2 @c This is part of the GNU Emacs Lisp Reference Manual.
3 @c Copyright (C) 1990, 1991, 1992, 1993, 1998, 1999
4 @c Free Software Foundation, Inc.
5 @c See the file elisp.texi for copying conditions.
6 @setfilename ../info/internals
7 @node GNU Emacs Internals, Standard Errors, Tips, Top
8 @comment node-name, next, previous, up
9 @appendix GNU Emacs Internals
11 This chapter describes how the runnable Emacs executable is dumped with
12 the preloaded Lisp libraries in it, how storage is allocated, and some
13 internal aspects of GNU Emacs that may be of interest to C programmers.
16 * Building Emacs:: How to the dumped Emacs is made.
17 * Pure Storage:: A kludge to make preloaded Lisp functions sharable.
18 * Garbage Collection:: Reclaiming space for Lisp objects no longer used.
19 * Memory Usage:: Info about total size of Lisp objects made so far.
20 * Writing Emacs Primitives:: Writing C code for Emacs.
21 * Object Internals:: Data formats of buffers, windows, processes.
25 @appendixsec Building Emacs
26 @cindex building Emacs
29 This section explains the steps involved in building the Emacs
30 executable. You don't have to know this material to build and install
31 Emacs, since the makefiles do all these things automatically. This
32 information is pertinent to Emacs maintenance.
34 Compilation of the C source files in the @file{src} directory
35 produces an executable file called @file{temacs}, also called a
36 @dfn{bare impure Emacs}. It contains the Emacs Lisp interpreter and I/O
37 routines, but not the editing commands.
39 @cindex @file{loadup.el}
40 The command @w{@samp{temacs -l loadup}} uses @file{temacs} to create
41 the real runnable Emacs executable. These arguments direct
42 @file{temacs} to evaluate the Lisp files specified in the file
43 @file{loadup.el}. These files set up the normal Emacs editing
44 environment, resulting in an Emacs that is still impure but no longer
47 It takes a substantial time to load the standard Lisp files. Luckily,
48 you don't have to do this each time you run Emacs; @file{temacs} can
49 dump out an executable program called @file{emacs} that has these files
50 preloaded. @file{emacs} starts more quickly because it does not need to
51 load the files. This is the Emacs executable that is normally
54 To create @file{emacs}, use the command @samp{temacs -batch -l loadup
55 dump}. The purpose of @samp{-batch} here is to prevent @file{temacs}
56 from trying to initialize any of its data on the terminal; this ensures
57 that the tables of terminal information are empty in the dumped Emacs.
58 The argument @samp{dump} tells @file{loadup.el} to dump a new executable
61 Some operating systems don't support dumping. On those systems, you
62 must start Emacs with the @samp{temacs -l loadup} command each time you
63 use it. This takes a substantial time, but since you need to start
64 Emacs once a day at most---or once a week if you never log out---the
65 extra time is not too severe a problem.
67 @cindex @file{site-load.el}
69 You can specify additional files to preload by writing a library named
70 @file{site-load.el} that loads them. You may need to add a definition
73 #define SITELOAD_PURESIZE_EXTRA @var{n}
77 to make @var{n} added bytes of pure space to hold the additional files.
78 (Try adding increments of 20000 until it is big enough.) However, the
79 advantage of preloading additional files decreases as machines get
80 faster. On modern machines, it is usually not advisable.
82 After @file{loadup.el} reads @file{site-load.el}, it finds the
83 documentation strings for primitive and preloaded functions (and
84 variables) in the file @file{etc/DOC} where they are stored, by calling
85 @code{Snarf-documentation} (@pxref{Accessing Documentation}).
87 @cindex @file{site-init.el}
88 You can specify other Lisp expressions to execute just before dumping
89 by putting them in a library named @file{site-init.el}. This file is
90 executed after the documentation strings are found.
92 If you want to preload function or variable definitions, there are
93 three ways you can do this and make their documentation strings
94 accessible when you subsequently run Emacs:
98 Arrange to scan these files when producing the @file{etc/DOC} file,
99 and load them with @file{site-load.el}.
102 Load the files with @file{site-init.el}, then copy the files into the
103 installation directory for Lisp files when you install Emacs.
106 Specify a non-@code{nil} value for
107 @code{byte-compile-dynamic-docstrings} as a local variable in each of these
108 files, and load them with either @file{site-load.el} or
109 @file{site-init.el}. (This method has the drawback that the
110 documentation strings take up space in Emacs all the time.)
113 It is not advisable to put anything in @file{site-load.el} or
114 @file{site-init.el} that would alter any of the features that users
115 expect in an ordinary unmodified Emacs. If you feel you must override
116 normal features for your site, do it with @file{default.el}, so that
117 users can override your changes if they wish. @xref{Startup Summary}.
119 @defun dump-emacs to-file from-file
121 This function dumps the current state of Emacs into an executable file
122 @var{to-file}. It takes symbols from @var{from-file} (this is normally
123 the executable file @file{temacs}).
125 If you want to use this function in an Emacs that was already dumped,
126 you must run Emacs with @samp{-batch}.
130 @appendixsec Pure Storage
133 Emacs Lisp uses two kinds of storage for user-created Lisp objects:
134 @dfn{normal storage} and @dfn{pure storage}. Normal storage is where
135 all the new data created during an Emacs session are kept; see the
136 following section for information on normal storage. Pure storage is
137 used for certain data in the preloaded standard Lisp files---data that
138 should never change during actual use of Emacs.
140 Pure storage is allocated only while @file{temacs} is loading the
141 standard preloaded Lisp libraries. In the file @file{emacs}, it is
142 marked as read-only (on operating systems that permit this), so that
143 the memory space can be shared by all the Emacs jobs running on the
144 machine at once. Pure storage is not expandable; a fixed amount is
145 allocated when Emacs is compiled, and if that is not sufficient for the
146 preloaded libraries, @file{temacs} crashes. If that happens, you must
147 increase the compilation parameter @code{PURESIZE} in the file
148 @file{src/puresize.h}. This normally won't happen unless you try to
149 preload additional libraries or add features to the standard ones.
151 @defun purecopy object
152 This function makes a copy in pure storage of @var{object}, and returns
153 it. It copies a string by simply making a new string with the same
154 characters in pure storage. It recursively copies the contents of
155 vectors and cons cells. It does not make copies of other objects such
156 as symbols, but just returns them unchanged. It signals an error if
157 asked to copy markers.
159 This function is a no-op except while Emacs is being built and dumped;
160 it is usually called only in the file @file{emacs/lisp/loaddefs.el}, but
161 a few packages call it just in case you decide to preload them.
164 @defvar pure-bytes-used
165 The value of this variable is the number of bytes of pure storage
166 allocated so far. Typically, in a dumped Emacs, this number is very
167 close to the total amount of pure storage available---if it were not,
168 we would preallocate less.
172 This variable determines whether @code{defun} should make a copy of the
173 function definition in pure storage. If it is non-@code{nil}, then the
174 function definition is copied into pure storage.
176 This flag is @code{t} while loading all of the basic functions for
177 building Emacs initially (allowing those functions to be sharable and
178 non-collectible). Dumping Emacs as an executable always writes
179 @code{nil} in this variable, regardless of the value it actually has
180 before and after dumping.
182 You should not change this flag in a running Emacs.
185 @node Garbage Collection
186 @appendixsec Garbage Collection
187 @cindex garbage collector
189 @cindex memory allocation
190 When a program creates a list or the user defines a new function (such
191 as by loading a library), that data is placed in normal storage. If
192 normal storage runs low, then Emacs asks the operating system to
193 allocate more memory in blocks of 1k bytes. Each block is used for one
194 type of Lisp object, so symbols, cons cells, markers, etc., are
195 segregated in distinct blocks in memory. (Vectors, long strings,
196 buffers and certain other editing types, which are fairly large, are
197 allocated in individual blocks, one per object, while small strings are
198 packed into blocks of 8k bytes.)
200 It is quite common to use some storage for a while, then release it by
201 (for example) killing a buffer or deleting the last pointer to an
202 object. Emacs provides a @dfn{garbage collector} to reclaim this
203 abandoned storage. (This name is traditional, but ``garbage recycler''
204 might be a more intuitive metaphor for this facility.)
206 The garbage collector operates by finding and marking all Lisp objects
207 that are still accessible to Lisp programs. To begin with, it assumes
208 all the symbols, their values and associated function definitions, and
209 any data presently on the stack, are accessible. Any objects that can
210 be reached indirectly through other accessible objects are also
213 When marking is finished, all objects still unmarked are garbage. No
214 matter what the Lisp program or the user does, it is impossible to refer
215 to them, since there is no longer a way to reach them. Their space
216 might as well be reused, since no one will miss them. The second
217 (``sweep'') phase of the garbage collector arranges to reuse them.
219 @c ??? Maybe add something describing weak hash tables here?
222 The sweep phase puts unused cons cells onto a @dfn{free list}
223 for future allocation; likewise for symbols and markers. It compacts
224 the accessible strings so they occupy fewer 8k blocks; then it frees the
225 other 8k blocks. Vectors, buffers, windows, and other large objects are
226 individually allocated and freed using @code{malloc} and @code{free}.
228 @cindex CL note---allocate more storage
230 @b{Common Lisp note:} Unlike other Lisps, GNU Emacs Lisp does not
231 call the garbage collector when the free list is empty. Instead, it
232 simply requests the operating system to allocate more storage, and
233 processing continues until @code{gc-cons-threshold} bytes have been
236 This means that you can make sure that the garbage collector will not
237 run during a certain portion of a Lisp program by calling the garbage
238 collector explicitly just before it (provided that portion of the
239 program does not use so much space as to force a second garbage
243 @deffn Command garbage-collect
244 This command runs a garbage collection, and returns information on
245 the amount of space in use. (Garbage collection can also occur
246 spontaneously if you use more than @code{gc-cons-threshold} bytes of
247 Lisp data since the previous garbage collection.)
249 @code{garbage-collect} returns a list containing the following
254 ((@var{used-conses} . @var{free-conses})
255 (@var{used-syms} . @var{free-syms})
257 (@var{used-miscs} . @var{free-miscs})
258 @var{used-string-chars}
259 @var{used-vector-slots}
260 (@var{used-floats} . @var{free-floats})
261 (@var{used-intervals} . @var{free-intervals})
262 (@var{used-strings} . @var{free-strings}))
270 @result{} ((106886 . 13184) (9769 . 0)
271 (7731 . 4651) 347543 121628
272 (31 . 94) (1273 . 168)
277 Here is a table explaining each element:
281 The number of cons cells in use.
284 The number of cons cells for which space has been obtained from the
285 operating system, but that are not currently being used.
288 The number of symbols in use.
291 The number of symbols for which space has been obtained from the
292 operating system, but that are not currently being used.
295 The number of miscellaneous objects in use. These include markers and
296 overlays, plus certain objects not visible to users.
299 The number of miscellaneous objects for which space has been obtained
300 from the operating system, but that are not currently being used.
302 @item used-string-chars
303 The total size of all strings, in characters.
305 @item used-vector-slots
306 The total number of elements of existing vectors.
310 The number of floats in use.
314 The number of floats for which space has been obtained from the
315 operating system, but that are not currently being used.
318 The number of intervals in use. Intervals are an internal
319 data structure used for representing text properties.
322 The number of intervals for which space has been obtained
323 from the operating system, but that are not currently being used.
326 The number of strings in use.
329 The number of string headers for which the space was obtained from the
330 operating system, but which are currently not in use. (A string
331 object consists of a header and the storage for the string text
332 itself; the latter is only allocated when the string is created.)
336 @defopt garbage-collection-messages
337 If this variable is non-@code{nil}, Emacs displays a message at the
338 beginning and end of garbage collection. The default value is
339 @code{nil}, meaning there are no such messages.
343 This is a normal hook that is run at the end of garbage collection.
344 Garbage collection is inhibited while the hook functions run, so be
345 careful writing them.
348 @defopt gc-cons-threshold
349 The value of this variable is the number of bytes of storage that must
350 be allocated for Lisp objects after one garbage collection in order to
351 trigger another garbage collection. A cons cell counts as eight bytes,
352 a string as one byte per character plus a few bytes of overhead, and so
353 on; space allocated to the contents of buffers does not count. Note
354 that the subsequent garbage collection does not happen immediately when
355 the threshold is exhausted, but only the next time the Lisp evaluator is
358 The initial threshold value is 400,000. If you specify a larger
359 value, garbage collection will happen less often. This reduces the
360 amount of time spent garbage collecting, but increases total memory use.
361 You may want to do this when running a program that creates lots of
364 You can make collections more frequent by specifying a smaller value,
365 down to 10,000. A value less than 10,000 will remain in effect only
366 until the subsequent garbage collection, at which time
367 @code{garbage-collect} will set the threshold back to 10,000.
370 The value return by @code{garbage-collect} describes the amount of
371 memory used by Lisp data, broken down by data type. By contrast, the
372 function @code{memory-limit} provides information on the total amount of
373 memory Emacs is currently using.
377 This function returns the address of the last byte Emacs has allocated,
378 divided by 1024. We divide the value by 1024 to make sure it fits in a
381 You can use this to get a general idea of how your actions affect the
386 This variable contains the total number of garbage collections
387 done so far in this Emacs session.
391 This variable contains the total number of seconds of elapsed time
392 during garbage collection so far in this Emacs session, as a floating
397 @section Memory Usage
399 These functions and variables give information about the total amount
400 of memory allocation that Emacs has done, broken down by data type.
401 Note the difference between these and the values returned by
402 @code{(garbage-collect)}; those count objects that currently exist, but
403 these count the number or size of all allocations, including those for
404 objects that have since been freed.
406 @defvar cons-cells-consed
407 The total number of cons cells that have been allocated so far
408 in this Emacs session.
411 @defvar floats-consed
412 The total number of floats that have been allocated so far
413 in this Emacs session.
416 @defvar vector-cells-consed
417 The total number of vector cells that have been allocated so far
418 in this Emacs session.
421 @defvar symbols-consed
422 The total number of symbols that have been allocated so far
423 in this Emacs session.
426 @defvar string-chars-consed
427 The total number of string characters that have been allocated so far
428 in this Emacs session.
431 @defvar misc-objects-consed
432 The total number of miscellaneous objects that have been allocated so
433 far in this Emacs session. These include markers and overlays, plus
434 certain objects not visible to users.
437 @defvar intervals-consed
438 The total number of intervals that have been allocated so far
439 in this Emacs session.
442 @defvar strings-consed
443 The total number of strings that have been allocated so far in this
447 @node Writing Emacs Primitives
448 @appendixsec Writing Emacs Primitives
449 @cindex primitive function internals
451 Lisp primitives are Lisp functions implemented in C. The details of
452 interfacing the C function so that Lisp can call it are handled by a few
453 C macros. The only way to really understand how to write new C code is
454 to read the source, but we can explain some things here.
456 An example of a special form is the definition of @code{or}, from
457 @file{eval.c}. (An ordinary function would have the same general
460 @cindex garbage collection protection
463 DEFUN ("or", For, Sor, 0, UNEVALLED, 0,
464 doc: /* Eval args until one of them yields non-nil, then return that value.
465 The remaining args are not evalled at all.
466 If all args return nil, return nil.
469 usage: (or CONDITIONS ...) */)
473 register Lisp_Object val;
474 Lisp_Object args_left;
489 val = Feval (Fcar (args_left));
492 args_left = Fcdr (args_left);
494 while (!NILP (args_left));
504 Let's start with a precise explanation of the arguments to the
505 @code{DEFUN} macro. Here is a template for them:
508 DEFUN (@var{lname}, @var{fname}, @var{sname}, @var{min}, @var{max}, @var{interactive}, @var{doc})
513 This is the name of the Lisp symbol to define as the function name; in
514 the example above, it is @code{or}.
517 This is the C function name for this function. This is
518 the name that is used in C code for calling the function. The name is,
519 by convention, @samp{F} prepended to the Lisp name, with all dashes
520 (@samp{-}) in the Lisp name changed to underscores. Thus, to call this
521 function from C code, call @code{For}. Remember that the arguments must
522 be of type @code{Lisp_Object}; various macros and functions for creating
523 values of type @code{Lisp_Object} are declared in the file
527 This is a C variable name to use for a structure that holds the data for
528 the subr object that represents the function in Lisp. This structure
529 conveys the Lisp symbol name to the initialization routine that will
530 create the symbol and store the subr object as its definition. By
531 convention, this name is always @var{fname} with @samp{F} replaced with
535 This is the minimum number of arguments that the function requires. The
536 function @code{or} allows a minimum of zero arguments.
539 This is the maximum number of arguments that the function accepts, if
540 there is a fixed maximum. Alternatively, it can be @code{UNEVALLED},
541 indicating a special form that receives unevaluated arguments, or
542 @code{MANY}, indicating an unlimited number of evaluated arguments (the
543 equivalent of @code{&rest}). Both @code{UNEVALLED} and @code{MANY} are
544 macros. If @var{max} is a number, it may not be less than @var{min} and
545 it may not be greater than seven.
548 This is an interactive specification, a string such as might be used as
549 the argument of @code{interactive} in a Lisp function. In the case of
550 @code{or}, it is 0 (a null pointer), indicating that @code{or} cannot be
551 called interactively. A value of @code{""} indicates a function that
552 should receive no arguments when called interactively.
555 This is the documentation string. It uses C comment syntax rather
556 than C string syntax because comment syntax requires nothing special
557 to include multiple lines. The @samp{doc:} identifies the comment
558 that follows as the documentation string. The @samp{/*} and @samp{*/}
559 delimiters that begin and end the comment are not part of the
560 documentation string.
562 If the last line of the documentation string begins with the keyword
563 @samp{usage:}, the rest of the line is treated as the argument list
564 for documentation purposes. This way, you can use different argument
565 names in the documentation string from the ones used in the C code.
567 All the usual rules for documentation strings in Lisp code
568 (@pxref{Documentation Tips}) apply to C code documentation strings
572 After the call to the @code{DEFUN} macro, you must write the argument
573 name list that every C function must have, followed by ordinary C
574 declarations for the arguments. For a function with a fixed maximum
575 number of arguments, declare a C argument for each Lisp argument, and
576 give them all type @code{Lisp_Object}. When a Lisp function has no
577 upper limit on the number of arguments, its implementation in C actually
578 receives exactly two arguments: the first is the number of Lisp
579 arguments, and the second is the address of a block containing their
580 values. They have types @code{int} and @w{@code{Lisp_Object *}}.
582 Within the function @code{For} itself, note the use of the macros
583 @code{GCPRO1} and @code{UNGCPRO}. @code{GCPRO1} is used to ``protect''
584 a variable from garbage collection---to inform the garbage collector that
585 it must look in that variable and regard its contents as an accessible
586 object. This is necessary whenever you call @code{Feval} or anything
587 that can directly or indirectly call @code{Feval}. At such a time, any
588 Lisp object that you intend to refer to again must be protected somehow.
589 @code{UNGCPRO} cancels the protection of the variables that are
590 protected in the current function. It is necessary to do this explicitly.
592 It suffices to ensure that at least one pointer to each object is
593 GC-protected; as long as the object is not recycled, all pointers to
594 it remain valid. So if you are sure that a local variable points to
595 an object that will be preserved by some other pointer, that local
596 variable does not need a GCPRO. (Formerly, strings were an exception
597 to this rule; in older Emacs versions, every pointer to a string
598 needed to be marked by GC.)
600 The macro @code{GCPRO1} protects just one local variable. If you
601 want to protect two, use @code{GCPRO2} instead; repeating
602 @code{GCPRO1} will not work. Macros, @code{GCPRO3}, @code{GCPRO4},
603 @code{GCPRO5}, and @code{GCPRO6} also exist. These macros implicitly
604 use local variables such as @code{gcpro1}; you must declare these
605 explicitly, with type @code{struct gcpro}. Thus, if you use
606 @code{GCPRO2}, you must declare @code{gcpro1} and @code{gcpro2}.
607 Alas, we can't explain all the tricky details here.
609 Built-in functions that take a variable number of arguments actually
610 accept two arguments at the C level: the number of Lisp arguments, and
611 a @code{Lisp_Object *} pointer to a C vector containing those Lisp
612 arguments. This C vector may be part of a Lisp vector, but it need
613 not be. The responsibility for using GCPRO to protecting the Lisp
614 arguments from GC if necessary rests with the caller in this case,
615 since the caller allocated or found the storage for them.
617 You must not use C initializers for static or global variables unless
618 the variables are never written once Emacs is dumped. These variables
619 with initializers are allocated in an area of memory that becomes
620 read-only (on certain operating systems) as a result of dumping Emacs.
623 Do not use static variables within functions---place all static
624 variables at top level in the file. This is necessary because Emacs on
625 some operating systems defines the keyword @code{static} as a null
626 macro. (This definition is used because those systems put all variables
627 declared static in a place that becomes read-only after dumping, whether
628 they have initializers or not.)
630 Defining the C function is not enough to make a Lisp primitive
631 available; you must also create the Lisp symbol for the primitive and
632 store a suitable subr object in its function cell. The code looks like
636 defsubr (&@var{subr-structure-name});
640 Here @var{subr-structure-name} is the name you used as the third
641 argument to @code{DEFUN}.
643 If you add a new primitive to a file that already has Lisp primitives
644 defined in it, find the function (near the end of the file) named
645 @code{syms_of_@var{something}}, and add the call to @code{defsubr}
646 there. If the file doesn't have this function, or if you create a new
647 file, add to it a @code{syms_of_@var{filename}} (e.g.,
648 @code{syms_of_myfile}). Then find the spot in @file{emacs.c} where all
649 of these functions are called, and add a call to
650 @code{syms_of_@var{filename}} there.
652 @vindex byte-boolean-vars
653 The function @code{syms_of_@var{filename}} is also the place to define
654 any C variables that are to be visible as Lisp variables.
655 @code{DEFVAR_LISP} makes a C variable of type @code{Lisp_Object} visible
656 in Lisp. @code{DEFVAR_INT} makes a C variable of type @code{int}
657 visible in Lisp with a value that is always an integer.
658 @code{DEFVAR_BOOL} makes a C variable of type @code{int} visible in Lisp
659 with a value that is either @code{t} or @code{nil}. Note that variables
660 defined with @code{DEFVAR_BOOL} are automatically added to the list
661 @code{byte-boolean-vars} used by the byte compiler.
663 If you define a file-scope C variable of type @code{Lisp_Object},
664 you must protect it from garbage-collection by calling @code{staticpro}
665 in @code{syms_of_@var{filename}}, like this:
668 staticpro (&@var{variable});
671 Here is another example function, with more complicated arguments.
672 This comes from the code in @file{window.c}, and it demonstrates the use
673 of macros and functions to manipulate Lisp objects.
677 DEFUN ("coordinates-in-window-p", Fcoordinates_in_window_p,
678 Scoordinates_in_window_p, 2, 2,
679 "xSpecify coordinate pair: \nXExpression which evals to window: ",
680 "Return non-nil if COORDINATES is in WINDOW.\n\
681 COORDINATES is a cons of the form (X . Y), X and Y being distances\n\
685 If they are on the border between WINDOW and its right sibling,\n\
686 `vertical-line' is returned.")
687 (coordinates, window)
688 register Lisp_Object coordinates, window;
694 CHECK_LIVE_WINDOW (window, 0);
695 CHECK_CONS (coordinates, 1);
696 x = XINT (Fcar (coordinates));
697 y = XINT (Fcdr (coordinates));
701 switch (coordinates_in_window (XWINDOW (window), &x, &y))
703 case 0: /* NOT in window at all. */
708 case 1: /* In text part of window. */
709 return Fcons (make_number (x), make_number (y));
713 case 2: /* In mode line of window. */
718 case 3: /* On right border of window. */
719 return Qvertical_line;
730 Note that C code cannot call functions by name unless they are defined
731 in C. The way to call a function written in Lisp is to use
732 @code{Ffuncall}, which embodies the Lisp function @code{funcall}. Since
733 the Lisp function @code{funcall} accepts an unlimited number of
734 arguments, in C it takes two: the number of Lisp-level arguments, and a
735 one-dimensional array containing their values. The first Lisp-level
736 argument is the Lisp function to call, and the rest are the arguments to
737 pass to it. Since @code{Ffuncall} can call the evaluator, you must
738 protect pointers from garbage collection around the call to
741 The C functions @code{call0}, @code{call1}, @code{call2}, and so on,
742 provide handy ways to call a Lisp function conveniently with a fixed
743 number of arguments. They work by calling @code{Ffuncall}.
745 @file{eval.c} is a very good file to look through for examples;
746 @file{lisp.h} contains the definitions for some important macros and
749 If you define a function which is side-effect free, update the code in
750 @file{byte-opt.el} which binds @code{side-effect-free-fns} and
751 @code{side-effect-and-error-free-fns} to include it. This will help the
754 @node Object Internals
755 @appendixsec Object Internals
756 @cindex object internals
758 GNU Emacs Lisp manipulates many different types of data. The actual
759 data are stored in a heap and the only access that programs have to it
760 is through pointers. Pointers are thirty-two bits wide in most
761 implementations. Depending on the operating system and type of machine
762 for which you compile Emacs, twenty-eight bits are used to address the
763 object, and the remaining four bits are used for a GC mark bit and the
764 tag that identifies the object's type.
766 Because Lisp objects are represented as tagged pointers, it is always
767 possible to determine the Lisp data type of any object. The C data type
768 @code{Lisp_Object} can hold any Lisp object of any data type. Ordinary
769 variables have type @code{Lisp_Object}, which means they can hold any
770 type of Lisp value; you can determine the actual data type only at run
771 time. The same is true for function arguments; if you want a function
772 to accept only a certain type of argument, you must check the type
773 explicitly using a suitable predicate (@pxref{Type Predicates}).
774 @cindex type checking internals
777 * Buffer Internals:: Components of a buffer structure.
778 * Window Internals:: Components of a window structure.
779 * Process Internals:: Components of a process structure.
782 @node Buffer Internals
783 @appendixsubsec Buffer Internals
784 @cindex internals, of buffer
785 @cindex buffer internals
787 Buffers contain fields not directly accessible by the Lisp programmer.
788 We describe them here, naming them by the names used in the C code.
789 Many are accessible indirectly in Lisp programs via Lisp primitives.
791 Two structures are used to represent buffers in C. The
792 @code{buffer_text} structure contains fields describing the text of a
793 buffer; the @code{buffer} structure holds other fields. In the case
794 of indirect buffers, two or more @code{buffer} structures reference
795 the same @code{buffer_text} structure.
797 Here is a list of the @code{struct buffer_text} fields:
801 This field contains the actual address of the buffer contents.
804 This holds the character position of the gap in the buffer.
808 This field contains the character position of the end of the buffer
812 Contains the byte position of the gap.
815 Holds the byte position of the end of the buffer text.
818 Contains the size of buffer's gap. @xref{Buffer Gap}.
821 This field counts buffer-modification events for this buffer. It is
822 incremented for each such event, and never otherwise changed.
825 Contains the previous value of @code{modiff}, as of the last time a
826 buffer was visited or saved in a file.
829 Counts modifications to overlays analogous to @code{modiff}.
832 Holds the number of characters at the start of the text that are known
833 to be unchanged since the last redisplay that finished.
836 Holds the number of characters at the end of the text that are known to
837 be unchanged since the last redisplay that finished.
839 @item unchanged_modified
840 Contains the value of @code{modiff} at the time of the last redisplay
841 that finished. If this value matches @code{modiff},
842 @code{beg_unchanged} and @code{end_unchanged} contain no useful
845 @item overlay_unchanged_modified
846 Contains the value of @code{overlay_modiff} at the time of the last
847 redisplay that finished. If this value matches @code{overlay_modiff},
848 @code{beg_unchanged} and @code{end_unchanged} contain no useful
852 The markers that refer to this buffer. This is actually a single
853 marker, and successive elements in its marker @code{chain} are the other
854 markers referring to this buffer text.
857 Contains the interval tree which records the text properties of this
861 The fields of @code{struct buffer} are:
865 Points to the next buffer, in the chain of all buffers including killed
866 buffers. This chain is used only for garbage collection, in order to
867 collect killed buffers properly. Note that vectors, and most kinds of
868 objects allocated as vectors, are all on one chain, but buffers are on a
869 separate chain of their own.
872 This is a @code{struct buffer_text} structure. In an ordinary buffer,
873 it holds the buffer contents. In indirect buffers, this field is not
877 This points to the @code{buffer_text} structure that is used for this
878 buffer. In an ordinary buffer, this is the @code{own_text} field above.
879 In an indirect buffer, this is the @code{own_text} field of the base
883 Contains the character position of point in a buffer.
886 Contains the byte position of point in a buffer.
889 This field contains the character position of the beginning of the
890 accessible range of text in the buffer.
893 This field contains the byte position of the beginning of the
894 accessible range of text in the buffer.
897 This field contains the character position of the end of the
898 accessible range of text in the buffer.
901 This field contains the byte position of the end of the
902 accessible range of text in the buffer.
905 In an indirect buffer, this points to the base buffer. In an ordinary
908 @item local_var_flags
909 This field contains flags indicating that certain variables are local in
910 this buffer. Such variables are declared in the C code using
911 @code{DEFVAR_PER_BUFFER}, and their buffer-local bindings are stored in
912 fields in the buffer structure itself. (Some of these fields are
913 described in this table.)
916 This field contains the modification time of the visited file. It is
917 set when the file is written or read. Before writing the buffer into a
918 file, this field is compared to the modification time of the file to see
919 if the file has changed on disk. @xref{Buffer Modification}.
921 @item auto_save_modified
922 This field contains the time when the buffer was last auto-saved.
924 @item auto_save_failure_time
925 The time at which we detected a failure to auto-save, or -1 if we didn't
928 @item last_window_start
929 This field contains the @code{window-start} position in the buffer as of
930 the last time the buffer was displayed in a window.
933 This flag is set when narrowing changes in a buffer.
935 @item prevent_redisplay_optimizations_p
936 this flag indicates that redisplay optimizations should not be used
937 to display this buffer.
940 This field points to the buffer's undo list. @xref{Undo}.
943 The buffer name is a string that names the buffer. It is guaranteed to
944 be unique. @xref{Buffer Names}.
947 The name of the file visited in this buffer, or @code{nil}.
950 The directory for expanding relative file names.
953 Length of the file this buffer is visiting, when last read or saved.
954 This and other fields concerned with saving are not kept in the
955 @code{buffer_text} structure because indirect buffers are never saved.
957 @item auto_save_file_name
958 File name used for auto-saving this buffer. This is not in the
959 @code{buffer_text} because it's not used in indirect buffers at all.
962 Non-@code{nil} means this buffer is read-only.
965 This field contains the mark for the buffer. The mark is a marker,
966 hence it is also included on the list @code{markers}. @xref{The Mark}.
968 @item local_var_alist
969 This field contains the association list describing the buffer-local
970 variable bindings of this buffer, not including the built-in
971 buffer-local bindings that have special slots in the buffer object.
972 (Those slots are omitted from this table.) @xref{Buffer-Local
976 Symbol naming the major mode of this buffer, e.g., @code{lisp-mode}.
979 Pretty name of major mode, e.g., @code{"Lisp"}.
981 @item mode_line_format
982 Mode line element that controls the format of the mode line. If this
983 is @code{nil}, no mode line will be displayed.
985 @item header_line_format
986 This field is analoguous to @code{mode_line_format} for the mode
987 line displayed at the top of windows.
990 This field holds the buffer's local keymap. @xref{Keymaps}.
993 This buffer's local abbrevs.
996 This field contains the syntax table for the buffer. @xref{Syntax Tables}.
999 This field contains the category table for the buffer.
1001 @item case_fold_search
1002 The value of @code{case-fold-search} in this buffer.
1005 The value of @code{tab-width} in this buffer.
1008 The value of @code{fill-column} in this buffer.
1011 The value of @code{left-margin} in this buffer.
1013 @item auto_fill_function
1014 The value of @code{auto-fill-function} in this buffer.
1016 @item downcase_table
1017 This field contains the conversion table for converting text to lower case.
1021 This field contains the conversion table for converting text to upper case.
1024 @item case_canon_table
1025 This field contains the conversion table for canonicalizing text for
1026 case-folding search. @xref{Case Tables}.
1028 @item case_eqv_table
1029 This field contains the equivalence table for case-folding search.
1032 @item truncate_lines
1033 The value of @code{truncate-lines} in this buffer.
1036 The value of @code{ctl-arrow} in this buffer.
1038 @item selective_display
1039 The value of @code{selective-display} in this buffer.
1041 @item selective_display_ellipsis
1042 The value of @code{selective-display-ellipsis} in this buffer.
1045 An alist of the minor modes of this buffer.
1047 @item overwrite_mode
1048 The value of @code{overwrite_mode} in this buffer.
1051 The value of @code{abbrev-mode} in this buffer.
1054 This field contains the buffer's display table, or @code{nil} if it doesn't
1055 have one. @xref{Display Tables}.
1058 This field contains the time when the buffer was last saved, as an integer.
1059 @xref{Buffer Modification}.
1062 This field is non-@code{nil} if the buffer's mark is active.
1064 @item overlays_before
1065 This field holds a list of the overlays in this buffer that end at or
1066 before the current overlay center position. They are sorted in order of
1067 decreasing end position.
1069 @item overlays_after
1070 This field holds a list of the overlays in this buffer that end after
1071 the current overlay center position. They are sorted in order of
1072 increasing beginning position.
1074 @item overlay_center
1075 This field holds the current overlay center position. @xref{Overlays}.
1077 @item enable_multibyte_characters
1078 This field holds the buffer's local value of
1079 @code{enable-multibyte-characters}---either @code{t} or @code{nil}.
1081 @item buffer_file_coding_system
1082 The value of @code{buffer-file-coding-system} in this buffer.
1085 The value of @code{buffer-file-format} in this buffer.
1088 In an indirect buffer, or a buffer that is the base of an indirect
1089 buffer, this holds a marker that records point for this buffer when the
1090 buffer is not current.
1093 In an indirect buffer, or a buffer that is the base of an indirect
1094 buffer, this holds a marker that records @code{begv} for this buffer
1095 when the buffer is not current.
1098 In an indirect buffer, or a buffer that is the base of an indirect
1099 buffer, this holds a marker that records @code{zv} for this buffer when
1100 the buffer is not current.
1103 The truename of the visited file, or @code{nil}.
1105 @item invisibility_spec
1106 The value of @code{buffer-invisibility-spec} in this buffer.
1108 @item last_selected_window
1109 This is the last window that was selected with this buffer in it, or @code{nil}
1110 if that window no longer displays this buffer.
1113 This field is incremented each time the buffer is displayed in a window.
1115 @item left_margin_width
1116 The value of @code{left-margin-width} in this buffer.
1118 @item right_margin_width
1119 The value of @code{right-margin-width} in this buffer.
1121 @item indicate_empty_lines
1122 Non-@code{nil} means indicate empty lines (lines with no text) with a
1123 small bitmap in the fringe, when using a window system that can do it.
1126 This holds a time stamp that is updated each time this buffer is
1127 displayed in a window.
1129 @item scroll_up_aggressively
1130 The value of @code{scroll-up-aggressively} in this buffer.
1132 @item scroll_down_aggressively
1133 The value of @code{scroll-down-aggressively} in this buffer.
1136 @node Window Internals
1137 @appendixsubsec Window Internals
1138 @cindex internals, of window
1139 @cindex window internals
1141 Windows have the following accessible fields:
1145 The frame that this window is on.
1148 Non-@code{nil} if this window is a minibuffer window.
1151 Internally, Emacs arranges windows in a tree; each group of siblings has
1152 a parent window whose area includes all the siblings. This field points
1153 to a window's parent.
1155 Parent windows do not display buffers, and play little role in display
1156 except to shape their child windows. Emacs Lisp programs usually have
1157 no access to the parent windows; they operate on the windows at the
1158 leaves of the tree, which actually display buffers.
1160 The following four fields also describe the window tree structure.
1163 In a window subdivided horizontally by child windows, the leftmost child.
1164 Otherwise, @code{nil}.
1167 In a window subdivided vertically by child windows, the topmost child.
1168 Otherwise, @code{nil}.
1171 The next sibling of this window. It is @code{nil} in a window that is
1172 the rightmost or bottommost of a group of siblings.
1175 The previous sibling of this window. It is @code{nil} in a window that
1176 is the leftmost or topmost of a group of siblings.
1179 This is the left-hand edge of the window, measured in columns. (The
1180 leftmost column on the screen is @w{column 0}.)
1183 This is the top edge of the window, measured in lines. (The top line on
1184 the screen is @w{line 0}.)
1187 The height of the window, measured in lines.
1190 The width of the window, measured in columns. This width includes the
1191 scroll bar and fringes, and/or the separator line on the right of the
1195 The buffer that the window is displaying. This may change often during
1196 the life of the window.
1199 The position in the buffer that is the first character to be displayed
1203 @cindex window point internals
1204 This is the value of point in the current buffer when this window is
1205 selected; when it is not selected, it retains its previous value.
1208 If this flag is non-@code{nil}, it says that the window has been
1209 scrolled explicitly by the Lisp program. This affects what the next
1210 redisplay does if point is off the screen: instead of scrolling the
1211 window to show the text around point, it moves point to a location that
1214 @item frozen_window_start_p
1215 This field is set temporarily to 1 to indicate to redisplay that
1216 @code{start} of this window should not be changed, even if point
1219 @item start_at_line_beg
1220 Non-@code{nil} means current value of @code{start} was the beginning of a line
1224 Non-@code{nil} means don't delete this window for becoming ``too small''.
1226 @item height_fixed_p
1227 This field is temporarily set to 1 to fix the height of the selected
1228 window when the echo area is resized.
1231 This is the last time that the window was selected. The function
1232 @code{get-lru-window} uses this field.
1234 @item sequence_number
1235 A unique number assigned to this window when it was created.
1238 The @code{modiff} field of the window's buffer, as of the last time
1239 a redisplay completed in this window.
1241 @item last_overlay_modified
1242 The @code{overlay_modiff} field of the window's buffer, as of the last
1243 time a redisplay completed in this window.
1246 The buffer's value of point, as of the last time a redisplay completed
1250 A non-@code{nil} value means the window's buffer was ``modified'' when the
1251 window was last updated.
1253 @item vertical_scroll_bar
1254 This window's vertical scroll bar.
1256 @item left_margin_width
1257 The width of the left margin in this window, or @code{nil} not to
1258 specify it (in which case the buffer's value of @code{left-margin-width}
1261 @item right_margin_width
1262 Likewise for the right margin.
1270 @item window_end_pos
1271 This is computed as @code{z} minus the buffer position of the last glyph
1272 in the current matrix of the window. The value is only valid if
1273 @code{window_end_valid} is not @code{nil}.
1275 @item window_end_bytepos
1276 The byte position corresponding to @code{window_end_pos}.
1278 @item window_end_vpos
1279 The window-relative vertical position of the line containing
1280 @code{window_end_pos}.
1282 @item window_end_valid
1283 This field is set to a non-@code{nil} value if @code{window_end_pos} is truly
1284 valid. This is @code{nil} if nontrivial redisplay is preempted since in that
1285 case the display that @code{window_end_pos} was computed for did not get
1288 @item redisplay_end_trigger
1289 If redisplay in this window goes beyond this buffer position, it runs
1290 run the @code{redisplay-end-trigger-hook}.
1295 ??? Are temporary storage areas.
1299 A structure describing where the cursor is in this window.
1302 The value of @code{cursor} as of the last redisplay that finished.
1305 A structure describing where the cursor of this window physically is.
1307 @item phys_cursor_type
1308 The type of cursor that was last displayed on this window.
1310 @item phys_cursor_on_p
1311 This field is non-zero if the cursor is physically on.
1314 Non-zero means the cursor in this window is logically on.
1316 @item last_cursor_off_p
1317 This field contains the value of @code{cursor_off_p} as of the time of
1320 @item must_be_updated_p
1321 This is set to 1 during redisplay when this window must be updated.
1324 This is the number of columns that the display in the window is scrolled
1325 horizontally to the left. Normally, this is 0.
1328 Vertical scroll amount, in pixels. Normally, this is 0.
1331 Non-@code{nil} if this window is dedicated to its buffer.
1334 The window's display table, or @code{nil} if none is specified for it.
1336 @item update_mode_line
1337 Non-@code{nil} means this window's mode line needs to be updated.
1339 @item base_line_number
1340 The line number of a certain position in the buffer, or @code{nil}.
1341 This is used for displaying the line number of point in the mode line.
1344 The position in the buffer for which the line number is known, or
1345 @code{nil} meaning none is known.
1347 @item region_showing
1348 If the region (or part of it) is highlighted in this window, this field
1349 holds the mark position that made one end of that region. Otherwise,
1350 this field is @code{nil}.
1352 @item column_number_displayed
1353 The column number currently displayed in this window's mode line, or @code{nil}
1354 if column numbers are not being displayed.
1356 @item current_matrix
1357 A glyph matrix describing the current display of this window.
1359 @item desired_matrix
1360 A glyph matrix describing the desired display of this window.
1363 @node Process Internals
1364 @appendixsubsec Process Internals
1365 @cindex internals, of process
1366 @cindex process internals
1368 The fields of a process are:
1372 A string, the name of the process.
1375 A list containing the command arguments that were used to start this
1379 A function used to accept output from the process instead of a buffer,
1383 A function called whenever the process receives a signal, or @code{nil}.
1386 The associated buffer of the process.
1389 An integer, the Unix process @sc{id}.
1392 A flag, non-@code{nil} if this is really a child process.
1393 It is @code{nil} for a network connection.
1396 A marker indicating the position of the end of the last output from this
1397 process inserted into the buffer. This is often but not always the end
1400 @item kill_without_query
1401 If this is non-@code{nil}, killing Emacs while this process is still
1402 running does not ask for confirmation about killing the process.
1404 @item raw_status_low
1405 @itemx raw_status_high
1406 These two fields record 16 bits each of the process status returned by
1407 the @code{wait} system call.
1410 The process status, as @code{process-status} should return it.
1414 If these two fields are not equal, a change in the status of the process
1415 needs to be reported, either by running the sentinel or by inserting a
1416 message in the process buffer.
1419 Non-@code{nil} if communication with the subprocess uses a @sc{pty};
1420 @code{nil} if it uses a pipe.
1423 The file descriptor for input from the process.
1426 The file descriptor for output to the process.
1429 The file descriptor for the terminal that the subprocess is using. (On
1430 some systems, there is no need to record this, so the value is
1434 The name of the terminal that the subprocess is using,
1435 or @code{nil} if it is using pipes.
1437 @item decode_coding_system
1438 Coding-system for decoding the input from this process.
1441 A working buffer for decoding.
1443 @item decoding_carryover
1444 Size of carryover in decoding.
1446 @item encode_coding_system
1447 Coding-system for encoding the output to this process.
1450 A working buffer for encoding.
1452 @item encoding_carryover
1453 Size of carryover in encoding.
1455 @item inherit_coding_system_flag
1456 Flag to set @code{coding-system} of the process buffer from the
1457 coding system used to decode process output.
1461 arch-tag: 4b2c33bc-d7e4-43f5-bc20-27c0db52a53e