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 your init file! (~/.xemacs/init.el or ~/.emacs) If
25 you observe strange problems, invoke XEmacs with the `-vanilla'
26 option and see if you can repeat the problem.
29 * Problems with building XEmacs
30 ===============================
33 *** Don't use -O2 with gcc 2.8.1 and egcs 1.0 under SPARC architectures
34 without also using `-fno-schedule-insns'.
36 gcc will generate incorrect code otherwise, typically resulting in
37 crashes in the function skip-syntax-backward.
41 There have been reports of egcs-1.1 not compiling XEmacs correctly on
42 Alpha Linux. There have also been reports that egcs-1.0.3a is O.K.
44 *** Don't use -O2 or -O3 with Cygwin 1.0, CodeFusion-99070 or gcc 2.7.2 on x86
45 without also using `-fno-strength-reduce'.
47 gcc will generate incorrect code otherwise. This bug is present in at
48 least 2.6.x and 2.7.[0-2]. This bug has been fixed in GCC 2.7.2.1 and
49 later. This bug is O/S independent, but is limited to x86 architectures.
51 This problem is known to be fixed in egcs (or pgcc) 1.0 or later.
53 Unfortunately, later releases of Cygnus-released compilers (not the
54 Net-released ones) have a bug with the same `problem signature'.
56 If you're lucky, you'll get an error while compiling that looks like:
58 event-stream.c:3189: internal error--unrecognizable insn:
59 (insn 256 14 15 (set (reg/v:SI 24)
60 (minus:SI (reg/v:SI 25)
61 (const_int 2))) -1 (insn_list 11 (nil))
65 If you're unlucky, your code will simply execute incorrectly.
67 *** Don't use gcc-2.95.2 with -mcpu=ultrasparc on Solaris 2.6.
69 gcc will assume a 64-bit operating system, even though you've
70 merely told it to assume a 64-bit instruction set.
72 *** Don't use -O2 with gcc 2.7.2 under Intel architectures without also
73 using `-fno-caller-saves'.
75 gcc will generate incorrect code otherwise. This bug is still
76 present in gcc 2.7.2.3. There have been no reports to indicate the
77 bug is present in egcs 1.0 (or pgcc 1.0) or later. This bug is O/S
78 independent, but limited to x86 architectures.
80 This problem is known to be fixed in egcs (or pgcc) 1.0 or later.
82 *** When using gcc, you get the error message "undefined symbol __fixunsdfsi".
83 When using gcc, you get the error message "undefined symbol __main".
85 This means that you need to link with the gcc library. It may be called
86 "gcc-gnulib" or "libgcc.a"; figure out where it is, and define LIB_GCC in
87 config.h to point to it.
89 It may also work to use the GCC version of `ld' instead of the standard one.
91 *** Excessive optimization with pgcc can break XEmacs
93 It has been reported on some systems that compiling with -O6 can lead
94 to XEmacs failures. The workaround is to use a lower optimization
95 level. -O2 and -O4 have been tested extensively.
97 All of this depends heavily on the version of pgcc and the version
98 of libc. Snapshots near the release of pgcc-1.0 have been tested
99 extensively and no sign of breakage has been seen on systems using
102 *** src/Makefile and lib-src/Makefile are truncated--most of the file missing.
104 This can happen if configure uses GNU sed version 2.03. That version
105 had a bug. GNU sed version 2.05 works properly.
107 *** When compiling with X11, you get "undefined symbol _XtStrings".
109 This means that you are trying to link emacs against the X11r4 version of
110 libXt.a, but you have compiled either Emacs or the code in the lwlib
111 subdirectory with the X11r5 header files. That doesn't work.
113 Remember, you can't compile lwlib for r4 and emacs for r5, or vice versa.
114 They must be in sync.
116 *** test-distrib says that the distribution has been clobbered
117 or, temacs prints "Command key out of range 0-127"
118 or, temacs runs and dumps xemacs, but xemacs totally fails to work.
119 or, temacs gets errors dumping xemacs
121 This can be because the .elc files have been garbled. Do not be
122 fooled by the fact that most of a .elc file is text: these are binary
123 files and can contain all 256 byte values.
125 In particular `shar' cannot be used for transmitting GNU Emacs. It
126 typically truncates "lines". (this does not apply to GNU shar, which
127 uses uuencode to encode binary files.)
129 If you have a copy of Emacs that has been damaged in its nonprinting
130 characters, you can fix them by running:
134 This will rebuild all the needed .elc files.
136 *** `compress' and `uncompress' not found and XFree86
138 XFree86 installs a very old version of libz.a by default ahead of where
139 more modern version of libz might be installed. This will cause problems
140 when attempting to link against libMagick. The fix is to remove the old
141 libz.a in the X11 binary directory.
145 *** On AIX 4.3, you must specify --with-dialogs=athena with configure
147 *** The libXt shipped with AIX 4.3 up to 4.3.2 is broken. This causes
148 xemacs -nw to fail in various ways. The official APAR is this:
150 APAR NUMBER: <IX89470> RESOLVED AS: PROGRAM ERROR
153 <IX89470>: LIBXT.A INCORRECT HANDLING OF EXCEPTIONS IN XTAPPADDINPUT
155 The solution is to install X11.base.lib at version >=4.3.2.5.
157 *** On AIX, you get this compiler error message:
159 Processing include file ./XMenuInt.h
160 1501-106: (S) Include file X11/Xlib.h not found.
162 This means your system was installed with only the X11 runtime i.d
163 libraries. You have to find your sipo (bootable tape) and install
166 *** On AIX 4.1.2, linker error messages such as
167 ld: 0711-212 SEVERE ERROR: Symbol .__quous, found in the global symbol table
168 of archive /usr/lib/libIM.a, was not defined in archive member shr.o.
170 This is a problem in libIM.a. You can work around it by executing
171 these shell commands in the src subdirectory of the directory where
174 cp /usr/lib/libIM.a .
178 Then change -lIM to ./libIM.a in the command to link temacs (in
181 *** Excessive optimization on AIX 4.2 can lead to compiler failure.
183 Valdis.Kletnieks@vt.edu writes:
184 At least at the b34 level, and the latest-and-greatest IBM xlc
185 (3.1.4.4), there are problems with -O3. I haven't investigated
190 *** Dumping error when using GNU binutils / GNU ld on a Sun.
192 Errors similar to the following:
194 Dumping under the name xemacs unexec():
195 dldump(/space/rpluim/xemacs-obj/src/xemacs): ld.so.1: ./temacs:
196 fatal: /space/rpluim/xemacs-obj/src/xemacs: unknown dynamic entry:
199 are caused by using GNU ld. There are several workarounds available:
201 In XEmacs 21.2 or later, configure using the new portable dumper
204 Alternatively, you can link using the Sun version of ld, which is
205 normally held in /usr/ccs/bin. This can be done by one of:
207 - building gcc with these configure flags:
208 configure --with-ld=/usr/ccs/bin/ld --with-as=/usr/ccs/bin/as
210 - adding -B/usr/ccs/bin/ to CFLAGS used to configure XEmacs
211 (Note: The trailing '/' there is significant.)
213 - uninstalling GNU ld.
215 The Solaris2 FAQ claims:
217 When you install gcc, don't make the mistake of installing
218 GNU binutils or GNU libc, they are not as capable as their
219 counterparts you get with Solaris 2.x.
221 *** Link failure when using acc on a Sun.
223 To use acc, you need additional options just before the libraries, such as
225 /usr/lang/SC2.0.1/values-Xt.o -L/usr/lang/SC2.0.1/cg87 -L/usr/lang/SC2.0.1
227 and you need to add -lansi just before -lc.
229 The precise file names depend on the compiler version, so we
230 cannot easily arrange to supply them.
232 *** Problems finding X11 libraries on Solaris with Openwindows
234 Some users have reported problems in this area. The reported solution
235 is to define the environment variable OPENWINHOME, even if you must set
236 it to `/usr/openwin'.
238 *** Sed problems on Solaris 2.5
240 There have been reports of Sun sed truncating very lines in the
241 Makefile during configuration. The workaround is to use GNU sed or,
242 even better, think of a better way to generate Makefile, and send us a
245 *** On Solaris 2 I get undefined symbols from libcurses.a.
247 You probably have /usr/ucblib/ on your LD_LIBRARY_PATH. Do the link with
248 LD_LIBRARY_PATH unset. Generally, avoid using any ucb* stuff when
251 *** On Solaris 2 I cannot make alloc.o, glyphs.o or process.o.
253 The SparcWorks C compiler may have difficulty building those modules
254 with optimization level -xO4. Try using only "-fast" optimization
255 for just those modules. (Or use gcc).
257 *** Solaris 2.3 /bin/sh coredumps during configuration.
259 This only occurs if you have LANG != C. This is a known bug with
260 /bin/sh fixed by installing Patch-ID# 101613-01. Or, you can use
261 bash, as a workaround.
263 *** On SunOS, you get linker errors
265 _get_wmShellWidgetClass
266 _get_applicationShellWidgetClass
268 The fix to this is to install patch 100573 for OpenWindows 3.0
269 or link libXmu statically.
271 *** On Sunos 4, you get the error ld: Undefined symbol __lib_version.
273 This is the result of using cc or gcc with the shared library meant
274 for acc (the Sunpro compiler). Check your LD_LIBRARY_PATH and delete
275 /usr/lang/SC2.0.1 or some similar directory.
277 *** Undefined symbols when linking on Sunos 4.1.
279 If you get the undefined symbols _atowc _wcslen, _iswprint, _iswspace,
280 _iswcntrl, _wcscpy, and _wcsncpy, then you need to add -lXwchar after
281 -lXaw in the command that links temacs.
283 This problem seems to arise only when the international language
284 extensions to X11R5 are installed.
286 *** On a Sun running SunOS 4.1.1, you get this error message from GNU ld:
288 /lib/libc.a(_Q_sub.o): Undefined symbol __Q_get_rp_rd referenced from text segment
290 The problem is in the Sun shared C library, not in GNU ld.
292 The solution is to install Patch-ID# 100267-03 from Sun.
294 *** SunOS 4.1.2: undefined symbol _get_wmShellWidgetClass
296 Apparently the version of libXmu.so.a that Sun ships is hosed: it's missing
297 some stuff that is in libXmu.a (the static version). Sun has a patch for
298 this, but a workaround is to use the static version of libXmu, by changing
299 the link command from "-lXmu" to "-Bstatic -lXmu -Bdynamic". If you have
300 OpenWindows 3.0, ask Sun for these patches:
301 100512-02 4.1.x OpenWindows 3.0 libXt Jumbo patch
302 100573-03 4.1.x OpenWindows 3.0 undefined symbols with shared libXmu
304 *** Random other SunOS 4.1.[12] link errors.
306 The X headers and libraries that Sun ships in /usr/{include,lib}/X11 are
307 broken. Use the ones in /usr/openwin/{include,lib} instead.
310 *** Under Linux, you get "too many arguments to function `getpgrp'".
312 You have probably installed LessTiff under `/usr/local' and `libXm.so'
313 could not be found when linking `getpgrp()' test program, making XEmacs
314 think that `getpgrp()' takes an argument. Try adding `/usr/local/lib'
315 in `/etc/ld.so.conf' and run `ldconfig'. Then run XEmacs's `configure'
316 again. As with all problems of this type, reading the config.log file
317 generated from configure and seeing the log of how the test failed can
320 *** `Error: No ExtNode to pop!' on Linux systems with Lesstif.
322 This error message has been observed with lesstif-0.75a. It does not
323 appear to cause any harm.
325 *** xemacs: can't resolve symbol '__malloc_hook'
327 This is a Linux problem where you've compiled the XEmacs binary on a libc
328 5.4 with version higher than 5.4.19 and attempted to run the binary against
329 an earlier version. The solution is to upgrade your old library.
333 *** On Irix 6.5, the MIPSpro compiler gets an internal compiler error
335 The MIPSpro Compiler (at least version 7.2.1) can't seem to handle the
336 union type properly, and fails to compile src/glyphs.c. To avoid this
337 problem, always build ---use-union-type=no (but that's the default, so
338 you should only see this problem if you're an XEmacs maintainer).
340 *** Linking with -rpath on IRIX.
342 Darrell Kindred <dkindred@cmu.edu> writes:
343 There are a couple of problems [with use of -rpath with Irix ld], though:
345 1. The ld in IRIX 5.3 ignores all but the last -rpath
346 spec, so the patched configure spits out a warning
347 if --x-libraries or --site-runtime-libraries are
348 specified under irix 5.x, and it only adds -rpath
349 entries for the --site-runtime-libraries. This bug was
350 fixed sometime between 5.3 and 6.2.
352 2. IRIX gcc 2.7.2 doesn't accept -rpath directly, so
353 it would have to be prefixed by -Xlinker or "-Wl,".
354 This would be fine, except that configure compiles with
355 ${CC-cc} $CFLAGS $LDFLAGS ...
356 rather than quoting $LDFLAGS with prefix-args, like
357 src/Makefile does. So if you specify --x-libraries
358 or --site-runtime-libraries, you must use --use-gcc=no,
359 or configure will fail.
361 *** On Irix 6.3, the SGI ld quits with segmentation fault when linking temacs
363 This occurs if you use the SGI linker version 7.1. Installing the
364 patch SG0001872 fixes this problem.
366 *** On Irix 6.0, make tries (and fails) to build a program named unexelfsgi
368 A compiler bug inserts spaces into the string "unexelfsgi . o"
369 in src/Makefile. Edit src/Makefile, after configure is run,
370 find that string, and take out the spaces.
372 Compiler fixes in Irix 6.0.1 should eliminate this problem.
374 *** On Irix 5.2, unexelfsgi.c can't find cmplrs/stsupport.h.
376 The file cmplrs/stsupport.h was included in the wrong file set in the
377 Irix 5.2 distribution. You can find it in the optional fileset
378 compiler_dev, or copy it from some other Irix 5.2 system. A kludgy
379 workaround is to change unexelfsgi.c to include sym.h instead of
382 *** Coredumping in Irix 6.2
384 Pete Forman <gsez020@compo.bedford.waii.com> writes:
385 A problem noted by myself and others (I've lost the references) was
386 that XEmacs coredumped when the cut or copy toolbar buttons were
387 pressed. This has been fixed by loading the SGI patchset (Feb 98)
388 without having to recompile XEmacs.
390 My versions are XEmacs 20.3 (problem first noted in 19.15) and IRIX
391 6.2, compiled using -n32. I'd guess that the relevant individual
392 patch was "SG0002580: multiple fixes for X libraries". SGI recommends
393 that the complete patch set be installed rather than parts of it.
395 ** Digital UNIX/OSF/VMS
396 *** On Digital UNIX, the DEC C compiler might have a problem compiling
399 In particular, src/extents.c and src/faces.c might cause the DEC C
400 compiler to abort. When this happens: cd src, compile the files by
401 hand, cd .., and redo the "make" command. When recompiling the files by
402 hand, use the old C compiler for the following versions of Digital UNIX:
403 - V3.n: Remove "-migrate" from the compile command.
404 - V4.n: Add "-oldc" to the compile command.
406 A related compiler bug has been fixed by the DEC compiler team. The
407 new versions of the compiler should run fine.
409 *** Under some versions of OSF XEmacs runs fine if built without
410 optimization but will crash randomly if built with optimization.
412 Using 'cc -g' is not sufficient to eliminate all optimization. Try
415 *** Compilation errors on VMS.
417 Sorry, XEmacs does not work under VMS. You might consider working on
418 the port if you really want to have XEmacs work under VMS.
421 *** On HPUX, the HP C compiler might have a problem compiling some files
424 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
426 Had to drop once again to level 2 optimization, at least to
427 compile lstream.c. Otherwise, I get a "variable is void: \if"
428 problem while dumping (this is a problem I already reported
429 with vanilla hpux 10.01 and 9.07, which went away after
430 applying patches for the C compiler). Trouble is I still
431 haven't found the same patch for hpux 10.10, and I don't
432 remember the patch numbers. I think potential XEmacs builders
433 on HP should be warned about this.
435 *** I don't have `xmkmf' and `imake' on my HP.
437 You can get these standard X tools by anonymous FTP to
438 hpcvaaz.cv.hp.com. Essentially all X programs need these.
440 *** On HP-UX, problems with make
442 Marcus Thiessel <marcus@xemacs.org>
444 Some releases of XEmacs (e.g. 20.4) require GNU make to build
445 successfully. You don't need GNU make when building 21.x.
447 *** On HP-UX 9.05 XEmacs won't compile or coredump during the build.
449 Marcus Thiessel <marcus@xemacs.org>
451 This might be a sed problem. For your own safety make sure to use
452 GNU sed while dumping XEmacs.
454 *** On HP-UX 11.0 XEmacs causes excessive X11 errors when running.
455 (also appears on AIX as reported in comp.emacs.xemacs)
457 Marcus Thiessel <marcus@xemacs.org>
459 Unfortunately, XEmacs releases prior to 21.0 don't work with
460 Motif2.1. It will compile but you will get excessive X11 errors like
462 xemacs: X Error of failed request: BadGC (invalid GC parameter)
464 and finally XEmacs gets killed. A workaround is to use the
465 Motif1.2_R6 libraries. You can the following line to your call to
468 --x-libraries="/usr/lib/Motif1.2_R6 -L/usr/lib/X11R6"
470 Make sure /usr/lib/Motif1.2_R6/libXm.sl is a link to
471 /usr/lib/Motif1.2_R6/libXm.3.
473 *** On HP-UX 11.0: Object "" does not have windowed ancestor
475 Marcus Thiessel <marcus@xemacs.org>
477 XEmacs dies without core file and reports:
479 Error: Object "" does not have windowed ancestor.
481 This is a bug. Please apply the patch PHSS_19964 (check if
482 superseded). The other alternative is to link with Motif1.2_R6 (see
487 *** Native cc on SCO OpenServer 5 is now OK. Icc may still throw you
488 a curve. Here is what Robert Lipe <robertl@arnet.com> says:
490 Unlike XEmacs 19.13, building with the native cc on SCO OpenServer 5
491 now produces a functional binary. I will typically build this
492 configuration for COFF with:
494 /path_to_xemacs_source/configure --with-gcc=no \
495 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
496 --with-xpm --with-xface --with-sound=nas
498 This version now supports ELF builds. I highly recommend this to
499 reduce the in-core footprint of XEmacs. This is now how I compile
500 all my test releases. Build it like this:
502 /path_to_XEmacs_source/configure --with-gcc=no \
503 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
504 --with-xpm --with-xface --with-sound=nas --dynamic
506 The compiler known as icc [ supplied with the OpenServer 5 Development
507 System ] generates a working binary, but it takes forever to generate
508 XEmacs. ICC also whines more about the code than /bin/cc does. I do
509 believe all its whining is legitimate, however. Note that you do
510 have to 'cd src ; make LD=icc' to avoid linker errors.
512 The way I handle the build procedure is:
514 /path_to_XEmacs_source/configure --with-gcc=no \
515 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
516 --with-xpm --with-xface --with-sound=nas --dynamic --compiler="icc"
518 NOTE I have the xpm, xface, and audio libraries and includes in
519 /usr/local/lib, /usr/local/include. If you don't have these,
520 don't include the "--with-*" arguments in any of my examples.
522 In previous versions of XEmacs, you had to override the defaults while
523 compiling font-lock.o and extents.o when building with icc. This seems
524 to no longer be true, but I'm including this old information in case it
525 resurfaces. The process I used was:
528 [ procure pizza, beer, repeat ]
530 make CC="icc -W0,-mP1COPT_max_tree_size=3000" font-lock.o extents.o
533 If you want sound support, get the tls566 supplement from
534 ftp.sco.com:/TLS or any of its mirrors. It works just groovy
537 The M-x manual-entry is known not to work. If you know Lisp and would
538 like help in making it work, e-mail me at <robertl@dgii.com>.
539 (UNCHECKED for 19.15 -- it might work).
541 In earlier releases, gnuserv/gnuclient/gnudoit would open a frame
542 just fine, but the client would lock up and the server would
543 terminate when you used C-x # to close the frame. This is now
546 In etc/ there are two files of note. emacskeys.sco and emacsstrs.sco.
547 The comments at the top of emacskeys.sco describe its function, and
548 the emacstrs.sco is a suitable candidate for /usr/lib/keyboard/strings
549 to take advantage of the keyboard map in emacskeys.sco.
551 Note: Much of the above entry is probably not valid for XEmacs 21.0
556 *** In general use etc/check_cygwin_setup.sh to trap environment problems.
558 The script etc/check_cygwin_setup.sh will attempt to detect whether
559 you have a suitable environment for building. This script may not work
560 correctly if you are using ash instead of bash (see below).
562 *** Syntax errors running configure scripts, make failing with exit code 127
563 in inexplicable situations, etc.
565 This may be because you are using the default cygwin shell. The
566 default cygwin shell (/bin/sh.exe) is ash which appears to work in
567 most circumstances but has some weird failure modes. You need to
568 replace the symlink with bash.exe.
570 *** Lots of compile errors, esp. on lines containing macro definitions
571 terminated by backslashes.
573 Your partition holding the source files is mounted binary. It needs
574 to be mounted text. (This will not screw up any binary files because
575 the Cygwin utilities specify explicitly whether they want binary or
576 text mode when working with source vs. binary files, which overrides
577 the mount type.) To fix this, you just need to run the appropriate
578 mount command once -- afterwards, the settings are remembered in the
581 *** Errors from make like /c:not found.
583 make sure you set the environment variable MAKE_MODE to UNIX in your
584 .bashrc, Control Panel (Windows 2000/NT), or AUTOEXEC.BAT (Windows
587 *** X11 not detected.
589 This is usually because xmkmf is not in your path or because you are
590 using the default cygwin shell. The default cygwin shell (/bin/sh.exe)
591 is ash which appears to work in most circumstances but has some weird
592 failure modes. I recommend replacing sh.exe with bash.exe, this will
593 mean configure is slower but more reliable.
595 *** Subprocesses do not work.
597 You do not have "tty" in your CYGWIN environment variable. This must
598 be set in your autoexec.bat (win95) or the system properties (winnt)
599 as it must be read before the cygwin DLL initializes.
601 *** ^G does not work on hung subprocesses.
603 This is a known problem. It can be remedied with cygwin b20 or greater
604 by defining BROKEN_SIGIO in src/s/cygwin32.h, however this currently
605 leads to instability in XEmacs.
607 *** The info files will not build.
609 makeinfo that ships with cygwin (all versions) is a noop. You need to
610 obtain makeinfo from somewhere or build it yourself.
612 *** I have no graphics.
614 You need to obtain the various graphics libraries. Pre-built versions
615 of these and the X libraries are located on the XEmacs website in
616 ftp://ftp.xemacs.org/pub/aux/cygwin/.
618 *** There are no images in the toolbar buttons.
620 You need version 4.71 of commctrl.dll which does not ship with windows
621 95. You can get this by installing IE 4.0 or downloading it from the
625 * Problems with running XEmacs
626 ==============================
628 *** gnuserv is running, some clients can connect, but others cannot.
630 The code in gnuslib.c respects the value of TMPDIR. If the server and
631 the client have different values in their environment, you lose.
632 One program known to set TMPDIR and manifest this problem is exmh.
633 You can defeat the use of TMPDIR by unsetting USE_TMPDIR at the top of
634 gnuserv.h at build time.
636 *** C-z just refreshes the screen instead of suspending Emacs.
638 You are probably using a shell that doesn't support job control, even
639 though the system itself is capable of it. Try using a different
642 *** You type Control-H (Backspace) expecting to delete characters.
644 Emacs has traditionally used Control-H for help; unfortunately this
645 interferes with its use as Backspace on TTY's. As of XEmacs 21,
646 XEmacs looks at the "erase" setting of TTY structures and maps C-h to
647 backspace when erase is set to C-h. This is sort of a special hack,
648 but it makes it possible for you to use the standard:
652 to get your backspace key to erase characters. The erase setting is
653 recorded in the Lisp variable `tty-erase-char', which you can use to
654 tune the settings in your .emacs.
656 A major drawback of this is that when C-h becomes backspace, it no
657 longer invokes help. In that case, you need to use f1 for help, or
658 bind another key. An example of the latter is the following code,
659 which moves help to Meta-? (ESC ?):
661 (global-set-key "\M-?" 'help-command)
663 *** Mail agents (VM, Gnus, rmail) cannot get new mail
665 rmail and VM get new mail from /usr/spool/mail/$USER using a program
666 called `movemail'. This program interlocks with /bin/mail using the
667 protocol defined by /bin/mail.
669 There are two different protocols in general use. One of them uses
670 the `flock' system call. The other involves creating a lock file;
671 `movemail' must be able to write in /usr/spool/mail in order to do
672 this. You control which one is used by defining, or not defining, the
673 macro MAIL_USE_FLOCK in config.h or the m- or s- file it includes. IF
674 YOU DON'T USE THE FORM OF INTERLOCKING THAT IS NORMAL ON YOUR SYSTEM,
677 If your system uses the lock file protocol, and fascist restrictions
678 prevent ordinary users from writing the lock files in /usr/spool/mail,
679 you may need to make `movemail' setgid to a suitable group such as
680 `mail'. To do this, use the following commands (as root) after doing
686 Installation normally copies movemail from the build directory to an
687 installation directory which is usually under /usr/local/lib. The
688 installed copy of movemail is usually in the directory
689 /usr/local/lib/emacs/VERSION/TARGET. You must change the group and
690 mode of the installed copy; changing the group and mode of the build
691 directory copy is ineffective.
693 *** VM appears to hang in large folders.
695 This is normal (trust us) when upgrading to VM-6.22 from earlier
696 versions. Let VM finish what it is doing and all will be well.
698 *** Changes made to .el files do not take effect.
700 You may have forgotten to recompile them into .elc files. Then the
701 old .elc files will be loaded, and your changes will not be seen. To
702 fix this, do `M-x byte-recompile-directory' and specify the directory
703 that contains the Lisp files.
705 Note that you will get a warning when loading a .elc file that is
706 older than the corresponding .el file.
708 *** Things which should be bold or italic (such as the initial
709 copyright notice) are not.
711 The fonts of the "bold" and "italic" faces are generated from the font
712 of the "default" face; in this way, your bold and italic fonts will
713 have the appropriate size and family. However, emacs can only be
714 clever in this way if you have specified the default font using the
715 XLFD (X Logical Font Description) format, which looks like
717 *-courier-medium-r-*-*-*-120-*-*-*-*-*-*
719 if you use any of the other, less strict font name formats, some of
722 lucidasanstypewriter-12
726 then emacs won't be able to guess the names of the "bold" and "italic"
727 versions. All X fonts can be referred to via XLFD-style names, so you
728 should use those forms. See the man pages for X(1), xlsfonts(1), and
731 *** The dumped Emacs crashes when run, trying to write pure data.
733 Two causes have been seen for such problems.
735 1) On a system where getpagesize is not a system call, it is defined
736 as a macro. If the definition (in both unexec.c and malloc.c) is wrong,
737 it can cause problems like this. You might be able to find the correct
738 value in the man page for a.out (5).
740 2) Some systems allocate variables declared static among the
741 initialized variables. Emacs makes all initialized variables in most
742 of its files pure after dumping, but the variables declared static and
743 not initialized are not supposed to be pure. On these systems you
744 may need to add "#define static" to the m- or the s- file.
746 *** Reading and writing files is very very slow.
748 Try evaluating the form (setq lock-directory nil) and see if that helps.
749 There is a problem with file-locking on some systems (possibly related
750 to NFS) that I don't understand. Please send mail to the address
751 xemacs-beta@xemacs.org if you figure this one out.
753 *** When emacs starts up, I get lots of warnings about unknown keysyms.
755 If you are running the prebuilt binaries, the Motif library expects to find
756 certain thing in the XKeysymDB file. This file is normally in /usr/lib/X11/
757 or in /usr/openwin/lib/. If you keep yours in a different place, set the
758 environment variable $XKEYSYMDB to point to it before starting emacs. If
759 you still have the problem after doing that, perhaps your version of X is
760 too old. There is a copy of the MIT X11R5 XKeysymDB file in the emacs `etc'
761 directory. Try using that one.
763 *** My X resources used to work, and now some of them are being ignored.
765 Check the resources in .../etc/Emacs.ad (which is the same as the file
766 sample.Xdefaults). Perhaps some of the default resources built in to
767 emacs are now overriding your existing resources. Copy and edit the
768 resources in Emacs.ad as necessary.
770 *** I have focus problems when I use `M-o' to switch to another screen
771 without using the mouse.
773 The focus issues with a program like XEmacs, which has multiple
774 homogeneous top-level windows, are very complicated, and as a result,
775 most window managers don't implement them correctly.
777 The R4/R5 version of twm (and all of its descendants) had buggy focus
778 handling. Sufficiently recent versions of tvtwm have been fixed. In
779 addition, if you're using twm, make sure you have not specified
780 "NoTitleFocus" in your .tvtwmrc file. The very nature of this option
781 makes twm do some illegal focus tricks, even with the patch.
783 It is known that olwm and olvwm are buggy, and in different ways. If
784 you're using click-to-type mode, try using point-to-type, or vice
787 In older versions of NCDwm, one could not even type at XEmacs windows.
788 This has been fixed in newer versions (2.4.3, and possibly earlier).
790 (Many people suggest that XEmacs should warp the mouse when focusing
791 on another screen in point-to-type mode. This is not ICCCM-compliant
792 behavior. Implementing such policy is the responsibility of the
793 window manager itself, it is not legal for a client to do this.)
795 *** Emacs spontaneously displays "I-search: " at the bottom of the screen.
797 This means that Control-S/Control-Q (XON/XOFF) "flow control" is being
798 used. C-s/C-q flow control is bad for Emacs editors because it takes
799 away C-s and C-q as user commands. Since editors do not output long
800 streams of text without user commands, there is no need for a
801 user-issuable "stop output" command in an editor; therefore, a
802 properly designed flow control mechanism would transmit all possible
803 input characters without interference. Designing such a mechanism is
804 easy, for a person with at least half a brain.
806 There are three possible reasons why flow control could be taking place:
808 1) Terminal has not been told to disable flow control
809 2) Insufficient padding for the terminal in use
810 3) Some sort of terminal concentrator or line switch is responsible
812 First of all, many terminals have a set-up mode which controls whether
813 they generate XON/XOFF flow control characters. This must be set to
814 "no XON/XOFF" in order for Emacs to work. Sometimes there is an
815 escape sequence that the computer can send to turn flow control off
816 and on. If so, perhaps the termcap `ti' string should turn flow
817 control off, and the `te' string should turn it on.
819 Once the terminal has been told "no flow control", you may find it
820 needs more padding. The amount of padding Emacs sends is controlled
821 by the termcap entry for the terminal in use, and by the output baud
822 rate as known by the kernel. The shell command `stty' will print
823 your output baud rate; `stty' with suitable arguments will set it if
824 it is wrong. Setting to a higher speed causes increased padding. If
825 the results are wrong for the correct speed, there is probably a
826 problem in the termcap entry. You must speak to a local Unix wizard
827 to fix this. Perhaps you are just using the wrong terminal type.
829 For terminals that lack a "no flow control" mode, sometimes just
830 giving lots of padding will prevent actual generation of flow control
831 codes. You might as well try it.
833 If you are really unlucky, your terminal is connected to the computer
834 through a concentrator which sends XON/XOFF flow control to the
835 computer, or it insists on sending flow control itself no matter how
836 much padding you give it. Unless you can figure out how to turn flow
837 control off on this concentrator (again, refer to your local wizard),
838 you are screwed! You should have the terminal or concentrator
839 replaced with a properly designed one. In the mean time, some drastic
840 measures can make Emacs semi-work.
842 You can make Emacs ignore C-s and C-q and let the operating system
843 handle them. To do this on a per-session basis, just type M-x
844 enable-flow-control RET. You will see a message that C-\ and C-^ are
845 now translated to C-s and C-q. (Use the same command M-x
846 enable-flow-control to turn *off* this special mode. It toggles flow
849 If C-\ and C-^ are inconvenient for you (for example, if one of them
850 is the escape character of your terminal concentrator), you can choose
851 other characters by setting the variables flow-control-c-s-replacement
852 and flow-control-c-q-replacement. But choose carefully, since all
853 other control characters are already used by emacs.
855 IMPORTANT: if you type C-s by accident while flow control is enabled,
856 Emacs output will freeze, and you will have to remember to type C-q in
859 If you work in an environment where a majority of terminals of a
860 certain type are flow control hobbled, you can use the function
861 `enable-flow-control-on' to turn on this flow control avoidance scheme
862 automatically. Here is an example:
864 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
866 If this isn't quite correct (e.g. you have a mixture of flow-control hobbled
867 and good vt200 terminals), you can still run enable-flow-control
870 I have no intention of ever redesigning the Emacs command set for the
871 assumption that terminals use C-s/C-q flow control. XON/XOFF flow
872 control technique is a bad design, and terminals that need it are bad
873 merchandise and should not be purchased. Now that X is becoming
874 widespread, XON/XOFF seems to be on the way out. If you can get some
875 use out of GNU Emacs on inferior terminals, more power to you, but I
876 will not make Emacs worse for properly designed systems for the sake
879 *** Control-S and Control-Q commands are ignored completely.
881 For some reason, your system is using brain-damaged C-s/C-q flow
882 control despite Emacs's attempts to turn it off. Perhaps your
883 terminal is connected to the computer through a concentrator
884 that wants to use flow control.
886 You should first try to tell the concentrator not to use flow control.
887 If you succeed in this, try making the terminal work without
888 flow control, as described in the preceding section.
890 If that line of approach is not successful, map some other characters
891 into C-s and C-q using keyboard-translate-table. The example above
892 shows how to do this with C-^ and C-\.
894 *** Control-S and Control-Q commands are ignored completely on a net
897 Some versions of rlogin (and possibly telnet) do not pass flow
898 control characters to the remote system to which they connect.
899 On such systems, emacs on the remote system cannot disable flow
900 control on the local system.
902 One way to cure this is to disable flow control on the local host
903 (the one running rlogin, not the one running rlogind) using the
904 stty command, before starting the rlogin process. On many systems,
905 `stty start u stop u' will do this.
907 Some versions of tcsh will prevent even this from working. One way
908 around this is to start another shell before starting rlogin, and
909 issue the stty command to disable flow control from that shell.
911 If none of these methods work, the best solution is to type
912 `M-x enable-flow-control' at the beginning of your emacs session, or
913 if you expect the problem to continue, add a line such as the
914 following to your .emacs (on the host running rlogind):
916 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
918 See the entry about spontaneous display of I-search (above) for more
921 *** TTY redisplay is slow.
923 XEmacs has fairly new TTY redisplay support (beginning from 19.12),
924 which doesn't include some basic TTY optimizations -- like using
925 scrolling regions to move around blocks of text. This is why
926 redisplay on the traditional terminals, or over slow lines can be very
929 If you are interested in fixing this, please let us know at
930 <xemacs-beta@xemacs.org>.
932 *** Screen is updated wrong, but only on one kind of terminal.
934 This could mean that the termcap entry you are using for that terminal
935 is wrong, or it could mean that Emacs has a bug handing the
936 combination of features specified for that terminal.
938 The first step in tracking this down is to record what characters
939 Emacs is sending to the terminal. Execute the Lisp expression
940 (open-termscript "./emacs-script") to make Emacs write all terminal
941 output into the file ~/emacs-script as well; then do what makes the
942 screen update wrong, and look at the file and decode the characters
943 using the manual for the terminal. There are several possibilities:
945 1) The characters sent are correct, according to the terminal manual.
947 In this case, there is no obvious bug in Emacs, and most likely you
948 need more padding, or possibly the terminal manual is wrong.
950 2) The characters sent are incorrect, due to an obscure aspect of the
951 terminal behavior not described in an obvious way by termcap.
953 This case is hard. It will be necessary to think of a way for Emacs
954 to distinguish between terminals with this kind of behavior and other
955 terminals that behave subtly differently but are classified the same
956 by termcap; or else find an algorithm for Emacs to use that avoids the
957 difference. Such changes must be tested on many kinds of terminals.
959 3) The termcap entry is wrong.
961 See the file etc/TERMS for information on changes that are known to be
962 needed in commonly used termcap entries for certain terminals.
964 4) The characters sent are incorrect, and clearly cannot be right for
965 any terminal with the termcap entry you were using.
967 This is unambiguously an Emacs bug, and can probably be fixed in
968 termcap.c, terminfo.c, tparam.c, cm.c, redisplay-tty.c,
969 redisplay-output.c, or redisplay.c.
971 *** My buffers are full of \000 characters or otherwise corrupt.
973 Some compilers have trouble with gmalloc.c and ralloc.c; try recompiling
974 without optimization. If that doesn't work, try recompiling with
975 SYSTEM_MALLOC defined, and/or with REL_ALLOC undefined.
977 *** A position you specified in .Xdefaults is ignored, using twm.
979 twm normally ignores "program-specified" positions.
980 You can tell it to obey them with this command in your `.twmrc' file:
982 UsePPosition "on" #allow clents to request a position
984 *** With M-x enable-flow-control, you need to type C-\ twice to do
985 incremental search--a single C-\ gets no response.
987 This has been traced to communicating with your machine via kermit,
988 with C-\ as the kermit escape character. One solution is to use
989 another escape character in kermit. One user did
991 set escape-character 17
993 in his .kermrc file, to make C-q the kermit escape character.
995 *** The Motif version of Emacs paints the screen a solid color.
997 This has been observed to result from the following X resource:
999 Emacs*default.attributeFont: -*-courier-medium-r-*-*-*-140-*-*-*-*-iso8859-*
1001 That the resource has this effect indicates a bug in something, but we
1002 do not yet know what. If it is an Emacs bug, we hope someone can
1003 explain what the bug is so we can fix it. In the mean time, removing
1004 the resource prevents the problem.
1006 *** After running emacs once, subsequent invocations crash.
1008 Some versions of SVR4 have a serious bug in the implementation of the
1009 mmap () system call in the kernel; this causes emacs to run correctly
1010 the first time, and then crash when run a second time.
1012 Contact your vendor and ask for the mmap bug fix; in the mean time,
1013 you may be able to work around the problem by adding a line to your
1014 operating system description file (whose name is reported by the
1015 configure script) that reads:
1016 #define SYSTEM_MALLOC
1017 This makes Emacs use memory less efficiently, but seems to work around
1020 *** Inability to send an Alt-modified key, when Emacs is communicating
1021 directly with an X server.
1023 If you have tried to bind an Alt-modified key as a command, and it
1024 does not work to type the command, the first thing you should check is
1025 whether the key is getting through to Emacs. To do this, type C-h c
1026 followed by the Alt-modified key. C-h c should say what kind of event
1027 it read. If it says it read an Alt-modified key, then make sure you
1028 have made the key binding correctly.
1030 If C-h c reports an event that doesn't have the Alt modifier, it may
1031 be because your X server has no key for the Alt modifier. The X
1032 server that comes from MIT does not set up the Alt modifier by
1035 If your keyboard has keys named Alt, you can enable them as follows:
1037 xmodmap -e 'add mod2 = Alt_L'
1038 xmodmap -e 'add mod2 = Alt_R'
1040 If the keyboard has just one key named Alt, then only one of those
1041 commands is needed. The modifier `mod2' is a reasonable choice if you
1042 are using an unmodified MIT version of X. Otherwise, choose any
1043 modifier bit not otherwise used.
1045 If your keyboard does not have keys named Alt, you can use some other
1046 keys. Use the keysym command in xmodmap to turn a function key (or
1047 some other 'spare' key) into Alt_L or into Alt_R, and then use the
1048 commands show above to make them modifier keys.
1050 Note that if you have Alt keys but no Meta keys, Emacs translates Alt
1051 into Meta. This is because of the great importance of Meta in Emacs.
1053 *** In Shell mode, you get a ^M at the end of every line.
1055 This happens to people who use tcsh, because it is trying to be too
1056 smart. It sees that the Shell uses terminal type `unknown' and turns
1057 on the flag to output ^M at the end of each line. You can fix the
1058 problem by adding this to your .cshrc file:
1061 if ($EMACS == "t") then
1063 stty -icrnl -onlcr -echo susp ^Z
1067 *** An error message such as `X protocol error: BadMatch (invalid
1068 parameter attributes) on protocol request 93'.
1070 This comes from having an invalid X resource, such as
1072 (which is invalid because it specifies a color name for something
1073 that isn't a color.)
1075 The fix is to correct your X resources.
1077 *** Once you pull down a menu from the menubar, it won't go away.
1079 It has been claimed that this is caused by a bug in certain very old
1080 (1990?) versions of the twm window manager. It doesn't happen with
1081 recent vintages, or with other window managers.
1083 *** Emacs ignores the "help" key when running OLWM.
1085 OLWM grabs the help key, and retransmits it to the appropriate client
1086 using XSendEvent. Allowing emacs to react to synthetic events is a
1087 security hole, so this is turned off by default. You can enable it by
1088 setting the variable x-allow-sendevents to t. You can also cause fix
1089 this by telling OLWM to not grab the help key, with the null binding
1090 "OpenWindows.KeyboardCommand.Help:".
1092 *** Programs running under terminal emulator do not recognize `emacs'
1095 The cause of this is a shell startup file that sets the TERMCAP
1096 environment variable. The terminal emulator uses that variable to
1097 provide the information on the special terminal type that Emacs
1100 Rewrite your shell startup file so that it does not change TERMCAP
1101 in such a case. You could use the following conditional which sets
1102 it only if it is undefined.
1104 if ( ! ${?TERMCAP} ) setenv TERMCAP ~/my-termcap-file
1106 Or you could set TERMCAP only when you set TERM--which should not
1107 happen in a non-login shell.
1109 *** The popup menu appears at the bottom/right of my screen.
1111 You probably have something like the following in your ~/.Xdefaults
1113 Emacs.geometry: 81x56--9--1
1115 Use the following instead
1117 Emacs*EmacsFrame.geometry: 81x56--9--1
1121 *** Your Delete key sends a Backspace to the terminal, using an AIXterm.
1123 The solution is to include in your .Xdefaults the lines:
1125 *aixterm.Translations: #override <Key>BackSpace: string(0x7f)
1126 aixterm*ttyModes: erase ^?
1128 This makes your Backspace key send DEL (ASCII 127).
1130 *** On AIX 4, some programs fail when run in a Shell buffer
1131 with an error message like No terminfo entry for "unknown".
1133 On AIX, many terminal type definitions are not installed by default.
1134 `unknown' is one of them. Install the "Special Generic Terminal
1135 Definitions" to make them defined.
1137 *** On AIX, you get this message when running Emacs:
1139 Could not load program emacs
1140 Symbol smtcheckinit in csh is undefined
1141 Error was: Exec format error
1145 Could not load program .emacs
1146 Symbol _system_con in csh is undefined
1147 Symbol _fp_trapsta in csh is undefined
1148 Error was: Exec format error
1150 These can happen when you try to run on AIX 3.2.5 a program that was
1151 compiled with 3.2.4. The fix is to recompile.
1153 *** Trouble using ptys on AIX.
1155 People often install the pty devices on AIX incorrectly.
1156 Use `smit pty' to reinstall them properly.
1160 *** The Emacs window disappears when you type M-q.
1162 Some versions of the Open Look window manager interpret M-q as a quit
1163 command for whatever window you are typing at. If you want to use
1164 Emacs with that window manager, you should try to configure the window
1165 manager to use some other command. You can disable the
1166 shortcut keys entirely by adding this line to ~/.OWdefaults:
1168 OpenWindows.WindowMenuAccelerators: False
1170 *** When Emacs tries to ring the bell, you get an error like
1172 audio: sst_open: SETQSIZE" Invalid argument
1173 audio: sst_close: SETREG MMR2, Invalid argument
1175 you have probably compiled using an ANSI C compiler, but with non-ANSI
1176 include files. In particular, on Suns, the file
1177 /usr/include/sun/audioio.h uses the _IOW macro to define the constant
1178 AUDIOSETQSIZE. _IOW in turn uses a K&R preprocessor feature that is
1179 now explicitly forbidden in ANSI preprocessors, namely substitution
1180 inside character constants. All ANSI C compilers must provide a
1181 workaround for this problem. Lucid's C compiler is shipped with a new
1182 set of system include files. If you are using GCC, there is a script
1183 called fixincludes that creates new versions of some system include
1184 files that use this obsolete feature.
1186 *** On Solaris 2.6, XEmacs dumps core when exiting.
1188 This happens if you're XEmacs is running on the same machine as the X
1189 server, and the optimized memory transport has been turned on by
1190 setting the environment variable XSUNTRANSPORT. The crash occurs
1191 during the call to XCloseDisplay.
1193 If this describes your situation, you need to undefine the
1194 XSUNTRANSPORT environment variable.
1196 *** On Solaris, C-x doesn't get through to Emacs when you use the console.
1198 This is a Solaris feature (at least on Intel x86 cpus). Type C-r
1199 C-r C-t, to toggle whether C-x gets through to Emacs.
1201 *** On Solaris 2.4, Dired hangs and C-g does not work. Or Emacs hangs
1202 forever waiting for termination of a subprocess that is a zombie.
1204 casper@fwi.uva.nl says the problem is in X11R6. Rebuild libX11.so
1205 after changing the file xc/config/cf/sunLib.tmpl. Change the lines
1208 #define SharedX11Reqs -lthread
1213 #if OSMinorVersion < 4
1215 #define SharedX11Reqs -lthread
1219 Be sure also to edit x/config/cf/sun.cf so that OSMinorVersion is 4
1220 (as it should be for Solaris 2.4). The file has three definitions for
1221 OSMinorVersion: the first is for x86, the second for SPARC under
1222 Solaris, and the third for SunOS 4. Make sure to update the
1223 definition for your type of machine and system.
1225 Then do `make Everything' in the top directory of X11R6, to rebuild
1226 the makefiles and rebuild X. The X built this way work only on
1227 Solaris 2.4, not on 2.3.
1229 For multithreaded X to work it necessary to install patch
1230 101925-02 to fix problems in header files [2.4]. You need
1231 to reinstall gcc or re-run just-fixinc after installing that
1234 However, Frank Rust <frust@iti.cs.tu-bs.de> used a simpler solution:
1236 #define ThreadedX YES
1238 #define ThreadedX NO
1239 in sun.cf and did `make World' to rebuild X11R6. Removing all
1240 `-DXTHREAD*' flags and `-lthread' entries from lib/X11/Makefile and
1241 typing 'make install' in that directory also seemed to work.
1243 *** On SunOS 4.1.3, Emacs unpredictably crashes in _yp_dobind_soft.
1245 This happens if you configure Emacs specifying just `sparc-sun-sunos4'
1246 on a system that is version 4.1.3. You must specify the precise
1247 version number (or let configure figure out the configuration, which
1248 it can do perfectly well for SunOS).
1250 *** Mail is lost when sent to local aliases.
1252 Many emacs mail user agents (VM and rmail, for instance) use the
1253 sendmail.el library. This library can arrange for mail to be
1254 delivered by passing messages to the /usr/lib/sendmail (usually)
1255 program . In doing so, it passes the '-t' flag to sendmail, which
1256 means that the name of the recipient of the message is not on the
1257 command line and, therefore, that sendmail must parse the message to
1258 obtain the destination address.
1260 There is a bug in the SunOS4.1.1 and SunOS4.1.3 versions of sendmail.
1261 In short, when given the -t flag, the SunOS sendmail won't recognize
1262 non-local (i.e. NIS) aliases. It has been reported that the Solaris
1263 2.x versions of sendmail do not have this bug. For those using SunOS
1264 4.1, the best fix is to install sendmail V8 or IDA sendmail (which
1265 have other advantages over the regular sendmail as well). At the time
1266 of this writing, these official versions are available:
1268 Sendmail V8 on ftp.cs.berkeley.edu in /ucb/sendmail:
1269 sendmail.8.6.9.base.tar.Z (the base system source & documentation)
1270 sendmail.8.6.9.cf.tar.Z (configuration files)
1271 sendmail.8.6.9.misc.tar.Z (miscellaneous support programs)
1272 sendmail.8.6.9.xdoc.tar.Z (extended documentation, with postscript)
1274 IDA sendmail on vixen.cso.uiuc.edu in /pub:
1275 sendmail-5.67b+IDA-1.5.tar.gz
1277 *** Emacs fails to understand most Internet host names, even though
1278 the names work properly with other programs on the same system.
1279 Emacs won't work with X-windows if the value of DISPLAY is HOSTNAME:0.
1280 Gnus can't make contact with the specified host for nntp.
1282 This typically happens on Suns and other systems that use shared
1283 libraries. The cause is that the site has installed a version of the
1284 shared library which uses a name server--but has not installed a
1285 similar version of the unshared library which Emacs uses.
1287 The result is that most programs, using the shared library, work with
1288 the nameserver, but Emacs does not.
1290 The fix is to install an unshared library that corresponds to what you
1291 installed in the shared library, and then relink Emacs.
1293 On SunOS 4.1, simply define HAVE_RES_INIT.
1295 If you have already installed the name resolver in the file libresolv.a,
1296 then you need to compile Emacs to use that library. The easiest way to
1297 do this is to add to config.h a definition of LIBS_SYSTEM, LIBS_MACHINE
1298 or LIB_STANDARD which uses -lresolv. Watch out! If you redefine a macro
1299 that is already in use in your configuration to supply some other libraries,
1300 be careful not to lose the others.
1302 Thus, you could start by adding this to config.h:
1304 #define LIBS_SYSTEM -lresolv
1306 Then if this gives you an error for redefining a macro, and you see that
1307 the s- file defines LIBS_SYSTEM as -lfoo -lbar, you could change config.h
1310 #define LIBS_SYSTEM -lresolv -lfoo -lbar
1312 *** With process-connection-type set to t, each line of subprocess
1313 output is terminated with a ^M, making ange-ftp and GNUS not work.
1315 On SunOS systems, this problem has been seen to be a result of an
1316 incomplete installation of gcc 2.2 which allowed some non-ANSI
1317 compatible include files into the compilation. In particular this
1318 affected virtually all ioctl() calls.
1323 **** XEmacs warns "Symbol `toggleClassRec' has different size in shared
1324 object, consider re-linking / Symbol `labelClassRec' has different
1325 size in shared object, consider re-linking / Warning: Representation
1326 size 4 must match superclass's to override value"
1328 Sometimes this results in segfaults when using the tab control widget
1329 or a progress bar widget.
1331 Some versions of Debian install 3D versions of the Athena widget
1332 library as /usr/X11R6/lib/libXaw.so. We have not yet solved the
1333 problem of identifying the actual library in use in ./configure, so it
1334 is possible for XEmacs to be compiled with reference to headers for
1335 "flat" Xaw but find a "3D" Xaw when loading.
1337 The straightforward solution is to rebuild XEmacs with additional
1338 configure options: --with-widgets=athena --with-athena=3d.
1340 There are several 3D Athena widget sets available; to see which ones
1341 are supported by XEmacs, use ./configure --usage.
1345 The Mandrake Linux distribution is attempting to comprehensively
1346 update the user interface, and make it consistent across
1347 applications. This is very difficult, and will occasionally cause
1348 conflicts with applications like Emacs with their own long-established
1349 interfaces. Known issues specific to Mandrake or especially common:
1351 Some versions of XEmacs (21.1.9 is known) distributed with Mandrake
1352 were patched to make the Meta and Alt keysyms synonymous. These
1353 normally work as expected in the Mandrake environment. However,
1354 custom-built XEmacsen (including all 21.2 betas) will "inexplicably"
1355 not respect the "Alt-invokes-Meta-commands" convention. See "I want
1356 XEmacs to use the Alt key" below.
1358 The color-gcc wrapper (see below) is in common use on the Mandrake
1361 *** XEmacs configured with ESD crashes with a segmentation violation
1363 This often occurs when a progress bar pops up.
1365 The problem is that the ESD sound daemon manipulates interrupts in a
1366 way that disagrees with XEmacs. The currently available workaround is
1367 not very satisfactory: remove ESD support. Rebuild after
1368 reconfiguring with the option --with-sound=none[,native][,nas]
1370 The funny syntax (requiring the initial "none") is for backward
1371 compatibility, and may change. Native sound and NAS do not cause the
1372 problem, so they may be added to the option to get some sound support.
1374 *** I want XEmacs to use the Alt key, not the XXX key, for Meta commands
1376 For historical reasons, XEmacs looks for a Meta key, then an Alt key.
1377 It binds Meta commands to the X11 modifier bit attached to the first
1378 of these it finds. On PCs, the Windows key is often assigned the Meta
1379 bit, but many desktop environments go to great lengths to get all apps
1380 to use the Alt key, and reserve the Windows key to (sensibly enough)
1383 One correct way to implement this was suggested on comp.emacs.xemacs
1384 (by Kilian Foth and in more detail by Michael Piotrowski): unmap the
1385 Meta modifier using xmodmap or xkb, and then map the Meta/Windows key
1386 to the Super or Hyper keysym and an appropriate mod bit. XEmacs will
1387 not find the Meta keysym, and default to using the Alt key for Meta
1388 keybindings. Typically few applications use the (X11) Meta modifier;
1389 it is tedious but not too much so to teach the ones you need to use
1390 Super instead of Meta. There may be further useful hints in the
1391 discussion of keymapping on non-Linux platforms.
1393 *** The color-gcc wrapper
1395 This wrapper colorizes the error messages from gcc. By default XEmacs
1396 does not interpret the escape sequences used to generate colors,
1397 resulting in a cluttered, hard-to-read buffer. You can remove the
1398 wrapper, or defeat the wrapper colorization in Emacs process buffers
1399 by editing the "nocolor" attribute in /etc/colorgccrc:
1401 $ diff -u /etc/colorgccrc.old /etc/colorgccrc
1402 --- /etc/colorgccrc.old Tue Dec 26 02:17:46 2000
1403 +++ /etc/colorgccrc Tue Dec 26 02:15:48 2000
1406 +nocolor: dumb emacs
1408 If you want colorization in your Emacs buffers, you may get good
1409 results from the ansi-color.el library:
1411 http://www.geocities.com/kensanata/color-emacs.html#ansicolors
1413 This is written for the mainline GNU Emacs but the author has made
1414 efforts to adapt it to XEmacs. YMMV.
1416 *** You get crashes in a non-C locale with Linux GNU Libc 2.0.
1418 Internationalization was not the top priority for GNU Libc 2.0.
1419 As of this writing (1998-12-28) you may get crashes while running
1420 XEmacs in a non-C locale. For example, `LC_ALL=en_US xemacs' crashes
1421 while `LC_ALL=C xemacs' runs fine. This happens for example with GNU
1422 libc 2.0.7. Installing libintl.a and libintl.h built from gettext
1423 0.10.35 and re-building XEmacs solves the crashes. Presumably soon
1424 everyone will upgrade to GNU Libc 2.1 and this problem will go away.
1426 *** `C-z', or `M-x suspend-emacs' hangs instead of suspending.
1428 If you build with `gpm' support on Linux, you cannot suspend XEmacs
1429 because gpm installs a buggy SIGTSTP handler. Either compile with
1430 `--with-gpm=no', or don't suspend XEmacs on the Linux console until
1433 *** With certain fonts, when the cursor appears on a character, the
1434 character doesn't appear--you get a solid box instead.
1436 One user on a Linux system reported that this problem went away with
1437 installation of a new X server. The failing server was XFree86 3.1.1.
1438 XFree86 3.1.2 works.
1440 *** Slow startup on Linux.
1442 People using systems based on the Linux kernel sometimes report that
1443 startup takes 10 to 15 seconds longer than `usual'. There are two
1444 problems, one older, one newer.
1446 **** Old problem: IPv4 host lookup
1448 On older systems, this is because Emacs looks up the host name when it
1449 starts. Normally, this takes negligible time; the extra delay is due
1450 to improper system configuration. (Recent Linux distros usually have
1451 this configuration correct "out of the box".) This problem can occur
1452 for both networked and non-networked machines.
1454 Here is how to fix the configuration. It requires being root.
1456 ***** Networked Case
1458 First, make sure the files `/etc/hosts' and `/etc/host.conf' both
1459 exist. The first line in the `/etc/hosts' file should look like this
1460 (replace HOSTNAME with your host name):
1462 127.0.0.1 localhost HOSTNAME
1464 Also make sure that the `/etc/host.conf' files contains the following
1470 Any changes, permanent and temporary, to the host name should be
1471 indicated in the `/etc/hosts' file, since it acts a limited local
1472 database of addresses and names (e.g., some SLIP connections
1473 dynamically allocate ip addresses).
1475 ***** Non-Networked Case
1477 The solution described in the networked case applies here as well.
1478 However, if you never intend to network your machine, you can use a
1479 simpler solution: create an empty `/etc/host.conf' file. The command
1480 `touch /etc/host.conf' suffices to create the file. The `/etc/hosts'
1481 file is not necessary with this approach.
1483 **** New problem: IPv6 CNAME lookup
1485 A newer problem is due to XEmacs changing to use the modern
1486 getaddrinfo() interface from the older gethostbyname() interface. The
1487 solution above is insufficient, because getaddrinfo() by default tries
1488 to get IPv6 information for localhost. This always involves a dns
1489 lookup to get the CNAME, and the strategies above don't work. It then
1490 falls back to IPv4 behavior.
1492 ***** Robust network case
1494 Configure your network so that there are no nameservers configured
1495 until the network is actually running. getaddrinfo() will not try to
1496 access a nameserver that isn't configured.
1498 ***** Flaky network case
1500 If you have a flaky modem or DSL connection that can be relied on only
1501 to go down whenever you want to bring XEmacs up, you need to force
1502 IPv4 behavior. Explicitly setting DISPLAY=127.0.0.1:0.0 (or whatever
1503 is appropriate) works in most cases.
1505 If you cannot or do not want to do that, you can hard code IPv4
1506 behavior in src/process-unix.c. This is bad[tm], on your own head be
1507 it. In the function unix_canonicalize_host_name() about ten lines
1508 down, change the statement
1510 hints.ai_family = AF_UNSPEC;
1514 hints.ai_family = PF_INET;
1518 getaddrinfo() is also called in src/sysdep.c:init_system_name() and in
1519 src/process-unix.c:unix_open_network_stream(). It should not be
1520 useful to make this change in either of those places.
1524 *** On Irix, I don't see the toolbar icons and I'm getting lots of
1525 entries in the warnings buffer.
1527 SGI ships a really old Xpm library in /usr/lib which does not work at
1528 all well with XEmacs. The solution is to install your own copy of the
1529 latest version of Xpm somewhere and then use the --site-includes and
1530 --site-libraries flags to tell configure where to find it.
1532 *** Trouble using ptys on IRIX, or running out of ptys.
1534 The program mkpts (which may be in `/usr/adm' or `/usr/sbin') needs to
1535 be set-UID to root, or non-root programs like Emacs will not be able
1536 to allocate ptys reliably.
1538 *** Motif dialog boxes lose on Irix.
1540 Larry Auton <lda@control.att.com> writes:
1541 Beware of not specifying
1543 --with-dialogs=athena
1545 if it builds with the motif dialogs [boom!] you're a dead man.
1547 *** Beware of the default image & graphics library on Irix
1549 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
1551 You *have* to compile your own jpeg lib. The one delivered with SGI
1552 systems is a C++ lib, which apparently XEmacs cannot cope with.
1555 ** Digital UNIX/OSF/VMS/Ultrix
1556 *** XEmacs crashes on Digital Unix within font-lock, or when dealing
1557 with large compilation buffers.
1559 The default stack size under Digital Unix is rather small (2M as
1560 opposed to Solaris 8M), hosing the regexp code, which uses alloca()
1561 extensively, overflowing the stack when complex regexps are used.
1564 1) Increase your stack size, using `ulimit -s 8192' or a (t)csh
1567 2) Recompile regex.c with REGEX_MALLOC defined.
1569 *** The `Alt' key doesn't behave as `Meta' when running DECwindows.
1571 The default DEC keyboard mapping has the Alt keys set up to generate the
1572 keysym `Multi_key', which has a meaning to xemacs which is distinct from that
1573 of the `Meta_L' and `Meta-R' keysyms. A second problem is that certain keys
1574 have the Mod2 modifier attached to them for no adequately explored reason.
1575 The correct fix is to pass this file to xmodmap upon starting X:
1578 keysym Multi_key = Alt_L
1582 *** The Compose key on a DEC keyboard does not work as Meta key.
1584 This shell command should fix it:
1586 xmodmap -e 'keycode 0xb1 = Meta_L'
1588 *** `expand-file-name' fails to work on any but the machine you dumped
1591 On Ultrix, if you use any of the functions which look up information
1592 in the passwd database before dumping Emacs (say, by using
1593 expand-file-name in site-init.el), then those functions will not work
1594 in the dumped Emacs on any host but the one Emacs was dumped on.
1596 The solution? Don't use expand-file-name in site-init.el, or in
1597 anything it loads. Yuck - some solution.
1599 I'm not sure why this happens; if you can find out exactly what is
1600 going on, and perhaps find a fix or a workaround, please let us know.
1601 Perhaps the YP functions cache some information, the cache is included
1602 in the dumped Emacs, and is then inaccurate on any other host.
1606 *** I get complaints about the mapping of my HP keyboard at startup,
1607 but I haven't changed anything.
1609 The default HP keymap is set up to have Mod1 assigned to two different keys:
1610 Meta_L and Mode_switch (even though there is not actually a Mode_switch key on
1611 the keyboard -- it uses an "imaginary" keycode.) There actually is a reason
1612 for this, but it's not a good one. The correct fix is to execute this command
1615 xmodmap -e 'remove mod1 = Mode_switch'
1617 *** On HP-UX, you get "poll: Interrupted system call" message in the
1618 window where XEmacs was launched.
1620 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
1622 I get a very strange problem when linking libc.a dynamically: every
1623 event (mouse, keyboard, expose...) results in a "poll: Interrupted
1624 system call" message in the window where XEmacs was
1625 launched. Forcing a static link of libc.a alone by adding
1626 /usr/lib/libc.a at the end of the link line solves this. Note that
1627 my 9.07 build of 19.14b17 and my (old) build of 19.13 both exhibit
1628 the same behavior. I've tried various hpux patches to no avail. If
1629 this problem cannot be solved before the release date, binary kits
1630 for HP *must* be linked statically against libc, otherwise this
1631 problem will show up. (This is directed at whoever will volunteer
1632 for this kit, as I won't be available to do it, unless 19.14 gets
1633 delayed until mid-june ;-). I think this problem will be an FAQ soon
1634 after the release otherwise.
1636 Note: The above entry is probably not valid for XEmacs 21.0 and
1639 *** The right Alt key works wrong on German HP keyboards (and perhaps
1640 other non-English HP keyboards too).
1642 This is because HP-UX defines the modifiers wrong in X. Here is a
1643 shell script to fix the problem; be sure that it is run after VUE
1644 configures the X server.
1646 xmodmap 2> /dev/null - << EOF
1647 keysym Alt_L = Meta_L
1648 keysym Alt_R = Meta_R
1653 keysym Mode_switch = NoSymbol
1655 keysym Meta_R = Mode_switch
1656 add mod2 = Mode_switch
1660 *** XEmacs dumps core at startup when native audio is used. Native
1661 audio does not work with recent versions of HP-UX.
1663 Under HP-UX 10.20 and later (e.g., HP-UX 11.XX), with native audio
1664 enabled, the dumped XEmacs binary ("xemacs") core dumps at startup if
1665 recent versions of the libAlib.sl audio shared library is used. Note
1666 that "temacs" will run, but "xemacs" will dump core. This, of course,
1667 causes the XEmacs build to fail. If GNU malloc is enabled, a stack
1668 trace will show XEmacs to have crashed in the "first" call to malloc().
1670 This bug currently exists in all versions of XEmacs, when the undump
1671 mechanism is used. It is not known if using the experimental portable
1672 dumper will allow native audio to work.
1676 Recent versions of the HP-UX 10.20 (and later) audio shared library (in
1677 /opt/audio/lib), pulls in the libdce shared library, which pulls in a
1678 thread (libcma) library. This prevents the HP-UX undump() routine (in
1679 unexhp9k800.c) from properly working. What's happening is that some
1680 initialization routines are being called in the libcma library, *BEFORE*
1681 main() is called, and these initialization routines are calling
1682 malloc(). Unfortunately, in order for the undumper to work, XEmacs must
1683 adjust (move upwards) the sbrk() value *BEFORE* the first call to
1684 malloc(); if malloc() is called before XEmacs has properly adjusted sbrk
1685 (which is what is happening), dumped memory that is being used by
1686 XEmacs, is improperly re-allocated for use by malloc() and the dumped
1687 memory is corrupted. This causes XEmacs to die an horrible death.
1689 It is believed that versions of the audio library past December 1998
1690 will trigger this problem. Under HP-UX 10.20, you probably have to
1691 install audio library patches to encounter this. It's probable that
1692 recent "fresh, out-of-the-box" HP-UX 11.XX workstations also have this
1693 problem. For HP-UX 10.20, it's believed that audio patch PHSS_17121 (or
1694 a superceeding one, like PHSS_17554, PHSS_17971, PHSS_18777, PHSS_21481,
1695 or PHSS_21662, etc.) will trigger this.
1697 To check if your audio library will cause problems for XEmacs, run
1698 "chatr /opt/audio/lib/libAlib.sl". If "libdce" appears in the displayed
1699 shared library list, XEmacs will probably encounter problems if audio is
1704 Don't enable native audio. Re-run configure without native audio
1707 If your site supports it, try using NAS (Network Audio Support).
1709 Try using the experimental portable dumper. It may work, or it may
1713 *** `Pid xxx killed due to text modification or page I/O error'
1715 On HP-UX, you can get that error when the Emacs executable is on an NFS
1716 file system. HP-UX responds this way if it tries to swap in a page and
1717 does not get a response from the server within a timeout whose default
1718 value is just ten seconds.
1720 If this happens to you, extend the timeout period.
1722 *** Shell mode on HP-UX gives the message, "`tty`: Ambiguous".
1724 christos@theory.tn.cornell.edu says:
1726 The problem is that in your .cshrc you have something that tries to
1727 execute `tty`. If you are not running the shell on a real tty then tty
1728 will print "not a tty". Csh expects one word in some places, but tty
1729 is giving it back 3.
1731 The solution is to add a pair of quotes around `tty` to make it a
1734 if (`tty` == "/dev/console")
1736 should be changed to:
1738 if ("`tty`" == "/dev/console")
1740 Even better, move things that set up terminal sections out of .cshrc
1745 *** Regular expressions matching bugs on SCO systems.
1747 On SCO, there are problems in regexp matching when Emacs is compiled
1748 with the system compiler. The compiler version is "Microsoft C
1749 version 6", SCO 4.2.0h Dev Sys Maintenance Supplement 01/06/93; Quick
1750 C Compiler Version 1.00.46 (Beta). The solution is to compile with
1755 *** Emacs exits with "X protocol error" when run with an X server for
1758 A certain X server for Windows had a bug which caused this.
1759 Supposedly the newer 32-bit version of this server doesn't have the
1764 * Compatibility problems (with Emacs 18, GNU Emacs, or previous XEmacs/lemacs)
1765 ==============================================================================
1767 *** "Symbol's value as variable is void: unread-command-char".
1768 "Wrong type argument: arrayp, #<keymap 143 entries>"
1769 "Wrong type argument: stringp, [#<keypress-event return>]"
1771 There are a few incompatible changes in XEmacs, and these are the
1772 symptoms. Some of the emacs-lisp code you are running needs to be
1773 updated to be compatible with XEmacs.
1775 The code should not treat keymaps as arrays (use `define-key', etc.),
1776 should not use obsolete variables like `unread-command-char' (use
1777 `unread-command-events'). Many (most) of the new ways of doing things
1778 are compatible in GNU Emacs and XEmacs.
1780 Modern Emacs packages (Gnus, VM, W3, efs, etc) are written to support
1781 GNU Emacs and XEmacs. We have provided modified versions of several
1782 popular emacs packages (dired, etc) which are compatible with this
1783 version of emacs. Check to make sure you have not set your load-path
1784 so that your private copies of these packages are being found before
1785 the versions in the lisp directory.
1787 Make sure that your load-path and your $EMACSLOADPATH environment
1788 variable are not pointing at an Emacs18 lisp directory. This will
1791 ** Some packages that worked before now cause the error
1792 Wrong type argument: arrayp, #<face ... >
1794 Code which uses the `face' accessor functions must be recompiled with
1795 xemacs 19.9 or later. The functions whose callers must be recompiled
1796 are: face-font, face-foreground, face-background,
1797 face-background-pixmap, and face-underline-p. The .elc files
1798 generated by version 19.9 will work in 19.6 and 19.8, but older .elc
1799 files which contain calls to these functions will not work in 19.9.
1801 ** Signaling: (error "Byte code stack underflow (byte compiler bug), pc 38")
1803 This error is given when XEmacs 20 is compiled without MULE support
1804 but is attempting to load a .elc which requires MULE support. The fix
1805 is to rebytecompile the offending file.
1807 ** Signaling: (wrong-type-argument ...) when loading mail-abbrevs
1809 The is seen when installing the Insidious Big Brother Data Base (bbdb)
1810 which includes an outdated copy of mail-abbrevs.el. Remove the copy
1811 that comes with bbdb and use the one that comes with XEmacs.
1817 ** A reminder: XEmacs/Mule work does not currently receive *any*
1818 funding, and all work is done by volunteers. If you think you can
1819 help, please contact the XEmacs maintainers.
1821 ** XEmacs/Mule doesn't support TTY's satisfactorily.
1823 This is a major problem, which we plan to address in a future release
1824 of XEmacs. Basically, XEmacs should have primitives to be told
1825 whether the terminal can handle international output, and which
1826 locale. Also, it should be able to do approximations of characters to
1827 the nearest supported by the locale.
1829 ** Internationalized (Asian) Isearch doesn't work.
1831 Currently, Isearch doesn't directly support any of the input methods
1832 that are not XIM based (like egg, canna and quail) (and there are
1833 potential problems with XIM version too...). If you're using egg
1834 there is a workaround. Hitting <RET> right after C-s to invoke
1835 Isearch will put Isearch in string mode, where a complete string can
1836 be typed into the minibuffer and then processed by Isearch afterwards.
1837 Since egg is now supported in the minibuffer using string mode you can
1838 now use egg to input your Japanese, Korean or Chinese string, then hit
1839 return to send that to Isearch and then use standard Isearch commands
1842 ** Using egg and mousing around while in 'fence' mode screws up my
1845 Don't do this. The fence modes of egg and canna are currently very
1846 modal, and messing with where they expect point to be and what they
1847 think is the current buffer is just asking for trouble. If you're
1848 lucky they will realize that something is awry, and simply delete the
1849 fence, but worst case can trash other buffers too. We've tried to
1850 protect against this where we can, but there still are many ways to
1851 shoot yourself in the foot. So just finish what you are typing into
1852 the fence before reaching for the mouse.
1854 ** Not all languages in Quail are supported like Devanagari and Indian
1855 languages, Lao and Tibetan.
1857 Quail requires more work and testing. Although it has been ported to
1858 XEmacs, it works really well for Japanese and for the European
1861 ** Right-to-left mode is not yet implemented, so languages like
1862 Arabic, Hebrew and Thai don't work.
1864 Getting this right requires more work. It may be implemented in a
1865 future XEmacs version, but don't hold your breath. If you know
1866 someone who is ready to implement this, please let us know.
1868 ** We need more developers and native language testers. It's extremely
1869 difficult (and not particularly productive) to address languages that
1870 nobody is using and testing.
1872 ** The kWnn and cWnn support for Chinese and Korean needs developers
1873 and testers. It probably doesn't work.
1875 ** There are no `native XEmacs' TUTORIALs for any Asian languages,
1876 including Japanese. FSF Emacs and XEmacs tutorials are quite similar,
1877 so it should be sufficient to skim through the differences and apply
1878 them to the Japanese version.
1880 ** We only have localized menus translated for Japanese, and the
1881 Japanese menus are developing bitrot (the Mule menu appears in
1884 ** XIM is untested for any language other than Japanese.