3 This file describes various problems that have been encountered
4 in compiling, installing and running XEmacs. It has been updated for
7 This file is rather large, but we have tried to sort the entries by
8 their respective relevance for XEmacs, but may have not succeeded
9 completely in that task. The file is divided into four parts:
11 - Problems with building XEmacs
12 - Problems with running XEmacs
13 - Compatibility problems
16 Use `C-c C-f' to move to the next equal level of outline, and
17 `C-c C-b' to move to previous equal level. `C-h m' will give more
18 info about the Outline mode.
20 Also, Try finding the things you need using one of the search commands
21 XEmacs provides (e.g. `C-s').
24 WATCH OUT for .emacs file! ~/.emacs is your Emacs init file. If
25 you observe strange problems, invoke XEmacs with the `-q' option
26 and see if you can repeat the problem.
29 * Problems with building XEmacs
30 ===============================
33 *** Don't use -O2 with gcc 2.7.2 under Intel/XXX without also using
34 `-fno-strength-reduce'.
36 gcc will generate incorrect code otherwise. This bug is present in at
37 least 2.6.x and 2.7.[0-2]. This bug has been fixed in GCC 2.7.2.1 and
38 later. This bug is O/S independent, but is limited to x86 architectures.
40 This problem is known to be fixed in egcs (or pgcc) 1.0 or later.
42 *** Don't use -O2 with gcc 2.7.2 under Intel architectures without also
43 using `-fno-caller-saves'.
45 gcc will generate incorrect code otherwise. This bug is still
46 present in gcc 2.7.2.3. There have been no reports to indicate the
47 bug is present in egcs 1.0 (or pgcc 1.0) or later. This bug is O/S
48 independent, but limited to x86 architectures.
50 This problem is known to be fixed in egcs (or pgcc) 1.0 or later.
52 *** When using gcc, you get the error message "undefined symbol __fixunsdfsi".
53 When using gcc, you get the error message "undefined symbol __main".
55 This means that you need to link with the gcc library. It may be called
56 "gcc-gnulib" or "libgcc.a"; figure out where it is, and define LIB_GCC in
57 config.h to point to it.
59 It may also work to use the GCC version of `ld' instead of the standard one.
61 *** Excessive optimization with pgcc can break XEmacs
63 It has been reported on some systems that compiling with -O6 can lead
64 to XEmacs failures. The workaround is to use a lower optimization
65 level. -O2 and -O4 have been tested extensively.
67 All of this depends heavily on the version of pgcc and the version
68 of libc. Snapshots near the release of pgcc-1.0 have been tested
69 extensively and no sign of breakage has been seen on systems using
72 *** src/Makefile and lib-src/Makefile are truncated--most of the file missing.
74 This can happen if configure uses GNU sed version 2.03. That version
75 had a bug. GNU sed version 2.05 works properly.
77 *** When compiling with X11, you get "undefined symbol _XtStrings".
79 This means that you are trying to link emacs against the X11r4 version of
80 libXt.a, but you have compiled either Emacs or the code in the lwlib
81 subdirectory with the X11r5 header files. That doesn't work.
83 Remember, you can't compile lwlib for r4 and emacs for r5, or vice versa.
86 *** test-distrib says that the distribution has been clobbered
87 or, temacs prints "Command key out of range 0-127"
88 or, temacs runs and dumps xemacs, but xemacs totally fails to work.
89 or, temacs gets errors dumping xemacs
91 This can be because the .elc files have been garbled. Do not be
92 fooled by the fact that most of a .elc file is text: these are binary
93 files and can contain all 256 byte values.
95 In particular `shar' cannot be used for transmitting GNU Emacs. It
96 typically truncates "lines". (this does not apply to GNU shar, which
97 uses uuencode to encode binary files.)
99 If you have a copy of Emacs that has been damaged in its nonprinting
100 characters, you can fix them by running:
104 This will rebuild all the needed .elc files.
106 *** `compress' and `uncompress' not found and XFree86
108 XFree86 installs a very old version of libz.a by default ahead of where
109 more modern version of libz might be installed. This will cause problems
110 when attempting to link against libMagick. The fix is to remove the old
111 libz.a in the X11 binary directory.
115 *** On AIX 4.3, you must specify --with-dialogs=athena with configure
117 *** On AIX, you get this compiler error message:
119 Processing include file ./XMenuInt.h
120 1501-106: (S) Include file X11/Xlib.h not found.
122 This means your system was installed with only the X11 runtime i.d
123 libraries. You have to find your sipo (bootable tape) and install
126 *** On AIX 4.1.2, linker error messages such as
127 ld: 0711-212 SEVERE ERROR: Symbol .__quous, found in the global symbol table
128 of archive /usr/lib/libIM.a, was not defined in archive member shr.o.
130 This is a problem in libIM.a. You can work around it by executing
131 these shell commands in the src subdirectory of the directory where
134 cp /usr/lib/libIM.a .
138 Then change -lIM to ./libIM.a in the command to link temacs (in
141 *** Excessive optimization on AIX 4.2 can lead to compiler failure.
143 Valdis.Kletnieks@vt.edu writes:
144 At least at the b34 level, and the latest-and-greatest IBM xlc
145 (3.1.4.4), there are problems with -O3. I haven't investigated
150 *** Link failure when using acc on a Sun.
152 To use acc, you need additional options just before the libraries, such as
154 /usr/lang/SC2.0.1/values-Xt.o -L/usr/lang/SC2.0.1/cg87 -L/usr/lang/SC2.0.1
156 and you need to add -lansi just before -lc.
158 The precise file names depend on the compiler version, so we
159 cannot easily arrange to supply them.
161 *** Problems finding X11 libraries on Solaris with Openwindows
163 Some users have reported problems in this area. The reported solution
164 is to define the environment variable OPENWINHOME, even if you must set
165 it to `/usr/openwin'.
167 *** Sed problems on Solaris 2.5
169 There have been reports of Sun sed truncating very lines in the
170 Makefile during configuration. The workaround is to use GNU sed or,
171 even better, think of a better way to generate Makefile, and send us a
174 *** On Solaris 2 I get undefined symbols from libcurses.a.
176 You probably have /usr/ucblib/ on your LD_LIBRARY_PATH. Do the link with
177 LD_LIBRARY_PATH unset. Generally, avoid using any ucb* stuff when
180 *** On Solaris 2 I cannot make alloc.o, glyphs.o or process.o.
182 The SparcWorks C compiler may have difficulty building those modules
183 with optimization level -xO4. Try using only "-fast" optimization
184 for just those modules. (Or use gcc).
186 *** Solaris 2.3 /bin/sh coredumps during configuration.
188 This only occurs if you have LANG != C. This is a known bug with
189 /bin/sh fixed by installing Patch-ID# 101613-01. Or, you can use
190 bash, as a workaround.
192 *** On SunOS, you get linker errors
194 _get_wmShellWidgetClass
195 _get_applicationShellWidgetClass
197 The fix to this is to install patch 100573 for OpenWindows 3.0
198 or link libXmu statically.
200 *** On Sunos 4, you get the error ld: Undefined symbol __lib_version.
202 This is the result of using cc or gcc with the shared library meant
203 for acc (the Sunpro compiler). Check your LD_LIBRARY_PATH and delete
204 /usr/lang/SC2.0.1 or some similar directory.
206 *** Undefined symbols when linking on Sunos 4.1.
208 If you get the undefined symbols _atowc _wcslen, _iswprint, _iswspace,
209 _iswcntrl, _wcscpy, and _wcsncpy, then you need to add -lXwchar after
210 -lXaw in the command that links temacs.
212 This problem seems to arise only when the international language
213 extensions to X11R5 are installed.
215 *** On a Sun running SunOS 4.1.1, you get this error message from GNU ld:
217 /lib/libc.a(_Q_sub.o): Undefined symbol __Q_get_rp_rd referenced from text segment
219 The problem is in the Sun shared C library, not in GNU ld.
221 The solution is to install Patch-ID# 100267-03 from Sun.
223 *** SunOS 4.1.2: undefined symbol _get_wmShellWidgetClass
225 Apparently the version of libXmu.so.a that Sun ships is hosed: it's missing
226 some stuff that is in libXmu.a (the static version). Sun has a patch for
227 this, but a workaround is to use the static version of libXmu, by changing
228 the link command from "-lXmu" to "-Bstatic -lXmu -Bdynamic". If you have
229 OpenWindows 3.0, ask Sun for these patches:
230 100512-02 4.1.x OpenWindows 3.0 libXt Jumbo patch
231 100573-03 4.1.x OpenWindows 3.0 undefined symbols with shared libXmu
233 *** Random other SunOS 4.1.[12] link errors.
235 The X headers and libraries that Sun ships in /usr/{include,lib}/X11 are
236 broken. Use the ones in /usr/openwin/{include,lib} instead.
239 *** Under Linux, you get "too many arguments to function `getpgrp'".
241 You have probably installed LessTiff under `/usr/local' and `libXm.so'
242 could not be found when linking `getpgrp()' test program, making XEmacs
243 think that `getpgrp()' takes an argument. Try adding `/usr/local/lib'
244 in `/etc/ld.so.conf' and run `ldconfig'. Then run XEmacs's `configure'
245 again. As with all problems of this type, reading the config.log file
246 generated from configure and seeing the log of how the test failed can
249 *** `Error: No ExtNode to pop!' on Linux systems with Lesstif.
251 This error message has been observed with lesstif-0.75a. It does not
252 appear to cause any harm.
254 *** xemacs: can't resolve symbol '__malloc_hook'
256 This is a Linux problem where you've compiled the XEmacs binary on a libc
257 5.4 with version higher than 5.4.19 and attempted to run the binary against
258 an earlier version. The solution is to upgrade your old library.
261 *** Linking with -rpath on IRIX.
263 Darrell Kindred <dkindred@cmu.edu> writes:
264 There are a couple of problems [with use of -rpath with Irix ld], though:
266 1. The ld in IRIX 5.3 ignores all but the last -rpath
267 spec, so the patched configure spits out a warning
268 if --x-libraries or --site-runtime-libraries are
269 specified under irix 5.x, and it only adds -rpath
270 entries for the --site-runtime-libraries. This bug was
271 fixed sometime between 5.3 and 6.2.
273 2. IRIX gcc 2.7.2 doesn't accept -rpath directly, so
274 it would have to be prefixed by -Xlinker or "-Wl,".
275 This would be fine, except that configure compiles with
276 ${CC-cc} $CFLAGS $LDFLAGS ...
277 rather than quoting $LDFLAGS with prefix-args, like
278 src/Makefile does. So if you specify --x-libraries
279 or --site-runtime-libraries, you must use --use-gcc=no,
280 or configure will fail.
282 *** On Irix 6.3, the SGI ld quits with segmentation fault when linking temacs
284 This occurs if you use the SGI linker version 7.1. Installing the
285 patch SG0001872 fixes this problem.
287 *** On Irix 6.0, make tries (and fails) to build a program named unexelfsgi
289 A compiler bug inserts spaces into the string "unexelfsgi . o"
290 in src/Makefile. Edit src/Makefile, after configure is run,
291 find that string, and take out the spaces.
293 Compiler fixes in Irix 6.0.1 should eliminate this problem.
295 *** On Irix 5.2, unexelfsgi.c can't find cmplrs/stsupport.h.
297 The file cmplrs/stsupport.h was included in the wrong file set in the
298 Irix 5.2 distribution. You can find it in the optional fileset
299 compiler_dev, or copy it from some other Irix 5.2 system. A kludgy
300 workaround is to change unexelfsgi.c to include sym.h instead of
303 *** Coredumping in Irix 6.2
305 Pete Forman <gsez020@compo.bedford.waii.com> writes:
306 A problem noted by myself and others (I've lost the references) was
307 that XEmacs coredumped when the cut or copy toolbar buttons were
308 pressed. This has been fixed by loading the SGI patchset (Feb 98)
309 without having to recompile XEmacs.
311 My versions are XEmacs 20.3 (problem first noted in 19.15) and IRIX
312 6.2, compiled using -n32. I'd guess that the relevant individual
313 patch was "SG0002580: multiple fixes for X libraries". SGI recommends
314 that the complete patch set be installed rather than parts of it.
316 ** Digital UNIX/OSF/VMS
317 *** On Digital UNIX, the DEC C compiler might have a problem compiling
320 In particular, src/extents.c and src/faces.c might cause the DEC C
321 compiler to abort. When this happens: cd src, compile the files by
322 hand, cd .., and redo the "make" command. When recompiling the files by
323 hand, use the old C compiler for the following versions of Digital UNIX:
324 - V3.n: Remove "-migrate" from the compile command.
325 - V4.n: Add "-oldc" to the compile command.
327 A related compiler bug has been fixed by the DEC compiler team. The
328 new versions of the compiler should run fine.
330 *** Under some versions of OSF XEmacs runs fine if built without
331 optimization but will crash randomly if built with optimization.
333 Using 'cc -g' is not sufficient to eliminate all optimization. Try
336 *** Compilation errors on VMS.
338 Sorry, XEmacs does not work under VMS. You might consider working on
339 the port if you really want to have XEmacs work under VMS.
342 *** On HPUX, the HP C compiler might have a problem compiling some files
345 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
347 Had to drop once again to level 2 optimization, at least to
348 compile lstream.c. Otherwise, I get a "variable is void: \if"
349 problem while dumping (this is a problem I already reported
350 with vanilla hpux 10.01 and 9.07, which went away after
351 applying patches for the C compiler). Trouble is I still
352 haven't found the same patch for hpux 10.10, and I don't
353 remember the patch numbers. I think potential XEmacs builders
354 on HP should be warned about this.
356 *** I don't have `xmkmf' and `imake' on my HP.
358 You can get these standard X tools by anonymous FTP to
359 hpcvaaz.cv.hp.com. Essentially all X programs need these.
361 *** On HP-UX, problems with make
363 Marcus Thiessel <marcus_thiessel@hp.com>
365 Some releases of XEmacs (e.g. 20.4) require GNU make to build
366 successfully. You don't need GNU make when building 21.x.
368 *** On HP-UX 9.05 XEmacs won't compile or coredump during the build.
370 Marcus Thiessel <marcus_thiessel@hp.com>
372 This might be a sed problem. For your own safety make sure to use
373 GNU sed while dumping XEmacs.
375 *** On HP-UX 11.0 XEmacs causes excessive X11 errors when running.
377 Marcus Thiessel <marcus_thiessel@hp.com>
379 Unfortunately, XEmacs releases <21.0 don't work with Motif2.1. It
380 will compile but you will get excessive X11 errors like
382 xemacs: X Error of failed request: BadGC (invalid GC parameter)
384 and finally XEmacs gets killed. A workaround is to use the
385 Motif1.2_R6 libraries. You can the following line to your call to
388 --x-libraries="/usr/lib/Motif1.2_R6 -L/usr/lib/X11R6"
390 Make sure /usr/lib/Motif1.2_R6/libXm.sl is a link to
391 /usr/lib/Motif1.2_R6/libXm.3.
394 *** Native cc on SCO OpenServer 5 is now OK. Icc may still throw you
395 a curve. Here is what Robert Lipe <robertl@arnet.com> says:
397 Unlike XEmacs 19.13, building with the native cc on SCO OpenServer 5
398 now produces a functional binary. I will typically build this
399 configuration for COFF with:
401 /path_to_xemacs_source/configure --with-gcc=no \
402 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
403 --with-xpm --with-xface --with-sound=nas
405 This version now supports ELF builds. I highly recommend this to
406 reduce the in-core footprint of XEmacs. This is now how I compile
407 all my test releases. Build it like this:
409 /path_to_XEmacs_source/configure --with-gcc=no \
410 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
411 --with-xpm --with-xface --with-sound=nas --dynamic
413 The compiler known as icc [ supplied with the OpenServer 5 Development
414 System ] generates a working binary, but it takes forever to generate
415 XEmacs. ICC also whines more about the code than /bin/cc does. I do
416 believe all its whining is legitimate, however. Note that you do
417 have to 'cd src ; make LD=icc' to avoid linker errors.
419 The way I handle the build procedure is:
421 /path_to_XEmacs_source/configure --with-gcc=no \
422 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
423 --with-xpm --with-xface --with-sound=nas --dynamic --compiler="icc"
425 NOTE I have the xpm, xface, and audio libraries and includes in
426 /usr/local/lib, /usr/local/include. If you don't have these,
427 don't include the "--with-*" arguments in any of my examples.
429 In previous versions of XEmacs, you had to override the defaults while
430 compiling font-lock.o and extents.o when building with icc. This seems
431 to no longer be true, but I'm including this old information in case it
432 resurfaces. The process I used was:
435 [ procure pizza, beer, repeat ]
437 make CC="icc -W0,-mP1COPT_max_tree_size=3000" font-lock.o extents.o
440 If you want sound support, get the tls566 supplement from
441 ftp.sco.com:/TLS or any of its mirrors. It works just groovy
444 The M-x manual-entry is known not to work. If you know Lisp and would
445 like help in making it work, e-mail me at <robertl@dgii.com>.
446 (UNCHECKED for 19.15 -- it might work).
448 In earlier releases, gnuserv/gnuclient/gnudoit would open a frame
449 just fine, but the client would lock up and the server would
450 terminate when you used C-x # to close the frame. This is now
453 In etc/ there are two files of note. emacskeys.sco and emacsstrs.sco.
454 The comments at the top of emacskeys.sco describe its function, and
455 the emacstrs.sco is a suitable candidate for /usr/lib/keyboard/strings
456 to take advantage of the keyboard map in emacskeys.sco.
458 Note: Much of the above entry is probably not valid for XEmacs 21.0
463 * Problems with running XEmacs
464 ==============================
466 *** C-z just refreshes the screen instead of suspending Emacs.
468 You are probably using a shell that doesn't support job control, even
469 though the system itself is capable of it. Try using a different
472 *** You type Control-H (Backspace) expecting to delete characters.
474 Emacs has traditionally used Control-H for help; unfortunately this
475 interferes with its use as Backspace on TTY's. One way to solve this
476 problem is to put this in your .emacs:
478 (when (eq tty-erase-char ?\C-h)
479 (keyboard-translate ?\C-h ?\C-?)
480 (global-set-key "\M-?" 'help-command))
482 This checks whether the TTY erase char is C-h, and if it is, makes
483 Control-H (Backspace) work sensibly, and moves help to Meta-? (ESC ?).
485 Note that you can probably also access help using F1.
487 *** Mail agents (VM, Gnus, rmail) cannot get new mail
489 rmail and VM get new mail from /usr/spool/mail/$USER using a program
490 called `movemail'. This program interlocks with /bin/mail using the
491 protocol defined by /bin/mail.
493 There are two different protocols in general use. One of them uses
494 the `flock' system call. The other involves creating a lock file;
495 `movemail' must be able to write in /usr/spool/mail in order to do
496 this. You control which one is used by defining, or not defining, the
497 macro MAIL_USE_FLOCK in config.h or the m- or s- file it includes. IF
498 YOU DON'T USE THE FORM OF INTERLOCKING THAT IS NORMAL ON YOUR SYSTEM,
501 If your system uses the lock file protocol, and fascist restrictions
502 prevent ordinary users from writing the lock files in /usr/spool/mail,
503 you may need to make `movemail' setgid to a suitable group such as
504 `mail'. To do this, use the following commands (as root) after doing
510 Installation normally copies movemail from the build directory to an
511 installation directory which is usually under /usr/local/lib. The
512 installed copy of movemail is usually in the directory
513 /usr/local/lib/emacs/VERSION/TARGET. You must change the group and
514 mode of the installed copy; changing the group and mode of the build
515 directory copy is ineffective.
517 *** VM appears to hang in large folders.
519 This is normal (trust us) when upgrading to VM-6.22 from earlier
520 versions. Let VM finish what it is doing and all will be well.
522 *** Changes made to .el files do not take effect.
524 You may have forgotten to recompile them into .elc files. Then the
525 old .elc files will be loaded, and your changes will not be seen. To
526 fix this, do `M-x byte-recompile-directory' and specify the directory
527 that contains the Lisp files.
529 Note that you will get a warning when loading a .elc file that is
530 older than the corresponding .el file.
532 *** Things which should be bold or italic (such as the initial
533 copyright notice) are not.
535 The fonts of the "bold" and "italic" faces are generated from the font
536 of the "default" face; in this way, your bold and italic fonts will
537 have the appropriate size and family. However, emacs can only be
538 clever in this way if you have specified the default font using the
539 XLFD (X Logical Font Description) format, which looks like
541 *-courier-medium-r-*-*-*-120-*-*-*-*-*-*
543 if you use any of the other, less strict font name formats, some of
546 lucidasanstypewriter-12
550 then emacs won't be able to guess the names of the "bold" and "italic"
551 versions. All X fonts can be referred to via XLFD-style names, so you
552 should use those forms. See the man pages for X(1), xlsfonts(1), and
555 *** The dumped Emacs crashes when run, trying to write pure data.
557 Two causes have been seen for such problems.
559 1) On a system where getpagesize is not a system call, it is defined
560 as a macro. If the definition (in both unexec.c and malloc.c) is wrong,
561 it can cause problems like this. You might be able to find the correct
562 value in the man page for a.out (5).
564 2) Some systems allocate variables declared static among the
565 initialized variables. Emacs makes all initialized variables in most
566 of its files pure after dumping, but the variables declared static and
567 not initialized are not supposed to be pure. On these systems you
568 may need to add "#define static" to the m- or the s- file.
570 *** Reading and writing files is very very slow.
572 Try evaluating the form (setq lock-directory nil) and see if that helps.
573 There is a problem with file-locking on some systems (possibly related
574 to NFS) that I don't understand. Please send mail to the address
575 xemacs@xemacs.org if you figure this one out.
577 *** When emacs starts up, I get lots of warnings about unknown keysyms.
579 If you are running the prebuilt binaries, the Motif library expects to find
580 certain thing in the XKeysymDB file. This file is normally in /usr/lib/X11/
581 or in /usr/openwin/lib/. If you keep yours in a different place, set the
582 environment variable $XKEYSYMDB to point to it before starting emacs. If
583 you still have the problem after doing that, perhaps your version of X is
584 too old. There is a copy of the MIT X11R5 XKeysymDB file in the emacs `etc'
585 directory. Try using that one.
587 *** My X resources used to work, and now some of them are being ignored.
589 Check the resources in .../etc/Emacs.ad (which is the same as the file
590 sample.Xdefaults). Perhaps some of the default resources built in to
591 emacs are now overriding your existing resources. Copy and edit the
592 resources in Emacs.ad as necessary.
594 *** I have focus problems when I use `M-o' to switch to another screen
595 without using the mouse.
597 The focus issues with a program like XEmacs, which has multiple
598 homogeneous top-level windows, are very complicated, and as a result,
599 most window managers don't implement them correctly.
601 The R4/R5 version of twm (and all of its descendants) had buggy focus
602 handling. Sufficiently recent versions of tvtwm have been fixed. In
603 addition, if you're using twm, make sure you have not specified
604 "NoTitleFocus" in your .tvtwmrc file. The very nature of this option
605 makes twm do some illegal focus tricks, even with the patch.
607 It is known that olwm and olvwm are buggy, and in different ways. If
608 you're using click-to-type mode, try using point-to-type, or vice
611 In older versions of NCDwm, one could not even type at XEmacs windows.
612 This has been fixed in newer versions (2.4.3, and possibly earlier).
614 (Many people suggest that XEmacs should warp the mouse when focusing
615 on another screen in point-to-type mode. This is not ICCCM-compliant
616 behavior. Implementing such policy is the responsibility of the
617 window manager itself, it is not legal for a client to do this.)
619 *** Emacs spontaneously displays "I-search: " at the bottom of the screen.
621 This means that Control-S/Control-Q (XON/XOFF) "flow control" is being
622 used. C-s/C-q flow control is bad for Emacs editors because it takes
623 away C-s and C-q as user commands. Since editors do not output long
624 streams of text without user commands, there is no need for a
625 user-issuable "stop output" command in an editor; therefore, a
626 properly designed flow control mechanism would transmit all possible
627 input characters without interference. Designing such a mechanism is
628 easy, for a person with at least half a brain.
630 There are three possible reasons why flow control could be taking place:
632 1) Terminal has not been told to disable flow control
633 2) Insufficient padding for the terminal in use
634 3) Some sort of terminal concentrator or line switch is responsible
636 First of all, many terminals have a set-up mode which controls whether
637 they generate XON/XOFF flow control characters. This must be set to
638 "no XON/XOFF" in order for Emacs to work. Sometimes there is an
639 escape sequence that the computer can send to turn flow control off
640 and on. If so, perhaps the termcap `ti' string should turn flow
641 control off, and the `te' string should turn it on.
643 Once the terminal has been told "no flow control", you may find it
644 needs more padding. The amount of padding Emacs sends is controlled
645 by the termcap entry for the terminal in use, and by the output baud
646 rate as known by the kernel. The shell command `stty' will print
647 your output baud rate; `stty' with suitable arguments will set it if
648 it is wrong. Setting to a higher speed causes increased padding. If
649 the results are wrong for the correct speed, there is probably a
650 problem in the termcap entry. You must speak to a local Unix wizard
651 to fix this. Perhaps you are just using the wrong terminal type.
653 For terminals that lack a "no flow control" mode, sometimes just
654 giving lots of padding will prevent actual generation of flow control
655 codes. You might as well try it.
657 If you are really unlucky, your terminal is connected to the computer
658 through a concentrator which sends XON/XOFF flow control to the
659 computer, or it insists on sending flow control itself no matter how
660 much padding you give it. Unless you can figure out how to turn flow
661 control off on this concentrator (again, refer to your local wizard),
662 you are screwed! You should have the terminal or concentrator
663 replaced with a properly designed one. In the mean time, some drastic
664 measures can make Emacs semi-work.
666 You can make Emacs ignore C-s and C-q and let the operating system
667 handle them. To do this on a per-session basis, just type M-x
668 enable-flow-control RET. You will see a message that C-\ and C-^ are
669 now translated to C-s and C-q. (Use the same command M-x
670 enable-flow-control to turn *off* this special mode. It toggles flow
673 If C-\ and C-^ are inconvenient for you (for example, if one of them
674 is the escape character of your terminal concentrator), you can choose
675 other characters by setting the variables flow-control-c-s-replacement
676 and flow-control-c-q-replacement. But choose carefully, since all
677 other control characters are already used by emacs.
679 IMPORTANT: if you type C-s by accident while flow control is enabled,
680 Emacs output will freeze, and you will have to remember to type C-q in
683 If you work in an environment where a majority of terminals of a
684 certain type are flow control hobbled, you can use the function
685 `enable-flow-control-on' to turn on this flow control avoidance scheme
686 automatically. Here is an example:
688 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
690 If this isn't quite correct (e.g. you have a mixture of flow-control hobbled
691 and good vt200 terminals), you can still run enable-flow-control
694 I have no intention of ever redesigning the Emacs command set for the
695 assumption that terminals use C-s/C-q flow control. XON/XOFF flow
696 control technique is a bad design, and terminals that need it are bad
697 merchandise and should not be purchased. Now that X is becoming
698 widespread, XON/XOFF seems to be on the way out. If you can get some
699 use out of GNU Emacs on inferior terminals, more power to you, but I
700 will not make Emacs worse for properly designed systems for the sake
703 *** Control-S and Control-Q commands are ignored completely.
705 For some reason, your system is using brain-damaged C-s/C-q flow
706 control despite Emacs's attempts to turn it off. Perhaps your
707 terminal is connected to the computer through a concentrator
708 that wants to use flow control.
710 You should first try to tell the concentrator not to use flow control.
711 If you succeed in this, try making the terminal work without
712 flow control, as described in the preceding section.
714 If that line of approach is not successful, map some other characters
715 into C-s and C-q using keyboard-translate-table. The example above
716 shows how to do this with C-^ and C-\.
718 *** Control-S and Control-Q commands are ignored completely on a net
721 Some versions of rlogin (and possibly telnet) do not pass flow
722 control characters to the remote system to which they connect.
723 On such systems, emacs on the remote system cannot disable flow
724 control on the local system.
726 One way to cure this is to disable flow control on the local host
727 (the one running rlogin, not the one running rlogind) using the
728 stty command, before starting the rlogin process. On many systems,
729 `stty start u stop u' will do this.
731 Some versions of tcsh will prevent even this from working. One way
732 around this is to start another shell before starting rlogin, and
733 issue the stty command to disable flow control from that shell.
735 If none of these methods work, the best solution is to type
736 `M-x enable-flow-control' at the beginning of your emacs session, or
737 if you expect the problem to continue, add a line such as the
738 following to your .emacs (on the host running rlogind):
740 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
742 See the entry about spontaneous display of I-search (above) for more
745 *** TTY redisplay is slow.
747 XEmacs has fairly new TTY redisplay support (beginning from 19.12),
748 which doesn't include some basic TTY optimizations -- like using
749 scrolling regions to move around blocks of text. This is why
750 redisplay on the traditional terminals, or over slow lines can be very
753 If you are interested in fixing this, please let us know at
756 *** Screen is updated wrong, but only on one kind of terminal.
758 This could mean that the termcap entry you are using for that terminal
759 is wrong, or it could mean that Emacs has a bug handing the
760 combination of features specified for that terminal.
762 The first step in tracking this down is to record what characters
763 Emacs is sending to the terminal. Execute the Lisp expression
764 (open-termscript "./emacs-script") to make Emacs write all terminal
765 output into the file ~/emacs-script as well; then do what makes the
766 screen update wrong, and look at the file and decode the characters
767 using the manual for the terminal. There are several possibilities:
769 1) The characters sent are correct, according to the terminal manual.
771 In this case, there is no obvious bug in Emacs, and most likely you
772 need more padding, or possibly the terminal manual is wrong.
774 2) The characters sent are incorrect, due to an obscure aspect of the
775 terminal behavior not described in an obvious way by termcap.
777 This case is hard. It will be necessary to think of a way for Emacs
778 to distinguish between terminals with this kind of behavior and other
779 terminals that behave subtly differently but are classified the same
780 by termcap; or else find an algorithm for Emacs to use that avoids the
781 difference. Such changes must be tested on many kinds of terminals.
783 3) The termcap entry is wrong.
785 See the file etc/TERMS for information on changes that are known to be
786 needed in commonly used termcap entries for certain terminals.
788 4) The characters sent are incorrect, and clearly cannot be right for
789 any terminal with the termcap entry you were using.
791 This is unambiguously an Emacs bug, and can probably be fixed in
792 termcap.c, terminfo.c, tparam.c, cm.c, redisplay-tty.c,
793 redisplay-output.c, or redisplay.c.
795 *** My buffers are full of \000 characters or otherwise corrupt.
797 Some compilers have trouble with gmalloc.c and ralloc.c; try recompiling
798 without optimization. If that doesn't work, try recompiling with
799 SYSTEM_MALLOC defined, and/or with REL_ALLOC undefined.
801 *** A position you specified in .Xdefaults is ignored, using twm.
803 twm normally ignores "program-specified" positions.
804 You can tell it to obey them with this command in your `.twmrc' file:
806 UsePPosition "on" #allow clents to request a position
808 *** With M-x enable-flow-control, you need to type C-\ twice to do
809 incremental search--a single C-\ gets no response.
811 This has been traced to communicating with your machine via kermit,
812 with C-\ as the kermit escape character. One solution is to use
813 another escape character in kermit. One user did
815 set escape-character 17
817 in his .kermrc file, to make C-q the kermit escape character.
819 *** The Motif version of Emacs paints the screen a solid color.
821 This has been observed to result from the following X resource:
823 Emacs*default.attributeFont: -*-courier-medium-r-*-*-*-140-*-*-*-*-iso8859-*
825 That the resource has this effect indicates a bug in something, but we
826 do not yet know what. If it is an Emacs bug, we hope someone can
827 explain what the bug is so we can fix it. In the mean time, removing
828 the resource prevents the problem.
830 *** After running emacs once, subsequent invocations crash.
832 Some versions of SVR4 have a serious bug in the implementation of the
833 mmap () system call in the kernel; this causes emacs to run correctly
834 the first time, and then crash when run a second time.
836 Contact your vendor and ask for the mmap bug fix; in the mean time,
837 you may be able to work around the problem by adding a line to your
838 operating system description file (whose name is reported by the
839 configure script) that reads:
840 #define SYSTEM_MALLOC
841 This makes Emacs use memory less efficiently, but seems to work around
844 *** Inability to send an Alt-modified key, when Emacs is communicating
845 directly with an X server.
847 If you have tried to bind an Alt-modified key as a command, and it
848 does not work to type the command, the first thing you should check is
849 whether the key is getting through to Emacs. To do this, type C-h c
850 followed by the Alt-modified key. C-h c should say what kind of event
851 it read. If it says it read an Alt-modified key, then make sure you
852 have made the key binding correctly.
854 If C-h c reports an event that doesn't have the Alt modifier, it may
855 be because your X server has no key for the Alt modifier. The X
856 server that comes from MIT does not set up the Alt modifier by
859 If your keyboard has keys named Alt, you can enable them as follows:
861 xmodmap -e 'add mod2 = Alt_L'
862 xmodmap -e 'add mod2 = Alt_R'
864 If the keyboard has just one key named Alt, then only one of those
865 commands is needed. The modifier `mod2' is a reasonable choice if you
866 are using an unmodified MIT version of X. Otherwise, choose any
867 modifier bit not otherwise used.
869 If your keyboard does not have keys named Alt, you can use some other
870 keys. Use the keysym command in xmodmap to turn a function key (or
871 some other 'spare' key) into Alt_L or into Alt_R, and then use the
872 commands show above to make them modifier keys.
874 Note that if you have Alt keys but no Meta keys, Emacs translates Alt
875 into Meta. This is because of the great importance of Meta in Emacs.
877 *** In Shell mode, you get a ^M at the end of every line.
879 This happens to people who use tcsh, because it is trying to be too
880 smart. It sees that the Shell uses terminal type `unknown' and turns
881 on the flag to output ^M at the end of each line. You can fix the
882 problem by adding this to your .cshrc file:
885 if ($EMACS == "t") then
887 stty -icrnl -onlcr -echo susp ^Z
891 *** An error message such as `X protocol error: BadMatch (invalid
892 parameter attributes) on protocol request 93'.
894 This comes from having an invalid X resource, such as
896 (which is invalid because it specifies a color name for something
899 The fix is to correct your X resources.
901 *** Once you pull down a menu from the menubar, it won't go away.
903 It has been claimed that this is caused by a bug in certain very old
904 (1990?) versions of the twm window manager. It doesn't happen with
905 recent vintages, or with other window managers.
907 *** Emacs ignores the "help" key when running OLWM.
909 OLWM grabs the help key, and retransmits it to the appropriate client
910 using XSendEvent. Allowing emacs to react to synthetic events is a
911 security hole, so this is turned off by default. You can enable it by
912 setting the variable x-allow-sendevents to t. You can also cause fix
913 this by telling OLWM to not grab the help key, with the null binding
914 "OpenWindows.KeyboardCommand.Help:".
916 *** Programs running under terminal emulator do not recognize `emacs'
919 The cause of this is a shell startup file that sets the TERMCAP
920 environment variable. The terminal emulator uses that variable to
921 provide the information on the special terminal type that Emacs
924 Rewrite your shell startup file so that it does not change TERMCAP
925 in such a case. You could use the following conditional which sets
926 it only if it is undefined.
928 if ( ! ${?TERMCAP} ) setenv TERMCAP ~/my-termcap-file
930 Or you could set TERMCAP only when you set TERM--which should not
931 happen in a non-login shell.
933 *** The popup menu appears at the buttom/right of my screen.
935 You probably have something like the following in your ~/.Xdefaults
937 Emacs.geometry: 81x56--9--1
939 Use the following instead
941 Emacs*EmacsFrame.geometry: 81x56--9--1
945 *** Your Delete key sends a Backspace to the terminal, using an AIXterm.
947 The solution is to include in your .Xdefaults the lines:
949 *aixterm.Translations: #override <Key>BackSpace: string(0x7f)
950 aixterm*ttyModes: erase ^?
952 This makes your Backspace key send DEL (ASCII 127).
954 *** On AIX 4, some programs fail when run in a Shell buffer
955 with an error message like No terminfo entry for "unknown".
957 On AIX, many terminal type definitions are not installed by default.
958 `unknown' is one of them. Install the "Special Generic Terminal
959 Definitions" to make them defined.
961 *** On AIX, you get this message when running Emacs:
963 Could not load program emacs
964 Symbol smtcheckinit in csh is undefined
965 Error was: Exec format error
969 Could not load program .emacs
970 Symbol _system_con in csh is undefined
971 Symbol _fp_trapsta in csh is undefined
972 Error was: Exec format error
974 These can happen when you try to run on AIX 3.2.5 a program that was
975 compiled with 3.2.4. The fix is to recompile.
977 *** Trouble using ptys on AIX.
979 People often install the pty devices on AIX incorrectly.
980 Use `smit pty' to reinstall them properly.
984 *** The Emacs window disappears when you type M-q.
986 Some versions of the Open Look window manager interpret M-q as a quit
987 command for whatever window you are typing at. If you want to use
988 Emacs with that window manager, you should try to configure the window
989 manager to use some other command. You can disable the
990 shortcut keys entirely by adding this line to ~/.OWdefaults:
992 OpenWindows.WindowMenuAccelerators: False
994 *** When Emacs tries to ring the bell, you get an error like
996 audio: sst_open: SETQSIZE" Invalid argument
997 audio: sst_close: SETREG MMR2, Invalid argument
999 you have probably compiled using an ANSI C compiler, but with non-ANSI
1000 include files. In particular, on Suns, the file
1001 /usr/include/sun/audioio.h uses the _IOW macro to define the constant
1002 AUDIOSETQSIZE. _IOW in turn uses a K&R preprocessor feature that is
1003 now explicitly forbidden in ANSI preprocessors, namely substitution
1004 inside character constants. All ANSI C compilers must provide a
1005 workaround for this problem. Lucid's C compiler is shipped with a new
1006 set of system include files. If you are using GCC, there is a script
1007 called fixincludes that creates new versions of some system include
1008 files that use this obsolete feature.
1010 *** On Solaris 2.6, XEmacs dumps core when exiting.
1012 This happens if you're XEmacs is running on the same machine as the X
1013 server, and the optimized memory transport has been turned on by
1014 setting the environment variable XSUNTRANSPORT. The crash occurs
1015 during the call to XCloseDisplay.
1017 If this describes your situation, you need to undefine the
1018 XSUNTRANSPORT environment variable.
1020 *** On Solaris, C-x doesn't get through to Emacs when you use the console.
1022 This is a Solaris feature (at least on Intel x86 cpus). Type C-r
1023 C-r C-t, to toggle whether C-x gets through to Emacs.
1025 *** On Solaris 2.4, Dired hangs and C-g does not work. Or Emacs hangs
1026 forever waiting for termination of a subprocess that is a zombie.
1028 casper@fwi.uva.nl says the problem is in X11R6. Rebuild libX11.so
1029 after changing the file xc/config/cf/sunLib.tmpl. Change the lines
1032 #define SharedX11Reqs -lthread
1037 #if OSMinorVersion < 4
1039 #define SharedX11Reqs -lthread
1043 Be sure also to edit x/config/cf/sun.cf so that OSMinorVersion is 4
1044 (as it should be for Solaris 2.4). The file has three definitions for
1045 OSMinorVersion: the first is for x86, the second for SPARC under
1046 Solaris, and the third for SunOS 4. Make sure to update the
1047 definition for your type of machine and system.
1049 Then do `make Everything' in the top directory of X11R6, to rebuild
1050 the makefiles and rebuild X. The X built this way work only on
1051 Solaris 2.4, not on 2.3.
1053 For multithreaded X to work it necessary to install patch
1054 101925-02 to fix problems in header files [2.4]. You need
1055 to reinstall gcc or re-run just-fixinc after installing that
1058 However, Frank Rust <frust@iti.cs.tu-bs.de> used a simpler solution:
1060 #define ThreadedX YES
1062 #define ThreadedX NO
1063 in sun.cf and did `make World' to rebuild X11R6. Removing all
1064 `-DXTHREAD*' flags and `-lthread' entries from lib/X11/Makefile and
1065 typing 'make install' in that directory also seemed to work.
1067 *** On SunOS 4.1.3, Emacs unpredictably crashes in _yp_dobind_soft.
1069 This happens if you configure Emacs specifying just `sparc-sun-sunos4'
1070 on a system that is version 4.1.3. You must specify the precise
1071 version number (or let configure figure out the configuration, which
1072 it can do perfectly well for SunOS).
1074 *** Mail is lost when sent to local aliases.
1076 Many emacs mail user agents (VM and rmail, for instance) use the
1077 sendmail.el library. This library can arrange for mail to be
1078 delivered by passing messages to the /usr/lib/sendmail (usually)
1079 program . In doing so, it passes the '-t' flag to sendmail, which
1080 means that the name of the recipient of the message is not on the
1081 command line and, therefore, that sendmail must parse the message to
1082 obtain the destination address.
1084 There is a bug in the SunOS4.1.1 and SunOS4.1.3 versions of sendmail.
1085 In short, when given the -t flag, the SunOS sendmail won't recognize
1086 non-local (i.e. NIS) aliases. It has been reported that the Solaris
1087 2.x versions of sendmail do not have this bug. For those using SunOS
1088 4.1, the best fix is to install sendmail V8 or IDA sendmail (which
1089 have other advantages over the regular sendmail as well). At the time
1090 of this writing, these official versions are available:
1092 Sendmail V8 on ftp.cs.berkeley.edu in /ucb/sendmail:
1093 sendmail.8.6.9.base.tar.Z (the base system source & documentation)
1094 sendmail.8.6.9.cf.tar.Z (configuration files)
1095 sendmail.8.6.9.misc.tar.Z (miscellaneous support programs)
1096 sendmail.8.6.9.xdoc.tar.Z (extended documentation, with postscript)
1098 IDA sendmail on vixen.cso.uiuc.edu in /pub:
1099 sendmail-5.67b+IDA-1.5.tar.gz
1101 *** Emacs fails to understand most Internet host names, even though
1102 the names work properly with other programs on the same system.
1103 Emacs won't work with X-windows if the value of DISPLAY is HOSTNAME:0.
1104 Gnus can't make contact with the specified host for nntp.
1106 This typically happens on Suns and other systems that use shared
1107 libraries. The cause is that the site has installed a version of the
1108 shared library which uses a name server--but has not installed a
1109 similar version of the unshared library which Emacs uses.
1111 The result is that most programs, using the shared library, work with
1112 the nameserver, but Emacs does not.
1114 The fix is to install an unshared library that corresponds to what you
1115 installed in the shared library, and then relink Emacs.
1117 On SunOS 4.1, simply define HAVE_RES_INIT.
1119 If you have already installed the name resolver in the file libresolv.a,
1120 then you need to compile Emacs to use that library. The easiest way to
1121 do this is to add to config.h a definition of LIBS_SYSTEM, LIBS_MACHINE
1122 or LIB_STANDARD which uses -lresolv. Watch out! If you redefine a macro
1123 that is already in use in your configuration to supply some other libraries,
1124 be careful not to lose the others.
1126 Thus, you could start by adding this to config.h:
1128 #define LIBS_SYSTEM -lresolv
1130 Then if this gives you an error for redefining a macro, and you see that
1131 the s- file defines LIBS_SYSTEM as -lfoo -lbar, you could change config.h
1134 #define LIBS_SYSTEM -lresolv -lfoo -lbar
1136 *** With process-connection-type set to t, each line of subprocess
1137 output is terminated with a ^M, making ange-ftp and GNUS not work.
1139 On SunOS systems, this problem has been seen to be a result of an
1140 incomplete installation of gcc 2.2 which allowed some non-ANSI
1141 compatible include files into the compilation. In particular this
1142 affected virtually all ioctl() calls.
1146 *** `C-z', or `M-x suspend-emacs' hangs instead of suspending.
1148 If you build with `gpm' support on Linux, you cannot suspend XEmacs
1149 because gpm installs a buggy SIGTSTP handler. Either compile with
1150 `--with-gpm=no', or don't suspend XEmacs on the Linux console until
1153 *** With certain fonts, when the cursor appears on a character, the
1154 character doesn't appear--you get a solid box instead.
1156 One user on a Linux system reported that this problem went away with
1157 installation of a new X server. The failing server was XFree86 3.1.1.
1158 XFree86 3.1.2 works.
1160 *** Slow startup on Linux.
1162 People using systems based on the Linux kernel sometimes report that
1163 startup takes 10 to 15 seconds longer than `usual'.
1165 This is because Emacs looks up the host name when it starts.
1166 Normally, this takes negligible time; the extra delay is due to
1167 improper system configuration. This problem can occur for both
1168 networked and non-networked machines.
1170 Here is how to fix the configuration. It requires being root.
1174 First, make sure the files `/etc/hosts' and `/etc/host.conf' both
1175 exist. The first line in the `/etc/hosts' file should look like this
1176 (replace HOSTNAME with your host name):
1178 127.0.0.1 localhost HOSTNAME
1180 Also make sure that the `/etc/host.conf' files contains the following
1186 Any changes, permanent and temporary, to the host name should be
1187 indicated in the `/etc/hosts' file, since it acts a limited local
1188 database of addresses and names (e.g., some SLIP connections
1189 dynamically allocate ip addresses).
1191 **** Non-Networked Case
1193 The solution described in the networked case applies here as well.
1194 However, if you never intend to network your machine, you can use a
1195 simpler solution: create an empty `/etc/host.conf' file. The command
1196 `touch /etc/host.conf' suffices to create the file. The `/etc/hosts'
1197 file is not necessary with this approach.
1201 *** On Irix, I don't see the toolbar icons and I'm getting lots of
1202 entries in the warnings buffer.
1204 SGI ships a really old Xpm library in /usr/lib which does not work at
1205 all well with XEmacs. The solution is to install your own copy of the
1206 latest version of Xpm somewhere and then use the --site-includes and
1207 --site-libraries flags to tell configure where to find it.
1209 *** Trouble using ptys on IRIX, or running out of ptys.
1211 The program mkpts (which may be in `/usr/adm' or `/usr/sbin') needs to
1212 be set-UID to root, or non-root programs like Emacs will not be able
1213 to allocate ptys reliably.
1215 *** Motif dialog boxes lose on Irix.
1217 Larry Auton <lda@control.att.com> writes:
1218 Beware of not specifying
1220 --with-dialogs=athena
1222 if it builds with the motif dialogs [boom!] you're a dead man.
1224 *** Beware of the default image & graphics library on Irix
1226 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
1228 You *have* to compile your own jpeg lib. The one delivered with SGI
1229 systems is a C++ lib, which apparently XEmacs cannot cope with.
1232 ** Digital UNIX/OSF/VMS/Ultrix
1233 *** XEmacs crashes on Digital Unix within font-lock, or when dealing
1234 with large compilation buffers.
1236 The default stack size under Digital Unix is rather small (2M as
1237 opposed to Solaris 8M), hosing the regexp code, which uses alloca()
1238 extensively, overflowing the stack when complex regexps are used.
1241 1) Increase your stack size, using `ulimit -s 8192' or a (t)csh
1244 2) Recompile regex.c with REGEX_MALLOC defined.
1246 *** The `Alt' key doesn't behave as `Meta' when running DECwindows.
1248 The default DEC keyboard mapping has the Alt keys set up to generate the
1249 keysym `Multi_key', which has a meaning to xemacs which is distinct from that
1250 of the `Meta_L' and `Meta-R' keysyms. A second problem is that certain keys
1251 have the Mod2 modifier attached to them for no adequately explored reason.
1252 The correct fix is to pass this file to xmodmap upon starting X:
1255 keysym Multi_key = Alt_L
1259 *** The Compose key on a DEC keyboard does not work as Meta key.
1261 This shell command should fix it:
1263 xmodmap -e 'keycode 0xb1 = Meta_L'
1265 *** `expand-file-name' fails to work on any but the machine you dumped
1268 On Ultrix, if you use any of the functions which look up information
1269 in the passwd database before dumping Emacs (say, by using
1270 expand-file-name in site-init.el), then those functions will not work
1271 in the dumped Emacs on any host but the one Emacs was dumped on.
1273 The solution? Don't use expand-file-name in site-init.el, or in
1274 anything it loads. Yuck - some solution.
1276 I'm not sure why this happens; if you can find out exactly what is
1277 going on, and perhaps find a fix or a workaround, please let us know.
1278 Perhaps the YP functions cache some information, the cache is included
1279 in the dumped Emacs, and is then inaccurate on any other host.
1283 *** I get complaints about the mapping of my HP keyboard at startup,
1284 but I haven't changed anything.
1286 The default HP keymap is set up to have Mod1 assigned to two different keys:
1287 Meta_L and Mode_switch (even though there is not actually a Mode_switch key on
1288 the keyboard -- it uses an "imaginary" keycode.) There actually is a reason
1289 for this, but it's not a good one. The correct fix is to execute this command
1292 xmodmap -e 'remove mod1 = Mode_switch'
1294 *** On HP-UX, you get "poll: Interrupted system call" message in the
1295 window where XEmacs was launched.
1297 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
1299 I get a very strange problem when linking libc.a dynamically: every
1300 event (mouse, keyboard, expose...) results in a "poll: Interrupted
1301 system call" message in the window where XEmacs was
1302 launched. Forcing a static link of libc.a alone by adding
1303 /usr/lib/libc.a at the end of the link line solves this. Note that
1304 my 9.07 build of 19.14b17 and my (old) build of 19.13 both exhibit
1305 the same behaviour. I've tried various hpux patches to no avail. If
1306 this problem cannot be solved before the release date, binary kits
1307 for HP *must* be linked statically against libc, otherwise this
1308 problem will show up. (This is directed at whoever will volunteer
1309 for this kit, as I won't be available to do it, unless 19.14 gets
1310 delayed until mid-june ;-). I think this problem will be an FAQ soon
1311 after the release otherwise.
1313 Note: The above entry is probably not valid for XEmacs 21.0 and
1316 *** The right Alt key works wrong on German HP keyboards (and perhaps
1317 other non-English HP keyboards too).
1319 This is because HP-UX defines the modifiers wrong in X. Here is a
1320 shell script to fix the problem; be sure that it is run after VUE
1321 configures the X server.
1323 xmodmap 2> /dev/null - << EOF
1324 keysym Alt_L = Meta_L
1325 keysym Alt_R = Meta_R
1330 keysym Mode_switch = NoSymbol
1332 keysym Meta_R = Mode_switch
1333 add mod2 = Mode_switch
1336 *** `Pid xxx killed due to text modification or page I/O error'
1338 On HP-UX, you can get that error when the Emacs executable is on an NFS
1339 file system. HP-UX responds this way if it tries to swap in a page and
1340 does not get a response from the server within a timeout whose default
1341 value is just ten seconds.
1343 If this happens to you, extend the timeout period.
1345 *** Shell mode on HP-UX gives the message, "`tty`: Ambiguous".
1347 christos@theory.tn.cornell.edu says:
1349 The problem is that in your .cshrc you have something that tries to
1350 execute `tty`. If you are not running the shell on a real tty then tty
1351 will print "not a tty". Csh expects one word in some places, but tty
1352 is giving it back 3.
1354 The solution is to add a pair of quotes around `tty` to make it a
1357 if (`tty` == "/dev/console")
1359 should be changed to:
1361 if ("`tty`" == "/dev/console")
1363 Even better, move things that set up terminal sections out of .cshrc
1368 *** Regular expressions matching bugs on SCO systems.
1370 On SCO, there are problems in regexp matching when Emacs is compiled
1371 with the system compiler. The compiler version is "Microsoft C
1372 version 6", SCO 4.2.0h Dev Sys Maintenance Supplement 01/06/93; Quick
1373 C Compiler Version 1.00.46 (Beta). The solution is to compile with
1378 *** Emacs exits with "X protocol error" when run with an X server for
1381 A certain X server for Windows had a bug which caused this.
1382 Supposedly the newer 32-bit version of this server doesn't have the
1387 * Compatibility problems (with Emacs 18, GNU Emacs, or previous XEmacs/lemacs)
1388 ==============================================================================
1390 *** "Symbol's value as variable is void: unread-command-char".
1391 "Wrong type argument: arrayp, #<keymap 143 entries>"
1392 "Wrong type argument: stringp, [#<keypress-event return>]"
1394 There are a few incompatible changes in XEmacs, and these are the
1395 symptoms. Some of the emacs-lisp code you are running needs to be
1396 updated to be compatible with XEmacs.
1398 The code should not treat keymaps as arrays (use `define-key', etc.),
1399 should not use obsolete variables like `unread-command-char' (use
1400 `unread-command-events'). Many (most) of the new ways of doing things
1401 are compatible in GNU Emacs and XEmacs.
1403 Modern Emacs packages (Gnus, VM, W3, efs, etc) are written to support
1404 GNU Emacs and XEmacs. We have provided modified versions of several
1405 popular emacs packages (dired, etc) which are compatible with this
1406 version of emacs. Check to make sure you have not set your load-path
1407 so that your private copies of these packages are being found before
1408 the versions in the lisp directory.
1410 Make sure that your load-path and your $EMACSLOADPATH environment
1411 variable are not pointing at an Emacs18 lisp directory. This will
1414 ** Some packages that worked before now cause the error
1415 Wrong type argument: arrayp, #<face ... >
1417 Code which uses the `face' accessor functions must be recompiled with
1418 xemacs 19.9 or later. The functions whose callers must be recompiled
1419 are: face-font, face-foreground, face-background,
1420 face-background-pixmap, and face-underline-p. The .elc files
1421 generated by version 19.9 will work in 19.6 and 19.8, but older .elc
1422 files which contain calls to these functions will not work in 19.9.
1424 ** Signaling: (error "Byte code stack underflow (byte compiler bug), pc 38")
1426 This error is given when XEmacs 20 is compiled without MULE support
1427 but is attempting to load a .elc which requires MULE support. The fix
1428 is to rebytecompile the offending file.
1430 ** Signaling: (wrong-type-argument ...) when loading mail-abbrevs
1432 The is seen when installing the Insidious Big Brother Data Base (bbdb)
1433 which includes an outdated copy of mail-abbrevs.el. Remove the copy
1434 that comes with bbdb and use the one that comes with XEmacs.
1440 ** A reminder: XEmacs/Mule work does not currently receive *any*
1441 funding, and all work is done by volunteers. If you think you can
1442 help, please contact the XEmacs maintainers.
1444 ** XEmacs/Mule doesn't support TTY's satisfactorily.
1446 This is a major problem, which we plan to address in a future release
1447 of XEmacs. Basically, XEmacs should have primitives to be told
1448 whether the terminal can handle international output, and which
1449 locale. Also, it should be able to do approximations of characters to
1450 the nearest supported by the locale.
1452 ** Internationalized (Asian) Isearch doesn't work.
1454 Currently, Isearch doesn't directly support any of the input methods
1455 that are not XIM based (like egg, canna and quail) (and there are
1456 potential problems with XIM version too...). If you're using egg
1457 there is a workaround. Hitting <RET> right after C-s to invoke
1458 Isearch will put Isearch in string mode, where a complete string can
1459 be typed into the minibuffer and then processed by Isearch afterwards.
1460 Since egg is now supported in the minibuffer using string mode you can
1461 now use egg to input your Japanese, Korean or Chinese string, then hit
1462 return to send that to Isearch and then use standard Isearch commands
1465 ** Using egg and mousing around while in 'fence' mode screws up my
1468 Don't do this. The fence modes of egg and canna are currently very
1469 modal, and messing with where they expect point to be and what they
1470 think is the current buffer is just asking for trouble. If you're
1471 lucky they will realize that something is awry, and simply delete the
1472 fence, but worst case can trash other buffers too. We've tried to
1473 protect against this where we can, but there still are many ways to
1474 shoot yourself in the foot. So just finish what you are typing into
1475 the fence before reaching for the mouse.
1477 ** Not all languages in Quail are supported like Devanagari and Indian
1478 languages, Lao and Tibetan.
1480 Quail requires more work and testing. Although it has been ported to
1481 XEmacs, it works really well for Japanese and for the European
1484 ** Right-to-left mode is not yet implemented, so languages like
1485 Arabic, Hebrew and Thai don't work.
1487 Getting this right requires more work. It may be implemented in a
1488 future XEmacs version, but don't hold your breath. If you know
1489 someone who is ready to implement this, please let us know.
1491 ** We need more developers and native language testers. It's extremely
1492 difficult (and not particularly productive) to address languages that
1493 nobody is using and testing.
1495 ** The kWnn and cWnn support for Chinese and Korean needs developers
1496 and testers. It probably doesn't work.
1498 ** There are no `native XEmacs' TUTORIALs for any Asian languages,
1499 including Japanese. FSF Emacs and XEmacs tutorials are quite similar,
1500 so it should be sufficient to skim through the differences and apply
1501 them to the Japanese version.
1503 ** We only have localized menus translated for Japanese, and the
1504 Japanese menus are developing bitrot (the Mule menu appears in
1507 ** XIM is untested for any language other than Japanese.