1 This file describes various problems that have been encountered
2 in compiling, installing and running GNU Emacs.
5 * Characters from the mule-unicode charsets aren't displayed under X.
7 XFree86 4 contains many fonts in iso10646-1 encoding which have
8 minimal character repertoires (whereas the encoding is meant to be a
9 reasonable indication of the repertoire). Emacs may choose one of
10 these to display characters from the mule-unicode charsets and then
11 typically won't be able to find the glyphs to display many characters.
12 (Check with C-u C-x = .) To avoid this, you may need to use a fontset
13 which sets the font for the mule-unicode sets explicitly. E.g. to use
14 GNU unifont, include in the fontset spec:
16 mule-unicode-2500-33ff:-gnu-unifont-*-iso10646-1,\
17 mule-unicode-e000-ffff:-gnu-unifont-*-iso10646-1,\
18 mule-unicode-0100-24ff:-gnu-unifont-*-iso10646-1
20 * Problems with file dialogs in Emacs built with Open Motif.
22 When Emacs 21 is built with Open Motif 2.1, it can happen that the
23 graphical file dialog boxes do not work properly. The "OK", "Filter"
24 and "Cancel" buttons do not respond to mouse clicks. Dragging the
25 file dialog window usually causes the buttons to work again.
27 The solution is to use LessTif instead. LessTif is a free replacement
28 for Motif. See the file INSTALL for information on how to do this.
30 Another workaround is not to use the mouse to trigger file prompts,
31 but to use the keyboard. This way, you will be prompted for a file in
32 the minibuffer instead of a graphical file dialog.
34 * Emacs reports a BadAtom error (from X) running on Solaris 7 or 8.
36 This happens when Emacs was built on some other version of Solaris.
37 Rebuild it on Solaris 8.
39 * Mule-UCS loads very slowly.
41 Changes to Emacs internals interact badly with Mule-UCS's `un-define'
42 library, which is the usual interface to Mule-UCS. Apply the
43 following patch to Mule-UCS 0.84 and rebuild it. That will help,
44 though loading will still be slower than in Emacs 20. (Some
45 distributions, such as Debian, may already have applied such a patch.)
47 --- lisp/un-define.el 6 Mar 2001 22:41:38 -0000 1.30
48 +++ lisp/un-define.el 19 Apr 2002 18:34:26 -0000
49 @@ -610,13 +624,21 @@ by calling post-read-conversion and pre-
55 - (mucs-define-coding-system
56 - (nth 0 y) (nth 1 y) (nth 2 y)
57 - (nth 3 y) (nth 4 y) (nth 5 y) (nth 6 y))
58 - (coding-system-put (car y) 'alias-coding-systems (list (car x))))
60 + (if (fboundp 'register-char-codings)
61 + ;; Mule 5, where we don't need the eol-type specified and
62 + ;; register-char-codings may be very slow for these coding
63 + ;; system definitions.
65 + (mucs-define-coding-system
66 + (car x) (nth 1 y) (nth 2 y)
67 + (nth 3 y) (nth 4 y) (nth 5 y)))
70 + (mucs-define-coding-system
71 + (nth 0 y) (nth 1 y) (nth 2 y)
72 + (nth 3 y) (nth 4 y) (nth 5 y) (nth 6 y))
73 + (coding-system-put (car y) 'alias-coding-systems (list (car x)))))
77 ?u "UTF-8 coding system"
79 Note that Emacs has native support for Unicode, roughly equivalent to
80 Mule-UCS's, so you may not need it.
82 * Building Emacs with GCC 2.9x fails in the `src' directory.
84 This may happen if you use a development version of GNU `cpp' from one
85 of the GCC snapshots between Oct 2000 and Feb 2001, or from a released
86 version of GCC newer than 2.95.2 which was prepared around those
87 dates; similar problems were reported with some snapshots of GCC 3.1
88 around Sep 30 2001. The preprocessor in those versions is
89 incompatible with a traditional Unix cpp (e.g., it expands ".." into
90 ". .", which breaks relative file names that reference the parent
91 directory; or inserts TAB characters before lines that set Make
94 The solution is to make sure the preprocessor is run with the
95 `-traditional' option. The `configure' script does that automatically
96 when it detects the known problems in your cpp, but you might hit some
97 unknown ones. To force the `configure' script to use `-traditional',
98 run the script like this:
100 CPP='gcc -E -traditional' ./configure ...
102 (replace the ellipsis "..." with any additional arguments you pass to
105 Note that this problem does not pertain to the MS-Windows port of
106 Emacs, since it doesn't use the preprocessor to generate Makefiles.
108 * Building Emacs with a system compiler fails to link because of an
109 undefined symbol such as __eprintf which does not appear in Emacs.
111 This can happen if some of the libraries linked into Emacs were built
112 with GCC, but Emacs itself is being linked with a compiler other than
113 GCC. Object files compiled with GCC might need some helper functions
114 from libgcc.a, the library which comes with GCC, but the system
115 compiler does not instruct the linker to search libgcc.a during the
118 A solution is to link with GCC, like this:
122 Since the .o object files already exist, this will not recompile Emacs
123 with GCC, but just restart by trying again to link temacs.
125 * Building the MS-Windows port with Cygwin GCC can fail.
127 Emacs may not build using recent Cygwin builds of GCC, such as Cygwin
128 version 1.1.8, using the default configure settings. It appears to be
129 necessary to specify the -mwin32 flag when compiling, and define
132 configure --with-gcc --cflags -mwin32 --cflags -D__MSVCRT__
134 * Building the MS-Windows port with Leim fails in the `leim' directory.
136 The error message might be something like this:
138 Converting d:/emacs-21.3/leim/CXTERM-DIC/4Corner.tit to quail-package...
139 Invalid ENCODE: value in TIT dictionary
140 NMAKE : fatal error U1077: '"../src/obj-spd/i386/emacs.exe"' : return code
144 This can happen if the Leim distribution is unpacked with a program
145 which converts the `*.tit' files to DOS-style CR-LF text format. The
146 `*.tit' files in the leim/CXTERM-DIC directory require Unix-style line
147 endings to compile properly, because Emacs reads them without any code
150 The solution is to make sure the program used to unpack Leim does not
151 change the files' line endings behind your back. The GNU FTP site has
152 in the `/gnu/emacs/windows' directory a program called `djtarnt.exe'
153 which can be used to unpack `.tar.gz' and `.zip' archives without
156 * Emacs crashes when dumping itself on Mac PPC running Yellow Dog GNU/Linux.
158 The crashes happen inside the function Fmake_symbol; here's a typical
159 C backtrace printed by GDB:
161 0x190c0c0 in Fmake_symbol ()
163 #0 0x190c0c0 in Fmake_symbol ()
164 #1 0x1942ca4 in init_obarray ()
165 #2 0x18b3500 in main ()
166 #3 0x114371c in __libc_start_main (argc=5, argv=0x7ffff5b4, envp=0x7ffff5cc,
168 This could happen because GCC version 2.95 and later changed the base
169 of the load address to 0x10000000. Emacs needs to be told about this,
170 but we currently cannot do that automatically, because that breaks
171 other versions of GNU/Linux on the MacPPC. Until we find a way to
172 distinguish between the Yellow Dog and the other varieties of
173 GNU/Linux systems on the PPC, you will have to manually uncomment the
174 following section near the end of the file src/m/macppc.h in the Emacs
177 #if 0 /* This breaks things on PPC GNU/Linux except for Yellowdog,
178 even with identical GCC, as, ld. Let's take it out until we
179 know what's really going on here. */
180 /* GCC 2.95 and newer on GNU/Linux PPC changed the load address to
182 #if defined __linux__
183 #if __GNUC__ > 2 || (__GNUC__ == 2 && __GNUC_MINOR__ >= 95)
184 #define DATA_SEG_BITS 0x10000000
189 Remove the "#if 0" and "#endif" directives which surround this, save
190 the file, and then reconfigure and rebuild Emacs. The dumping process
193 * JPEG images aren't displayed.
195 This has been reported when Emacs is built with jpeg-6a library.
196 Upgrading to jpeg-6b solves the problem. Configure checks for the
197 correct version, but this problem could occur if a binary built
198 against a shared libjpeg is run on a system with an older version.
200 * Building `ctags' for MS-Windows with the MinGW port of GCC fails.
202 This might happen due to a bug in the MinGW header assert.h, which
203 defines the `assert' macro with a trailing semi-colon. The following
204 patch to assert.h should solve this:
206 *** include/assert.h.orig Sun Nov 7 02:41:36 1999
207 --- include/assert.h Mon Jan 29 11:49:10 2001
211 * If not debugging, assert does nothing.
213 ! #define assert(x) ((void)0);
215 #else /* debugging enabled */
219 * If not debugging, assert does nothing.
221 ! #define assert(x) ((void)0)
223 #else /* debugging enabled */
227 * Improving performance with slow X connections
229 There are several ways to improve this performance, any subset of which can
230 be carried out at the same time:
232 1) If you don't need X Input Methods (XIM) for entering text in some
233 language you use, you can improve performance on WAN links by
234 configuring Emacs with option `--without-xim'. Configuring Emacs
235 without XIM does not affect the use of Emacs' own input methods, which
236 are part of the Leim package.
238 2) If the connection is very slow, you might also want to consider
239 switching off scroll bars, menu bar, and tool bar.
241 3) Use ssh to forward the X connection, and enable compression on this
242 forwarded X connection (ssh -XC remotehostname emacs ...).
244 4) Use lbxproxy on the remote end of the connection. This is an interface
245 to the low bandwidth X extension in most modern X servers, which
246 improves performance dramatically, at the slight expense of correctness
247 of the X protocol. lbxproxy acheives the performance gain by grouping
248 several X requests in one TCP packet and sending them off together,
249 instead of requiring a round-trip for each X request in a seperate
250 packet. The switches that seem to work best for emacs are:
251 -noatomsfile -nowinattr -cheaterrors -cheatevents
252 Note that the -nograbcmap option is known to cause problems.
253 For more about lbxproxy, see:
254 http://www.xfree86.org/4.2.0/lbxproxy.1.html
256 * Getting a Meta key on the FreeBSD console
258 By default, neither Alt nor any other key acts as a Meta key on
259 FreeBSD, but this can be changed using kbdcontrol(1). Dump the
260 current keymap to a file with the command
262 $ kbdcontrol -d >emacs.kbd
264 Edit emacs.kbd, and give the key you want to be the Meta key the
265 definition `meta'. For instance, if your keyboard has a ``Windows''
266 key with scan code 105, change the line for scan code 105 in emacs.kbd
269 105 meta meta meta meta meta meta meta meta O
271 to make the Windows key the Meta key. Load the new keymap with
273 $ kbdcontrol -l emacs.kbd
275 * Emacs' xterm-mouse-mode doesn't work on the Gnome terminal.
277 A symptom of this bug is that double-clicks insert a control sequence
278 into the buffer. The reason this happens is an apparent
279 incompatibility of the Gnome terminal with Xterm, which also affects
280 other programs using the Xterm mouse interface. A problem report has
283 * Emacs pauses for several seconds when changing the default font
285 This has been reported for fvwm 2.2.5 and the window manager of KDE
286 2.1. The reason for the pause is Xt waiting for a ConfigureNotify
287 event from the window manager, which the window manager doesn't send.
288 Xt stops waiting after a default timeout of usually 5 seconds.
290 A workaround for this is to add something like
292 emacs.waitForWM: false
294 to your X resources. Alternatively, add `(wait-for-wm . nil)' to a
295 frame's parameter list, like this:
297 (modify-frame-parameters nil '((wait-for-wm . nil)))
299 (this should go into your `.emacs' file).
301 * Underlines appear at the wrong position.
303 This is caused by fonts having a wrong UNDERLINE_POSITION property.
304 Examples are the font 7x13 on XFree prior to version 4.1, or the jmk
305 neep font from the Debian xfonts-jmk package. To circumvent this
306 problem, set x-use-underline-position-properties to nil in your
309 To see what is the value of UNDERLINE_POSITION defined by the font,
310 type `xlsfonts -lll FONT' and look at the font's UNDERLINE_POSITION
313 * When using Xaw3d scroll bars without arrows, the very first mouse
314 click in a scroll bar might be ignored by the scroll bar widget. This
315 is probably a bug in Xaw3d; when Xaw3d is compiled with arrows, the
318 * There are known binary incompatibilities between Xaw, Xaw3d, neXtaw,
319 XawM and the few other derivatives of Xaw. So when you compile with
320 one of these, it may not work to dynamically link with another one.
321 For example, strange problems, such as Emacs exiting when you type
322 "C-x 1", were reported when Emacs compiled with Xaw3d and libXaw was
323 used with neXtaw at run time.
325 The solution is to rebuild Emacs with the toolkit version you actually
326 want to use, or set LD_PRELOAD to preload the same toolkit version you
329 * Clicking C-mouse-2 in the scroll bar doesn't split the window.
331 This currently doesn't work with scroll-bar widgets (and we don't know
332 a good way of implementing it with widgets). If Emacs is configured
333 --without-toolkit-scroll-bars, C-mouse-2 on the scroll bar does work.
335 * Emacs aborts inside the function `tparam1'.
337 This can happen if Emacs was built without terminfo support, but the
338 terminal's capabilities use format that is only supported by terminfo.
339 If your system has ncurses installed, this might happen if your
340 version of ncurses is broken; upgrading to a newer version of ncurses
341 and reconfiguring and rebuilding Emacs should solve this.
343 All modern systems support terminfo, so even if ncurses is not the
344 problem, you should look for a way to configure Emacs so that it uses
347 * Error messages about undefined colors on X.
349 The messages might say something like this:
351 Unable to load color "grey95"
353 (typically, in the `*Messages*' buffer), or something like this:
355 Error while displaying tooltip: (error Undefined color lightyellow)
357 These problems could happen if some other X program has used up too
358 many colors of the X palette, leaving Emacs with insufficient system
359 resources to load all the colors it needs.
361 A solution is to exit the offending X programs before starting Emacs.
363 * Colors are not available on a tty or in xterm.
365 Emacs 21 supports colors on character terminals and terminal
366 emulators, but this support relies on the terminfo or termcap database
367 entry to specify that the display supports color. Emacs looks at the
368 "Co" capability for the terminal to find out how many colors are
369 supported; it should be non-zero to activate the color support within
370 Emacs. (Most color terminals support 8 or 16 colors.) If your system
371 uses terminfo, the name of the capability equivalent to "Co" is
374 In addition to the "Co" capability, Emacs needs the "op" (for
375 ``original pair'') capability, which tells how to switch the terminal
376 back to the default foreground and background colors. Emacs will not
377 use colors if this capability is not defined. If your terminal entry
378 doesn't provide such a capability, try using the ANSI standard escape
379 sequence \E[00m (that is, define a new termcap/terminfo entry and make
380 it use your current terminal's entry plus \E[00m for the "op"
383 Finally, the "NC" capability (terminfo name: "ncv") tells Emacs which
384 attributes cannot be used with colors. Setting this capability
385 incorrectly might have the effect of disabling colors; try setting
386 this capability to `0' (zero) and see if that helps.
388 Emacs uses the database entry for the terminal whose name is the value
389 of the environment variable TERM. With `xterm', a common terminal
390 entry that supports color is `xterm-color', so setting TERM's value to
391 `xterm-color' might activate the color support on an xterm-compatible
394 Beginning with version 21.3, Emacs supports the --color command-line
395 option which may be used to force Emacs to use one of a few popular
396 modes for getting colors on a tty. For example, --color=ansi8 sets up
397 for using the ANSI-standard escape sequences that support 8 colors.
399 Some modes do not use colors unless you turn on the Font-lock mode.
400 Some people have long ago set their `~/.emacs' files to turn on
401 Font-lock on X only, so they won't see colors on a tty. The
402 recommended way of turning on Font-lock is by typing "M-x
403 global-font-lock-mode RET" or by customizing the variable
404 `global-font-lock-mode'.
406 * Emacs on a tty switches the cursor to large blinking block.
408 This was reported to happen on some GNU/Linux systems which use
409 ncurses version 5.0, but could be relevant for other versions as well.
410 These versions of ncurses come with a `linux' terminfo entry, where
411 the "cvvis" capability (termcap "vs") is defined as "\E[?25h\E[?8c"
412 (show cursor, change size). This escape sequence switches on a
413 blinking hardware text-mode cursor whose size is a full character
414 cell. This blinking cannot be stopped, since a hardware cursor
417 A work-around is to redefine the "cvvis" capability so that it
418 enables a *software* cursor. The software cursor works by inverting
419 the colors of the character at point, so what you see is a block
420 cursor that doesn't blink. For this to work, you need to redefine
421 the "cnorm" capability as well, so that it operates on the software
422 cursor instead of the hardware cursor.
424 To this end, run "infocmp linux > linux-term", edit the file
425 `linux-term' to make both the "cnorm" and "cvvis" capabilities send
426 the sequence "\E[?25h\E[?17;0;64c", and then run "tic linux-term" to
427 produce a modified terminfo entry.
429 Alternatively, if you want a blinking underscore as your Emacs cursor,
430 change the "cvvis" capability to send the "\E[?25h\E[?0c" command.
432 * Problems in Emacs built with LessTif.
434 The problems seem to depend on the version of LessTif and the Motif
435 emulation for which it is set up.
437 Only the Motif 1.2 emulation seems to be stable enough in LessTif.
438 Lesstif 0.92-17's Motif 1.2 emulation seems to work okay on FreeBSD.
439 On GNU/Linux systems, lesstif-0.92.6 configured with "./configure
440 --enable-build-12 --enable-default-12" is reported to be the most
441 successful. The binary GNU/Linux package
442 lesstif-devel-0.92.0-1.i386.rpm was reported to have problems with
445 On some systems, even with Motif 1.2 emulation, Emacs occasionally
446 locks up, grabbing all mouse and keyboard events. We still don't know
447 what causes these problems; they are not reproducible by Emacs
450 * Known problems with the MS-Windows port of Emacs 21.2.
452 Frames are not refreshed while the File or Font dialog or a pop-up menu
453 is displayed. This also means help text for pop-up menus is not
454 displayed at all. This is because message handling under Windows is
455 synchronous, so we cannot handle repaint (or any other) messages while
456 waiting for a system function to return the result of the dialog or
457 pop-up menu interaction.
459 Windows 95 and Windows NT up to version 4.0 do not support help text
460 for menus. Help text is only available in later versions of Windows.
462 There are problems with display if mouse-tracking is enabled and the
463 mouse is moved off a frame, over another frame then back over the first
464 frame. A workaround is to click the left mouse button inside the frame
465 after moving back into it.
467 Some minor flickering still persists during mouse-tracking, although
468 not as severely as in 21.1.
470 Emacs can sometimes abort when non-ASCII text, possibly with null
471 characters, is copied and pasted into a buffer.
473 An inactive cursor remains in an active window after the Windows
474 Manager driven switch of the focus, until a key is pressed.
476 Windows input methods are not recognized by Emacs (as of v21.2). Some
477 of these input methods cause the keyboard to send characters encoded
478 in the appropriate coding system (e.g., ISO 8859-1 for Latin-1
479 characters, ISO 8859-8 for Hebrew characters, etc.). To make this
480 work, set the keyboard coding system to the appropriate value after
481 you activate the Windows input method. For example, if you activate
482 the Hebrew input method, type "C-x RET k iso-8859-8 RET". (Emacs
483 ought to recognize the Windows language-change event and set up the
484 appropriate keyboard encoding automatically, but it doesn't do that
487 Windows uses UTF-16 encoding to deal with multilingual text (text not
488 encodable in the `system codepage') in the clipboard. To deal with
489 this, load the library `utf-16' and use `set-selection-coding-system'
490 to set the clipboard coding system to `utf-16-le-dos'. This won't
491 cope with Far Eastern (`CJK') text; if necessary, install the Mule-UCS
492 package (see etc/MORE.STUFF), whose `utf-16-le-dos' coding system does
493 encode a lot of CJK characters.
495 The %b specifier for format-time-string does not produce abbreviated
496 month names with consistent widths for some locales on some versions
497 of Windows. This is caused by a deficiency in the underlying system
500 * The `configure' script doesn't find the jpeg library.
502 There are reports that this happens on some systems because the linker
503 by default only looks for shared libraries, but jpeg distribution by
504 default only installs a nonshared version of the library, `libjpeg.a'.
506 If this is the problem, you can configure the jpeg library with the
507 `--enable-shared' option and then rebuild libjpeg. This produces a
508 shared version of libjpeg, which you need to install. Finally, rerun
509 the Emacs configure script, which should now find the jpeg library.
510 Alternatively, modify the generated src/Makefile to link the .a file
511 explicitly, and edit src/config.h to define HAVE_JPEG.
513 * Building Emacs over NFS fails with ``Text file busy''.
515 This was reported to happen when building Emacs on a GNU/Linux system
516 (RedHat Linux 6.2) using a build directory automounted from Solaris
517 (SunOS 5.6) file server, but it might not be limited to that
518 configuration alone. Presumably, the NFS server doesn't commit the
519 files' data to disk quickly enough, and the Emacs executable file is
520 left ``busy'' for several seconds after Emacs has finished dumping
521 itself. This causes the subsequent commands which invoke the dumped
522 Emacs executable to fail with the above message.
524 In some of these cases, a time skew between the NFS server and the
525 machine where Emacs is built is detected and reported by GNU Make
526 (it says that some of the files have modification time in the future).
527 This might be a symptom of NFS-related problems.
529 If the NFS server runs on Solaris, apply the Solaris patch 105379-05
530 (Sunos 5.6: /kernel/misc/nfssrv patch). If that doesn't work, or if
531 you have a different version of the OS or the NFS server, you can
532 force the NFS server to use 1KB blocks, which was reported to fix the
533 problem albeit at a price of slowing down file I/O. You can force 1KB
534 blocks by specifying the "-o rsize=1024,wsize=1024" options to the
535 `mount' command, or by adding ",rsize=1024,wsize=1024" to the mount
536 options in the appropriate system configuration file, such as
539 Alternatively, when Make fails due to this problem, you could wait for
540 a few seconds and then invoke Make again. In one particular case,
541 waiting for 10 or more seconds between the two Make invocations seemed
542 to work around the problem.
544 Similar problems can happen if your machine NFS-mounts a directory
545 onto itself. Suppose the Emacs sources live in `/usr/local/src' and
546 you are working on the host called `marvin'. Then an entry in the
547 `/etc/fstab' file like the following is asking for trouble:
549 marvin:/usr/local/src /usr/local/src ...options.omitted...
551 The solution is to remove this line from `etc/fstab'.
553 * Emacs binary is not in executable format, and cannot be run.
555 This was reported to happen when Emacs is built in a directory mounted
556 via NFS. Usually, the file `emacs' produced in these cases is full of
557 binary null characters, and the `file' utility says:
559 emacs: ASCII text, with no line terminators
561 We don't know what exactly causes this failure. A work-around is to
562 build Emacs in a directory on a local disk.
564 * Accented ISO-8859-1 characters are displayed as | or _.
566 Try other font set sizes (S-mouse-1). If the problem persists with
567 other sizes as well, your text is corrupted, probably through software
568 that is not 8-bit clean. If the problem goes away with another font
569 size, it's probably because some fonts pretend to be ISO-8859-1 fonts
570 when they are really ASCII fonts. In particular the schumacher-clean
571 fonts have this bug in some versions of X.
573 To see what glyphs are included in a font, use `xfd', like this:
575 xfd -fn -schumacher-clean-medium-r-normal--12-120-75-75-c-60-iso8859-1
577 If this shows only ASCII glyphs, the font is indeed the source of the
580 The solution is to remove the corresponding lines from the appropriate
581 `fonts.alias' file, then run `mkfontdir' in that directory, and then run
584 * Large file support is disabled on HP-UX. See the comments in
587 * Crashes when displaying GIF images in Emacs built with version
588 libungif-4.1.0 are resolved by using version libungif-4.1.0b1.
589 Configure checks for the correct version, but this problem could occur
590 if a binary built against a shared libungif is run on a system with an
593 * Font Lock displays portions of the buffer in incorrect faces.
595 By far the most frequent cause of this is a parenthesis `(' or a brace
596 `{' in column zero. Font Lock assumes that such a paren is outside of
597 any comment or string. This is of course not true in general, but the
598 vast majority of well-formatted program source files don't have such
599 parens, and therefore this assumption is used to allow optimizations
600 in Font Lock's syntactical analysis. These optimizations avoid some
601 pathological cases where jit-lock, the Just-in-Time fontification
602 introduced with Emacs 21.1, could significantly slow down scrolling
603 through the buffer, especially scrolling backwards, and also jumping
604 to the end of a very large buffer.
606 Beginning with version 21.3, a parenthesis or a brace in column zero
607 is highlighted in bold-red face if it is inside a string or a comment,
608 to indicate that it could interfere with Font Lock (and also with
609 indentation) and should be moved or escaped with a backslash.
611 If you don't use large buffers, or have a very fast machine which
612 makes the delays insignificant, you can avoid the incorrect
613 fontification by setting the variable
614 `font-lock-beginning-of-syntax-function' to a nil value. (This must
615 be done _after_ turning on Font Lock.)
617 Another alternative is to avoid a paren in column zero. For example,
618 in a Lisp string you could precede the paren with a backslash.
620 * When running on KDE, colors or fonts are not as specified for Emacs,
623 For example, you could see background you set for Emacs only in the
624 empty portions of the Emacs display, while characters have some other
627 This happens because KDE's defaults apply its color and font
628 definitions even to applications that weren't compiled for KDE. The
629 solution is to uncheck the "Apply fonts and colors to non-KDE apps"
630 option in Preferences->Look&Feel->Style.
632 Alternatively, if you do want the KDE defaults to apply to other
633 applications, but not to Emacs, you could modify the file `Emacs.ad'
634 (should be in the `/usr/share/apps/kdisplay/app-defaults/' directory)
635 so that it doesn't set the default background and foreground only for
636 Emacs. For example, make sure the following resources are either not
637 present or commented out:
639 Emacs.default.attributeForeground
640 Emacs.default.attributeBackground
644 * Interrupting Cygwin port of Bash from Emacs doesn't work.
646 Cygwin 1.x builds of the ported Bash cannot be interrupted from the
647 MS-Windows version of Emacs. This is due to some change in the Bash
648 port or in the Cygwin library which apparently make Bash ignore the
649 keyboard interrupt event sent by Emacs to Bash. (Older Cygwin ports
650 of Bash, up to b20.1, did receive SIGINT from Emacs.)
652 * Dired is very slow.
654 This could happen if invocation of the `df' program takes a long
655 time. Possible reasons for this include:
657 - ClearCase mounted filesystems (VOBs) that sometimes make `df'
658 response time extremely slow (dozens of seconds);
660 - slow automounters on some old versions of Unix;
662 - slow operation of some versions of `df'.
664 To work around the problem, you could either (a) set the variable
665 `directory-free-space-program' to nil, and thus prevent Emacs from
666 invoking `df'; (b) use `df' from the GNU Fileutils package; or
667 (c) use CVS, which is Free Software, instead of ClearCase.
669 * Accessing remote files with ange-ftp hangs the MS-Windows version of Emacs.
671 If the FTP client is the Cygwin port of GNU `ftp', this appears to be
672 due to some bug in the Cygwin DLL or some incompatibility between it
673 and the implementation of asynchronous subprocesses in the Windows
674 port of Emacs. Specifically, some parts of the FTP server responses
675 are not flushed out, apparently due to buffering issues, which
678 The solution is to downgrade to an older version of the Cygwin DLL
679 (version 1.3.2 was reported to solve the problem), or use the stock
680 Windows FTP client, usually found in the `C:\WINDOWS' or 'C:\WINNT'
681 directory. To force ange-ftp use the stock Windows client, set the
682 variable `ange-ftp-ftp-program-name' to the absolute file name of the
683 client's executable. For example:
685 (setq ange-ftp-ftp-program-name "c:/windows/ftp.exe")
687 If you want to stick with the Cygwin FTP client, you can work around
688 this problem by putting this in your `.emacs' file:
690 (setq ange-ftp-ftp-program-args '("-i" "-n" "-g" "-v" "--prompt" "")
692 * Versions of the W3 package released before Emacs 21.1 don't run
693 under Emacs 21. This fixed in W3 version 4.0pre.47.
695 * On AIX, if linking fails because libXbsd isn't found, check if you
696 are compiling with the system's `cc' and CFLAGS containing `-O5'. If
697 so, you have hit a compiler bug. Please make sure to re-configure
698 Emacs so that it isn't compiled with `-O5'.
700 * Compiling on AIX 4.3.x or 4.4 fails.
702 This could happen if you use /bin/c89 as your compiler, instead of
703 the default `cc'. /bin/c89 treats certain warnings, such as benign
704 redefinitions of macros, as errors, and fails the build. A solution
705 is to use the default compiler `cc'.
707 * Old versions of the PSGML package use the obsolete variables
708 `before-change-function' and `after-change-function', which are no
709 longer used by Emacs. Please use PSGML 1.2.3 or later.
711 * PSGML conflicts with sgml-mode.
713 PSGML package uses the same names of some variables (like keymap)
714 as built-in sgml-mode.el because it was created as a replacement
715 of that package. The conflict will be shown if you load
716 sgml-mode.el before psgml.el. E.g. this could happen if you edit
717 HTML page and then start to work with SGML or XML file. html-mode
718 (from sgml-mode.el) is used for HTML file and loading of psgml.el
719 (for sgml-mode or xml-mode) will cause an error.
721 * The LDAP support rely on ldapsearch program from OpenLDAP version 2.
723 It can fail to work with ldapsearch program from OpenLDAP version 1.
724 Version 1 of OpenLDAP is now deprecated. If you are still using it,
725 please upgrade to version 2. As a temporary workaround, remove
726 argument "-x" from the variable `ldap-ldapsearch-args'.
728 * The `oc-unicode' package doesn't work with Emacs 21.
730 This package tries to define more private charsets than there are free
731 slots now. If the built-in Unicode/UTF-8 support is insufficient,
732 e.g. if you need more CJK coverage, use the current Mule-UCS package.
733 Any files encoded as emacs-mule using oc-unicode won't be read
734 correctly by Emacs 21.
736 * Using epop3.el package causes Emacs to signal an error.
738 The error message might be something like this:
740 "Lisp nesting exceeds max-lisp-eval-depth"
742 This happens because epop3 redefines the function gethash, which is a
743 built-in primitive beginning with Emacs 21.1. We don't have a patch
744 for epop3 that fixes this, but perhaps a newer version of epop3
747 * ps-print commands fail to find prologue files ps-prin*.ps.
749 This can happen if you use an old version of X-Symbol package: it
750 defines compatibility functions which trick ps-print into thinking it
751 runs in XEmacs, and look for the prologue files in a wrong directory.
753 The solution is to upgrade X-Symbol to a later version.
755 * On systems with shared libraries you might encounter run-time errors
756 from the dynamic linker telling you that it is unable to find some
757 shared libraries, for instance those for Xaw3d or image support.
758 These errors mean Emacs has been linked with a library whose shared
759 library is not in the default search path of the dynamic linker.
761 Similar problems could prevent Emacs from building, since the build
762 process invokes Emacs several times.
764 On many systems, it is possible to set LD_LIBRARY_PATH in your
765 environment to specify additional directories where shared libraries
768 Other systems allow to set LD_RUN_PATH in a similar way, but before
769 Emacs is linked. With LD_RUN_PATH set, the linker will include a
770 specified run-time search path in the executable.
772 On some systems, Emacs can crash due to problems with dynamic
773 linking. Specifically, on SGI Irix 6.5, crashes were reported with
774 backtraces like this:
777 0 strcmp(0xf49239d, 0x4031184, 0x40302b4, 0x12, 0xf0000000, 0xf4923aa, 0x0, 0x492ddb2) ["/xlv22/ficus-jan23/work/irix/lib/libc/libc_n32_M3_ns/strings/strcmp.s":35, 0xfb7e480]
778 1 general_find_symbol(0xf49239d, 0x0, 0x0, 0x0, 0xf0000000, 0xf4923aa, 0x0, 0x492ddb2)
779 ["/comp2/mtibuild/v73/workarea/v7.3/rld/rld.c":2140, 0xfb65a98]
780 2 resolve_symbol(0xf49239d, 0x4031184, 0x0, 0xfbdd438, 0x0, 0xf4923aa, 0x0, 0x492ddb2)
781 ["/comp2/mtibuild/v73/workarea/v7.3/rld/rld.c":1947, 0xfb657e4]
782 3 lazy_text_resolve(0xd18, 0x1a3, 0x40302b4, 0x12, 0xf0000000, 0xf4923aa, 0x0, 0x492ddb2)
783 ["/comp2/mtibuild/v73/workarea/v7.3/rld/rld.c":997, 0xfb64d44]
784 4 _rld_text_resolve(0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0)
785 ["/comp2/mtibuild/v73/workarea/v7.3/rld/rld_bridge.s":175, 0xfb6032c]
787 (`rld' is the dynamic linker.) We don't know yet why this
788 happens, but setting the environment variable LD_BIND_NOW to 1 (which
789 forces the dynamic linker to bind all shared objects early on) seems
790 to work around the problem.
792 Please refer to the documentation of your dynamic linker for details.
794 * On Solaris 2.7, building Emacs with WorkShop Compilers 5.0 98/12/15
795 C 5.0 failed, apparently with non-default CFLAGS, most probably due to
796 compiler bugs. Using Sun Solaris 2.7 Sun WorkShop 6 update 1 C
797 release was reported to work without problems. It worked OK on
798 another system with Solaris 8 using apparently the same 5.0 compiler
799 and the default CFLAGS.
801 * Compiling syntax.c with the OPENSTEP 4.2 compiler gcc 2.7.2.1 fails.
803 The compiler was reported to crash while compiling syntax.c with the
806 cc: Internal compiler error: program cc1obj got fatal signal 11
808 To work around this, replace the macros UPDATE_SYNTAX_TABLE_FORWARD,
809 INC_BOTH, and INC_FROM with functions. To this end, first define 3
810 functions, one each for every macro. Here's an example:
812 static int update_syntax_table_forward(int from)
814 return(UPDATE_SYNTAX_TABLE_FORWARD(from));
815 }/*update_syntax_table_forward*/
817 Then replace all references to UPDATE_SYNTAX_TABLE_FORWARD in syntax.c
818 with a call to the function update_syntax_table_forward.
820 * Emacs fails to start, complaining about missing fonts.
822 A typical error message might be something like
824 No fonts match `-*-fixed-medium-r-*--6-*-*-*-*-*-iso8859-1'
826 This happens because some X resource specifies a bad font family for
827 Emacs to use. The possible places where this specification might be
830 - in your ~/.Xdefaults file
832 - client-side X resource file, such as ~/Emacs or
833 /usr/X11R6/lib/app-defaults/Emacs or
834 /usr/X11R6/lib/X11/app-defaults/Emacs
836 One of these files might have bad or malformed specification of a
837 fontset that Emacs should use. To fix the problem, you need to find
838 the problematic line(s) and correct them.
840 * Emacs 20 and later fails to load Lisp files at startup.
842 The typical error message might be like this:
844 "Cannot open load file: fontset"
846 This could happen if you compress the file lisp/subdirs.el. That file
847 tells Emacs what are the directories where it should look for Lisp
848 files. Emacs cannot work with subdirs.el compressed, since the
849 Auto-compress mode it needs for this will not be loaded until later,
850 when your .emacs file is processed. (The package `fontset.el' is
851 required to set up fonts used to display text on window systems, and
852 it's loaded very early in the startup procedure.)
854 Similarly, any other .el file for which there's no corresponding .elc
855 file could fail to load if it is compressed.
857 The solution is to uncompress all .el files which don't have a .elc
860 Another possible reason for such failures is stale *.elc files
861 lurking somewhere on your load-path. The following command will
862 print any duplicate Lisp files that are present in load-path:
864 emacs -q -batch -f list-load-path-shadows
866 If this command prints any file names, some of these files are stale,
867 and should be deleted or their directories removed from your
870 * Emacs prints an error at startup after upgrading from an earlier version.
872 An example of such an error is:
874 x-complement-fontset-spec: "Wrong type argument: stringp, nil"
876 This can be another symptom of stale *.elc files in your classpath.
877 The following command will print any duplicate Lisp files that are
878 present in load-path:
880 emacs -q -batch -f list-load-path-shadows
882 If this command prints any file names, some of these files are stale,
883 and should be deleted or their directories removed from your
886 * Attempting to visit remote files via ange-ftp fails.
888 If the error message is "ange-ftp-file-modtime: Specified time is not
889 representable", then this could happen when `lukemftp' is used as the
890 ftp client. This was reported to happen on Debian GNU/Linux, kernel
891 version 2.4.3, with `lukemftp' 1.5-5, but might happen on other
892 systems as well. To avoid this problem, switch to using the standard
893 ftp client. On a Debian system, type
895 update-alternatives --config ftp
897 and then choose /usr/bin/netkit-ftp.
899 * Emacs built on Windows 9x/ME crashes at startup on Windows XP,
900 or Emacs built on XP crashes at startup on Windows 9x/ME.
902 There appear to be general problems running programs compiled on
903 Windows 9x/ME on Windows XP and vice-versa, at least when compilation
904 is done with MSVC 6.0. This affects other programs as well as Emacs.
905 The compatibility options in the program properties on Windows XP may
908 * Antivirus software interacts badly with the MS-Windows version of Emacs.
910 The usual manifestation of these problems is that subprocesses don't
911 work or even wedge the entire system. In particular, "M-x shell RET"
912 was reported to fail to work. But other commands also sometimes don't
913 work when an antivirus package is installed.
915 The solution is to switch the antivirus software to a less aggressive
916 mode (e.g., disable the ``auto-protect'' feature), or even uninstall
917 or disable it entirely.
919 * On Windows 95/98/ME, subprocesses do not terminate properly.
921 This is a limitation of the Operating System, and can cause problems
922 when shutting down Windows. Ensure that all subprocesses are exited
923 cleanly before exiting Emacs. For more details, see the FAQ at
924 http://www.gnu.org/software/emacs/windows/.
926 * Windows 95/98/ME crashes when Emacs invokes non-existent programs.
928 When a program you are trying to run is not found on the PATH,
929 Windows might respond by crashing or locking up your system. In
930 particular, this has been reported when trying to compile a Java
931 program in JDEE when javac.exe is installed, but not on the system
934 * Mail sent through Microsoft Exchange in some encodings appears to be
935 mangled and is not seen correctly in Rmail or Gnus. We don't know
936 exactly what happens, but it isn't an Emacs problem in cases we've
939 * After upgrading to a newer version of Emacs, the Meta key stops working.
941 This was reported to happen on a GNU/Linux system distributed by
942 Mandrake. The reason is that the previous version of Emacs was
943 modified by Mandrake to make the Alt key act as the Meta key, on a
944 keyboard where the Windows key is the one which produces the Meta
945 modifier. A user who started using a newer version of Emacs, which
946 was not hacked by Mandrake, expected the Alt key to continue to act as
947 Meta, and was astonished when that didn't happen.
949 The solution is to find out what key on your keyboard produces the Meta
950 modifier, and use that key instead. Try all of the keys to the left
951 and to the right of the space bar, together with the `x' key, and see
952 which combination produces "M-x" in the echo area. You can also use
953 the `xmodmap' utility to show all the keys which produce a Meta
956 xmodmap -pk | egrep -i "meta|alt"
958 A more convenient way of finding out which keys produce a Meta modifier
959 is to use the `xkbprint' utility, if it's available on your system:
961 xkbprint 0:0 /tmp/k.ps
963 This produces a PostScript file `/tmp/k.ps' with a picture of your
964 keyboard; printing that file on a PostScript printer will show what
965 keys can serve as Meta.
967 The `xkeycaps' also shows a visual representation of the current
968 keyboard settings. It also allows to modify them.
970 * On OSF/Dec Unix/Tru64/<whatever it is this year> under X locally or
971 remotely, M-SPC acts as a `compose' key with strange results. See
974 Changing Alt_L to Meta_L fixes it:
975 % xmodmap -e 'keysym Alt_L = Meta_L Alt_L'
976 % xmodmap -e 'keysym Alt_R = Meta_R Alt_R'
978 * Error "conflicting types for `initstate'" compiling with GCC on Irix 6.
980 Install GCC 2.95 or a newer version, and this problem should go away.
981 It is possible that this problem results from upgrading the operating
982 system without reinstalling GCC; so you could also try reinstalling
983 the same version of GCC, and telling us whether that fixes the problem.
985 * Emacs dumps core on Solaris in function IMCheckWindow.
987 This was reported to happen when Emacs runs with more than one frame,
988 and one of them is closed, either with "C-x 5 0" or from the window
991 This bug was reported to Sun as
993 Gtk apps dump core in ximlocal.so.2:IMCheckIMWindow()
996 Installing Solaris 8 patch 108773-12 for Sparc and 108774-12 for x86
997 reportedly fixes the bug, which appears to be inside the shared
1000 Alternatively, you can configure Emacs with `--with-xim=no' to prevent
1001 the core dump, but will loose X input method support, of course. (You
1002 can use Emacs's own input methods instead, if you install Leim.)
1004 * On Solaris 7, Emacs gets a segmentation fault when starting up using X.
1006 This results from Sun patch 107058-01 (SunOS 5.7: Patch for
1007 assembler) if you use GCC version 2.7 or later.
1008 To work around it, either install patch 106950-03 or later,
1009 or uninstall patch 107058-01, or install the GNU Binutils.
1010 Then recompile Emacs, and it should work.
1012 * With X11R6.4, public-patch-3, Emacs crashes at startup.
1014 Reportedly this patch in X fixes the problem.
1016 --- xc/lib/X11/imInt.c~ Wed Jun 30 13:31:56 1999
1017 +++ xc/lib/X11/imInt.c Thu Jul 1 15:10:27 1999
1019 -/* $TOG: imInt.c /main/5 1998/05/30 21:11:16 kaleb $ */
1020 +/* $TOG: imInt.c /main/5 1998/05/30 21:11:16 kaleb $ */
1021 /******************************************************************
1023 Copyright 1992, 1993, 1994 by FUJITSU LIMITED
1030 + char* begin = NULL;
1034 char* ximmodifier = XIMMODIFIER;
1035 @@ -182,7 +182,11 @@
1037 ret = Xmalloc(end - begin + 2);
1039 - (void)strncpy(ret, begin, end - begin + 1);
1040 + if (begin != NULL) {
1041 + (void)strncpy(ret, begin, end - begin + 1);
1045 ret[end - begin + 1] = '\0';
1050 * Emacs crashes on Irix 6.5 on the SGI R10K, when compiled with GCC.
1052 This seems to be fixed in GCC 2.95.
1054 * Emacs crashes in utmpname on Irix 5.3.
1056 This problem is fixed in Patch 3175 for Irix 5.3.
1057 It is also fixed in Irix versions 6.2 and up.
1059 * The S-C-t key combination doesn't get passed to Emacs on X.
1061 This happens because some X configurations assign the Ctrl-Shift-t
1062 combination the same meaning as the Multi_key. The offending
1063 definition is in the file `...lib/X11/locale/iso8859-1/Compose'; there
1064 might be other similar combinations which are grabbed by X for similar
1067 We think that this can be countermanded with the `xmodmap' utility, if
1068 you want to be able to bind one of these key sequences within Emacs.
1070 * On Solaris, CTRL-t is ignored by Emacs when you use
1071 the fr.ISO-8859-15 locale (and maybe other related locales).
1073 You can fix this by editing the file:
1075 /usr/openwin/lib/locale/iso8859-15/Compose
1077 Near the bottom there is a line that reads:
1079 Ctrl<t> <quotedbl> <Y> : "\276" threequarters
1083 Ctrl<T> <quotedbl> <Y> : "\276" threequarters
1085 Note the lower case <t>. Changing this line should make C-t work.
1087 * Emacs on Digital Unix 4.0 fails to build, giving error message
1088 Invalid dimension for the charset-ID 160
1090 This is due to a bug or an installation problem in GCC 2.8.0.
1091 Installing a more recent version of GCC fixes the problem.
1093 * Buffers from `with-output-to-temp-buffer' get set up in Help mode.
1095 Changes in Emacs 20.4 to the hooks used by that function cause
1096 problems for some packages, specifically BBDB. See the function's
1097 documentation for the hooks involved. BBDB 2.00.06 fixes the problem.
1099 * Under X, C-v and/or other keys don't work.
1101 These may have been intercepted by your window manager. In
1102 particular, AfterStep 1.6 is reported to steal C-v in its default
1103 configuration. Various Meta keys are also likely to be taken by the
1104 configuration of the `feel'. See the WM's documentation for how to
1107 * When using Exceed, fonts sometimes appear too tall.
1109 When the display is set to an Exceed X-server and fonts are specified
1110 (either explicitly with the -fn option or implicitly with X resources)
1111 then the fonts may appear "too tall". The actual character sizes are
1112 correct but there is too much vertical spacing between rows, which
1113 gives the appearance of "double spacing".
1115 To prevent this, turn off the Exceed's "automatic font substitution"
1116 feature (in the font part of the configuration window).
1118 * Failure in unexec while dumping emacs on Digital Unix 4.0
1120 This problem manifests itself as an error message
1122 unexec: Bad address, writing data section to ...
1124 The user suspects that this happened because his X libraries
1125 were built for an older system version,
1127 ./configure --x-includes=/usr/include --x-libraries=/usr/shlib
1129 made the problem go away.
1131 * No visible display on mips-sgi-irix6.2 when compiling with GCC 2.8.1.
1133 This problem went away after installing the latest IRIX patches
1136 The same problem has been reported on Irix 6.3.
1138 * As of version 20.4, Emacs doesn't work properly if configured for
1139 the Motif toolkit and linked against the free LessTif library. The
1140 next Emacs release is expected to work with LessTif.
1142 * Emacs gives the error, Couldn't find per display information.
1144 This can result if the X server runs out of memory because Emacs uses
1145 a large number of fonts. On systems where this happens, C-h h is
1148 We do not know of a way to prevent the problem.
1150 * Emacs makes HPUX 11.0 crash.
1152 This is a bug in HPUX; HPUX patch PHKL_16260 is said to fix it.
1154 * Emacs crashes during dumping on the HPPA machine (HPUX 10.20).
1156 This seems to be due to a GCC bug; it is fixed in GCC 2.8.1.
1158 * The Hyperbole package causes *Help* buffers not to be displayed in
1159 Help mode due to setting `temp-buffer-show-hook' rather than using
1160 `add-hook'. Using `(add-hook 'temp-buffer-show-hook
1161 'help-mode-maybe)' after loading Hyperbole should fix this.
1163 * Versions of the PSGML package earlier than 1.0.3 (stable) or 1.1.2
1164 (alpha) fail to parse DTD files correctly in Emacs 20.3 and later.
1165 Here is a patch for psgml-parse.el from PSGML 1.0.1 and, probably,
1168 --- psgml-parse.el 1998/08/21 19:18:18 1.1
1169 +++ psgml-parse.el 1998/08/21 19:20:00
1170 @@ -2383,7 +2383,7 @@ (defun sgml-push-to-entity (entity &opti
1171 (setq sgml-buffer-parse-state nil))
1173 ((stringp entity) ; a file name
1174 - (save-excursion (insert-file-contents entity))
1175 + (insert-file-contents entity)
1176 (setq default-directory (file-name-directory entity)))
1177 ((consp (sgml-entity-text entity)) ; external id?
1178 (let* ((extid (sgml-entity-text entity))
1180 * Emacs 21 freezes when visiting a TeX file with AUC TeX installed.
1182 Emacs 21 needs version 10 or later of AUC TeX; upgrading should solve
1185 * No colors in AUC TeX with Emacs 21.
1187 Upgrade to AUC TeX version 10 or later, and make sure it is
1188 byte-compiled with Emacs 21.
1190 * Running TeX from AUC TeX package with Emacs 20.3 gives a Lisp error
1191 about a read-only tex output buffer.
1193 This problem appeared for AUC TeX version 9.9j and some earlier
1194 versions. Here is a patch for the file tex-buf.el in the AUC TeX
1197 diff -c auctex/tex-buf.el~ auctex/tex-buf.el
1198 *** auctex/tex-buf.el~ Wed Jul 29 18:35:32 1998
1199 --- auctex/tex-buf.el Sat Sep 5 15:20:38 1998
1202 (dir (TeX-master-directory)))
1203 (TeX-process-check file) ; Check that no process is running
1204 (setq TeX-command-buffer (current-buffer))
1205 ! (with-output-to-temp-buffer buffer)
1208 (insert "Running `" name "' on `" file "' with ``" command "''\n")
1210 (dir (TeX-master-directory)))
1211 (TeX-process-check file) ; Check that no process is running
1212 (setq TeX-command-buffer (current-buffer))
1213 ! (let (temp-buffer-show-function temp-buffer-show-hook)
1214 ! (with-output-to-temp-buffer buffer))
1217 (insert "Running `" name "' on `" file "' with ``" command "''\n")
1219 * On Irix 6.3, substituting environment variables in file names
1220 in the minibuffer gives peculiar error messages such as
1222 Substituting nonexistent environment variable ""
1224 This is not an Emacs bug; it is caused by something in SGI patch
1225 003082 August 11, 1998.
1227 * After a while, Emacs slips into unibyte mode.
1229 The VM mail package, which is not part of Emacs, sometimes does
1230 (standard-display-european t)
1231 That should be changed to
1232 (standard-display-european 1 t)
1234 * Installing Emacs gets an error running `install-info'.
1236 You need to install a recent version of Texinfo; that package
1237 supplies the `install-info' command.
1239 * Emacs does not recognize the AltGr key, on HPUX.
1241 To fix this, set up a file ~/.dt/sessions/sessionetc with executable
1242 rights, containing this text:
1244 --------------------------------
1245 xmodmap 2> /dev/null - << EOF
1246 keysym Alt_L = Meta_L
1247 keysym Alt_R = Meta_R
1252 keysym Mode_switch = NoSymbol
1254 keysym Meta_R = Mode_switch
1255 add mod2 = Mode_switch
1257 --------------------------------
1259 * Emacs hangs on KDE when a large portion of text is killed.
1261 This is caused by a bug in the KDE applet `klipper' which periodically
1262 requests the X clipboard contents from applications. Early versions
1263 of klipper don't implement the ICCM protocol for large selections,
1264 which leads to Emacs being flooded with selection requests. After a
1265 while, Emacs will print a message:
1267 Timed out waiting for property-notify event
1269 A workaround is to not use `klipper'.
1271 * Emacs compiled with DJGPP for MS-DOS/MS-Windows cannot access files
1272 in the directory with the special name `dev' under the root of any
1273 drive, e.g. `c:/dev'.
1275 This is an unfortunate side-effect of the support for Unix-style
1276 device names such as /dev/null in the DJGPP runtime library. A
1277 work-around is to rename the problem directory to another name.
1279 * M-SPC seems to be ignored as input.
1281 See if your X server is set up to use this as a command
1282 for character composition.
1284 * Emacs startup on GNU/Linux systems (and possibly other systems) is slow.
1286 This can happen if the system is misconfigured and Emacs can't get the
1287 full qualified domain name, FQDN. You should have your FQDN in the
1288 /etc/hosts file, something like this:
1291 129.187.137.82 nuc04.t30.physik.tu-muenchen.de nuc04
1293 The way to set this up may vary on non-GNU systems.
1295 * Garbled display on non-X terminals when Emacs runs on Digital Unix 4.0.
1297 So far it appears that running `tset' triggers this problem (when TERM
1298 is vt100, at least). If you do not run `tset', then Emacs displays
1299 properly. If someone can tell us precisely which effect of running
1300 `tset' actually causes the problem, we may be able to implement a fix
1303 * When you run Ispell from Emacs, it reports a "misalignment" error.
1305 This can happen if you compiled the Ispell program to use ASCII
1306 characters only and then try to use it from Emacs with non-ASCII
1307 characters, like Latin-1. The solution is to recompile Ispell with
1308 support for 8-bit characters.
1310 To see whether your Ispell program supports 8-bit characters, type
1311 this at your shell's prompt:
1315 and look in the output for the string "NO8BIT". If Ispell says
1316 "!NO8BIT (8BIT)", your speller supports 8-bit characters; otherwise it
1319 To rebuild Ispell with 8-bit character support, edit the local.h file
1320 in the Ispell distribution and make sure it does _not_ define NO8BIT.
1321 Then rebuild the speller.
1323 Another possible cause for "misalignment" error messages is that the
1324 version of Ispell installed on your machine is old. Upgrade.
1326 Yet another possibility is that you are trying to spell-check a word
1327 in a language that doesn't fit the dictionary you choose for use by
1328 Ispell. (Ispell can only spell-check one language at a time, because
1329 it uses a single dictionary.) Make sure that the text you are
1330 spelling and the dictionary used by Ispell conform to each other.
1332 If your spell-checking program is Aspell, it has been reported that if
1333 you have a personal configuration file (normally ~/.aspell.conf), it
1334 can cause this error. Remove that file, execute `ispell-kill-ispell'
1335 in Emacs, and then try spell-checking again.
1337 * On Linux-based GNU systems using libc versions 5.4.19 through
1338 5.4.22, Emacs crashes at startup with a segmentation fault.
1340 This problem happens if libc defines the symbol __malloc_initialized.
1341 One known solution is to upgrade to a newer libc version. 5.4.33 is
1344 * On Windows, you cannot use the right-hand ALT key and the left-hand
1345 CTRL key together to type a Control-Meta character.
1347 This is a consequence of a misfeature beyond Emacs's control.
1349 Under Windows, the AltGr key on international keyboards generates key
1350 events with the modifiers Right-Alt and Left-Ctrl. Since Emacs cannot
1351 distinguish AltGr from an explicit Right-Alt and Left-Ctrl
1352 combination, whenever it sees Right-Alt and Left-Ctrl it assumes that
1353 AltGr has been pressed. The variable `w32-recognize-altgr' can be set
1354 to nil to tell Emacs that AltGr is really Ctrl and Alt.
1356 * Emacs crashes when using the Exceed 6.0 X server
1358 If you are using Exceed 6.1, upgrade to a later version. This was
1359 reported to prevent the crashes.
1361 * Under some Windows X-servers, Emacs' display is incorrect
1363 The symptoms are that Emacs does not completely erase blank areas of the
1364 screen during scrolling or some other screen operations (e.g., selective
1365 display or when killing a region). M-x recenter will cause the screen
1366 to be completely redisplayed and the "extra" characters will disappear.
1368 This is known to occur under Exceed 6, and possibly earlier versions
1369 as well; it is reportedly solved in version 6.2.0.16 and later. The
1370 problem lies in the X-server settings.
1372 There are reports that you can solve the problem with Exceed by
1373 running `Xconfig' from within NT, choosing "X selection", then
1374 un-checking the boxes "auto-copy X selection" and "auto-paste to X
1377 Of this does not work, please inform bug-gnu-emacs@gnu.org. Then
1378 please call support for your X-server and see if you can get a fix.
1379 If you do, please send it to bug-gnu-emacs@gnu.org so we can list it
1382 * On Solaris 2, Emacs dumps core when built with Motif.
1384 The Solaris Motif libraries are buggy, at least up through Solaris 2.5.1.
1385 Install the current Motif runtime library patch appropriate for your host.
1386 (Make sure the patch is current; some older patch versions still have the bug.)
1387 You should install the other patches recommended by Sun for your host, too.
1388 You can obtain Sun patches from ftp://sunsolve.sun.com/pub/patches/;
1389 look for files with names ending in `.PatchReport' to see which patches
1390 are currently recommended for your host.
1392 On Solaris 2.6, Emacs is said to work with Motif when Solaris patch
1393 105284-12 is installed, but fail when 105284-15 is installed.
1394 105284-18 might fix it again.
1396 * On Solaris 2.6 and 7, the Compose key does not work.
1398 This is a bug in Motif in Solaris. Supposedly it has been fixed for
1399 the next major release of Solaris. However, if someone with Sun
1400 support complains to Sun about the bug, they may release a patch.
1401 If you do this, mention Sun bug #4188711.
1403 One workaround is to use a locale that allows non-ASCII characters.
1404 For example, before invoking emacs, set the LC_ALL environment
1405 variable to "en_US" (American English). The directory /usr/lib/locale
1406 lists the supported locales; any locale other than "C" or "POSIX"
1409 pen@lysator.liu.se says (Feb 1998) that the Compose key does work
1410 if you link with the MIT X11 libraries instead of the Solaris X11
1413 * Frames may cover dialogs they created when using CDE.
1415 This can happen if you have "Allow Primary Windows On Top" enabled which
1416 seems to be the default in the Common Desktop Environment.
1417 To change, go in to "Desktop Controls" -> "Window Style Manager"
1418 and uncheck "Allow Primary Windows On Top".
1420 * Emacs does not know your host's fully-qualified domain name.
1422 You need to configure your machine with a fully qualified domain name,
1423 either in /etc/hosts, /etc/hostname, the NIS, or wherever your system
1424 calls for specifying this.
1426 If you cannot fix the configuration, you can set the Lisp variable
1427 mail-host-address to the value you want.
1429 * Error 12 (virtual memory exceeded) when dumping Emacs, on UnixWare 2.1
1431 Paul Abrahams (abrahams@acm.org) reports that with the installed
1432 virtual memory settings for UnixWare 2.1.2, an Error 12 occurs during
1433 the "make" that builds Emacs, when running temacs to dump emacs. That
1434 error indicates that the per-process virtual memory limit has been
1435 exceeded. The default limit is probably 32MB. Raising the virtual
1436 memory limit to 40MB should make it possible to finish building Emacs.
1438 You can do this with the command `ulimit' (sh) or `limit' (csh).
1439 But you have to be root to do it.
1441 According to Martin Sohnius, you can also retune this in the kernel:
1443 # /etc/conf/bin/idtune SDATLIM 33554432 ## soft data size limit
1444 # /etc/conf/bin/idtune HDATLIM 33554432 ## hard "
1445 # /etc/conf/bin/idtune SVMMSIZE unlimited ## soft process size limit
1446 # /etc/conf/bin/idtune HVMMSIZE unlimited ## hard "
1447 # /etc/conf/bin/idbuild -B
1449 (He recommends you not change the stack limit, though.)
1450 These changes take effect when you reboot.
1452 * Redisplay using X11 is much slower than previous Emacs versions.
1454 We've noticed that certain X servers draw the text much slower when
1455 scroll bars are on the left. We don't know why this happens. If this
1456 happens to you, you can work around it by putting the scroll bars
1457 on the right (as they were in Emacs 19).
1459 Here's how to do this:
1461 (set-scroll-bar-mode 'right)
1463 If you're not sure whether (or how much) this problem affects you,
1464 try that and see how much difference it makes. To set things back
1467 (set-scroll-bar-mode 'left)
1469 * Under X11, some characters appear as hollow boxes.
1471 Each X11 font covers just a fraction of the characters that Emacs
1472 supports. To display the whole range of Emacs characters requires
1473 many different fonts, collected into a fontset.
1475 If some of the fonts called for in your fontset do not exist on your X
1476 server, then the characters that have no font appear as hollow boxes.
1477 You can remedy the problem by installing additional fonts.
1479 The intlfonts distribution includes a full spectrum of fonts that can
1480 display all the characters Emacs supports.
1482 Another cause of this for specific characters is fonts which have a
1483 missing glyph and no default character. This is known ot occur for
1484 character number 160 (no-break space) in some fonts, such as Lucida
1485 but Emacs sets the display table for the unibyte and Latin-1 version
1486 of this character to display a space.
1488 * Under X11, some characters appear improperly aligned in their lines.
1490 You may have bad X11 fonts; try installing the intlfonts distribution.
1492 * Certain fonts make each line take one pixel more than it "should".
1494 This is because these fonts contain characters a little taller
1495 than the font's nominal height. Emacs needs to make sure that
1496 lines do not overlap.
1498 * You request inverse video, and the first Emacs frame is in inverse
1499 video, but later frames are not in inverse video.
1501 This can happen if you have an old version of the custom library in
1502 your search path for Lisp packages. Use M-x list-load-path-shadows to
1503 check whether this is true. If it is, delete the old custom library.
1505 * In FreeBSD 2.1.5, useless symbolic links remain in /tmp or other
1506 directories that have the +t bit.
1508 This is because of a kernel bug in FreeBSD 2.1.5 (fixed in 2.2).
1509 Emacs uses symbolic links to implement file locks. In a directory
1510 with +t bit, the directory owner becomes the owner of the symbolic
1511 link, so that it cannot be removed by anyone else.
1513 If you don't like those useless links, you can let Emacs not to using
1514 file lock by adding #undef CLASH_DETECTION to config.h.
1516 * When using M-x dbx with the SparcWorks debugger, the `up' and `down'
1517 commands do not move the arrow in Emacs.
1519 You can fix this by adding the following line to `~/.dbxinit':
1521 dbxenv output_short_file_name off
1523 * Emacs says it has saved a file, but the file does not actually
1526 This can happen on certain systems when you are using NFS, if the
1527 remote disk is full. It is due to a bug in NFS (or certain NFS
1528 implementations), and there is apparently nothing Emacs can do to
1529 detect the problem. Emacs checks the failure codes of all the system
1530 calls involved in writing a file, including `close'; but in the case
1531 where the problem occurs, none of those system calls fails.
1533 * "Compose Character" key does strange things when used as a Meta key.
1535 If you define one key to serve as both Meta and Compose Character, you
1536 will get strange results. In previous Emacs versions, this "worked"
1537 in that the key acted as Meta--that's because the older Emacs versions
1538 did not try to support Compose Character. Now Emacs tries to do
1539 character composition in the standard X way. This means that you
1540 must pick one meaning or the other for any given key.
1542 You can use both functions (Meta, and Compose Character) if you assign
1543 them to two different keys.
1545 * Emacs gets a segmentation fault at startup, on AIX4.2.
1547 If you are using IBM's xlc compiler, compile emacs.c
1548 without optimization; that should avoid the problem.
1550 * movemail compiled with POP support can't connect to the POP server.
1552 Make sure that the `pop' entry in /etc/services, or in the services
1553 NIS map if your machine uses NIS, has the same port number as the
1554 entry on the POP server. A common error is for the POP server to be
1555 listening on port 110, the assigned port for the POP3 protocol, while
1556 the client is trying to connect on port 109, the assigned port for the
1559 * Emacs crashes in x-popup-dialog.
1561 This can happen if the dialog widget cannot find the font it wants to
1562 use. You can work around the problem by specifying another font with
1563 an X resource--for example, `Emacs.dialog*.font: 9x15' (or any font that
1564 happens to exist on your X server).
1566 * Emacs crashes when you use Bibtex mode.
1568 This happens if your system puts a small limit on stack size. You can
1569 prevent the problem by using a suitable shell command (often `ulimit')
1570 to raise the stack size limit before you run Emacs.
1572 Patches to raise the stack size limit automatically in `main'
1573 (src/emacs.c) on various systems would be greatly appreciated.
1575 * Emacs crashes with SIGBUS or SIGSEGV on HPUX 9 after you delete a frame.
1577 We think this is due to a bug in the X libraries provided by HP. With
1578 the alternative X libraries in /usr/contrib/mitX11R5/lib, the problem
1581 * Emacs crashes with SIGBUS or SIGSEGV on Solaris after you delete a frame.
1583 We suspect that this is a similar bug in the X libraries provided by
1584 Sun. There is a report that one of these patches fixes the bug and
1585 makes the problem stop:
1587 105216-01 105393-01 105518-01 105621-01 105665-01 105615-02 105216-02
1588 105667-01 105401-08 105615-03 105621-02 105686-02 105736-01 105755-03
1589 106033-01 105379-01 105786-01 105181-04 105379-03 105786-04 105845-01
1590 105284-05 105669-02 105837-01 105837-02 105558-01 106125-02 105407-01
1592 Another person using a newer system (kernel patch level Generic_105181-06)
1593 suspects that the bug was fixed by one of these more recent patches:
1595 106040-07 SunOS 5.6: X Input & Output Method patch
1596 106222-01 OpenWindows 3.6: filemgr (ff.core) fixes
1597 105284-12 Motif 1.2.7: sparc Runtime library patch
1599 * Problems running Perl under Emacs on Windows NT/95.
1601 `perl -de 0' just hangs when executed in an Emacs subshell.
1602 The fault lies with Perl (indirectly with Windows NT/95).
1604 The problem is that the Perl debugger explicitly opens a connection to
1605 "CON", which is the DOS/NT equivalent of "/dev/tty", for interacting
1608 On Unix, this is okay, because Emacs (or the shell?) creates a
1609 pseudo-tty so that /dev/tty is really the pipe Emacs is using to
1610 communicate with the subprocess.
1612 On NT, this fails because CON always refers to the handle for the
1613 relevant console (approximately equivalent to a tty), and cannot be
1614 redirected to refer to the pipe Emacs assigned to the subprocess as
1617 A workaround is to modify perldb.pl to use STDIN/STDOUT instead of CON.
1621 *** PERL/LIB/PERLDB.PL.orig Wed May 26 08:24:18 1993
1622 --- PERL/LIB/PERLDB.PL Mon Jul 01 15:28:16 1996
1629 $rcfile="perldb.ini";
1637 $rcfile="perldb.ini";
1642 *** perl/5.001/lib/perl5db.pl.orig Sun Jun 04 21:13:40 1995
1643 --- perl/5.001/lib/perl5db.pl Mon Jul 01 17:00:08 1996
1650 $rcfile="perldb.ini";
1658 $rcfile="perldb.ini";
1662 * Problems on MS-DOG if DJGPP v2.0 is used to compile Emacs:
1664 There are two DJGPP library bugs which cause problems:
1666 * Running `shell-command' (or `compile', or `grep') you get
1667 `Searching for program: permission denied (EACCES), c:/command.com';
1668 * After you shell to DOS, Ctrl-Break kills Emacs.
1670 To work around these bugs, you can use two files in the msdos
1671 subdirectory: `is_exec.c' and `sigaction.c'. Compile them and link
1672 them into the Emacs executable `temacs'; then they will replace the
1673 incorrect library functions.
1675 * When compiling with DJGPP on Windows NT, "config msdos" fails.
1677 If the error message is "VDM has been already loaded", this is because
1678 Windows has a program called `redir.exe' that is incompatible with a
1679 program by the same name supplied with DJGPP, which is used by
1680 config.bat. To resolve this, move the DJGPP's `bin' subdirectory to
1681 the front of your PATH environment variable.
1683 * When compiling with DJGPP on Windows 95, Make fails for some targets
1686 This can happen if long file name support (the setting of environment
1687 variable LFN) when Emacs distribution was unpacked and during
1688 compilation are not the same. See the MSDOG section of INSTALL for
1689 the explanation of how to avoid this problem.
1691 * Emacs compiled for MSDOS cannot find some Lisp files, or other
1692 run-time support files, when long filename support is enabled.
1694 Usually, this problem will manifest itself when Emacs exits
1695 immediately after flashing the startup screen, because it cannot find
1696 the Lisp files it needs to load at startup. Redirect Emacs stdout
1697 and stderr to a file to see the error message printed by Emacs.
1699 Another manifestation of this problem is that Emacs is unable to load
1700 the support for editing program sources in languages such as C and
1703 This can happen if the Emacs distribution was unzipped without LFN
1704 support, thus causing long filenames to be truncated to the first 6
1705 characters and a numeric tail that Windows 95 normally attaches to it.
1706 You should unzip the files again with a utility that supports long
1707 filenames (such as djtar from DJGPP or InfoZip's UnZip program
1708 compiled with DJGPP v2). The MSDOG section of the file INSTALL
1709 explains this issue in more detail.
1711 Another possible reason for such failures is that Emacs compiled for
1712 MSDOS is used on Windows NT, where long file names are not supported
1713 by this version of Emacs, but the distribution was unpacked by an
1714 unzip program that preserved the long file names instead of truncating
1715 them to DOS 8+3 limits. To be useful on NT, the MSDOS port of Emacs
1716 must be unzipped by a DOS utility, so that long file names are
1719 * Emacs compiled with DJGPP complains at startup:
1721 "Wrong type of argument: internal-facep, msdos-menu-active-face"
1723 This can happen if you define an environment variable `TERM'. Emacs
1724 on MSDOS uses an internal terminal emulator which is disabled if the
1725 value of `TERM' is anything but the string "internal". Emacs then
1726 works as if its terminal were a dumb glass teletype that doesn't
1727 support faces. To work around this, arrange for `TERM' to be
1728 undefined when Emacs runs. The best way to do that is to add an
1729 [emacs] section to the DJGPP.ENV file which defines an empty value for
1730 `TERM'; this way, only Emacs gets the empty value, while the rest of
1731 your system works as before.
1733 * On Windows 95, Alt-f6 does not get through to Emacs.
1735 This character seems to be trapped by the kernel in Windows 95.
1736 You can enter M-f6 by typing ESC f6.
1738 * Typing Alt-Shift has strange effects on Windows.
1740 This combination of keys is a command to change keyboard layout. If
1741 you proceed to type another non-modifier key before you let go of Alt
1742 and Shift, the Alt and Shift act as modifiers in the usual way. A
1743 more permanent work around is to change it to another key combination,
1744 or disable it in the keyboard control panel.
1746 * `tparam' reported as a multiply-defined symbol when linking with ncurses.
1748 This problem results from an incompatible change in ncurses, in
1749 version 1.9.9e approximately. This version is unable to provide a
1750 definition of tparm without also defining tparam. This is also
1751 incompatible with Terminfo; as a result, the Emacs Terminfo support
1752 does not work with this version of ncurses.
1754 The fix is to install a newer version of ncurses, such as version 4.2.
1756 * Emacs does not start complaining that it cannot open termcap database file.
1758 If your system uses Terminfo rather than termcap (most modern
1759 systems do), this could happen if the proper version of
1760 ncurses is not visible to the Emacs configure script (i.e. it
1761 cannot be found along the usual path the linker looks for
1762 libraries). It can happen because your version of ncurses is
1763 obsolete, or is available only in form of binaries.
1765 The solution is to install an up-to-date version of ncurses in
1766 the developer's form (header files, static libraries and
1767 symbolic links); in some GNU/Linux distributions (e.g. Debian)
1768 it constitutes a separate package.
1770 * Strange results from format %d in a few cases, on a Sun.
1772 Sun compiler version SC3.0 has been found to miscompile part of
1773 editfns.c. The workaround is to compile with some other compiler such
1776 * Output from subprocess (such as man or diff) is randomly truncated
1777 on GNU/Linux systems.
1779 This is due to a kernel bug which seems to be fixed in Linux version
1782 * Error messages `internal facep []' happen on GNU/Linux systems.
1784 There is a report that replacing libc.so.5.0.9 with libc.so.5.2.16
1785 caused this to start happening. People are not sure why, but the
1786 problem seems unlikely to be in Emacs itself. Some suspect that it
1787 is actually Xlib which won't work with libc.so.5.2.16.
1789 Using the old library version is a workaround.
1791 * On Solaris, Emacs crashes if you use (display-time).
1793 This can happen if you configure Emacs without specifying the precise
1794 version of Solaris that you are using.
1796 * Emacs dumps core on startup, on Solaris.
1798 Bill Sebok says that the cause of this is Solaris 2.4 vendor patch
1799 102303-05, which extends the Solaris linker to deal with the Solaris
1800 Common Desktop Environment's linking needs. You can fix the problem
1801 by removing this patch and installing patch 102049-02 instead.
1802 However, that linker version won't work with CDE.
1804 Solaris 2.5 comes with a linker that has this bug. It is reported that if
1805 you install all the latest patches (as of June 1996), the bug is fixed.
1806 We suspect the crucial patch is one of these, but we don't know
1809 103093-03: [README] SunOS 5.5: kernel patch (2140557 bytes)
1810 102832-01: [README] OpenWindows 3.5: Xview Jumbo Patch (4181613 bytes)
1811 103242-04: [README] SunOS 5.5: linker patch (595363 bytes)
1813 (One user reports that the bug was fixed by those patches together
1814 with patches 102980-04, 103279-01, 103300-02, and 103468-01.)
1816 If you can determine which patch does fix the bug, please tell
1817 bug-gnu-emacs@gnu.org.
1819 Meanwhile, the GNU linker links Emacs properly on both Solaris 2.4 and
1822 * Emacs dumps core if lisp-complete-symbol is called, on Solaris.
1824 If you compile Emacs with the -fast or -xO4 option with version 3.0.2
1825 of the Sun C compiler, Emacs dumps core when lisp-complete-symbol is
1826 called. The problem does not happen if you compile with GCC.
1828 * "Cannot find callback list" messages from dialog boxes on HPUX, in
1829 Emacs built with Motif.
1831 This problem resulted from a bug in GCC 2.4.5. Newer GCC versions
1832 such as 2.7.0 fix the problem.
1834 * On Irix 6.0, make tries (and fails) to build a program named unexelfsgi
1836 A compiler bug inserts spaces into the string "unexelfsgi . o"
1837 in src/Makefile. Edit src/Makefile, after configure is run,
1838 find that string, and take out the spaces.
1840 Compiler fixes in Irix 6.0.1 should eliminate this problem.
1842 * "out of virtual swap space" on Irix 5.3
1844 This message occurs when the system runs out of swap space due to too
1845 many large programs running. The solution is either to provide more
1846 swap space or to reduce the number of large programs being run. You
1847 can check the current status of the swap space by executing the
1850 You can increase swap space by changing the file /etc/fstab. Adding a
1853 /usr/swap/swap.more swap swap pri=3 0 0
1855 where /usr/swap/swap.more is a file previously created (for instance
1856 by using /etc/mkfile), will increase the swap space by the size of
1857 that file. Execute `swap -m' or reboot the machine to activate the
1858 new swap area. See the manpages for `swap' and `fstab' for further
1861 The objectserver daemon can use up lots of memory because it can be
1862 swamped with NIS information. It collects information about all users
1863 on the network that can log on to the host.
1865 If you want to disable the objectserver completely, you can execute
1866 the command `chkconfig objectserver off' and reboot. That may disable
1867 some of the window system functionality, such as responding CDROM
1870 You can also remove NIS support from the objectserver. The SGI `admin'
1871 FAQ has a detailed description on how to do that; see question 35
1872 ("Why isn't the objectserver working?"). The admin FAQ can be found at
1873 ftp://viz.tamu.edu/pub/sgi/faq/.
1875 * With certain fonts, when the cursor appears on a character, the
1876 character doesn't appear--you get a solid box instead.
1878 One user on a Linux-based GNU system reported that this problem went
1879 away with installation of a new X server. The failing server was
1880 XFree86 3.1.1. XFree86 3.1.2 works.
1882 * On SunOS 4.1.3, Emacs unpredictably crashes in _yp_dobind_soft.
1884 This happens if you configure Emacs specifying just `sparc-sun-sunos4'
1885 on a system that is version 4.1.3. You must specify the precise
1886 version number (or let configure figure out the configuration, which
1887 it can do perfectly well for SunOS).
1889 * On SunOS 4, Emacs processes keep going after you kill the X server
1890 (or log out, if you logged in using X).
1892 Someone reported that recompiling with GCC 2.7.0 fixed this problem.
1894 * On AIX 4, some programs fail when run in a Shell buffer
1895 with an error message like No terminfo entry for "unknown".
1897 On AIX, many terminal type definitions are not installed by default.
1898 `unknown' is one of them. Install the "Special Generic Terminal
1899 Definitions" to make them defined.
1901 * On SunOS, you get linker errors
1902 ld: Undefined symbol
1903 _get_wmShellWidgetClass
1904 _get_applicationShellWidgetClass
1906 The fix to this is to install patch 100573 for OpenWindows 3.0
1907 or link libXmu statically.
1909 * On AIX 4.1.2, linker error messages such as
1910 ld: 0711-212 SEVERE ERROR: Symbol .__quous, found in the global symbol table
1911 of archive /usr/lib/libIM.a, was not defined in archive member shr.o.
1913 This is a problem in libIM.a. You can work around it by executing
1914 these shell commands in the src subdirectory of the directory where
1917 cp /usr/lib/libIM.a .
1921 Then change -lIM to ./libIM.a in the command to link temacs (in
1924 * Unpredictable segmentation faults on Solaris 2.3 and 2.4.
1926 A user reported that this happened in 19.29 when it was compiled with
1927 the Sun compiler, but not when he recompiled with GCC 2.7.0.
1929 We do not know whether something in Emacs is partly to blame for this.
1931 * Emacs exits with "X protocol error" when run with an X server for
1934 A certain X server for Windows had a bug which caused this.
1935 Supposedly the newer 32-bit version of this server doesn't have the
1938 * Emacs crashes at startup on MSDOS.
1940 Some users report that Emacs 19.29 requires dpmi memory management,
1941 and crashes on startup if the system does not have it. We don't yet
1942 know why this happens--perhaps these machines don't have enough real
1943 memory, or perhaps something is wrong in Emacs or the compiler.
1944 However, arranging to use dpmi support is a workaround.
1946 You can find out if you have a dpmi host by running go32 without
1947 arguments; it will tell you if it uses dpmi memory. For more
1948 information about dpmi memory, consult the djgpp FAQ. (djgpp
1949 is the GNU C compiler as packaged for MSDOS.)
1951 Compiling Emacs under MSDOS is extremely sensitive for proper memory
1952 configuration. If you experience problems during compilation, consider
1953 removing some or all memory resident programs (notably disk caches)
1954 and make sure that your memory managers are properly configured. See
1955 the djgpp faq for configuration hints.
1957 * A position you specified in .Xdefaults is ignored, using twm.
1959 twm normally ignores "program-specified" positions.
1960 You can tell it to obey them with this command in your `.twmrc' file:
1962 UsePPosition "on" #allow clients to request a position
1964 * Compiling lib-src says there is no rule to make test-distrib.c.
1966 This results from a bug in a VERY old version of GNU Sed. To solve
1967 the problem, install the current version of GNU Sed, then rerun
1968 Emacs's configure script.
1970 * Compiling wakeup, in lib-src, says it can't make wakeup.c.
1972 This results from a bug in GNU Sed version 2.03. To solve the
1973 problem, install the current version of GNU Sed, then rerun Emacs's
1976 * On Sunos 4.1.1, there are errors compiling sysdep.c.
1978 If you get errors such as
1980 "sysdep.c", line 2017: undefined structure or union
1981 "sysdep.c", line 2017: undefined structure or union
1982 "sysdep.c", line 2019: nodename undefined
1984 This can result from defining LD_LIBRARY_PATH. It is very tricky
1985 to use that environment variable with Emacs. The Emacs configure
1986 script links many test programs with the system libraries; you must
1987 make sure that the libraries available to configure are the same
1988 ones available when you build Emacs.
1990 * The right Alt key works wrong on German HP keyboards (and perhaps
1991 other non-English HP keyboards too).
1993 This is because HPUX defines the modifiers wrong in X. Here is a
1994 shell script to fix the problem; be sure that it is run after VUE
1995 configures the X server.
1997 xmodmap 2> /dev/null - << EOF
1998 keysym Alt_L = Meta_L
1999 keysym Alt_R = Meta_R
2004 keysym Mode_switch = NoSymbol
2006 keysym Meta_R = Mode_switch
2007 add mod2 = Mode_switch
2010 * The Emacs window disappears when you type M-q.
2012 Some versions of the Open Look window manager interpret M-q as a quit
2013 command for whatever window you are typing at. If you want to use
2014 Emacs with that window manager, you should try to configure the window
2015 manager to use some other command. You can disable the
2016 shortcut keys entirely by adding this line to ~/.OWdefaults:
2018 OpenWindows.WindowMenuAccelerators: False
2020 * Emacs does not notice when you release the mouse.
2022 There are reports that this happened with (some) Microsoft mice and
2023 that replacing the mouse made it stop.
2025 * Trouble using ptys on IRIX, or running out of ptys.
2027 The program mkpts (which may be in `/usr/adm' or `/usr/sbin') needs to
2028 be set-UID to root, or non-root programs like Emacs will not be able
2029 to allocate ptys reliably.
2031 * On Irix 5.2, unexelfsgi.c can't find cmplrs/stsupport.h.
2033 The file cmplrs/stsupport.h was included in the wrong file set in the
2034 Irix 5.2 distribution. You can find it in the optional fileset
2035 compiler_dev, or copy it from some other Irix 5.2 system. A kludgy
2036 workaround is to change unexelfsgi.c to include sym.h instead of
2039 * Slow startup on Linux-based GNU systems.
2041 People using systems based on the Linux kernel sometimes report that
2042 startup takes 10 to 15 seconds longer than `usual'.
2044 This is because Emacs looks up the host name when it starts.
2045 Normally, this takes negligible time; the extra delay is due to
2046 improper system configuration. This problem can occur for both
2047 networked and non-networked machines.
2049 Here is how to fix the configuration. It requires being root.
2053 First, make sure the files `/etc/hosts' and `/etc/host.conf' both
2054 exist. The first line in the `/etc/hosts' file should look like this
2055 (replace HOSTNAME with your host name):
2059 Also make sure that the `/etc/host.conf' files contains the following
2065 Any changes, permanent and temporary, to the host name should be
2066 indicated in the `/etc/hosts' file, since it acts a limited local
2067 database of addresses and names (e.g., some SLIP connections
2068 dynamically allocate ip addresses).
2070 ** Non-Networked Case
2072 The solution described in the networked case applies here as well.
2073 However, if you never intend to network your machine, you can use a
2074 simpler solution: create an empty `/etc/host.conf' file. The command
2075 `touch /etc/host.conf' suffices to create the file. The `/etc/hosts'
2076 file is not necessary with this approach.
2078 * On Solaris 2.4, Dired hangs and C-g does not work. Or Emacs hangs
2079 forever waiting for termination of a subprocess that is a zombie.
2081 casper@fwi.uva.nl says the problem is in X11R6. Rebuild libX11.so
2082 after changing the file xc/config/cf/sunLib.tmpl. Change the lines
2085 #define SharedX11Reqs -lthread
2090 #if OSMinorVersion < 4
2092 #define SharedX11Reqs -lthread
2096 Be sure also to edit x/config/cf/sun.cf so that OSMinorVersion is 4
2097 (as it should be for Solaris 2.4). The file has three definitions for
2098 OSMinorVersion: the first is for x86, the second for SPARC under
2099 Solaris, and the third for SunOS 4. Make sure to update the
2100 definition for your type of machine and system.
2102 Then do `make Everything' in the top directory of X11R6, to rebuild
2103 the makefiles and rebuild X. The X built this way work only on
2104 Solaris 2.4, not on 2.3.
2106 For multithreaded X to work it is necessary to install patch
2107 101925-02 to fix problems in header files [2.4]. You need
2108 to reinstall gcc or re-run just-fixinc after installing that
2111 However, Frank Rust <frust@iti.cs.tu-bs.de> used a simpler solution:
2113 #define ThreadedX YES
2115 #define ThreadedX NO
2116 in sun.cf and did `make World' to rebuild X11R6. Removing all
2117 `-DXTHREAD*' flags and `-lthread' entries from lib/X11/Makefile and
2118 typing 'make install' in that directory also seemed to work.
2120 * With M-x enable-flow-control, you need to type C-\ twice
2121 to do incremental search--a single C-\ gets no response.
2123 This has been traced to communicating with your machine via kermit,
2124 with C-\ as the kermit escape character. One solution is to use
2125 another escape character in kermit. One user did
2127 set escape-character 17
2129 in his .kermrc file, to make C-q the kermit escape character.
2131 * The Motif version of Emacs paints the screen a solid color.
2133 This has been observed to result from the following X resource:
2135 Emacs*default.attributeFont: -*-courier-medium-r-*-*-*-140-*-*-*-*-iso8859-*
2137 That the resource has this effect indicates a bug in something, but we
2138 do not yet know what. If it is an Emacs bug, we hope someone can
2139 explain what the bug is so we can fix it. In the mean time, removing
2140 the resource prevents the problem.
2142 * Emacs gets hung shortly after startup, on Sunos 4.1.3.
2144 We think this is due to a bug in Sunos. The word is that
2145 one of these Sunos patches fixes the bug:
2147 100075-11 100224-06 100347-03 100482-05 100557-02 100623-03 100804-03 101080-01
2148 100103-12 100249-09 100496-02 100564-07 100630-02 100891-10 101134-01
2149 100170-09 100296-04 100377-09 100507-04 100567-04 100650-02 101070-01 101145-01
2150 100173-10 100305-15 100383-06 100513-04 100570-05 100689-01 101071-03 101200-02
2151 100178-09 100338-05 100421-03 100536-02 100584-05 100784-01 101072-01 101207-01
2153 We don't know which of these patches really matter. If you find out
2154 which ones, please inform bug-gnu-emacs@gnu.org.
2156 * Emacs aborts while starting up, only when run without X.
2158 This problem often results from compiling Emacs with GCC when GCC was
2159 installed incorrectly. The usual error in installing GCC is to
2160 specify --includedir=/usr/include. Installation of GCC makes
2161 corrected copies of the system header files. GCC is supposed to use
2162 the corrected copies in preference to the original system headers.
2163 Specifying --includedir=/usr/include causes the original system header
2164 files to be used. On some systems, the definition of ioctl in the
2165 original system header files is invalid for ANSI C and causes Emacs
2168 The fix is to reinstall GCC, and this time do not specify --includedir
2169 when you configure it. Then recompile Emacs. Specifying --includedir
2170 is appropriate only in very special cases and it should *never* be the
2171 same directory where system header files are kept.
2173 * On Solaris 2.x, GCC complains "64 bit integer types not supported"
2175 This suggests that GCC is not installed correctly. Most likely you
2176 are using GCC 2.7.2.3 (or earlier) on Solaris 2.6 (or later); this
2177 does not work without patching. To run GCC 2.7.2.3 on Solaris 2.6 or
2178 later, you must patch fixinc.svr4 and reinstall GCC from scratch as
2179 described in the Solaris FAQ
2180 <http://www.wins.uva.nl/pub/solaris/solaris2.html>. A better fix is
2181 to upgrade to GCC 2.8.1 or later.
2183 * The Compose key on a DEC keyboard does not work as Meta key.
2185 This shell command should fix it:
2187 xmodmap -e 'keycode 0xb1 = Meta_L'
2189 * Regular expressions matching bugs on SCO systems.
2191 On SCO, there are problems in regexp matching when Emacs is compiled
2192 with the system compiler. The compiler version is "Microsoft C
2193 version 6", SCO 4.2.0h Dev Sys Maintenance Supplement 01/06/93; Quick
2194 C Compiler Version 1.00.46 (Beta). The solution is to compile with
2197 * On Sunos 4, you get the error ld: Undefined symbol __lib_version.
2199 This is the result of using cc or gcc with the shared library meant
2200 for acc (the Sunpro compiler). Check your LD_LIBRARY_PATH and delete
2201 /usr/lang/SC2.0.1 or some similar directory.
2203 * You can't select from submenus (in the X toolkit version).
2205 On certain systems, mouse-tracking and selection in top-level menus
2206 works properly with the X toolkit, but neither of them works when you
2207 bring up a submenu (such as Bookmarks or Compare or Apply Patch, in
2210 This works on most systems. There is speculation that the failure is
2211 due to bugs in old versions of X toolkit libraries, but no one really
2212 knows. If someone debugs this and finds the precise cause, perhaps a
2213 workaround can be found.
2215 * Unusable default font on SCO 3.2v4.
2217 The Open Desktop environment comes with default X resource settings
2218 that tell Emacs to use a variable-width font. Emacs cannot use such
2219 fonts, so it does not work.
2221 This is caused by the file /usr/lib/X11/app-defaults/ScoTerm, which is
2222 the application-specific resource file for the `scoterm' terminal
2223 emulator program. It contains several extremely general X resources
2224 that affect other programs besides `scoterm'. In particular, these
2225 resources affect Emacs also:
2227 *Font: -*-helvetica-medium-r-*--12-*-p-*
2228 *Background: scoBackground
2229 *Foreground: scoForeground
2231 The best solution is to create an application-specific resource file for
2232 Emacs, /usr/lib/X11/sco/startup/Emacs, with the following contents:
2234 Emacs*Font: -*-courier-medium-r-*-*-*-120-*-*-*-*-iso8859-1
2235 Emacs*Background: white
2236 Emacs*Foreground: black
2238 (These settings mimic the Emacs defaults, but you can change them to
2239 suit your needs.) This resource file is only read when the X server
2240 starts up, so you should restart it by logging out of the Open Desktop
2241 environment or by running `scologin stop; scologin start` from the shell
2242 as root. Alternatively, you can put these settings in the
2243 /usr/lib/X11/app-defaults/Emacs resource file and simply restart Emacs,
2244 but then they will not affect remote invocations of Emacs that use the
2245 Open Desktop display.
2247 These resource files are not normally shared across a network of SCO
2248 machines; you must create the file on each machine individually.
2250 * rcs2log gives you the awk error message "too many fields".
2252 This is due to an arbitrary limit in certain versions of awk.
2253 The solution is to use gawk (GNU awk).
2255 * Emacs is slow using X11R5 on HP/UX.
2257 This happens if you use the MIT versions of the X libraries--it
2258 doesn't run as fast as HP's version. People sometimes use the version
2259 because they see the HP version doesn't have the libraries libXaw.a,
2260 libXmu.a, libXext.a and others. HP/UX normally doesn't come with
2261 those libraries installed. To get good performance, you need to
2262 install them and rebuild Emacs.
2264 * Loading fonts is very slow.
2266 You might be getting scalable fonts instead of precomputed bitmaps.
2267 Known scalable font directories are "Type1" and "Speedo". A font
2268 directory contains scalable fonts if it contains the file
2271 If this is so, re-order your X windows font path to put the scalable
2272 font directories last. See the documentation of `xset' for details.
2274 With some X servers, it may be necessary to take the scalable font
2275 directories out of your path entirely, at least for Emacs 19.26.
2276 Changes in the future may make this unnecessary.
2278 * On AIX 3.2.4, releasing Ctrl/Act key has no effect, if Shift is down.
2280 Due to a feature of AIX, pressing or releasing the Ctrl/Act key is
2281 ignored when the Shift, Alt or AltGr keys are held down. This can
2282 lead to the keyboard being "control-locked"--ordinary letters are
2283 treated as control characters.
2285 You can get out of this "control-locked" state by pressing and
2286 releasing Ctrl/Act while not pressing or holding any other keys.
2288 * display-time causes kernel problems on ISC systems.
2290 Under Interactive Unix versions 3.0.1 and 4.0 (and probably other
2291 versions), display-time causes the loss of large numbers of STREVENT
2292 cells. Eventually the kernel's supply of these cells is exhausted.
2293 This makes emacs and the whole system run slow, and can make other
2294 processes die, in particular pcnfsd.
2296 Other emacs functions that communicate with remote processes may have
2297 the same problem. Display-time seems to be far the worst.
2299 The only known fix: Don't run display-time.
2301 * On Solaris, C-x doesn't get through to Emacs when you use the console.
2303 This is a Solaris feature (at least on Intel x86 cpus). Type C-r
2304 C-r C-t, to toggle whether C-x gets through to Emacs.
2306 * Error message `Symbol's value as variable is void: x', followed by
2307 segmentation fault and core dump.
2309 This has been tracked to a bug in tar! People report that tar erroneously
2310 added a line like this at the beginning of files of Lisp code:
2312 x FILENAME, N bytes, B tape blocks
2314 If your tar has this problem, install GNU tar--if you can manage to
2317 * Link failure when using acc on a Sun.
2319 To use acc, you need additional options just before the libraries, such as
2321 /usr/lang/SC2.0.1/values-Xt.o -L/usr/lang/SC2.0.1/cg87 -L/usr/lang/SC2.0.1
2323 and you need to add -lansi just before -lc.
2325 The precise file names depend on the compiler version, so we
2326 cannot easily arrange to supply them.
2328 * Link failure on IBM AIX 1.3 ptf 0013.
2330 There is a real duplicate definition of the function `_slibc_free' in
2331 the library /lib/libc_s.a (just do nm on it to verify). The
2335 ar xv libc_s.a NLtmtime.o
2336 ar dv libc_s.a NLtmtime.o
2338 * Undefined symbols _dlopen, _dlsym and/or _dlclose on a Sun.
2340 If you see undefined symbols _dlopen, _dlsym, or _dlclose when linking
2341 with -lX11, compile and link against the file mit/util/misc/dlsym.c in
2342 the MIT X11R5 distribution. Alternatively, link temacs using shared
2343 libraries with s/sunos4shr.h. (This doesn't work if you use the X
2346 If you get the additional error that the linker could not find
2347 lib_version.o, try extracting it from X11/usr/lib/X11/libvim.a in
2348 X11R4, then use it in the link.
2350 * Error messages `Wrong number of arguments: #<subr where-is-internal>, 5'
2352 This typically results from having the powerkey library loaded.
2353 Powerkey was designed for Emacs 19.22. It is obsolete now because
2354 Emacs 19 now has this feature built in; and powerkey also calls
2355 where-is-internal in an obsolete way.
2357 So the fix is to arrange not to load powerkey.
2359 * In Shell mode, you get a ^M at the end of every line.
2361 This happens to people who use tcsh, because it is trying to be too
2362 smart. It sees that the Shell uses terminal type `unknown' and turns
2363 on the flag to output ^M at the end of each line. You can fix the
2364 problem by adding this to your .cshrc file:
2367 if ($EMACS == "t") then
2369 stty -icrnl -onlcr -echo susp ^Z
2373 * An error message such as `X protocol error: BadMatch (invalid
2374 parameter attributes) on protocol request 93'.
2376 This comes from having an invalid X resource, such as
2378 (which is invalid because it specifies a color name for something
2379 that isn't a color.)
2381 The fix is to correct your X resources.
2383 * Undefined symbols when linking on Sunos 4.1 using --with-x-toolkit.
2385 If you get the undefined symbols _atowc _wcslen, _iswprint, _iswspace,
2386 _iswcntrl, _wcscpy, and _wcsncpy, then you need to add -lXwchar after
2387 -lXaw in the command that links temacs.
2389 This problem seems to arise only when the international language
2390 extensions to X11R5 are installed.
2392 * Typing C-c C-c in Shell mode kills your X server.
2394 This happens with Linux kernel 1.0 thru 1.04, approximately. The workaround is
2395 to define SIGNALS_VIA_CHARACTERS in config.h and recompile Emacs.
2396 Newer Linux kernel versions don't have this problem.
2398 * src/Makefile and lib-src/Makefile are truncated--most of the file missing.
2400 This can happen if configure uses GNU sed version 2.03. That version
2401 had a bug. GNU sed version 2.05 works properly.
2403 * Slow startup on X11R6 with X windows.
2405 If Emacs takes two minutes to start up on X11R6, see if your X
2406 resources specify any Adobe fonts. That causes the type-1 font
2407 renderer to start up, even if the font you asked for is not a type-1
2410 One way to avoid this problem is to eliminate the type-1 fonts from
2411 your font path, like this:
2413 xset -fp /usr/X11R6/lib/X11/fonts/Type1/
2415 * Pull-down menus appear in the wrong place, in the toolkit version of Emacs.
2417 An X resource of this form can cause the problem:
2419 Emacs*geometry: 80x55+0+0
2421 This resource is supposed to apply, and does apply, to the menus
2422 individually as well as to Emacs frames. If that is not what you
2423 want, rewrite the resource.
2425 To check thoroughly for such resource specifications, use `xrdb
2426 -query' to see what resources the X server records, and also look at
2427 the user's ~/.Xdefaults and ~/.Xdefaults-* files.
2429 * --with-x-toolkit version crashes when used with shared libraries.
2431 On some systems, including Sunos 4 and DGUX 5.4.2 and perhaps others,
2432 unexec doesn't work properly with the shared library for the X
2433 toolkit. You might be able to work around this by using a nonshared
2434 libXt.a library. The real fix is to upgrade the various versions of
2435 unexec and/or ralloc. We think this has been fixed on Sunos 4
2436 and Solaris in version 19.29.
2438 * `make install' fails on install-doc with `Error 141'.
2440 This happens on Ultrix 4.2 due to failure of a pipeline of tar
2441 commands. We don't know why they fail, but the bug seems not to be in
2442 Emacs. The workaround is to run the shell command in install-doc by
2445 * --with-x-toolkit option configures wrong on BSD/386.
2447 This problem is due to bugs in the shell in version 1.0 of BSD/386.
2448 The workaround is to edit the configure file to use some other shell,
2451 * Subprocesses remain, hanging but not zombies, on Sunos 5.3.
2453 A bug in Sunos 5.3 causes Emacs subprocesses to remain after Emacs
2454 exits. Sun patch # 101415-02 is part of the fix for this, but it only
2455 applies to ptys, and doesn't fix the problem with subprocesses
2456 communicating through pipes.
2458 * Mail is lost when sent to local aliases.
2460 Many emacs mail user agents (VM and rmail, for instance) use the
2461 sendmail.el library. This library can arrange for mail to be
2462 delivered by passing messages to the /usr/lib/sendmail (usually)
2463 program . In doing so, it passes the '-t' flag to sendmail, which
2464 means that the name of the recipient of the message is not on the
2465 command line and, therefore, that sendmail must parse the message to
2466 obtain the destination address.
2468 There is a bug in the SunOS4.1.1 and SunOS4.1.3 versions of sendmail.
2469 In short, when given the -t flag, the SunOS sendmail won't recognize
2470 non-local (i.e. NIS) aliases. It has been reported that the Solaris
2471 2.x versions of sendmail do not have this bug. For those using SunOS
2472 4.1, the best fix is to install sendmail V8 or IDA sendmail (which
2473 have other advantages over the regular sendmail as well). At the time
2474 of this writing, these official versions are available:
2476 Sendmail V8 on ftp.cs.berkeley.edu in /ucb/sendmail:
2477 sendmail.8.6.9.base.tar.Z (the base system source & documentation)
2478 sendmail.8.6.9.cf.tar.Z (configuration files)
2479 sendmail.8.6.9.misc.tar.Z (miscellaneous support programs)
2480 sendmail.8.6.9.xdoc.tar.Z (extended documentation, with postscript)
2482 IDA sendmail on vixen.cso.uiuc.edu in /pub:
2483 sendmail-5.67b+IDA-1.5.tar.gz
2485 * On AIX, you get this message when running Emacs:
2487 Could not load program emacs
2488 Symbol smtcheckinit in csh is undefined
2489 Error was: Exec format error
2493 Could not load program .emacs
2494 Symbol _system_con in csh is undefined
2495 Symbol _fp_trapsta in csh is undefined
2496 Error was: Exec format error
2498 These can happen when you try to run on AIX 3.2.5 a program that was
2499 compiled with 3.2.4. The fix is to recompile.
2501 * On AIX, you get this compiler error message:
2503 Processing include file ./XMenuInt.h
2504 1501-106: (S) Include file X11/Xlib.h not found.
2506 This means your system was installed with only the X11 runtime i.d
2507 libraries. You have to find your sipo (bootable tape) and install
2508 X11Dev... with smit.
2510 * You "lose characters" after typing Compose Character key.
2512 This is because the Compose Character key is defined as the keysym
2513 Multi_key, and Emacs (seeing that) does the proper X11
2514 character-composition processing. If you don't want your Compose key
2515 to do that, you can redefine it with xmodmap.
2517 For example, here's one way to turn it into a Meta key:
2519 xmodmap -e "keysym Multi_key = Meta_L"
2521 If all users at your site of a particular keyboard prefer Meta to
2522 Compose, you can make the remapping happen automatically by adding the
2523 xmodmap command to the xdm setup script for that display.
2525 * C-z just refreshes the screen instead of suspending Emacs.
2527 You are probably using a shell that doesn't support job control, even
2528 though the system itself is capable of it. Either use a different shell,
2529 or set the variable `cannot-suspend' to a non-nil value.
2531 * Watch out for .emacs files and EMACSLOADPATH environment vars
2533 These control the actions of Emacs.
2534 ~/.emacs is your Emacs init file.
2535 EMACSLOADPATH overrides which directories the function
2538 If you observe strange problems, check for these and get rid
2539 of them, then try again.
2541 * After running emacs once, subsequent invocations crash.
2543 Some versions of SVR4 have a serious bug in the implementation of the
2544 mmap () system call in the kernel; this causes emacs to run correctly
2545 the first time, and then crash when run a second time.
2547 Contact your vendor and ask for the mmap bug fix; in the mean time,
2548 you may be able to work around the problem by adding a line to your
2549 operating system description file (whose name is reported by the
2550 configure script) that reads:
2551 #define SYSTEM_MALLOC
2552 This makes Emacs use memory less efficiently, but seems to work around
2555 * Inability to send an Alt-modified key, when Emacs is communicating
2556 directly with an X server.
2558 If you have tried to bind an Alt-modified key as a command, and it
2559 does not work to type the command, the first thing you should check is
2560 whether the key is getting through to Emacs. To do this, type C-h c
2561 followed by the Alt-modified key. C-h c should say what kind of event
2562 it read. If it says it read an Alt-modified key, then make sure you
2563 have made the key binding correctly.
2565 If C-h c reports an event that doesn't have the Alt modifier, it may
2566 be because your X server has no key for the Alt modifier. The X
2567 server that comes from MIT does not set up the Alt modifier by
2570 If your keyboard has keys named Alt, you can enable them as follows:
2572 xmodmap -e 'add mod2 = Alt_L'
2573 xmodmap -e 'add mod2 = Alt_R'
2575 If the keyboard has just one key named Alt, then only one of those
2576 commands is needed. The modifier `mod2' is a reasonable choice if you
2577 are using an unmodified MIT version of X. Otherwise, choose any
2578 modifier bit not otherwise used.
2580 If your keyboard does not have keys named Alt, you can use some other
2581 keys. Use the keysym command in xmodmap to turn a function key (or
2582 some other 'spare' key) into Alt_L or into Alt_R, and then use the
2583 commands show above to make them modifier keys.
2585 Note that if you have Alt keys but no Meta keys, Emacs translates Alt
2586 into Meta. This is because of the great importance of Meta in Emacs.
2588 * `Pid xxx killed due to text modification or page I/O error'
2590 On HP/UX, you can get that error when the Emacs executable is on an NFS
2591 file system. HP/UX responds this way if it tries to swap in a page and
2592 does not get a response from the server within a timeout whose default
2593 value is just ten seconds.
2595 If this happens to you, extend the timeout period.
2597 * `expand-file-name' fails to work on any but the machine you dumped Emacs on.
2599 On Ultrix, if you use any of the functions which look up information
2600 in the passwd database before dumping Emacs (say, by using
2601 expand-file-name in site-init.el), then those functions will not work
2602 in the dumped Emacs on any host but the one Emacs was dumped on.
2604 The solution? Don't use expand-file-name in site-init.el, or in
2605 anything it loads. Yuck - some solution.
2607 I'm not sure why this happens; if you can find out exactly what is
2608 going on, and perhaps find a fix or a workaround, please let us know.
2609 Perhaps the YP functions cache some information, the cache is included
2610 in the dumped Emacs, and is then inaccurate on any other host.
2612 * On some variants of SVR4, Emacs does not work at all with X.
2614 Try defining BROKEN_FIONREAD in your config.h file. If this solves
2615 the problem, please send a bug report to tell us this is needed; be
2616 sure to say exactly what type of machine and system you are using.
2618 * Linking says that the functions insque and remque are undefined.
2620 Change oldXMenu/Makefile by adding insque.o to the variable OBJS.
2622 * Emacs fails to understand most Internet host names, even though
2623 the names work properly with other programs on the same system.
2624 * Emacs won't work with X-windows if the value of DISPLAY is HOSTNAME:0.
2625 * GNUs can't make contact with the specified host for nntp.
2627 This typically happens on Suns and other systems that use shared
2628 libraries. The cause is that the site has installed a version of the
2629 shared library which uses a name server--but has not installed a
2630 similar version of the unshared library which Emacs uses.
2632 The result is that most programs, using the shared library, work with
2633 the nameserver, but Emacs does not.
2635 The fix is to install an unshared library that corresponds to what you
2636 installed in the shared library, and then relink Emacs.
2638 On SunOS 4.1, simply define HAVE_RES_INIT.
2640 If you have already installed the name resolver in the file libresolv.a,
2641 then you need to compile Emacs to use that library. The easiest way to
2642 do this is to add to config.h a definition of LIBS_SYSTEM, LIBS_MACHINE
2643 or LIB_STANDARD which uses -lresolv. Watch out! If you redefine a macro
2644 that is already in use in your configuration to supply some other libraries,
2645 be careful not to lose the others.
2647 Thus, you could start by adding this to config.h:
2649 #define LIBS_SYSTEM -lresolv
2651 Then if this gives you an error for redefining a macro, and you see that
2652 the s- file defines LIBS_SYSTEM as -lfoo -lbar, you could change config.h
2655 #define LIBS_SYSTEM -lresolv -lfoo -lbar
2657 * On a Sun running SunOS 4.1.1, you get this error message from GNU ld:
2659 /lib/libc.a(_Q_sub.o): Undefined symbol __Q_get_rp_rd referenced from text segment
2661 The problem is in the Sun shared C library, not in GNU ld.
2663 The solution is to install Patch-ID# 100267-03 from Sun.
2665 * Self documentation messages are garbled.
2667 This means that the file `etc/DOC-...' doesn't properly correspond
2668 with the Emacs executable. Redumping Emacs and then installing the
2669 corresponding pair of files should fix the problem.
2671 * Trouble using ptys on AIX.
2673 People often install the pty devices on AIX incorrectly.
2674 Use `smit pty' to reinstall them properly.
2676 * Shell mode on HP/UX gives the message, "`tty`: Ambiguous".
2678 christos@theory.tn.cornell.edu says:
2680 The problem is that in your .cshrc you have something that tries to
2681 execute `tty`. If you are not running the shell on a real tty then
2682 tty will print "not a tty". Csh expects one word in some places,
2683 but tty is giving it back 3.
2685 The solution is to add a pair of quotes around `tty` to make it a single
2688 if (`tty` == "/dev/console")
2690 should be changed to:
2692 if ("`tty`" == "/dev/console")
2694 Even better, move things that set up terminal sections out of .cshrc
2697 * Using X Windows, control-shift-leftbutton makes Emacs hang.
2699 Use the shell command `xset bc' to make the old X Menu package work.
2701 * Emacs running under X Windows does not handle mouse clicks.
2702 * `emacs -geometry 80x20' finds a file named `80x20'.
2704 One cause of such problems is having (setq term-file-prefix nil) in
2705 your .emacs file. Another cause is a bad value of EMACSLOADPATH in
2708 * Emacs gets error message from linker on Sun.
2710 If the error message says that a symbol such as `f68881_used' or
2711 `ffpa_used' or `start_float' is undefined, this probably indicates
2712 that you have compiled some libraries, such as the X libraries,
2713 with a floating point option other than the default.
2715 It's not terribly hard to make this work with small changes in
2716 crt0.c together with linking with Fcrt1.o, Wcrt1.o or Mcrt1.o.
2717 However, the easiest approach is to build Xlib with the default
2718 floating point option: -fsoft.
2720 * Emacs fails to get default settings from X Windows server.
2722 The X library in X11R4 has a bug; it interchanges the 2nd and 3rd
2723 arguments to XGetDefaults. Define the macro XBACKWARDS in config.h to
2724 tell Emacs to compensate for this.
2726 I don't believe there is any way Emacs can determine for itself
2727 whether this problem is present on a given system.
2729 * Keyboard input gets confused after a beep when using a DECserver
2732 This problem seems to be a matter of configuring the DECserver to use
2733 7 bit characters rather than 8 bit characters.
2735 * M-x shell persistently reports "Process shell exited abnormally with code 1".
2737 This happened on Suns as a result of what is said to be a bug in Sunos
2738 version 4.0.x. The only fix was to reboot the machine.
2740 * Programs running under terminal emulator do not recognize `emacs'
2743 The cause of this is a shell startup file that sets the TERMCAP
2744 environment variable. The terminal emulator uses that variable to
2745 provide the information on the special terminal type that Emacs
2748 Rewrite your shell startup file so that it does not change TERMCAP
2749 in such a case. You could use the following conditional which sets
2750 it only if it is undefined.
2752 if ( ! ${?TERMCAP} ) setenv TERMCAP ~/my-termcap-file
2754 Or you could set TERMCAP only when you set TERM--which should not
2755 happen in a non-login shell.
2757 * X Windows doesn't work if DISPLAY uses a hostname.
2759 People have reported kernel bugs in certain systems that cause Emacs
2760 not to work with X Windows if DISPLAY is set using a host name. But
2761 the problem does not occur if DISPLAY is set to `unix:0.0'. I think
2762 the bug has to do with SIGIO or FIONREAD.
2764 You may be able to compensate for the bug by doing (set-input-mode nil nil).
2765 However, that has the disadvantage of turning off interrupts, so that
2766 you are unable to quit out of a Lisp program by typing C-g.
2768 The easy way to do this is to put
2770 (setq x-sigio-bug t)
2772 in your site-init.el file.
2774 * Problem with remote X server on Suns.
2776 On a Sun, running Emacs on one machine with the X server on another
2777 may not work if you have used the unshared system libraries. This
2778 is because the unshared libraries fail to use YP for host name lookup.
2779 As a result, the host name you specify may not be recognized.
2781 * Shell mode ignores interrupts on Apollo Domain
2783 You may find that M-x shell prints the following message:
2785 Warning: no access to tty; thus no job control in this shell...
2787 This can happen if there are not enough ptys on your system.
2788 Here is how to make more of them.
2792 # shows how many pty's you have. I had 8, named pty0 to pty7)
2794 # creates eight new pty's
2796 * Fatal signal in the command temacs -l loadup inc dump
2798 This command is the final stage of building Emacs. It is run by the
2799 Makefile in the src subdirectory, or by build.com on VMS.
2801 It has been known to get fatal errors due to insufficient swapping
2802 space available on the machine.
2804 On 68000's, it has also happened because of bugs in the
2805 subroutine `alloca'. Verify that `alloca' works right, even
2806 for large blocks (many pages).
2808 * test-distrib says that the distribution has been clobbered
2809 * or, temacs prints "Command key out of range 0-127"
2810 * or, temacs runs and dumps emacs, but emacs totally fails to work.
2811 * or, temacs gets errors dumping emacs
2813 This can be because the .elc files have been garbled. Do not be
2814 fooled by the fact that most of a .elc file is text: these are
2815 binary files and can contain all 256 byte values.
2817 In particular `shar' cannot be used for transmitting GNU Emacs.
2818 It typically truncates "lines". What appear to be "lines" in
2819 a binary file can of course be of any length. Even once `shar'
2820 itself is made to work correctly, `sh' discards null characters
2821 when unpacking the shell archive.
2823 I have also seen character \177 changed into \377. I do not know
2824 what transfer means caused this problem. Various network
2825 file transfer programs are suspected of clobbering the high bit.
2827 If you have a copy of Emacs that has been damaged in its
2828 nonprinting characters, you can fix them:
2830 1) Record the names of all the .elc files.
2831 2) Delete all the .elc files.
2832 3) Recompile alloc.c with a value of PURESIZE twice as large.
2833 (See puresize.h.) You might as well save the old alloc.o.
2834 4) Remake emacs. It should work now.
2835 5) Running emacs, do Meta-x byte-compile-file repeatedly
2836 to recreate all the .elc files that used to exist.
2837 You may need to increase the value of the variable
2838 max-lisp-eval-depth to succeed in running the compiler interpreted
2839 on certain .el files. 400 was sufficient as of last report.
2840 6) Reinstall the old alloc.o (undoing changes to alloc.c if any)
2842 7) Remake emacs. It should work now, with valid .elc files.
2844 * temacs prints "Pure Lisp storage exhausted"
2846 This means that the Lisp code loaded from the .elc and .el
2847 files during temacs -l loadup inc dump took up more
2848 space than was allocated.
2850 This could be caused by
2851 1) adding code to the preloaded Lisp files
2852 2) adding more preloaded files in loadup.el
2853 3) having a site-init.el or site-load.el which loads files.
2854 Note that ANY site-init.el or site-load.el is nonstandard;
2855 if you have received Emacs from some other site
2856 and it contains a site-init.el or site-load.el file, consider
2858 4) getting the wrong .el or .elc files
2859 (not from the directory you expected).
2860 5) deleting some .elc files that are supposed to exist.
2861 This would cause the source files (.el files) to be
2862 loaded instead. They take up more room, so you lose.
2863 6) a bug in the Emacs distribution which underestimates
2866 If the need for more space is legitimate, change the definition
2867 of PURESIZE in puresize.h.
2869 But in some of the cases listed above, this problem is a consequence
2870 of something else that is wrong. Be sure to check and fix the real
2873 * Changes made to .el files do not take effect.
2875 You may have forgotten to recompile them into .elc files.
2876 Then the old .elc files will be loaded, and your changes
2877 will not be seen. To fix this, do M-x byte-recompile-directory
2878 and specify the directory that contains the Lisp files.
2880 Emacs should print a warning when loading a .elc file which is older
2881 than the corresponding .el file.
2883 * The dumped Emacs crashes when run, trying to write pure data.
2885 Two causes have been seen for such problems.
2887 1) On a system where getpagesize is not a system call, it is defined
2888 as a macro. If the definition (in both unexec.c and malloc.c) is wrong,
2889 it can cause problems like this. You might be able to find the correct
2890 value in the man page for a.out (5).
2892 2) Some systems allocate variables declared static among the
2893 initialized variables. Emacs makes all initialized variables in most
2894 of its files pure after dumping, but the variables declared static and
2895 not initialized are not supposed to be pure. On these systems you
2896 may need to add "#define static" to the m- or the s- file.
2898 * Compilation errors on VMS.
2900 You will get warnings when compiling on VMS because there are
2901 variable names longer than 32 (or whatever it is) characters.
2902 This is not an error. Ignore it.
2904 VAX C does not support #if defined(foo). Uses of this construct
2905 were removed, but some may have crept back in. They must be rewritten.
2907 There is a bug in the C compiler which fails to sign extend characters
2908 in conditional expressions. The bug is:
2913 The result is i == 255; the fix is to typecast the char in the
2914 conditional expression as an (int). Known occurrences of such
2915 constructs in Emacs have been fixed.
2917 * rmail gets error getting new mail
2919 rmail gets new mail from /usr/spool/mail/$USER using a program
2920 called `movemail'. This program interlocks with /bin/mail using
2921 the protocol defined by /bin/mail.
2923 There are two different protocols in general use. One of them uses
2924 the `flock' system call. The other involves creating a lock file;
2925 `movemail' must be able to write in /usr/spool/mail in order to do
2926 this. You control which one is used by defining, or not defining,
2927 the macro MAIL_USE_FLOCK in config.h or the m- or s- file it includes.
2928 IF YOU DON'T USE THE FORM OF INTERLOCKING THAT IS NORMAL ON YOUR
2929 SYSTEM, YOU CAN LOSE MAIL!
2931 If your system uses the lock file protocol, and fascist restrictions
2932 prevent ordinary users from writing the lock files in /usr/spool/mail,
2933 you may need to make `movemail' setgid to a suitable group such as
2934 `mail'. You can use these commands (as root):
2939 If your system uses the lock file protocol, and fascist restrictions
2940 prevent ordinary users from writing the lock files in /usr/spool/mail,
2941 you may need to make `movemail' setgid to a suitable group such as
2942 `mail'. To do this, use the following commands (as root) after doing the
2948 Installation normally copies movemail from the build directory to an
2949 installation directory which is usually under /usr/local/lib. The
2950 installed copy of movemail is usually in the directory
2951 /usr/local/lib/emacs/VERSION/TARGET. You must change the group and
2952 mode of the installed copy; changing the group and mode of the build
2953 directory copy is ineffective.
2955 * Emacs spontaneously displays "I-search: " at the bottom of the screen.
2957 This means that Control-S/Control-Q (XON/XOFF) "flow control" is being
2958 used. C-s/C-q flow control is bad for Emacs editors because it takes
2959 away C-s and C-q as user commands. Since editors do not output long
2960 streams of text without user commands, there is no need for a
2961 user-issuable "stop output" command in an editor; therefore, a
2962 properly designed flow control mechanism would transmit all possible
2963 input characters without interference. Designing such a mechanism is
2964 easy, for a person with at least half a brain.
2966 There are three possible reasons why flow control could be taking place:
2968 1) Terminal has not been told to disable flow control
2969 2) Insufficient padding for the terminal in use
2970 3) Some sort of terminal concentrator or line switch is responsible
2972 First of all, many terminals have a set-up mode which controls whether
2973 they generate XON/XOFF flow control characters. This must be set to
2974 "no XON/XOFF" in order for Emacs to work. Sometimes there is an
2975 escape sequence that the computer can send to turn flow control off
2976 and on. If so, perhaps the termcap `ti' string should turn flow
2977 control off, and the `te' string should turn it on.
2979 Once the terminal has been told "no flow control", you may find it
2980 needs more padding. The amount of padding Emacs sends is controlled
2981 by the termcap entry for the terminal in use, and by the output baud
2982 rate as known by the kernel. The shell command `stty' will print
2983 your output baud rate; `stty' with suitable arguments will set it if
2984 it is wrong. Setting to a higher speed causes increased padding. If
2985 the results are wrong for the correct speed, there is probably a
2986 problem in the termcap entry. You must speak to a local Unix wizard
2987 to fix this. Perhaps you are just using the wrong terminal type.
2989 For terminals that lack a "no flow control" mode, sometimes just
2990 giving lots of padding will prevent actual generation of flow control
2991 codes. You might as well try it.
2993 If you are really unlucky, your terminal is connected to the computer
2994 through a concentrator which sends XON/XOFF flow control to the
2995 computer, or it insists on sending flow control itself no matter how
2996 much padding you give it. Unless you can figure out how to turn flow
2997 control off on this concentrator (again, refer to your local wizard),
2998 you are screwed! You should have the terminal or concentrator
2999 replaced with a properly designed one. In the mean time, some drastic
3000 measures can make Emacs semi-work.
3002 You can make Emacs ignore C-s and C-q and let the operating system
3003 handle them. To do this on a per-session basis, just type M-x
3004 enable-flow-control RET. You will see a message that C-\ and C-^ are
3005 now translated to C-s and C-q. (Use the same command M-x
3006 enable-flow-control to turn *off* this special mode. It toggles flow
3009 If C-\ and C-^ are inconvenient for you (for example, if one of them
3010 is the escape character of your terminal concentrator), you can choose
3011 other characters by setting the variables flow-control-c-s-replacement
3012 and flow-control-c-q-replacement. But choose carefully, since all
3013 other control characters are already used by emacs.
3015 IMPORTANT: if you type C-s by accident while flow control is enabled,
3016 Emacs output will freeze, and you will have to remember to type C-q in
3019 If you work in an environment where a majority of terminals of a
3020 certain type are flow control hobbled, you can use the function
3021 `enable-flow-control-on' to turn on this flow control avoidance scheme
3022 automatically. Here is an example:
3024 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
3026 If this isn't quite correct (e.g. you have a mixture of flow-control hobbled
3027 and good vt200 terminals), you can still run enable-flow-control
3030 I have no intention of ever redesigning the Emacs command set for the
3031 assumption that terminals use C-s/C-q flow control. XON/XOFF flow
3032 control technique is a bad design, and terminals that need it are bad
3033 merchandise and should not be purchased. Now that X is becoming
3034 widespread, XON/XOFF seems to be on the way out. If you can get some
3035 use out of GNU Emacs on inferior terminals, more power to you, but I
3036 will not make Emacs worse for properly designed systems for the sake
3037 of inferior systems.
3039 * Control-S and Control-Q commands are ignored completely.
3041 For some reason, your system is using brain-damaged C-s/C-q flow
3042 control despite Emacs's attempts to turn it off. Perhaps your
3043 terminal is connected to the computer through a concentrator
3044 that wants to use flow control.
3046 You should first try to tell the concentrator not to use flow control.
3047 If you succeed in this, try making the terminal work without
3048 flow control, as described in the preceding section.
3050 If that line of approach is not successful, map some other characters
3051 into C-s and C-q using keyboard-translate-table. The example above
3052 shows how to do this with C-^ and C-\.
3054 * Control-S and Control-Q commands are ignored completely on a net connection.
3056 Some versions of rlogin (and possibly telnet) do not pass flow
3057 control characters to the remote system to which they connect.
3058 On such systems, emacs on the remote system cannot disable flow
3059 control on the local system.
3061 One way to cure this is to disable flow control on the local host
3062 (the one running rlogin, not the one running rlogind) using the
3063 stty command, before starting the rlogin process. On many systems,
3064 "stty start u stop u" will do this.
3066 Some versions of tcsh will prevent even this from working. One way
3067 around this is to start another shell before starting rlogin, and
3068 issue the stty command to disable flow control from that shell.
3070 If none of these methods work, the best solution is to type
3071 M-x enable-flow-control at the beginning of your emacs session, or
3072 if you expect the problem to continue, add a line such as the
3073 following to your .emacs (on the host running rlogind):
3075 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
3077 See the entry about spontaneous display of I-search (above) for more
3080 * Screen is updated wrong, but only on one kind of terminal.
3082 This could mean that the termcap entry you are using for that
3083 terminal is wrong, or it could mean that Emacs has a bug handing
3084 the combination of features specified for that terminal.
3086 The first step in tracking this down is to record what characters
3087 Emacs is sending to the terminal. Execute the Lisp expression
3088 (open-termscript "./emacs-script") to make Emacs write all
3089 terminal output into the file ~/emacs-script as well; then do
3090 what makes the screen update wrong, and look at the file
3091 and decode the characters using the manual for the terminal.
3092 There are several possibilities:
3094 1) The characters sent are correct, according to the terminal manual.
3096 In this case, there is no obvious bug in Emacs, and most likely you
3097 need more padding, or possibly the terminal manual is wrong.
3099 2) The characters sent are incorrect, due to an obscure aspect
3100 of the terminal behavior not described in an obvious way
3103 This case is hard. It will be necessary to think of a way for
3104 Emacs to distinguish between terminals with this kind of behavior
3105 and other terminals that behave subtly differently but are
3106 classified the same by termcap; or else find an algorithm for
3107 Emacs to use that avoids the difference. Such changes must be
3108 tested on many kinds of terminals.
3110 3) The termcap entry is wrong.
3112 See the file etc/TERMS for information on changes
3113 that are known to be needed in commonly used termcap entries
3114 for certain terminals.
3116 4) The characters sent are incorrect, and clearly cannot be
3117 right for any terminal with the termcap entry you were using.
3119 This is unambiguously an Emacs bug, and can probably be fixed
3120 in termcap.c, tparam.c, term.c, scroll.c, cm.c or dispnew.c.
3122 * Output from Control-V is slow.
3124 On many bit-map terminals, scrolling operations are fairly slow.
3125 Often the termcap entry for the type of terminal in use fails
3126 to inform Emacs of this. The two lines at the bottom of the screen
3127 before a Control-V command are supposed to appear at the top after
3128 the Control-V command. If Emacs thinks scrolling the lines is fast,
3129 it will scroll them to the top of the screen.
3131 If scrolling is slow but Emacs thinks it is fast, the usual reason is
3132 that the termcap entry for the terminal you are using does not
3133 specify any padding time for the `al' and `dl' strings. Emacs
3134 concludes that these operations take only as much time as it takes to
3135 send the commands at whatever line speed you are using. You must
3136 fix the termcap entry to specify, for the `al' and `dl', as much
3137 time as the operations really take.
3139 Currently Emacs thinks in terms of serial lines which send characters
3140 at a fixed rate, so that any operation which takes time for the
3141 terminal to execute must also be padded. With bit-map terminals
3142 operated across networks, often the network provides some sort of
3143 flow control so that padding is never needed no matter how slow
3144 an operation is. You must still specify a padding time if you want
3145 Emacs to realize that the operation takes a long time. This will
3146 cause padding characters to be sent unnecessarily, but they do
3147 not really cost much. They will be transmitted while the scrolling
3148 is happening and then discarded quickly by the terminal.
3150 Most bit-map terminals provide commands for inserting or deleting
3151 multiple lines at once. Define the `AL' and `DL' strings in the
3152 termcap entry to say how to do these things, and you will have
3153 fast output without wasted padding characters. These strings should
3154 each contain a single %-spec saying how to send the number of lines
3155 to be scrolled. These %-specs are like those in the termcap
3158 You should also define the `IC' and `DC' strings if your terminal
3159 has a command to insert or delete multiple characters. These
3160 take the number of positions to insert or delete as an argument.
3162 A `cs' string to set the scrolling region will reduce the amount
3163 of motion you see on the screen when part of the screen is scrolled.
3165 * Your Delete key sends a Backspace to the terminal, using an AIXterm.
3167 The solution is to include in your .Xdefaults the lines:
3169 *aixterm.Translations: #override <Key>BackSpace: string(0x7f)
3170 aixterm*ttyModes: erase ^?
3172 This makes your Backspace key send DEL (ASCII 127).
3174 * You type Control-H (Backspace) expecting to delete characters.
3176 Put `stty dec' in your .login file and your problems will disappear
3179 The choice of Backspace for erasure was based on confusion, caused by
3180 the fact that backspacing causes erasure (later, when you type another
3181 character) on most display terminals. But it is a mistake. Deletion
3182 of text is not the same thing as backspacing followed by failure to
3183 overprint. I do not wish to propagate this confusion by conforming
3186 For this reason, I believe `stty dec' is the right mode to use,
3187 and I have designed Emacs to go with that. If there were a thousand
3188 other control characters, I would define Control-h to delete as well;
3189 but there are not very many other control characters, and I think
3190 that providing the most mnemonic possible Help character is more
3191 important than adapting to people who don't use `stty dec'.
3193 If you are obstinate about confusing buggy overprinting with deletion,
3194 you can redefine Backspace in your .emacs file:
3195 (global-set-key "\b" 'delete-backward-char)
3196 You can probably access help-command via f1.
3198 * Editing files through RFS gives spurious "file has changed" warnings.
3199 It is possible that a change in Emacs 18.37 gets around this problem,
3200 but in case not, here is a description of how to fix the RFS bug that
3203 There was a serious pair of bugs in the handling of the fsync() system
3204 call in the RFS server.
3206 The first is that the fsync() call is handled as another name for the
3207 close() system call (!!). It appears that fsync() is not used by very
3208 many programs; Emacs version 18 does an fsync() before closing files
3209 to make sure that the bits are on the disk.
3211 This is fixed by the enclosed patch to the RFS server.
3213 The second, more serious problem, is that fsync() is treated as a
3214 non-blocking system call (i.e., it's implemented as a message that
3215 gets sent to the remote system without waiting for a reply). Fsync is
3216 a useful tool for building atomic file transactions. Implementing it
3217 as a non-blocking RPC call (when the local call blocks until the sync
3218 is done) is a bad idea; unfortunately, changing it will break the RFS
3219 protocol. No fix was supplied for this problem.
3221 (as always, your line numbers may vary)
3223 % rcsdiff -c -r1.2 serversyscall.c
3224 RCS file: RCS/serversyscall.c,v
3225 retrieving revision 1.2
3226 diff -c -r1.2 serversyscall.c
3227 *** /tmp/,RCSt1003677 Wed Jan 28 15:15:02 1987
3228 --- serversyscall.c Wed Jan 28 15:14:48 1987
3232 * No return sent for close or fsync!
3234 ! if (syscall == RSYS_close || syscall == RSYS_fsync)
3235 proc->p_returnval = deallocate_fd(proc, msg->m_args[0]);
3240 * No return sent for close or fsync!
3242 ! if (syscall == RSYS_close)
3243 proc->p_returnval = deallocate_fd(proc, msg->m_args[0]);
3247 * Vax C compiler bugs affecting Emacs.
3249 You may get one of these problems compiling Emacs:
3251 foo.c line nnn: compiler error: no table entry for op STASG
3252 foo.c: fatal error in /lib/ccom
3254 These are due to bugs in the C compiler; the code is valid C.
3255 Unfortunately, the bugs are unpredictable: the same construct
3256 may compile properly or trigger one of these bugs, depending
3257 on what else is in the source file being compiled. Even changes
3258 in header files that should not affect the file being compiled
3259 can affect whether the bug happens. In addition, sometimes files
3260 that compile correctly on one machine get this bug on another machine.
3262 As a result, it is hard for me to make sure this bug will not affect
3263 you. I have attempted to find and alter these constructs, but more
3264 can always appear. However, I can tell you how to deal with it if it
3265 should happen. The bug comes from having an indexed reference to an
3266 array of Lisp_Objects, as an argument in a function call:
3269 ... foo (5, args[i], ...)...
3270 putting the argument into a temporary variable first, as in
3275 ... foo (r, tem, ...)...
3276 causes the problem to go away.
3277 The `contents' field of a Lisp vector is an array of Lisp_Objects,
3278 so you may see the problem happening with indexed references to that.
3280 * 68000 C compiler problems
3282 Various 68000 compilers have different problems.
3283 These are some that have been observed.
3285 ** Using value of assignment expression on union type loses.
3286 This means that x = y = z; or foo (x = z); does not work
3287 if x is of type Lisp_Object.
3289 ** "cannot reclaim" error.
3291 This means that an expression is too complicated. You get the correct
3292 line number in the error message. The code must be rewritten with
3293 simpler expressions.
3295 ** XCONS, XSTRING, etc macros produce incorrect code.
3297 If temacs fails to run at all, this may be the cause.
3298 Compile this test program and look at the assembler code:
3300 struct foo { char x; unsigned int y : 24; };
3305 test ((int *) arg.y);
3308 If the code is incorrect, your compiler has this problem.
3309 In the XCONS, etc., macros in lisp.h you must replace (a).u.val with
3310 ((a).u.val + coercedummy) where coercedummy is declared as int.
3312 This problem will not happen if the m-...h file for your type
3313 of machine defines NO_UNION_TYPE. That is the recommended setting now.
3315 * C compilers lose on returning unions
3317 I hear that some C compilers cannot handle returning a union type.
3318 Most of the functions in GNU Emacs return type Lisp_Object, which is
3319 defined as a union on some rare architectures.
3321 This problem will not happen if the m-...h file for your type
3322 of machine defines NO_UNION_TYPE.
3325 Copyright 1987,88,89,93,94,95,96,97,98,1999,2001,2002
3326 Free Software Foundation, Inc.
3328 Copying and redistribution of this file with or without modification
3329 are permitted without royalty provided this notice is preserved.
3333 paragraph-separate: "[
\f]*$"