1 Copyright (C) 1987, 1988, 1989, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
2 2001, 2002, 2003, 2004, 2005, 2006, 2007
3 Free Software Foundation, Inc.
4 See the end of the file for license conditions.
7 This file describes various problems that have been encountered
8 in compiling, installing and running GNU Emacs. Try doing Ctl-C Ctl-t
9 and browsing through the outline headers.
11 * Emacs startup failures
13 ** Emacs fails to start, complaining about missing fonts.
15 A typical error message might be something like
17 No fonts match `-*-fixed-medium-r-*--6-*-*-*-*-*-iso8859-1'
19 This happens because some X resource specifies a bad font family for
20 Emacs to use. The possible places where this specification might be
23 - in your ~/.Xdefaults file
25 - client-side X resource file, such as ~/Emacs or
26 /usr/X11R6/lib/app-defaults/Emacs or
27 /usr/X11R6/lib/X11/app-defaults/Emacs
29 One of these files might have bad or malformed specification of a
30 fontset that Emacs should use. To fix the problem, you need to find
31 the problematic line(s) and correct them.
33 ** Emacs aborts while starting up, only when run without X.
35 This problem often results from compiling Emacs with GCC when GCC was
36 installed incorrectly. The usual error in installing GCC is to
37 specify --includedir=/usr/include. Installation of GCC makes
38 corrected copies of the system header files. GCC is supposed to use
39 the corrected copies in preference to the original system headers.
40 Specifying --includedir=/usr/include causes the original system header
41 files to be used. On some systems, the definition of ioctl in the
42 original system header files is invalid for ANSI C and causes Emacs
45 The fix is to reinstall GCC, and this time do not specify --includedir
46 when you configure it. Then recompile Emacs. Specifying --includedir
47 is appropriate only in very special cases and it should *never* be the
48 same directory where system header files are kept.
50 ** Emacs does not start, complaining that it cannot open termcap database file.
52 If your system uses Terminfo rather than termcap (most modern
53 systems do), this could happen if the proper version of
54 ncurses is not visible to the Emacs configure script (i.e. it
55 cannot be found along the usual path the linker looks for
56 libraries). It can happen because your version of ncurses is
57 obsolete, or is available only in form of binaries.
59 The solution is to install an up-to-date version of ncurses in
60 the developer's form (header files, static libraries and
61 symbolic links); in some GNU/Linux distributions (e.g. Debian)
62 it constitutes a separate package.
64 ** Emacs 20 and later fails to load Lisp files at startup.
66 The typical error message might be like this:
68 "Cannot open load file: fontset"
70 This could happen if you compress the file lisp/subdirs.el. That file
71 tells Emacs what are the directories where it should look for Lisp
72 files. Emacs cannot work with subdirs.el compressed, since the
73 Auto-compress mode it needs for this will not be loaded until later,
74 when your .emacs file is processed. (The package `fontset.el' is
75 required to set up fonts used to display text on window systems, and
76 it's loaded very early in the startup procedure.)
78 Similarly, any other .el file for which there's no corresponding .elc
79 file could fail to load if it is compressed.
81 The solution is to uncompress all .el files which don't have a .elc
84 Another possible reason for such failures is stale *.elc files
85 lurking somewhere on your load-path. The following command will
86 print any duplicate Lisp files that are present in load-path:
88 emacs -q -batch -f list-load-path-shadows
90 If this command prints any file names, some of these files are stale,
91 and should be deleted or their directories removed from your
94 ** Emacs prints an error at startup after upgrading from an earlier version.
96 An example of such an error is:
98 x-complement-fontset-spec: "Wrong type argument: stringp, nil"
100 This can be another symptom of stale *.elc files in your load-path.
101 The following command will print any duplicate Lisp files that are
102 present in load-path:
104 emacs -q -batch -f list-load-path-shadows
106 If this command prints any file names, some of these files are stale,
107 and should be deleted or their directories removed from your
110 ** With X11R6.4, public-patch-3, Emacs crashes at startup.
112 Reportedly this patch in X fixes the problem.
114 --- xc/lib/X11/imInt.c~ Wed Jun 30 13:31:56 1999
115 +++ xc/lib/X11/imInt.c Thu Jul 1 15:10:27 1999
117 -/* $TOG: imInt.c /main/5 1998/05/30 21:11:16 kaleb $ */
118 +/* $TOG: imInt.c /main/5 1998/05/30 21:11:16 kaleb $ */
119 /******************************************************************
121 Copyright 1992, 1993, 1994 by FUJITSU LIMITED
128 + char* begin = NULL;
132 char* ximmodifier = XIMMODIFIER;
135 ret = Xmalloc(end - begin + 2);
137 - (void)strncpy(ret, begin, end - begin + 1);
138 + if (begin != NULL) {
139 + (void)strncpy(ret, begin, end - begin + 1);
143 ret[end - begin + 1] = '\0';
149 ** Emacs crashes in x-popup-dialog.
151 This can happen if the dialog widget cannot find the font it wants to
152 use. You can work around the problem by specifying another font with
153 an X resource--for example, `Emacs.dialog*.font: 9x15' (or any font that
154 happens to exist on your X server).
156 ** Emacs crashes when you use Bibtex mode.
158 This happens if your system puts a small limit on stack size. You can
159 prevent the problem by using a suitable shell command (often `ulimit')
160 to raise the stack size limit before you run Emacs.
162 Patches to raise the stack size limit automatically in `main'
163 (src/emacs.c) on various systems would be greatly appreciated.
165 ** Error message `Symbol's value as variable is void: x', followed by
166 a segmentation fault and core dump.
168 This has been tracked to a bug in tar! People report that tar erroneously
169 added a line like this at the beginning of files of Lisp code:
171 x FILENAME, N bytes, B tape blocks
173 If your tar has this problem, install GNU tar--if you can manage to
176 ** Crashes when displaying GIF images in Emacs built with version
177 libungif-4.1.0 are resolved by using version libungif-4.1.0b1.
178 Configure checks for the correct version, but this problem could occur
179 if a binary built against a shared libungif is run on a system with an
182 ** Emacs aborts inside the function `tparam1'.
184 This can happen if Emacs was built without terminfo support, but the
185 terminal's capabilities use format that is only supported by terminfo.
186 If your system has ncurses installed, this might happen if your
187 version of ncurses is broken; upgrading to a newer version of ncurses
188 and reconfiguring and rebuilding Emacs should solve this.
190 All modern systems support terminfo, so even if ncurses is not the
191 problem, you should look for a way to configure Emacs so that it uses
194 ** Emacs crashes when using the Exceed 6.0 X server.
196 If you are using Exceed 6.1, upgrade to a later version. This was
197 reported to prevent the crashes.
199 ** Emacs crashes with SIGSEGV in XtInitializeWidgetClass.
201 It crashes on X, but runs fine when called with option "-nw".
203 This has been observed when Emacs is linked with GNU ld but without passing
204 the -z nocombreloc flag. Emacs normally knows to pass the -z nocombreloc
205 flag when needed, so if you come across a situation where the flag is
206 necessary but missing, please report it via M-x report-emacs-bug.
208 On platforms such as Solaris, you can also work around this problem by
209 configuring your compiler to use the native linker instead of GNU ld.
211 ** Emacs compiled with Gtk+ crashes when closing a display (x-close-connection).
213 This happens because of bugs in Gtk+. Gtk+ 2.10 seems to be OK. See bug
214 http://bugzilla.gnome.org/show_bug.cgi?id=85715.
216 ** Emacs compiled with Gtk+ crashes on startup on cygwin.
218 A typical error message is
219 ***MEMORY-ERROR***: emacs[5172]: GSlice: failed to allocate 504 bytes
220 (alignment: 512): Function not implemented
222 Emacs supplies its own malloc, but glib (part of Gtk+) calls memalign and on
223 cygwin that becomes the cygwin supplied memalign. As malloc is not the
224 cygwin malloc, the cygwin memalign always returns ENOSYS. A fix for this
225 problem would be welcome.
227 * General runtime problems
231 *** Changes made to .el files do not take effect.
233 You may have forgotten to recompile them into .elc files.
234 Then the old .elc files will be loaded, and your changes
235 will not be seen. To fix this, do M-x byte-recompile-directory
236 and specify the directory that contains the Lisp files.
238 Emacs should print a warning when loading a .elc file which is older
239 than the corresponding .el file.
241 *** Watch out for .emacs files and EMACSLOADPATH environment vars.
243 These control the actions of Emacs.
244 ~/.emacs is your Emacs init file.
245 EMACSLOADPATH overrides which directories the function
248 If you observe strange problems, check for these and get rid
249 of them, then try again.
251 *** Using epop3.el package causes Emacs to signal an error.
253 The error message might be something like this:
255 "Lisp nesting exceeds max-lisp-eval-depth"
257 This happens because epop3 redefines the function gethash, which is a
258 built-in primitive beginning with Emacs 21.1. We don't have a patch
259 for epop3 that fixes this, but perhaps a newer version of epop3
262 *** Buffers from `with-output-to-temp-buffer' get set up in Help mode.
264 Changes in Emacs 20.4 to the hooks used by that function cause
265 problems for some packages, specifically BBDB. See the function's
266 documentation for the hooks involved. BBDB 2.00.06 fixes the problem.
268 *** The Hyperbole package causes *Help* buffers not to be displayed in
269 Help mode due to setting `temp-buffer-show-hook' rather than using
270 `add-hook'. Using `(add-hook 'temp-buffer-show-hook
271 'help-mode-maybe)' after loading Hyperbole should fix this.
275 *** "Compose Character" key does strange things when used as a Meta key.
277 If you define one key to serve as both Meta and Compose Character, you
278 will get strange results. In previous Emacs versions, this "worked"
279 in that the key acted as Meta--that's because the older Emacs versions
280 did not try to support Compose Character. Now Emacs tries to do
281 character composition in the standard X way. This means that you
282 must pick one meaning or the other for any given key.
284 You can use both functions (Meta, and Compose Character) if you assign
285 them to two different keys.
287 *** C-z just refreshes the screen instead of suspending Emacs.
289 You are probably using a shell that doesn't support job control, even
290 though the system itself is capable of it. Either use a different shell,
291 or set the variable `cannot-suspend' to a non-nil value.
293 *** With M-x enable-flow-control, you need to type C-\ twice
294 to do incremental search--a single C-\ gets no response.
296 This has been traced to communicating with your machine via kermit,
297 with C-\ as the kermit escape character. One solution is to use
298 another escape character in kermit. One user did
300 set escape-character 17
302 in his .kermrc file, to make C-q the kermit escape character.
304 ** Mailers and other helper programs
306 *** movemail compiled with POP support can't connect to the POP server.
308 Make sure that the `pop' entry in /etc/services, or in the services
309 NIS map if your machine uses NIS, has the same port number as the
310 entry on the POP server. A common error is for the POP server to be
311 listening on port 110, the assigned port for the POP3 protocol, while
312 the client is trying to connect on port 109, the assigned port for the
315 *** RMAIL gets error getting new mail.
317 RMAIL gets new mail from /usr/spool/mail/$USER using a program
318 called `movemail'. This program interlocks with /bin/mail using
319 the protocol defined by /bin/mail.
321 There are two different protocols in general use. One of them uses
322 the `flock' system call. The other involves creating a lock file;
323 `movemail' must be able to write in /usr/spool/mail in order to do
324 this. You control which one is used by defining, or not defining,
325 the macro MAIL_USE_FLOCK in config.h or the m- or s- file it includes.
326 IF YOU DON'T USE THE FORM OF INTERLOCKING THAT IS NORMAL ON YOUR
327 SYSTEM, YOU CAN LOSE MAIL!
329 If your system uses the lock file protocol, and fascist restrictions
330 prevent ordinary users from writing the lock files in /usr/spool/mail,
331 you may need to make `movemail' setgid to a suitable group such as
332 `mail'. To do this, use the following commands (as root) after doing the
338 Installation normally copies movemail from the build directory to an
339 installation directory which is usually under /usr/local/lib. The
340 installed copy of movemail is usually in the directory
341 /usr/local/lib/emacs/VERSION/TARGET. You must change the group and
342 mode of the installed copy; changing the group and mode of the build
343 directory copy is ineffective.
345 *** rcs2log gives you the awk error message "too many fields".
347 This is due to an arbitrary limit in certain versions of awk.
348 The solution is to use gawk (GNU awk).
350 ** Problems with hostname resolution
352 *** Emacs fails to understand most Internet host names, even though
353 the names work properly with other programs on the same system.
354 *** Emacs won't work with X-windows if the value of DISPLAY is HOSTNAME:0.
355 *** Gnus can't make contact with the specified host for nntp.
357 This typically happens on Suns and other systems that use shared
358 libraries. The cause is that the site has installed a version of the
359 shared library which uses a name server--but has not installed a
360 similar version of the unshared library which Emacs uses.
362 The result is that most programs, using the shared library, work with
363 the nameserver, but Emacs does not.
365 The fix is to install an unshared library that corresponds to what you
366 installed in the shared library, and then relink Emacs.
368 On SunOS 4.1, simply define HAVE_RES_INIT.
370 If you have already installed the name resolver in the file libresolv.a,
371 then you need to compile Emacs to use that library. The easiest way to
372 do this is to add to config.h a definition of LIBS_SYSTEM, LIBS_MACHINE
373 or LIB_STANDARD which uses -lresolv. Watch out! If you redefine a macro
374 that is already in use in your configuration to supply some other libraries,
375 be careful not to lose the others.
377 Thus, you could start by adding this to config.h:
379 #define LIBS_SYSTEM -lresolv
381 Then if this gives you an error for redefining a macro, and you see that
382 the s- file defines LIBS_SYSTEM as -lfoo -lbar, you could change config.h
385 #define LIBS_SYSTEM -lresolv -lfoo -lbar
387 *** Emacs does not know your host's fully-qualified domain name.
389 You need to configure your machine with a fully qualified domain name,
390 either in /etc/hosts, /etc/hostname, the NIS, or wherever your system
391 calls for specifying this.
393 If you cannot fix the configuration, you can set the Lisp variable
394 mail-host-address to the value you want.
398 *** Emacs says it has saved a file, but the file does not actually
401 This can happen on certain systems when you are using NFS, if the
402 remote disk is full. It is due to a bug in NFS (or certain NFS
403 implementations), and there is apparently nothing Emacs can do to
404 detect the problem. Emacs checks the failure codes of all the system
405 calls involved in writing a file, including `close'; but in the case
406 where the problem occurs, none of those system calls fails.
408 *** Editing files through RFS gives spurious "file has changed" warnings.
409 It is possible that a change in Emacs 18.37 gets around this problem,
410 but in case not, here is a description of how to fix the RFS bug that
413 There was a serious pair of bugs in the handling of the fsync() system
414 call in the RFS server.
416 The first is that the fsync() call is handled as another name for the
417 close() system call (!!). It appears that fsync() is not used by very
418 many programs; Emacs version 18 does an fsync() before closing files
419 to make sure that the bits are on the disk.
421 This is fixed by the enclosed patch to the RFS server.
423 The second, more serious problem, is that fsync() is treated as a
424 non-blocking system call (i.e., it's implemented as a message that
425 gets sent to the remote system without waiting for a reply). Fsync is
426 a useful tool for building atomic file transactions. Implementing it
427 as a non-blocking RPC call (when the local call blocks until the sync
428 is done) is a bad idea; unfortunately, changing it will break the RFS
429 protocol. No fix was supplied for this problem.
431 (as always, your line numbers may vary)
433 % rcsdiff -c -r1.2 serversyscall.c
434 RCS file: RCS/serversyscall.c,v
435 retrieving revision 1.2
436 diff -c -r1.2 serversyscall.c
437 *** /tmp/,RCSt1003677 Wed Jan 28 15:15:02 1987
438 --- serversyscall.c Wed Jan 28 15:14:48 1987
442 * No return sent for close or fsync!
444 ! if (syscall == RSYS_close || syscall == RSYS_fsync)
445 proc->p_returnval = deallocate_fd(proc, msg->m_args[0]);
450 * No return sent for close or fsync!
452 ! if (syscall == RSYS_close)
453 proc->p_returnval = deallocate_fd(proc, msg->m_args[0]);
459 *** Old versions of the PSGML package use the obsolete variables
460 `before-change-function' and `after-change-function', which are no
461 longer used by Emacs. Please use PSGML 1.2.3 or later.
463 *** PSGML conflicts with sgml-mode.
465 PSGML package uses the same names of some variables (like keymap)
466 as built-in sgml-mode.el because it was created as a replacement
467 of that package. The conflict will be shown if you load
468 sgml-mode.el before psgml.el. E.g. this could happen if you edit
469 HTML page and then start to work with SGML or XML file. html-mode
470 (from sgml-mode.el) is used for HTML file and loading of psgml.el
471 (for sgml-mode or xml-mode) will cause an error.
473 *** Versions of the PSGML package earlier than 1.0.3 (stable) or 1.1.2
474 (alpha) fail to parse DTD files correctly in Emacs 20.3 and later.
475 Here is a patch for psgml-parse.el from PSGML 1.0.1 and, probably,
478 --- psgml-parse.el 1998/08/21 19:18:18 1.1
479 +++ psgml-parse.el 1998/08/21 19:20:00
480 @@ -2383,7 +2383,7 @@ (defun sgml-push-to-entity (entity &opti
481 (setq sgml-buffer-parse-state nil))
483 ((stringp entity) ; a file name
484 - (save-excursion (insert-file-contents entity))
485 + (insert-file-contents entity)
486 (setq default-directory (file-name-directory entity)))
487 ((consp (sgml-entity-text entity)) ; external id?
488 (let* ((extid (sgml-entity-text entity))
492 You should not be using a version older than 11.52 if you can avoid
495 *** Emacs 21 freezes when visiting a TeX file with AUCTeX installed.
497 Emacs 21 needs version 10 or later of AUCTeX; upgrading should solve
500 *** No colors in AUCTeX with Emacs 21.
502 Upgrade to AUC TeX version 10 or later, and make sure it is
503 byte-compiled with Emacs 21.
507 *** Lines are not updated or new lines are added in the buffer upon commit.
509 When committing files located higher in the hierarchy than the examined
510 directory, some versions of the CVS program return an ambiguous message
511 from which PCL-CVS cannot extract the full location of the committed
512 files. As a result, the corresponding lines in the PCL-CVS buffer are
513 not updated with the new revision of these files, and new lines are
514 added to the top-level directory.
516 This can happen with CVS versions 1.12.8 and 1.12.9. Upgrade to CVS
517 1.12.10 or newer to fix this problem.
519 ** Miscellaneous problems
521 *** Self-documentation messages are garbled.
523 This means that the file `etc/DOC-...' doesn't properly correspond
524 with the Emacs executable. Redumping Emacs and then installing the
525 corresponding pair of files should fix the problem.
527 *** Programs running under terminal emulator do not recognize `emacs'
530 The cause of this is a shell startup file that sets the TERMCAP
531 environment variable. The terminal emulator uses that variable to
532 provide the information on the special terminal type that Emacs
535 Rewrite your shell startup file so that it does not change TERMCAP
536 in such a case. You could use the following conditional which sets
537 it only if it is undefined.
539 if ( ! ${?TERMCAP} ) setenv TERMCAP ~/my-termcap-file
541 Or you could set TERMCAP only when you set TERM--which should not
542 happen in a non-login shell.
544 *** In Shell mode, you get a ^M at the end of every line.
546 This happens to people who use tcsh, because it is trying to be too
547 smart. It sees that the Shell uses terminal type `unknown' and turns
548 on the flag to output ^M at the end of each line. You can fix the
549 problem by adding this to your .cshrc file:
552 if ("$EMACS" =~ /*) then
554 stty -icrnl -onlcr -echo susp ^Z
558 *** Emacs startup on GNU/Linux systems (and possibly other systems) is slow.
560 This can happen if the system is misconfigured and Emacs can't get the
561 full qualified domain name, FQDN. You should have your FQDN in the
562 /etc/hosts file, something like this:
565 129.187.137.82 nuc04.t30.physik.tu-muenchen.de nuc04
567 The way to set this up may vary on non-GNU systems.
569 *** Attempting to visit remote files via ange-ftp fails.
571 If the error message is "ange-ftp-file-modtime: Specified time is not
572 representable", then this could happen when `lukemftp' is used as the
573 ftp client. This was reported to happen on Debian GNU/Linux, kernel
574 version 2.4.3, with `lukemftp' 1.5-5, but might happen on other
575 systems as well. To avoid this problem, switch to using the standard
576 ftp client. On a Debian system, type
578 update-alternatives --config ftp
580 and then choose /usr/bin/netkit-ftp.
582 *** JPEG images aren't displayed.
584 This has been reported when Emacs is built with jpeg-6a library.
585 Upgrading to jpeg-6b solves the problem. Configure checks for the
586 correct version, but this problem could occur if a binary built
587 against a shared libjpeg is run on a system with an older version.
589 *** Dired is very slow.
591 This could happen if invocation of the `df' program takes a long
592 time. Possible reasons for this include:
594 - ClearCase mounted filesystems (VOBs) that sometimes make `df'
595 response time extremely slow (dozens of seconds);
597 - slow automounters on some old versions of Unix;
599 - slow operation of some versions of `df'.
601 To work around the problem, you could either (a) set the variable
602 `directory-free-space-program' to nil, and thus prevent Emacs from
603 invoking `df'; (b) use `df' from the GNU Fileutils package; or
604 (c) use CVS, which is Free Software, instead of ClearCase.
606 *** Versions of the W3 package released before Emacs 21.1 don't run
607 under Emacs 21. This fixed in W3 version 4.0pre.47.
609 *** The LDAP support rely on ldapsearch program from OpenLDAP version 2.
611 It can fail to work with ldapsearch program from OpenLDAP version 1.
612 Version 1 of OpenLDAP is now deprecated. If you are still using it,
613 please upgrade to version 2. As a temporary workaround, remove
614 argument "-x" from the variable `ldap-ldapsearch-args'.
616 *** ps-print commands fail to find prologue files ps-prin*.ps.
618 This can happen if you use an old version of X-Symbol package: it
619 defines compatibility functions which trick ps-print into thinking it
620 runs in XEmacs, and look for the prologue files in a wrong directory.
622 The solution is to upgrade X-Symbol to a later version.
624 *** On systems with shared libraries you might encounter run-time errors
625 from the dynamic linker telling you that it is unable to find some
626 shared libraries, for instance those for Xaw3d or image support.
627 These errors mean Emacs has been linked with a library whose shared
628 library is not in the default search path of the dynamic linker.
630 Similar problems could prevent Emacs from building, since the build
631 process invokes Emacs several times.
633 On many systems, it is possible to set LD_LIBRARY_PATH in your
634 environment to specify additional directories where shared libraries
637 Other systems allow to set LD_RUN_PATH in a similar way, but before
638 Emacs is linked. With LD_RUN_PATH set, the linker will include a
639 specified run-time search path in the executable.
641 On some systems, Emacs can crash due to problems with dynamic
642 linking. Specifically, on SGI Irix 6.5, crashes were reported with
643 backtraces like this:
646 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]
647 1 general_find_symbol(0xf49239d, 0x0, 0x0, 0x0, 0xf0000000, 0xf4923aa, 0x0, 0x492ddb2)
648 ["/comp2/mtibuild/v73/workarea/v7.3/rld/rld.c":2140, 0xfb65a98]
649 2 resolve_symbol(0xf49239d, 0x4031184, 0x0, 0xfbdd438, 0x0, 0xf4923aa, 0x0, 0x492ddb2)
650 ["/comp2/mtibuild/v73/workarea/v7.3/rld/rld.c":1947, 0xfb657e4]
651 3 lazy_text_resolve(0xd18, 0x1a3, 0x40302b4, 0x12, 0xf0000000, 0xf4923aa, 0x0, 0x492ddb2)
652 ["/comp2/mtibuild/v73/workarea/v7.3/rld/rld.c":997, 0xfb64d44]
653 4 _rld_text_resolve(0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0)
654 ["/comp2/mtibuild/v73/workarea/v7.3/rld/rld_bridge.s":175, 0xfb6032c]
656 (`rld' is the dynamic linker.) We don't know yet why this
657 happens, but setting the environment variable LD_BIND_NOW to 1 (which
658 forces the dynamic linker to bind all shared objects early on) seems
659 to work around the problem.
661 Please refer to the documentation of your dynamic linker for details.
663 *** You request inverse video, and the first Emacs frame is in inverse
664 video, but later frames are not in inverse video.
666 This can happen if you have an old version of the custom library in
667 your search path for Lisp packages. Use M-x list-load-path-shadows to
668 check whether this is true. If it is, delete the old custom library.
670 *** When you run Ispell from Emacs, it reports a "misalignment" error.
672 This can happen if you compiled the Ispell program to use ASCII
673 characters only and then try to use it from Emacs with non-ASCII
674 characters, like Latin-1. The solution is to recompile Ispell with
675 support for 8-bit characters.
677 To see whether your Ispell program supports 8-bit characters, type
678 this at your shell's prompt:
682 and look in the output for the string "NO8BIT". If Ispell says
683 "!NO8BIT (8BIT)", your speller supports 8-bit characters; otherwise it
686 To rebuild Ispell with 8-bit character support, edit the local.h file
687 in the Ispell distribution and make sure it does _not_ define NO8BIT.
688 Then rebuild the speller.
690 Another possible cause for "misalignment" error messages is that the
691 version of Ispell installed on your machine is old. Upgrade.
693 Yet another possibility is that you are trying to spell-check a word
694 in a language that doesn't fit the dictionary you choose for use by
695 Ispell. (Ispell can only spell-check one language at a time, because
696 it uses a single dictionary.) Make sure that the text you are
697 spelling and the dictionary used by Ispell conform to each other.
699 If your spell-checking program is Aspell, it has been reported that if
700 you have a personal configuration file (normally ~/.aspell.conf), it
701 can cause this error. Remove that file, execute `ispell-kill-ispell'
702 in Emacs, and then try spell-checking again.
704 * Runtime problems related to font handling
706 ** Under X11, some characters appear as hollow boxes.
708 Each X11 font covers just a fraction of the characters that Emacs
709 supports. To display the whole range of Emacs characters requires
710 many different fonts, collected into a fontset.
712 If some of the fonts called for in your fontset do not exist on your X
713 server, then the characters that have no font appear as hollow boxes.
714 You can remedy the problem by installing additional fonts.
716 The intlfonts distribution includes a full spectrum of fonts that can
717 display all the characters Emacs supports. The etl-unicode collection
718 of fonts (available from <URL:ftp://ftp.x.org/contrib/fonts/> and
719 <URL:ftp://ftp.xfree86.org/pub/mirror/X.Org/contrib/fonts/>) includes
720 fonts that can display many Unicode characters; they can also be used
721 by ps-print and ps-mule to print Unicode characters.
723 Another cause of this for specific characters is fonts which have a
724 missing glyph and no default character. This is known to occur for
725 character number 160 (no-break space) in some fonts, such as Lucida
726 but Emacs sets the display table for the unibyte and Latin-1 version
727 of this character to display a space.
729 ** Under X11, some characters appear improperly aligned in their lines.
731 You may have bad X11 fonts; try installing the intlfonts distribution
732 or the etl-unicode collection (see the previous entry).
734 ** Certain fonts make each line take one pixel more than it "should".
736 This is because these fonts contain characters a little taller
737 than the font's nominal height. Emacs needs to make sure that
738 lines do not overlap.
740 ** Loading fonts is very slow.
742 You might be getting scalable fonts instead of precomputed bitmaps.
743 Known scalable font directories are "Type1" and "Speedo". A font
744 directory contains scalable fonts if it contains the file
747 If this is so, re-order your X windows font path to put the scalable
748 font directories last. See the documentation of `xset' for details.
750 With some X servers, it may be necessary to take the scalable font
751 directories out of your path entirely, at least for Emacs 19.26.
752 Changes in the future may make this unnecessary.
754 ** Font Lock displays portions of the buffer in incorrect faces.
756 By far the most frequent cause of this is a parenthesis `(' or a brace
757 `{' in column zero. Font Lock assumes that such a paren is outside of
758 any comment or string. This is of course not true in general, but the
759 vast majority of well-formatted program source files don't have such
760 parens, and therefore this assumption is used to allow optimizations
761 in Font Lock's syntactical analysis. These optimizations avoid some
762 pathological cases where jit-lock, the Just-in-Time fontification
763 introduced with Emacs 21.1, could significantly slow down scrolling
764 through the buffer, especially scrolling backwards, and also jumping
765 to the end of a very large buffer.
767 Beginning with version 22.1, a parenthesis or a brace in column zero
768 is highlighted in bold-red face if it is inside a string or a comment,
769 to indicate that it could interfere with Font Lock (and also with
770 indentation) and should be moved or escaped with a backslash.
772 If you don't use large buffers, or have a very fast machine which
773 makes the delays insignificant, you can avoid the incorrect
774 fontification by setting the variable
775 `font-lock-beginning-of-syntax-function' to a nil value. (This must
776 be done _after_ turning on Font Lock.)
778 Another alternative is to avoid a paren in column zero. For example,
779 in a Lisp string you could precede the paren with a backslash.
781 ** With certain fonts, when the cursor appears on a character, the
782 character doesn't appear--you get a solid box instead.
784 One user on a Linux-based GNU system reported that this problem went
785 away with installation of a new X server. The failing server was
786 XFree86 3.1.1. XFree86 3.1.2 works.
788 ** Characters are displayed as empty boxes or with wrong font under X.
790 This can occur when two different versions of FontConfig are used.
791 For example, XFree86 4.3.0 has one version and Gnome usually comes
792 with a newer version. Emacs compiled with --with-gtk will then use
793 the newer version. In most cases the problem can be temporarily
794 fixed by stopping the application that has the error (it can be
795 Emacs or any other application), removing ~/.fonts.cache-1,
796 and then start the application again.
797 If removing ~/.fonts.cache-1 and restarting doesn't help, the
798 application with problem must be recompiled with the same version
799 of FontConfig as the rest of the system uses. For KDE, it is
800 sufficient to recompile Qt.
802 ** Emacs pauses for several seconds when changing the default font.
804 This has been reported for fvwm 2.2.5 and the window manager of KDE
805 2.1. The reason for the pause is Xt waiting for a ConfigureNotify
806 event from the window manager, which the window manager doesn't send.
807 Xt stops waiting after a default timeout of usually 5 seconds.
809 A workaround for this is to add something like
811 emacs.waitForWM: false
813 to your X resources. Alternatively, add `(wait-for-wm . nil)' to a
814 frame's parameter list, like this:
816 (modify-frame-parameters nil '((wait-for-wm . nil)))
818 (this should go into your `.emacs' file).
820 ** Underlines appear at the wrong position.
822 This is caused by fonts having a wrong UNDERLINE_POSITION property.
823 Examples are the font 7x13 on XFree prior to version 4.1, or the jmk
824 neep font from the Debian xfonts-jmk package. To circumvent this
825 problem, set x-use-underline-position-properties to nil in your
828 To see what is the value of UNDERLINE_POSITION defined by the font,
829 type `xlsfonts -lll FONT' and look at the font's UNDERLINE_POSITION
832 ** When using Exceed, fonts sometimes appear too tall.
834 When the display is set to an Exceed X-server and fonts are specified
835 (either explicitly with the -fn option or implicitly with X resources)
836 then the fonts may appear "too tall". The actual character sizes are
837 correct but there is too much vertical spacing between rows, which
838 gives the appearance of "double spacing".
840 To prevent this, turn off the Exceed's "automatic font substitution"
841 feature (in the font part of the configuration window).
843 * Internationalization problems
845 ** M-{ does not work on a Spanish PC keyboard.
847 Many Spanish keyboards seem to ignore that combination. Emacs can't
848 do anything about it.
850 ** Characters from the mule-unicode charsets aren't displayed under X.
852 XFree86 4 contains many fonts in iso10646-1 encoding which have
853 minimal character repertoires (whereas the encoding part of the font
854 name is meant to be a reasonable indication of the repertoire
855 according to the XLFD spec). Emacs may choose one of these to display
856 characters from the mule-unicode charsets and then typically won't be
857 able to find the glyphs to display many characters. (Check with C-u
858 C-x = .) To avoid this, you may need to use a fontset which sets the
859 font for the mule-unicode sets explicitly. E.g. to use GNU unifont,
860 include in the fontset spec:
862 mule-unicode-2500-33ff:-gnu-unifont-*-iso10646-1,\
863 mule-unicode-e000-ffff:-gnu-unifont-*-iso10646-1,\
864 mule-unicode-0100-24ff:-gnu-unifont-*-iso10646-1
866 ** The UTF-8/16/7 coding systems don't encode CJK (Far Eastern) characters.
868 Emacs directly supports the Unicode BMP whose code points are in the
869 ranges 0000-33ff and e000-ffff, and indirectly supports the parts of
870 CJK characters belonging to these legacy charsets:
872 GB2312, Big5, JISX0208, JISX0212, JISX0213-1, JISX0213-2, KSC5601
874 The latter support is done in Utf-Translate-Cjk mode (turned on by
875 default). Which Unicode CJK characters are decoded into which Emacs
876 charset is decided by the current language environment. For instance,
877 in Chinese-GB, most of them are decoded into chinese-gb2312.
879 If you read UTF-8 data with code points outside these ranges, the
880 characters appear in the buffer as raw bytes of the original UTF-8
881 (composed into a single quasi-character) and they will be written back
882 correctly as UTF-8, assuming you don't break the composed sequences.
883 If you read such characters from UTF-16 or UTF-7 data, they are
884 substituted with the Unicode `replacement character', and you lose
887 ** Mule-UCS loads very slowly.
889 Changes to Emacs internals interact badly with Mule-UCS's `un-define'
890 library, which is the usual interface to Mule-UCS. Apply the
891 following patch to Mule-UCS 0.84 and rebuild it. That will help,
892 though loading will still be slower than in Emacs 20. (Some
893 distributions, such as Debian, may already have applied such a patch.)
895 --- lisp/un-define.el 6 Mar 2001 22:41:38 -0000 1.30
896 +++ lisp/un-define.el 19 Apr 2002 18:34:26 -0000
897 @@ -610,13 +624,21 @@ by calling post-read-conversion and pre-
903 - (mucs-define-coding-system
904 - (nth 0 y) (nth 1 y) (nth 2 y)
905 - (nth 3 y) (nth 4 y) (nth 5 y) (nth 6 y))
906 - (coding-system-put (car y) 'alias-coding-systems (list (car x))))
908 + (if (fboundp 'register-char-codings)
909 + ;; Mule 5, where we don't need the eol-type specified and
910 + ;; register-char-codings may be very slow for these coding
911 + ;; system definitions.
912 + (let ((y (cadr x)))
913 + (mucs-define-coding-system
914 + (car x) (nth 1 y) (nth 2 y)
915 + (nth 3 y) (nth 4 y) (nth 5 y)))
918 + (mucs-define-coding-system
919 + (nth 0 y) (nth 1 y) (nth 2 y)
920 + (nth 3 y) (nth 4 y) (nth 5 y) (nth 6 y))
921 + (coding-system-put (car y) 'alias-coding-systems (list (car x)))))
925 ?u "UTF-8 coding system"
927 Note that Emacs has native support for Unicode, roughly equivalent to
928 Mule-UCS's, so you may not need it.
930 ** Mule-UCS compilation problem.
932 Emacs of old versions and XEmacs byte-compile the form `(progn progn
933 ...)' the same way as `(progn ...)', but Emacs of version 21.3 and the
934 later process that form just as interpreter does, that is, as `progn'
935 variable reference. Apply the following patch to Mule-UCS 0.84 to
936 make it compiled by the latest Emacs.
938 --- mucs-ccl.el 2 Sep 2005 00:42:23 -0000 1.1.1.1
939 +++ mucs-ccl.el 2 Sep 2005 01:31:51 -0000 1.3
940 @@ -639,10 +639,14 @@
941 (mucs-notify-embedment 'mucs-ccl-required name)
942 (setq ccl-pgm-list (cdr ccl-pgm-list)))
943 ; (message "MCCLREGFIN:%S" result)
945 - (setq mucs-ccl-facility-alist
946 - (quote ,mucs-ccl-facility-alist))
948 + ;; The only way the function is used in this package is included
949 + ;; in `mucs-package-definition-end-hook' value, where it must
950 + ;; return (possibly empty) *list* of forms. Do this. Do not rely
951 + ;; on byte compiler to remove extra `progn's in `(progn ...)'
953 + `((setq mucs-ccl-facility-alist
954 + (quote ,mucs-ccl-facility-alist))
957 ;;; Add hook for embedding translation informations to a package.
958 (add-hook 'mucs-package-definition-end-hook
960 ** Accented ISO-8859-1 characters are displayed as | or _.
962 Try other font set sizes (S-mouse-1). If the problem persists with
963 other sizes as well, your text is corrupted, probably through software
964 that is not 8-bit clean. If the problem goes away with another font
965 size, it's probably because some fonts pretend to be ISO-8859-1 fonts
966 when they are really ASCII fonts. In particular the schumacher-clean
967 fonts have this bug in some versions of X.
969 To see what glyphs are included in a font, use `xfd', like this:
971 xfd -fn -schumacher-clean-medium-r-normal--12-120-75-75-c-60-iso8859-1
973 If this shows only ASCII glyphs, the font is indeed the source of the
976 The solution is to remove the corresponding lines from the appropriate
977 `fonts.alias' file, then run `mkfontdir' in that directory, and then run
980 ** The `oc-unicode' package doesn't work with Emacs 21.
982 This package tries to define more private charsets than there are free
983 slots now. The current built-in Unicode support is actually more
984 flexible. (Use option `utf-translate-cjk-mode' if you need CJK
985 support.) Files encoded as emacs-mule using oc-unicode aren't
986 generally read correctly by Emacs 21.
988 ** After a while, Emacs slips into unibyte mode.
990 The VM mail package, which is not part of Emacs, sometimes does
991 (standard-display-european t)
992 That should be changed to
993 (standard-display-european 1 t)
997 ** X keyboard problems
999 *** You "lose characters" after typing Compose Character key.
1001 This is because the Compose Character key is defined as the keysym
1002 Multi_key, and Emacs (seeing that) does the proper X11
1003 character-composition processing. If you don't want your Compose key
1004 to do that, you can redefine it with xmodmap.
1006 For example, here's one way to turn it into a Meta key:
1008 xmodmap -e "keysym Multi_key = Meta_L"
1010 If all users at your site of a particular keyboard prefer Meta to
1011 Compose, you can make the remapping happen automatically by adding the
1012 xmodmap command to the xdm setup script for that display.
1014 *** Using X Windows, control-shift-leftbutton makes Emacs hang.
1016 Use the shell command `xset bc' to make the old X Menu package work.
1018 *** C-SPC fails to work on Fedora GNU/Linux (or with fcitx input method).
1020 Fedora Core 4 steals the C-SPC key by default for the `iiimx' program
1021 which is the input method for some languages. It blocks Emacs users
1022 from using the C-SPC key for `set-mark-command'.
1024 One solutions is to remove the `<Ctrl>space' from the `Iiimx' file
1025 which can be found in the `/usr/lib/X11/app-defaults' directory.
1026 However, that requires root access.
1028 Another is to specify `Emacs*useXIM: false' in your X resources.
1030 Another is to build Emacs with the `--without-xim' configure option.
1032 The same problem happens on any other system if you are using fcitx
1033 (Chinese input method) which by default use C-SPC for toggling. If
1034 you want to use fcitx with Emacs, you have two choices. Toggle fcitx
1035 by another key (e.g. C-\) by modifying ~/.fcitx/config, or be
1036 accustomed to use C-@ for `set-mark-command'.
1038 *** M-SPC seems to be ignored as input.
1040 See if your X server is set up to use this as a command
1041 for character composition.
1043 *** The S-C-t key combination doesn't get passed to Emacs on X.
1045 This happens because some X configurations assign the Ctrl-Shift-t
1046 combination the same meaning as the Multi_key. The offending
1047 definition is in the file `...lib/X11/locale/iso8859-1/Compose'; there
1048 might be other similar combinations which are grabbed by X for similar
1051 We think that this can be countermanded with the `xmodmap' utility, if
1052 you want to be able to bind one of these key sequences within Emacs.
1054 *** Under X, C-v and/or other keys don't work.
1056 These may have been intercepted by your window manager. In
1057 particular, AfterStep 1.6 is reported to steal C-v in its default
1058 configuration. Various Meta keys are also likely to be taken by the
1059 configuration of the `feel'. See the WM's documentation for how to
1062 *** Clicking C-mouse-2 in the scroll bar doesn't split the window.
1064 This currently doesn't work with scroll-bar widgets (and we don't know
1065 a good way of implementing it with widgets). If Emacs is configured
1066 --without-toolkit-scroll-bars, C-mouse-2 on the scroll bar does work.
1068 *** Inability to send an Alt-modified key, when Emacs is communicating
1069 directly with an X server.
1071 If you have tried to bind an Alt-modified key as a command, and it
1072 does not work to type the command, the first thing you should check is
1073 whether the key is getting through to Emacs. To do this, type C-h c
1074 followed by the Alt-modified key. C-h c should say what kind of event
1075 it read. If it says it read an Alt-modified key, then make sure you
1076 have made the key binding correctly.
1078 If C-h c reports an event that doesn't have the Alt modifier, it may
1079 be because your X server has no key for the Alt modifier. The X
1080 server that comes from MIT does not set up the Alt modifier by
1083 If your keyboard has keys named Alt, you can enable them as follows:
1085 xmodmap -e 'add mod2 = Alt_L'
1086 xmodmap -e 'add mod2 = Alt_R'
1088 If the keyboard has just one key named Alt, then only one of those
1089 commands is needed. The modifier `mod2' is a reasonable choice if you
1090 are using an unmodified MIT version of X. Otherwise, choose any
1091 modifier bit not otherwise used.
1093 If your keyboard does not have keys named Alt, you can use some other
1094 keys. Use the keysym command in xmodmap to turn a function key (or
1095 some other 'spare' key) into Alt_L or into Alt_R, and then use the
1096 commands show above to make them modifier keys.
1098 Note that if you have Alt keys but no Meta keys, Emacs translates Alt
1099 into Meta. This is because of the great importance of Meta in Emacs.
1101 ** Window-manager and toolkit-related problems
1103 *** Gnome: Emacs receives input directly from the keyboard, bypassing XIM.
1105 This seems to happen when gnome-settings-daemon version 2.12 or later
1106 is running. If gnome-settings-daemon is not running, Emacs receives
1107 input through XIM without any problem. Furthermore, this seems only
1108 to happen in *.UTF-8 locales; zh_CN.GB2312 and zh_CN.GBK locales, for
1109 example, work fine. A bug report has been filed in the Gnome
1110 bugzilla: http://bugzilla.gnome.org/show_bug.cgi?id=357032
1112 *** Gnome: Emacs' xterm-mouse-mode doesn't work on the Gnome terminal.
1114 A symptom of this bug is that double-clicks insert a control sequence
1115 into the buffer. The reason this happens is an apparent
1116 incompatibility of the Gnome terminal with Xterm, which also affects
1117 other programs using the Xterm mouse interface. A problem report has
1120 *** KDE: When running on KDE, colors or fonts are not as specified for Emacs,
1123 For example, you could see background you set for Emacs only in the
1124 empty portions of the Emacs display, while characters have some other
1127 This happens because KDE's defaults apply its color and font
1128 definitions even to applications that weren't compiled for KDE. The
1129 solution is to uncheck the "Apply fonts and colors to non-KDE apps"
1130 option in Preferences->Look&Feel->Style (KDE 2). In KDE 3, this option
1131 is in the "Colors" section, rather than "Style".
1133 Alternatively, if you do want the KDE defaults to apply to other
1134 applications, but not to Emacs, you could modify the file `Emacs.ad'
1135 (should be in the `/usr/share/apps/kdisplay/app-defaults/' directory)
1136 so that it doesn't set the default background and foreground only for
1137 Emacs. For example, make sure the following resources are either not
1138 present or commented out:
1140 Emacs.default.attributeForeground
1141 Emacs.default.attributeBackground
1145 *** KDE: Emacs hangs on KDE when a large portion of text is killed.
1147 This is caused by a bug in the KDE applet `klipper' which periodically
1148 requests the X clipboard contents from applications. Early versions
1149 of klipper don't implement the ICCCM protocol for large selections,
1150 which leads to Emacs being flooded with selection requests. After a
1151 while, Emacs may print a message:
1153 Timed out waiting for property-notify event
1155 A workaround is to not use `klipper'. An upgrade to the `klipper' that
1156 comes with KDE 3.3 or later also solves the problem.
1158 *** CDE: Frames may cover dialogs they created when using CDE.
1160 This can happen if you have "Allow Primary Windows On Top" enabled which
1161 seems to be the default in the Common Desktop Environment.
1162 To change, go in to "Desktop Controls" -> "Window Style Manager"
1163 and uncheck "Allow Primary Windows On Top".
1165 *** Xaw3d : When using Xaw3d scroll bars without arrows, the very first mouse
1166 click in a scroll bar might be ignored by the scroll bar widget. This
1167 is probably a bug in Xaw3d; when Xaw3d is compiled with arrows, the
1170 *** Xaw: There are known binary incompatibilities between Xaw, Xaw3d, neXtaw,
1171 XawM and the few other derivatives of Xaw. So when you compile with
1172 one of these, it may not work to dynamically link with another one.
1173 For example, strange problems, such as Emacs exiting when you type
1174 "C-x 1", were reported when Emacs compiled with Xaw3d and libXaw was
1175 used with neXtaw at run time.
1177 The solution is to rebuild Emacs with the toolkit version you actually
1178 want to use, or set LD_PRELOAD to preload the same toolkit version you
1181 *** Open Motif: Problems with file dialogs in Emacs built with Open Motif.
1183 When Emacs 21 is built with Open Motif 2.1, it can happen that the
1184 graphical file dialog boxes do not work properly. The "OK", "Filter"
1185 and "Cancel" buttons do not respond to mouse clicks. Dragging the
1186 file dialog window usually causes the buttons to work again.
1188 The solution is to use LessTif instead. LessTif is a free replacement
1189 for Motif. See the file INSTALL for information on how to do this.
1191 Another workaround is not to use the mouse to trigger file prompts,
1192 but to use the keyboard. This way, you will be prompted for a file in
1193 the minibuffer instead of a graphical file dialog.
1195 *** LessTif: Problems in Emacs built with LessTif.
1197 The problems seem to depend on the version of LessTif and the Motif
1198 emulation for which it is set up.
1200 Only the Motif 1.2 emulation seems to be stable enough in LessTif.
1201 Lesstif 0.92-17's Motif 1.2 emulation seems to work okay on FreeBSD.
1202 On GNU/Linux systems, lesstif-0.92.6 configured with "./configure
1203 --enable-build-12 --enable-default-12" is reported to be the most
1204 successful. The binary GNU/Linux package
1205 lesstif-devel-0.92.0-1.i386.rpm was reported to have problems with
1208 On some systems, even with Motif 1.2 emulation, Emacs occasionally
1209 locks up, grabbing all mouse and keyboard events. We still don't know
1210 what causes these problems; they are not reproducible by Emacs
1213 *** Motif: The Motif version of Emacs paints the screen a solid color.
1215 This has been observed to result from the following X resource:
1217 Emacs*default.attributeFont: -*-courier-medium-r-*-*-*-140-*-*-*-*-iso8859-*
1219 That the resource has this effect indicates a bug in something, but we
1220 do not yet know what. If it is an Emacs bug, we hope someone can
1221 explain what the bug is so we can fix it. In the mean time, removing
1222 the resource prevents the problem.
1224 ** General X problems
1226 *** Redisplay using X11 is much slower than previous Emacs versions.
1228 We've noticed that certain X servers draw the text much slower when
1229 scroll bars are on the left. We don't know why this happens. If this
1230 happens to you, you can work around it by putting the scroll bars
1231 on the right (as they were in Emacs 19).
1233 Here's how to do this:
1235 (set-scroll-bar-mode 'right)
1237 If you're not sure whether (or how much) this problem affects you,
1238 try that and see how much difference it makes. To set things back
1241 (set-scroll-bar-mode 'left)
1243 *** Error messages about undefined colors on X.
1245 The messages might say something like this:
1247 Unable to load color "grey95"
1249 (typically, in the `*Messages*' buffer), or something like this:
1251 Error while displaying tooltip: (error Undefined color lightyellow)
1253 These problems could happen if some other X program has used up too
1254 many colors of the X palette, leaving Emacs with insufficient system
1255 resources to load all the colors it needs.
1257 A solution is to exit the offending X programs before starting Emacs.
1259 "undefined color" messages can also occur if the RgbPath entry in the
1260 X configuration file is incorrect, or the rgb.txt file is not where
1261 X expects to find it.
1263 *** Improving performance with slow X connections.
1265 There are several ways to improve this performance, any subset of which can
1266 be carried out at the same time:
1268 1) If you don't need X Input Methods (XIM) for entering text in some
1269 language you use, you can improve performance on WAN links by using
1270 the X resource useXIM to turn off use of XIM. This does not affect
1271 the use of Emacs' own input methods, which are part of the Leim
1274 2) If the connection is very slow, you might also want to consider
1275 switching off scroll bars, menu bar, and tool bar. Adding the
1276 following forms to your .emacs file will accomplish that, but only
1277 after the the initial frame is displayed:
1279 (scroll-bar-mode -1)
1283 For still quicker startup, put these X resources in your .Xdefaults
1286 Emacs.verticalScrollBars: off
1290 3) Use ssh to forward the X connection, and enable compression on this
1291 forwarded X connection (ssh -XC remotehostname emacs ...).
1293 4) Use lbxproxy on the remote end of the connection. This is an interface
1294 to the low bandwidth X extension in most modern X servers, which
1295 improves performance dramatically, at the slight expense of correctness
1296 of the X protocol. lbxproxy acheives the performance gain by grouping
1297 several X requests in one TCP packet and sending them off together,
1298 instead of requiring a round-trip for each X request in a seperate
1299 packet. The switches that seem to work best for emacs are:
1300 -noatomsfile -nowinattr -cheaterrors -cheatevents
1301 Note that the -nograbcmap option is known to cause problems.
1302 For more about lbxproxy, see:
1303 http://www.xfree86.org/4.3.0/lbxproxy.1.html
1305 5) If copying and killing is slow, try to disable the interaction with the
1306 native system's clipboard by adding these lines to your .emacs file:
1307 (setq interprogram-cut-function nil)
1308 (setq interprogram-paste-function nil)
1310 *** Emacs gives the error, Couldn't find per display information.
1312 This can result if the X server runs out of memory because Emacs uses
1313 a large number of fonts. On systems where this happens, C-h h is
1316 We do not know of a way to prevent the problem.
1318 *** Emacs does not notice when you release the mouse.
1320 There are reports that this happened with (some) Microsoft mice and
1321 that replacing the mouse made it stop.
1323 *** You can't select from submenus (in the X toolkit version).
1325 On certain systems, mouse-tracking and selection in top-level menus
1326 works properly with the X toolkit, but neither of them works when you
1327 bring up a submenu (such as Bookmarks or Compare or Apply Patch, in
1330 This works on most systems. There is speculation that the failure is
1331 due to bugs in old versions of X toolkit libraries, but no one really
1332 knows. If someone debugs this and finds the precise cause, perhaps a
1333 workaround can be found.
1335 *** An error message such as `X protocol error: BadMatch (invalid
1336 parameter attributes) on protocol request 93'.
1338 This comes from having an invalid X resource, such as
1340 (which is invalid because it specifies a color name for something
1341 that isn't a color.)
1343 The fix is to correct your X resources.
1345 *** Slow startup on X11R6 with X windows.
1347 If Emacs takes two minutes to start up on X11R6, see if your X
1348 resources specify any Adobe fonts. That causes the type-1 font
1349 renderer to start up, even if the font you asked for is not a type-1
1352 One way to avoid this problem is to eliminate the type-1 fonts from
1353 your font path, like this:
1355 xset -fp /usr/X11R6/lib/X11/fonts/Type1/
1357 *** Pull-down menus appear in the wrong place, in the toolkit version of Emacs.
1359 An X resource of this form can cause the problem:
1361 Emacs*geometry: 80x55+0+0
1363 This resource is supposed to apply, and does apply, to the menus
1364 individually as well as to Emacs frames. If that is not what you
1365 want, rewrite the resource.
1367 To check thoroughly for such resource specifications, use `xrdb
1368 -query' to see what resources the X server records, and also look at
1369 the user's ~/.Xdefaults and ~/.Xdefaults-* files.
1371 *** Emacs running under X Windows does not handle mouse clicks.
1372 *** `emacs -geometry 80x20' finds a file named `80x20'.
1374 One cause of such problems is having (setq term-file-prefix nil) in
1375 your .emacs file. Another cause is a bad value of EMACSLOADPATH in
1378 *** Emacs fails to get default settings from X Windows server.
1380 The X library in X11R4 has a bug; it interchanges the 2nd and 3rd
1381 arguments to XGetDefaults. Define the macro XBACKWARDS in config.h to
1382 tell Emacs to compensate for this.
1384 I don't believe there is any way Emacs can determine for itself
1385 whether this problem is present on a given system.
1387 *** X Windows doesn't work if DISPLAY uses a hostname.
1389 People have reported kernel bugs in certain systems that cause Emacs
1390 not to work with X Windows if DISPLAY is set using a host name. But
1391 the problem does not occur if DISPLAY is set to `unix:0.0'. I think
1392 the bug has to do with SIGIO or FIONREAD.
1394 You may be able to compensate for the bug by doing (set-input-mode nil nil).
1395 However, that has the disadvantage of turning off interrupts, so that
1396 you are unable to quit out of a Lisp program by typing C-g.
1398 The easy way to do this is to put
1400 (setq x-sigio-bug t)
1402 in your site-init.el file.
1404 * Runtime problems on character termunals
1406 ** Emacs spontaneously displays "I-search: " at the bottom of the screen.
1408 This means that Control-S/Control-Q (XON/XOFF) "flow control" is being
1409 used. C-s/C-q flow control is bad for Emacs editors because it takes
1410 away C-s and C-q as user commands. Since editors do not output long
1411 streams of text without user commands, there is no need for a
1412 user-issuable "stop output" command in an editor; therefore, a
1413 properly designed flow control mechanism would transmit all possible
1414 input characters without interference. Designing such a mechanism is
1415 easy, for a person with at least half a brain.
1417 There are three possible reasons why flow control could be taking place:
1419 1) Terminal has not been told to disable flow control
1420 2) Insufficient padding for the terminal in use
1421 3) Some sort of terminal concentrator or line switch is responsible
1423 First of all, many terminals have a set-up mode which controls whether
1424 they generate XON/XOFF flow control characters. This must be set to
1425 "no XON/XOFF" in order for Emacs to work. Sometimes there is an
1426 escape sequence that the computer can send to turn flow control off
1427 and on. If so, perhaps the termcap `ti' string should turn flow
1428 control off, and the `te' string should turn it on.
1430 Once the terminal has been told "no flow control", you may find it
1431 needs more padding. The amount of padding Emacs sends is controlled
1432 by the termcap entry for the terminal in use, and by the output baud
1433 rate as known by the kernel. The shell command `stty' will print
1434 your output baud rate; `stty' with suitable arguments will set it if
1435 it is wrong. Setting to a higher speed causes increased padding. If
1436 the results are wrong for the correct speed, there is probably a
1437 problem in the termcap entry. You must speak to a local Unix wizard
1438 to fix this. Perhaps you are just using the wrong terminal type.
1440 For terminals that lack a "no flow control" mode, sometimes just
1441 giving lots of padding will prevent actual generation of flow control
1442 codes. You might as well try it.
1444 If you are really unlucky, your terminal is connected to the computer
1445 through a concentrator which sends XON/XOFF flow control to the
1446 computer, or it insists on sending flow control itself no matter how
1447 much padding you give it. Unless you can figure out how to turn flow
1448 control off on this concentrator (again, refer to your local wizard),
1449 you are screwed! You should have the terminal or concentrator
1450 replaced with a properly designed one. In the mean time, some drastic
1451 measures can make Emacs semi-work.
1453 You can make Emacs ignore C-s and C-q and let the operating system
1454 handle them. To do this on a per-session basis, just type M-x
1455 enable-flow-control RET. You will see a message that C-\ and C-^ are
1456 now translated to C-s and C-q. (Use the same command M-x
1457 enable-flow-control to turn *off* this special mode. It toggles flow
1460 If C-\ and C-^ are inconvenient for you (for example, if one of them
1461 is the escape character of your terminal concentrator), you can choose
1462 other characters by setting the variables flow-control-c-s-replacement
1463 and flow-control-c-q-replacement. But choose carefully, since all
1464 other control characters are already used by emacs.
1466 IMPORTANT: if you type C-s by accident while flow control is enabled,
1467 Emacs output will freeze, and you will have to remember to type C-q in
1470 If you work in an environment where a majority of terminals of a
1471 certain type are flow control hobbled, you can use the function
1472 `enable-flow-control-on' to turn on this flow control avoidance scheme
1473 automatically. Here is an example:
1475 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
1477 If this isn't quite correct (e.g. you have a mixture of flow-control hobbled
1478 and good vt200 terminals), you can still run enable-flow-control
1481 I have no intention of ever redesigning the Emacs command set for the
1482 assumption that terminals use C-s/C-q flow control. XON/XOFF flow
1483 control technique is a bad design, and terminals that need it are bad
1484 merchandise and should not be purchased. Now that X is becoming
1485 widespread, XON/XOFF seems to be on the way out. If you can get some
1486 use out of GNU Emacs on inferior terminals, more power to you, but I
1487 will not make Emacs worse for properly designed systems for the sake
1488 of inferior systems.
1490 ** Control-S and Control-Q commands are ignored completely.
1492 For some reason, your system is using brain-damaged C-s/C-q flow
1493 control despite Emacs's attempts to turn it off. Perhaps your
1494 terminal is connected to the computer through a concentrator
1495 that wants to use flow control.
1497 You should first try to tell the concentrator not to use flow control.
1498 If you succeed in this, try making the terminal work without
1499 flow control, as described in the preceding section.
1501 If that line of approach is not successful, map some other characters
1502 into C-s and C-q using keyboard-translate-table. The example above
1503 shows how to do this with C-^ and C-\.
1505 ** Screen is updated wrong, but only on one kind of terminal.
1507 This could mean that the termcap entry you are using for that
1508 terminal is wrong, or it could mean that Emacs has a bug handing
1509 the combination of features specified for that terminal.
1511 The first step in tracking this down is to record what characters
1512 Emacs is sending to the terminal. Execute the Lisp expression
1513 (open-termscript "./emacs-script") to make Emacs write all
1514 terminal output into the file ~/emacs-script as well; then do
1515 what makes the screen update wrong, and look at the file
1516 and decode the characters using the manual for the terminal.
1517 There are several possibilities:
1519 1) The characters sent are correct, according to the terminal manual.
1521 In this case, there is no obvious bug in Emacs, and most likely you
1522 need more padding, or possibly the terminal manual is wrong.
1524 2) The characters sent are incorrect, due to an obscure aspect
1525 of the terminal behavior not described in an obvious way
1528 This case is hard. It will be necessary to think of a way for
1529 Emacs to distinguish between terminals with this kind of behavior
1530 and other terminals that behave subtly differently but are
1531 classified the same by termcap; or else find an algorithm for
1532 Emacs to use that avoids the difference. Such changes must be
1533 tested on many kinds of terminals.
1535 3) The termcap entry is wrong.
1537 See the file etc/TERMS for information on changes
1538 that are known to be needed in commonly used termcap entries
1539 for certain terminals.
1541 4) The characters sent are incorrect, and clearly cannot be
1542 right for any terminal with the termcap entry you were using.
1544 This is unambiguously an Emacs bug, and can probably be fixed
1545 in termcap.c, tparam.c, term.c, scroll.c, cm.c or dispnew.c.
1547 ** Control-S and Control-Q commands are ignored completely on a net connection.
1549 Some versions of rlogin (and possibly telnet) do not pass flow
1550 control characters to the remote system to which they connect.
1551 On such systems, emacs on the remote system cannot disable flow
1552 control on the local system.
1554 One way to cure this is to disable flow control on the local host
1555 (the one running rlogin, not the one running rlogind) using the
1556 stty command, before starting the rlogin process. On many systems,
1557 "stty start u stop u" will do this.
1559 Some versions of tcsh will prevent even this from working. One way
1560 around this is to start another shell before starting rlogin, and
1561 issue the stty command to disable flow control from that shell.
1563 If none of these methods work, the best solution is to type
1564 M-x enable-flow-control at the beginning of your emacs session, or
1565 if you expect the problem to continue, add a line such as the
1566 following to your .emacs (on the host running rlogind):
1568 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
1570 See the entry about spontaneous display of I-search (above) for more
1573 ** Output from Control-V is slow.
1575 On many bit-map terminals, scrolling operations are fairly slow.
1576 Often the termcap entry for the type of terminal in use fails
1577 to inform Emacs of this. The two lines at the bottom of the screen
1578 before a Control-V command are supposed to appear at the top after
1579 the Control-V command. If Emacs thinks scrolling the lines is fast,
1580 it will scroll them to the top of the screen.
1582 If scrolling is slow but Emacs thinks it is fast, the usual reason is
1583 that the termcap entry for the terminal you are using does not
1584 specify any padding time for the `al' and `dl' strings. Emacs
1585 concludes that these operations take only as much time as it takes to
1586 send the commands at whatever line speed you are using. You must
1587 fix the termcap entry to specify, for the `al' and `dl', as much
1588 time as the operations really take.
1590 Currently Emacs thinks in terms of serial lines which send characters
1591 at a fixed rate, so that any operation which takes time for the
1592 terminal to execute must also be padded. With bit-map terminals
1593 operated across networks, often the network provides some sort of
1594 flow control so that padding is never needed no matter how slow
1595 an operation is. You must still specify a padding time if you want
1596 Emacs to realize that the operation takes a long time. This will
1597 cause padding characters to be sent unnecessarily, but they do
1598 not really cost much. They will be transmitted while the scrolling
1599 is happening and then discarded quickly by the terminal.
1601 Most bit-map terminals provide commands for inserting or deleting
1602 multiple lines at once. Define the `AL' and `DL' strings in the
1603 termcap entry to say how to do these things, and you will have
1604 fast output without wasted padding characters. These strings should
1605 each contain a single %-spec saying how to send the number of lines
1606 to be scrolled. These %-specs are like those in the termcap
1609 You should also define the `IC' and `DC' strings if your terminal
1610 has a command to insert or delete multiple characters. These
1611 take the number of positions to insert or delete as an argument.
1613 A `cs' string to set the scrolling region will reduce the amount
1614 of motion you see on the screen when part of the screen is scrolled.
1616 ** You type Control-H (Backspace) expecting to delete characters.
1618 Put `stty dec' in your .login file and your problems will disappear
1621 The choice of Backspace for erasure was based on confusion, caused by
1622 the fact that backspacing causes erasure (later, when you type another
1623 character) on most display terminals. But it is a mistake. Deletion
1624 of text is not the same thing as backspacing followed by failure to
1625 overprint. I do not wish to propagate this confusion by conforming
1628 For this reason, I believe `stty dec' is the right mode to use,
1629 and I have designed Emacs to go with that. If there were a thousand
1630 other control characters, I would define Control-h to delete as well;
1631 but there are not very many other control characters, and I think
1632 that providing the most mnemonic possible Help character is more
1633 important than adapting to people who don't use `stty dec'.
1635 If you are obstinate about confusing buggy overprinting with deletion,
1636 you can redefine Backspace in your .emacs file:
1637 (global-set-key "\b" 'delete-backward-char)
1638 You can probably access help-command via f1.
1640 ** Colors are not available on a tty or in xterm.
1642 Emacs 21 supports colors on character terminals and terminal
1643 emulators, but this support relies on the terminfo or termcap database
1644 entry to specify that the display supports color. Emacs looks at the
1645 "Co" capability for the terminal to find out how many colors are
1646 supported; it should be non-zero to activate the color support within
1647 Emacs. (Most color terminals support 8 or 16 colors.) If your system
1648 uses terminfo, the name of the capability equivalent to "Co" is
1651 In addition to the "Co" capability, Emacs needs the "op" (for
1652 ``original pair'') capability, which tells how to switch the terminal
1653 back to the default foreground and background colors. Emacs will not
1654 use colors if this capability is not defined. If your terminal entry
1655 doesn't provide such a capability, try using the ANSI standard escape
1656 sequence \E[00m (that is, define a new termcap/terminfo entry and make
1657 it use your current terminal's entry plus \E[00m for the "op"
1660 Finally, the "NC" capability (terminfo name: "ncv") tells Emacs which
1661 attributes cannot be used with colors. Setting this capability
1662 incorrectly might have the effect of disabling colors; try setting
1663 this capability to `0' (zero) and see if that helps.
1665 Emacs uses the database entry for the terminal whose name is the value
1666 of the environment variable TERM. With `xterm', a common terminal
1667 entry that supports color is `xterm-color', so setting TERM's value to
1668 `xterm-color' might activate the color support on an xterm-compatible
1671 Beginning with version 22.1, Emacs supports the --color command-line
1672 option which may be used to force Emacs to use one of a few popular
1673 modes for getting colors on a tty. For example, --color=ansi8 sets up
1674 for using the ANSI-standard escape sequences that support 8 colors.
1676 Some modes do not use colors unless you turn on the Font-lock mode.
1677 Some people have long ago set their `~/.emacs' files to turn on
1678 Font-lock on X only, so they won't see colors on a tty. The
1679 recommended way of turning on Font-lock is by typing "M-x
1680 global-font-lock-mode RET" or by customizing the variable
1681 `global-font-lock-mode'.
1683 * Runtime problems specific to individual Unix variants
1687 *** GNU/Linux: Process output is corrupted.
1689 There is a bug in Linux kernel 2.6.10 PTYs that can cause emacs to
1690 read corrupted process output.
1692 *** GNU/Linux: Remote access to CVS with SSH causes file corruption.
1694 If you access a remote CVS repository via SSH, files may be corrupted
1695 due to bad interaction between CVS, SSH, and libc.
1697 To fix the problem, save the following script into a file, make it
1698 executable, and set CVS_RSH environment variable to the file name of
1702 exec 2> >(exec cat >&2 2>/dev/null)
1705 *** GNU/Linux: On Linux-based GNU systems using libc versions 5.4.19 through
1706 5.4.22, Emacs crashes at startup with a segmentation fault.
1708 This problem happens if libc defines the symbol __malloc_initialized.
1709 One known solution is to upgrade to a newer libc version. 5.4.33 is
1712 *** GNU/Linux: After upgrading to a newer version of Emacs,
1713 the Meta key stops working.
1715 This was reported to happen on a GNU/Linux system distributed by
1716 Mandrake. The reason is that the previous version of Emacs was
1717 modified by Mandrake to make the Alt key act as the Meta key, on a
1718 keyboard where the Windows key is the one which produces the Meta
1719 modifier. A user who started using a newer version of Emacs, which
1720 was not hacked by Mandrake, expected the Alt key to continue to act as
1721 Meta, and was astonished when that didn't happen.
1723 The solution is to find out what key on your keyboard produces the Meta
1724 modifier, and use that key instead. Try all of the keys to the left
1725 and to the right of the space bar, together with the `x' key, and see
1726 which combination produces "M-x" in the echo area. You can also use
1727 the `xmodmap' utility to show all the keys which produce a Meta
1730 xmodmap -pk | egrep -i "meta|alt"
1732 A more convenient way of finding out which keys produce a Meta modifier
1733 is to use the `xkbprint' utility, if it's available on your system:
1735 xkbprint 0:0 /tmp/k.ps
1737 This produces a PostScript file `/tmp/k.ps' with a picture of your
1738 keyboard; printing that file on a PostScript printer will show what
1739 keys can serve as Meta.
1741 The `xkeycaps' also shows a visual representation of the current
1742 keyboard settings. It also allows to modify them.
1744 *** GNU/Linux: slow startup on Linux-based GNU systems.
1746 People using systems based on the Linux kernel sometimes report that
1747 startup takes 10 to 15 seconds longer than `usual'.
1749 This is because Emacs looks up the host name when it starts.
1750 Normally, this takes negligible time; the extra delay is due to
1751 improper system configuration. This problem can occur for both
1752 networked and non-networked machines.
1754 Here is how to fix the configuration. It requires being root.
1756 **** Networked Case.
1758 First, make sure the files `/etc/hosts' and `/etc/host.conf' both
1759 exist. The first line in the `/etc/hosts' file should look like this
1760 (replace HOSTNAME with your host name):
1764 Also make sure that the `/etc/host.conf' files contains the following
1770 Any changes, permanent and temporary, to the host name should be
1771 indicated in the `/etc/hosts' file, since it acts a limited local
1772 database of addresses and names (e.g., some SLIP connections
1773 dynamically allocate ip addresses).
1775 **** Non-Networked Case.
1777 The solution described in the networked case applies here as well.
1778 However, if you never intend to network your machine, you can use a
1779 simpler solution: create an empty `/etc/host.conf' file. The command
1780 `touch /etc/host.conf' suffices to create the file. The `/etc/hosts'
1781 file is not necessary with this approach.
1783 *** GNU/Linux: Emacs on a tty switches the cursor to large blinking block.
1785 This was reported to happen on some GNU/Linux systems which use
1786 ncurses version 5.0, but could be relevant for other versions as well.
1787 These versions of ncurses come with a `linux' terminfo entry, where
1788 the "cvvis" capability (termcap "vs") is defined as "\E[?25h\E[?8c"
1789 (show cursor, change size). This escape sequence switches on a
1790 blinking hardware text-mode cursor whose size is a full character
1791 cell. This blinking cannot be stopped, since a hardware cursor
1794 A work-around is to redefine the "cvvis" capability so that it
1795 enables a *software* cursor. The software cursor works by inverting
1796 the colors of the character at point, so what you see is a block
1797 cursor that doesn't blink. For this to work, you need to redefine
1798 the "cnorm" capability as well, so that it operates on the software
1799 cursor instead of the hardware cursor.
1801 To this end, run "infocmp linux > linux-term", edit the file
1802 `linux-term' to make both the "cnorm" and "cvvis" capabilities send
1803 the sequence "\E[?25h\E[?17;0;64c", and then run "tic linux-term" to
1804 produce a modified terminfo entry.
1806 Alternatively, if you want a blinking underscore as your Emacs cursor,
1807 change the "cvvis" capability to send the "\E[?25h\E[?0c" command.
1809 *** GNU/Linux: Error messages `internal facep []' happen on GNU/Linux systems.
1811 There is a report that replacing libc.so.5.0.9 with libc.so.5.2.16
1812 caused this to start happening. People are not sure why, but the
1813 problem seems unlikely to be in Emacs itself. Some suspect that it
1814 is actually Xlib which won't work with libc.so.5.2.16.
1816 Using the old library version is a workaround.
1820 *** Mac OS X (Carbon): Environment Variables from dotfiles are ignored.
1822 When starting Emacs from the Dock or the Finder on Mac OS X, the
1823 environment variables that are set up in dotfiles, such as .cshrc or
1824 .profile, are ignored. This is because the Finder and Dock are not
1825 started from a shell, but instead from the Window Manager itself.
1827 The workaround for this is to create a .MacOSX/environment.plist file to
1828 setup these environment variables. These environment variables will
1829 apply to all processes regardless of where they are started.
1830 For me information, see http://developer.apple.com/qa/qa2001/qa1067.html.
1832 *** Mac OS X (Carbon): Process output truncated when using ptys.
1834 There appears to be a problem with the implementation of pty's on the
1835 Mac OS X that causes process output to be truncated. To avoid this,
1836 leave process-connection-type set to its default value of nil.
1838 *** Mac OS X 10.3.9 (Carbon): QuickTime 7.0.4 updater breaks build.
1840 On the above environment, build fails at the link stage with the
1841 message like "Undefined symbols: _HICopyAccessibilityActionDescription
1842 referenced from QuickTime expected to be defined in Carbon". A
1843 workaround is to use QuickTime 7.0.1 reinstaller.
1847 *** FreeBSD 2.1.5: useless symbolic links remain in /tmp or other
1848 directories that have the +t bit.
1850 This is because of a kernel bug in FreeBSD 2.1.5 (fixed in 2.2).
1851 Emacs uses symbolic links to implement file locks. In a directory
1852 with +t bit, the directory owner becomes the owner of the symbolic
1853 link, so that it cannot be removed by anyone else.
1855 If you don't like those useless links, you can let Emacs not to using
1856 file lock by adding #undef CLASH_DETECTION to config.h.
1858 *** FreeBSD: Getting a Meta key on the console.
1860 By default, neither Alt nor any other key acts as a Meta key on
1861 FreeBSD, but this can be changed using kbdcontrol(1). Dump the
1862 current keymap to a file with the command
1864 $ kbdcontrol -d >emacs.kbd
1866 Edit emacs.kbd, and give the key you want to be the Meta key the
1867 definition `meta'. For instance, if your keyboard has a ``Windows''
1868 key with scan code 105, change the line for scan code 105 in emacs.kbd
1871 105 meta meta meta meta meta meta meta meta O
1873 to make the Windows key the Meta key. Load the new keymap with
1875 $ kbdcontrol -l emacs.kbd
1879 *** HP/UX : Shell mode gives the message, "`tty`: Ambiguous".
1881 christos@theory.tn.cornell.edu says:
1883 The problem is that in your .cshrc you have something that tries to
1884 execute `tty`. If you are not running the shell on a real tty then
1885 tty will print "not a tty". Csh expects one word in some places,
1886 but tty is giving it back 3.
1888 The solution is to add a pair of quotes around `tty` to make it a single
1891 if (`tty` == "/dev/console")
1893 should be changed to:
1895 if ("`tty`" == "/dev/console")
1897 Even better, move things that set up terminal sections out of .cshrc
1900 *** HP/UX: `Pid xxx killed due to text modification or page I/O error'.
1902 On HP/UX, you can get that error when the Emacs executable is on an NFS
1903 file system. HP/UX responds this way if it tries to swap in a page and
1904 does not get a response from the server within a timeout whose default
1905 value is just ten seconds.
1907 If this happens to you, extend the timeout period.
1909 *** HP/UX: The right Alt key works wrong on German HP keyboards (and perhaps
1910 other non-English HP keyboards too).
1912 This is because HP-UX defines the modifiers wrong in X. Here is a
1913 shell script to fix the problem; be sure that it is run after VUE
1914 configures the X server.
1916 xmodmap 2> /dev/null - << EOF
1917 keysym Alt_L = Meta_L
1918 keysym Alt_R = Meta_R
1923 keysym Mode_switch = NoSymbol
1925 keysym Meta_R = Mode_switch
1926 add mod2 = Mode_switch
1929 *** HP/UX: "Cannot find callback list" messages from dialog boxes in
1930 Emacs built with Motif.
1932 This problem resulted from a bug in GCC 2.4.5. Newer GCC versions
1933 such as 2.7.0 fix the problem.
1935 *** HP/UX: Emacs does not recognize the AltGr key.
1937 To fix this, set up a file ~/.dt/sessions/sessionetc with executable
1938 rights, containing this text:
1940 --------------------------------
1941 xmodmap 2> /dev/null - << EOF
1942 keysym Alt_L = Meta_L
1943 keysym Alt_R = Meta_R
1948 keysym Mode_switch = NoSymbol
1950 keysym Meta_R = Mode_switch
1951 add mod2 = Mode_switch
1953 --------------------------------
1955 *** HP/UX 11.0: Emacs makes HP/UX 11.0 crash.
1957 This is a bug in HPUX; HPUX patch PHKL_16260 is said to fix it.
1961 *** AIX: Trouble using ptys.
1963 People often install the pty devices on AIX incorrectly.
1964 Use `smit pty' to reinstall them properly.
1966 *** AIXterm: Your Delete key sends a Backspace to the terminal.
1968 The solution is to include in your .Xdefaults the lines:
1970 *aixterm.Translations: #override <Key>BackSpace: string(0x7f)
1971 aixterm*ttyModes: erase ^?
1973 This makes your Backspace key send DEL (ASCII 127).
1975 *** AIX: If linking fails because libXbsd isn't found, check if you
1976 are compiling with the system's `cc' and CFLAGS containing `-O5'. If
1977 so, you have hit a compiler bug. Please make sure to re-configure
1978 Emacs so that it isn't compiled with `-O5'.
1980 *** AIX 4.3.x or 4.4: Compiling fails.
1982 This could happen if you use /bin/c89 as your compiler, instead of
1983 the default `cc'. /bin/c89 treats certain warnings, such as benign
1984 redefinitions of macros, as errors, and fails the build. A solution
1985 is to use the default compiler `cc'.
1987 *** AIX 4: Some programs fail when run in a Shell buffer
1988 with an error message like No terminfo entry for "unknown".
1990 On AIX, many terminal type definitions are not installed by default.
1991 `unknown' is one of them. Install the "Special Generic Terminal
1992 Definitions" to make them defined.
1996 We list bugs in current versions here. Solaris 2.x and 4.x are covered in the
1997 section on legacy systems.
1999 *** On Solaris, C-x doesn't get through to Emacs when you use the console.
2001 This is a Solaris feature (at least on Intel x86 cpus). Type C-r
2002 C-r C-t, to toggle whether C-x gets through to Emacs.
2004 *** Problem with remote X server on Suns.
2006 On a Sun, running Emacs on one machine with the X server on another
2007 may not work if you have used the unshared system libraries. This
2008 is because the unshared libraries fail to use YP for host name lookup.
2009 As a result, the host name you specify may not be recognized.
2011 *** Solaris 2,6: Emacs crashes with SIGBUS or SIGSEGV on Solaris after you delete a frame.
2013 We suspect that this is a bug in the X libraries provided by
2014 Sun. There is a report that one of these patches fixes the bug and
2015 makes the problem stop:
2017 105216-01 105393-01 105518-01 105621-01 105665-01 105615-02 105216-02
2018 105667-01 105401-08 105615-03 105621-02 105686-02 105736-01 105755-03
2019 106033-01 105379-01 105786-01 105181-04 105379-03 105786-04 105845-01
2020 105284-05 105669-02 105837-01 105837-02 105558-01 106125-02 105407-01
2022 Another person using a newer system (kernel patch level Generic_105181-06)
2023 suspects that the bug was fixed by one of these more recent patches:
2025 106040-07 SunOS 5.6: X Input & Output Method patch
2026 106222-01 OpenWindows 3.6: filemgr (ff.core) fixes
2027 105284-12 Motif 1.2.7: sparc Runtime library patch
2029 *** Solaris 7 or 8: Emacs reports a BadAtom error (from X)
2031 This happens when Emacs was built on some other version of Solaris.
2032 Rebuild it on Solaris 8.
2034 *** When using M-x dbx with the SparcWorks debugger, the `up' and `down'
2035 commands do not move the arrow in Emacs.
2037 You can fix this by adding the following line to `~/.dbxinit':
2039 dbxenv output_short_file_name off
2041 *** On Solaris, CTRL-t is ignored by Emacs when you use
2042 the fr.ISO-8859-15 locale (and maybe other related locales).
2044 You can fix this by editing the file:
2046 /usr/openwin/lib/locale/iso8859-15/Compose
2048 Near the bottom there is a line that reads:
2050 Ctrl<t> <quotedbl> <Y> : "\276" threequarters
2054 Ctrl<T> <quotedbl> <Y> : "\276" threequarters
2056 Note the lower case <t>. Changing this line should make C-t work.
2060 *** Irix 6.5: Emacs crashes on the SGI R10K, when compiled with GCC.
2062 This seems to be fixed in GCC 2.95.
2064 *** Irix: Trouble using ptys, or running out of ptys.
2066 The program mkpts (which may be in `/usr/adm' or `/usr/sbin') needs to
2067 be set-UID to root, or non-root programs like Emacs will not be able
2068 to allocate ptys reliably.
2070 * Runtime problems specific to MS-Windows
2072 ** Windows 95 and networking.
2074 To support server sockets, Emacs 22.1 loads ws2_32.dll. If this file
2075 is missing, all Emacs networking features are disabled.
2077 Old versions of Windows 95 may not have the required DLL. To use
2078 Emacs' networking features on Windows 95, you must install the
2079 "Windows Socket 2" update available from MicroSoft's support Web.
2081 ** Emacs exits with "X protocol error" when run with an X server for MS-Windows.
2083 A certain X server for Windows had a bug which caused this.
2084 Supposedly the newer 32-bit version of this server doesn't have the
2087 ** Known problems with the MS-Windows port of Emacs 22.1
2089 Using create-fontset-from-ascii-font or the --font startup parameter
2090 with a Chinese, Japanese or Korean font leads to display problems.
2091 Use a Latin-only font as your default font. If you want control over
2092 which font is used to display Chinese, Japanese or Korean character,
2093 use create-fontset-from-fontset-spec to define a fontset.
2095 Frames are not refreshed while the File or Font dialog or a pop-up menu
2096 is displayed. This also means help text for pop-up menus is not
2097 displayed at all. This is because message handling under Windows is
2098 synchronous, so we cannot handle repaint (or any other) messages while
2099 waiting for a system function to return the result of the dialog or
2100 pop-up menu interaction.
2102 Windows 95 and Windows NT up to version 4.0 do not support help text
2103 for menus. Help text is only available in later versions of Windows.
2105 When "ClearType" method is selected as the "method to smooth edges of
2106 screen fonts" (in Display Properties, Appearance tab, under
2107 "Effects"), there are various problems related to display of
2108 characters: 2-pixel trace is left behind when moving overlays, bold
2109 fonts can be hard to read, small portions of some characters could
2110 appear chopped, etc. This happens because, under ClearType,
2111 characters are drawn outside their advertised bounding box. Emacs 21
2112 disabled the use of ClearType, whereas Emacs 22 allows it and has some
2113 code to enlarge the width of the bounding box. Apparently, this
2114 display feature needs more changes to get it 100% right. A workaround
2115 is to disable ClearType.
2117 There are problems with display if mouse-tracking is enabled and the
2118 mouse is moved off a frame, over another frame then back over the first
2119 frame. A workaround is to click the left mouse button inside the frame
2120 after moving back into it.
2122 Some minor flickering still persists during mouse-tracking, although
2123 not as severely as in 21.1.
2125 An inactive cursor remains in an active window after the Windows
2126 Manager driven switch of the focus, until a key is pressed.
2128 Windows input methods are not recognized by Emacs. However, some
2129 of these input methods cause the keyboard to send characters encoded
2130 in the appropriate coding system (e.g., ISO 8859-1 for Latin-1
2131 characters, ISO 8859-8 for Hebrew characters, etc.). To make these
2132 input methods work with Emacs, set the keyboard coding system to the
2133 appropriate value after you activate the Windows input method. For
2134 example, if you activate the Hebrew input method, type this:
2136 C-x RET k hebrew-iso-8bit RET
2138 (Emacs ought to recognize the Windows language-change event and set up
2139 the appropriate keyboard encoding automatically, but it doesn't do
2140 that yet.) In addition, to use these Windows input methods, you
2141 should set your "Language for non-Unicode programs" (on Windows XP,
2142 this is on the Advanced tab of Regional Settings) to the language of
2145 To bind keys that produce non-ASCII characters with modifiers, you
2146 must specify raw byte codes. For instance, if you want to bind
2147 META-a-grave to a command, you need to specify this in your `~/.emacs':
2149 (global-set-key [?\M-\340] ...)
2151 The above example is for the Latin-1 environment where the byte code
2152 of the encoded a-grave is 340 octal. For other environments, use the
2153 encoding appropriate to that environment.
2155 The %b specifier for format-time-string does not produce abbreviated
2156 month names with consistent widths for some locales on some versions
2157 of Windows. This is caused by a deficiency in the underlying system
2160 ** Typing Alt-Shift has strange effects on MS-Windows.
2162 This combination of keys is a command to change keyboard layout. If
2163 you proceed to type another non-modifier key before you let go of Alt
2164 and Shift, the Alt and Shift act as modifiers in the usual way. A
2165 more permanent work around is to change it to another key combination,
2166 or disable it in the keyboard control panel.
2168 ** Cygwin build of Emacs hangs after rebasing Cygwin DLLs
2170 Usually, on Cygwin, one needs to rebase the DLLs if an application
2171 aborts with a message like this:
2173 C:\cygwin\bin\python.exe: *** unable to remap C:\cygwin\bin\cygssl.dll to
2174 same address as parent(0xDF0000) != 0xE00000
2176 However, since Cygwin DLL 1.5.17 was released, after such rebasing,
2179 This was reported to happen for Emacs 21.2 and also for the pretest of
2180 Emacs 22.1 on Cygwin.
2182 To work around this, build Emacs like this:
2184 LDFLAGS='-Wl,--enable-auto-import -Wl,--enable-auto-image-base' ./configure
2186 make LD='$(CC)' install
2188 This produces an Emacs binary that is independent of rebasing.
2190 Note that you _must_ use LD='$(CC)' in the last two commands above, to
2191 prevent GCC from passing the "--image-base 0x20000000" option to the
2192 linker, which is what it does by default. That option produces an
2193 Emacs binary with the base address 0x20000000, which will cause Emacs
2194 to hang after Cygwin DLLs are rebased.
2196 ** Interrupting Cygwin port of Bash from Emacs doesn't work.
2198 Cygwin 1.x builds of the ported Bash cannot be interrupted from the
2199 MS-Windows version of Emacs. This is due to some change in the Bash
2200 port or in the Cygwin library which apparently make Bash ignore the
2201 keyboard interrupt event sent by Emacs to Bash. (Older Cygwin ports
2202 of Bash, up to b20.1, did receive SIGINT from Emacs.)
2204 ** Accessing remote files with ange-ftp hangs the MS-Windows version of Emacs.
2206 If the FTP client is the Cygwin port of GNU `ftp', this appears to be
2207 due to some bug in the Cygwin DLL or some incompatibility between it
2208 and the implementation of asynchronous subprocesses in the Windows
2209 port of Emacs. Specifically, some parts of the FTP server responses
2210 are not flushed out, apparently due to buffering issues, which
2213 The solution is to downgrade to an older version of the Cygwin DLL
2214 (version 1.3.2 was reported to solve the problem), or use the stock
2215 Windows FTP client, usually found in the `C:\WINDOWS' or 'C:\WINNT'
2216 directory. To force ange-ftp use the stock Windows client, set the
2217 variable `ange-ftp-ftp-program-name' to the absolute file name of the
2218 client's executable. For example:
2220 (setq ange-ftp-ftp-program-name "c:/windows/ftp.exe")
2222 If you want to stick with the Cygwin FTP client, you can work around
2223 this problem by putting this in your `.emacs' file:
2225 (setq ange-ftp-ftp-program-args '("-i" "-n" "-g" "-v" "--prompt" "")
2227 ** lpr commands don't work on MS-Windows with some cheap printers.
2229 This problem may also strike other platforms, but the solution is
2230 likely to be a global one, and not Emacs specific.
2232 Many cheap inkjet, and even some cheap laser printers, do not
2233 print plain text anymore, they will only print through graphical
2234 printer drivers. A workaround on MS-Windows is to use Windows' basic
2235 built in editor to print (this is possibly the only useful purpose it
2238 (setq printer-name "") ;; notepad takes the default
2239 (setq lpr-command "notepad") ;; notepad
2240 (setq lpr-switches nil) ;; not needed
2241 (setq lpr-printer-switch "/P") ;; run notepad as batch printer
2243 ** Antivirus software interacts badly with the MS-Windows version of Emacs.
2245 The usual manifestation of these problems is that subprocesses don't
2246 work or even wedge the entire system. In particular, "M-x shell RET"
2247 was reported to fail to work. But other commands also sometimes don't
2248 work when an antivirus package is installed.
2250 The solution is to switch the antivirus software to a less aggressive
2251 mode (e.g., disable the ``auto-protect'' feature), or even uninstall
2252 or disable it entirely.
2254 ** Pressing the mouse button on MS-Windows does not give a mouse-2 event.
2256 This is usually a problem with the mouse driver. Because most Windows
2257 programs do not do anything useful with the middle mouse button, many
2258 mouse drivers allow you to define the wheel press to do something
2259 different. Some drivers do not even have the option to generate a
2260 middle button press. In such cases, setting the wheel press to
2261 "scroll" sometimes works if you press the button twice. Trying a
2262 generic mouse driver might help.
2264 ** Scrolling the mouse wheel on MS-Windows always scrolls the top window.
2266 This is another common problem with mouse drivers. Instead of
2267 generating scroll events, some mouse drivers try to fake scroll bar
2268 movement. But they are not intelligent enough to handle multiple
2269 scroll bars within a frame. Trying a generic mouse driver might help.
2271 ** Mail sent through Microsoft Exchange in some encodings appears to be
2272 mangled and is not seen correctly in Rmail or Gnus. We don't know
2273 exactly what happens, but it isn't an Emacs problem in cases we've
2276 ** On MS-Windows, you cannot use the right-hand ALT key and the left-hand
2277 CTRL key together to type a Control-Meta character.
2279 This is a consequence of a misfeature beyond Emacs's control.
2281 Under Windows, the AltGr key on international keyboards generates key
2282 events with the modifiers Right-Alt and Left-Ctrl. Since Emacs cannot
2283 distinguish AltGr from an explicit Right-Alt and Left-Ctrl
2284 combination, whenever it sees Right-Alt and Left-Ctrl it assumes that
2285 AltGr has been pressed. The variable `w32-recognize-altgr' can be set
2286 to nil to tell Emacs that AltGr is really Ctrl and Alt.
2288 ** Under some X-servers running on MS-Windows, Emacs' display is incorrect.
2290 The symptoms are that Emacs does not completely erase blank areas of the
2291 screen during scrolling or some other screen operations (e.g., selective
2292 display or when killing a region). M-x recenter will cause the screen
2293 to be completely redisplayed and the "extra" characters will disappear.
2295 This is known to occur under Exceed 6, and possibly earlier versions
2296 as well; it is reportedly solved in version 6.2.0.16 and later. The
2297 problem lies in the X-server settings.
2299 There are reports that you can solve the problem with Exceed by
2300 running `Xconfig' from within NT, choosing "X selection", then
2301 un-checking the boxes "auto-copy X selection" and "auto-paste to X
2304 Of this does not work, please inform bug-gnu-emacs@gnu.org. Then
2305 please call support for your X-server and see if you can get a fix.
2306 If you do, please send it to bug-gnu-emacs@gnu.org so we can list it
2309 * Build-time problems
2313 *** The `configure' script doesn't find the jpeg library.
2315 There are reports that this happens on some systems because the linker
2316 by default only looks for shared libraries, but jpeg distribution by
2317 default only installs a nonshared version of the library, `libjpeg.a'.
2319 If this is the problem, you can configure the jpeg library with the
2320 `--enable-shared' option and then rebuild libjpeg. This produces a
2321 shared version of libjpeg, which you need to install. Finally, rerun
2322 the Emacs configure script, which should now find the jpeg library.
2323 Alternatively, modify the generated src/Makefile to link the .a file
2324 explicitly, and edit src/config.h to define HAVE_JPEG.
2326 *** `configure' warns ``accepted by the compiler, rejected by the preprocessor''.
2328 This indicates a mismatch between the C compiler and preprocessor that
2329 configure is using. For example, on Solaris 10 trying to use
2330 CC=/opt/SUNWspro/bin/cc (the Sun Studio compiler) together with
2331 CPP=/usr/ccs/lib/cpp can result in errors of this form (you may also
2332 see the error ``"/usr/include/sys/isa_defs.h", line 500: undefined control'').
2334 The solution is to tell configure to use the correct C preprocessor
2335 for your C compiler (CPP="/opt/SUNWspro/bin/cc -E" in the above
2338 *** `configure' fails with ``"junk.c", line 660: invalid input token: 8.elc''
2340 The final stage of the Emacs configure process uses the C preprocessor
2341 to generate the Makefiles. Errors of this form can occur if the C
2342 preprocessor inserts extra whitespace into its output. The solution
2343 is to find the switches that stop your preprocessor from inserting extra
2344 whitespace, add them to CPPFLAGS, and re-run configure. For example,
2345 this error can occur on Solaris 10 when using the Sun Studio compiler
2346 ``Sun C 5.8'' with its preprocessor CPP="/opt/SUNWspro/bin/cc -E".
2347 The relevant switch in this case is "-Xs" (``compile assuming
2348 (pre-ANSI) K & R C style code'').
2352 *** Building Emacs over NFS fails with ``Text file busy''.
2354 This was reported to happen when building Emacs on a GNU/Linux system
2355 (RedHat Linux 6.2) using a build directory automounted from Solaris
2356 (SunOS 5.6) file server, but it might not be limited to that
2357 configuration alone. Presumably, the NFS server doesn't commit the
2358 files' data to disk quickly enough, and the Emacs executable file is
2359 left ``busy'' for several seconds after Emacs has finished dumping
2360 itself. This causes the subsequent commands which invoke the dumped
2361 Emacs executable to fail with the above message.
2363 In some of these cases, a time skew between the NFS server and the
2364 machine where Emacs is built is detected and reported by GNU Make
2365 (it says that some of the files have modification time in the future).
2366 This might be a symptom of NFS-related problems.
2368 If the NFS server runs on Solaris, apply the Solaris patch 105379-05
2369 (Sunos 5.6: /kernel/misc/nfssrv patch). If that doesn't work, or if
2370 you have a different version of the OS or the NFS server, you can
2371 force the NFS server to use 1KB blocks, which was reported to fix the
2372 problem albeit at a price of slowing down file I/O. You can force 1KB
2373 blocks by specifying the "-o rsize=1024,wsize=1024" options to the
2374 `mount' command, or by adding ",rsize=1024,wsize=1024" to the mount
2375 options in the appropriate system configuration file, such as
2378 Alternatively, when Make fails due to this problem, you could wait for
2379 a few seconds and then invoke Make again. In one particular case,
2380 waiting for 10 or more seconds between the two Make invocations seemed
2381 to work around the problem.
2383 Similar problems can happen if your machine NFS-mounts a directory
2384 onto itself. Suppose the Emacs sources live in `/usr/local/src' and
2385 you are working on the host called `marvin'. Then an entry in the
2386 `/etc/fstab' file like the following is asking for trouble:
2388 marvin:/usr/local/src /usr/local/src ...options.omitted...
2390 The solution is to remove this line from `etc/fstab'.
2392 *** Building Emacs with GCC 2.9x fails in the `src' directory.
2394 This may happen if you use a development version of GNU `cpp' from one
2395 of the GCC snapshots between Oct 2000 and Feb 2001, or from a released
2396 version of GCC newer than 2.95.2 which was prepared around those
2397 dates; similar problems were reported with some snapshots of GCC 3.1
2398 around Sep 30 2001. The preprocessor in those versions is
2399 incompatible with a traditional Unix cpp (e.g., it expands ".." into
2400 ". .", which breaks relative file names that reference the parent
2401 directory; or inserts TAB characters before lines that set Make
2404 The solution is to make sure the preprocessor is run with the
2405 `-traditional' option. The `configure' script does that automatically
2406 when it detects the known problems in your cpp, but you might hit some
2407 unknown ones. To force the `configure' script to use `-traditional',
2408 run the script like this:
2410 CPP='gcc -E -traditional' ./configure ...
2412 (replace the ellipsis "..." with any additional arguments you pass to
2415 Note that this problem does not pertain to the MS-Windows port of
2416 Emacs, since it doesn't use the preprocessor to generate Makefiles.
2418 *** src/Makefile and lib-src/Makefile are truncated--most of the file missing.
2419 *** Compiling wakeup, in lib-src, says it can't make wakeup.c.
2421 This can happen if configure uses GNU sed version 2.03. That version
2422 had a bug. GNU sed version 2.05 works properly.To solve the
2423 problem, install the current version of GNU Sed, then rerun Emacs's
2426 *** Compiling lib-src says there is no rule to make test-distrib.c.
2428 This results from a bug in a VERY old version of GNU Sed. To solve
2429 the problem, install the current version of GNU Sed, then rerun
2430 Emacs's configure script.
2432 *** Building a 32-bit executable on a 64-bit GNU/Linux architecture.
2434 First ensure that the necessary 32-bit system libraries and include
2435 files are installed. Then use:
2437 env CC="gcc -m32" ./configure --build=i386-linux-gnu \
2438 --x-libraries=/usr/X11R6/lib
2440 (using the location of the 32-bit X libraries on your system).
2442 *** Building the Cygwin port for MS-Windows can fail with some GCC version
2444 Building Emacs 22 with Cygwin builds of GCC 3.4.4-1 and 3.4.4-2 is
2445 reported to either fail or cause Emacs to segfault at run time. In
2446 addition, the Cygwin GCC 3.4.4-2 has problems with generating debug
2447 info. Cygwin users are advised not to use these versions of GCC for
2448 compiling Emacs. GCC versions 4.0.3, 4.1.1, and 4.1.2 reportedly
2449 build a working Cygwin binary of Emacs, so we recommend these GCC
2450 versions. Note that these three versions of GCC, 4.0.3, 4.1.1, and
2451 4.1.2, are currently the _only_ versions known to succeed in building
2452 Emacs (as of v22.1).
2454 *** Building the native MS-Windows port with Cygwin GCC can fail.
2456 Emacs may not build using recent Cygwin builds of GCC, such as Cygwin
2457 version 1.1.8, using the default configure settings. It appears to be
2458 necessary to specify the -mwin32 flag when compiling, and define
2459 __MSVCRT__, like so:
2461 configure --with-gcc --cflags -mwin32 --cflags -D__MSVCRT__
2463 *** Building the MS-Windows port fails with a CreateProcess failure.
2465 Some versions of mingw32 make on some versions of Windows do not seem
2466 to detect the shell correctly. Try "make SHELL=cmd.exe", or if that
2467 fails, try running make from Cygwin bash instead.
2469 *** Building the MS-Windows port with Leim fails in the `leim' directory.
2471 The error message might be something like this:
2473 Converting d:/emacs-21.3/leim/CXTERM-DIC/4Corner.tit to quail-package...
2474 Invalid ENCODE: value in TIT dictionary
2475 NMAKE : fatal error U1077: '"../src/obj-spd/i386/emacs.exe"' : return code
2479 This can happen if the Leim distribution is unpacked with a program
2480 which converts the `*.tit' files to DOS-style CR-LF text format. The
2481 `*.tit' files in the leim/CXTERM-DIC directory require Unix-style line
2482 endings to compile properly, because Emacs reads them without any code
2485 The solution is to make sure the program used to unpack Leim does not
2486 change the files' line endings behind your back. The GNU FTP site has
2487 in the `/gnu/emacs/windows' directory a program called `djtarnt.exe'
2488 which can be used to unpack `.tar.gz' and `.zip' archives without
2491 *** Building `ctags' for MS-Windows with the MinGW port of GCC fails.
2493 This might happen due to a bug in the MinGW header assert.h, which
2494 defines the `assert' macro with a trailing semi-colon. The following
2495 patch to assert.h should solve this:
2497 *** include/assert.h.orig Sun Nov 7 02:41:36 1999
2498 --- include/assert.h Mon Jan 29 11:49:10 2001
2502 * If not debugging, assert does nothing.
2504 ! #define assert(x) ((void)0);
2506 #else /* debugging enabled */
2510 * If not debugging, assert does nothing.
2512 ! #define assert(x) ((void)0)
2514 #else /* debugging enabled */
2517 *** Building the MS-Windows port with Visual Studio 2005 fails.
2519 Microsoft no longer ships the single threaded version of the C library
2520 with their compiler, and the multithreaded static library is missing
2521 some functions that Microsoft have deemed non-threadsafe. The
2522 dynamically linked C library has all the functions, but there is a
2523 conflict between the versions of malloc in the DLL and in Emacs, which
2524 is not resolvable due to the way Windows does dynamic linking.
2526 We recommend the use of the MingW port of GCC for compiling Emacs, as
2527 not only does it not suffer these problems, but it is also Free
2528 software like Emacs.
2532 *** Building Emacs with a system compiler fails to link because of an
2533 undefined symbol such as __eprintf which does not appear in Emacs.
2535 This can happen if some of the libraries linked into Emacs were built
2536 with GCC, but Emacs itself is being linked with a compiler other than
2537 GCC. Object files compiled with GCC might need some helper functions
2538 from libgcc.a, the library which comes with GCC, but the system
2539 compiler does not instruct the linker to search libgcc.a during the
2542 A solution is to link with GCC, like this:
2546 Since the .o object files already exist, this will not recompile Emacs
2547 with GCC, but just restart by trying again to link temacs.
2549 *** AIX 1.3 ptf 0013: Link failure.
2551 There is a real duplicate definition of the function `_slibc_free' in
2552 the library /lib/libc_s.a (just do nm on it to verify). The
2556 ar xv libc_s.a NLtmtime.o
2557 ar dv libc_s.a NLtmtime.o
2559 *** AIX 4.1.2: Linker error messages such as
2560 ld: 0711-212 SEVERE ERROR: Symbol .__quous, found in the global symbol table
2561 of archive /usr/lib/libIM.a, was not defined in archive member shr.o.
2563 This is a problem in libIM.a. You can work around it by executing
2564 these shell commands in the src subdirectory of the directory where
2567 cp /usr/lib/libIM.a .
2571 Then change -lIM to ./libIM.a in the command to link temacs (in
2574 *** Sun with acc: Link failure when using acc on a Sun.
2576 To use acc, you need additional options just before the libraries, such as
2578 /usr/lang/SC2.0.1/values-Xt.o -L/usr/lang/SC2.0.1/cg87 -L/usr/lang/SC2.0.1
2580 and you need to add -lansi just before -lc.
2582 The precise file names depend on the compiler version, so we
2583 cannot easily arrange to supply them.
2585 *** Linking says that the functions insque and remque are undefined.
2587 Change oldXMenu/Makefile by adding insque.o to the variable OBJS.
2589 *** `tparam' reported as a multiply-defined symbol when linking with ncurses.
2591 This problem results from an incompatible change in ncurses, in
2592 version 1.9.9e approximately. This version is unable to provide a
2593 definition of tparm without also defining tparam. This is also
2594 incompatible with Terminfo; as a result, the Emacs Terminfo support
2595 does not work with this version of ncurses.
2597 The fix is to install a newer version of ncurses, such as version 4.2.
2601 *** Linux: Segfault during `make bootstrap' under certain recent versions of the Linux kernel.
2603 With certain recent Linux kernels (like the one of Redhat Fedora Core
2604 1 and newer), the new "Exec-shield" functionality is enabled by default, which
2605 creates a different memory layout that breaks the emacs dumper. Emacs tries
2606 to handle this at build time, but if the workaround used fails, these
2607 instructions can be useful.
2608 The work-around explained here is not enough on Fedora Core 4 (and possible
2609 newer). Read the next item.
2611 Configure can overcome the problem of exec-shield if the architecture is
2612 x86 and the program setarch is present. On other architectures no
2613 workaround is known.
2615 You can check the Exec-shield state like this:
2617 cat /proc/sys/kernel/exec-shield
2619 It returns non-zero when Exec-shield is enabled, 0 otherwise. Please
2620 read your system documentation for more details on Exec-shield and
2621 associated commands. Exec-shield can be turned off with this command:
2623 echo "0" > /proc/sys/kernel/exec-shield
2625 When Exec-shield is enabled, building Emacs will segfault during the
2626 execution of this command:
2628 ./temacs --batch --load loadup [dump|bootstrap]
2630 To work around this problem, it is necessary to temporarily disable
2631 Exec-shield while building Emacs, or, on x86, by using the `setarch'
2632 command when running temacs like this:
2634 setarch i386 ./temacs --batch --load loadup [dump|bootstrap]
2637 *** Fedora Core 4 GNU/Linux: Segfault during dumping.
2639 In addition to exec-shield explained above "Linux: Segfault during
2640 `make bootstrap' under certain recent versions of the Linux kernel"
2641 item, Linux kernel shipped with Fedora Core 4 randomizes the virtual
2642 address space of a process. As the result dumping may fail even if
2643 you turn off exec-shield. In this case, use the -R option to the setarch
2646 setarch i386 -R ./temacs --batch --load loadup [dump|bootstrap]
2650 setarch i386 -R make bootstrap
2652 *** Fatal signal in the command temacs -l loadup inc dump.
2654 This command is the final stage of building Emacs. It is run by the
2655 Makefile in the src subdirectory, or by build.com on VMS.
2657 It has been known to get fatal errors due to insufficient swapping
2658 space available on the machine.
2660 On 68000s, it has also happened because of bugs in the
2661 subroutine `alloca'. Verify that `alloca' works right, even
2662 for large blocks (many pages).
2664 *** test-distrib says that the distribution has been clobbered.
2665 *** or, temacs prints "Command key out of range 0-127".
2666 *** or, temacs runs and dumps emacs, but emacs totally fails to work.
2667 *** or, temacs gets errors dumping emacs.
2669 This can be because the .elc files have been garbled. Do not be
2670 fooled by the fact that most of a .elc file is text: these are
2671 binary files and can contain all 256 byte values.
2673 In particular `shar' cannot be used for transmitting GNU Emacs.
2674 It typically truncates "lines". What appear to be "lines" in
2675 a binary file can of course be of any length. Even once `shar'
2676 itself is made to work correctly, `sh' discards null characters
2677 when unpacking the shell archive.
2679 I have also seen character \177 changed into \377. I do not know
2680 what transfer means caused this problem. Various network
2681 file transfer programs are suspected of clobbering the high bit.
2683 If you have a copy of Emacs that has been damaged in its
2684 nonprinting characters, you can fix them:
2686 1) Record the names of all the .elc files.
2687 2) Delete all the .elc files.
2688 3) Recompile alloc.c with a value of PURESIZE twice as large.
2689 (See puresize.h.) You might as well save the old alloc.o.
2690 4) Remake emacs. It should work now.
2691 5) Running emacs, do Meta-x byte-compile-file repeatedly
2692 to recreate all the .elc files that used to exist.
2693 You may need to increase the value of the variable
2694 max-lisp-eval-depth to succeed in running the compiler interpreted
2695 on certain .el files. 400 was sufficient as of last report.
2696 6) Reinstall the old alloc.o (undoing changes to alloc.c if any)
2698 7) Remake emacs. It should work now, with valid .elc files.
2700 *** temacs prints "Pure Lisp storage exhausted".
2702 This means that the Lisp code loaded from the .elc and .el
2703 files during temacs -l loadup inc dump took up more
2704 space than was allocated.
2706 This could be caused by
2707 1) adding code to the preloaded Lisp files
2708 2) adding more preloaded files in loadup.el
2709 3) having a site-init.el or site-load.el which loads files.
2710 Note that ANY site-init.el or site-load.el is nonstandard;
2711 if you have received Emacs from some other site
2712 and it contains a site-init.el or site-load.el file, consider
2714 4) getting the wrong .el or .elc files
2715 (not from the directory you expected).
2716 5) deleting some .elc files that are supposed to exist.
2717 This would cause the source files (.el files) to be
2718 loaded instead. They take up more room, so you lose.
2719 6) a bug in the Emacs distribution which underestimates
2722 If the need for more space is legitimate, change the definition
2723 of PURESIZE in puresize.h.
2725 But in some of the cases listed above, this problem is a consequence
2726 of something else that is wrong. Be sure to check and fix the real
2729 *** Linux: Emacs crashes when dumping itself on Mac PPC running Yellow Dog GNU/Linux.
2731 The crashes happen inside the function Fmake_symbol; here's a typical
2732 C backtrace printed by GDB:
2734 0x190c0c0 in Fmake_symbol ()
2736 #0 0x190c0c0 in Fmake_symbol ()
2737 #1 0x1942ca4 in init_obarray ()
2738 #2 0x18b3500 in main ()
2739 #3 0x114371c in __libc_start_main (argc=5, argv=0x7ffff5b4, envp=0x7ffff5cc,
2741 This could happen because GCC version 2.95 and later changed the base
2742 of the load address to 0x10000000. Emacs needs to be told about this,
2743 but we currently cannot do that automatically, because that breaks
2744 other versions of GNU/Linux on the MacPPC. Until we find a way to
2745 distinguish between the Yellow Dog and the other varieties of
2746 GNU/Linux systems on the PPC, you will have to manually uncomment the
2747 following section near the end of the file src/m/macppc.h in the Emacs
2750 #if 0 /* This breaks things on PPC GNU/Linux except for Yellowdog,
2751 even with identical GCC, as, ld. Let's take it out until we
2752 know what's really going on here. */
2753 /* GCC 2.95 and newer on GNU/Linux PPC changed the load address to
2755 #if defined __linux__
2756 #if __GNUC__ > 2 || (__GNUC__ == 2 && __GNUC_MINOR__ >= 95)
2757 #define DATA_SEG_BITS 0x10000000
2762 Remove the "#if 0" and "#endif" directives which surround this, save
2763 the file, and then reconfigure and rebuild Emacs. The dumping process
2768 *** Installing Emacs gets an error running `install-info'.
2770 You need to install a recent version of Texinfo; that package
2771 supplies the `install-info' command.
2773 *** Installing to a directory with spaces in the name fails.
2775 For example, if you call configure with a directory-related option
2776 with spaces in the value, eg --enable-locallisppath='/path/with\ spaces'.
2777 Using directory paths with spaces is not supported at this time: you
2778 must re-configure without using spaces.
2782 *** Emacs binary is not in executable format, and cannot be run.
2784 This was reported to happen when Emacs is built in a directory mounted
2785 via NFS, for some combinations of NFS client and NFS server.
2786 Usually, the file `emacs' produced in these cases is full of
2787 binary null characters, and the `file' utility says:
2789 emacs: ASCII text, with no line terminators
2791 We don't know what exactly causes this failure. A work-around is to
2792 build Emacs in a directory on a local disk.
2794 *** The dumped Emacs crashes when run, trying to write pure data.
2796 Two causes have been seen for such problems.
2798 1) On a system where getpagesize is not a system call, it is defined
2799 as a macro. If the definition (in both unexec.c and malloc.c) is wrong,
2800 it can cause problems like this. You might be able to find the correct
2801 value in the man page for a.out (5).
2803 2) Some systems allocate variables declared static among the
2804 initialized variables. Emacs makes all initialized variables in most
2805 of its files pure after dumping, but the variables declared static and
2806 not initialized are not supposed to be pure. On these systems you
2807 may need to add "#define static" to the m- or the s- file.
2811 ** Error messages `Wrong number of arguments: #<subr where-is-internal>, 5'.
2813 This typically results from having the powerkey library loaded.
2814 Powerkey was designed for Emacs 19.22. It is obsolete now because
2815 Emacs 19 now has this feature built in; and powerkey also calls
2816 where-is-internal in an obsolete way.
2818 So the fix is to arrange not to load powerkey.
2820 * Runtime problems on legacy systems
2822 This section covers bugs reported on very old hardware or software.
2823 If you are using hardware and an operating system shipped after 2000,
2824 it is unlikely you will see any of these.
2826 ** Ancient operating systems
2828 AIX 4.2 was end-of-lifed on Dec 31st, 1999.
2830 *** AIX: You get this compiler error message:
2832 Processing include file ./XMenuInt.h
2833 1501-106: (S) Include file X11/Xlib.h not found.
2835 This means your system was installed with only the X11 runtime i.d
2836 libraries. You have to find your sipo (bootable tape) and install
2837 X11Dev... with smit.
2839 (This report must be ancient. Bootable tapes are long dead.)
2841 *** AIX 3.2.4: Releasing Ctrl/Act key has no effect, if Shift is down.
2843 Due to a feature of AIX, pressing or releasing the Ctrl/Act key is
2844 ignored when the Shift, Alt or AltGr keys are held down. This can
2845 lead to the keyboard being "control-locked"--ordinary letters are
2846 treated as control characters.
2848 You can get out of this "control-locked" state by pressing and
2849 releasing Ctrl/Act while not pressing or holding any other keys.
2851 *** AIX 3.2.5: You get this message when running Emacs:
2853 Could not load program emacs
2854 Symbol smtcheckinit in csh is undefined
2855 Error was: Exec format error
2859 Could not load program .emacs
2860 Symbol _system_con in csh is undefined
2861 Symbol _fp_trapsta in csh is undefined
2862 Error was: Exec format error
2864 These can happen when you try to run on AIX 3.2.5 a program that was
2865 compiled with 3.2.4. The fix is to recompile.
2867 *** AIX 4.2: Emacs gets a segmentation fault at startup.
2869 If you are using IBM's xlc compiler, compile emacs.c
2870 without optimization; that should avoid the problem.
2874 **** ISC: display-time causes kernel problems on ISC systems.
2876 Under Interactive Unix versions 3.0.1 and 4.0 (and probably other
2877 versions), display-time causes the loss of large numbers of STREVENT
2878 cells. Eventually the kernel's supply of these cells is exhausted.
2879 This makes emacs and the whole system run slow, and can make other
2880 processes die, in particular pcnfsd.
2882 Other emacs functions that communicate with remote processes may have
2883 the same problem. Display-time seems to be far the worst.
2885 The only known fix: Don't run display-time.
2889 SunOS 4.1.4 stopped shipping on Sep 30 1998.
2891 **** SunOS: You get linker errors
2892 ld: Undefined symbol
2893 _get_wmShellWidgetClass
2894 _get_applicationShellWidgetClass
2896 **** Sun 4.0.x: M-x shell persistently reports "Process shell exited abnormally with code 1".
2898 This happened on Suns as a result of what is said to be a bug in Sunos
2899 version 4.0.x. The only fix was to reboot the machine.
2901 **** SunOS4.1.1 and SunOS4.1.3: Mail is lost when sent to local aliases.
2903 Many emacs mail user agents (VM and rmail, for instance) use the
2904 sendmail.el library. This library can arrange for mail to be
2905 delivered by passing messages to the /usr/lib/sendmail (usually)
2906 program . In doing so, it passes the '-t' flag to sendmail, which
2907 means that the name of the recipient of the message is not on the
2908 command line and, therefore, that sendmail must parse the message to
2909 obtain the destination address.
2911 There is a bug in the SunOS4.1.1 and SunOS4.1.3 versions of sendmail.
2912 In short, when given the -t flag, the SunOS sendmail won't recognize
2913 non-local (i.e. NIS) aliases. It has been reported that the Solaris
2914 2.x versions of sendmail do not have this bug. For those using SunOS
2915 4.1, the best fix is to install sendmail V8 or IDA sendmail (which
2916 have other advantages over the regular sendmail as well). At the time
2917 of this writing, these official versions are available:
2919 Sendmail V8 on ftp.cs.berkeley.edu in /ucb/sendmail:
2920 sendmail.8.6.9.base.tar.Z (the base system source & documentation)
2921 sendmail.8.6.9.cf.tar.Z (configuration files)
2922 sendmail.8.6.9.misc.tar.Z (miscellaneous support programs)
2923 sendmail.8.6.9.xdoc.tar.Z (extended documentation, with postscript)
2925 IDA sendmail on vixen.cso.uiuc.edu in /pub:
2926 sendmail-5.67b+IDA-1.5.tar.gz
2928 **** Sunos 4: You get the error ld: Undefined symbol __lib_version.
2930 This is the result of using cc or gcc with the shared library meant
2931 for acc (the Sunpro compiler). Check your LD_LIBRARY_PATH and delete
2932 /usr/lang/SC2.0.1 or some similar directory.
2934 **** SunOS 4.1.3: Emacs unpredictably crashes in _yp_dobind_soft.
2936 This happens if you configure Emacs specifying just `sparc-sun-sunos4'
2937 on a system that is version 4.1.3. You must specify the precise
2938 version number (or let configure figure out the configuration, which
2939 it can do perfectly well for SunOS).
2941 **** Sunos 4.1.3: Emacs gets hung shortly after startup.
2943 We think this is due to a bug in Sunos. The word is that
2944 one of these Sunos patches fixes the bug:
2946 100075-11 100224-06 100347-03 100482-05 100557-02 100623-03 100804-03 101080-01
2947 100103-12 100249-09 100496-02 100564-07 100630-02 100891-10 101134-01
2948 100170-09 100296-04 100377-09 100507-04 100567-04 100650-02 101070-01 101145-01
2949 100173-10 100305-15 100383-06 100513-04 100570-05 100689-01 101071-03 101200-02
2950 100178-09 100338-05 100421-03 100536-02 100584-05 100784-01 101072-01 101207-01
2952 We don't know which of these patches really matter. If you find out
2953 which ones, please inform bug-gnu-emacs@gnu.org.
2955 **** SunOS 4: Emacs processes keep going after you kill the X server
2956 (or log out, if you logged in using X).
2958 Someone reported that recompiling with GCC 2.7.0 fixed this problem.
2960 The fix to this is to install patch 100573 for OpenWindows 3.0
2961 or link libXmu statically.
2963 **** Sunos 5.3: Subprocesses remain, hanging but not zombies.
2965 A bug in Sunos 5.3 causes Emacs subprocesses to remain after Emacs
2966 exits. Sun patch # 101415-02 is part of the fix for this, but it only
2967 applies to ptys, and doesn't fix the problem with subprocesses
2968 communicating through pipes.
2972 **** Shell mode ignores interrupts on Apollo Domain.
2974 You may find that M-x shell prints the following message:
2976 Warning: no access to tty; thus no job control in this shell...
2978 This can happen if there are not enough ptys on your system.
2979 Here is how to make more of them.
2983 # shows how many pty's you have. I had 8, named pty0 to pty7)
2985 # creates eight new pty's
2989 *** Irix 6.2: No visible display on mips-sgi-irix6.2 when compiling with GCC 2.8.1.
2991 This problem went away after installing the latest IRIX patches
2994 The same problem has been reported on Irix 6.3.
2996 *** Irix 6.3: substituting environment variables in file names
2997 in the minibuffer gives peculiar error messages such as
2999 Substituting nonexistent environment variable ""
3001 This is not an Emacs bug; it is caused by something in SGI patch
3002 003082 August 11, 1998.
3006 **** OPENSTEP 4.2: Compiling syntax.c with gcc 2.7.2.1 fails.
3008 The compiler was reported to crash while compiling syntax.c with the
3011 cc: Internal compiler error: program cc1obj got fatal signal 11
3013 To work around this, replace the macros UPDATE_SYNTAX_TABLE_FORWARD,
3014 INC_BOTH, and INC_FROM with functions. To this end, first define 3
3015 functions, one each for every macro. Here's an example:
3017 static int update_syntax_table_forward(int from)
3019 return(UPDATE_SYNTAX_TABLE_FORWARD(from));
3020 }/*update_syntax_table_forward*/
3022 Then replace all references to UPDATE_SYNTAX_TABLE_FORWARD in syntax.c
3023 with a call to the function update_syntax_table_forward.
3027 **** Strange results from format %d in a few cases, on a Sun.
3029 Sun compiler version SC3.0 has been found to miscompile part of
3030 editfns.c. The workaround is to compile with some other compiler such
3033 **** On Solaris, Emacs dumps core if lisp-complete-symbol is called.
3035 If you compile Emacs with the -fast or -xO4 option with version 3.0.2
3036 of the Sun C compiler, Emacs dumps core when lisp-complete-symbol is
3037 called. The problem does not happen if you compile with GCC.
3039 **** On Solaris, Emacs crashes if you use (display-time).
3041 This can happen if you configure Emacs without specifying the precise
3042 version of Solaris that you are using.
3044 **** Solaris 2.3 and 2.4: Unpredictable segmentation faults.
3046 A user reported that this happened in 19.29 when it was compiled with
3047 the Sun compiler, but not when he recompiled with GCC 2.7.0.
3049 We do not know whether something in Emacs is partly to blame for this.
3051 **** Solaris 2.4: Emacs dumps core on startup.
3053 Bill Sebok says that the cause of this is Solaris 2.4 vendor patch
3054 102303-05, which extends the Solaris linker to deal with the Solaris
3055 Common Desktop Environment's linking needs. You can fix the problem
3056 by removing this patch and installing patch 102049-02 instead.
3057 However, that linker version won't work with CDE.
3059 Solaris 2.5 comes with a linker that has this bug. It is reported that if
3060 you install all the latest patches (as of June 1996), the bug is fixed.
3061 We suspect the crucial patch is one of these, but we don't know
3064 103093-03: [README] SunOS 5.5: kernel patch (2140557 bytes)
3065 102832-01: [README] OpenWindows 3.5: Xview Jumbo Patch (4181613 bytes)
3066 103242-04: [README] SunOS 5.5: linker patch (595363 bytes)
3068 (One user reports that the bug was fixed by those patches together
3069 with patches 102980-04, 103279-01, 103300-02, and 103468-01.)
3071 If you can determine which patch does fix the bug, please tell
3072 bug-gnu-emacs@gnu.org.
3074 Meanwhile, the GNU linker links Emacs properly on both Solaris 2.4 and
3077 **** Solaris 2.4: Dired hangs and C-g does not work. Or Emacs hangs
3078 forever waiting for termination of a subprocess that is a zombie.
3080 casper@fwi.uva.nl says the problem is in X11R6. Rebuild libX11.so
3081 after changing the file xc/config/cf/sunLib.tmpl. Change the lines
3084 #define SharedX11Reqs -lthread
3089 #if OSMinorVersion < 4
3091 #define SharedX11Reqs -lthread
3095 Be sure also to edit x/config/cf/sun.cf so that OSMinorVersion is 4
3096 (as it should be for Solaris 2.4). The file has three definitions for
3097 OSMinorVersion: the first is for x86, the second for SPARC under
3098 Solaris, and the third for SunOS 4. Make sure to update the
3099 definition for your type of machine and system.
3101 Then do `make Everything' in the top directory of X11R6, to rebuild
3102 the makefiles and rebuild X. The X built this way work only on
3103 Solaris 2.4, not on 2.3.
3105 For multithreaded X to work it is necessary to install patch
3106 101925-02 to fix problems in header files [2.4]. You need
3107 to reinstall gcc or re-run just-fixinc after installing that
3110 However, Frank Rust <frust@iti.cs.tu-bs.de> used a simpler solution:
3112 #define ThreadedX YES
3114 #define ThreadedX NO
3115 in sun.cf and did `make World' to rebuild X11R6. Removing all
3116 `-DXTHREAD*' flags and `-lthread' entries from lib/X11/Makefile and
3117 typing 'make install' in that directory also seemed to work.
3119 **** Solaris 2.x: GCC complains "64 bit integer types not supported".
3121 This suggests that GCC is not installed correctly. Most likely you
3122 are using GCC 2.7.2.3 (or earlier) on Solaris 2.6 (or later); this
3123 does not work without patching. To run GCC 2.7.2.3 on Solaris 2.6 or
3124 later, you must patch fixinc.svr4 and reinstall GCC from scratch as
3125 described in the Solaris FAQ
3126 <http://www.wins.uva.nl/pub/solaris/solaris2.html>. A better fix is
3127 to upgrade to GCC 2.8.1 or later.
3129 **** Solaris 2.7: Building Emacs with WorkShop Compilers 5.0 98/12/15
3130 C 5.0 failed, apparently with non-default CFLAGS, most probably due to
3131 compiler bugs. Using Sun Solaris 2.7 Sun WorkShop 6 update 1 C
3132 release was reported to work without problems. It worked OK on
3133 another system with Solaris 8 using apparently the same 5.0 compiler
3134 and the default CFLAGS.
3136 **** Solaris 2.x: Emacs dumps core when built with Motif.
3138 The Solaris Motif libraries are buggy, at least up through Solaris 2.5.1.
3139 Install the current Motif runtime library patch appropriate for your host.
3140 (Make sure the patch is current; some older patch versions still have the bug.)
3141 You should install the other patches recommended by Sun for your host, too.
3142 You can obtain Sun patches from ftp://sunsolve.sun.com/pub/patches/;
3143 look for files with names ending in `.PatchReport' to see which patches
3144 are currently recommended for your host.
3146 On Solaris 2.6, Emacs is said to work with Motif when Solaris patch
3147 105284-12 is installed, but fail when 105284-15 is installed.
3148 105284-18 might fix it again.
3150 **** Solaris 2.6 and 7: the Compose key does not work.
3152 This is a bug in Motif in Solaris. Supposedly it has been fixed for
3153 the next major release of Solaris. However, if someone with Sun
3154 support complains to Sun about the bug, they may release a patch.
3155 If you do this, mention Sun bug #4188711.
3157 One workaround is to use a locale that allows non-ASCII characters.
3158 For example, before invoking emacs, set the LC_ALL environment
3159 variable to "en_US" (American English). The directory /usr/lib/locale
3160 lists the supported locales; any locale other than "C" or "POSIX"
3163 pen@lysator.liu.se says (Feb 1998) that the Compose key does work
3164 if you link with the MIT X11 libraries instead of the Solaris X11
3167 *** HP/UX versions before 11.0
3169 HP/UX 9 was end-of-lifed in December 1998.
3170 HP/UX 10 was end-of-lifed in May 1999.
3172 **** HP/UX 9: Emacs crashes with SIGBUS or SIGSEGV after you delete a frame.
3174 We think this is due to a bug in the X libraries provided by HP. With
3175 the alternative X libraries in /usr/contrib/mitX11R5/lib, the problem
3178 *** HP/UX 10: Large file support is disabled.
3180 See the comments in src/s/hpux10.h.
3182 *** HP/UX: Emacs is slow using X11R5.
3184 This happens if you use the MIT versions of the X libraries--it
3185 doesn't run as fast as HP's version. People sometimes use the version
3186 because they see the HP version doesn't have the libraries libXaw.a,
3187 libXmu.a, libXext.a and others. HP/UX normally doesn't come with
3188 those libraries installed. To get good performance, you need to
3189 install them and rebuild Emacs.
3191 *** Ultrix and Digital Unix
3193 **** Ultrix 4.2: `make install' fails on install-doc with `Error 141'.
3195 This happens on Ultrix 4.2 due to failure of a pipeline of tar
3196 commands. We don't know why they fail, but the bug seems not to be in
3197 Emacs. The workaround is to run the shell command in install-doc by
3200 **** Digital Unix 4.0: Garbled display on non-X terminals when Emacs runs.
3202 So far it appears that running `tset' triggers this problem (when TERM
3203 is vt100, at least). If you do not run `tset', then Emacs displays
3204 properly. If someone can tell us precisely which effect of running
3205 `tset' actually causes the problem, we may be able to implement a fix
3208 **** Ultrix: `expand-file-name' fails to work on any but the machine you dumped Emacs on.
3210 On Ultrix, if you use any of the functions which look up information
3211 in the passwd database before dumping Emacs (say, by using
3212 expand-file-name in site-init.el), then those functions will not work
3213 in the dumped Emacs on any host but the one Emacs was dumped on.
3215 The solution? Don't use expand-file-name in site-init.el, or in
3216 anything it loads. Yuck - some solution.
3218 I'm not sure why this happens; if you can find out exactly what is
3219 going on, and perhaps find a fix or a workaround, please let us know.
3220 Perhaps the YP functions cache some information, the cache is included
3221 in the dumped Emacs, and is then inaccurate on any other host.
3225 **** SVr4: On some variants of SVR4, Emacs does not work at all with X.
3227 Try defining BROKEN_FIONREAD in your config.h file. If this solves
3228 the problem, please send a bug report to tell us this is needed; be
3229 sure to say exactly what type of machine and system you are using.
3231 **** SVr4: After running emacs once, subsequent invocations crash.
3233 Some versions of SVR4 have a serious bug in the implementation of the
3234 mmap () system call in the kernel; this causes emacs to run correctly
3235 the first time, and then crash when run a second time.
3237 Contact your vendor and ask for the mmap bug fix; in the mean time,
3238 you may be able to work around the problem by adding a line to your
3239 operating system description file (whose name is reported by the
3240 configure script) that reads:
3241 #define SYSTEM_MALLOC
3242 This makes Emacs use memory less efficiently, but seems to work around
3245 *** Irix 5 and earlier
3247 Exactly when Irix-5 end-of-lifed is obscure. But since Irix 6.0
3248 shipped in 1994, it has been some years.
3250 **** Irix 5.2: unexelfsgi.c can't find cmplrs/stsupport.h.
3252 The file cmplrs/stsupport.h was included in the wrong file set in the
3253 Irix 5.2 distribution. You can find it in the optional fileset
3254 compiler_dev, or copy it from some other Irix 5.2 system. A kludgy
3255 workaround is to change unexelfsgi.c to include sym.h instead of
3258 **** Irix 5.3: "out of virtual swap space".
3260 This message occurs when the system runs out of swap space due to too
3261 many large programs running. The solution is either to provide more
3262 swap space or to reduce the number of large programs being run. You
3263 can check the current status of the swap space by executing the
3266 You can increase swap space by changing the file /etc/fstab. Adding a
3269 /usr/swap/swap.more swap swap pri=3 0 0
3271 where /usr/swap/swap.more is a file previously created (for instance
3272 by using /etc/mkfile), will increase the swap space by the size of
3273 that file. Execute `swap -m' or reboot the machine to activate the
3274 new swap area. See the manpages for `swap' and `fstab' for further
3277 The objectserver daemon can use up lots of memory because it can be
3278 swamped with NIS information. It collects information about all users
3279 on the network that can log on to the host.
3281 If you want to disable the objectserver completely, you can execute
3282 the command `chkconfig objectserver off' and reboot. That may disable
3283 some of the window system functionality, such as responding CDROM
3286 You can also remove NIS support from the objectserver. The SGI `admin'
3287 FAQ has a detailed description on how to do that; see question 35
3288 ("Why isn't the objectserver working?"). The admin FAQ can be found at
3289 ftp://viz.tamu.edu/pub/sgi/faq/.
3291 **** Irix 5.3: Emacs crashes in utmpname.
3293 This problem is fixed in Patch 3175 for Irix 5.3.
3294 It is also fixed in Irix versions 6.2 and up.
3296 **** Irix 6.0: Make tries (and fails) to build a program named unexelfsgi.
3298 A compiler bug inserts spaces into the string "unexelfsgi . o"
3299 in src/Makefile. Edit src/Makefile, after configure is run,
3300 find that string, and take out the spaces.
3302 Compiler fixes in Irix 6.0.1 should eliminate this problem.
3304 *** SCO Unix and UnixWare
3306 **** SCO 3.2v4: Unusable default font.
3308 The Open Desktop environment comes with default X resource settings
3309 that tell Emacs to use a variable-width font. Emacs cannot use such
3310 fonts, so it does not work.
3312 This is caused by the file /usr/lib/X11/app-defaults/ScoTerm, which is
3313 the application-specific resource file for the `scoterm' terminal
3314 emulator program. It contains several extremely general X resources
3315 that affect other programs besides `scoterm'. In particular, these
3316 resources affect Emacs also:
3318 *Font: -*-helvetica-medium-r-*--12-*-p-*
3319 *Background: scoBackground
3320 *Foreground: scoForeground
3322 The best solution is to create an application-specific resource file for
3323 Emacs, /usr/lib/X11/sco/startup/Emacs, with the following contents:
3325 Emacs*Font: -*-courier-medium-r-*-*-*-120-*-*-*-*-iso8859-1
3326 Emacs*Background: white
3327 Emacs*Foreground: black
3329 (These settings mimic the Emacs defaults, but you can change them to
3330 suit your needs.) This resource file is only read when the X server
3331 starts up, so you should restart it by logging out of the Open Desktop
3332 environment or by running `scologin stop; scologin start` from the shell
3333 as root. Alternatively, you can put these settings in the
3334 /usr/lib/X11/app-defaults/Emacs resource file and simply restart Emacs,
3335 but then they will not affect remote invocations of Emacs that use the
3336 Open Desktop display.
3338 These resource files are not normally shared across a network of SCO
3339 machines; you must create the file on each machine individually.
3341 **** SCO 4.2.0: Regular expressions matching bugs on SCO systems.
3343 On SCO, there are problems in regexp matching when Emacs is compiled
3344 with the system compiler. The compiler version is "Microsoft C
3345 version 6", SCO 4.2.0h Dev Sys Maintenance Supplement 01/06/93; Quick
3346 C Compiler Version 1.00.46 (Beta). The solution is to compile with
3349 **** UnixWare 2.1: Error 12 (virtual memory exceeded) when dumping Emacs.
3351 Paul Abrahams (abrahams@acm.org) reports that with the installed
3352 virtual memory settings for UnixWare 2.1.2, an Error 12 occurs during
3353 the "make" that builds Emacs, when running temacs to dump emacs. That
3354 error indicates that the per-process virtual memory limit has been
3355 exceeded. The default limit is probably 32MB. Raising the virtual
3356 memory limit to 40MB should make it possible to finish building Emacs.
3358 You can do this with the command `ulimit' (sh) or `limit' (csh).
3359 But you have to be root to do it.
3361 According to Martin Sohnius, you can also retune this in the kernel:
3363 # /etc/conf/bin/idtune SDATLIM 33554432 ## soft data size limit
3364 # /etc/conf/bin/idtune HDATLIM 33554432 ## hard "
3365 # /etc/conf/bin/idtune SVMMSIZE unlimited ## soft process size limit
3366 # /etc/conf/bin/idtune HVMMSIZE unlimited ## hard "
3367 # /etc/conf/bin/idbuild -B
3369 (He recommends you not change the stack limit, though.)
3370 These changes take effect when you reboot.
3374 **** Linux 1.0-1.04: Typing C-c C-c in Shell mode kills your X server.
3376 This happens with Linux kernel 1.0 thru 1.04, approximately. The workaround is
3377 to define SIGNALS_VIA_CHARACTERS in config.h and recompile Emacs.
3378 Newer Linux kernel versions don't have this problem.
3380 **** Linux 1.3: Output from subprocess (such as man or diff) is randomly
3381 truncated on GNU/Linux systems.
3383 This is due to a kernel bug which seems to be fixed in Linux version
3386 ** Windows 3.1, 95, 98, and ME
3388 *** MS-Windows NT/95: Problems running Perl under Emacs
3390 `perl -de 0' just hangs when executed in an Emacs subshell.
3391 The fault lies with Perl (indirectly with Windows NT/95).
3393 The problem is that the Perl debugger explicitly opens a connection to
3394 "CON", which is the DOS/NT equivalent of "/dev/tty", for interacting
3397 On Unix, this is okay, because Emacs (or the shell?) creates a
3398 pseudo-tty so that /dev/tty is really the pipe Emacs is using to
3399 communicate with the subprocess.
3401 On NT, this fails because CON always refers to the handle for the
3402 relevant console (approximately equivalent to a tty), and cannot be
3403 redirected to refer to the pipe Emacs assigned to the subprocess as
3406 A workaround is to modify perldb.pl to use STDIN/STDOUT instead of CON.
3410 *** PERL/LIB/PERLDB.PL.orig Wed May 26 08:24:18 1993
3411 --- PERL/LIB/PERLDB.PL Mon Jul 01 15:28:16 1996
3418 $rcfile="perldb.ini";
3426 $rcfile="perldb.ini";
3431 *** perl/5.001/lib/perl5db.pl.orig Sun Jun 04 21:13:40 1995
3432 --- perl/5.001/lib/perl5db.pl Mon Jul 01 17:00:08 1996
3439 $rcfile="perldb.ini";
3447 $rcfile="perldb.ini";
3451 *** MS-Windows 95: Alt-f6 does not get through to Emacs.
3453 This character seems to be trapped by the kernel in Windows 95.
3454 You can enter M-f6 by typing ESC f6.
3456 *** MS-Windows 95/98/ME: subprocesses do not terminate properly.
3458 This is a limitation of the Operating System, and can cause problems
3459 when shutting down Windows. Ensure that all subprocesses are exited
3460 cleanly before exiting Emacs. For more details, see the FAQ at
3461 http://www.gnu.org/software/emacs/windows/.
3463 *** MS-Windows 95/98/ME: crashes when Emacs invokes non-existent programs.
3465 When a program you are trying to run is not found on the PATH,
3466 Windows might respond by crashing or locking up your system. In
3467 particular, this has been reported when trying to compile a Java
3468 program in JDEE when javac.exe is installed, but not on the system
3473 *** When compiling with DJGPP on MS-Windows NT, "config msdos" fails.
3475 If the error message is "VDM has been already loaded", this is because
3476 Windows has a program called `redir.exe' that is incompatible with a
3477 program by the same name supplied with DJGPP, which is used by
3478 config.bat. To resolve this, move the DJGPP's `bin' subdirectory to
3479 the front of your PATH environment variable.
3481 *** When compiling with DJGPP on MS-Windows 95, Make fails for some targets
3484 This can happen if long file name support (the setting of environment
3485 variable LFN) when Emacs distribution was unpacked and during
3486 compilation are not the same. See the MSDOG section of INSTALL for
3487 the explanation of how to avoid this problem.
3489 *** Emacs compiled with DJGPP complains at startup:
3491 "Wrong type of argument: internal-facep, msdos-menu-active-face"
3493 This can happen if you define an environment variable `TERM'. Emacs
3494 on MSDOS uses an internal terminal emulator which is disabled if the
3495 value of `TERM' is anything but the string "internal". Emacs then
3496 works as if its terminal were a dumb glass teletype that doesn't
3497 support faces. To work around this, arrange for `TERM' to be
3498 undefined when Emacs runs. The best way to do that is to add an
3499 [emacs] section to the DJGPP.ENV file which defines an empty value for
3500 `TERM'; this way, only Emacs gets the empty value, while the rest of
3501 your system works as before.
3503 *** MS-DOS: Emacs crashes at startup.
3505 Some users report that Emacs 19.29 requires dpmi memory management,
3506 and crashes on startup if the system does not have it. We don't yet
3507 know why this happens--perhaps these machines don't have enough real
3508 memory, or perhaps something is wrong in Emacs or the compiler.
3509 However, arranging to use dpmi support is a workaround.
3511 You can find out if you have a dpmi host by running go32 without
3512 arguments; it will tell you if it uses dpmi memory. For more
3513 information about dpmi memory, consult the djgpp FAQ. (djgpp
3514 is the GNU C compiler as packaged for MSDOS.)
3516 Compiling Emacs under MSDOS is extremely sensitive for proper memory
3517 configuration. If you experience problems during compilation, consider
3518 removing some or all memory resident programs (notably disk caches)
3519 and make sure that your memory managers are properly configured. See
3520 the djgpp faq for configuration hints.
3522 *** Emacs compiled with DJGPP for MS-DOS/MS-Windows cannot access files
3523 in the directory with the special name `dev' under the root of any
3524 drive, e.g. `c:/dev'.
3526 This is an unfortunate side-effect of the support for Unix-style
3527 device names such as /dev/null in the DJGPP runtime library. A
3528 work-around is to rename the problem directory to another name.
3530 *** MS-DOS+DJGPP: Problems on MS-DOG if DJGPP v2.0 is used to compile Emacs.
3532 There are two DJGPP library bugs which cause problems:
3534 * Running `shell-command' (or `compile', or `grep') you get
3535 `Searching for program: permission denied (EACCES), c:/command.com';
3536 * After you shell to DOS, Ctrl-Break kills Emacs.
3538 To work around these bugs, you can use two files in the msdos
3539 subdirectory: `is_exec.c' and `sigaction.c'. Compile them and link
3540 them into the Emacs executable `temacs'; then they will replace the
3541 incorrect library functions.
3543 *** MS-DOS: Emacs compiled for MSDOS cannot find some Lisp files, or other
3544 run-time support files, when long filename support is enabled.
3546 Usually, this problem will manifest itself when Emacs exits
3547 immediately after flashing the startup screen, because it cannot find
3548 the Lisp files it needs to load at startup. Redirect Emacs stdout
3549 and stderr to a file to see the error message printed by Emacs.
3551 Another manifestation of this problem is that Emacs is unable to load
3552 the support for editing program sources in languages such as C and
3555 This can happen if the Emacs distribution was unzipped without LFN
3556 support, thus causing long filenames to be truncated to the first 6
3557 characters and a numeric tail that Windows 95 normally attaches to it.
3558 You should unzip the files again with a utility that supports long
3559 filenames (such as djtar from DJGPP or InfoZip's UnZip program
3560 compiled with DJGPP v2). The MSDOG section of the file INSTALL
3561 explains this issue in more detail.
3563 Another possible reason for such failures is that Emacs compiled for
3564 MSDOS is used on Windows NT, where long file names are not supported
3565 by this version of Emacs, but the distribution was unpacked by an
3566 unzip program that preserved the long file names instead of truncating
3567 them to DOS 8+3 limits. To be useful on NT, the MSDOS port of Emacs
3568 must be unzipped by a DOS utility, so that long file names are
3571 ** Archaic window managers and toolkits
3573 *** OpenLook: Under OpenLook, the Emacs window disappears when you type M-q.
3575 Some versions of the Open Look window manager interpret M-q as a quit
3576 command for whatever window you are typing at. If you want to use
3577 Emacs with that window manager, you should try to configure the window
3578 manager to use some other command. You can disable the
3579 shortcut keys entirely by adding this line to ~/.OWdefaults:
3581 OpenWindows.WindowMenuAccelerators: False
3583 **** twm: A position you specified in .Xdefaults is ignored, using twm.
3585 twm normally ignores "program-specified" positions.
3586 You can tell it to obey them with this command in your `.twmrc' file:
3588 UsePPosition "on" #allow clients to request a position
3590 ** Bugs related to old DEC hardware
3592 *** The Compose key on a DEC keyboard does not work as Meta key.
3594 This shell command should fix it:
3596 xmodmap -e 'keycode 0xb1 = Meta_L'
3598 *** Keyboard input gets confused after a beep when using a DECserver
3601 This problem seems to be a matter of configuring the DECserver to use
3602 7 bit characters rather than 8 bit characters.
3604 * Build problems on legacy systems
3606 ** BSD/386 1.0: --with-x-toolkit option configures wrong.
3608 This problem is due to bugs in the shell in version 1.0 of BSD/386.
3609 The workaround is to edit the configure file to use some other shell,
3612 ** Digital Unix 4.0: Emacs fails to build, giving error message
3613 Invalid dimension for the charset-ID 160
3615 This is due to a bug or an installation problem in GCC 2.8.0.
3616 Installing a more recent version of GCC fixes the problem.
3618 ** Digital Unix 4.0: Failure in unexec while dumping emacs.
3620 This problem manifests itself as an error message
3622 unexec: Bad address, writing data section to ...
3624 The user suspects that this happened because his X libraries
3625 were built for an older system version,
3627 ./configure --x-includes=/usr/include --x-libraries=/usr/shlib
3629 made the problem go away.
3631 ** Sunos 4.1.1: there are errors compiling sysdep.c.
3633 If you get errors such as
3635 "sysdep.c", line 2017: undefined structure or union
3636 "sysdep.c", line 2017: undefined structure or union
3637 "sysdep.c", line 2019: nodename undefined
3639 This can result from defining LD_LIBRARY_PATH. It is very tricky
3640 to use that environment variable with Emacs. The Emacs configure
3641 script links many test programs with the system libraries; you must
3642 make sure that the libraries available to configure are the same
3643 ones available when you build Emacs.
3645 ** SunOS 4.1.1: You get this error message from GNU ld:
3647 /lib/libc.a(_Q_sub.o): Undefined symbol __Q_get_rp_rd referenced from text segment
3649 The problem is in the Sun shared C library, not in GNU ld.
3651 The solution is to install Patch-ID# 100267-03 from Sun.
3653 ** Sunos 4.1: Undefined symbols when linking using --with-x-toolkit.
3655 If you get the undefined symbols _atowc _wcslen, _iswprint, _iswspace,
3656 _iswcntrl, _wcscpy, and _wcsncpy, then you need to add -lXwchar after
3657 -lXaw in the command that links temacs.
3659 This problem seems to arise only when the international language
3660 extensions to X11R5 are installed.
3662 ** SunOS: Emacs gets error message from linker on Sun.
3664 If the error message says that a symbol such as `f68881_used' or
3665 `ffpa_used' or `start_float' is undefined, this probably indicates
3666 that you have compiled some libraries, such as the X libraries,
3667 with a floating point option other than the default.
3669 It's not terribly hard to make this work with small changes in
3670 crt0.c together with linking with Fcrt1.o, Wcrt1.o or Mcrt1.o.
3671 However, the easiest approach is to build Xlib with the default
3672 floating point option: -fsoft.
3674 ** SunOS: Undefined symbols _dlopen, _dlsym and/or _dlclose.
3676 If you see undefined symbols _dlopen, _dlsym, or _dlclose when linking
3677 with -lX11, compile and link against the file mit/util/misc/dlsym.c in
3678 the MIT X11R5 distribution. Alternatively, link temacs using shared
3679 libraries with s/sunos4shr.h. (This doesn't work if you use the X
3682 If you get the additional error that the linker could not find
3683 lib_version.o, try extracting it from X11/usr/lib/X11/libvim.a in
3684 X11R4, then use it in the link.
3686 ** SunOS4, DGUX 5.4.2: --with-x-toolkit version crashes when used with shared libraries.
3688 On some systems, including Sunos 4 and DGUX 5.4.2 and perhaps others,
3689 unexec doesn't work properly with the shared library for the X
3690 toolkit. You might be able to work around this by using a nonshared
3691 libXt.a library. The real fix is to upgrade the various versions of
3692 unexec and/or ralloc. We think this has been fixed on Sunos 4
3693 and Solaris in version 19.29.
3695 ** HPUX 10.20: Emacs crashes during dumping on the HPPA machine.
3697 This seems to be due to a GCC bug; it is fixed in GCC 2.8.1.
3699 ** VMS: Compilation errors on VMS.
3701 You will get warnings when compiling on VMS because there are
3702 variable names longer than 32 (or whatever it is) characters.
3703 This is not an error. Ignore it.
3705 VAX C does not support #if defined(foo). Uses of this construct
3706 were removed, but some may have crept back in. They must be rewritten.
3708 There is a bug in the C compiler which fails to sign extend characters
3709 in conditional expressions. The bug is:
3714 The result is i == 255; the fix is to typecast the char in the
3715 conditional expression as an (int). Known occurrences of such
3716 constructs in Emacs have been fixed.
3718 ** Vax C compiler bugs affecting Emacs.
3720 You may get one of these problems compiling Emacs:
3722 foo.c line nnn: compiler error: no table entry for op STASG
3723 foo.c: fatal error in /lib/ccom
3725 These are due to bugs in the C compiler; the code is valid C.
3726 Unfortunately, the bugs are unpredictable: the same construct
3727 may compile properly or trigger one of these bugs, depending
3728 on what else is in the source file being compiled. Even changes
3729 in header files that should not affect the file being compiled
3730 can affect whether the bug happens. In addition, sometimes files
3731 that compile correctly on one machine get this bug on another machine.
3733 As a result, it is hard for me to make sure this bug will not affect
3734 you. I have attempted to find and alter these constructs, but more
3735 can always appear. However, I can tell you how to deal with it if it
3736 should happen. The bug comes from having an indexed reference to an
3737 array of Lisp_Objects, as an argument in a function call:
3740 ... foo (5, args[i], ...)...
3741 putting the argument into a temporary variable first, as in
3746 ... foo (r, tem, ...)...
3747 causes the problem to go away.
3748 The `contents' field of a Lisp vector is an array of Lisp_Objects,
3749 so you may see the problem happening with indexed references to that.
3751 ** 68000 C compiler problems
3753 Various 68000 compilers have different problems.
3754 These are some that have been observed.
3756 *** Using value of assignment expression on union type loses.
3757 This means that x = y = z; or foo (x = z); does not work
3758 if x is of type Lisp_Object.
3760 *** "cannot reclaim" error.
3762 This means that an expression is too complicated. You get the correct
3763 line number in the error message. The code must be rewritten with
3764 simpler expressions.
3766 *** XCONS, XSTRING, etc macros produce incorrect code.
3768 If temacs fails to run at all, this may be the cause.
3769 Compile this test program and look at the assembler code:
3771 struct foo { char x; unsigned int y : 24; };
3776 test ((int *) arg.y);
3779 If the code is incorrect, your compiler has this problem.
3780 In the XCONS, etc., macros in lisp.h you must replace (a).u.val with
3781 ((a).u.val + coercedummy) where coercedummy is declared as int.
3783 This problem will not happen if the m-...h file for your type
3784 of machine defines NO_UNION_TYPE. That is the recommended setting now.
3786 *** C compilers lose on returning unions.
3788 I hear that some C compilers cannot handle returning a union type.
3789 Most of the functions in GNU Emacs return type Lisp_Object, which is
3790 defined as a union on some rare architectures.
3792 This problem will not happen if the m-...h file for your type
3793 of machine defines NO_UNION_TYPE.
3796 This file is part of GNU Emacs.
3798 GNU Emacs is free software; you can redistribute it and/or modify
3799 it under the terms of the GNU General Public License as published by
3800 the Free Software Foundation; either version 2, or (at your option)
3803 GNU Emacs is distributed in the hope that it will be useful,
3804 but WITHOUT ANY WARRANTY; without even the implied warranty of
3805 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
3806 GNU General Public License for more details.
3808 You should have received a copy of the GNU General Public License
3809 along with GNU Emacs; see the file COPYING. If not, write to the
3810 Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
3811 Boston, MA 02110-1301, USA.
3816 paragraph-separate: "[
\f]*$"
3819 arch-tag: 49fc0d95-88cb-4715-b21c-f27fb5a4764a