1486 client's custom builder).
1488 @section @code{libgdb} components
1490 @subheading Observer - @file{gdb-events.h}
1491 @file{gdb-events} provides the client with a very raw mechanism that can
1492 be used to implement an observer. At present it only allows for one
1493 observer and that observer must, internally, handle the need to delay
1494 the processing of any event notifications until after @code{libgdb} has
1495 finished the current command.
1497 @subheading Builder - @file{ui-out.h}
1498 @file{ui-out} provides the infrastructure necessary for a client to
1499 create a builder. That builder is then passed down to @code{libgdb}
1500 when doing any queries.
1502 @subheading Event Loop - @file{event-loop.h}
1503 @c There could be an entire section on the event-loop
1504 @file{event-loop}, currently non-re-entrant, provides a simple event
1505 loop. A client would need to either plug its self into this loop or,
1506 implement a new event-loop that GDB would use.
1508 The event-loop will eventually be made re-entrant. This is so that
1509 @value{GDB} can better handle the problem of some commands blocking
1510 instead of returning.
1512 @subheading Library - @file{gdb.h}
1513 @file{libgdb} is the most obvious component of this system. It provides
1514 the query interface. Each function is parameterized by a @code{ui-out}
1515 builder. The result of the query is constructed using that builder
1516 before the query function returns.
1518 @node Symbol Handling
1520 @chapter Symbol Handling
1522 Symbols are a key part of @value{GDBN}'s operation. Symbols include variables,
1523 functions, and types.
1525 @section Symbol Reading
1527 @cindex symbol reading
1528 @cindex reading of symbols
1529 @cindex symbol files
1530 @value{GDBN} reads symbols from @dfn{symbol files}. The usual symbol
1531 file is the file containing the program which @value{GDBN} is
1532 debugging. @value{GDBN} can be directed to use a different file for
1533 symbols (with the @samp{symbol-file} command), and it can also read
1534 more symbols via the @samp{add-file} and @samp{load} commands, or while
1535 reading symbols from shared libraries.
1537 @findex find_sym_fns
1538 Symbol files are initially opened by code in @file{symfile.c} using
1539 the BFD library (@pxref{Support Libraries}). BFD identifies the type
1540 of the file by examining its header. @code{find_sym_fns} then uses
1541 this identification to locate a set of symbol-reading functions.
1543 @findex add_symtab_fns
1544 @cindex @code{sym_fns} structure
1545 @cindex adding a symbol-reading module
1546 Symbol-reading modules identify themselves to @value{GDBN} by calling
1547 @code{add_symtab_fns} during their module initialization. The argument
1548 to @code{add_symtab_fns} is a @code{struct sym_fns} which contains the
1549 name (or name prefix) of the symbol format, the length of the prefix,
1550 and pointers to four functions. These functions are called at various
1551 times to process symbol files whose identification matches the specified
1554 The functions supplied by each module are:
1557 @item @var{xyz}_symfile_init(struct sym_fns *sf)
1559 @cindex secondary symbol file
1560 Called from @code{symbol_file_add} when we are about to read a new
1561 symbol file. This function should clean up any internal state (possibly
1562 resulting from half-read previous files, for example) and prepare to
1563 read a new symbol file. Note that the symbol file which we are reading
1564 might be a new ``main'' symbol file, or might be a secondary symbol file
1565 whose symbols are being added to the existing symbol table.
1567 The argument to @code{@var{xyz}_symfile_init} is a newly allocated
1568 @code{struct sym_fns} whose @code{bfd} field contains the BFD for the
1569 new symbol file being read. Its @code{private} field has been zeroed,
1570 and can be modified as desired. Typically, a struct of private
1571 information will be @code{malloc}'d, and a pointer to it will be placed
1572 in the @code{private} field.
1574 There is no result from @code{@var{xyz}_symfile_init}, but it can call
1575 @code{error} if it detects an unavoidable problem.
1577 @item @var{xyz}_new_init()
1579 Called from @code{symbol_file_add} when discarding existing symbols.
1580 This function needs only handle the symbol-reading module's internal
1581 state; the symbol table data structures visible to the rest of
1582 @value{GDBN} will be discarded by @code{symbol_file_add}. It has no
1583 arguments and no result. It may be called after
1584 @code{@var{xyz}_symfile_init}, if a new symbol table is being read, or
1585 may be called alone if all symbols are simply being discarded.
1587 @item @var{xyz}_symfile_read(struct sym_fns *sf, CORE_ADDR addr, int mainline)
1589 Called from @code{symbol_file_add} to actually read the symbols from a
1590 symbol-file into a set of psymtabs or symtabs.
1592 @code{sf} points to the @code{struct sym_fns} originally passed to
1593 @code{@var{xyz}_sym_init} for possible initialization. @code{addr} is
1594 the offset between the file's specified start address and its true
1595 address in memory. @code{mainline} is 1 if this is the main symbol
1596 table being read, and 0 if a secondary symbol file (e.g. shared library
1597 or dynamically loaded file) is being read.@refill
1600 In addition, if a symbol-reading module creates psymtabs when
1601 @var{xyz}_symfile_read is called, these psymtabs will contain a pointer
1602 to a function @code{@var{xyz}_psymtab_to_symtab}, which can be called
1603 from any point in the @value{GDBN} symbol-handling code.
1606 @item @var{xyz}_psymtab_to_symtab (struct partial_symtab *pst)
1608 Called from @code{psymtab_to_symtab} (or the @code{PSYMTAB_TO_SYMTAB} macro) if
1609 the psymtab has not already been read in and had its @code{pst->symtab}
1610 pointer set. The argument is the psymtab to be fleshed-out into a
1611 symtab. Upon return, @code{pst->readin} should have been set to 1, and
1612 @code{pst->symtab} should contain a pointer to the new corresponding symtab, or
1613 zero if there were no symbols in that part of the symbol file.
1616 @section Partial Symbol Tables
1618 @value{GDBN} has three types of symbol tables:
1621 @cindex full symbol table
1624 Full symbol tables (@dfn{symtabs}). These contain the main
1625 information about symbols and addresses.
1629 Partial symbol tables (@dfn{psymtabs}). These contain enough
1630 information to know when to read the corresponding part of the full
1633 @cindex minimal symbol table
1636 Minimal symbol tables (@dfn{msymtabs}). These contain information
1637 gleaned from non-debugging symbols.
1640 @cindex partial symbol table
1641 This section describes partial symbol tables.
1643 A psymtab is constructed by doing a very quick pass over an executable
1644 file's debugging information. Small amounts of information are
1645 extracted---enough to identify which parts of the symbol table will
1646 need to be re-read and fully digested later, when the user needs the
1647 information. The speed of this pass causes @value{GDBN} to start up very
1648 quickly. Later, as the detailed rereading occurs, it occurs in small
1649 pieces, at various times, and the delay therefrom is mostly invisible to
1651 @c (@xref{Symbol Reading}.)
1653 The symbols that show up in a file's psymtab should be, roughly, those
1654 visible to the debugger's user when the program is not running code from
1655 that file. These include external symbols and types, static symbols and
1656 types, and @code{enum} values declared at file scope.
1658 The psymtab also contains the range of instruction addresses that the
1659 full symbol table would represent.
1661 @cindex finding a symbol
1662 @cindex symbol lookup
1663 The idea is that there are only two ways for the user (or much of the
1664 code in the debugger) to reference a symbol:
1667 @findex find_pc_function
1668 @findex find_pc_line
1670 By its address (e.g. execution stops at some address which is inside a
1671 function in this file). The address will be noticed to be in the
1672 range of this psymtab, and the full symtab will be read in.
1673 @code{find_pc_function}, @code{find_pc_line}, and other
1674 @code{find_pc_@dots{}} functions handle this.
1676 @cindex lookup_symbol
1679 (e.g. the user asks to print a variable, or set a breakpoint on a
1680 function). Global names and file-scope names will be found in the
1681 psymtab, which will cause the symtab to be pulled in. Local names will
1682 have to be qualified by a global name, or a file-scope name, in which
1683 case we will have already read in the symtab as we evaluated the
1684 qualifier. Or, a local symbol can be referenced when we are ``in'' a
1685 local scope, in which case the first case applies. @code{lookup_symbol}
1686 does most of the work here.
1689 The only reason that psymtabs exist is to cause a symtab to be read in
1690 at the right moment. Any symbol that can be elided from a psymtab,
1691 while still causing that to happen, should not appear in it. Since
1692 psymtabs don't have the idea of scope, you can't put local symbols in
1693 them anyway. Psymtabs don't have the idea of the type of a symbol,
1694 either, so types need not appear, unless they will be referenced by
1697 It is a bug for @value{GDBN} to behave one way when only a psymtab has
1698 been read, and another way if the corresponding symtab has been read
1699 in. Such bugs are typically caused by a psymtab that does not contain
1700 all the visible symbols, or which has the wrong instruction address
1703 The psymtab for a particular section of a symbol file (objfile) could be
1704 thrown away after the symtab has been read in. The symtab should always
1705 be searched before the psymtab, so the psymtab will never be used (in a
1706 bug-free environment). Currently, psymtabs are allocated on an obstack,
1707 and all the psymbols themselves are allocated in a pair of large arrays
1708 on an obstack, so there is little to be gained by trying to free them
1709 unless you want to do a lot more work.
1713 @unnumberedsubsec Fundamental Types (e.g., @code{FT_VOID}, @code{FT_BOOLEAN}).
1715 @cindex fundamental types
1716 These are the fundamental types that @value{GDBN} uses internally. Fundamental
1717 types from the various debugging formats (stabs, ELF, etc) are mapped
1718 into one of these. They are basically a union of all fundamental types
1719 that @value{GDBN} knows about for all the languages that @value{GDBN}
1722 @unnumberedsubsec Type Codes (e.g., @code{TYPE_CODE_PTR}, @code{TYPE_CODE_ARRAY}).
1725 Each time @value{GDBN} builds an internal type, it marks it with one
1726 of these types. The type may be a fundamental type, such as
1727 @code{TYPE_CODE_INT}, or a derived type, such as @code{TYPE_CODE_PTR}
1728 which is a pointer to another type. Typically, several @code{FT_*}
1729 types map to one @code{TYPE_CODE_*} type, and are distinguished by
1730 other members of the type struct, such as whether the type is signed
1731 or unsigned, and how many bits it uses.
1733 @unnumberedsubsec Builtin Types (e.g., @code{builtin_type_void}, @code{builtin_type_char}).
1735 These are instances of type structs that roughly correspond to
1736 fundamental types and are created as global types for @value{GDBN} to
1737 use for various ugly historical reasons. We eventually want to
1738 eliminate these. Note for example that @code{builtin_type_int}
1739 initialized in @file{gdbtypes.c} is basically the same as a
1740 @code{TYPE_CODE_INT} type that is initialized in @file{c-lang.c} for
1741 an @code{FT_INTEGER} fundamental type. The difference is that the
1742 @code{builtin_type} is not associated with any particular objfile, and
1743 only one instance exists, while @file{c-lang.c} builds as many
1744 @code{TYPE_CODE_INT} types as needed, with each one associated with
1745 some particular objfile.
1747 @section Object File Formats
1748 @cindex object file formats
1752 @cindex @code{a.out} format
1753 The @code{a.out} format is the original file format for Unix. It
1754 consists of three sections: @code{text}, @code{data}, and @code{bss},
1755 which are for program code, initialized data, and uninitialized data,
1758 The @code{a.out} format is so simple that it doesn't have any reserved
1759 place for debugging information. (Hey, the original Unix hackers used
1760 @samp{adb}, which is a machine-language debugger!) The only debugging
1761 format for @code{a.out} is stabs, which is encoded as a set of normal
1762 symbols with distinctive attributes.
1764 The basic @code{a.out} reader is in @file{dbxread.c}.
1769 The COFF format was introduced with System V Release 3 (SVR3) Unix.
1770 COFF files may have multiple sections, each prefixed by a header. The
1771 number of sections is limited.
1773 The COFF specification includes support for debugging. Although this
1774 was a step forward, the debugging information was woefully limited. For
1775 instance, it was not possible to represent code that came from an
1778 The COFF reader is in @file{coffread.c}.
1782 @cindex ECOFF format
1783 ECOFF is an extended COFF originally introduced for Mips and Alpha
1786 The basic ECOFF reader is in @file{mipsread.c}.
1790 @cindex XCOFF format
1791 The IBM RS/6000 running AIX uses an object file format called XCOFF.
1792 The COFF sections, symbols, and line numbers are used, but debugging
1793 symbols are @code{dbx}-style stabs whose strings are located in the
1794 @code{.debug} section (rather than the string table). For more
1795 information, see @ref{Top,,,stabs,The Stabs Debugging Format}.
1797 The shared library scheme has a clean interface for figuring out what
1798 shared libraries are in use, but the catch is that everything which
1799 refers to addresses (symbol tables and breakpoints at least) needs to be
1800 relocated for both shared libraries and the main executable. At least
1801 using the standard mechanism this can only be done once the program has
1802 been run (or the core file has been read).
1806 @cindex PE-COFF format
1807 Windows 95 and NT use the PE (@dfn{Portable Executable}) format for their
1808 executables. PE is basically COFF with additional headers.
1810 While BFD includes special PE support, @value{GDBN} needs only the basic
1816 The ELF format came with System V Release 4 (SVR4) Unix. ELF is similar
1817 to COFF in being organized into a number of sections, but it removes
1818 many of COFF's limitations.
1820 The basic ELF reader is in @file{elfread.c}.
1825 SOM is HP's object file and debug format (not to be confused with IBM's
1826 SOM, which is a cross-language ABI).
1828 The SOM reader is in @file{hpread.c}.
1830 @subsection Other File Formats
1832 @cindex Netware Loadable Module format
1833 Other file formats that have been supported by @value{GDBN} include Netware
1834 Loadable Modules (@file{nlmread.c}).
1836 @section Debugging File Formats
1838 This section describes characteristics of debugging information that
1839 are independent of the object file format.
1843 @cindex stabs debugging info
1844 @code{stabs} started out as special symbols within the @code{a.out}
1845 format. Since then, it has been encapsulated into other file
1846 formats, such as COFF and ELF.
1848 While @file{dbxread.c} does some of the basic stab processing,
1849 including for encapsulated versions, @file{stabsread.c} does
1854 @cindex COFF debugging info
1855 The basic COFF definition includes debugging information. The level
1856 of support is minimal and non-extensible, and is not often used.
1858 @subsection Mips debug (Third Eye)
1860 @cindex ECOFF debugging info
1861 ECOFF includes a definition of a special debug format.
1863 The file @file{mdebugread.c} implements reading for this format.
1867 @cindex DWARF 1 debugging info
1868 DWARF 1 is a debugging format that was originally designed to be
1869 used with ELF in SVR4 systems.
1874 @c If defined, these are the producer strings in a DWARF 1 file. All of
1875 @c these have reasonable defaults already.
1877 The DWARF 1 reader is in @file{dwarfread.c}.
1881 @cindex DWARF 2 debugging info
1882 DWARF 2 is an improved but incompatible version of DWARF 1.
1884 The DWARF 2 reader is in @file{dwarf2read.c}.
1888 @cindex SOM debugging info
1889 Like COFF, the SOM definition includes debugging information.
1891 @section Adding a New Symbol Reader to @value{GDBN}
1893 @cindex adding debugging info reader
1894 If you are using an existing object file format (@code{a.out}, COFF, ELF, etc),
1895 there is probably little to be done.
1897 If you need to add a new object file format, you must first add it to
1898 BFD. This is beyond the scope of this document.
1900 You must then arrange for the BFD code to provide access to the
1901 debugging symbols. Generally @value{GDBN} will have to call swapping routines
1902 from BFD and a few other BFD internal routines to locate the debugging
1903 information. As much as possible, @value{GDBN} should not depend on the BFD
1904 internal data structures.
1906 For some targets (e.g., COFF), there is a special transfer vector used
1907 to call swapping routines, since the external data structures on various
1908 platforms have different sizes and layouts. Specialized routines that
1909 will only ever be implemented by one object file format may be called
1910 directly. This interface should be described in a file
1911 @file{bfd/lib@var{xyz}.h}, which is included by @value{GDBN}.
1914 @node Language Support
1916 @chapter Language Support
1918 @cindex language support
1919 @value{GDBN}'s language support is mainly driven by the symbol reader,
1920 although it is possible for the user to set the source language
1923 @value{GDBN} chooses the source language by looking at the extension
1924 of the file recorded in the debug info; @file{.c} means C, @file{.f}
1925 means Fortran, etc. It may also use a special-purpose language
1926 identifier if the debug format supports it, like with DWARF.
1928 @section Adding a Source Language to @value{GDBN}
1930 @cindex adding source language
1931 To add other languages to @value{GDBN}'s expression parser, follow the
1935 @item Create the expression parser.
1937 @cindex expression parser
1938 This should reside in a file @file{@var{lang}-exp.y}. Routines for
1939 building parsed expressions into a @code{union exp_element} list are in
1942 @cindex language parser
1943 Since we can't depend upon everyone having Bison, and YACC produces
1944 parsers that define a bunch of global names, the following lines
1945 @strong{must} be included at the top of the YACC parser, to prevent the
1946 various parsers from defining the same global names:
1949 #define yyparse @var{lang}_parse
1950 #define yylex @var{lang}_lex
1951 #define yyerror @var{lang}_error
1952 #define yylval @var{lang}_lval
1953 #define yychar @var{lang}_char
1954 #define yydebug @var{lang}_debug
1955 #define yypact @var{lang}_pact
1956 #define yyr1 @var{lang}_r1
1957 #define yyr2 @var{lang}_r2
1958 #define yydef @var{lang}_def
1959 #define yychk @var{lang}_chk
1960 #define yypgo @var{lang}_pgo
1961 #define yyact @var{lang}_act
1962 #define yyexca @var{lang}_exca
1963 #define yyerrflag @var{lang}_errflag
1964 #define yynerrs @var{lang}_nerrs
1967 At the bottom of your parser, define a @code{struct language_defn} and
1968 initialize it with the right values for your language. Define an
1969 @code{initialize_@var{lang}} routine and have it call
1970 @samp{add_language(@var{lang}_language_defn)} to tell the rest of @value{GDBN}
1971 that your language exists. You'll need some other supporting variables
1972 and functions, which will be used via pointers from your
1973 @code{@var{lang}_language_defn}. See the declaration of @code{struct
1974 language_defn} in @file{language.h}, and the other @file{*-exp.y} files,
1975 for more information.
1977 @item Add any evaluation routines, if necessary
1979 @cindex expression evaluation routines
1980 @findex evaluate_subexp
1981 @findex prefixify_subexp
1982 @findex length_of_subexp
1983 If you need new opcodes (that represent the operations of the language),
1984 add them to the enumerated type in @file{expression.h}. Add support
1985 code for these operations in the @code{evaluate_subexp} function
1986 defined in the file @file{eval.c}. Add cases
1987 for new opcodes in two functions from @file{parse.c}:
1988 @code{prefixify_subexp} and @code{length_of_subexp}. These compute
1989 the number of @code{exp_element}s that a given operation takes up.
1991 @item Update some existing code
1993 Add an enumerated identifier for your language to the enumerated type
1994 @code{enum language} in @file{defs.h}.
1996 Update the routines in @file{language.c} so your language is included.
1997 These routines include type predicates and such, which (in some cases)
1998 are language dependent. If your language does not appear in the switch
1999 statement, an error is reported.
2001 @vindex current_language
2002 Also included in @file{language.c} is the code that updates the variable
2003 @code{current_language}, and the routines that translate the
2004 @code{language_@var{lang}} enumerated identifier into a printable
2007 @findex _initialize_language
2008 Update the function @code{_initialize_language} to include your
2009 language. This function picks the default language upon startup, so is
2010 dependent upon which languages that @value{GDBN} is built for.
2012 @findex allocate_symtab
2013 Update @code{allocate_symtab} in @file{symfile.c} and/or symbol-reading
2014 code so that the language of each symtab (source file) is set properly.
2015 This is used to determine the language to use at each stack frame level.
2016 Currently, the language is set based upon the extension of the source
2017 file. If the language can be better inferred from the symbol
2018 information, please set the language of the symtab in the symbol-reading
2021 @findex print_subexp
2022 @findex op_print_tab
2023 Add helper code to @code{print_subexp} (in @file{expprint.c}) to handle any new
2024 expression opcodes you have added to @file{expression.h}. Also, add the
2025 printed representations of your operators to @code{op_print_tab}.
2027 @item Add a place of call
2030 Add a call to @code{@var{lang}_parse()} and @code{@var{lang}_error} in
2031 @code{parse_exp_1} (defined in @file{parse.c}).
2033 @item Use macros to trim code
2035 @cindex trimming language-dependent code
2036 The user has the option of building @value{GDBN} for some or all of the
2037 languages. If the user decides to build @value{GDBN} for the language
2038 @var{lang}, then every file dependent on @file{language.h} will have the
2039 macro @code{_LANG_@var{lang}} defined in it. Use @code{#ifdef}s to
2040 leave out large routines that the user won't need if he or she is not
2041 using your language.
2043 Note that you do not need to do this in your YACC parser, since if @value{GDBN}
2044 is not build for @var{lang}, then @file{@var{lang}-exp.tab.o} (the
2045 compiled form of your parser) is not linked into @value{GDBN} at all.
2047 See the file @file{configure.in} for how @value{GDBN} is configured
2048 for different languages.
2050 @item Edit @file{Makefile.in}
2052 Add dependencies in @file{Makefile.in}. Make sure you update the macro
2053 variables such as @code{HFILES} and @code{OBJS}, otherwise your code may
2054 not get linked in, or, worse yet, it may not get @code{tar}red into the
2059 @node Host Definition
2061 @chapter Host Definition
2063 With the advent of Autoconf, it's rarely necessary to have host
2064 definition machinery anymore. The following information is provided,
2065 mainly, as an historical reference.
2067 @section Adding a New Host
2069 @cindex adding a new host
2070 @cindex host, adding
2071 @value{GDBN}'s host configuration support normally happens via Autoconf.
2072 New host-specific definitions should not be needed. Older hosts
2073 @value{GDBN} still use the host-specific definitions and files listed
2074 below, but these mostly exist for historical reasons, and will
2075 eventually disappear.
2078 @item gdb/config/@var{arch}/@var{xyz}.mh
2079 This file once contained both host and native configuration information
2080 (@pxref{Native Debugging}) for the machine @var{xyz}. The host
2081 configuration information is now handed by Autoconf.
2083 Host configuration information included a definition of
2084 @code{XM_FILE=xm-@var{xyz}.h} and possibly definitions for @code{CC},
2085 @code{SYSV_DEFINE}, @code{XM_CFLAGS}, @code{XM_ADD_FILES},
2086 @code{XM_CLIBS}, @code{XM_CDEPS}, etc.; see @file{Makefile.in}.
2088 New host only configurations do not need this file.
2090 @item gdb/config/@var{arch}/xm-@var{xyz}.h
2091 This file once contained definitions and includes required when hosting
2092 gdb on machine @var{xyz}. Those definitions and includes are now
2093 handled by Autoconf.
2095 New host and native configurations do not need this file.
2097 @emph{Maintainer's note: Some hosts continue to use the @file{xm-xyz.h}
2098 file to define the macros @var{HOST_FLOAT_FORMAT},
2099 @var{HOST_DOUBLE_FORMAT} and @var{HOST_LONG_DOUBLE_FORMAT}. That code
2100 also needs to be replaced with either an Autoconf or run-time test.}
2104 @subheading Generic Host Support Files
2106 @cindex generic host support
2107 There are some ``generic'' versions of routines that can be used by
2108 various systems. These can be customized in various ways by macros
2109 defined in your @file{xm-@var{xyz}.h} file. If these routines work for
2110 the @var{xyz} host, you can just include the generic file's name (with
2111 @samp{.o}, not @samp{.c}) in @code{XDEPFILES}.
2113 Otherwise, if your machine needs custom support routines, you will need
2114 to write routines that perform the same functions as the generic file.
2115 Put them into @code{@var{xyz}-xdep.c}, and put @code{@var{xyz}-xdep.o}
2116 into @code{XDEPFILES}.
2119 @cindex remote debugging support
2120 @cindex serial line support
2122 This contains serial line support for Unix systems. This is always
2123 included, via the makefile variable @code{SER_HARDWIRE}; override this
2124 variable in the @file{.mh} file to avoid it.
2127 This contains serial line support for 32-bit programs running under DOS,
2128 using the DJGPP (a.k.a.@: GO32) execution environment.
2130 @cindex TCP remote support
2132 This contains generic TCP support using sockets.
2135 @section Host Conditionals
2137 When @value{GDBN} is configured and compiled, various macros are
2138 defined or left undefined, to control compilation based on the
2139 attributes of the host system. These macros and their meanings (or if
2140 the meaning is not documented here, then one of the source files where
2141 they are used is indicated) are:
2144 @item @value{GDBN}INIT_FILENAME
2145 The default name of @value{GDBN}'s initialization file (normally
2149 This macro is deprecated.
2152 Define this if your system does not have a @code{<sys/file.h>}.
2154 @item SIGWINCH_HANDLER
2155 If your host defines @code{SIGWINCH}, you can define this to be the name
2156 of a function to be called if @code{SIGWINCH} is received.
2158 @item SIGWINCH_HANDLER_BODY
2159 Define this to expand into code that will define the function named by
2160 the expansion of @code{SIGWINCH_HANDLER}.
2162 @item ALIGN_STACK_ON_STARTUP
2163 @cindex stack alignment
2164 Define this if your system is of a sort that will crash in
2165 @code{tgetent} if the stack happens not to be longword-aligned when
2166 @code{main} is called. This is a rare situation, but is known to occur
2167 on several different types of systems.
2169 @item CRLF_SOURCE_FILES
2170 @cindex DOS text files
2171 Define this if host files use @code{\r\n} rather than @code{\n} as a
2172 line terminator. This will cause source file listings to omit @code{\r}
2173 characters when printing and it will allow @code{\r\n} line endings of files
2174 which are ``sourced'' by gdb. It must be possible to open files in binary
2175 mode using @code{O_BINARY} or, for fopen, @code{"rb"}.
2177 @item DEFAULT_PROMPT
2179 The default value of the prompt string (normally @code{"(gdb) "}).
2182 @cindex terminal device
2183 The name of the generic TTY device, defaults to @code{"/dev/tty"}.
2185 @item FCLOSE_PROVIDED
2186 Define this if the system declares @code{fclose} in the headers included
2187 in @code{defs.h}. This isn't needed unless your compiler is unusually
2191 Define this if binary files are opened the same way as text files.
2193 @item GETENV_PROVIDED
2194 Define this if the system declares @code{getenv} in its headers included
2195 in @code{defs.h}. This isn't needed unless your compiler is unusually
2200 In some cases, use the system call @code{mmap} for reading symbol
2201 tables. For some machines this allows for sharing and quick updates.
2204 Define this if the host system has @code{termio.h}.
2211 Values for host-side constants.
2214 Substitute for isatty, if not available.
2217 This is the longest integer type available on the host. If not defined,
2218 it will default to @code{long long} or @code{long}, depending on
2219 @code{CC_HAS_LONG_LONG}.
2221 @item CC_HAS_LONG_LONG
2222 @cindex @code{long long} data type
2223 Define this if the host C compiler supports @code{long long}. This is set
2224 by the @code{configure} script.
2226 @item PRINTF_HAS_LONG_LONG
2227 Define this if the host can handle printing of long long integers via
2228 the printf format conversion specifier @code{ll}. This is set by the
2229 @code{configure} script.
2231 @item HAVE_LONG_DOUBLE
2232 Define this if the host C compiler supports @code{long double}. This is
2233 set by the @code{configure} script.
2235 @item PRINTF_HAS_LONG_DOUBLE
2236 Define this if the host can handle printing of long double float-point
2237 numbers via the printf format conversion specifier @code{Lg}. This is
2238 set by the @code{configure} script.
2240 @item SCANF_HAS_LONG_DOUBLE
2241 Define this if the host can handle the parsing of long double
2242 float-point numbers via the scanf format conversion specifier
2243 @code{Lg}. This is set by the @code{configure} script.
2245 @item LSEEK_NOT_LINEAR
2246 Define this if @code{lseek (n)} does not necessarily move to byte number
2247 @code{n} in the file. This is only used when reading source files. It
2248 is normally faster to define @code{CRLF_SOURCE_FILES} when possible.
2251 This macro is used as the argument to @code{lseek} (or, most commonly,
2252 @code{bfd_seek}). FIXME, should be replaced by SEEK_SET instead,
2253 which is the POSIX equivalent.
2255 @item MMAP_BASE_ADDRESS
2256 When using HAVE_MMAP, the first mapping should go at this address.
2258 @item MMAP_INCREMENT
2259 when using HAVE_MMAP, this is the increment between mappings.
2262 If defined, this should be one or more tokens, such as @code{volatile},
2263 that can be used in both the declaration and definition of functions to
2264 indicate that they never return. The default is already set correctly
2265 if compiling with GCC. This will almost never need to be defined.
2268 If defined, this should be one or more tokens, such as
2269 @code{__attribute__ ((noreturn))}, that can be used in the declarations
2270 of functions to indicate that they never return. The default is already
2271 set correctly if compiling with GCC. This will almost never need to be
2276 @value{GDBN} will use the @code{mmalloc} library for memory allocation
2277 for symbol reading if this symbol is defined. Be careful defining it
2278 since there are systems on which @code{mmalloc} does not work for some
2279 reason. One example is the DECstation, where its RPC library can't
2280 cope with our redefinition of @code{malloc} to call @code{mmalloc}.
2281 When defining @code{USE_MMALLOC}, you will also have to set
2282 @code{MMALLOC} in the Makefile, to point to the @code{mmalloc} library. This
2283 define is set when you configure with @samp{--with-mmalloc}.
2287 Define this if you are using @code{mmalloc}, but don't want the overhead
2288 of checking the heap with @code{mmcheck}. Note that on some systems,
2289 the C runtime makes calls to @code{malloc} prior to calling @code{main}, and if
2290 @code{free} is ever called with these pointers after calling
2291 @code{mmcheck} to enable checking, a memory corruption abort is certain
2292 to occur. These systems can still use @code{mmalloc}, but must define
2296 Define this to 1 if the C runtime allocates memory prior to
2297 @code{mmcheck} being called, but that memory is never freed so we don't
2298 have to worry about it triggering a memory corruption abort. The
2299 default is 0, which means that @code{mmcheck} will only install the heap
2300 checking functions if there has not yet been any memory allocation
2301 calls, and if it fails to install the functions, @value{GDBN} will issue a
2302 warning. This is currently defined if you configure using
2303 @samp{--with-mmalloc}.
2305 @item NO_SIGINTERRUPT
2306 @findex siginterrupt
2307 Define this to indicate that @code{siginterrupt} is not available.
2311 Define these to appropriate value for the system @code{lseek}, if not already
2315 This is the signal for stopping @value{GDBN}. Defaults to
2316 @code{SIGTSTP}. (Only redefined for the Convex.)
2319 Define this if the interior's tty should be opened with the @code{O_NOCTTY}
2320 flag. (FIXME: This should be a native-only flag, but @file{inflow.c} is
2324 Means that System V (prior to SVR4) include files are in use. (FIXME:
2325 This symbol is abused in @file{infrun.c}, @file{regex.c},
2326 @file{remote-nindy.c}, and @file{utils.c} for other things, at the
2330 Define this to help placate @code{lint} in some situations.
2333 Define this to override the defaults of @code{__volatile__} or
2338 @node Target Architecture Definition
2340 @chapter Target Architecture Definition
2342 @cindex target architecture definition
2343 @value{GDBN}'s target architecture defines what sort of
2344 machine-language programs @value{GDBN} can work with, and how it works
2347 The target architecture object is implemented as the C structure
2348 @code{struct gdbarch *}. The structure, and its methods, are generated
2349 using the Bourne shell script @file{gdbarch.sh}.
2351 @section Operating System ABI Variant Handling
2352 @cindex OS ABI variants
2354 @value{GDBN} provides a mechanism for handling variations in OS
2355 ABIs. An OS ABI variant may have influence over any number of
2356 variables in the target architecture definition. There are two major
2357 components in the OS ABI mechanism: sniffers and handlers.
2359 A @dfn{sniffer} examines a file matching a BFD architecture/flavour pair
2360 (the architecture may be wildcarded) in an attempt to determine the
2361 OS ABI of that file. Sniffers with a wildcarded architecture are considered
2362 to be @dfn{generic}, while sniffers for a specific architecture are
2363 considered to be @dfn{specific}. A match from a specific sniffer
2364 overrides a match from a generic sniffer. Multiple sniffers for an
2365 architecture/flavour may exist, in order to differentiate between two
2366 different operating systems which use the same basic file format. The
2367 OS ABI framework provides a generic sniffer for ELF-format files which
2368 examines the @code{EI_OSABI} field of the ELF header, as well as note
2369 sections known to be used by several operating systems.
2371 @cindex fine-tuning @code{gdbarch} structure
2372 A @dfn{handler} is used to fine-tune the @code{gdbarch} structure for the
2373 selected OS ABI. There may be only one handler for a given OS ABI
2374 for each BFD architecture.
2376 The following OS ABI variants are defined in @file{osabi.h}:
2380 @findex GDB_OSABI_UNKNOWN
2381 @item GDB_OSABI_UNKNOWN
2382 The ABI of the inferior is unknown. The default @code{gdbarch}
2383 settings for the architecture will be used.
2385 @findex GDB_OSABI_SVR4
2386 @item GDB_OSABI_SVR4
2387 UNIX System V Release 4
2389 @findex GDB_OSABI_HURD
2390 @item GDB_OSABI_HURD
2391 GNU using the Hurd kernel
2393 @findex GDB_OSABI_SOLARIS
2394 @item GDB_OSABI_SOLARIS
2397 @findex GDB_OSABI_OSF1
2398 @item GDB_OSABI_OSF1
2399 OSF/1, including Digital UNIX and Compaq Tru64 UNIX
2401 @findex GDB_OSABI_LINUX
2402 @item GDB_OSABI_LINUX
2403 GNU using the Linux kernel
2405 @findex GDB_OSABI_FREEBSD_AOUT
2406 @item GDB_OSABI_FREEBSD_AOUT
2407 FreeBSD using the a.out executable format
2409 @findex GDB_OSABI_FREEBSD_ELF
2410 @item GDB_OSABI_FREEBSD_ELF
2411 FreeBSD using the ELF executable format
2413 @findex GDB_OSABI_NETBSD_AOUT
2414 @item GDB_OSABI_NETBSD_AOUT
2415 NetBSD using the a.out executable format
2417 @findex GDB_OSABI_NETBSD_ELF
2418 @item GDB_OSABI_NETBSD_ELF
2419 NetBSD using the ELF executable format
2421 @findex GDB_OSABI_WINCE
2422 @item GDB_OSABI_WINCE
2425 @findex GDB_OSABI_GO32
2426 @item GDB_OSABI_GO32
2429 @findex GDB_OSABI_NETWARE
2430 @item GDB_OSABI_NETWARE
2433 @findex GDB_OSABI_ARM_EABI_V1
2434 @item GDB_OSABI_ARM_EABI_V1
2435 ARM Embedded ABI version 1
2437 @findex GDB_OSABI_ARM_EABI_V2
2438 @item GDB_OSABI_ARM_EABI_V2
2439 ARM Embedded ABI version 2
2441 @findex GDB_OSABI_ARM_APCS
2442 @item GDB_OSABI_ARM_APCS
2443 Generic ARM Procedure Call Standard
2447 Here are the functions that make up the OS ABI framework:
2449 @deftypefun const char *gdbarch_osabi_name (enum gdb_osabi @var{osabi})
2450 Return the name of the OS ABI corresponding to @var{osabi}.
2453 @deftypefun void gdbarch_register_osabi (enum bfd_architecture @var{arch}, unsigned long @var{machine}, enum gdb_osabi @var{osabi}, void (*@var{init_osabi})(struct gdbarch_info @var{info}, struct gdbarch *@var{gdbarch}))
2454 Register the OS ABI handler specified by @var{init_osabi} for the
2455 architecture, machine type and OS ABI specified by @var{arch},
2456 @var{machine} and @var{osabi}. In most cases, a value of zero for the
2457 machine type, which implies the architecture's default machine type,
2461 @deftypefun void gdbarch_register_osabi_sniffer (enum bfd_architecture @var{arch}, enum bfd_flavour @var{flavour}, enum gdb_osabi (*@var{sniffer})(bfd *@var{abfd}))
2462 Register the OS ABI file sniffer specified by @var{sniffer} for the
2463 BFD architecture/flavour pair specified by @var{arch} and @var{flavour}.
2464 If @var{arch} is @code{bfd_arch_unknown}, the sniffer is considered to
2465 be generic, and is allowed to examine @var{flavour}-flavoured files for
2469 @deftypefun enum gdb_osabi gdbarch_lookup_osabi (bfd *@var{abfd})
2470 Examine the file described by @var{abfd} to determine its OS ABI.
2471 The value @code{GDB_OSABI_UNKNOWN} is returned if the OS ABI cannot
2475 @deftypefun void gdbarch_init_osabi (struct gdbarch info @var{info}, struct gdbarch *@var{gdbarch}, enum gdb_osabi @var{osabi})
2476 Invoke the OS ABI handler corresponding to @var{osabi} to fine-tune the
2477 @code{gdbarch} structure specified by @var{gdbarch}. If a handler
2478 corresponding to @var{osabi} has not been registered for @var{gdbarch}'s
2479 architecture, a warning will be issued and the debugging session will continue
2480 with the defaults already established for @var{gdbarch}.
2483 @section Registers and Memory
2485 @value{GDBN}'s model of the target machine is rather simple.
2486 @value{GDBN} assumes the machine includes a bank of registers and a
2487 block of memory. Each register may have a different size.
2489 @value{GDBN} does not have a magical way to match up with the
2490 compiler's idea of which registers are which; however, it is critical
2491 that they do match up accurately. The only way to make this work is
2492 to get accurate information about the order that the compiler uses,
2493 and to reflect that in the @code{REGISTER_NAME} and related macros.
2495 @value{GDBN} can handle big-endian, little-endian, and bi-endian architectures.
2497 @section Pointers Are Not Always Addresses
2498 @cindex pointer representation
2499 @cindex address representation
2500 @cindex word-addressed machines
2501 @cindex separate data and code address spaces
2502 @cindex spaces, separate data and code address
2503 @cindex address spaces, separate data and code
2504 @cindex code pointers, word-addressed
2505 @cindex converting between pointers and addresses
2506 @cindex D10V addresses
2508 On almost all 32-bit architectures, the representation of a pointer is
2509 indistinguishable from the representation of some fixed-length number
2510 whose value is the byte address of the object pointed to. On such
2511 machines, the words ``pointer'' and ``address'' can be used interchangeably.
2512 However, architectures with smaller word sizes are often cramped for
2513 address space, so they may choose a pointer representation that breaks this
2514 identity, and allows a larger code address space.
2516 For example, the Mitsubishi D10V is a 16-bit VLIW processor whose
2517 instructions are 32 bits long@footnote{Some D10V instructions are
2518 actually pairs of 16-bit sub-instructions. However, since you can't
2519 jump into the middle of such a pair, code addresses can only refer to
2520 full 32 bit instructions, which is what matters in this explanation.}.
2521 If the D10V used ordinary byte addresses to refer to code locations,
2522 then the processor would only be able to address 64kb of instructions.
2523 However, since instructions must be aligned on four-byte boundaries, the
2524 low two bits of any valid instruction's byte address are always
2525 zero---byte addresses waste two bits. So instead of byte addresses,
2526 the D10V uses word addresses---byte addresses shifted right two bits---to
2527 refer to code. Thus, the D10V can use 16-bit words to address 256kb of
2530 However, this means that code pointers and data pointers have different
2531 forms on the D10V. The 16-bit word @code{0xC020} refers to byte address
2532 @code{0xC020} when used as a data address, but refers to byte address
2533 @code{0x30080} when used as a code address.
2535 (The D10V also uses separate code and data address spaces, which also
2536 affects the correspondence between pointers and addresses, but we're
2537 going to ignore that here; this example is already too long.)
2539 To cope with architectures like this---the D10V is not the only
2540 one!---@value{GDBN} tries to distinguish between @dfn{addresses}, which are
2541 byte numbers, and @dfn{pointers}, which are the target's representation
2542 of an address of a particular type of data. In the example above,
2543 @code{0xC020} is the pointer, which refers to one of the addresses
2544 @code{0xC020} or @code{0x30080}, depending on the type imposed upon it.
2545 @value{GDBN} provides functions for turning a pointer into an address
2546 and vice versa, in the appropriate way for the current architecture.
2548 Unfortunately, since addresses and pointers are identical on almost all
2549 processors, this distinction tends to bit-rot pretty quickly. Thus,
2550 each time you port @value{GDBN} to an architecture which does
2551 distinguish between pointers and addresses, you'll probably need to
2552 clean up some architecture-independent code.
2554 Here are functions which convert between pointers and addresses:
2556 @deftypefun CORE_ADDR extract_typed_address (void *@var{buf}, struct type *@var{type})
2557 Treat the bytes at @var{buf} as a pointer or reference of type
2558 @var{type}, and return the address it represents, in a manner
2559 appropriate for the current architecture. This yields an address
2560 @value{GDBN} can use to read target memory, disassemble, etc. Note that
2561 @var{buf} refers to a buffer in @value{GDBN}'s memory, not the
2564 For example, if the current architecture is the Intel x86, this function
2565 extracts a little-endian integer of the appropriate length from
2566 @var{buf} and returns it. However, if the current architecture is the
2567 D10V, this function will return a 16-bit integer extracted from
2568 @var{buf}, multiplied by four if @var{type} is a pointer to a function.
2570 If @var{type} is not a pointer or reference type, then this function
2571 will signal an internal error.
2574 @deftypefun CORE_ADDR store_typed_address (void *@var{buf}, struct type *@var{type}, CORE_ADDR @var{addr})
2575 Store the address @var{addr} in @var{buf}, in the proper format for a
2576 pointer of type @var{type} in the current architecture. Note that
2577 @var{buf} refers to a buffer in @value{GDBN}'s memory, not the
2580 For example, if the current architecture is the Intel x86, this function
2581 stores @var{addr} unmodified as a little-endian integer of the
2582 appropriate length in @var{buf}. However, if the current architecture
2583 is the D10V, this function divides @var{addr} by four if @var{type} is
2584 a pointer to a function, and then stores it in @var{buf}.
2586 If @var{type} is not a pointer or reference type, then this function
2587 will signal an internal error.
2590 @deftypefun CORE_ADDR value_as_address (struct value *@var{val})
2591 Assuming that @var{val} is a pointer, return the address it represents,
2592 as appropriate for the current architecture.
2594 This function actually works on integral values, as well as pointers.
2595 For pointers, it performs architecture-specific conversions as
2596 described above for @code{extract_typed_address}.
2599 @deftypefun CORE_ADDR value_from_pointer (struct type *@var{type}, CORE_ADDR @var{addr})
2600 Create and return a value representing a pointer of type @var{type} to
2601 the address @var{addr}, as appropriate for the current architecture.
2602 This function performs architecture-specific conversions as described
2603 above for @code{store_typed_address}.
2607 @value{GDBN} also provides functions that do the same tasks, but assume
2608 that pointers are simply byte addresses; they aren't sensitive to the
2609 current architecture, beyond knowing the appropriate endianness.
2611 @deftypefun CORE_ADDR extract_address (void *@var{addr}, int len)
2612 Extract a @var{len}-byte number from @var{addr} in the appropriate
2613 endianness for the current architecture, and return it. Note that
2614 @var{addr} refers to @value{GDBN}'s memory, not the inferior's.
2616 This function should only be used in architecture-specific code; it
2617 doesn't have enough information to turn bits into a true address in the
2618 appropriate way for the current architecture. If you can, use
2619 @code{extract_typed_address} instead.
2622 @deftypefun void store_address (void *@var{addr}, int @var{len}, LONGEST @var{val})
2623 Store @var{val} at @var{addr} as a @var{len}-byte integer, in the
2624 appropriate endianness for the current architecture. Note that
2625 @var{addr} refers to a buffer in @value{GDBN}'s memory, not the
2628 This function should only be used in architecture-specific code; it
2629 doesn't have enough information to turn a true address into bits in the
2630 appropriate way for the current architecture. If you can, use
2631 @code{store_typed_address} instead.
2635 Here are some macros which architectures can define to indicate the
2636 relationship between pointers and addresses. These have default
2637 definitions, appropriate for architectures on which all pointers are
2638 simple unsigned byte addresses.
2640 @deftypefn {Target Macro} CORE_ADDR POINTER_TO_ADDRESS (struct type *@var{type}, char *@var{buf})
2641 Assume that @var{buf} holds a pointer of type @var{type}, in the
2642 appropriate format for the current architecture. Return the byte
2643 address the pointer refers to.
2645 This function may safely assume that @var{type} is either a pointer or a
2646 C@t{++} reference type.
2649 @deftypefn {Target Macro} void ADDRESS_TO_POINTER (struct type *@var{type}, char *@var{buf}, CORE_ADDR @var{addr})
2650 Store in @var{buf} a pointer of type @var{type} representing the address
2651 @var{addr}, in the appropriate format for the current architecture.
2653 This function may safely assume that @var{type} is either a pointer or a
2654 C@t{++} reference type.
2657 @section Address Classes
2658 @cindex address classes
2659 @cindex DW_AT_byte_size
2660 @cindex DW_AT_address_class
2662 Sometimes information about different kinds of addresses is available
2663 via the debug information. For example, some programming environments
2664 define addresses of several different sizes. If the debug information
2665 distinguishes these kinds of address classes through either the size
2666 info (e.g, @code{DW_AT_byte_size} in @w{DWARF 2}) or through an explicit
2667 address class attribute (e.g, @code{DW_AT_address_class} in @w{DWARF 2}), the
2668 following macros should be defined in order to disambiguate these
2669 types within @value{GDBN} as well as provide the added information to
2670 a @value{GDBN} user when printing type expressions.
2672 @deftypefn {Target Macro} int ADDRESS_CLASS_TYPE_FLAGS (int @var{byte_size}, int @var{dwarf2_addr_class})
2673 Returns the type flags needed to construct a pointer type whose size
2674 is @var{byte_size} and whose address class is @var{dwarf2_addr_class}.
2675 This function is normally called from within a symbol reader. See
2676 @file{dwarf2read.c}.
2679 @deftypefn {Target Macro} char *ADDRESS_CLASS_TYPE_FLAGS_TO_NAME (int @var{type_flags})
2680 Given the type flags representing an address class qualifier, return
2683 @deftypefn {Target Macro} int ADDRESS_CLASS_NAME_to_TYPE_FLAGS (int @var{name}, int *var{type_flags_ptr})
2684 Given an address qualifier name, set the @code{int} refererenced by @var{type_flags_ptr} to the type flags
2685 for that address class qualifier.
2688 Since the need for address classes is rather rare, none of
2689 the address class macros defined by default. Predicate
2690 macros are provided to detect when they are defined.
2692 Consider a hypothetical architecture in which addresses are normally
2693 32-bits wide, but 16-bit addresses are also supported. Furthermore,
2694 suppose that the @w{DWARF 2} information for this architecture simply
2695 uses a @code{DW_AT_byte_size} value of 2 to indicate the use of one
2696 of these "short" pointers. The following functions could be defined
2697 to implement the address class macros:
2700 somearch_address_class_type_flags (int byte_size,
2701 int dwarf2_addr_class)
2704 return TYPE_FLAG_ADDRESS_CLASS_1;
2710 somearch_address_class_type_flags_to_name (int type_flags)
2712 if (type_flags & TYPE_FLAG_ADDRESS_CLASS_1)
2719 somearch_address_class_name_to_type_flags (char *name,
2720 int *type_flags_ptr)
2722 if (strcmp (name, "short") == 0)
2724 *type_flags_ptr = TYPE_FLAG_ADDRESS_CLASS_1;
2732 The qualifier @code{@@short} is used in @value{GDBN}'s type expressions
2733 to indicate the presence of one of these "short" pointers. E.g, if
2734 the debug information indicates that @code{short_ptr_var} is one of these
2735 short pointers, @value{GDBN} might show the following behavior:
2738 (gdb) ptype short_ptr_var
2739 type = int * @@short
2743 @section Raw and Virtual Register Representations
2744 @cindex raw register representation
2745 @cindex virtual register representation
2746 @cindex representations, raw and virtual registers
2748 @emph{Maintainer note: This section is pretty much obsolete. The
2749 functionality described here has largely been replaced by
2750 pseudo-registers and the mechanisms described in @ref{Target
2751 Architecture Definition, , Using Different Register and Memory Data
2752 Representations}. See also @uref{http://www.gnu.org/software/gdb/bugs/,
2753 Bug Tracking Database} and
2754 @uref{http://sources.redhat.com/gdb/current/ari/, ARI Index} for more
2755 up-to-date information.}
2757 Some architectures use one representation for a value when it lives in a
2758 register, but use a different representation when it lives in memory.
2759 In @value{GDBN}'s terminology, the @dfn{raw} representation is the one used in
2760 the target registers, and the @dfn{virtual} representation is the one
2761 used in memory, and within @value{GDBN} @code{struct value} objects.
2763 @emph{Maintainer note: Notice that the same mechanism is being used to
2764 both convert a register to a @code{struct value} and alternative
2767 For almost all data types on almost all architectures, the virtual and
2768 raw representations are identical, and no special handling is needed.
2769 However, they do occasionally differ. For example:
2773 The x86 architecture supports an 80-bit @code{long double} type. However, when
2774 we store those values in memory, they occupy twelve bytes: the
2775 floating-point number occupies the first ten, and the final two bytes
2776 are unused. This keeps the values aligned on four-byte boundaries,
2777 allowing more efficient access. Thus, the x86 80-bit floating-point
2778 type is the raw representation, and the twelve-byte loosely-packed
2779 arrangement is the virtual representation.
2782 Some 64-bit MIPS targets present 32-bit registers to @value{GDBN} as 64-bit
2783 registers, with garbage in their upper bits. @value{GDBN} ignores the top 32
2784 bits. Thus, the 64-bit form, with garbage in the upper 32 bits, is the
2785 raw representation, and the trimmed 32-bit representation is the
2786 virtual representation.
2789 In general, the raw representation is determined by the architecture, or
2790 @value{GDBN}'s interface to the architecture, while the virtual representation
2791 can be chosen for @value{GDBN}'s convenience. @value{GDBN}'s register file,
2792 @code{registers}, holds the register contents in raw format, and the
2793 @value{GDBN} remote protocol transmits register values in raw format.
2795 Your architecture may define the following macros to request
2796 conversions between the raw and virtual format:
2798 @deftypefn {Target Macro} int REGISTER_CONVERTIBLE (int @var{reg})
2799 Return non-zero if register number @var{reg}'s value needs different raw
2800 and virtual formats.
2802 You should not use @code{REGISTER_CONVERT_TO_VIRTUAL} for a register
2803 unless this macro returns a non-zero value for that register.
2806 @deftypefn {Target Macro} int REGISTER_RAW_SIZE (int @var{reg})
2807 The size of register number @var{reg}'s raw value. This is the number
2808 of bytes the register will occupy in @code{registers}, or in a @value{GDBN}
2809 remote protocol packet.
2812 @deftypefn {Target Macro} int REGISTER_VIRTUAL_SIZE (int @var{reg})
2813 The size of register number @var{reg}'s value, in its virtual format.
2814 This is the size a @code{struct value}'s buffer will have, holding that
2818 @deftypefn {Target Macro} struct type *REGISTER_VIRTUAL_TYPE (int @var{reg})
2819 This is the type of the virtual representation of register number
2820 @var{reg}. Note that there is no need for a macro giving a type for the
2821 register's raw form; once the register's value has been obtained, @value{GDBN}
2822 always uses the virtual form.
2825 @deftypefn {Target Macro} void REGISTER_CONVERT_TO_VIRTUAL (int @var{reg}, struct type *@var{type}, char *@var{from}, char *@var{to})
2826 Convert the value of register number @var{reg} to @var{type}, which
2827 should always be @code{REGISTER_VIRTUAL_TYPE (@var{reg})}. The buffer
2828 at @var{from} holds the register's value in raw format; the macro should
2829 convert the value to virtual format, and place it at @var{to}.
2831 Note that @code{REGISTER_CONVERT_TO_VIRTUAL} and
2832 @code{REGISTER_CONVERT_TO_RAW} take their @var{reg} and @var{type}
2833 arguments in different orders.
2835 You should only use @code{REGISTER_CONVERT_TO_VIRTUAL} with registers
2836 for which the @code{REGISTER_CONVERTIBLE} macro returns a non-zero
2840 @deftypefn {Target Macro} void REGISTER_CONVERT_TO_RAW (struct type *@var{type}, int @var{reg}, char *@var{from}, char *@var{to})
2841 Convert the value of register number @var{reg} to @var{type}, which
2842 should always be @code{REGISTER_VIRTUAL_TYPE (@var{reg})}. The buffer
2843 at @var{from} holds the register's value in raw format; the macro should
2844 convert the value to virtual format, and place it at @var{to}.
2846 Note that REGISTER_CONVERT_TO_VIRTUAL and REGISTER_CONVERT_TO_RAW take
2847 their @var{reg} and @var{type} arguments in different orders.
2851 @section Using Different Register and Memory Data Representations
2852 @cindex register representation
2853 @cindex memory representation
2854 @cindex representations, register and memory
2855 @cindex register data formats, converting
2856 @cindex @code{struct value}, converting register contents to
2858 @emph{Maintainer's note: The way GDB manipulates registers is undergoing
2859 significant change. Many of the macros and functions refered to in this
2860 section are likely to be subject to further revision. See
2861 @uref{http://sources.redhat.com/gdb/current/ari/, A.R. Index} and
2862 @uref{http://www.gnu.org/software/gdb/bugs, Bug Tracking Database} for
2863 further information. cagney/2002-05-06.}
2865 Some architectures can represent a data object in a register using a
2866 form that is different to the objects more normal memory representation.
2872 The Alpha architecture can represent 32 bit integer values in
2873 floating-point registers.
2876 The x86 architecture supports 80-bit floating-point registers. The
2877 @code{long double} data type occupies 96 bits in memory but only 80 bits
2878 when stored in a register.
2882 In general, the register representation of a data type is determined by
2883 the architecture, or @value{GDBN}'s interface to the architecture, while
2884 the memory representation is determined by the Application Binary
2887 For almost all data types on almost all architectures, the two
2888 representations are identical, and no special handling is needed.
2889 However, they do occasionally differ. Your architecture may define the
2890 following macros to request conversions between the register and memory
2891 representations of a data type:
2893 @deftypefn {Target Macro} int CONVERT_REGISTER_P (int @var{reg})
2894 Return non-zero if the representation of a data value stored in this
2895 register may be different to the representation of that same data value
2896 when stored in memory.
2898 When non-zero, the macros @code{REGISTER_TO_VALUE} and
2899 @code{VALUE_TO_REGISTER} are used to perform any necessary conversion.
2902 @deftypefn {Target Macro} void REGISTER_TO_VALUE (int @var{reg}, struct type *@var{type}, char *@var{from}, char *@var{to})
2903 Convert the value of register number @var{reg} to a data object of type
2904 @var{type}. The buffer at @var{from} holds the register's value in raw
2905 format; the converted value should be placed in the buffer at @var{to}.
2907 Note that @code{REGISTER_TO_VALUE} and @code{VALUE_TO_REGISTER} take
2908 their @var{reg} and @var{type} arguments in different orders.
2910 You should only use @code{REGISTER_TO_VALUE} with registers for which
2911 the @code{CONVERT_REGISTER_P} macro returns a non-zero value.
2914 @deftypefn {Target Macro} void VALUE_TO_REGISTER (struct type *@var{type}, int @var{reg}, char *@var{from}, char *@var{to})
2915 Convert a data value of type @var{type} to register number @var{reg}'
2918 Note that @code{REGISTER_TO_VALUE} and @code{VALUE_TO_REGISTER} take
2919 their @var{reg} and @var{type} arguments in different orders.
2921 You should only use @code{VALUE_TO_REGISTER} with registers for which
2922 the @code{CONVERT_REGISTER_P} macro returns a non-zero value.
2925 @deftypefn {Target Macro} void REGISTER_CONVERT_TO_TYPE (int @var{regnum}, struct type *@var{type}, char *@var{buf})
2926 See @file{mips-tdep.c}. It does not do what you want.
2930 @section Frame Interpretation
2932 @section Inferior Call Setup
2934 @section Compiler Characteristics
2936 @section Target Conditionals
2938 This section describes the macros that you can use to define the target
2943 @item ADDITIONAL_OPTIONS
2944 @itemx ADDITIONAL_OPTION_CASES
2945 @itemx ADDITIONAL_OPTION_HANDLER
2946 @itemx ADDITIONAL_OPTION_HELP
2947 @findex ADDITIONAL_OPTION_HELP
2948 @findex ADDITIONAL_OPTION_HANDLER
2949 @findex ADDITIONAL_OPTION_CASES
2950 @findex ADDITIONAL_OPTIONS
2951 These are a set of macros that allow the addition of additional command
2952 line options to @value{GDBN}. They are currently used only for the unsupported
2953 i960 Nindy target, and should not be used in any other configuration.
2955 @item ADDR_BITS_REMOVE (addr)
2956 @findex ADDR_BITS_REMOVE
2957 If a raw machine instruction address includes any bits that are not
2958 really part of the address, then define this macro to expand into an
2959 expression that zeroes those bits in @var{addr}. This is only used for
2960 addresses of instructions, and even then not in all contexts.
2962 For example, the two low-order bits of the PC on the Hewlett-Packard PA
2963 2.0 architecture contain the privilege level of the corresponding
2964 instruction. Since instructions must always be aligned on four-byte
2965 boundaries, the processor masks out these bits to generate the actual
2966 address of the instruction. ADDR_BITS_REMOVE should filter out these
2967 bits with an expression such as @code{((addr) & ~3)}.
2969 @item ADDRESS_CLASS_NAME_TO_TYPE_FLAGS (@var{name}, @var{type_flags_ptr})
2970 @findex ADDRESS_CLASS_NAME_TO_TYPE_FLAGS
2971 If @var{name} is a valid address class qualifier name, set the @code{int}
2972 referenced by @var{type_flags_ptr} to the mask representing the qualifier
2973 and return 1. If @var{name} is not a valid address class qualifier name,
2976 The value for @var{type_flags_ptr} should be one of
2977 @code{TYPE_FLAG_ADDRESS_CLASS_1}, @code{TYPE_FLAG_ADDRESS_CLASS_2}, or
2978 possibly some combination of these values or'd together.
2979 @xref{Target Architecture Definition, , Address Classes}.
2981 @item ADDRESS_CLASS_NAME_TO_TYPE_FLAGS_P ()
2982 @findex ADDRESS_CLASS_NAME_TO_TYPE_FLAGS_P
2983 Predicate which indicates whether @code{ADDRESS_CLASS_NAME_TO_TYPE_FLAGS}
2986 @item ADDRESS_CLASS_TYPE_FLAGS (@var{byte_size}, @var{dwarf2_addr_class})
2987 @findex ADDRESS_CLASS_TYPE_FLAGS (@var{byte_size}, @var{dwarf2_addr_class})
2988 Given a pointers byte size (as described by the debug information) and
2989 the possible @code{DW_AT_address_class} value, return the type flags
2990 used by @value{GDBN} to represent this address class. The value
2991 returned should be one of @code{TYPE_FLAG_ADDRESS_CLASS_1},
2992 @code{TYPE_FLAG_ADDRESS_CLASS_2}, or possibly some combination of these
2993 values or'd together.
2994 @xref{Target Architecture Definition, , Address Classes}.
2996 @item ADDRESS_CLASS_TYPE_FLAGS_P ()
2997 @findex ADDRESS_CLASS_TYPE_FLAGS_P
2998 Predicate which indicates whether @code{ADDRESS_CLASS_TYPE_FLAGS} has
3001 @item ADDRESS_CLASS_TYPE_FLAGS_TO_NAME (@var{type_flags})
3002 @findex ADDRESS_CLASS_TYPE_FLAGS_TO_NAME
3003 Return the name of the address class qualifier associated with the type
3004 flags given by @var{type_flags}.
3006 @item ADDRESS_CLASS_TYPE_FLAGS_TO_NAME_P ()
3007 @findex ADDRESS_CLASS_TYPE_FLAGS_TO_NAME_P
3008 Predicate which indicates whether @code{ADDRESS_CLASS_TYPE_FLAGS_TO_NAME} has
3010 @xref{Target Architecture Definition, , Address Classes}.
3012 @item ADDRESS_TO_POINTER (@var{type}, @var{buf}, @var{addr})
3013 @findex ADDRESS_TO_POINTER
3014 Store in @var{buf} a pointer of type @var{type} representing the address
3015 @var{addr}, in the appropriate format for the current architecture.
3016 This macro may safely assume that @var{type} is either a pointer or a
3017 C@t{++} reference type.
3018 @xref{Target Architecture Definition, , Pointers Are Not Always Addresses}.
3020 @item BEFORE_MAIN_LOOP_HOOK
3021 @findex BEFORE_MAIN_LOOP_HOOK
3022 Define this to expand into any code that you want to execute before the
3023 main loop starts. Although this is not, strictly speaking, a target
3024 conditional, that is how it is currently being used. Note that if a
3025 configuration were to define it one way for a host and a different way
3026 for the target, @value{GDBN} will probably not compile, let alone run
3027 correctly. This macro is currently used only for the unsupported i960 Nindy
3028 target, and should not be used in any other configuration.
3030 @item BELIEVE_PCC_PROMOTION
3031 @findex BELIEVE_PCC_PROMOTION
3032 Define if the compiler promotes a @code{short} or @code{char}
3033 parameter to an @code{int}, but still reports the parameter as its
3034 original type, rather than the promoted type.
3036 @item BELIEVE_PCC_PROMOTION_TYPE
3037 @findex BELIEVE_PCC_PROMOTION_TYPE
3038 Define this if @value{GDBN} should believe the type of a @code{short}
3039 argument when compiled by @code{pcc}, but look within a full int space to get
3040 its value. Only defined for Sun-3 at present.
3042 @item BITS_BIG_ENDIAN
3043 @findex BITS_BIG_ENDIAN
3044 Define this if the numbering of bits in the targets does @strong{not} match the
3045 endianness of the target byte order. A value of 1 means that the bits
3046 are numbered in a big-endian bit order, 0 means little-endian.
3050 This is the character array initializer for the bit pattern to put into
3051 memory where a breakpoint is set. Although it's common to use a trap
3052 instruction for a breakpoint, it's not required; for instance, the bit
3053 pattern could be an invalid instruction. The breakpoint must be no
3054 longer than the shortest instruction of the architecture.
3056 @code{BREAKPOINT} has been deprecated in favor of
3057 @code{BREAKPOINT_FROM_PC}.
3059 @item BIG_BREAKPOINT
3060 @itemx LITTLE_BREAKPOINT
3061 @findex LITTLE_BREAKPOINT
3062 @findex BIG_BREAKPOINT
3063 Similar to BREAKPOINT, but used for bi-endian targets.
3065 @code{BIG_BREAKPOINT} and @code{LITTLE_BREAKPOINT} have been deprecated in
3066 favor of @code{BREAKPOINT_FROM_PC}.
3068 @item REMOTE_BREAKPOINT
3069 @itemx LITTLE_REMOTE_BREAKPOINT
3070 @itemx BIG_REMOTE_BREAKPOINT
3071 @findex BIG_REMOTE_BREAKPOINT
3072 @findex LITTLE_REMOTE_BREAKPOINT
3073 @findex REMOTE_BREAKPOINT
3074 Similar to BREAKPOINT, but used for remote targets.
3076 @code{BIG_REMOTE_BREAKPOINT} and @code{LITTLE_REMOTE_BREAKPOINT} have been
3077 deprecated in favor of @code{BREAKPOINT_FROM_PC}.
3079 @item BREAKPOINT_FROM_PC (@var{pcptr}, @var{lenptr})
3080 @findex BREAKPOINT_FROM_PC
3081 Use the program counter to determine the contents and size of a
3082 breakpoint instruction. It returns a pointer to a string of bytes
3083 that encode a breakpoint instruction, stores the length of the string
3084 to *@var{lenptr}, and adjusts pc (if necessary) to point to the actual
3085 memory location where the breakpoint should be inserted.
3087 Although it is common to use a trap instruction for a breakpoint, it's
3088 not required; for instance, the bit pattern could be an invalid
3089 instruction. The breakpoint must be no longer than the shortest
3090 instruction of the architecture.
3092 Replaces all the other @var{BREAKPOINT} macros.
3094 @item MEMORY_INSERT_BREAKPOINT (@var{addr}, @var{contents_cache})
3095 @itemx MEMORY_REMOVE_BREAKPOINT (@var{addr}, @var{contents_cache})
3096 @findex MEMORY_REMOVE_BREAKPOINT
3097 @findex MEMORY_INSERT_BREAKPOINT
3098 Insert or remove memory based breakpoints. Reasonable defaults
3099 (@code{default_memory_insert_breakpoint} and
3100 @code{default_memory_remove_breakpoint} respectively) have been
3101 provided so that it is not necessary to define these for most
3102 architectures. Architectures which may want to define
3103 @code{MEMORY_INSERT_BREAKPOINT} and @code{MEMORY_REMOVE_BREAKPOINT} will
3104 likely have instructions that are oddly sized or are not stored in a
3105 conventional manner.
3107 It may also be desirable (from an efficiency standpoint) to define
3108 custom breakpoint insertion and removal routines if
3109 @code{BREAKPOINT_FROM_PC} needs to read the target's memory for some
3113 @findex CALL_DUMMY_P
3114 A C expression that is non-zero when the target supports inferior function
3117 @item CALL_DUMMY_WORDS
3118 @findex CALL_DUMMY_WORDS
3119 Pointer to an array of @code{LONGEST} words of data containing
3120 host-byte-ordered @code{REGISTER_BYTES} sized values that partially
3121 specify the sequence of instructions needed for an inferior function
3124 Should be deprecated in favor of a macro that uses target-byte-ordered
3127 @item SIZEOF_CALL_DUMMY_WORDS
3128 @findex SIZEOF_CALL_DUMMY_WORDS
3129 The size of @code{CALL_DUMMY_WORDS}. When @code{CALL_DUMMY_P} this must
3130 return a positive value. See also @code{CALL_DUMMY_LENGTH}.
3134 A static initializer for @code{CALL_DUMMY_WORDS}. Deprecated.
3136 @item CALL_DUMMY_LOCATION
3137 @findex CALL_DUMMY_LOCATION
3138 See the file @file{inferior.h}.
3140 @item CALL_DUMMY_STACK_ADJUST
3141 @findex CALL_DUMMY_STACK_ADJUST
3142 Stack adjustment needed when performing an inferior function call.
3144 Should be deprecated in favor of something like @code{STACK_ALIGN}.
3146 @item CALL_DUMMY_STACK_ADJUST_P
3147 @findex CALL_DUMMY_STACK_ADJUST_P
3148 Predicate for use of @code{CALL_DUMMY_STACK_ADJUST}.
3150 Should be deprecated in favor of something like @code{STACK_ALIGN}.
3152 @item CANNOT_FETCH_REGISTER (@var{regno})
3153 @findex CANNOT_FETCH_REGISTER
3154 A C expression that should be nonzero if @var{regno} cannot be fetched
3155 from an inferior process. This is only relevant if
3156 @code{FETCH_INFERIOR_REGISTERS} is not defined.
3158 @item CANNOT_STORE_REGISTER (@var{regno})
3159 @findex CANNOT_STORE_REGISTER
3160 A C expression that should be nonzero if @var{regno} should not be
3161 written to the target. This is often the case for program counters,
3162 status words, and other special registers. If this is not defined,
3163 @value{GDBN} will assume that all registers may be written.
3165 @item DO_DEFERRED_STORES
3166 @itemx CLEAR_DEFERRED_STORES
3167 @findex CLEAR_DEFERRED_STORES
3168 @findex DO_DEFERRED_STORES
3169 Define this to execute any deferred stores of registers into the inferior,
3170 and to cancel any deferred stores.
3172 Currently only implemented correctly for native Sparc configurations?
3174 @item int CONVERT_REGISTER_P(@var{regnum})
3175 @findex CONVERT_REGISTER_P
3176 Return non-zero if register @var{regnum} can represent data values in a
3178 @xref{Target Architecture Definition, , Using Different Register and Memory Data Representations}.
3180 @item DBX_PARM_SYMBOL_CLASS
3181 @findex DBX_PARM_SYMBOL_CLASS
3182 Hook for the @code{SYMBOL_CLASS} of a parameter when decoding DBX symbol
3183 information. In the i960, parameters can be stored as locals or as
3184 args, depending on the type of the debug record.
3186 @item DECR_PC_AFTER_BREAK
3187 @findex DECR_PC_AFTER_BREAK
3188 Define this to be the amount by which to decrement the PC after the
3189 program encounters a breakpoint. This is often the number of bytes in
3190 @code{BREAKPOINT}, though not always. For most targets this value will be 0.
3192 @item DECR_PC_AFTER_HW_BREAK
3193 @findex DECR_PC_AFTER_HW_BREAK
3194 Similarly, for hardware breakpoints.
3196 @item DISABLE_UNSETTABLE_BREAK (@var{addr})
3197 @findex DISABLE_UNSETTABLE_BREAK
3198 If defined, this should evaluate to 1 if @var{addr} is in a shared
3199 library in which breakpoints cannot be set and so should be disabled.
3201 @item PRINT_FLOAT_INFO()
3202 @findex PRINT_FLOAT_INFO
3203 If defined, then the @samp{info float} command will print information about
3204 the processor's floating point unit.
3206 @item print_registers_info (@var{gdbarch}, @var{frame}, @var{regnum}, @var{all})
3207 @findex print_registers_info
3208 If defined, pretty print the value of the register @var{regnum} for the
3209 specified @var{frame}. If the value of @var{regnum} is -1, pretty print
3210 either all registers (@var{all} is non zero) or a select subset of
3211 registers (@var{all} is zero).
3213 The default method prints one register per line, and if @var{all} is
3214 zero omits floating-point registers.
3216 @item PRINT_VECTOR_INFO()
3217 @findex PRINT_VECTOR_INFO
3218 If defined, then the @samp{info vector} command will call this function
3219 to print information about the processor's vector unit.
3221 By default, the @samp{info vector} command will print all vector
3222 registers (the register's type having the vector attribute).
3224 @item DWARF_REG_TO_REGNUM
3225 @findex DWARF_REG_TO_REGNUM
3226 Convert DWARF register number into @value{GDBN} regnum. If not defined,
3227 no conversion will be performed.
3229 @item DWARF2_REG_TO_REGNUM
3230 @findex DWARF2_REG_TO_REGNUM
3231 Convert DWARF2 register number into @value{GDBN} regnum. If not
3232 defined, no conversion will be performed.
3234 @item ECOFF_REG_TO_REGNUM
3235 @findex ECOFF_REG_TO_REGNUM
3236 Convert ECOFF register number into @value{GDBN} regnum. If not defined,
3237 no conversion will be performed.
3239 @item END_OF_TEXT_DEFAULT
3240 @findex END_OF_TEXT_DEFAULT
3241 This is an expression that should designate the end of the text section.
3244 @item EXTRACT_RETURN_VALUE(@var{type}, @var{regbuf}, @var{valbuf})
3245 @findex EXTRACT_RETURN_VALUE
3246 Define this to extract a function's return value of type @var{type} from
3247 the raw register state @var{regbuf} and copy that, in virtual format,
3250 @item EXTRACT_STRUCT_VALUE_ADDRESS(@var{regbuf})
3251 @findex EXTRACT_STRUCT_VALUE_ADDRESS
3252 When defined, extract from the array @var{regbuf} (containing the raw
3253 register state) the @code{CORE_ADDR} at which a function should return
3254 its structure value.
3256 If not defined, @code{EXTRACT_RETURN_VALUE} is used.
3258 @item EXTRACT_STRUCT_VALUE_ADDRESS_P()
3259 @findex EXTRACT_STRUCT_VALUE_ADDRESS_P
3260 Predicate for @code{EXTRACT_STRUCT_VALUE_ADDRESS}.
3264 Deprecated in favor of @code{PRINT_FLOAT_INFO}.
3268 If the virtual frame pointer is kept in a register, then define this
3269 macro to be the number (greater than or equal to zero) of that register.
3271 This should only need to be defined if @code{TARGET_READ_FP} is not
3274 @item FRAMELESS_FUNCTION_INVOCATION(@var{fi})
3275 @findex FRAMELESS_FUNCTION_INVOCATION
3276 Define this to an expression that returns 1 if the function invocation
3277 represented by @var{fi} does not have a stack frame associated with it.
3280 @item frame_align (@var{address})
3281 @anchor{frame_align}
3283 Define this to adjust @var{address} so that it meets the alignment
3284 requirements for the start of a new stack frame. A stack frame's
3285 alignment requirements are typically stronger than a target processors
3286 stack alignment requirements (@pxref{STACK_ALIGN}).
3288 This function is used to ensure that, when creating a dummy frame, both
3289 the initial stack pointer and (if needed) the address of the return
3290 value are correctly aligned.
3292 Unlike @code{STACK_ALIGN}, this function always adjusts the address in
3293 the direction of stack growth.
3295 By default, no frame based stack alignment is performed.
3297 @item FRAME_ARGS_ADDRESS_CORRECT
3298 @findex FRAME_ARGS_ADDRESS_CORRECT
3301 @item FRAME_CHAIN(@var{frame})
3303 Given @var{frame}, return a pointer to the calling frame.
3305 @item FRAME_CHAIN_VALID(@var{chain}, @var{thisframe})
3306 @findex FRAME_CHAIN_VALID
3307 Define this to be an expression that returns zero if the given frame is an
3308 outermost frame, with no caller, and nonzero otherwise. Most normal
3309 situations can be handled without defining this macro, including @code{NULL}
3310 chain pointers, dummy frames, and frames whose PC values are inside the
3311 startup file (e.g.@: @file{crt0.o}), inside @code{main}, or inside
3314 @item FRAME_INIT_SAVED_REGS(@var{frame})
3315 @findex FRAME_INIT_SAVED_REGS
3316 See @file{frame.h}. Determines the address of all registers in the
3317 current stack frame storing each in @code{frame->saved_regs}. Space for
3318 @code{frame->saved_regs} shall be allocated by
3319 @code{FRAME_INIT_SAVED_REGS} using @code{frame_saved_regs_zalloc}.
3321 @code{FRAME_FIND_SAVED_REGS} and @code{EXTRA_FRAME_INFO} are deprecated.
3323 @item FRAME_NUM_ARGS (@var{fi})
3324 @findex FRAME_NUM_ARGS
3325 For the frame described by @var{fi} return the number of arguments that
3326 are being passed. If the number of arguments is not known, return
3329 @item FRAME_SAVED_PC(@var{frame})
3330 @findex FRAME_SAVED_PC
3331 Given @var{frame}, return the pc saved there. This is the return
3334 @item FUNCTION_EPILOGUE_SIZE
3335 @findex FUNCTION_EPILOGUE_SIZE
3336 For some COFF targets, the @code{x_sym.x_misc.x_fsize} field of the
3337 function end symbol is 0. For such targets, you must define
3338 @code{FUNCTION_EPILOGUE_SIZE} to expand into the standard size of a
3339 function's epilogue.
3341 @item FUNCTION_START_OFFSET
3342 @findex FUNCTION_START_OFFSET
3343 An integer, giving the offset in bytes from a function's address (as
3344 used in the values of symbols, function pointers, etc.), and the
3345 function's first genuine instruction.
3347 This is zero on almost all machines: the function's address is usually
3348 the address of its first instruction. However, on the VAX, for example,
3349 each function starts with two bytes containing a bitmask indicating
3350 which registers to save upon entry to the function. The VAX @code{call}
3351 instructions check this value, and save the appropriate registers
3352 automatically. Thus, since the offset from the function's address to
3353 its first instruction is two bytes, @code{FUNCTION_START_OFFSET} would
3356 @item GCC_COMPILED_FLAG_SYMBOL
3357 @itemx GCC2_COMPILED_FLAG_SYMBOL
3358 @findex GCC2_COMPILED_FLAG_SYMBOL
3359 @findex GCC_COMPILED_FLAG_SYMBOL
3360 If defined, these are the names of the symbols that @value{GDBN} will
3361 look for to detect that GCC compiled the file. The default symbols
3362 are @code{gcc_compiled.} and @code{gcc2_compiled.},
3363 respectively. (Currently only defined for the Delta 68.)
3365 @item @value{GDBN}_MULTI_ARCH
3366 @findex @value{GDBN}_MULTI_ARCH
3367 If defined and non-zero, enables support for multiple architectures
3368 within @value{GDBN}.
3370 This support can be enabled at two levels. At level one, only
3371 definitions for previously undefined macros are provided; at level two,
3372 a multi-arch definition of all architecture dependent macros will be
3375 @item @value{GDBN}_TARGET_IS_HPPA
3376 @findex @value{GDBN}_TARGET_IS_HPPA
3377 This determines whether horrible kludge code in @file{dbxread.c} and
3378 @file{partial-stab.h} is used to mangle multiple-symbol-table files from
3379 HPPA's. This should all be ripped out, and a scheme like @file{elfread.c}
3382 @item GET_LONGJMP_TARGET
3383 @findex GET_LONGJMP_TARGET
3384 For most machines, this is a target-dependent parameter. On the
3385 DECstation and the Iris, this is a native-dependent parameter, since
3386 the header file @file{setjmp.h} is needed to define it.
3388 This macro determines the target PC address that @code{longjmp} will jump to,
3389 assuming that we have just stopped at a @code{longjmp} breakpoint. It takes a
3390 @code{CORE_ADDR *} as argument, and stores the target PC value through this
3391 pointer. It examines the current state of the machine as needed.
3393 @item GET_SAVED_REGISTER
3394 @findex GET_SAVED_REGISTER
3395 @findex get_saved_register
3396 Define this if you need to supply your own definition for the function
3397 @code{get_saved_register}.
3399 @item IBM6000_TARGET
3400 @findex IBM6000_TARGET
3401 Shows that we are configured for an IBM RS/6000 target. This
3402 conditional should be eliminated (FIXME) and replaced by
3403 feature-specific macros. It was introduced in a haste and we are
3404 repenting at leisure.
3406 @item I386_USE_GENERIC_WATCHPOINTS
3407 An x86-based target can define this to use the generic x86 watchpoint
3408 support; see @ref{Algorithms, I386_USE_GENERIC_WATCHPOINTS}.
3410 @item SYMBOLS_CAN_START_WITH_DOLLAR
3411 @findex SYMBOLS_CAN_START_WITH_DOLLAR
3412 Some systems have routines whose names start with @samp{$}. Giving this
3413 macro a non-zero value tells @value{GDBN}'s expression parser to check for such
3414 routines when parsing tokens that begin with @samp{$}.
3416 On HP-UX, certain system routines (millicode) have names beginning with
3417 @samp{$} or @samp{$$}. For example, @code{$$dyncall} is a millicode
3418 routine that handles inter-space procedure calls on PA-RISC.
3420 @item INIT_EXTRA_FRAME_INFO (@var{fromleaf}, @var{frame})
3421 @findex INIT_EXTRA_FRAME_INFO
3422 If additional information about the frame is required this should be
3423 stored in @code{frame->extra_info}. Space for @code{frame->extra_info}
3424 is allocated using @code{frame_extra_info_zalloc}.
3426 @item DEPRECATED_INIT_FRAME_PC (@var{fromleaf}, @var{prev})
3427 @findex DEPRECATED_INIT_FRAME_PC
3428 This is a C statement that sets the pc of the frame pointed to by
3429 @var{prev}. [By default...]
3431 @item INNER_THAN (@var{lhs}, @var{rhs})
3433 Returns non-zero if stack address @var{lhs} is inner than (nearer to the
3434 stack top) stack address @var{rhs}. Define this as @code{lhs < rhs} if
3435 the target's stack grows downward in memory, or @code{lhs > rsh} if the
3438 @item gdbarch_in_function_epilogue_p (@var{gdbarch}, @var{pc})
3439 @findex gdbarch_in_function_epilogue_p
3440 Returns non-zero if the given @var{pc} is in the epilogue of a function.
3441 The epilogue of a function is defined as the part of a function where
3442 the stack frame of the function already has been destroyed up to the
3443 final `return from function call' instruction.
3445 @item SIGTRAMP_START (@var{pc})
3446 @findex SIGTRAMP_START
3447 @itemx SIGTRAMP_END (@var{pc})
3448 @findex SIGTRAMP_END
3449 Define these to be the start and end address of the @code{sigtramp} for the
3450 given @var{pc}. On machines where the address is just a compile time
3451 constant, the macro expansion will typically just ignore the supplied
3454 @item IN_SOLIB_CALL_TRAMPOLINE (@var{pc}, @var{name})
3455 @findex IN_SOLIB_CALL_TRAMPOLINE
3456 Define this to evaluate to nonzero if the program is stopped in the
3457 trampoline that connects to a shared library.
3459 @item IN_SOLIB_RETURN_TRAMPOLINE (@var{pc}, @var{name})
3460 @findex IN_SOLIB_RETURN_TRAMPOLINE
3461 Define this to evaluate to nonzero if the program is stopped in the
3462 trampoline that returns from a shared library.
3464 @item IN_SOLIB_DYNSYM_RESOLVE_CODE (@var{pc})
3465 @findex IN_SOLIB_DYNSYM_RESOLVE_CODE
3466 Define this to evaluate to nonzero if the program is stopped in the
3469 @item SKIP_SOLIB_RESOLVER (@var{pc})
3470 @findex SKIP_SOLIB_RESOLVER
3471 Define this to evaluate to the (nonzero) address at which execution
3472 should continue to get past the dynamic linker's symbol resolution
3473 function. A zero value indicates that it is not important or necessary
3474 to set a breakpoint to get through the dynamic linker and that single
3475 stepping will suffice.
3477 @item INTEGER_TO_ADDRESS (@var{type}, @var{buf})
3478 @findex INTEGER_TO_ADDRESS
3479 @cindex converting integers to addresses
3480 Define this when the architecture needs to handle non-pointer to address
3481 conversions specially. Converts that value to an address according to
3482 the current architectures conventions.
3484 @emph{Pragmatics: When the user copies a well defined expression from
3485 their source code and passes it, as a parameter, to @value{GDBN}'s
3486 @code{print} command, they should get the same value as would have been
3487 computed by the target program. Any deviation from this rule can cause
3488 major confusion and annoyance, and needs to be justified carefully. In
3489 other words, @value{GDBN} doesn't really have the freedom to do these
3490 conversions in clever and useful ways. It has, however, been pointed
3491 out that users aren't complaining about how @value{GDBN} casts integers
3492 to pointers; they are complaining that they can't take an address from a
3493 disassembly listing and give it to @code{x/i}. Adding an architecture
3494 method like @code{INTEGER_TO_ADDRESS} certainly makes it possible for
3495 @value{GDBN} to ``get it right'' in all circumstances.}
3497 @xref{Target Architecture Definition, , Pointers Are Not Always
3500 @item IS_TRAPPED_INTERNALVAR (@var{name})
3501 @findex IS_TRAPPED_INTERNALVAR
3502 This is an ugly hook to allow the specification of special actions that
3503 should occur as a side-effect of setting the value of a variable
3504 internal to @value{GDBN}. Currently only used by the h8500. Note that this
3505 could be either a host or target conditional.
3507 @item NEED_TEXT_START_END
3508 @findex NEED_TEXT_START_END
3509 Define this if @value{GDBN} should determine the start and end addresses of the
3510 text section. (Seems dubious.)
3512 @item NO_HIF_SUPPORT
3513 @findex NO_HIF_SUPPORT
3514 (Specific to the a29k.)
3516 @item POINTER_TO_ADDRESS (@var{type}, @var{buf})
3517 @findex POINTER_TO_ADDRESS
3518 Assume that @var{buf} holds a pointer of type @var{type}, in the
3519 appropriate format for the current architecture. Return the byte
3520 address the pointer refers to.
3521 @xref{Target Architecture Definition, , Pointers Are Not Always Addresses}.
3523 @item REGISTER_CONVERTIBLE (@var{reg})
3524 @findex REGISTER_CONVERTIBLE
3525 Return non-zero if @var{reg} uses different raw and virtual formats.
3526 @xref{Target Architecture Definition, , Raw and Virtual Register Representations}.
3528 @item REGISTER_TO_VALUE(@var{regnum}, @var{type}, @var{from}, @var{to})
3529 @findex REGISTER_TO_VALUE
3530 Convert the raw contents of register @var{regnum} into a value of type
3532 @xref{Target Architecture Definition, , Using Different Register and Memory Data Representations}.
3534 @item REGISTER_RAW_SIZE (@var{reg})
3535 @findex REGISTER_RAW_SIZE
3536 Return the raw size of @var{reg}; defaults to the size of the register's
3538 @xref{Target Architecture Definition, , Raw and Virtual Register Representations}.
3540 @item REGISTER_VIRTUAL_SIZE (@var{reg})
3541 @findex REGISTER_VIRTUAL_SIZE
3542 Return the virtual size of @var{reg}; defaults to the size of the
3543 register's virtual type.
3544 Return the virtual size of @var{reg}.
3545 @xref{Target Architecture Definition, , Raw and Virtual Register Representations}.
3547 @item REGISTER_VIRTUAL_TYPE (@var{reg})
3548 @findex REGISTER_VIRTUAL_TYPE
3549 Return the virtual type of @var{reg}.
3550 @xref{Target Architecture Definition, , Raw and Virtual Register Representations}.
3552 @item REGISTER_CONVERT_TO_VIRTUAL(@var{reg}, @var{type}, @var{from}, @var{to})
3553 @findex REGISTER_CONVERT_TO_VIRTUAL
3554 Convert the value of register @var{reg} from its raw form to its virtual
3556 @xref{Target Architecture Definition, , Raw and Virtual Register Representations}.
3558 @item REGISTER_CONVERT_TO_RAW(@var{type}, @var{reg}, @var{from}, @var{to})
3559 @findex REGISTER_CONVERT_TO_RAW
3560 Convert the value of register @var{reg} from its virtual form to its raw
3562 @xref{Target Architecture Definition, , Raw and Virtual Register Representations}.
3564 @item RETURN_VALUE_ON_STACK(@var{type})
3565 @findex RETURN_VALUE_ON_STACK
3566 @cindex returning structures by value
3567 @cindex structures, returning by value
3569 Return non-zero if values of type TYPE are returned on the stack, using
3570 the ``struct convention'' (i.e., the caller provides a pointer to a
3571 buffer in which the callee should store the return value). This
3572 controls how the @samp{finish} command finds a function's return value,
3573 and whether an inferior function call reserves space on the stack for
3576 The full logic @value{GDBN} uses here is kind of odd.
3580 If the type being returned by value is not a structure, union, or array,
3581 and @code{RETURN_VALUE_ON_STACK} returns zero, then @value{GDBN}
3582 concludes the value is not returned using the struct convention.
3585 Otherwise, @value{GDBN} calls @code{USE_STRUCT_CONVENTION} (see below).
3586 If that returns non-zero, @value{GDBN} assumes the struct convention is
3590 In other words, to indicate that a given type is returned by value using
3591 the struct convention, that type must be either a struct, union, array,
3592 or something @code{RETURN_VALUE_ON_STACK} likes, @emph{and} something
3593 that @code{USE_STRUCT_CONVENTION} likes.
3595 Note that, in C and C@t{++}, arrays are never returned by value. In those
3596 languages, these predicates will always see a pointer type, never an
3597 array type. All the references above to arrays being returned by value
3598 apply only to other languages.
3600 @item SOFTWARE_SINGLE_STEP_P()
3601 @findex SOFTWARE_SINGLE_STEP_P
3602 Define this as 1 if the target does not have a hardware single-step
3603 mechanism. The macro @code{SOFTWARE_SINGLE_STEP} must also be defined.
3605 @item SOFTWARE_SINGLE_STEP(@var{signal}, @var{insert_breapoints_p})
3606 @findex SOFTWARE_SINGLE_STEP
3607 A function that inserts or removes (depending on
3608 @var{insert_breapoints_p}) breakpoints at each possible destinations of
3609 the next instruction. See @file{sparc-tdep.c} and @file{rs6000-tdep.c}
3612 @item SOFUN_ADDRESS_MAYBE_MISSING
3613 @findex SOFUN_ADDRESS_MAYBE_MISSING
3614 Somebody clever observed that, the more actual addresses you have in the
3615 debug information, the more time the linker has to spend relocating
3616 them. So whenever there's some other way the debugger could find the
3617 address it needs, you should omit it from the debug info, to make
3620 @code{SOFUN_ADDRESS_MAYBE_MISSING} indicates that a particular set of
3621 hacks of this sort are in use, affecting @code{N_SO} and @code{N_FUN}
3622 entries in stabs-format debugging information. @code{N_SO} stabs mark
3623 the beginning and ending addresses of compilation units in the text
3624 segment. @code{N_FUN} stabs mark the starts and ends of functions.
3626 @code{SOFUN_ADDRESS_MAYBE_MISSING} means two things:
3630 @code{N_FUN} stabs have an address of zero. Instead, you should find the
3631 addresses where the function starts by taking the function name from
3632 the stab, and then looking that up in the minsyms (the
3633 linker/assembler symbol table). In other words, the stab has the
3634 name, and the linker/assembler symbol table is the only place that carries
3638 @code{N_SO} stabs have an address of zero, too. You just look at the
3639 @code{N_FUN} stabs that appear before and after the @code{N_SO} stab,
3640 and guess the starting and ending addresses of the compilation unit from
3644 @item PCC_SOL_BROKEN
3645 @findex PCC_SOL_BROKEN
3646 (Used only in the Convex target.)
3648 @item PC_IN_SIGTRAMP (@var{pc}, @var{name})
3649 @findex PC_IN_SIGTRAMP
3651 The @dfn{sigtramp} is a routine that the kernel calls (which then calls
3652 the signal handler). On most machines it is a library routine that is
3653 linked into the executable.
3655 This function, given a program counter value in @var{pc} and the
3656 (possibly NULL) name of the function in which that @var{pc} resides,
3657 returns nonzero if the @var{pc} and/or @var{name} show that we are in
3660 @item PC_LOAD_SEGMENT
3661 @findex PC_LOAD_SEGMENT
3662 If defined, print information about the load segment for the program
3663 counter. (Defined only for the RS/6000.)
3667 If the program counter is kept in a register, then define this macro to
3668 be the number (greater than or equal to zero) of that register.
3670 This should only need to be defined if @code{TARGET_READ_PC} and
3671 @code{TARGET_WRITE_PC} are not defined.
3675 The number of the ``next program counter'' register, if defined.
3678 @findex PARM_BOUNDARY
3679 If non-zero, round arguments to a boundary of this many bits before
3680 pushing them on the stack.
3682 @item PRINT_REGISTER_HOOK (@var{regno})
3683 @findex PRINT_REGISTER_HOOK
3684 If defined, this must be a function that prints the contents of the
3685 given register to standard output.
3687 @item PRINT_TYPELESS_INTEGER
3688 @findex PRINT_TYPELESS_INTEGER
3689 This is an obscure substitute for @code{print_longest} that seems to
3690 have been defined for the Convex target.
3692 @item PROCESS_LINENUMBER_HOOK
3693 @findex PROCESS_LINENUMBER_HOOK
3694 A hook defined for XCOFF reading.
3696 @item PROLOGUE_FIRSTLINE_OVERLAP
3697 @findex PROLOGUE_FIRSTLINE_OVERLAP
3698 (Only used in unsupported Convex configuration.)
3702 If defined, this is the number of the processor status register. (This
3703 definition is only used in generic code when parsing "$ps".)
3707 @findex call_function_by_hand
3708 @findex return_command
3709 Used in @samp{call_function_by_hand} to remove an artificial stack
3710 frame and in @samp{return_command} to remove a real stack frame.
3712 @item PUSH_ARGUMENTS (@var{nargs}, @var{args}, @var{sp}, @var{struct_return}, @var{struct_addr})
3713 @findex PUSH_ARGUMENTS
3714 Define this to push arguments onto the stack for inferior function
3715 call. Returns the updated stack pointer value.
3717 @item PUSH_DUMMY_FRAME
3718 @findex PUSH_DUMMY_FRAME
3719 Used in @samp{call_function_by_hand} to create an artificial stack frame.
3721 @item REGISTER_BYTES
3722 @findex REGISTER_BYTES
3723 The total amount of space needed to store @value{GDBN}'s copy of the machine's
3726 @item REGISTER_NAME(@var{i})
3727 @findex REGISTER_NAME
3728 Return the name of register @var{i} as a string. May return @code{NULL}
3729 or @code{NUL} to indicate that register @var{i} is not valid.
3731 @item REGISTER_NAMES
3732 @findex REGISTER_NAMES
3733 Deprecated in favor of @code{REGISTER_NAME}.
3735 @item REG_STRUCT_HAS_ADDR (@var{gcc_p}, @var{type})
3736 @findex REG_STRUCT_HAS_ADDR
3737 Define this to return 1 if the given type will be passed by pointer
3738 rather than directly.
3740 @item SAVE_DUMMY_FRAME_TOS (@var{sp})
3741 @findex SAVE_DUMMY_FRAME_TOS
3742 Used in @samp{call_function_by_hand} to notify the target dependent code
3743 of the top-of-stack value that will be passed to the the inferior code.
3744 This is the value of the @code{SP} after both the dummy frame and space
3745 for parameters/results have been allocated on the stack.
3747 @item SDB_REG_TO_REGNUM
3748 @findex SDB_REG_TO_REGNUM
3749 Define this to convert sdb register numbers into @value{GDBN} regnums. If not
3750 defined, no conversion will be done.
3752 @item SKIP_PERMANENT_BREAKPOINT
3753 @findex SKIP_PERMANENT_BREAKPOINT
3754 Advance the inferior's PC past a permanent breakpoint. @value{GDBN} normally
3755 steps over a breakpoint by removing it, stepping one instruction, and
3756 re-inserting the breakpoint. However, permanent breakpoints are
3757 hardwired into the inferior, and can't be removed, so this strategy
3758 doesn't work. Calling @code{SKIP_PERMANENT_BREAKPOINT} adjusts the processor's
3759 state so that execution will resume just after the breakpoint. This
3760 macro does the right thing even when the breakpoint is in the delay slot
3761 of a branch or jump.
3763 @item SKIP_PROLOGUE (@var{pc})
3764 @findex SKIP_PROLOGUE
3765 A C expression that returns the address of the ``real'' code beyond the
3766 function entry prologue found at @var{pc}.
3768 @item SKIP_TRAMPOLINE_CODE (@var{pc})
3769 @findex SKIP_TRAMPOLINE_CODE
3770 If the target machine has trampoline code that sits between callers and
3771 the functions being called, then define this macro to return a new PC
3772 that is at the start of the real function.
3776 If the stack-pointer is kept in a register, then define this macro to be
3777 the number (greater than or equal to zero) of that register.
3779 This should only need to be defined if @code{TARGET_WRITE_SP} and
3780 @code{TARGET_WRITE_SP} are not defined.
3782 @item STAB_REG_TO_REGNUM
3783 @findex STAB_REG_TO_REGNUM
3784 Define this to convert stab register numbers (as gotten from `r'
3785 declarations) into @value{GDBN} regnums. If not defined, no conversion will be
3788 @item STACK_ALIGN (@var{addr})
3789 @anchor{STACK_ALIGN}
3791 Define this to increase @var{addr} so that it meets the alignment
3792 requirements for the processor's stack.
3794 Unlike @ref{frame_align}, this function always adjusts @var{addr}
3797 By default, no stack alignment is performed.
3799 @item STEP_SKIPS_DELAY (@var{addr})
3800 @findex STEP_SKIPS_DELAY
3801 Define this to return true if the address is of an instruction with a
3802 delay slot. If a breakpoint has been placed in the instruction's delay
3803 slot, @value{GDBN} will single-step over that instruction before resuming
3804 normally. Currently only defined for the Mips.
3806 @item STORE_RETURN_VALUE (@var{type}, @var{regcache}, @var{valbuf})
3807 @findex STORE_RETURN_VALUE
3808 A C expression that writes the function return value, found in
3809 @var{valbuf}, into the @var{regcache}. @var{type} is the type of the
3810 value that is to be returned.
3812 @item SUN_FIXED_LBRAC_BUG
3813 @findex SUN_FIXED_LBRAC_BUG
3814 (Used only for Sun-3 and Sun-4 targets.)
3816 @item SYMBOL_RELOADING_DEFAULT
3817 @findex SYMBOL_RELOADING_DEFAULT
3818 The default value of the ``symbol-reloading'' variable. (Never defined in
3821 @item TARGET_CHAR_BIT
3822 @findex TARGET_CHAR_BIT
3823 Number of bits in a char; defaults to 8.
3825 @item TARGET_CHAR_SIGNED
3826 @findex TARGET_CHAR_SIGNED
3827 Non-zero if @code{char} is normally signed on this architecture; zero if
3828 it should be unsigned.
3830 The ISO C standard requires the compiler to treat @code{char} as
3831 equivalent to either @code{signed char} or @code{unsigned char}; any
3832 character in the standard execution set is supposed to be positive.
3833 Most compilers treat @code{char} as signed, but @code{char} is unsigned
3834 on the IBM S/390, RS6000, and PowerPC targets.
3836 @item TARGET_COMPLEX_BIT
3837 @findex TARGET_COMPLEX_BIT
3838 Number of bits in a complex number; defaults to @code{2 * TARGET_FLOAT_BIT}.
3840 At present this macro is not used.
3842 @item TARGET_DOUBLE_BIT
3843 @findex TARGET_DOUBLE_BIT
3844 Number of bits in a double float; defaults to @code{8 * TARGET_CHAR_BIT}.
3846 @item TARGET_DOUBLE_COMPLEX_BIT
3847 @findex TARGET_DOUBLE_COMPLEX_BIT
3848 Number of bits in a double complex; defaults to @code{2 * TARGET_DOUBLE_BIT}.
3850 At present this macro is not used.
3852 @item TARGET_FLOAT_BIT
3853 @findex TARGET_FLOAT_BIT
3854 Number of bits in a float; defaults to @code{4 * TARGET_CHAR_BIT}.
3856 @item TARGET_INT_BIT
3857 @findex TARGET_INT_BIT
3858 Number of bits in an integer; defaults to @code{4 * TARGET_CHAR_BIT}.
3860 @item TARGET_LONG_BIT
3861 @findex TARGET_LONG_BIT
3862 Number of bits in a long integer; defaults to @code{4 * TARGET_CHAR_BIT}.
3864 @item TARGET_LONG_DOUBLE_BIT
3865 @findex TARGET_LONG_DOUBLE_BIT
3866 Number of bits in a long double float;
3867 defaults to @code{2 * TARGET_DOUBLE_BIT}.
3869 @item TARGET_LONG_LONG_BIT
3870 @findex TARGET_LONG_LONG_BIT
3871 Number of bits in a long long integer; defaults to @code{2 * TARGET_LONG_BIT}.
3873 @item TARGET_PTR_BIT
3874 @findex TARGET_PTR_BIT
3875 Number of bits in a pointer; defaults to @code{TARGET_INT_BIT}.
3877 @item TARGET_SHORT_BIT
3878 @findex TARGET_SHORT_BIT
3879 Number of bits in a short integer; defaults to @code{2 * TARGET_CHAR_BIT}.
3881 @item TARGET_READ_PC
3882 @findex TARGET_READ_PC
3883 @itemx TARGET_WRITE_PC (@var{val}, @var{pid})
3884 @findex TARGET_WRITE_PC
3885 @itemx TARGET_READ_SP
3886 @findex TARGET_READ_SP
3887 @itemx TARGET_WRITE_SP
3888 @findex TARGET_WRITE_SP
3889 @itemx TARGET_READ_FP
3890 @findex TARGET_READ_FP
3896 These change the behavior of @code{read_pc}, @code{write_pc},
3897 @code{read_sp}, @code{write_sp} and @code{read_fp}. For most targets,
3898 these may be left undefined. @value{GDBN} will call the read and write
3899 register functions with the relevant @code{_REGNUM} argument.
3901 These macros are useful when a target keeps one of these registers in a
3902 hard to get at place; for example, part in a segment register and part
3903 in an ordinary register.
3905 @item TARGET_VIRTUAL_FRAME_POINTER(@var{pc}, @var{regp}, @var{offsetp})
3906 @findex TARGET_VIRTUAL_FRAME_POINTER
3907 Returns a @code{(register, offset)} pair representing the virtual
3908 frame pointer in use at the code address @var{pc}. If virtual
3909 frame pointers are not used, a default definition simply returns
3910 @code{FP_REGNUM}, with an offset of zero.
3912 @item TARGET_HAS_HARDWARE_WATCHPOINTS
3913 If non-zero, the target has support for hardware-assisted
3914 watchpoints. @xref{Algorithms, watchpoints}, for more details and
3915 other related macros.
3917 @item TARGET_PRINT_INSN (@var{addr}, @var{info})
3918 @findex TARGET_PRINT_INSN
3919 This is the function used by @value{GDBN} to print an assembly
3920 instruction. It prints the instruction at address @var{addr} in
3921 debugged memory and returns the length of the instruction, in bytes. If
3922 a target doesn't define its own printing routine, it defaults to an
3923 accessor function for the global pointer @code{tm_print_insn}. This
3924 usually points to a function in the @code{opcodes} library (@pxref{Support
3925 Libraries, ,Opcodes}). @var{info} is a structure (of type
3926 @code{disassemble_info}) defined in @file{include/dis-asm.h} used to
3927 pass information to the instruction decoding routine.
3929 @item USE_STRUCT_CONVENTION (@var{gcc_p}, @var{type})
3930 @findex USE_STRUCT_CONVENTION
3931 If defined, this must be an expression that is nonzero if a value of the
3932 given @var{type} being returned from a function must have space
3933 allocated for it on the stack. @var{gcc_p} is true if the function
3934 being considered is known to have been compiled by GCC; this is helpful
3935 for systems where GCC is known to use different calling convention than
3938 @item VALUE_TO_REGISTER(@var{type}, @var{regnum}, @var{from}, @var{to})
3939 @findex VALUE_TO_REGISTER
3940 Convert a value of type @var{type} into the raw contents of register
3942 @xref{Target Architecture Definition, , Using Different Register and Memory Data Representations}.
3944 @item VARIABLES_INSIDE_BLOCK (@var{desc}, @var{gcc_p})
3945 @findex VARIABLES_INSIDE_BLOCK
3946 For dbx-style debugging information, if the compiler puts variable
3947 declarations inside LBRAC/RBRAC blocks, this should be defined to be
3948 nonzero. @var{desc} is the value of @code{n_desc} from the
3949 @code{N_RBRAC} symbol, and @var{gcc_p} is true if @value{GDBN} has noticed the
3950 presence of either the @code{GCC_COMPILED_SYMBOL} or the
3951 @code{GCC2_COMPILED_SYMBOL}. By default, this is 0.
3953 @item OS9K_VARIABLES_INSIDE_BLOCK (@var{desc}, @var{gcc_p})
3954 @findex OS9K_VARIABLES_INSIDE_BLOCK
3955 Similarly, for OS/9000. Defaults to 1.
3958 Motorola M68K target conditionals.
3962 Define this to be the 4-bit location of the breakpoint trap vector. If
3963 not defined, it will default to @code{0xf}.
3965 @item REMOTE_BPT_VECTOR
3966 Defaults to @code{1}.
3968 @item NAME_OF_MALLOC
3969 @findex NAME_OF_MALLOC
3970 A string containing the name of the function to call in order to
3971 allocate some memory in the inferior. The default value is "malloc".
3975 @section Adding a New Target
3977 @cindex adding a target
3978 The following files add a target to @value{GDBN}:
3982 @item gdb/config/@var{arch}/@var{ttt}.mt
3983 Contains a Makefile fragment specific to this target. Specifies what
3984 object files are needed for target @var{ttt}, by defining
3985 @samp{TDEPFILES=@dots{}} and @samp{TDEPLIBS=@dots{}}. Also specifies
3986 the header file which describes @var{ttt}, by defining @samp{TM_FILE=
3989 You can also define @samp{TM_CFLAGS}, @samp{TM_CLIBS}, @samp{TM_CDEPS},
3990 but these are now deprecated, replaced by autoconf, and may go away in
3991 future versions of @value{GDBN}.
3993 @item gdb/@var{ttt}-tdep.c
3994 Contains any miscellaneous code required for this target machine. On
3995 some machines it doesn't exist at all. Sometimes the macros in
3996 @file{tm-@var{ttt}.h} become very complicated, so they are implemented
3997 as functions here instead, and the macro is simply defined to call the
3998 function. This is vastly preferable, since it is easier to understand
4001 @item gdb/@var{arch}-tdep.c
4002 @itemx gdb/@var{arch}-tdep.h
4003 This often exists to describe the basic layout of the target machine's
4004 processor chip (registers, stack, etc.). If used, it is included by
4005 @file{@var{ttt}-tdep.h}. It can be shared among many targets that use
4008 @item gdb/config/@var{arch}/tm-@var{ttt}.h
4009 (@file{tm.h} is a link to this file, created by @code{configure}). Contains
4010 macro definitions about the target machine's registers, stack frame
4011 format and instructions.
4013 New targets do not need this file and should not create it.
4015 @item gdb/config/@var{arch}/tm-@var{arch}.h
4016 This often exists to describe the basic layout of the target machine's
4017 processor chip (registers, stack, etc.). If used, it is included by
4018 @file{tm-@var{ttt}.h}. It can be shared among many targets that use the
4021 New targets do not need this file and should not create it.
4025 If you are adding a new operating system for an existing CPU chip, add a
4026 @file{config/tm-@var{os}.h} file that describes the operating system
4027 facilities that are unusual (extra symbol table info; the breakpoint
4028 instruction needed; etc.). Then write a @file{@var{arch}/tm-@var{os}.h}
4029 that just @code{#include}s @file{tm-@var{arch}.h} and
4030 @file{config/tm-@var{os}.h}.
4033 @section Converting an existing Target Architecture to Multi-arch
4034 @cindex converting targets to multi-arch
4036 This section describes the current accepted best practice for converting
4037 an existing target architecture to the multi-arch framework.
4039 The process consists of generating, testing, posting and committing a
4040 sequence of patches. Each patch must contain a single change, for
4046 Directly convert a group of functions into macros (the conversion does
4047 not change the behavior of any of the functions).
4050 Replace a non-multi-arch with a multi-arch mechanism (e.g.,
4054 Enable multi-arch level one.
4057 Delete one or more files.
4062 There isn't a size limit on a patch, however, a developer is strongly
4063 encouraged to keep the patch size down.
4065 Since each patch is well defined, and since each change has been tested
4066 and shows no regressions, the patches are considered @emph{fairly}
4067 obvious. Such patches, when submitted by developers listed in the
4068 @file{MAINTAINERS} file, do not need approval. Occasional steps in the
4069 process may be more complicated and less clear. The developer is
4070 expected to use their judgment and is encouraged to seek advice as
4073 @subsection Preparation
4075 The first step is to establish control. Build (with @option{-Werror}
4076 enabled) and test the target so that there is a baseline against which
4077 the debugger can be compared.
4079 At no stage can the test results regress or @value{GDBN} stop compiling
4080 with @option{-Werror}.
4082 @subsection Add the multi-arch initialization code
4084 The objective of this step is to establish the basic multi-arch
4085 framework. It involves
4090 The addition of a @code{@var{arch}_gdbarch_init} function@footnote{The
4091 above is from the original example and uses K&R C. @value{GDBN}
4092 has since converted to ISO C but lets ignore that.} that creates
4095 static struct gdbarch *
4096 d10v_gdbarch_init (info, arches)
4097 struct gdbarch_info info;
4098 struct gdbarch_list *arches;
4100 struct gdbarch *gdbarch;
4101 /* there is only one d10v architecture */
4103 return arches->gdbarch;
4104 gdbarch = gdbarch_alloc (&info, NULL);
4112 A per-architecture dump function to print any architecture specific
4116 mips_dump_tdep (struct gdbarch *current_gdbarch,
4117 struct ui_file *file)
4119 @dots{} code to print architecture specific info @dots{}
4124 A change to @code{_initialize_@var{arch}_tdep} to register this new
4128 _initialize_mips_tdep (void)
4130 gdbarch_register (bfd_arch_mips, mips_gdbarch_init,
4135 Add the macro @code{GDB_MULTI_ARCH}, defined as 0 (zero), to the file@*
4136 @file{config/@var{arch}/tm-@var{arch}.h}.
4140 @subsection Update multi-arch incompatible mechanisms
4142 Some mechanisms do not work with multi-arch. They include:
4145 @item EXTRA_FRAME_INFO
4147 @item FRAME_FIND_SAVED_REGS
4148 Replaced with @code{FRAME_INIT_SAVED_REGS}
4152 At this stage you could also consider converting the macros into
4155 @subsection Prepare for multi-arch level to one
4157 Temporally set @code{GDB_MULTI_ARCH} to @code{GDB_MULTI_ARCH_PARTIAL}
4158 and then build and start @value{GDBN} (the change should not be
4159 committed). @value{GDBN} may not build, and once built, it may die with
4160 an internal error listing the architecture methods that must be
4163 Fix any build problems (patch(es)).
4165 Convert all the architecture methods listed, which are only macros, into
4166 functions (patch(es)).
4168 Update @code{@var{arch}_gdbarch_init} to set all the missing
4169 architecture methods and wrap the corresponding macros in @code{#if
4170 !GDB_MULTI_ARCH} (patch(es)).
4172 @subsection Set multi-arch level one
4174 Change the value of @code{GDB_MULTI_ARCH} to GDB_MULTI_ARCH_PARTIAL (a
4177 Any problems with throwing ``the switch'' should have been fixed
4180 @subsection Convert remaining macros
4182 Suggest converting macros into functions (and setting the corresponding
4183 architecture method) in small batches.
4185 @subsection Set multi-arch level to two
4187 This should go smoothly.
4189 @subsection Delete the TM file
4191 The @file{tm-@var{arch}.h} can be deleted. @file{@var{arch}.mt} and
4192 @file{configure.in} updated.
4195 @node Target Vector Definition
4197 @chapter Target Vector Definition
4198 @cindex target vector
4200 The target vector defines the interface between @value{GDBN}'s
4201 abstract handling of target systems, and the nitty-gritty code that
4202 actually exercises control over a process or a serial port.
4203 @value{GDBN} includes some 30-40 different target vectors; however,
4204 each configuration of @value{GDBN} includes only a few of them.
4206 @section File Targets
4208 Both executables and core files have target vectors.
4210 @section Standard Protocol and Remote Stubs
4212 @value{GDBN}'s file @file{remote.c} talks a serial protocol to code
4213 that runs in the target system. @value{GDBN} provides several sample
4214 @dfn{stubs} that can be integrated into target programs or operating
4215 systems for this purpose; they are named @file{*-stub.c}.
4217 The @value{GDBN} user's manual describes how to put such a stub into
4218 your target code. What follows is a discussion of integrating the
4219 SPARC stub into a complicated operating system (rather than a simple
4220 program), by Stu Grossman, the author of this stub.
4222 The trap handling code in the stub assumes the following upon entry to
4227 %l1 and %l2 contain pc and npc respectively at the time of the trap;
4233 you are in the correct trap window.
4236 As long as your trap handler can guarantee those conditions, then there
4237 is no reason why you shouldn't be able to ``share'' traps with the stub.
4238 The stub has no requirement that it be jumped to directly from the
4239 hardware trap vector. That is why it calls @code{exceptionHandler()},
4240 which is provided by the external environment. For instance, this could
4241 set up the hardware traps to actually execute code which calls the stub
4242 first, and then transfers to its own trap handler.
4244 For the most point, there probably won't be much of an issue with
4245 ``sharing'' traps, as the traps we use are usually not used by the kernel,
4246 and often indicate unrecoverable error conditions. Anyway, this is all
4247 controlled by a table, and is trivial to modify. The most important
4248 trap for us is for @code{ta 1}. Without that, we can't single step or
4249 do breakpoints. Everything else is unnecessary for the proper operation
4250 of the debugger/stub.
4252 From reading the stub, it's probably not obvious how breakpoints work.
4253 They are simply done by deposit/examine operations from @value{GDBN}.
4255 @section ROM Monitor Interface
4257 @section Custom Protocols
4259 @section Transport Layer
4261 @section Builtin Simulator
4264 @node Native Debugging
4266 @chapter Native Debugging
4267 @cindex native debugging
4269 Several files control @value{GDBN}'s configuration for native support:
4273 @item gdb/config/@var{arch}/@var{xyz}.mh
4274 Specifies Makefile fragments needed by a @emph{native} configuration on
4275 machine @var{xyz}. In particular, this lists the required
4276 native-dependent object files, by defining @samp{NATDEPFILES=@dots{}}.
4277 Also specifies the header file which describes native support on
4278 @var{xyz}, by defining @samp{NAT_FILE= nm-@var{xyz}.h}. You can also
4279 define @samp{NAT_CFLAGS}, @samp{NAT_ADD_FILES}, @samp{NAT_CLIBS},
4280 @samp{NAT_CDEPS}, etc.; see @file{Makefile.in}.
4282 @emph{Maintainer's note: The @file{.mh} suffix is because this file
4283 originally contained @file{Makefile} fragments for hosting @value{GDBN}
4284 on machine @var{xyz}. While the file is no longer used for this
4285 purpose, the @file{.mh} suffix remains. Perhaps someone will
4286 eventually rename these fragments so that they have a @file{.mn}
4289 @item gdb/config/@var{arch}/nm-@var{xyz}.h
4290 (@file{nm.h} is a link to this file, created by @code{configure}). Contains C
4291 macro definitions describing the native system environment, such as
4292 child process control and core file support.
4294 @item gdb/@var{xyz}-nat.c
4295 Contains any miscellaneous C code required for this native support of
4296 this machine. On some machines it doesn't exist at all.
4299 There are some ``generic'' versions of routines that can be used by
4300 various systems. These can be customized in various ways by macros
4301 defined in your @file{nm-@var{xyz}.h} file. If these routines work for
4302 the @var{xyz} host, you can just include the generic file's name (with
4303 @samp{.o}, not @samp{.c}) in @code{NATDEPFILES}.
4305 Otherwise, if your machine needs custom support routines, you will need
4306 to write routines that perform the same functions as the generic file.
4307 Put them into @file{@var{xyz}-nat.c}, and put @file{@var{xyz}-nat.o}
4308 into @code{NATDEPFILES}.
4312 This contains the @emph{target_ops vector} that supports Unix child
4313 processes on systems which use ptrace and wait to control the child.
4316 This contains the @emph{target_ops vector} that supports Unix child
4317 processes on systems which use /proc to control the child.
4320 This does the low-level grunge that uses Unix system calls to do a ``fork
4321 and exec'' to start up a child process.
4324 This is the low level interface to inferior processes for systems using
4325 the Unix @code{ptrace} call in a vanilla way.
4328 @section Native core file Support
4329 @cindex native core files
4332 @findex fetch_core_registers
4333 @item core-aout.c::fetch_core_registers()
4334 Support for reading registers out of a core file. This routine calls
4335 @code{register_addr()}, see below. Now that BFD is used to read core
4336 files, virtually all machines should use @code{core-aout.c}, and should
4337 just provide @code{fetch_core_registers} in @code{@var{xyz}-nat.c} (or
4338 @code{REGISTER_U_ADDR} in @code{nm-@var{xyz}.h}).
4340 @item core-aout.c::register_addr()
4341 If your @code{nm-@var{xyz}.h} file defines the macro
4342 @code{REGISTER_U_ADDR(addr, blockend, regno)}, it should be defined to
4343 set @code{addr} to the offset within the @samp{user} struct of @value{GDBN}
4344 register number @code{regno}. @code{blockend} is the offset within the
4345 ``upage'' of @code{u.u_ar0}. If @code{REGISTER_U_ADDR} is defined,
4346 @file{core-aout.c} will define the @code{register_addr()} function and
4347 use the macro in it. If you do not define @code{REGISTER_U_ADDR}, but
4348 you are using the standard @code{fetch_core_registers()}, you will need
4349 to define your own version of @code{register_addr()}, put it into your
4350 @code{@var{xyz}-nat.c} file, and be sure @code{@var{xyz}-nat.o} is in
4351 the @code{NATDEPFILES} list. If you have your own
4352 @code{fetch_core_registers()}, you may not need a separate
4353 @code{register_addr()}. Many custom @code{fetch_core_registers()}
4354 implementations simply locate the registers themselves.@refill
4357 When making @value{GDBN} run native on a new operating system, to make it
4358 possible to debug core files, you will need to either write specific
4359 code for parsing your OS's core files, or customize
4360 @file{bfd/trad-core.c}. First, use whatever @code{#include} files your
4361 machine uses to define the struct of registers that is accessible
4362 (possibly in the u-area) in a core file (rather than
4363 @file{machine/reg.h}), and an include file that defines whatever header
4364 exists on a core file (e.g. the u-area or a @code{struct core}). Then
4365 modify @code{trad_unix_core_file_p} to use these values to set up the
4366 section information for the data segment, stack segment, any other
4367 segments in the core file (perhaps shared library contents or control
4368 information), ``registers'' segment, and if there are two discontiguous
4369 sets of registers (e.g. integer and float), the ``reg2'' segment. This
4370 section information basically delimits areas in the core file in a
4371 standard way, which the section-reading routines in BFD know how to seek
4374 Then back in @value{GDBN}, you need a matching routine called
4375 @code{fetch_core_registers}. If you can use the generic one, it's in
4376 @file{core-aout.c}; if not, it's in your @file{@var{xyz}-nat.c} file.
4377 It will be passed a char pointer to the entire ``registers'' segment,
4378 its length, and a zero; or a char pointer to the entire ``regs2''
4379 segment, its length, and a 2. The routine should suck out the supplied
4380 register values and install them into @value{GDBN}'s ``registers'' array.
4382 If your system uses @file{/proc} to control processes, and uses ELF
4383 format core files, then you may be able to use the same routines for
4384 reading the registers out of processes and out of core files.
4392 @section shared libraries
4394 @section Native Conditionals
4395 @cindex native conditionals
4397 When @value{GDBN} is configured and compiled, various macros are
4398 defined or left undefined, to control compilation when the host and
4399 target systems are the same. These macros should be defined (or left
4400 undefined) in @file{nm-@var{system}.h}.
4404 @findex ATTACH_DETACH
4405 If defined, then @value{GDBN} will include support for the @code{attach} and
4406 @code{detach} commands.
4408 @item CHILD_PREPARE_TO_STORE
4409 @findex CHILD_PREPARE_TO_STORE
4410 If the machine stores all registers at once in the child process, then
4411 define this to ensure that all values are correct. This usually entails
4412 a read from the child.
4414 [Note that this is incorrectly defined in @file{xm-@var{system}.h} files
4417 @item FETCH_INFERIOR_REGISTERS
4418 @findex FETCH_INFERIOR_REGISTERS
4419 Define this if the native-dependent code will provide its own routines
4420 @code{fetch_inferior_registers} and @code{store_inferior_registers} in
4421 @file{@var{host}-nat.c}. If this symbol is @emph{not} defined, and
4422 @file{infptrace.c} is included in this configuration, the default
4423 routines in @file{infptrace.c} are used for these functions.
4425 @item FILES_INFO_HOOK
4426 @findex FILES_INFO_HOOK
4427 (Only defined for Convex.)
4431 This macro is normally defined to be the number of the first floating
4432 point register, if the machine has such registers. As such, it would
4433 appear only in target-specific code. However, @file{/proc} support uses this
4434 to decide whether floats are in use on this target.
4436 @item GET_LONGJMP_TARGET
4437 @findex GET_LONGJMP_TARGET
4438 For most machines, this is a target-dependent parameter. On the
4439 DECstation and the Iris, this is a native-dependent parameter, since
4440 @file{setjmp.h} is needed to define it.
4442 This macro determines the target PC address that @code{longjmp} will jump to,
4443 assuming that we have just stopped at a longjmp breakpoint. It takes a
4444 @code{CORE_ADDR *} as argument, and stores the target PC value through this
4445 pointer. It examines the current state of the machine as needed.
4447 @item I386_USE_GENERIC_WATCHPOINTS
4448 An x86-based machine can define this to use the generic x86 watchpoint
4449 support; see @ref{Algorithms, I386_USE_GENERIC_WATCHPOINTS}.
4452 @findex KERNEL_U_ADDR
4453 Define this to the address of the @code{u} structure (the ``user
4454 struct'', also known as the ``u-page'') in kernel virtual memory. @value{GDBN}
4455 needs to know this so that it can subtract this address from absolute
4456 addresses in the upage, that are obtained via ptrace or from core files.
4457 On systems that don't need this value, set it to zero.
4459 @item KERNEL_U_ADDR_BSD
4460 @findex KERNEL_U_ADDR_BSD
4461 Define this to cause @value{GDBN} to determine the address of @code{u} at
4462 runtime, by using Berkeley-style @code{nlist} on the kernel's image in
4465 @item KERNEL_U_ADDR_HPUX
4466 @findex KERNEL_U_ADDR_HPUX
4467 Define this to cause @value{GDBN} to determine the address of @code{u} at
4468 runtime, by using HP-style @code{nlist} on the kernel's image in the
4471 @item ONE_PROCESS_WRITETEXT
4472 @findex ONE_PROCESS_WRITETEXT
4473 Define this to be able to, when a breakpoint insertion fails, warn the
4474 user that another process may be running with the same executable.
4476 @item PREPARE_TO_PROCEED (@var{select_it})
4477 @findex PREPARE_TO_PROCEED
4478 This (ugly) macro allows a native configuration to customize the way the
4479 @code{proceed} function in @file{infrun.c} deals with switching between
4482 In a multi-threaded task we may select another thread and then continue
4483 or step. But if the old thread was stopped at a breakpoint, it will
4484 immediately cause another breakpoint stop without any execution (i.e. it
4485 will report a breakpoint hit incorrectly). So @value{GDBN} must step over it
4488 If defined, @code{PREPARE_TO_PROCEED} should check the current thread
4489 against the thread that reported the most recent event. If a step-over
4490 is required, it returns TRUE. If @var{select_it} is non-zero, it should
4491 reselect the old thread.
4494 @findex PROC_NAME_FMT
4495 Defines the format for the name of a @file{/proc} device. Should be
4496 defined in @file{nm.h} @emph{only} in order to override the default
4497 definition in @file{procfs.c}.
4500 @findex PTRACE_FP_BUG
4501 See @file{mach386-xdep.c}.
4503 @item PTRACE_ARG3_TYPE
4504 @findex PTRACE_ARG3_TYPE
4505 The type of the third argument to the @code{ptrace} system call, if it
4506 exists and is different from @code{int}.
4508 @item REGISTER_U_ADDR
4509 @findex REGISTER_U_ADDR
4510 Defines the offset of the registers in the ``u area''.
4512 @item SHELL_COMMAND_CONCAT
4513 @findex SHELL_COMMAND_CONCAT
4514 If defined, is a string to prefix on the shell command used to start the
4519 If defined, this is the name of the shell to use to run the inferior.
4520 Defaults to @code{"/bin/sh"}.
4522 @item SOLIB_ADD (@var{filename}, @var{from_tty}, @var{targ}, @var{readsyms})
4524 Define this to expand into an expression that will cause the symbols in
4525 @var{filename} to be added to @value{GDBN}'s symbol table. If
4526 @var{readsyms} is zero symbols are not read but any necessary low level
4527 processing for @var{filename} is still done.
4529 @item SOLIB_CREATE_INFERIOR_HOOK
4530 @findex SOLIB_CREATE_INFERIOR_HOOK
4531 Define this to expand into any shared-library-relocation code that you
4532 want to be run just after the child process has been forked.
4534 @item START_INFERIOR_TRAPS_EXPECTED
4535 @findex START_INFERIOR_TRAPS_EXPECTED
4536 When starting an inferior, @value{GDBN} normally expects to trap
4538 the shell execs, and once when the program itself execs. If the actual
4539 number of traps is something other than 2, then define this macro to
4540 expand into the number expected.
4542 @item SVR4_SHARED_LIBS
4543 @findex SVR4_SHARED_LIBS
4544 Define this to indicate that SVR4-style shared libraries are in use.
4548 This determines whether small routines in @file{*-tdep.c}, which
4549 translate register values between @value{GDBN}'s internal
4550 representation and the @file{/proc} representation, are compiled.
4553 @findex U_REGS_OFFSET
4554 This is the offset of the registers in the upage. It need only be
4555 defined if the generic ptrace register access routines in
4556 @file{infptrace.c} are being used (that is, @file{infptrace.c} is
4557 configured in, and @code{FETCH_INFERIOR_REGISTERS} is not defined). If
4558 the default value from @file{infptrace.c} is good enough, leave it
4561 The default value means that u.u_ar0 @emph{points to} the location of
4562 the registers. I'm guessing that @code{#define U_REGS_OFFSET 0} means
4563 that @code{u.u_ar0} @emph{is} the location of the registers.
4567 See @file{objfiles.c}.
4570 @findex DEBUG_PTRACE
4571 Define this to debug @code{ptrace} calls.
4575 @node Support Libraries
4577 @chapter Support Libraries
4582 BFD provides support for @value{GDBN} in several ways:
4585 @item identifying executable and core files
4586 BFD will identify a variety of file types, including a.out, coff, and
4587 several variants thereof, as well as several kinds of core files.
4589 @item access to sections of files
4590 BFD parses the file headers to determine the names, virtual addresses,
4591 sizes, and file locations of all the various named sections in files
4592 (such as the text section or the data section). @value{GDBN} simply
4593 calls BFD to read or write section @var{x} at byte offset @var{y} for
4596 @item specialized core file support
4597 BFD provides routines to determine the failing command name stored in a
4598 core file, the signal with which the program failed, and whether a core
4599 file matches (i.e.@: could be a core dump of) a particular executable
4602 @item locating the symbol information
4603 @value{GDBN} uses an internal interface of BFD to determine where to find the
4604 symbol information in an executable file or symbol-file. @value{GDBN} itself
4605 handles the reading of symbols, since BFD does not ``understand'' debug
4606 symbols, but @value{GDBN} uses BFD's cached information to find the symbols,
4611 @cindex opcodes library
4613 The opcodes library provides @value{GDBN}'s disassembler. (It's a separate
4614 library because it's also used in binutils, for @file{objdump}).
4623 @cindex regular expressions library
4634 @item SIGN_EXTEND_CHAR
4636 @item SWITCH_ENUM_BUG
4651 This chapter covers topics that are lower-level than the major
4652 algorithms of @value{GDBN}.
4657 Cleanups are a structured way to deal with things that need to be done
4660 When your code does something (e.g., @code{xmalloc} some memory, or
4661 @code{open} a file) that needs to be undone later (e.g., @code{xfree}
4662 the memory or @code{close} the file), it can make a cleanup. The
4663 cleanup will be done at some future point: when the command is finished
4664 and control returns to the top level; when an error occurs and the stack
4665 is unwound; or when your code decides it's time to explicitly perform
4666 cleanups. Alternatively you can elect to discard the cleanups you
4672 @item struct cleanup *@var{old_chain};
4673 Declare a variable which will hold a cleanup chain handle.
4675 @findex make_cleanup
4676 @item @var{old_chain} = make_cleanup (@var{function}, @var{arg});
4677 Make a cleanup which will cause @var{function} to be called with
4678 @var{arg} (a @code{char *}) later. The result, @var{old_chain}, is a
4679 handle that can later be passed to @code{do_cleanups} or
4680 @code{discard_cleanups}. Unless you are going to call
4681 @code{do_cleanups} or @code{discard_cleanups}, you can ignore the result
4682 from @code{make_cleanup}.
4685 @item do_cleanups (@var{old_chain});
4686 Do all cleanups added to the chain since the corresponding
4687 @code{make_cleanup} call was made.
4689 @findex discard_cleanups
4690 @item discard_cleanups (@var{old_chain});
4691 Same as @code{do_cleanups} except that it just removes the cleanups from
4692 the chain and does not call the specified functions.
4695 Cleanups are implemented as a chain. The handle returned by
4696 @code{make_cleanups} includes the cleanup passed to the call and any
4697 later cleanups appended to the chain (but not yet discarded or
4701 make_cleanup (a, 0);
4703 struct cleanup *old = make_cleanup (b, 0);
4711 will call @code{c()} and @code{b()} but will not call @code{a()}. The
4712 cleanup that calls @code{a()} will remain in the cleanup chain, and will
4713 be done later unless otherwise discarded.@refill
4715 Your function should explicitly do or discard the cleanups it creates.
4716 Failing to do this leads to non-deterministic behavior since the caller
4717 will arbitrarily do or discard your functions cleanups. This need leads
4718 to two common cleanup styles.
4720 The first style is try/finally. Before it exits, your code-block calls
4721 @code{do_cleanups} with the old cleanup chain and thus ensures that your
4722 code-block's cleanups are always performed. For instance, the following
4723 code-segment avoids a memory leak problem (even when @code{error} is
4724 called and a forced stack unwind occurs) by ensuring that the
4725 @code{xfree} will always be called:
4728 struct cleanup *old = make_cleanup (null_cleanup, 0);
4729 data = xmalloc (sizeof blah);
4730 make_cleanup (xfree, data);
4735 The second style is try/except. Before it exits, your code-block calls
4736 @code{discard_cleanups} with the old cleanup chain and thus ensures that
4737 any created cleanups are not performed. For instance, the following
4738 code segment, ensures that the file will be closed but only if there is
4742 FILE *file = fopen ("afile", "r");
4743 struct cleanup *old = make_cleanup (close_file, file);
4745 discard_cleanups (old);
4749 Some functions, e.g. @code{fputs_filtered()} or @code{error()}, specify
4750 that they ``should not be called when cleanups are not in place''. This
4751 means that any actions you need to reverse in the case of an error or
4752 interruption must be on the cleanup chain before you call these
4753 functions, since they might never return to your code (they
4754 @samp{longjmp} instead).
4756 @section Per-architecture module data
4757 @cindex per-architecture module data
4758 @cindex multi-arch data
4759 @cindex data-pointer, per-architecture/per-module
4761 The multi-arch framework includes a mechanism for adding module specific
4762 per-architecture data-pointers to the @code{struct gdbarch} architecture
4765 A module registers one or more per-architecture data-pointers using the
4766 function @code{register_gdbarch_data}:
4768 @deftypefun struct gdbarch_data *register_gdbarch_data (gdbarch_data_init_ftype *@var{init}, gdbarch_data_free_ftype *@var{free})
4770 The @var{init} function is used to obtain an initial value for a
4771 per-architecture data-pointer. The function is called, after the
4772 architecture has been created, when the data-pointer is still
4773 uninitialized (@code{NULL}) and its value has been requested via a call
4774 to @code{gdbarch_data}. A data-pointer can also be initialize
4775 explicitly using @code{set_gdbarch_data}.
4777 The @var{free} function is called when a data-pointer needs to be
4778 destroyed. This occurs when either the corresponding @code{struct
4779 gdbarch} object is being destroyed or when @code{set_gdbarch_data} is
4780 overriding a non-@code{NULL} data-pointer value.
4782 The function @code{register_gdbarch_data} returns a @code{struct
4783 gdbarch_data} that is used to identify the data-pointer that was added
4788 A typical module has @code{init} and @code{free} functions of the form:
4791 static struct gdbarch_data *nozel_handle;
4793 nozel_init (struct gdbarch *gdbarch)
4795 struct nozel *data = XMALLOC (struct nozel);
4801 nozel_free (struct gdbarch *gdbarch, void *data)
4807 Since uninitialized (@code{NULL}) data-pointers are initialized
4808 on-demand, an @code{init} function is free to call other modules that
4809 use data-pointers. Those modules data-pointers will be initialized as
4810 needed. Care should be taken to ensure that the @code{init} call graph
4811 does not contain cycles.
4813 The data-pointer is registered with the call:
4817 _initialize_nozel (void)
4819 nozel_handle = register_gdbarch_data (nozel_init, nozel_free);
4823 The per-architecture data-pointer is accessed using the function:
4825 @deftypefun void *gdbarch_data (struct gdbarch *@var{gdbarch}, struct gdbarch_data *@var{data_handle})
4826 Given the architecture @var{arch} and module data handle
4827 @var{data_handle} (returned by @code{register_gdbarch_data}, this
4828 function returns the current value of the per-architecture data-pointer.
4831 The non-@code{NULL} data-pointer returned by @code{gdbarch_data} should
4832 be saved in a local variable and then used directly:
4836 nozel_total (struct gdbarch *gdbarch)
4839 struct nozel *data = gdbarch_data (gdbarch, nozel_handle);
4845 It is also possible to directly initialize the data-pointer using:
4847 @deftypefun void set_gdbarch_data (struct gdbarch *@var{gdbarch}, struct gdbarch_data *handle, void *@var{pointer})
4848 Update the data-pointer corresponding to @var{handle} with the value of
4849 @var{pointer}. If the previous data-pointer value is non-NULL, then it
4850 is freed using data-pointers @var{free} function.
4853 This function is used by modules that require a mechanism for explicitly
4854 setting the per-architecture data-pointer during architecture creation:
4857 /* Called during architecture creation. */
4859 set_gdbarch_nozel (struct gdbarch *gdbarch,
4862 struct nozel *data = XMALLOC (struct nozel);
4864 set_gdbarch_data (gdbarch, nozel_handle, nozel);
4869 /* Default, called when nozel not set by set_gdbarch_nozel(). */
4871 nozel_init (struct gdbarch *gdbarch)
4873 struct nozel *default_nozel = XMALLOC (struc nozel);
4875 return default_nozel;
4881 _initialize_nozel (void)
4883 nozel_handle = register_gdbarch_data (nozel_init, NULL);
4888 Note that an @code{init} function still needs to be registered. It is
4889 used to initialize the data-pointer when the architecture creation phase
4890 fail to set an initial value.
4893 @section Wrapping Output Lines
4894 @cindex line wrap in output
4897 Output that goes through @code{printf_filtered} or @code{fputs_filtered}
4898 or @code{fputs_demangled} needs only to have calls to @code{wrap_here}
4899 added in places that would be good breaking points. The utility
4900 routines will take care of actually wrapping if the line width is
4903 The argument to @code{wrap_here} is an indentation string which is
4904 printed @emph{only} if the line breaks there. This argument is saved
4905 away and used later. It must remain valid until the next call to
4906 @code{wrap_here} or until a newline has been printed through the
4907 @code{*_filtered} functions. Don't pass in a local variable and then
4910 It is usually best to call @code{wrap_here} after printing a comma or
4911 space. If you call it before printing a space, make sure that your
4912 indentation properly accounts for the leading space that will print if
4913 the line wraps there.
4915 Any function or set of functions that produce filtered output must
4916 finish by printing a newline, to flush the wrap buffer, before switching
4917 to unfiltered (@code{printf}) output. Symbol reading routines that
4918 print warnings are a good example.
4920 @section @value{GDBN} Coding Standards
4921 @cindex coding standards
4923 @value{GDBN} follows the GNU coding standards, as described in
4924 @file{etc/standards.texi}. This file is also available for anonymous
4925 FTP from GNU archive sites. @value{GDBN} takes a strict interpretation
4926 of the standard; in general, when the GNU standard recommends a practice
4927 but does not require it, @value{GDBN} requires it.
4929 @value{GDBN} follows an additional set of coding standards specific to
4930 @value{GDBN}, as described in the following sections.
4935 @value{GDBN} assumes an ISO/IEC 9899:1990 (a.k.a.@: ISO C90) compliant
4938 @value{GDBN} does not assume an ISO C or POSIX compliant C library.
4941 @subsection Memory Management
4943 @value{GDBN} does not use the functions @code{malloc}, @code{realloc},
4944 @code{calloc}, @code{free} and @code{asprintf}.
4946 @value{GDBN} uses the functions @code{xmalloc}, @code{xrealloc} and
4947 @code{xcalloc} when allocating memory. Unlike @code{malloc} et.al.@:
4948 these functions do not return when the memory pool is empty. Instead,
4949 they unwind the stack using cleanups. These functions return
4950 @code{NULL} when requested to allocate a chunk of memory of size zero.
4952 @emph{Pragmatics: By using these functions, the need to check every
4953 memory allocation is removed. These functions provide portable
4956 @value{GDBN} does not use the function @code{free}.
4958 @value{GDBN} uses the function @code{xfree} to return memory to the
4959 memory pool. Consistent with ISO-C, this function ignores a request to
4960 free a @code{NULL} pointer.
4962 @emph{Pragmatics: On some systems @code{free} fails when passed a
4963 @code{NULL} pointer.}
4965 @value{GDBN} can use the non-portable function @code{alloca} for the
4966 allocation of small temporary values (such as strings).
4968 @emph{Pragmatics: This function is very non-portable. Some systems
4969 restrict the memory being allocated to no more than a few kilobytes.}
4971 @value{GDBN} uses the string function @code{xstrdup} and the print
4972 function @code{xasprintf}.
4974 @emph{Pragmatics: @code{asprintf} and @code{strdup} can fail. Print
4975 functions such as @code{sprintf} are very prone to buffer overflow
4979 @subsection Compiler Warnings
4980 @cindex compiler warnings
4982 With few exceptions, developers should include the configuration option
4983 @samp{--enable-gdb-build-warnings=,-Werror} when building @value{GDBN}.
4984 The exceptions are listed in the file @file{gdb/MAINTAINERS}.
4986 This option causes @value{GDBN} (when built using GCC) to be compiled
4987 with a carefully selected list of compiler warning flags. Any warnings
4988 from those flags being treated as errors.
4990 The current list of warning flags includes:
4994 Since @value{GDBN} coding standard requires all functions to be declared
4995 using a prototype, the flag has the side effect of ensuring that
4996 prototyped functions are always visible with out resorting to
4997 @samp{-Wstrict-prototypes}.
5000 Such code often appears to work except on instruction set architectures
5001 that use register windows.
5008 Since @value{GDBN} uses the @code{format printf} attribute on all
5009 @code{printf} like functions this checks not just @code{printf} calls
5010 but also calls to functions such as @code{fprintf_unfiltered}.
5013 This warning includes uses of the assignment operator within an
5014 @code{if} statement.
5016 @item -Wpointer-arith
5018 @item -Wuninitialized
5021 @emph{Pragmatics: Due to the way that @value{GDBN} is implemented most
5022 functions have unused parameters. Consequently the warning
5023 @samp{-Wunused-parameter} is precluded from the list. The macro
5024 @code{ATTRIBUTE_UNUSED} is not used as it leads to false negatives ---
5025 it is not an error to have @code{ATTRIBUTE_UNUSED} on a parameter that
5026 is being used. The options @samp{-Wall} and @samp{-Wunused} are also
5027 precluded because they both include @samp{-Wunused-parameter}.}
5029 @emph{Pragmatics: @value{GDBN} has not simply accepted the warnings
5030 enabled by @samp{-Wall -Werror -W...}. Instead it is selecting warnings
5031 when and where their benefits can be demonstrated.}
5033 @subsection Formatting
5035 @cindex source code formatting
5036 The standard GNU recommendations for formatting must be followed
5039 A function declaration should not have its name in column zero. A
5040 function definition should have its name in column zero.
5044 static void foo (void);
5052 @emph{Pragmatics: This simplifies scripting. Function definitions can
5053 be found using @samp{^function-name}.}
5055 There must be a space between a function or macro name and the opening
5056 parenthesis of its argument list (except for macro definitions, as
5057 required by C). There must not be a space after an open paren/bracket
5058 or before a close paren/bracket.
5060 While additional whitespace is generally helpful for reading, do not use
5061 more than one blank line to separate blocks, and avoid adding whitespace
5062 after the end of a program line (as of 1/99, some 600 lines had
5063 whitespace after the semicolon). Excess whitespace causes difficulties
5064 for @code{diff} and @code{patch} utilities.
5066 Pointers are declared using the traditional K&R C style:
5080 @subsection Comments
5082 @cindex comment formatting
5083 The standard GNU requirements on comments must be followed strictly.
5085 Block comments must appear in the following form, with no @code{/*}- or
5086 @code{*/}-only lines, and no leading @code{*}:
5089 /* Wait for control to return from inferior to debugger. If inferior
5090 gets a signal, we may decide to start it up again instead of
5091 returning. That is why there is a loop in this function. When
5092 this function actually returns it means the inferior should be left
5093 stopped and @value{GDBN} should read more commands. */
5096 (Note that this format is encouraged by Emacs; tabbing for a multi-line
5097 comment works correctly, and @kbd{M-q} fills the block consistently.)
5099 Put a blank line between the block comments preceding function or
5100 variable definitions, and the definition itself.
5102 In general, put function-body comments on lines by themselves, rather
5103 than trying to fit them into the 20 characters left at the end of a
5104 line, since either the comment or the code will inevitably get longer
5105 than will fit, and then somebody will have to move it anyhow.
5109 @cindex C data types
5110 Code must not depend on the sizes of C data types, the format of the
5111 host's floating point numbers, the alignment of anything, or the order
5112 of evaluation of expressions.
5114 @cindex function usage
5115 Use functions freely. There are only a handful of compute-bound areas
5116 in @value{GDBN} that might be affected by the overhead of a function
5117 call, mainly in symbol reading. Most of @value{GDBN}'s performance is
5118 limited by the target interface (whether serial line or system call).
5120 However, use functions with moderation. A thousand one-line functions
5121 are just as hard to understand as a single thousand-line function.
5123 @emph{Macros are bad, M'kay.}
5124 (But if you have to use a macro, make sure that the macro arguments are
5125 protected with parentheses.)
5129 Declarations like @samp{struct foo *} should be used in preference to
5130 declarations like @samp{typedef struct foo @{ @dots{} @} *foo_ptr}.
5133 @subsection Function Prototypes
5134 @cindex function prototypes
5136 Prototypes must be used when both @emph{declaring} and @emph{defining}
5137 a function. Prototypes for @value{GDBN} functions must include both the
5138 argument type and name, with the name matching that used in the actual
5139 function definition.
5141 All external functions should have a declaration in a header file that
5142 callers include, except for @code{_initialize_*} functions, which must
5143 be external so that @file{init.c} construction works, but shouldn't be
5144 visible to random source files.
5146 Where a source file needs a forward declaration of a static function,
5147 that declaration must appear in a block near the top of the source file.
5150 @subsection Internal Error Recovery
5152 During its execution, @value{GDBN} can encounter two types of errors.
5153 User errors and internal errors. User errors include not only a user
5154 entering an incorrect command but also problems arising from corrupt
5155 object files and system errors when interacting with the target.
5156 Internal errors include situations where @value{GDBN} has detected, at
5157 run time, a corrupt or erroneous situation.
5159 When reporting an internal error, @value{GDBN} uses
5160 @code{internal_error} and @code{gdb_assert}.
5162 @value{GDBN} must not call @code{abort} or @code{assert}.
5164 @emph{Pragmatics: There is no @code{internal_warning} function. Either
5165 the code detected a user error, recovered from it and issued a
5166 @code{warning} or the code failed to correctly recover from the user
5167 error and issued an @code{internal_error}.}
5169 @subsection File Names
5171 Any file used when building the core of @value{GDBN} must be in lower
5172 case. Any file used when building the core of @value{GDBN} must be 8.3
5173 unique. These requirements apply to both source and generated files.
5175 @emph{Pragmatics: The core of @value{GDBN} must be buildable on many
5176 platforms including DJGPP and MacOS/HFS. Every time an unfriendly file
5177 is introduced to the build process both @file{Makefile.in} and
5178 @file{configure.in} need to be modified accordingly. Compare the
5179 convoluted conversion process needed to transform @file{COPYING} into
5180 @file{copying.c} with the conversion needed to transform
5181 @file{version.in} into @file{version.c}.}
5183 Any file non 8.3 compliant file (that is not used when building the core
5184 of @value{GDBN}) must be added to @file{gdb/config/djgpp/fnchange.lst}.
5186 @emph{Pragmatics: This is clearly a compromise.}
5188 When @value{GDBN} has a local version of a system header file (ex
5189 @file{string.h}) the file name based on the POSIX header prefixed with
5190 @file{gdb_} (@file{gdb_string.h}).
5192 For other files @samp{-} is used as the separator.
5195 @subsection Include Files
5197 A @file{.c} file should include @file{defs.h} first.
5199 A @file{.c} file should directly include the @code{.h} file of every
5200 declaration and/or definition it directly refers to. It cannot rely on
5203 A @file{.h} file should directly include the @code{.h} file of every
5204 declaration and/or definition it directly refers to. It cannot rely on
5205 indirect inclusion. Exception: The file @file{defs.h} does not need to
5206 be directly included.
5208 An external declaration should only appear in one include file.
5210 An external declaration should never appear in a @code{.c} file.
5211 Exception: a declaration for the @code{_initialize} function that
5212 pacifies @option{-Wmissing-declaration}.
5214 A @code{typedef} definition should only appear in one include file.
5216 An opaque @code{struct} declaration can appear in multiple @file{.h}
5217 files. Where possible, a @file{.h} file should use an opaque
5218 @code{struct} declaration instead of an include.
5220 All @file{.h} files should be wrapped in:
5223 #ifndef INCLUDE_FILE_NAME_H
5224 #define INCLUDE_FILE_NAME_H
5230 @subsection Clean Design and Portable Implementation
5233 In addition to getting the syntax right, there's the little question of
5234 semantics. Some things are done in certain ways in @value{GDBN} because long
5235 experience has shown that the more obvious ways caused various kinds of
5238 @cindex assumptions about targets
5239 You can't assume the byte order of anything that comes from a target
5240 (including @var{value}s, object files, and instructions). Such things
5241 must be byte-swapped using @code{SWAP_TARGET_AND_HOST} in
5242 @value{GDBN}, or one of the swap routines defined in @file{bfd.h},
5243 such as @code{bfd_get_32}.
5245 You can't assume that you know what interface is being used to talk to
5246 the target system. All references to the target must go through the
5247 current @code{target_ops} vector.
5249 You can't assume that the host and target machines are the same machine
5250 (except in the ``native'' support modules). In particular, you can't
5251 assume that the target machine's header files will be available on the
5252 host machine. Target code must bring along its own header files --
5253 written from scratch or explicitly donated by their owner, to avoid
5257 Insertion of new @code{#ifdef}'s will be frowned upon. It's much better
5258 to write the code portably than to conditionalize it for various
5261 @cindex system dependencies
5262 New @code{#ifdef}'s which test for specific compilers or manufacturers
5263 or operating systems are unacceptable. All @code{#ifdef}'s should test
5264 for features. The information about which configurations contain which
5265 features should be segregated into the configuration files. Experience
5266 has proven far too often that a feature unique to one particular system
5267 often creeps into other systems; and that a conditional based on some
5268 predefined macro for your current system will become worthless over
5269 time, as new versions of your system come out that behave differently
5270 with regard to this feature.
5272 Adding code that handles specific architectures, operating systems,
5273 target interfaces, or hosts, is not acceptable in generic code.
5275 @cindex portable file name handling
5276 @cindex file names, portability
5277 One particularly notorious area where system dependencies tend to
5278 creep in is handling of file names. The mainline @value{GDBN} code
5279 assumes Posix semantics of file names: absolute file names begin with
5280 a forward slash @file{/}, slashes are used to separate leading
5281 directories, case-sensitive file names. These assumptions are not
5282 necessarily true on non-Posix systems such as MS-Windows. To avoid
5283 system-dependent code where you need to take apart or construct a file
5284 name, use the following portable macros:
5287 @findex HAVE_DOS_BASED_FILE_SYSTEM
5288 @item HAVE_DOS_BASED_FILE_SYSTEM
5289 This preprocessing symbol is defined to a non-zero value on hosts
5290 whose filesystems belong to the MS-DOS/MS-Windows family. Use this
5291 symbol to write conditional code which should only be compiled for
5294 @findex IS_DIR_SEPARATOR
5295 @item IS_DIR_SEPARATOR (@var{c})
5296 Evaluates to a non-zero value if @var{c} is a directory separator
5297 character. On Unix and GNU/Linux systems, only a slash @file{/} is
5298 such a character, but on Windows, both @file{/} and @file{\} will
5301 @findex IS_ABSOLUTE_PATH
5302 @item IS_ABSOLUTE_PATH (@var{file})
5303 Evaluates to a non-zero value if @var{file} is an absolute file name.
5304 For Unix and GNU/Linux hosts, a name which begins with a slash
5305 @file{/} is absolute. On DOS and Windows, @file{d:/foo} and
5306 @file{x:\bar} are also absolute file names.
5308 @findex FILENAME_CMP
5309 @item FILENAME_CMP (@var{f1}, @var{f2})
5310 Calls a function which compares file names @var{f1} and @var{f2} as
5311 appropriate for the underlying host filesystem. For Posix systems,
5312 this simply calls @code{strcmp}; on case-insensitive filesystems it
5313 will call @code{strcasecmp} instead.
5315 @findex DIRNAME_SEPARATOR
5316 @item DIRNAME_SEPARATOR
5317 Evaluates to a character which separates directories in
5318 @code{PATH}-style lists, typically held in environment variables.
5319 This character is @samp{:} on Unix, @samp{;} on DOS and Windows.
5321 @findex SLASH_STRING
5323 This evaluates to a constant string you should use to produce an
5324 absolute filename from leading directories and the file's basename.
5325 @code{SLASH_STRING} is @code{"/"} on most systems, but might be
5326 @code{"\\"} for some Windows-based ports.
5329 In addition to using these macros, be sure to use portable library
5330 functions whenever possible. For example, to extract a directory or a
5331 basename part from a file name, use the @code{dirname} and
5332 @code{basename} library functions (available in @code{libiberty} for
5333 platforms which don't provide them), instead of searching for a slash
5334 with @code{strrchr}.
5336 Another way to generalize @value{GDBN} along a particular interface is with an
5337 attribute struct. For example, @value{GDBN} has been generalized to handle
5338 multiple kinds of remote interfaces---not by @code{#ifdef}s everywhere, but
5339 by defining the @code{target_ops} structure and having a current target (as
5340 well as a stack of targets below it, for memory references). Whenever
5341 something needs to be done that depends on which remote interface we are
5342 using, a flag in the current target_ops structure is tested (e.g.,
5343 @code{target_has_stack}), or a function is called through a pointer in the
5344 current target_ops structure. In this way, when a new remote interface
5345 is added, only one module needs to be touched---the one that actually
5346 implements the new remote interface. Other examples of
5347 attribute-structs are BFD access to multiple kinds of object file
5348 formats, or @value{GDBN}'s access to multiple source languages.
5350 Please avoid duplicating code. For example, in @value{GDBN} 3.x all
5351 the code interfacing between @code{ptrace} and the rest of
5352 @value{GDBN} was duplicated in @file{*-dep.c}, and so changing
5353 something was very painful. In @value{GDBN} 4.x, these have all been
5354 consolidated into @file{infptrace.c}. @file{infptrace.c} can deal
5355 with variations between systems the same way any system-independent
5356 file would (hooks, @code{#if defined}, etc.), and machines which are
5357 radically different don't need to use @file{infptrace.c} at all.
5359 All debugging code must be controllable using the @samp{set debug
5360 @var{module}} command. Do not use @code{printf} to print trace
5361 messages. Use @code{fprintf_unfiltered(gdb_stdlog, ...}. Do not use
5362 @code{#ifdef DEBUG}.
5367 @chapter Porting @value{GDBN}
5368 @cindex porting to new machines
5370 Most of the work in making @value{GDBN} compile on a new machine is in
5371 specifying the configuration of the machine. This is done in a
5372 dizzying variety of header files and configuration scripts, which we
5373 hope to make more sensible soon. Let's say your new host is called an
5374 @var{xyz} (e.g., @samp{sun4}), and its full three-part configuration
5375 name is @code{@var{arch}-@var{xvend}-@var{xos}} (e.g.,
5376 @samp{sparc-sun-sunos4}). In particular:
5380 In the top level directory, edit @file{config.sub} and add @var{arch},
5381 @var{xvend}, and @var{xos} to the lists of supported architectures,
5382 vendors, and operating systems near the bottom of the file. Also, add
5383 @var{xyz} as an alias that maps to
5384 @code{@var{arch}-@var{xvend}-@var{xos}}. You can test your changes by
5388 ./config.sub @var{xyz}
5395 ./config.sub @code{@var{arch}-@var{xvend}-@var{xos}}
5399 which should both respond with @code{@var{arch}-@var{xvend}-@var{xos}}
5400 and no error messages.
5403 You need to port BFD, if that hasn't been done already. Porting BFD is
5404 beyond the scope of this manual.
5407 To configure @value{GDBN} itself, edit @file{gdb/configure.host} to recognize
5408 your system and set @code{gdb_host} to @var{xyz}, and (unless your
5409 desired target is already available) also edit @file{gdb/configure.tgt},
5410 setting @code{gdb_target} to something appropriate (for instance,
5413 @emph{Maintainer's note: Work in progress. The file
5414 @file{gdb/configure.host} originally needed to be modified when either a
5415 new native target or a new host machine was being added to @value{GDBN}.
5416 Recent changes have removed this requirement. The file now only needs
5417 to be modified when adding a new native configuration. This will likely
5418 changed again in the future.}
5421 Finally, you'll need to specify and define @value{GDBN}'s host-, native-, and
5422 target-dependent @file{.h} and @file{.c} files used for your
5428 @chapter Releasing @value{GDBN}
5429 @cindex making a new release of gdb
5431 @section Versions and Branches
5433 @subsection Version Identifiers
5435 @value{GDBN}'s version is determined by the file @file{gdb/version.in}.
5437 @value{GDBN}'s mainline uses ISO dates to differentiate between
5438 versions. The CVS repository uses @var{YYYY}-@var{MM}-@var{DD}-cvs
5439 while the corresponding snapshot uses @var{YYYYMMDD}.
5441 @value{GDBN}'s release branch uses a slightly more complicated scheme.
5442 When the branch is first cut, the mainline version identifier is
5443 prefixed with the @var{major}.@var{minor} from of the previous release
5444 series but with .90 appended. As draft releases are drawn from the
5445 branch, the minor minor number (.90) is incremented. Once the first
5446 release (@var{M}.@var{N}) has been made, the version prefix is updated
5447 to @var{M}.@var{N}.0.90 (dot zero, dot ninety). Follow on releases have
5448 an incremented minor minor version number (.0).
5450 Using 5.1 (previous) and 5.2 (current), the example below illustrates a
5451 typical sequence of version identifiers:
5455 final release from previous branch
5456 @item 2002-03-03-cvs
5457 main-line the day the branch is cut
5458 @item 5.1.90-2002-03-03-cvs
5459 corresponding branch version
5461 first draft release candidate
5462 @item 5.1.91-2002-03-17-cvs
5463 updated branch version
5465 second draft release candidate
5466 @item 5.1.92-2002-03-31-cvs
5467 updated branch version
5469 final release candidate (see below)
5472 @item 5.2.0.90-2002-04-07-cvs
5473 updated CVS branch version
5475 second official release
5482 Minor minor minor draft release candidates such as 5.2.0.91 have been
5483 omitted from the example. Such release candidates are, typically, never
5486 For 5.1.93 the bziped tar ball @file{gdb-5.1.93.tar.bz2} is just the
5487 official @file{gdb-5.2.tar} renamed and compressed.
5490 To avoid version conflicts, vendors are expected to modify the file
5491 @file{gdb/version.in} to include a vendor unique alphabetic identifier
5492 (an official @value{GDBN} release never uses alphabetic characters in
5493 its version identifer).
5495 Since @value{GDBN} does not make minor minor minor releases (e.g.,
5496 5.1.0.1) the conflict between that and a minor minor draft release
5497 identifier (e.g., 5.1.0.90) is avoided.
5500 @subsection Branches
5502 @value{GDBN} draws a release series (5.2, 5.2.1, @dots{}) from a single
5503 release branch (gdb_5_2-branch). Since minor minor minor releases
5504 (5.1.0.1) are not made, the need to branch the release branch is avoided
5505 (it also turns out that the effort required for such a a branch and
5506 release is significantly greater than the effort needed to create a new
5507 release from the head of the release branch).
5509 Releases 5.0 and 5.1 used branch and release tags of the form:
5512 gdb_N_M-YYYY-MM-DD-branchpoint
5513 gdb_N_M-YYYY-MM-DD-branch
5514 gdb_M_N-YYYY-MM-DD-release
5517 Release 5.2 is trialing the branch and release tags:
5520 gdb_N_M-YYYY-MM-DD-branchpoint
5522 gdb_M_N-YYYY-MM-DD-release
5525 @emph{Pragmatics: The branchpoint and release tags need to identify when
5526 a branch and release are made. The branch tag, denoting the head of the
5527 branch, does not have this criteria.}
5530 @section Branch Commit Policy
5532 The branch commit policy is pretty slack. @value{GDBN} releases 5.0,
5533 5.1 and 5.2 all used the below:
5537 The @file{gdb/MAINTAINERS} file still holds.
5539 Don't fix something on the branch unless/until it is also fixed in the
5540 trunk. If this isn't possible, mentioning it in the @file{gdb/PROBLEMS}
5541 file is better than committing a hack.
5543 When considering a patch for the branch, suggested criteria include:
5544 Does it fix a build? Does it fix the sequence @kbd{break main; run}
5545 when debugging a static binary?
5547 The further a change is from the core of @value{GDBN}, the less likely
5548 the change will worry anyone (e.g., target specific code).
5550 Only post a proposal to change the core of @value{GDBN} after you've
5551 sent individual bribes to all the people listed in the
5552 @file{MAINTAINERS} file @t{;-)}
5555 @emph{Pragmatics: Provided updates are restricted to non-core
5556 functionality there is little chance that a broken change will be fatal.
5557 This means that changes such as adding a new architectures or (within
5558 reason) support for a new host are considered acceptable.}
5561 @section Obsoleting code
5563 Before anything else, poke the other developers (and around the source
5564 code) to see if there is anything that can be removed from @value{GDBN}
5565 (an old target, an unused file).
5567 Obsolete code is identified by adding an @code{OBSOLETE} prefix to every
5568 line. Doing this means that it is easy to identify something that has
5569 been obsoleted when greping through the sources.
5571 The process is done in stages --- this is mainly to ensure that the
5572 wider @value{GDBN} community has a reasonable opportunity to respond.
5573 Remember, everything on the Internet takes a week.
5577 Post the proposal on @email{gdb@@sources.redhat.com, the GDB mailing
5578 list} Creating a bug report to track the task's state, is also highly
5583 Post the proposal on @email{gdb-announce@@sources.redhat.com, the GDB
5584 Announcement mailing list}.
5588 Go through and edit all relevant files and lines so that they are
5589 prefixed with the word @code{OBSOLETE}.
5591 Wait until the next GDB version, containing this obsolete code, has been
5594 Remove the obsolete code.
5598 @emph{Maintainer note: While removing old code is regrettable it is
5599 hopefully better for @value{GDBN}'s long term development. Firstly it
5600 helps the developers by removing code that is either no longer relevant
5601 or simply wrong. Secondly since it removes any history associated with
5602 the file (effectively clearing the slate) the developer has a much freer
5603 hand when it comes to fixing broken files.}
5607 @section Before the Branch
5609 The most important objective at this stage is to find and fix simple
5610 changes that become a pain to track once the branch is created. For
5611 instance, configuration problems that stop @value{GDBN} from even
5612 building. If you can't get the problem fixed, document it in the
5613 @file{gdb/PROBLEMS} file.
5615 @subheading Prompt for @file{gdb/NEWS}
5617 People always forget. Send a post reminding them but also if you know
5618 something interesting happened add it yourself. The @code{schedule}
5619 script will mention this in its e-mail.
5621 @subheading Review @file{gdb/README}
5623 Grab one of the nightly snapshots and then walk through the
5624 @file{gdb/README} looking for anything that can be improved. The
5625 @code{schedule} script will mention this in its e-mail.
5627 @subheading Refresh any imported files.
5629 A number of files are taken from external repositories. They include:
5633 @file{texinfo/texinfo.tex}
5635 @file{config.guess} et.@: al.@: (see the top-level @file{MAINTAINERS}
5638 @file{etc/standards.texi}, @file{etc/make-stds.texi}
5641 @subheading Check the ARI
5643 @uref{http://sources.redhat.com/gdb/ari,,A.R.I.} is an @code{awk} script
5644 (Awk Regression Index ;-) that checks for a number of errors and coding
5645 conventions. The checks include things like using @code{malloc} instead
5646 of @code{xmalloc} and file naming problems. There shouldn't be any
5649 @subsection Review the bug data base
5651 Close anything obviously fixed.
5653 @subsection Check all cross targets build
5655 The targets are listed in @file{gdb/MAINTAINERS}.
5658 @section Cut the Branch
5660 @subheading Create the branch
5665 $ V=`echo $v | sed 's/\./_/g'`
5666 $ D=`date -u +%Y-%m-%d`
5669 $ echo cvs -f -d :ext:sources.redhat.com:/cvs/src rtag \
5670 -D $D-gmt gdb_$V-$D-branchpoint insight+dejagnu
5671 cvs -f -d :ext:sources.redhat.com:/cvs/src rtag
5672 -D 2002-03-03-gmt gdb_5_2-2002-03-03-branchpoint insight+dejagnu
5675 $ echo cvs -f -d :ext:sources.redhat.com:/cvs/src rtag \
5676 -b -r gdb_$V-$D-branchpoint gdb_$V-branch insight+dejagnu
5677 cvs -f -d :ext:sources.redhat.com:/cvs/src rtag \
5678 -b -r gdb_5_2-2002-03-03-branchpoint gdb_5_2-branch insight+dejagnu
5686 by using @kbd{-D YYYY-MM-DD-gmt} the branch is forced to an exact
5689 the trunk is first taged so that the branch point can easily be found
5691 Insight (which includes GDB) and dejagnu are all tagged at the same time
5693 @file{version.in} gets bumped to avoid version number conflicts
5695 the reading of @file{.cvsrc} is disabled using @file{-f}
5698 @subheading Update @file{version.in}
5703 $ V=`echo $v | sed 's/\./_/g'`
5707 $ echo cvs -f -d :ext:sources.redhat.com:/cvs/src co \
5708 -r gdb_$V-branch src/gdb/version.in
5709 cvs -f -d :ext:sources.redhat.com:/cvs/src co
5710 -r gdb_5_2-branch src/gdb/version.in
5712 U src/gdb/version.in
5714 $ echo $u.90-0000-00-00-cvs > version.in
5716 5.1.90-0000-00-00-cvs
5717 $ cvs -f commit version.in
5722 @file{0000-00-00} is used as a date to pump prime the version.in update
5725 @file{.90} and the previous branch version are used as fairly arbitrary
5726 initial branch version number
5730 @subheading Update the web and news pages
5734 @subheading Tweak cron to track the new branch
5736 The file @file{gdbadmin/cron/crontab} contains gdbadmin's cron table.
5737 This file needs to be updated so that:
5741 a daily timestamp is added to the file @file{version.in}
5743 the new branch is included in the snapshot process
5747 See the file @file{gdbadmin/cron/README} for how to install the updated
5750 The file @file{gdbadmin/ss/README} should also be reviewed to reflect
5751 any changes. That file is copied to both the branch/ and current/
5752 snapshot directories.
5755 @subheading Update the NEWS and README files
5757 The @file{NEWS} file needs to be updated so that on the branch it refers
5758 to @emph{changes in the current release} while on the trunk it also
5759 refers to @emph{changes since the current release}.
5761 The @file{README} file needs to be updated so that it refers to the
5764 @subheading Post the branch info
5766 Send an announcement to the mailing lists:
5770 @email{gdb-announce@@sources.redhat.com, GDB Announcement mailing list}
5772 @email{gdb@@sources.redhat.com, GDB Discsussion mailing list} and
5773 @email{gdb-testers@@sources.redhat.com, GDB Discsussion mailing list}
5776 @emph{Pragmatics: The branch creation is sent to the announce list to
5777 ensure that people people not subscribed to the higher volume discussion
5780 The announcement should include:
5786 how to check out the branch using CVS
5788 the date/number of weeks until the release
5790 the branch commit policy
5794 @section Stabilize the branch
5796 Something goes here.
5798 @section Create a Release
5800 The process of creating and then making available a release is broken
5801 down into a number of stages. The first part addresses the technical
5802 process of creating a releasable tar ball. The later stages address the
5803 process of releasing that tar ball.
5805 When making a release candidate just the first section is needed.
5807 @subsection Create a release candidate
5809 The objective at this stage is to create a set of tar balls that can be
5810 made available as a formal release (or as a less formal release
5813 @subsubheading Freeze the branch
5815 Send out an e-mail notifying everyone that the branch is frozen to
5816 @email{gdb-patches@@sources.redhat.com}.
5818 @subsubheading Establish a few defaults.
5823 $ t=/sourceware/snapshot-tmp/gdbadmin-tmp
5825 /sourceware/snapshot-tmp/gdbadmin-tmp/gdb_5_2-branch/5.2
5829 /sourceware/snapshot-tmp/gdbadmin-tmp/gdb_5_2-branch/5.2
5831 /home/gdbadmin/bin/autoconf
5840 Check the @code{autoconf} version carefully. You want to be using the
5841 version taken from the @file{binutils} snapshot directory, which can be
5842 found at @uref{ftp://sources.redhat.com/pub/binutils/}. It is very
5843 unlikely that a system installed version of @code{autoconf} (e.g.,
5844 @file{/usr/bin/autoconf}) is correct.
5847 @subsubheading Check out the relevant modules:
5850 $ for m in gdb insight dejagnu
5852 ( mkdir -p $m && cd $m && cvs -q -f -d /cvs/src co -P -r $b $m )
5862 The reading of @file{.cvsrc} is disabled (@file{-f}) so that there isn't
5863 any confusion between what is written here and what your local
5864 @code{cvs} really does.
5867 @subsubheading Update relevant files.
5873 Major releases get their comments added as part of the mainline. Minor
5874 releases should probably mention any significant bugs that were fixed.
5876 Don't forget to include the @file{ChangeLog} entry.
5879 $ emacs gdb/src/gdb/NEWS
5884 $ cp gdb/src/gdb/NEWS insight/src/gdb/NEWS
5885 $ cp gdb/src/gdb/ChangeLog insight/src/gdb/ChangeLog
5890 You'll need to update:
5902 $ emacs gdb/src/gdb/README
5907 $ cp gdb/src/gdb/README insight/src/gdb/README
5908 $ cp gdb/src/gdb/ChangeLog insight/src/gdb/ChangeLog
5911 @emph{Maintainer note: Hopefully the @file{README} file was reviewed
5912 before the initial branch was cut so just a simple substitute is needed
5915 @emph{Maintainer note: Other projects generate @file{README} and
5916 @file{INSTALL} from the core documentation. This might be worth
5919 @item gdb/version.in
5922 $ echo $v > gdb/src/gdb/version.in
5923 $ cat gdb/src/gdb/version.in
5925 $ emacs gdb/src/gdb/version.in
5928 ... Bump to version ...
5930 $ cp gdb/src/gdb/version.in insight/src/gdb/version.in
5931 $ cp gdb/src/gdb/ChangeLog insight/src/gdb/ChangeLog
5934 @item dejagnu/src/dejagnu/configure.in
5936 Dejagnu is more complicated. The version number is a parameter to
5937 @code{AM_INIT_AUTOMAKE}. Tweak it to read something like gdb-5.1.91.
5939 Don't forget to re-generate @file{configure}.
5941 Don't forget to include a @file{ChangeLog} entry.
5944 $ emacs dejagnu/src/dejagnu/configure.in
5949 $ ( cd dejagnu/src/dejagnu && autoconf )
5954 @subsubheading Do the dirty work
5956 This is identical to the process used to create the daily snapshot.
5959 $ for m in gdb insight
5961 ( cd $m/src && gmake -f src-release $m.tar )
5963 $ ( m=dejagnu; cd $m/src && gmake -f src-release $m.tar.bz2 )
5966 If the top level source directory does not have @file{src-release}
5967 (@value{GDBN} version 5.3.1 or earlier), try these commands instead:
5970 $ for m in gdb insight
5972 ( cd $m/src && gmake -f Makefile.in $m.tar )
5974 $ ( m=dejagnu; cd $m/src && gmake -f Makefile.in $m.tar.bz2 )
5977 @subsubheading Check the source files
5979 You're looking for files that have mysteriously disappeared.
5980 @kbd{distclean} has the habit of deleting files it shouldn't. Watch out
5981 for the @file{version.in} update @kbd{cronjob}.
5984 $ ( cd gdb/src && cvs -f -q -n update )
5988 @dots{} lots of generated files @dots{}
5993 @dots{} lots of generated files @dots{}
5998 @emph{Don't worry about the @file{gdb.info-??} or
5999 @file{gdb/p-exp.tab.c}. They were generated (and yes @file{gdb.info-1}
6000 was also generated only something strange with CVS means that they
6001 didn't get supressed). Fixing it would be nice though.}
6003 @subsubheading Create compressed versions of the release
6009 dejagnu/ dejagnu-gdb-5.2.tar.bz2 gdb/ gdb-5.2.tar insight/ insight-5.2.tar
6010 $ for m in gdb insight
6012 bzip2 -v -9 -c $m-$v.tar > $m-$v.tar.bz2
6013 gzip -v -9 -c $m-$v.tar > $m-$v.tar.gz
6023 A pipe such as @kbd{bunzip2 < xxx.bz2 | gzip -9 > xxx.gz} is not since,
6024 in that mode, @code{gzip} does not know the name of the file and, hence,
6025 can not include it in the compressed file. This is also why the release
6026 process runs @code{tar} and @code{bzip2} as separate passes.
6029 @subsection Sanity check the tar ball
6031 Pick a popular machine (Solaris/PPC?) and try the build on that.
6034 $ bunzip2 < gdb-5.2.tar.bz2 | tar xpf -
6039 $ ./gdb/gdb ./gdb/gdb
6043 Breakpoint 1 at 0x80732bc: file main.c, line 734.
6045 Starting program: /tmp/gdb-5.2/gdb/gdb
6047 Breakpoint 1, main (argc=1, argv=0xbffff8b4) at main.c:734
6048 734 catch_errors (captured_main, &args, "", RETURN_MASK_ALL);
6050 $1 = @{argc = 136426532, argv = 0x821b7f0@}
6054 @subsection Make a release candidate available
6056 If this is a release candidate then the only remaining steps are:
6060 Commit @file{version.in} and @file{ChangeLog}
6062 Tweak @file{version.in} (and @file{ChangeLog} to read
6063 @var{L}.@var{M}.@var{N}-0000-00-00-cvs so that the version update
6064 process can restart.
6066 Make the release candidate available in
6067 @uref{ftp://sources.redhat.com/pub/gdb/snapshots/branch}
6069 Notify the relevant mailing lists ( @email{gdb@@sources.redhat.com} and
6070 @email{gdb-testers@@sources.redhat.com} that the candidate is available.
6073 @subsection Make a formal release available
6075 (And you thought all that was required was to post an e-mail.)
6077 @subsubheading Install on sware
6079 Copy the new files to both the release and the old release directory:
6082 $ cp *.bz2 *.gz ~ftp/pub/gdb/old-releases/
6083 $ cp *.bz2 *.gz ~ftp/pub/gdb/releases
6087 Clean up the releases directory so that only the most recent releases
6088 are available (e.g. keep 5.2 and 5.2.1 but remove 5.1):
6091 $ cd ~ftp/pub/gdb/releases
6096 Update the file @file{README} and @file{.message} in the releases
6103 $ ln README .message
6106 @subsubheading Update the web pages.
6110 @item htdocs/download/ANNOUNCEMENT
6111 This file, which is posted as the official announcement, includes:
6114 General announcement
6116 News. If making an @var{M}.@var{N}.1 release, retain the news from
6117 earlier @var{M}.@var{N} release.
6122 @item htdocs/index.html
6123 @itemx htdocs/news/index.html
6124 @itemx htdocs/download/index.html
6125 These files include:
6128 announcement of the most recent release
6130 news entry (remember to update both the top level and the news directory).
6132 These pages also need to be regenerate using @code{index.sh}.
6134 @item download/onlinedocs/
6135 You need to find the magic command that is used to generate the online
6136 docs from the @file{.tar.bz2}. The best way is to look in the output
6137 from one of the nightly @code{cron} jobs and then just edit accordingly.
6141 $ ~/ss/update-web-docs \
6142 ~ftp/pub/gdb/releases/gdb-5.2.tar.bz2 \
6144 /www/sourceware/htdocs/gdb/download/onlinedocs \
6149 Just like the online documentation. Something like:
6152 $ /bin/sh ~/ss/update-web-ari \
6153 ~ftp/pub/gdb/releases/gdb-5.2.tar.bz2 \
6155 /www/sourceware/htdocs/gdb/download/ari \
6161 @subsubheading Shadow the pages onto gnu
6163 Something goes here.
6166 @subsubheading Install the @value{GDBN} tar ball on GNU
6168 At the time of writing, the GNU machine was @kbd{gnudist.gnu.org} in
6169 @file{~ftp/gnu/gdb}.
6171 @subsubheading Make the @file{ANNOUNCEMENT}
6173 Post the @file{ANNOUNCEMENT} file you created above to:
6177 @email{gdb-announce@@sources.redhat.com, GDB Announcement mailing list}
6179 @email{info-gnu@@gnu.org, General GNU Announcement list} (but delay it a
6180 day or so to let things get out)
6182 @email{bug-gdb@@gnu.org, GDB Bug Report mailing list}
6187 The release is out but you're still not finished.
6189 @subsubheading Commit outstanding changes
6191 In particular you'll need to commit any changes to:
6195 @file{gdb/ChangeLog}
6197 @file{gdb/version.in}
6204 @subsubheading Tag the release
6209 $ d=`date -u +%Y-%m-%d`
6212 $ ( cd insight/src/gdb && cvs -f -q update )
6213 $ ( cd insight/src && cvs -f -q tag gdb_5_2-$d-release )
6216 Insight is used since that contains more of the release than
6217 @value{GDBN} (@code{dejagnu} doesn't get tagged but I think we can live
6220 @subsubheading Mention the release on the trunk
6222 Just put something in the @file{ChangeLog} so that the trunk also
6223 indicates when the release was made.
6225 @subsubheading Restart @file{gdb/version.in}
6227 If @file{gdb/version.in} does not contain an ISO date such as
6228 @kbd{2002-01-24} then the daily @code{cronjob} won't update it. Having
6229 committed all the release changes it can be set to
6230 @file{5.2.0_0000-00-00-cvs} which will restart things (yes the @kbd{_}
6231 is important - it affects the snapshot process).
6233 Don't forget the @file{ChangeLog}.
6235 @subsubheading Merge into trunk
6237 The files committed to the branch may also need changes merged into the
6240 @subsubheading Revise the release schedule
6242 Post a revised release schedule to @email{gdb@@sources.redhat.com, GDB
6243 Discussion List} with an updated announcement. The schedule can be
6244 generated by running:
6247 $ ~/ss/schedule `date +%s` schedule
6251 The first parameter is approximate date/time in seconds (from the epoch)
6252 of the most recent release.
6254 Also update the schedule @code{cronjob}.
6256 @section Post release
6258 Remove any @code{OBSOLETE} code.
6265 The testsuite is an important component of the @value{GDBN} package.
6266 While it is always worthwhile to encourage user testing, in practice
6267 this is rarely sufficient; users typically use only a small subset of
6268 the available commands, and it has proven all too common for a change
6269 to cause a significant regression that went unnoticed for some time.
6271 The @value{GDBN} testsuite uses the DejaGNU testing framework.
6272 DejaGNU is built using @code{Tcl} and @code{expect}. The tests
6273 themselves are calls to various @code{Tcl} procs; the framework runs all the
6274 procs and summarizes the passes and fails.
6276 @section Using the Testsuite
6278 @cindex running the test suite
6279 To run the testsuite, simply go to the @value{GDBN} object directory (or to the
6280 testsuite's objdir) and type @code{make check}. This just sets up some
6281 environment variables and invokes DejaGNU's @code{runtest} script. While
6282 the testsuite is running, you'll get mentions of which test file is in use,
6283 and a mention of any unexpected passes or fails. When the testsuite is
6284 finished, you'll get a summary that looks like this:
6289 # of expected passes 6016
6290 # of unexpected failures 58
6291 # of unexpected successes 5
6292 # of expected failures 183
6293 # of unresolved testcases 3
6294 # of untested testcases 5
6297 The ideal test run consists of expected passes only; however, reality
6298 conspires to keep us from this ideal. Unexpected failures indicate
6299 real problems, whether in @value{GDBN} or in the testsuite. Expected
6300 failures are still failures, but ones which have been decided are too
6301 hard to deal with at the time; for instance, a test case might work
6302 everywhere except on AIX, and there is no prospect of the AIX case
6303 being fixed in the near future. Expected failures should not be added
6304 lightly, since you may be masking serious bugs in @value{GDBN}.
6305 Unexpected successes are expected fails that are passing for some
6306 reason, while unresolved and untested cases often indicate some minor
6307 catastrophe, such as the compiler being unable to deal with a test
6310 When making any significant change to @value{GDBN}, you should run the
6311 testsuite before and after the change, to confirm that there are no
6312 regressions. Note that truly complete testing would require that you
6313 run the testsuite with all supported configurations and a variety of
6314 compilers; however this is more than really necessary. In many cases
6315 testing with a single configuration is sufficient. Other useful
6316 options are to test one big-endian (Sparc) and one little-endian (x86)
6317 host, a cross config with a builtin simulator (powerpc-eabi,
6318 mips-elf), or a 64-bit host (Alpha).
6320 If you add new functionality to @value{GDBN}, please consider adding
6321 tests for it as well; this way future @value{GDBN} hackers can detect
6322 and fix their changes that break the functionality you added.
6323 Similarly, if you fix a bug that was not previously reported as a test
6324 failure, please add a test case for it. Some cases are extremely
6325 difficult to test, such as code that handles host OS failures or bugs
6326 in particular versions of compilers, and it's OK not to try to write
6327 tests for all of those.
6329 @section Testsuite Organization
6331 @cindex test suite organization
6332 The testsuite is entirely contained in @file{gdb/testsuite}. While the
6333 testsuite includes some makefiles and configury, these are very minimal,
6334 and used for little besides cleaning up, since the tests themselves
6335 handle the compilation of the programs that @value{GDBN} will run. The file
6336 @file{testsuite/lib/gdb.exp} contains common utility procs useful for
6337 all @value{GDBN} tests, while the directory @file{testsuite/config} contains
6338 configuration-specific files, typically used for special-purpose
6339 definitions of procs like @code{gdb_load} and @code{gdb_start}.
6341 The tests themselves are to be found in @file{testsuite/gdb.*} and
6342 subdirectories of those. The names of the test files must always end
6343 with @file{.exp}. DejaGNU collects the test files by wildcarding
6344 in the test directories, so both subdirectories and individual files
6345 get chosen and run in alphabetical order.
6347 The following table lists the main types of subdirectories and what they
6348 are for. Since DejaGNU finds test files no matter where they are
6349 located, and since each test file sets up its own compilation and
6350 execution environment, this organization is simply for convenience and
6355 This is the base testsuite. The tests in it should apply to all
6356 configurations of @value{GDBN} (but generic native-only tests may live here).
6357 The test programs should be in the subset of C that is valid K&R,
6358 ANSI/ISO, and C++ (@code{#ifdef}s are allowed if necessary, for instance
6361 @item gdb.@var{lang}
6362 Language-specific tests for any language @var{lang} besides C. Examples are
6363 @file{gdb.c++} and @file{gdb.java}.
6365 @item gdb.@var{platform}
6366 Non-portable tests. The tests are specific to a specific configuration
6367 (host or target), such as HP-UX or eCos. Example is @file{gdb.hp}, for
6370 @item gdb.@var{compiler}
6371 Tests specific to a particular compiler. As of this writing (June
6372 1999), there aren't currently any groups of tests in this category that
6373 couldn't just as sensibly be made platform-specific, but one could
6374 imagine a @file{gdb.gcc}, for tests of @value{GDBN}'s handling of GCC
6377 @item gdb.@var{subsystem}
6378 Tests that exercise a specific @value{GDBN} subsystem in more depth. For
6379 instance, @file{gdb.disasm} exercises various disassemblers, while
6380 @file{gdb.stabs} tests pathways through the stabs symbol reader.
6383 @section Writing Tests
6384 @cindex writing tests
6386 In many areas, the @value{GDBN} tests are already quite comprehensive; you
6387 should be able to copy existing tests to handle new cases.
6389 You should try to use @code{gdb_test} whenever possible, since it
6390 includes cases to handle all the unexpected errors that might happen.
6391 However, it doesn't cost anything to add new test procedures; for
6392 instance, @file{gdb.base/exprs.exp} defines a @code{test_expr} that
6393 calls @code{gdb_test} multiple times.
6395 Only use @code{send_gdb} and @code{gdb_expect} when absolutely
6396 necessary, such as when @value{GDBN} has several valid responses to a command.
6398 The source language programs do @emph{not} need to be in a consistent
6399 style. Since @value{GDBN} is used to debug programs written in many different
6400 styles, it's worth having a mix of styles in the testsuite; for
6401 instance, some @value{GDBN} bugs involving the display of source lines would
6402 never manifest themselves if the programs used GNU coding style
6409 Check the @file{README} file, it often has useful information that does not
6410 appear anywhere else in the directory.
6413 * Getting Started:: Getting started working on @value{GDBN}
6414 * Debugging GDB:: Debugging @value{GDBN} with itself
6417 @node Getting Started,,, Hints
6419 @section Getting Started
6421 @value{GDBN} is a large and complicated program, and if you first starting to
6422 work on it, it can be hard to know where to start. Fortunately, if you
6423 know how to go about it, there are ways to figure out what is going on.
6425 This manual, the @value{GDBN} Internals manual, has information which applies
6426 generally to many parts of @value{GDBN}.
6428 Information about particular functions or data structures are located in
6429 comments with those functions or data structures. If you run across a
6430 function or a global variable which does not have a comment correctly
6431 explaining what is does, this can be thought of as a bug in @value{GDBN}; feel
6432 free to submit a bug report, with a suggested comment if you can figure
6433 out what the comment should say. If you find a comment which is
6434 actually wrong, be especially sure to report that.
6436 Comments explaining the function of macros defined in host, target, or
6437 native dependent files can be in several places. Sometimes they are
6438 repeated every place the macro is defined. Sometimes they are where the
6439 macro is used. Sometimes there is a header file which supplies a
6440 default definition of the macro, and the comment is there. This manual
6441 also documents all the available macros.
6442 @c (@pxref{Host Conditionals}, @pxref{Target
6443 @c Conditionals}, @pxref{Native Conditionals}, and @pxref{Obsolete
6446 Start with the header files. Once you have some idea of how
6447 @value{GDBN}'s internal symbol tables are stored (see @file{symtab.h},
6448 @file{gdbtypes.h}), you will find it much easier to understand the
6449 code which uses and creates those symbol tables.
6451 You may wish to process the information you are getting somehow, to
6452 enhance your understanding of it. Summarize it, translate it to another
6453 language, add some (perhaps trivial or non-useful) feature to @value{GDBN}, use
6454 the code to predict what a test case would do and write the test case
6455 and verify your prediction, etc. If you are reading code and your eyes
6456 are starting to glaze over, this is a sign you need to use a more active
6459 Once you have a part of @value{GDBN} to start with, you can find more
6460 specifically the part you are looking for by stepping through each
6461 function with the @code{next} command. Do not use @code{step} or you
6462 will quickly get distracted; when the function you are stepping through
6463 calls another function try only to get a big-picture understanding
6464 (perhaps using the comment at the beginning of the function being
6465 called) of what it does. This way you can identify which of the
6466 functions being called by the function you are stepping through is the
6467 one which you are interested in. You may need to examine the data
6468 structures generated at each stage, with reference to the comments in
6469 the header files explaining what the data structures are supposed to
6472 Of course, this same technique can be used if you are just reading the
6473 code, rather than actually stepping through it. The same general
6474 principle applies---when the code you are looking at calls something
6475 else, just try to understand generally what the code being called does,
6476 rather than worrying about all its details.
6478 @cindex command implementation
6479 A good place to start when tracking down some particular area is with
6480 a command which invokes that feature. Suppose you want to know how
6481 single-stepping works. As a @value{GDBN} user, you know that the
6482 @code{step} command invokes single-stepping. The command is invoked
6483 via command tables (see @file{command.h}); by convention the function
6484 which actually performs the command is formed by taking the name of
6485 the command and adding @samp{_command}, or in the case of an
6486 @code{info} subcommand, @samp{_info}. For example, the @code{step}
6487 command invokes the @code{step_command} function and the @code{info
6488 display} command invokes @code{display_info}. When this convention is
6489 not followed, you might have to use @code{grep} or @kbd{M-x
6490 tags-search} in emacs, or run @value{GDBN} on itself and set a
6491 breakpoint in @code{execute_command}.
6493 @cindex @code{bug-gdb} mailing list
6494 If all of the above fail, it may be appropriate to ask for information
6495 on @code{bug-gdb}. But @emph{never} post a generic question like ``I was
6496 wondering if anyone could give me some tips about understanding
6497 @value{GDBN}''---if we had some magic secret we would put it in this manual.
6498 Suggestions for improving the manual are always welcome, of course.
6500 @node Debugging GDB,,,Hints
6502 @section Debugging @value{GDBN} with itself
6503 @cindex debugging @value{GDBN}
6505 If @value{GDBN} is limping on your machine, this is the preferred way to get it
6506 fully functional. Be warned that in some ancient Unix systems, like
6507 Ultrix 4.2, a program can't be running in one process while it is being
6508 debugged in another. Rather than typing the command @kbd{@w{./gdb
6509 ./gdb}}, which works on Suns and such, you can copy @file{gdb} to
6510 @file{gdb2} and then type @kbd{@w{./gdb ./gdb2}}.
6512 When you run @value{GDBN} in the @value{GDBN} source directory, it will read a
6513 @file{.gdbinit} file that sets up some simple things to make debugging
6514 gdb easier. The @code{info} command, when executed without a subcommand
6515 in a @value{GDBN} being debugged by gdb, will pop you back up to the top level
6516 gdb. See @file{.gdbinit} for details.
6518 If you use emacs, you will probably want to do a @code{make TAGS} after
6519 you configure your distribution; this will put the machine dependent
6520 routines for your local machine where they will be accessed first by
6523 Also, make sure that you've either compiled @value{GDBN} with your local cc, or
6524 have run @code{fixincludes} if you are compiling with gcc.
6526 @section Submitting Patches
6528 @cindex submitting patches
6529 Thanks for thinking of offering your changes back to the community of
6530 @value{GDBN} users. In general we like to get well designed enhancements.
6531 Thanks also for checking in advance about the best way to transfer the
6534 The @value{GDBN} maintainers will only install ``cleanly designed'' patches.
6535 This manual summarizes what we believe to be clean design for @value{GDBN}.
6537 If the maintainers don't have time to put the patch in when it arrives,
6538 or if there is any question about a patch, it goes into a large queue
6539 with everyone else's patches and bug reports.
6541 @cindex legal papers for code contributions
6542 The legal issue is that to incorporate substantial changes requires a
6543 copyright assignment from you and/or your employer, granting ownership
6544 of the changes to the Free Software Foundation. You can get the
6545 standard documents for doing this by sending mail to @code{gnu@@gnu.org}
6546 and asking for it. We recommend that people write in "All programs
6547 owned by the Free Software Foundation" as "NAME OF PROGRAM", so that
6548 changes in many programs (not just @value{GDBN}, but GAS, Emacs, GCC,
6550 contributed with only one piece of legalese pushed through the
6551 bureaucracy and filed with the FSF. We can't start merging changes until
6552 this paperwork is received by the FSF (their rules, which we follow
6553 since we maintain it for them).
6555 Technically, the easiest way to receive changes is to receive each
6556 feature as a small context diff or unidiff, suitable for @code{patch}.
6557 Each message sent to me should include the changes to C code and
6558 header files for a single feature, plus @file{ChangeLog} entries for
6559 each directory where files were modified, and diffs for any changes
6560 needed to the manuals (@file{gdb/doc/gdb.texinfo} or
6561 @file{gdb/doc/gdbint.texinfo}). If there are a lot of changes for a
6562 single feature, they can be split down into multiple messages.
6564 In this way, if we read and like the feature, we can add it to the
6565 sources with a single patch command, do some testing, and check it in.
6566 If you leave out the @file{ChangeLog}, we have to write one. If you leave
6567 out the doc, we have to puzzle out what needs documenting. Etc., etc.
6569 The reason to send each change in a separate message is that we will not
6570 install some of the changes. They'll be returned to you with questions
6571 or comments. If we're doing our job correctly, the message back to you
6572 will say what you have to fix in order to make the change acceptable.
6573 The reason to have separate messages for separate features is so that
6574 the acceptable changes can be installed while one or more changes are
6575 being reworked. If multiple features are sent in a single message, we
6576 tend to not put in the effort to sort out the acceptable changes from
6577 the unacceptable, so none of the features get installed until all are
6580 If this sounds painful or authoritarian, well, it is. But we get a lot
6581 of bug reports and a lot of patches, and many of them don't get
6582 installed because we don't have the time to finish the job that the bug
6583 reporter or the contributor could have done. Patches that arrive
6584 complete, working, and well designed, tend to get installed on the day
6585 they arrive. The others go into a queue and get installed as time
6586 permits, which, since the maintainers have many demands to meet, may not
6587 be for quite some time.
6589 Please send patches directly to
6590 @email{gdb-patches@@sources.redhat.com, the @value{GDBN} maintainers}.
6592 @section Obsolete Conditionals
6593 @cindex obsolete code
6595 Fragments of old code in @value{GDBN} sometimes reference or set the following
6596 configuration macros. They should not be used by new code, and old uses
6597 should be removed as those parts of the debugger are otherwise touched.
6600 @item STACK_END_ADDR
6601 This macro used to define where the end of the stack appeared, for use
6602 in interpreting core file formats that don't record this address in the
6603 core file itself. This information is now configured in BFD, and @value{GDBN}
6604 gets the info portably from there. The values in @value{GDBN}'s configuration
6605 files should be moved into BFD configuration files (if needed there),
6606 and deleted from all of @value{GDBN}'s config files.
6608 Any @file{@var{foo}-xdep.c} file that references STACK_END_ADDR
6609 is so old that it has never been converted to use BFD. Now that's old!