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').
25 WATCH OUT for your init file! (~/.xemacs/init.el or ~/.emacs) If
26 you observe strange problems, invoke XEmacs with the `-vanilla'
27 option and see if you can repeat the problem.
29 Note that most of the problems described here manifest at RUN
30 time, even those described as BUILD problems. It is quite unusual
31 for a released XEmacs to fail to build. So a "build problem"
32 requires you to tweak the build environment, then rebuild XEmacs.
33 A "runtime problem" is one that can be fixed by proper
34 configuration of the existing build. Compatibility problems and
35 Mule issues are generally runtime problems, but are treated
36 separately for convenience.
39 * Problems with building XEmacs
40 ===============================
43 Much general information is in INSTALL. If it's covered in
44 INSTALL, we don't repeat it here.
46 *** How does I configure to get the buffer tabs/progress bars?
48 These features depend on support for "native widgets". Use the
49 --with-widgets option to configure. Configuration of widgets is
50 automatic for "modern" toolkits (MS Windows, GTK, and Motif), but if
51 you are using Xt and the Athena widgets, you will probably want to
52 specify a "3d" widget set. See configure --usage, and don't forget to
53 install the corresponding development libraries.
55 *** I know I have libfoo installed, but configure doesn't find it.
57 Typical of Linux systems with package managers. To link with a shared
58 library, you only need the shared library. To compile objects that
59 link with it, you need the headers---and distros don't provide them with
60 the libraries. You need the additional "development" package, too.
62 If you have installed it yourself in a non-system location, you may also
63 need to specify --site-includes, --site-libraries, or --site-prefixes.
64 Note that many packages, especially X11, expect to find includes in a
65 subdirectory. Eg, XEmacs #includes <X11/xpm.h>, not <xpm.h>. See INSTALL.
67 *** Don't use -O2 with gcc 2.8.1 and egcs 1.0 under SPARC architectures
68 without also using `-fno-schedule-insns'.
70 gcc will generate incorrect code otherwise, typically resulting in
71 crashes in the function skip-syntax-backward.
75 There have been reports of egcs-1.1 not compiling XEmacs correctly on
76 Alpha Linux. There have also been reports that egcs-1.0.3a is O.K.
78 *** Don't use -O2 or -O3 with Cygwin 1.0, CodeFusion-99070 or gcc 2.7.2 on x86
79 without also using `-fno-strength-reduce'.
81 gcc will generate incorrect code otherwise. This bug is present in at
82 least 2.6.x and 2.7.[0-2]. This bug has been fixed in GCC 2.7.2.1 and
83 later. This bug is O/S independent, but is limited to x86 architectures.
85 This problem is known to be fixed in egcs (or pgcc) 1.0 or later.
87 Unfortunately, later releases of Cygnus-released compilers (not the
88 Net-released ones) have a bug with the same `problem signature'.
90 If you're lucky, you'll get an error while compiling that looks like:
92 event-stream.c:3189: internal error--unrecognizable insn:
93 (insn 256 14 15 (set (reg/v:SI 24)
94 (minus:SI (reg/v:SI 25)
95 (const_int 2))) -1 (insn_list 11 (nil))
99 If you're unlucky, your code will simply execute incorrectly.
101 *** Don't use gcc-2.95.2 with -mcpu=ultrasparc on Solaris 2.6.
103 gcc will assume a 64-bit operating system, even though you've
104 merely told it to assume a 64-bit instruction set.
106 *** Don't use -O2 with gcc 2.7.2 under Intel architectures without also
107 using `-fno-caller-saves'.
109 gcc will generate incorrect code otherwise. This bug is still
110 present in gcc 2.7.2.3. There have been no reports to indicate the
111 bug is present in egcs 1.0 (or pgcc 1.0) or later. This bug is O/S
112 independent, but limited to x86 architectures.
114 This problem is known to be fixed in egcs (or pgcc) 1.0 or later.
116 *** When using gcc, you get the error message "undefined symbol __fixunsdfsi".
117 When using gcc, you get the error message "undefined symbol __main".
119 This means that you need to link with the gcc library. It may be called
120 "gcc-gnulib" or "libgcc.a"; figure out where it is, and define LIB_GCC in
121 config.h to point to it.
123 It may also work to use the GCC version of `ld' instead of the standard one.
125 *** Excessive optimization with pgcc can break XEmacs
127 It has been reported on some systems that compiling with -O6 can lead
128 to XEmacs failures. The workaround is to use a lower optimization
129 level. -O2 and -O4 have been tested extensively.
131 All of this depends heavily on the version of pgcc and the version
132 of libc. Snapshots near the release of pgcc-1.0 have been tested
133 extensively and no sign of breakage has been seen on systems using
136 *** src/Makefile and lib-src/Makefile are truncated--most of the file missing.
138 This can happen if configure uses GNU sed version 2.03. That version
139 had a bug. GNU sed version 2.05 works properly.
141 *** When compiling with X11, you get "undefined symbol _XtStrings".
143 This means that you are trying to link emacs against the X11r4 version of
144 libXt.a, but you have compiled either Emacs or the code in the lwlib
145 subdirectory with the X11r5 header files. That doesn't work.
147 Remember, you can't compile lwlib for r4 and emacs for r5, or vice versa.
148 They must be in sync.
150 *** test-distrib says that the distribution has been clobbered
151 or, temacs prints "Command key out of range 0-127"
152 or, temacs runs and dumps xemacs, but xemacs totally fails to work.
153 or, temacs gets errors dumping xemacs
155 This can be because the .elc files have been garbled. Do not be
156 fooled by the fact that most of a .elc file is text: these are binary
157 files and can contain all 256 byte values.
159 In particular `shar' cannot be used for transmitting GNU Emacs. It
160 typically truncates "lines". (this does not apply to GNU shar, which
161 uses uuencode to encode binary files.)
163 If you have a copy of Emacs that has been damaged in its nonprinting
164 characters, you can fix them by running:
168 This will rebuild all the needed .elc files.
170 *** `compress' and `uncompress' not found and XFree86
172 XFree86 installs a very old version of libz.a by default ahead of where
173 more modern version of libz might be installed. This will cause problems
174 when attempting to link against libMagick. The fix is to remove the old
175 libz.a in the X11 binary directory.
179 *** On AIX 4.3, you must specify --with-dialogs=athena with configure
181 *** The libXt shipped with AIX 4.3 up to 4.3.2 is broken. This causes
182 xemacs -nw to fail in various ways. The official APAR is this:
184 APAR NUMBER: <IX89470> RESOLVED AS: PROGRAM ERROR
187 <IX89470>: LIBXT.A INCORRECT HANDLING OF EXCEPTIONS IN XTAPPADDINPUT
189 The solution is to install X11.base.lib at version >=4.3.2.5.
191 *** On AIX, you get this compiler error message:
193 Processing include file ./XMenuInt.h
194 1501-106: (S) Include file X11/Xlib.h not found.
196 This means your system was installed with only the X11 runtime i.d
197 libraries. You have to find your sipo (bootable tape) and install
200 *** On AIX 4.1.2, linker error messages such as
201 ld: 0711-212 SEVERE ERROR: Symbol .__quous, found in the global symbol table
202 of archive /usr/lib/libIM.a, was not defined in archive member shr.o.
204 This is a problem in libIM.a. You can work around it by executing
205 these shell commands in the src subdirectory of the directory where
208 cp /usr/lib/libIM.a .
212 Then change -lIM to ./libIM.a in the command to link temacs (in
215 *** Excessive optimization on AIX 4.2 can lead to compiler failure.
217 Valdis.Kletnieks@vt.edu writes:
218 At least at the b34 level, and the latest-and-greatest IBM xlc
219 (3.1.4.4), there are problems with -O3. I haven't investigated
224 *** Crashes when using Motif libraries, especially with multiple frames.
226 Crashes that produce C-backtraces like this:
228 #0 0xfec9a118 in _libc_kill () from /usr/lib/libc.so.1
229 #1 0x77f48 in fatal_error_signal (sig=11)
230 at /codes/rpluim/xemacs-21.4/src/emacs.c:539
231 #2 <signal handler called>
232 #3 0xfee929f4 in XFindContext () from /usr/openwin/lib/libX11.so.4
233 #4 0xfee92930 in XFindContext () from /usr/openwin/lib/libX11.so.4
234 #5 0xff297e54 in DisplayDestroy () from /usr/dt/lib/libXm.so.4
235 #6 0xfefbece0 in XtCallCallbackList () from /usr/openwin/lib/libXt.so.4
236 #7 0xfefc486c in XtPhase2Destroy () from /usr/openwin/lib/libXt.so.4
237 #8 0xfefc45d0 in _XtDoPhase2Destroy () from /usr/openwin/lib/libXt.so.4
238 #9 0xfefc43b4 in XtDestroyWidget () from /usr/openwin/lib/libXt.so.4
239 #10 0x15cf9c in x_delete_device (d=0x523f00)
241 are caused by buggy Motif libraries. Installing the following patches
242 has been reported to solve the problem on Solaris 2.7:
246 For information (although they have not been confirmed to work), the
247 equivalent patches for Solaris 2.8 are:
251 *** Dumping error when using GNU binutils / GNU ld on a Sun.
253 Errors similar to the following:
255 Dumping under the name xemacs unexec():
256 dldump(/space/rpluim/xemacs-obj/src/xemacs): ld.so.1: ./temacs:
257 fatal: /space/rpluim/xemacs-obj/src/xemacs: unknown dynamic entry:
260 are caused by using GNU ld. There are several workarounds available:
262 In XEmacs 21.2 or later, configure using the new portable dumper
265 Alternatively, you can link using the Sun version of ld, which is
266 normally held in /usr/ccs/bin. This can be done by one of:
268 - building gcc with these configure flags:
269 configure --with-ld=/usr/ccs/bin/ld --with-as=/usr/ccs/bin/as
271 - adding -B/usr/ccs/bin/ to CFLAGS used to configure XEmacs
272 (Note: The trailing '/' there is significant.)
274 - uninstalling GNU ld.
276 - configuring XEmacs with these environment settings (suggested by
277 Goran Koruga <goran.koruga@hermes.si>):
278 LD=/usr/ccs/bin/ld AS=/usr/ccs/bin/as configure
280 The Solaris2 FAQ claims:
282 When you install gcc, don't make the mistake of installing
283 GNU binutils or GNU libc, they are not as capable as their
284 counterparts you get with Solaris 2.x.
286 *** Link failure when using acc on a Sun.
288 To use acc, you need additional options just before the libraries, such as
290 /usr/lang/SC2.0.1/values-Xt.o -L/usr/lang/SC2.0.1/cg87 -L/usr/lang/SC2.0.1
292 and you need to add -lansi just before -lc.
294 The precise file names depend on the compiler version, so we
295 cannot easily arrange to supply them.
297 *** Problems finding X11 libraries on Solaris with Openwindows
299 Some users have reported problems in this area. The reported solution
300 is to define the environment variable OPENWINHOME, even if you must set
301 it to `/usr/openwin'.
303 *** Sed problems on Solaris 2.5
305 There have been reports of Sun sed truncating very lines in the
306 Makefile during configuration. The workaround is to use GNU sed or,
307 even better, think of a better way to generate Makefile, and send us a
310 *** On Solaris 2 I get undefined symbols from libcurses.a.
312 You probably have /usr/ucblib/ on your LD_LIBRARY_PATH. Do the link with
313 LD_LIBRARY_PATH unset. Generally, avoid using any ucb* stuff when
316 *** On Solaris 2 I cannot make alloc.o, glyphs.o or process.o.
318 The SparcWorks C compiler may have difficulty building those modules
319 with optimization level -xO4. Try using only "-fast" optimization
320 for just those modules. (Or use gcc).
322 *** Solaris 2.3 /bin/sh coredumps during configuration.
324 This only occurs if you have LANG != C. This is a known bug with
325 /bin/sh fixed by installing Patch-ID# 101613-01. Or, you can use
326 bash, as a workaround.
328 *** On SunOS, you get linker errors
330 _get_wmShellWidgetClass
331 _get_applicationShellWidgetClass
333 The fix to this is to install patch 100573 for OpenWindows 3.0
334 or link libXmu statically.
336 *** On Sunos 4, you get the error ld: Undefined symbol __lib_version.
338 This is the result of using cc or gcc with the shared library meant
339 for acc (the Sunpro compiler). Check your LD_LIBRARY_PATH and delete
340 /usr/lang/SC2.0.1 or some similar directory.
342 *** Undefined symbols when linking on Sunos 4.1.
344 If you get the undefined symbols _atowc _wcslen, _iswprint, _iswspace,
345 _iswcntrl, _wcscpy, and _wcsncpy, then you need to add -lXwchar after
346 -lXaw in the command that links temacs.
348 This problem seems to arise only when the international language
349 extensions to X11R5 are installed.
351 *** On a Sun running SunOS 4.1.1, you get this error message from GNU ld:
353 /lib/libc.a(_Q_sub.o): Undefined symbol __Q_get_rp_rd referenced from text segment
355 The problem is in the Sun shared C library, not in GNU ld.
357 The solution is to install Patch-ID# 100267-03 from Sun.
359 *** SunOS 4.1.2: undefined symbol _get_wmShellWidgetClass
361 Apparently the version of libXmu.so.a that Sun ships is hosed: it's missing
362 some stuff that is in libXmu.a (the static version). Sun has a patch for
363 this, but a workaround is to use the static version of libXmu, by changing
364 the link command from "-lXmu" to "-Bstatic -lXmu -Bdynamic". If you have
365 OpenWindows 3.0, ask Sun for these patches:
366 100512-02 4.1.x OpenWindows 3.0 libXt Jumbo patch
367 100573-03 4.1.x OpenWindows 3.0 undefined symbols with shared libXmu
369 *** Random other SunOS 4.1.[12] link errors.
371 The X headers and libraries that Sun ships in /usr/{include,lib}/X11 are
372 broken. Use the ones in /usr/openwin/{include,lib} instead.
375 *** Under Linux, you get "too many arguments to function `getpgrp'".
377 You have probably installed LessTiff under `/usr/local' and `libXm.so'
378 could not be found when linking `getpgrp()' test program, making XEmacs
379 think that `getpgrp()' takes an argument. Try adding `/usr/local/lib'
380 in `/etc/ld.so.conf' and run `ldconfig'. Then run XEmacs's `configure'
381 again. As with all problems of this type, reading the config.log file
382 generated from configure and seeing the log of how the test failed can
385 *** `Error: No ExtNode to pop!' on Linux systems with Lesstif.
387 This error message has been observed with lesstif-0.75a. It does not
388 appear to cause any harm.
390 *** xemacs: can't resolve symbol '__malloc_hook'
392 This is a Linux problem where you've compiled the XEmacs binary on a libc
393 5.4 with version higher than 5.4.19 and attempted to run the binary against
394 an earlier version. The solution is to upgrade your old library.
398 *** On Irix 6.5, the MIPSpro compiler gets an internal compiler error
400 The MIPSpro Compiler (at least version 7.2.1) can't seem to handle the
401 union type properly, and fails to compile src/glyphs.c. To avoid this
402 problem, always build ---use-union-type=no (but that's the default, so
403 you should only see this problem if you're an XEmacs maintainer).
405 *** Linking with -rpath on IRIX.
407 Darrell Kindred <dkindred@cmu.edu> writes:
408 There are a couple of problems [with use of -rpath with Irix ld], though:
410 1. The ld in IRIX 5.3 ignores all but the last -rpath
411 spec, so the patched configure spits out a warning
412 if --x-libraries or --site-runtime-libraries are
413 specified under irix 5.x, and it only adds -rpath
414 entries for the --site-runtime-libraries. This bug was
415 fixed sometime between 5.3 and 6.2.
417 2. IRIX gcc 2.7.2 doesn't accept -rpath directly, so
418 it would have to be prefixed by -Xlinker or "-Wl,".
419 This would be fine, except that configure compiles with
420 ${CC-cc} $CFLAGS $LDFLAGS ...
421 rather than quoting $LDFLAGS with prefix-args, like
422 src/Makefile does. So if you specify --x-libraries
423 or --site-runtime-libraries, you must use --use-gcc=no,
424 or configure will fail.
426 *** On Irix 6.3, the SGI ld quits with segmentation fault when linking temacs
428 This occurs if you use the SGI linker version 7.1. Installing the
429 patch SG0001872 fixes this problem.
431 *** On Irix 6.0, make tries (and fails) to build a program named unexelfsgi
433 A compiler bug inserts spaces into the string "unexelfsgi . o"
434 in src/Makefile. Edit src/Makefile, after configure is run,
435 find that string, and take out the spaces.
437 Compiler fixes in Irix 6.0.1 should eliminate this problem.
439 *** On Irix 5.2, unexelfsgi.c can't find cmplrs/stsupport.h.
441 The file cmplrs/stsupport.h was included in the wrong file set in the
442 Irix 5.2 distribution. You can find it in the optional fileset
443 compiler_dev, or copy it from some other Irix 5.2 system. A kludgy
444 workaround is to change unexelfsgi.c to include sym.h instead of
447 *** Coredumping in Irix 6.2
449 Pete Forman <gsez020@compo.bedford.waii.com> writes:
450 A problem noted by myself and others (I've lost the references) was
451 that XEmacs coredumped when the cut or copy toolbar buttons were
452 pressed. This has been fixed by loading the SGI patchset (Feb 98)
453 without having to recompile XEmacs.
455 My versions are XEmacs 20.3 (problem first noted in 19.15) and IRIX
456 6.2, compiled using -n32. I'd guess that the relevant individual
457 patch was "SG0002580: multiple fixes for X libraries". SGI recommends
458 that the complete patch set be installed rather than parts of it.
460 ** Digital UNIX/OSF/VMS
461 *** On Digital UNIX, the DEC C compiler might have a problem compiling
464 In particular, src/extents.c and src/faces.c might cause the DEC C
465 compiler to abort. When this happens: cd src, compile the files by
466 hand, cd .., and redo the "make" command. When recompiling the files by
467 hand, use the old C compiler for the following versions of Digital UNIX:
468 - V3.n: Remove "-migrate" from the compile command.
469 - V4.n: Add "-oldc" to the compile command.
471 A related compiler bug has been fixed by the DEC compiler team. The
472 new versions of the compiler should run fine.
474 *** Under some versions of OSF XEmacs runs fine if built without
475 optimization but will crash randomly if built with optimization.
477 Using 'cc -g' is not sufficient to eliminate all optimization. Try
480 *** Compilation errors on VMS.
482 Sorry, XEmacs does not work under VMS. You might consider working on
483 the port if you really want to have XEmacs work under VMS.
486 *** On HPUX, the HP C compiler might have a problem compiling some files
489 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
491 Had to drop once again to level 2 optimization, at least to
492 compile lstream.c. Otherwise, I get a "variable is void: \if"
493 problem while dumping (this is a problem I already reported
494 with vanilla hpux 10.01 and 9.07, which went away after
495 applying patches for the C compiler). Trouble is I still
496 haven't found the same patch for hpux 10.10, and I don't
497 remember the patch numbers. I think potential XEmacs builders
498 on HP should be warned about this.
500 *** I don't have `xmkmf' and `imake' on my HP.
502 You can get these standard X tools by anonymous FTP to
503 hpcvaaz.cv.hp.com. Essentially all X programs need these.
505 *** On HP-UX, problems with make
507 Marcus Thiessel <marcus@xemacs.org>
509 Some releases of XEmacs (e.g. 20.4) require GNU make to build
510 successfully. You don't need GNU make when building 21.x.
512 *** On HP-UX 9.05 XEmacs won't compile or coredump during the build.
514 Marcus Thiessel <marcus@xemacs.org>
516 This might be a sed problem. For your own safety make sure to use
517 GNU sed while dumping XEmacs.
519 *** On HP-UX 11.0 XEmacs causes excessive X11 errors when running.
520 (also appears on AIX as reported in comp.emacs.xemacs)
522 Marcus Thiessel <marcus@xemacs.org>
524 Unfortunately, XEmacs releases prior to 21.0 don't work with
525 Motif2.1. It will compile but you will get excessive X11 errors like
527 xemacs: X Error of failed request: BadGC (invalid GC parameter)
529 and finally XEmacs gets killed. A workaround is to use the
530 Motif1.2_R6 libraries. You can the following line to your call to
533 --x-libraries="/usr/lib/Motif1.2_R6 -L/usr/lib/X11R6"
535 Make sure /usr/lib/Motif1.2_R6/libXm.sl is a link to
536 /usr/lib/Motif1.2_R6/libXm.3.
538 *** On HP-UX 11.0: Object "" does not have windowed ancestor
540 Marcus Thiessel <marcus@xemacs.org>
542 XEmacs dies without core file and reports:
544 Error: Object "" does not have windowed ancestor.
546 This is a bug. Please apply the patch PHSS_19964 (check if
547 superseded). The other alternative is to link with Motif1.2_R6 (see
552 *** Native cc on SCO OpenServer 5 is now OK. Icc may still throw you
553 a curve. Here is what Robert Lipe <robertl@arnet.com> says:
555 Unlike XEmacs 19.13, building with the native cc on SCO OpenServer 5
556 now produces a functional binary. I will typically build this
557 configuration for COFF with:
559 /path_to_xemacs_source/configure --with-gcc=no \
560 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
561 --with-xpm --with-xface --with-sound=nas
563 This version now supports ELF builds. I highly recommend this to
564 reduce the in-core footprint of XEmacs. This is now how I compile
565 all my test releases. Build it like this:
567 /path_to_XEmacs_source/configure --with-gcc=no \
568 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
569 --with-xpm --with-xface --with-sound=nas --dynamic
571 The compiler known as icc [ supplied with the OpenServer 5 Development
572 System ] generates a working binary, but it takes forever to generate
573 XEmacs. ICC also whines more about the code than /bin/cc does. I do
574 believe all its whining is legitimate, however. Note that you do
575 have to 'cd src ; make LD=icc' to avoid linker errors.
577 The way I handle the build procedure is:
579 /path_to_XEmacs_source/configure --with-gcc=no \
580 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
581 --with-xpm --with-xface --with-sound=nas --dynamic --compiler="icc"
583 NOTE I have the xpm, xface, and audio libraries and includes in
584 /usr/local/lib, /usr/local/include. If you don't have these,
585 don't include the "--with-*" arguments in any of my examples.
587 In previous versions of XEmacs, you had to override the defaults while
588 compiling font-lock.o and extents.o when building with icc. This seems
589 to no longer be true, but I'm including this old information in case it
590 resurfaces. The process I used was:
593 [ procure pizza, beer, repeat ]
595 make CC="icc -W0,-mP1COPT_max_tree_size=3000" font-lock.o extents.o
598 If you want sound support, get the tls566 supplement from
599 ftp.sco.com:/TLS or any of its mirrors. It works just groovy
602 The M-x manual-entry is known not to work. If you know Lisp and would
603 like help in making it work, e-mail me at <robertl@dgii.com>.
604 (UNCHECKED for 19.15 -- it might work).
606 In earlier releases, gnuserv/gnuclient/gnudoit would open a frame
607 just fine, but the client would lock up and the server would
608 terminate when you used C-x # to close the frame. This is now
611 In etc/ there are two files of note. emacskeys.sco and emacsstrs.sco.
612 The comments at the top of emacskeys.sco describe its function, and
613 the emacstrs.sco is a suitable candidate for /usr/lib/keyboard/strings
614 to take advantage of the keyboard map in emacskeys.sco.
616 Note: Much of the above entry is probably not valid for XEmacs 21.0
621 *** In general use etc/check_cygwin_setup.sh to trap environment problems.
623 The script etc/check_cygwin_setup.sh will attempt to detect whether
624 you have a suitable environment for building. This script may not work
625 correctly if you are using ash instead of bash (see below).
627 *** Syntax errors running configure scripts, make failing with exit code 127
628 in inexplicable situations, etc.
630 This may be because you are using the default cygwin shell. The
631 default cygwin shell (/bin/sh.exe) is ash which appears to work in
632 most circumstances but has some weird failure modes. You need to
633 replace the symlink with bash.exe.
635 *** Lots of compile errors, esp. on lines containing macro definitions
636 terminated by backslashes.
638 Your partition holding the source files is mounted binary. It needs
639 to be mounted text. (This will not screw up any binary files because
640 the Cygwin utilities specify explicitly whether they want binary or
641 text mode when working with source vs. binary files, which overrides
642 the mount type.) To fix this, you just need to run the appropriate
643 mount command once -- afterwards, the settings are remembered in the
646 *** Errors from make like /c:not found.
648 make sure you set the environment variable MAKE_MODE to UNIX in your
649 .bashrc, Control Panel (Windows 2000/NT), or AUTOEXEC.BAT (Windows
652 *** X11 not detected.
654 This is usually because xmkmf is not in your path or because you are
655 using the default cygwin shell. The default cygwin shell (/bin/sh.exe)
656 is ash which appears to work in most circumstances but has some weird
657 failure modes. I recommend replacing sh.exe with bash.exe, this will
658 mean configure is slower but more reliable.
660 *** Subprocesses do not work.
662 You do not have "tty" in your CYGWIN environment variable. This must
663 be set in your autoexec.bat (win95) or the system properties (winnt)
664 as it must be read before the cygwin DLL initializes.
666 *** ^G does not work on hung subprocesses.
668 This is a known problem. It can be remedied with cygwin b20 or greater
669 by defining BROKEN_SIGIO in src/s/cygwin32.h, however this currently
670 leads to instability in XEmacs.
672 *** The info files will not build.
674 makeinfo that ships with cygwin (all versions) is a noop. You need to
675 obtain makeinfo from somewhere or build it yourself.
677 *** I have no graphics.
679 You need to obtain the various graphics libraries. Pre-built versions
680 of these and the X libraries are located on the XEmacs website in
681 ftp://ftp.xemacs.org/pub/aux/cygwin/.
683 *** There are no images in the toolbar buttons.
685 You need version 4.71 of commctrl.dll which does not ship with windows
686 95. You can get this by installing IE 4.0 or downloading it from the
690 * Problems with running XEmacs
691 ==============================
694 *** XEmacs crashes mysteriously.
696 Check whether XEmacs was configured --use-union-type. Many compilers
697 are known to treat union types incompatibly with proper functioning of
698 the Lisp_Object type. (Whether this is a compiler bug or nonstandard-
699 conforming code in XEmacs is a moot point.) Especially with
700 --with-mule, --pdump, and/or non-null --error-checking, this is known
701 to produce an unreliable build with many versions of MS VC++ and GCC,
702 and similar problems are likely to occur with other compilers.
704 Symptoms are similar to garbage collection and other "wild pointer"
705 bugs, ie, stack-smashing and other hard-to-debug crashes in unrelated
706 code. Try reconfiguring and building without --use-union-type.
708 --use-union-type _is_ useful to get improved _static_ type checking of
709 Lisp objects. It is theoretically possible that it might help with
710 aliasing bugs under optimization and improve runtime stability, but in
711 practice exactly the opposite seems to be true. If you don't work on
712 XEmacs C code directly, then avoid --use-union-type entirely for now.
714 *** XEmacs crashes mysteriously in regexp-intensive applications (eg, Gnus)
716 The regexp implementation used in XEmacs uses alloca by default for
717 efficiency. alloca provides no reliable way to check for out of
718 memory (in this case, stack). Normally not a problem, except for
719 systems with very small default stack allocations, and applications
720 that use multi-line regular expressions (ie, explicitly including ?\n)
721 in moderately large files (> 100kB or so).
723 You may get relief by increasing the amount of stack space allocated
724 to your XEmacs process (a system-dependent operation, ask your
725 administrator or local experts for help), or by recompiling the regexp
726 module regex.c with REGEX_MALLOC defined, relinking, and redumping.
728 *** At startup I get a warning on stderr about missing charsets:
730 Warning: Missing charsets in String to FontSet conversion
732 You need to specify appropriate charsets for your locale (usually the
733 value of the LANG environment variable) in .Xresources. See
734 etc/Emacs.ad for the relevant resources (mostly menubar fonts and
735 fontsets). Do not edit this file, it's purely informative.
737 If you have no satisfactory fonts for iso-8859-1, XEmacs will crash.
739 It looks like XFree86 4.x (the usual server on Linux and *BSD) has
740 some braindamage where .UTF-8 locales will always generate this
741 message, because the XFree86 (font)server doesn't know that UTF-8 will
742 use the ISO10646-1 font registry (or a Cmap or something).
744 If you are not using a .UTF-8 locale and see this warning for a
745 character set not listed in the default in Emacs.ad, please let
746 xemacs-beta@xemacs.org know about it, so we can add fonts to the
747 appropriate fontsets and stifle this warning. (Unfortunately it's
748 buried in Xlib, so we can't easily get rid of it otherwise.)
750 *** Whenever I try to retrieve a remote file, I have problems.
752 A typical error: FTP Error: USER request failed; 500 AUTH not understood.
753 Thanks to giacomo boffi <giacomo.boffi@polimi.it> on comp.emacs.xemacs:
755 tell your ftp client to not attempt AUTH authentication (or do not
756 use FTP servers that don't understand AUTH)
758 and notes that you need to add an element (often "-u") to
759 `efs-ftp-program-args'. Use M-x customize-variable, and verify the
760 needed flag with `man ftp' or other local documentation.
762 *** gnuserv is running, some clients can connect, but others cannot.
764 The code in gnuslib.c respects the value of TMPDIR. If the server and
765 the client have different values in their environment, you lose.
766 One program known to set TMPDIR and manifest this problem is exmh.
767 You can defeat the use of TMPDIR by unsetting USE_TMPDIR at the top of
768 gnuserv.h at build time.
770 *** C-z just refreshes the screen instead of suspending Emacs.
772 You are probably using a shell that doesn't support job control, even
773 though the system itself is capable of it. Try using a different
776 *** You type Control-H (Backspace) expecting to delete characters.
778 Emacs has traditionally used Control-H for help; unfortunately this
779 interferes with its use as Backspace on TTY's. As of XEmacs 21,
780 XEmacs looks at the "erase" setting of TTY structures and maps C-h to
781 backspace when erase is set to C-h. This is sort of a special hack,
782 but it makes it possible for you to use the standard:
786 to get your backspace key to erase characters. The erase setting is
787 recorded in the Lisp variable `tty-erase-char', which you can use to
788 tune the settings in your .emacs.
790 A major drawback of this is that when C-h becomes backspace, it no
791 longer invokes help. In that case, you need to use f1 for help, or
792 bind another key. An example of the latter is the following code,
793 which moves help to Meta-? (ESC ?):
795 (global-set-key "\M-?" 'help-command)
797 *** Mail agents (VM, Gnus, rmail) cannot get new mail
799 rmail and VM get new mail from /usr/spool/mail/$USER using a program
800 called `movemail'. This program interlocks with /bin/mail using the
801 protocol defined by /bin/mail.
803 There are two different protocols in general use. One of them uses
804 the `flock' system call. The other involves creating a lock file;
805 `movemail' must be able to write in /usr/spool/mail in order to do
806 this. You control which one is used by defining, or not defining, the
807 macro MAIL_USE_FLOCK in config.h or the m- or s- file it includes. IF
808 YOU DON'T USE THE FORM OF INTERLOCKING THAT IS NORMAL ON YOUR SYSTEM,
811 If your system uses the lock file protocol, and fascist restrictions
812 prevent ordinary users from writing the lock files in /usr/spool/mail,
813 you may need to make `movemail' setgid to a suitable group such as
814 `mail'. To do this, use the following commands (as root) after doing
820 Installation normally copies movemail from the build directory to an
821 installation directory which is usually under /usr/local/lib. The
822 installed copy of movemail is usually in the directory
823 /usr/local/lib/emacs/VERSION/TARGET. You must change the group and
824 mode of the installed copy; changing the group and mode of the build
825 directory copy is ineffective.
827 *** VM appears to hang in large folders.
829 This is normal (trust us) when upgrading to VM-6.22 from earlier
830 versions. Let VM finish what it is doing and all will be well.
832 *** Changes made to .el files do not take effect.
834 You may have forgotten to recompile them into .elc files. Then the
835 old .elc files will be loaded, and your changes will not be seen. To
836 fix this, do `M-x byte-recompile-directory' and specify the directory
837 that contains the Lisp files.
839 Note that you will get a warning when loading a .elc file that is
840 older than the corresponding .el file.
842 *** Things which should be bold or italic (such as the initial
843 copyright notice) are not.
845 The fonts of the "bold" and "italic" faces are generated from the font
846 of the "default" face; in this way, your bold and italic fonts will
847 have the appropriate size and family. However, emacs can only be
848 clever in this way if you have specified the default font using the
849 XLFD (X Logical Font Description) format, which looks like
851 *-courier-medium-r-*-*-*-120-*-*-*-*-*-*
853 if you use any of the other, less strict font name formats, some of
856 lucidasanstypewriter-12
860 then emacs won't be able to guess the names of the "bold" and "italic"
861 versions. All X fonts can be referred to via XLFD-style names, so you
862 should use those forms. See the man pages for X(1), xlsfonts(1), and
865 *** The dumped Emacs crashes when run, trying to write pure data.
867 Two causes have been seen for such problems.
869 1) On a system where getpagesize is not a system call, it is defined
870 as a macro. If the definition (in both unexec.c and malloc.c) is wrong,
871 it can cause problems like this. You might be able to find the correct
872 value in the man page for a.out (5).
874 2) Some systems allocate variables declared static among the
875 initialized variables. Emacs makes all initialized variables in most
876 of its files pure after dumping, but the variables declared static and
877 not initialized are not supposed to be pure. On these systems you
878 may need to add "#define static" to the m- or the s- file.
880 *** Reading and writing files is very very slow.
882 Try evaluating the form (setq lock-directory nil) and see if that helps.
883 There is a problem with file-locking on some systems (possibly related
884 to NFS) that I don't understand. Please send mail to the address
885 xemacs-beta@xemacs.org if you figure this one out.
887 *** When emacs starts up, I get lots of warnings about unknown keysyms.
889 If you are running the prebuilt binaries, the Motif library expects to find
890 certain thing in the XKeysymDB file. This file is normally in /usr/lib/X11/
891 or in /usr/openwin/lib/. If you keep yours in a different place, set the
892 environment variable $XKEYSYMDB to point to it before starting emacs. If
893 you still have the problem after doing that, perhaps your version of X is
894 too old. There is a copy of the MIT X11R5 XKeysymDB file in the emacs `etc'
895 directory. Try using that one.
897 *** My X resources used to work, and now some of them are being ignored.
899 Check the resources in .../etc/Emacs.ad (which is the same as the file
900 sample.Xdefaults). Perhaps some of the default resources built in to
901 emacs are now overriding your existing resources. Copy and edit the
902 resources in Emacs.ad as necessary.
904 *** I have focus problems when I use `M-o' to switch to another screen
905 without using the mouse.
907 The focus issues with a program like XEmacs, which has multiple
908 homogeneous top-level windows, are very complicated, and as a result,
909 most window managers don't implement them correctly.
911 The R4/R5 version of twm (and all of its descendants) had buggy focus
912 handling. Sufficiently recent versions of tvtwm have been fixed. In
913 addition, if you're using twm, make sure you have not specified
914 "NoTitleFocus" in your .tvtwmrc file. The very nature of this option
915 makes twm do some illegal focus tricks, even with the patch.
917 It is known that olwm and olvwm are buggy, and in different ways. If
918 you're using click-to-type mode, try using point-to-type, or vice
921 In older versions of NCDwm, one could not even type at XEmacs windows.
922 This has been fixed in newer versions (2.4.3, and possibly earlier).
924 (Many people suggest that XEmacs should warp the mouse when focusing
925 on another screen in point-to-type mode. This is not ICCCM-compliant
926 behavior. Implementing such policy is the responsibility of the
927 window manager itself, it is not legal for a client to do this.)
929 *** Emacs spontaneously displays "I-search: " at the bottom of the screen.
931 This means that Control-S/Control-Q (XON/XOFF) "flow control" is being
932 used. C-s/C-q flow control is bad for Emacs editors because it takes
933 away C-s and C-q as user commands. Since editors do not output long
934 streams of text without user commands, there is no need for a
935 user-issuable "stop output" command in an editor; therefore, a
936 properly designed flow control mechanism would transmit all possible
937 input characters without interference. Designing such a mechanism is
938 easy, for a person with at least half a brain.
940 There are three possible reasons why flow control could be taking place:
942 1) Terminal has not been told to disable flow control
943 2) Insufficient padding for the terminal in use
944 3) Some sort of terminal concentrator or line switch is responsible
946 First of all, many terminals have a set-up mode which controls whether
947 they generate XON/XOFF flow control characters. This must be set to
948 "no XON/XOFF" in order for Emacs to work. Sometimes there is an
949 escape sequence that the computer can send to turn flow control off
950 and on. If so, perhaps the termcap `ti' string should turn flow
951 control off, and the `te' string should turn it on.
953 Once the terminal has been told "no flow control", you may find it
954 needs more padding. The amount of padding Emacs sends is controlled
955 by the termcap entry for the terminal in use, and by the output baud
956 rate as known by the kernel. The shell command `stty' will print
957 your output baud rate; `stty' with suitable arguments will set it if
958 it is wrong. Setting to a higher speed causes increased padding. If
959 the results are wrong for the correct speed, there is probably a
960 problem in the termcap entry. You must speak to a local Unix wizard
961 to fix this. Perhaps you are just using the wrong terminal type.
963 For terminals that lack a "no flow control" mode, sometimes just
964 giving lots of padding will prevent actual generation of flow control
965 codes. You might as well try it.
967 If you are really unlucky, your terminal is connected to the computer
968 through a concentrator which sends XON/XOFF flow control to the
969 computer, or it insists on sending flow control itself no matter how
970 much padding you give it. Unless you can figure out how to turn flow
971 control off on this concentrator (again, refer to your local wizard),
972 you are screwed! You should have the terminal or concentrator
973 replaced with a properly designed one. In the mean time, some drastic
974 measures can make Emacs semi-work.
976 You can make Emacs ignore C-s and C-q and let the operating system
977 handle them. To do this on a per-session basis, just type M-x
978 enable-flow-control RET. You will see a message that C-\ and C-^ are
979 now translated to C-s and C-q. (Use the same command M-x
980 enable-flow-control to turn *off* this special mode. It toggles flow
983 If C-\ and C-^ are inconvenient for you (for example, if one of them
984 is the escape character of your terminal concentrator), you can choose
985 other characters by setting the variables flow-control-c-s-replacement
986 and flow-control-c-q-replacement. But choose carefully, since all
987 other control characters are already used by emacs.
989 IMPORTANT: if you type C-s by accident while flow control is enabled,
990 Emacs output will freeze, and you will have to remember to type C-q in
993 If you work in an environment where a majority of terminals of a
994 certain type are flow control hobbled, you can use the function
995 `enable-flow-control-on' to turn on this flow control avoidance scheme
996 automatically. Here is an example:
998 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
1000 If this isn't quite correct (e.g. you have a mixture of flow-control hobbled
1001 and good vt200 terminals), you can still run enable-flow-control
1004 I have no intention of ever redesigning the Emacs command set for the
1005 assumption that terminals use C-s/C-q flow control. XON/XOFF flow
1006 control technique is a bad design, and terminals that need it are bad
1007 merchandise and should not be purchased. Now that X is becoming
1008 widespread, XON/XOFF seems to be on the way out. If you can get some
1009 use out of GNU Emacs on inferior terminals, more power to you, but I
1010 will not make Emacs worse for properly designed systems for the sake
1011 of inferior systems.
1013 *** Control-S and Control-Q commands are ignored completely.
1015 For some reason, your system is using brain-damaged C-s/C-q flow
1016 control despite Emacs's attempts to turn it off. Perhaps your
1017 terminal is connected to the computer through a concentrator
1018 that wants to use flow control.
1020 You should first try to tell the concentrator not to use flow control.
1021 If you succeed in this, try making the terminal work without
1022 flow control, as described in the preceding section.
1024 If that line of approach is not successful, map some other characters
1025 into C-s and C-q using keyboard-translate-table. The example above
1026 shows how to do this with C-^ and C-\.
1028 *** Control-S and Control-Q commands are ignored completely on a net
1031 Some versions of rlogin (and possibly telnet) do not pass flow
1032 control characters to the remote system to which they connect.
1033 On such systems, emacs on the remote system cannot disable flow
1034 control on the local system.
1036 One way to cure this is to disable flow control on the local host
1037 (the one running rlogin, not the one running rlogind) using the
1038 stty command, before starting the rlogin process. On many systems,
1039 `stty start u stop u' will do this.
1041 Some versions of tcsh will prevent even this from working. One way
1042 around this is to start another shell before starting rlogin, and
1043 issue the stty command to disable flow control from that shell.
1045 If none of these methods work, the best solution is to type
1046 `M-x enable-flow-control' at the beginning of your emacs session, or
1047 if you expect the problem to continue, add a line such as the
1048 following to your .emacs (on the host running rlogind):
1050 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
1052 See the entry about spontaneous display of I-search (above) for more
1055 *** TTY redisplay is slow.
1057 XEmacs has fairly new TTY redisplay support (beginning from 19.12),
1058 which doesn't include some basic TTY optimizations -- like using
1059 scrolling regions to move around blocks of text. This is why
1060 redisplay on the traditional terminals, or over slow lines can be very
1063 If you are interested in fixing this, please let us know at
1064 <xemacs-beta@xemacs.org>.
1066 *** Screen is updated wrong, but only on one kind of terminal.
1068 This could mean that the termcap entry you are using for that terminal
1069 is wrong, or it could mean that Emacs has a bug handing the
1070 combination of features specified for that terminal.
1072 The first step in tracking this down is to record what characters
1073 Emacs is sending to the terminal. Execute the Lisp expression
1074 (open-termscript "./emacs-script") to make Emacs write all terminal
1075 output into the file ~/emacs-script as well; then do what makes the
1076 screen update wrong, and look at the file and decode the characters
1077 using the manual for the terminal. There are several possibilities:
1079 1) The characters sent are correct, according to the terminal manual.
1081 In this case, there is no obvious bug in Emacs, and most likely you
1082 need more padding, or possibly the terminal manual is wrong.
1084 2) The characters sent are incorrect, due to an obscure aspect of the
1085 terminal behavior not described in an obvious way by termcap.
1087 This case is hard. It will be necessary to think of a way for Emacs
1088 to distinguish between terminals with this kind of behavior and other
1089 terminals that behave subtly differently but are classified the same
1090 by termcap; or else find an algorithm for Emacs to use that avoids the
1091 difference. Such changes must be tested on many kinds of terminals.
1093 3) The termcap entry is wrong.
1095 See the file etc/TERMS for information on changes that are known to be
1096 needed in commonly used termcap entries for certain terminals.
1098 4) The characters sent are incorrect, and clearly cannot be right for
1099 any terminal with the termcap entry you were using.
1101 This is unambiguously an Emacs bug, and can probably be fixed in
1102 termcap.c, terminfo.c, tparam.c, cm.c, redisplay-tty.c,
1103 redisplay-output.c, or redisplay.c.
1105 *** My buffers are full of \000 characters or otherwise corrupt.
1107 Some compilers have trouble with gmalloc.c and ralloc.c; try recompiling
1108 without optimization. If that doesn't work, try recompiling with
1109 SYSTEM_MALLOC defined, and/or with REL_ALLOC undefined.
1111 *** A position you specified in .Xdefaults is ignored, using twm.
1113 twm normally ignores "program-specified" positions.
1114 You can tell it to obey them with this command in your `.twmrc' file:
1116 UsePPosition "on" #allow clents to request a position
1118 *** With M-x enable-flow-control, you need to type C-\ twice to do
1119 incremental search--a single C-\ gets no response.
1121 This has been traced to communicating with your machine via kermit,
1122 with C-\ as the kermit escape character. One solution is to use
1123 another escape character in kermit. One user did
1125 set escape-character 17
1127 in his .kermrc file, to make C-q the kermit escape character.
1129 *** The Motif version of Emacs paints the screen a solid color.
1131 This has been observed to result from the following X resource:
1133 Emacs*default.attributeFont: -*-courier-medium-r-*-*-*-140-*-*-*-*-iso8859-*
1135 That the resource has this effect indicates a bug in something, but we
1136 do not yet know what. If it is an Emacs bug, we hope someone can
1137 explain what the bug is so we can fix it. In the mean time, removing
1138 the resource prevents the problem.
1140 *** After running emacs once, subsequent invocations crash.
1142 Some versions of SVR4 have a serious bug in the implementation of the
1143 mmap () system call in the kernel; this causes emacs to run correctly
1144 the first time, and then crash when run a second time.
1146 Contact your vendor and ask for the mmap bug fix; in the mean time,
1147 you may be able to work around the problem by adding a line to your
1148 operating system description file (whose name is reported by the
1149 configure script) that reads:
1150 #define SYSTEM_MALLOC
1151 This makes Emacs use memory less efficiently, but seems to work around
1154 *** Inability to send an Alt-modified key, when Emacs is communicating
1155 directly with an X server.
1157 If you have tried to bind an Alt-modified key as a command, and it
1158 does not work to type the command, the first thing you should check is
1159 whether the key is getting through to Emacs. To do this, type C-h c
1160 followed by the Alt-modified key. C-h c should say what kind of event
1161 it read. If it says it read an Alt-modified key, then make sure you
1162 have made the key binding correctly.
1164 If C-h c reports an event that doesn't have the Alt modifier, it may
1165 be because your X server has no key for the Alt modifier. The X
1166 server that comes from MIT does not set up the Alt modifier by
1169 If your keyboard has keys named Alt, you can enable them as follows:
1171 xmodmap -e 'add mod2 = Alt_L'
1172 xmodmap -e 'add mod2 = Alt_R'
1174 If the keyboard has just one key named Alt, then only one of those
1175 commands is needed. The modifier `mod2' is a reasonable choice if you
1176 are using an unmodified MIT version of X. Otherwise, choose any
1177 modifier bit not otherwise used.
1179 If your keyboard does not have keys named Alt, you can use some other
1180 keys. Use the keysym command in xmodmap to turn a function key (or
1181 some other 'spare' key) into Alt_L or into Alt_R, and then use the
1182 commands show above to make them modifier keys.
1184 Note that if you have Alt keys but no Meta keys, Emacs translates Alt
1185 into Meta. This is because of the great importance of Meta in Emacs.
1187 *** In Shell mode, you get a ^M at the end of every line.
1189 This happens to people who use tcsh, because it is trying to be too
1190 smart. It sees that the Shell uses terminal type `unknown' and turns
1191 on the flag to output ^M at the end of each line. You can fix the
1192 problem by adding this to your .cshrc file:
1195 if ($EMACS == "t") then
1197 stty -icrnl -onlcr -echo susp ^Z
1201 *** An error message such as `X protocol error: BadMatch (invalid
1202 parameter attributes) on protocol request 93'.
1204 This comes from having an invalid X resource, such as
1206 (which is invalid because it specifies a color name for something
1207 that isn't a color.)
1209 The fix is to correct your X resources.
1211 *** Once you pull down a menu from the menubar, it won't go away.
1213 It has been claimed that this is caused by a bug in certain very old
1214 (1990?) versions of the twm window manager. It doesn't happen with
1215 recent vintages, or with other window managers.
1217 *** Emacs ignores the "help" key when running OLWM.
1219 OLWM grabs the help key, and retransmits it to the appropriate client
1220 using XSendEvent. Allowing emacs to react to synthetic events is a
1221 security hole, so this is turned off by default. You can enable it by
1222 setting the variable x-allow-sendevents to t. You can also cause fix
1223 this by telling OLWM to not grab the help key, with the null binding
1224 "OpenWindows.KeyboardCommand.Help:".
1226 *** Programs running under terminal emulator do not recognize `emacs'
1229 The cause of this is a shell startup file that sets the TERMCAP
1230 environment variable. The terminal emulator uses that variable to
1231 provide the information on the special terminal type that Emacs
1234 Rewrite your shell startup file so that it does not change TERMCAP
1235 in such a case. You could use the following conditional which sets
1236 it only if it is undefined.
1238 if ( ! ${?TERMCAP} ) setenv TERMCAP ~/my-termcap-file
1240 Or you could set TERMCAP only when you set TERM--which should not
1241 happen in a non-login shell.
1243 *** The popup menu appears at the bottom/right of my screen.
1245 You probably have something like the following in your ~/.Xdefaults
1247 Emacs.geometry: 81x56--9--1
1249 Use the following instead
1251 Emacs*EmacsFrame.geometry: 81x56--9--1
1255 *** XEmacs crashes on MacOS within font-lock, or when dealing
1256 with large compilation buffers, or in other regex applications.
1258 The default stack size under MacOS/X is rather small (512k as opposed
1259 to Solaris 8M), hosing the regexp code, which uses alloca()
1260 extensively, overflowing the stack when complex regexps are used.
1263 1) Increase your stack size, using `ulimit -s 8192' or a (t)csh
1266 2) Recompile regex.c with REGEX_MALLOC defined.
1269 *** Your Delete key sends a Backspace to the terminal, using an AIXterm.
1271 The solution is to include in your .Xdefaults the lines:
1273 *aixterm.Translations: #override <Key>BackSpace: string(0x7f)
1274 aixterm*ttyModes: erase ^?
1276 This makes your Backspace key send DEL (ASCII 127).
1278 *** On AIX 4, some programs fail when run in a Shell buffer
1279 with an error message like No terminfo entry for "unknown".
1281 On AIX, many terminal type definitions are not installed by default.
1282 `unknown' is one of them. Install the "Special Generic Terminal
1283 Definitions" to make them defined.
1285 *** On AIX, you get this message when running Emacs:
1287 Could not load program emacs
1288 Symbol smtcheckinit in csh is undefined
1289 Error was: Exec format error
1293 Could not load program .emacs
1294 Symbol _system_con in csh is undefined
1295 Symbol _fp_trapsta in csh is undefined
1296 Error was: Exec format error
1298 These can happen when you try to run on AIX 3.2.5 a program that was
1299 compiled with 3.2.4. The fix is to recompile.
1301 *** Trouble using ptys on AIX.
1303 People often install the pty devices on AIX incorrectly.
1304 Use `smit pty' to reinstall them properly.
1308 *** The Emacs window disappears when you type M-q.
1310 Some versions of the Open Look window manager interpret M-q as a quit
1311 command for whatever window you are typing at. If you want to use
1312 Emacs with that window manager, you should try to configure the window
1313 manager to use some other command. You can disable the
1314 shortcut keys entirely by adding this line to ~/.OWdefaults:
1316 OpenWindows.WindowMenuAccelerators: False
1318 *** When Emacs tries to ring the bell, you get an error like
1320 audio: sst_open: SETQSIZE" Invalid argument
1321 audio: sst_close: SETREG MMR2, Invalid argument
1323 you have probably compiled using an ANSI C compiler, but with non-ANSI
1324 include files. In particular, on Suns, the file
1325 /usr/include/sun/audioio.h uses the _IOW macro to define the constant
1326 AUDIOSETQSIZE. _IOW in turn uses a K&R preprocessor feature that is
1327 now explicitly forbidden in ANSI preprocessors, namely substitution
1328 inside character constants. All ANSI C compilers must provide a
1329 workaround for this problem. Lucid's C compiler is shipped with a new
1330 set of system include files. If you are using GCC, there is a script
1331 called fixincludes that creates new versions of some system include
1332 files that use this obsolete feature.
1334 *** On Solaris 2.6, XEmacs dumps core when exiting.
1336 This happens if you're XEmacs is running on the same machine as the X
1337 server, and the optimized memory transport has been turned on by
1338 setting the environment variable XSUNTRANSPORT. The crash occurs
1339 during the call to XCloseDisplay.
1341 If this describes your situation, you need to undefine the
1342 XSUNTRANSPORT environment variable.
1344 *** On Solaris, C-x doesn't get through to Emacs when you use the console.
1346 This is a Solaris feature (at least on Intel x86 cpus). Type C-r
1347 C-r C-t, to toggle whether C-x gets through to Emacs.
1349 *** On Solaris 2.4, Dired hangs and C-g does not work. Or Emacs hangs
1350 forever waiting for termination of a subprocess that is a zombie.
1352 casper@fwi.uva.nl says the problem is in X11R6. Rebuild libX11.so
1353 after changing the file xc/config/cf/sunLib.tmpl. Change the lines
1356 #define SharedX11Reqs -lthread
1361 #if OSMinorVersion < 4
1363 #define SharedX11Reqs -lthread
1367 Be sure also to edit x/config/cf/sun.cf so that OSMinorVersion is 4
1368 (as it should be for Solaris 2.4). The file has three definitions for
1369 OSMinorVersion: the first is for x86, the second for SPARC under
1370 Solaris, and the third for SunOS 4. Make sure to update the
1371 definition for your type of machine and system.
1373 Then do `make Everything' in the top directory of X11R6, to rebuild
1374 the makefiles and rebuild X. The X built this way work only on
1375 Solaris 2.4, not on 2.3.
1377 For multithreaded X to work it necessary to install patch
1378 101925-02 to fix problems in header files [2.4]. You need
1379 to reinstall gcc or re-run just-fixinc after installing that
1382 However, Frank Rust <frust@iti.cs.tu-bs.de> used a simpler solution:
1384 #define ThreadedX YES
1386 #define ThreadedX NO
1387 in sun.cf and did `make World' to rebuild X11R6. Removing all
1388 `-DXTHREAD*' flags and `-lthread' entries from lib/X11/Makefile and
1389 typing 'make install' in that directory also seemed to work.
1391 *** On SunOS 4.1.3, Emacs unpredictably crashes in _yp_dobind_soft.
1393 This happens if you configure Emacs specifying just `sparc-sun-sunos4'
1394 on a system that is version 4.1.3. You must specify the precise
1395 version number (or let configure figure out the configuration, which
1396 it can do perfectly well for SunOS).
1398 *** Mail is lost when sent to local aliases.
1400 Many emacs mail user agents (VM and rmail, for instance) use the
1401 sendmail.el library. This library can arrange for mail to be
1402 delivered by passing messages to the /usr/lib/sendmail (usually)
1403 program . In doing so, it passes the '-t' flag to sendmail, which
1404 means that the name of the recipient of the message is not on the
1405 command line and, therefore, that sendmail must parse the message to
1406 obtain the destination address.
1408 There is a bug in the SunOS4.1.1 and SunOS4.1.3 versions of sendmail.
1409 In short, when given the -t flag, the SunOS sendmail won't recognize
1410 non-local (i.e. NIS) aliases. It has been reported that the Solaris
1411 2.x versions of sendmail do not have this bug. For those using SunOS
1412 4.1, the best fix is to install sendmail V8 or IDA sendmail (which
1413 have other advantages over the regular sendmail as well). At the time
1414 of this writing, these official versions are available:
1416 Sendmail V8 on ftp.cs.berkeley.edu in /ucb/sendmail:
1417 sendmail.8.6.9.base.tar.Z (the base system source & documentation)
1418 sendmail.8.6.9.cf.tar.Z (configuration files)
1419 sendmail.8.6.9.misc.tar.Z (miscellaneous support programs)
1420 sendmail.8.6.9.xdoc.tar.Z (extended documentation, with postscript)
1422 IDA sendmail on vixen.cso.uiuc.edu in /pub:
1423 sendmail-5.67b+IDA-1.5.tar.gz
1425 *** Emacs fails to understand most Internet host names, even though
1426 the names work properly with other programs on the same system.
1427 Emacs won't work with X-windows if the value of DISPLAY is HOSTNAME:0.
1428 Gnus can't make contact with the specified host for nntp.
1430 This typically happens on Suns and other systems that use shared
1431 libraries. The cause is that the site has installed a version of the
1432 shared library which uses a name server--but has not installed a
1433 similar version of the unshared library which Emacs uses.
1435 The result is that most programs, using the shared library, work with
1436 the nameserver, but Emacs does not.
1438 The fix is to install an unshared library that corresponds to what you
1439 installed in the shared library, and then relink Emacs.
1441 On SunOS 4.1, simply define HAVE_RES_INIT.
1443 If you have already installed the name resolver in the file libresolv.a,
1444 then you need to compile Emacs to use that library. The easiest way to
1445 do this is to add to config.h a definition of LIBS_SYSTEM, LIBS_MACHINE
1446 or LIB_STANDARD which uses -lresolv. Watch out! If you redefine a macro
1447 that is already in use in your configuration to supply some other libraries,
1448 be careful not to lose the others.
1450 Thus, you could start by adding this to config.h:
1452 #define LIBS_SYSTEM -lresolv
1454 Then if this gives you an error for redefining a macro, and you see that
1455 the s- file defines LIBS_SYSTEM as -lfoo -lbar, you could change config.h
1458 #define LIBS_SYSTEM -lresolv -lfoo -lbar
1460 *** With process-connection-type set to t, each line of subprocess
1461 output is terminated with a ^M, making ange-ftp and GNUS not work.
1463 On SunOS systems, this problem has been seen to be a result of an
1464 incomplete installation of gcc 2.2 which allowed some non-ANSI
1465 compatible include files into the compilation. In particular this
1466 affected virtually all ioctl() calls.
1470 *** XEmacs crashes on exit.
1472 This is known to happen with Lesstif version 0.93.36. It is
1473 apparently due to breakage in Lesstif. There is a patch for Lesstif.
1475 Frank McIngvale <frankm@hiwaay.net> says:
1477 Ok, 0.93.34 works, and I tracked down the crash to a section
1478 marked "experimental" in 0.93.36. Patch attached, "works for me".
1480 diff -u -r lesstif-0.93.36/lib/Xm/ImageCache.c lesstif-0.93.36-mod/lib/Xm/ImageCache.c
1481 --- lesstif-0.93.36/lib/Xm/ImageCache.c 2002-08-05 14:53:24.000000000 -0500
1482 +++ lesstif-0.93.36-mod/lib/Xm/ImageCache.c 2002-11-11 11:13:12.000000000 -0600
1483 @@ -1166,5 +1166,4 @@
1484 DEBUGOUT(_LtDebug0(__FILE__, NULL, "_LtImageCacheScreenDestroy (XmGetPixmapByDepth) %p\n",
1487 - (void) _LTHashTableForEachItem(PixmapCache, YowIter, (XtPointer)s);
1491 *** XEmacs crashes on startup, in make-frame.
1493 Typically the Lisp backtrace includes
1495 make-frame(nil #<x-device on ":0.0" 0x2558>)
1497 somewhere near the top. One problem is due to an improvement in GNU
1498 ld that sorts the ELF reloc sections in the executable, giving
1499 dramatic speedups in startup for large executables. It also confuses
1500 the traditional unexec code in XEmacs, leading to the core dump. The
1501 solution is to use either the `--ldflags="-z nocombreloc" or the
1502 "--pdump" option to configure. "--pdump" is recommended.
1504 Recent 21.4 and 21.5 versions of XEmacs autodetect this feature of ld
1505 in configure. Unfortunately, Red Hat and SuSE (at least) distributed
1506 prerelease versions of ld (numbered around 2.11.90.x.y, nicknamed
1507 "Hannibal Lecter" at XEmacs.ORG) where autodetection fails but the
1508 feature is enabled by default. The recommended procedure is to
1509 upgrade to binutils >= 2.12 and rerun configure. Otherwise you must
1510 apply the flags by hand.
1512 Andrew Jaffe reported a problem on Red Hat 7.3 with identical
1513 symptoms, except that ld was already being invoked with -z
1514 nocombreloc. Switching dialogs and widgets from Motif to Athena
1515 eliminated the problem. Both LessTif and OpenMotif were installed,
1516 and a bad interaction is suspected. This problem has not yet been
1520 **** XEmacs warns "Symbol `toggleClassRec' has different size in shared
1521 object, consider re-linking / Symbol `labelClassRec' has different
1522 size in shared object, consider re-linking / Warning: Representation
1523 size 4 must match superclass's to override value"
1525 Sometimes this results in segfaults when using the tab control widget
1526 or a progress bar widget.
1528 Some versions of Debian install 3D versions of the Athena widget
1529 library as /usr/X11R6/lib/libXaw.so. We have not yet solved the
1530 problem of identifying the actual library in use in ./configure, so it
1531 is possible for XEmacs to be compiled with reference to headers for
1532 "flat" Xaw but find a "3D" Xaw when loading.
1534 The straightforward solution is to rebuild XEmacs with additional
1535 configure options: --with-widgets=athena --with-athena=3d.
1537 There are several 3D Athena widget sets available; to see which ones
1538 are supported by XEmacs, use ./configure --usage.
1542 The Mandrake Linux distribution is attempting to comprehensively
1543 update the user interface, and make it consistent across
1544 applications. This is very difficult, and will occasionally cause
1545 conflicts with applications like Emacs with their own long-established
1546 interfaces. Known issues specific to Mandrake or especially common:
1548 Some versions of XEmacs (21.1.9 is known) distributed with Mandrake
1549 were patched to make the Meta and Alt keysyms synonymous. These
1550 normally work as expected in the Mandrake environment. However,
1551 custom-built XEmacsen (including all 21.2 betas) will "inexplicably"
1552 not respect the "Alt-invokes-Meta-commands" convention. See "I want
1553 XEmacs to use the Alt key" below.
1555 The color-gcc wrapper (see below) is in common use on the Mandrake
1558 *** XEmacs configured with ESD crashes with a segmentation violation
1560 This often occurs when a progress bar pops up.
1562 The problem is that the ESD sound daemon manipulates interrupts in a
1563 way that disagrees with XEmacs. The currently available workaround is
1564 not very satisfactory: remove ESD support. Rebuild after
1565 reconfiguring with the option --with-sound=none[,native][,nas]
1567 The funny syntax (requiring the initial "none") is for backward
1568 compatibility, and may change. Native sound and NAS do not cause the
1569 problem, so they may be added to the option to get some sound support.
1571 *** I want XEmacs to use the Alt key, not the XXX key, for Meta commands
1573 For historical reasons, XEmacs looks for a Meta key, then an Alt key.
1574 It binds Meta commands to the X11 modifier bit attached to the first
1575 of these it finds. On PCs, the Windows key is often assigned the Meta
1576 bit, but many desktop environments go to great lengths to get all apps
1577 to use the Alt key, and reserve the Windows key to (sensibly enough)
1580 One correct way to implement this was suggested on comp.emacs.xemacs
1581 (by Kilian Foth and in more detail by Michael Piotrowski): unmap the
1582 Meta modifier using xmodmap or xkb, and then map the Meta/Windows key
1583 to the Super or Hyper keysym and an appropriate mod bit. XEmacs will
1584 not find the Meta keysym, and default to using the Alt key for Meta
1585 keybindings. Typically few applications use the (X11) Meta modifier;
1586 it is tedious but not too much so to teach the ones you need to use
1587 Super instead of Meta. There may be further useful hints in the
1588 discussion of keymapping on non-Linux platforms.
1590 *** The color-gcc wrapper
1592 This wrapper colorizes the error messages from gcc. By default XEmacs
1593 does not interpret the escape sequences used to generate colors,
1594 resulting in a cluttered, hard-to-read buffer. You can remove the
1595 wrapper, or defeat the wrapper colorization in Emacs process buffers
1596 by editing the "nocolor" attribute in /etc/colorgccrc:
1598 $ diff -u /etc/colorgccrc.old /etc/colorgccrc
1599 --- /etc/colorgccrc.old Tue Dec 26 02:17:46 2000
1600 +++ /etc/colorgccrc Tue Dec 26 02:15:48 2000
1603 +nocolor: dumb emacs
1605 If you want colorization in your Emacs buffers, you may get good
1606 results from the ansi-color.el library:
1608 http://www.geocities.com/kensanata/color-emacs.html#ansicolors
1610 This is written for the mainline GNU Emacs but the author has made
1611 efforts to adapt it to XEmacs. YMMV.
1613 *** Slow startup on Linux.
1615 People using systems based on the Linux kernel sometimes report that
1616 startup takes 10 to 15 seconds longer than `usual'. There are two
1617 problems, one older, one newer.
1619 **** Old problem: IPv4 host lookup
1621 On older systems, this is because Emacs looks up the host name when it
1622 starts. Normally, this takes negligible time; the extra delay is due
1623 to improper system configuration. (Recent Linux distros usually have
1624 this configuration correct "out of the box".) This problem can occur
1625 for both networked and non-networked machines.
1627 Here is how to fix the configuration. It requires being root.
1629 ***** Networked Case
1631 First, make sure the files `/etc/hosts' and `/etc/host.conf' both
1632 exist. The first line in the `/etc/hosts' file should look like this
1633 (replace HOSTNAME with your host name):
1635 127.0.0.1 localhost HOSTNAME
1637 Also make sure that the `/etc/host.conf' files contains the following
1643 Any changes, permanent and temporary, to the host name should be
1644 indicated in the `/etc/hosts' file, since it acts a limited local
1645 database of addresses and names (e.g., some SLIP connections
1646 dynamically allocate ip addresses).
1648 ***** Non-Networked Case
1650 The solution described in the networked case applies here as well.
1651 However, if you never intend to network your machine, you can use a
1652 simpler solution: create an empty `/etc/host.conf' file. The command
1653 `touch /etc/host.conf' suffices to create the file. The `/etc/hosts'
1654 file is not necessary with this approach.
1656 **** New problem: IPv6 CNAME lookup
1658 A newer problem is due to XEmacs changing to use the modern
1659 getaddrinfo() interface from the older gethostbyname() interface. The
1660 solution above is insufficient, because getaddrinfo() by default tries
1661 to get IPv6 information for localhost (including the local X server).
1662 This always involves a dns lookup to get the CNAME, and the strategies
1663 above don't work. It then falls back to IPv4 behavior.
1665 ***** Robust network case
1667 Configure your network so that there are no nameservers configured
1668 until the network is actually running. getaddrinfo() will not try to
1669 access a nameserver that isn't configured.
1671 ***** Flaky network case
1673 If you have a flaky modem or DSL connection that can be relied on only
1674 to go down whenever you want to bring XEmacs up, you need to force
1675 IPv4 behavior. Explicitly setting DISPLAY=127.0.0.1:0.0 (or whatever
1676 is appropriate) works in most cases.
1678 If you cannot or do not want to do that, you can hard code IPv4
1679 behavior in src/process-unix.c. Add `--with-ipv6-cname=no" to your
1680 configure options and rebuild XEmacs.
1682 *** XEmacs disappears after several sound events in succession
1684 If esd is not running (e.g. you unchecked the run esd checkbox in the
1685 GNOME Configuration tool), every sound event will cause the esound
1686 client library to signal SIGUSR1. It is not that difficult (hold down
1687 PgUp or PgDn, for example) to cause XEmacs to generate additional
1688 SIGUSR1 signals before the first handler has a chance to reset, so the
1689 second SIGUSR1 causes the default behavior of ending the process.
1691 The solution is either to build XEmacs with -sound=native,noesd or to make sure
1692 that esd is in fact running.
1694 *** You get crashes in a non-C locale with Linux GNU Libc 2.0.
1696 Internationalization was not the top priority for GNU Libc 2.0.
1697 As of this writing (1998-12-28) you may get crashes while running
1698 XEmacs in a non-C locale. For example, `LC_ALL=en_US xemacs' crashes
1699 while `LC_ALL=C xemacs' runs fine. This happens for example with GNU
1700 libc 2.0.7. Installing libintl.a and libintl.h built from gettext
1701 0.10.35 and re-building XEmacs solves the crashes. Presumably soon
1702 everyone will upgrade to GNU Libc 2.1 and this problem will go away.
1704 *** `C-z', or `M-x suspend-emacs' hangs instead of suspending.
1706 If you build with `gpm' support on Linux, you cannot suspend XEmacs
1707 because gpm installs a buggy SIGTSTP handler. Either compile with
1708 `--with-gpm=no', or don't suspend XEmacs on the Linux console until
1711 *** With certain fonts, when the cursor appears on a character, the
1712 character doesn't appear--you get a solid box instead.
1714 One user on a Linux system reported that this problem went away with
1715 installation of a new X server. The failing server was XFree86 3.1.1.
1716 XFree86 3.1.2 works.
1719 *** On Irix, I don't see the toolbar icons and I'm getting lots of
1720 entries in the warnings buffer.
1722 SGI ships a really old Xpm library in /usr/lib which does not work at
1723 all well with XEmacs. The solution is to install your own copy of the
1724 latest version of Xpm somewhere and then use the --site-includes and
1725 --site-libraries flags to tell configure where to find it.
1727 *** Trouble using ptys on IRIX, or running out of ptys.
1729 The program mkpts (which may be in `/usr/adm' or `/usr/sbin') needs to
1730 be set-UID to root, or non-root programs like Emacs will not be able
1731 to allocate ptys reliably.
1733 *** Motif dialog boxes lose on Irix.
1735 Larry Auton <lda@control.att.com> writes:
1736 Beware of not specifying
1738 --with-dialogs=athena
1740 if it builds with the motif dialogs [boom!] you're a dead man.
1742 *** Beware of the default image & graphics library on Irix
1744 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
1746 You *have* to compile your own jpeg lib. The one delivered with SGI
1747 systems is a C++ lib, which apparently XEmacs cannot cope with.
1750 ** Digital UNIX/OSF/VMS/Ultrix
1751 *** XEmacs crashes on Digital Unix within font-lock, or when dealing
1752 with large compilation buffers, or in other regex applications.
1754 The default stack size under Digital Unix is rather small (2M as
1755 opposed to Solaris 8M), hosing the regexp code, which uses alloca()
1756 extensively, overflowing the stack when complex regexps are used.
1759 1) Increase your stack size, using `ulimit -s 8192' or a (t)csh
1762 2) Recompile regex.c with REGEX_MALLOC defined.
1764 *** The `Alt' key doesn't behave as `Meta' when running DECwindows.
1766 The default DEC keyboard mapping has the Alt keys set up to generate the
1767 keysym `Multi_key', which has a meaning to xemacs which is distinct from that
1768 of the `Meta_L' and `Meta-R' keysyms. A second problem is that certain keys
1769 have the Mod2 modifier attached to them for no adequately explored reason.
1770 The correct fix is to pass this file to xmodmap upon starting X:
1773 keysym Multi_key = Alt_L
1777 *** The Compose key on a DEC keyboard does not work as Meta key.
1779 This shell command should fix it:
1781 xmodmap -e 'keycode 0xb1 = Meta_L'
1783 *** `expand-file-name' fails to work on any but the machine you dumped
1786 On Ultrix, if you use any of the functions which look up information
1787 in the passwd database before dumping Emacs (say, by using
1788 expand-file-name in site-init.el), then those functions will not work
1789 in the dumped Emacs on any host but the one Emacs was dumped on.
1791 The solution? Don't use expand-file-name in site-init.el, or in
1792 anything it loads. Yuck - some solution.
1794 I'm not sure why this happens; if you can find out exactly what is
1795 going on, and perhaps find a fix or a workaround, please let us know.
1796 Perhaps the YP functions cache some information, the cache is included
1797 in the dumped Emacs, and is then inaccurate on any other host.
1801 *** I get complaints about the mapping of my HP keyboard at startup,
1802 but I haven't changed anything.
1804 The default HP keymap is set up to have Mod1 assigned to two different keys:
1805 Meta_L and Mode_switch (even though there is not actually a Mode_switch key on
1806 the keyboard -- it uses an "imaginary" keycode.) There actually is a reason
1807 for this, but it's not a good one. The correct fix is to execute this command
1810 xmodmap -e 'remove mod1 = Mode_switch'
1812 *** On HP-UX, you get "poll: Interrupted system call" message in the
1813 window where XEmacs was launched.
1815 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
1817 I get a very strange problem when linking libc.a dynamically: every
1818 event (mouse, keyboard, expose...) results in a "poll: Interrupted
1819 system call" message in the window where XEmacs was
1820 launched. Forcing a static link of libc.a alone by adding
1821 /usr/lib/libc.a at the end of the link line solves this. Note that
1822 my 9.07 build of 19.14b17 and my (old) build of 19.13 both exhibit
1823 the same behavior. I've tried various hpux patches to no avail. If
1824 this problem cannot be solved before the release date, binary kits
1825 for HP *must* be linked statically against libc, otherwise this
1826 problem will show up. (This is directed at whoever will volunteer
1827 for this kit, as I won't be available to do it, unless 19.14 gets
1828 delayed until mid-june ;-). I think this problem will be an FAQ soon
1829 after the release otherwise.
1831 Note: The above entry is probably not valid for XEmacs 21.0 and
1834 *** The right Alt key works wrong on German HP keyboards (and perhaps
1835 other non-English HP keyboards too).
1837 This is because HP-UX defines the modifiers wrong in X. Here is a
1838 shell script to fix the problem; be sure that it is run after VUE
1839 configures the X server.
1841 xmodmap 2> /dev/null - << EOF
1842 keysym Alt_L = Meta_L
1843 keysym Alt_R = Meta_R
1848 keysym Mode_switch = NoSymbol
1850 keysym Meta_R = Mode_switch
1851 add mod2 = Mode_switch
1855 *** XEmacs dumps core at startup when native audio is used. Native
1856 audio does not work with recent versions of HP-UX.
1858 Under HP-UX 10.20 and later (e.g., HP-UX 11.XX), with native audio
1859 enabled, the dumped XEmacs binary ("xemacs") core dumps at startup if
1860 recent versions of the libAlib.sl audio shared library is used. Note
1861 that "temacs" will run, but "xemacs" will dump core. This, of course,
1862 causes the XEmacs build to fail. If GNU malloc is enabled, a stack
1863 trace will show XEmacs to have crashed in the "first" call to malloc().
1865 This bug currently exists in all versions of XEmacs, when the undump
1866 mechanism is used. It is not known if using the experimental portable
1867 dumper will allow native audio to work.
1871 Recent versions of the HP-UX 10.20 (and later) audio shared library (in
1872 /opt/audio/lib), pulls in the libdce shared library, which pulls in a
1873 thread (libcma) library. This prevents the HP-UX undump() routine (in
1874 unexhp9k800.c) from properly working. What's happening is that some
1875 initialization routines are being called in the libcma library, *BEFORE*
1876 main() is called, and these initialization routines are calling
1877 malloc(). Unfortunately, in order for the undumper to work, XEmacs must
1878 adjust (move upwards) the sbrk() value *BEFORE* the first call to
1879 malloc(); if malloc() is called before XEmacs has properly adjusted sbrk
1880 (which is what is happening), dumped memory that is being used by
1881 XEmacs, is improperly re-allocated for use by malloc() and the dumped
1882 memory is corrupted. This causes XEmacs to die an horrible death.
1884 It is believed that versions of the audio library past December 1998
1885 will trigger this problem. Under HP-UX 10.20, you probably have to
1886 install audio library patches to encounter this. It's probable that
1887 recent "fresh, out-of-the-box" HP-UX 11.XX workstations also have this
1888 problem. For HP-UX 10.20, it's believed that audio patch PHSS_17121 (or
1889 a superceeding one, like PHSS_17554, PHSS_17971, PHSS_18777, PHSS_21481,
1890 or PHSS_21662, etc.) will trigger this.
1892 To check if your audio library will cause problems for XEmacs, run
1893 "chatr /opt/audio/lib/libAlib.sl". If "libdce" appears in the displayed
1894 shared library list, XEmacs will probably encounter problems if audio is
1899 Don't enable native audio. Re-run configure without native audio
1902 If your site supports it, try using NAS (Network Audio Support).
1904 Try using the experimental portable dumper. It may work, or it may
1908 *** `Pid xxx killed due to text modification or page I/O error'
1910 On HP-UX, you can get that error when the Emacs executable is on an NFS
1911 file system. HP-UX responds this way if it tries to swap in a page and
1912 does not get a response from the server within a timeout whose default
1913 value is just ten seconds.
1915 If this happens to you, extend the timeout period.
1917 *** Shell mode on HP-UX gives the message, "`tty`: Ambiguous".
1919 christos@theory.tn.cornell.edu says:
1921 The problem is that in your .cshrc you have something that tries to
1922 execute `tty`. If you are not running the shell on a real tty then tty
1923 will print "not a tty". Csh expects one word in some places, but tty
1924 is giving it back 3.
1926 The solution is to add a pair of quotes around `tty` to make it a
1929 if (`tty` == "/dev/console")
1931 should be changed to:
1933 if ("`tty`" == "/dev/console")
1935 Even better, move things that set up terminal sections out of .cshrc
1940 *** Regular expressions matching bugs on SCO systems.
1942 On SCO, there are problems in regexp matching when Emacs is compiled
1943 with the system compiler. The compiler version is "Microsoft C
1944 version 6", SCO 4.2.0h Dev Sys Maintenance Supplement 01/06/93; Quick
1945 C Compiler Version 1.00.46 (Beta). The solution is to compile with
1950 *** In general, the Windows code is less mature than the Unix code.
1952 The Windows code base is still changing quickly. If you are
1953 experiencing problems, try the latest beta version to see if the
1954 problem still exists. Also ask on xemacs-nt@xemacs.org.
1958 *** XEmacs fails to start because cygXpm-noX4.dll was not found.
1960 Andy Piper <andy@xemacs.org> sez:
1962 cygXpm-noX4 is part of the cygwin distribution under libraries or
1963 graphics, but is not installed by default. You need to run the
1964 cygwin setup again and select this package.
1966 *** Subprocesses do not work.
1968 You do not have "tty" in your CYGWIN environment variable. This must
1969 be set in your autoexec.bat (win95) or the system properties (winnt)
1970 as it must be read before the cygwin DLL initializes.
1972 *** ^G does not work on hung subprocesses.
1974 This is a known problem. It can be remedied by defining BROKEN_SIGIO
1975 in src/s/cygwin.h, however this currently leads to instability in XEmacs.
1976 (#### is this still true?)
1978 *** Errors from make like `/c:not found' when running `M-x compile'.
1980 Make sure you set the environment variable MAKE_MODE to UNIX in your
1981 init file (.xemacs/init.el), Control Panel (Windows 2000/NT), or
1982 AUTOEXEC.BAT (Windows 98/95).
1984 *** There are no images in the toolbar buttons.
1986 You need version 4.71 of commctrl.dll which does not ship with windows
1987 95. You can get this by installing IE 4.0 or downloading it from the
1991 * Compatibility problems (with Emacs 18, GNU Emacs, or previous XEmacs/lemacs)
1992 ==============================================================================
1994 *** "Symbol's value as variable is void: unread-command-char".
1995 "Wrong type argument: arrayp, #<keymap 143 entries>"
1996 "Wrong type argument: stringp, [#<keypress-event return>]"
1998 There are a few incompatible changes in XEmacs, and these are the
1999 symptoms. Some of the emacs-lisp code you are running needs to be
2000 updated to be compatible with XEmacs.
2002 The code should not treat keymaps as arrays (use `define-key', etc.),
2003 should not use obsolete variables like `unread-command-char' (use
2004 `unread-command-events'). Many (most) of the new ways of doing things
2005 are compatible in GNU Emacs and XEmacs.
2007 Modern Emacs packages (Gnus, VM, W3, efs, etc) are written to support
2008 GNU Emacs and XEmacs. We have provided modified versions of several
2009 popular emacs packages (dired, etc) which are compatible with this
2010 version of emacs. Check to make sure you have not set your load-path
2011 so that your private copies of these packages are being found before
2012 the versions in the lisp directory.
2014 Make sure that your load-path and your $EMACSLOADPATH environment
2015 variable are not pointing at an Emacs18 lisp directory. This will
2018 ** Some packages that worked before now cause the error
2019 Wrong type argument: arrayp, #<face ... >
2021 Code which uses the `face' accessor functions must be recompiled with
2022 xemacs 19.9 or later. The functions whose callers must be recompiled
2023 are: face-font, face-foreground, face-background,
2024 face-background-pixmap, and face-underline-p. The .elc files
2025 generated by version 19.9 will work in 19.6 and 19.8, but older .elc
2026 files which contain calls to these functions will not work in 19.9.
2028 ** Signaling: (error "Byte code stack underflow (byte compiler bug), pc 38")
2030 This error is given when XEmacs 20 is compiled without MULE support
2031 but is attempting to load a .elc which requires MULE support. The fix
2032 is to rebytecompile the offending file.
2034 ** Signaling: (wrong-type-argument ...) when loading mail-abbrevs
2036 The is seen when installing the Insidious Big Brother Data Base (bbdb)
2037 which includes an outdated copy of mail-abbrevs.el. Remove the copy
2038 that comes with bbdb and use the one that comes with XEmacs.
2044 ** A reminder: XEmacs/Mule work does not currently receive *any*
2045 funding, and all work is done by volunteers. If you think you can
2046 help, please contact the XEmacs maintainers.
2048 ** XEmacs/Mule doesn't support TTY's satisfactorily.
2050 This is a major problem, which we plan to address in a future release
2051 of XEmacs. Basically, XEmacs should have primitives to be told
2052 whether the terminal can handle international output, and which
2053 locale. Also, it should be able to do approximations of characters to
2054 the nearest supported by the locale.
2056 ** Internationalized (Asian) Isearch doesn't work.
2058 Currently, Isearch doesn't directly support any of the input methods
2059 that are not XIM based (like egg, canna and quail) (and there are
2060 potential problems with XIM version too...). If you're using egg
2061 there is a workaround. Hitting <RET> right after C-s to invoke
2062 Isearch will put Isearch in string mode, where a complete string can
2063 be typed into the minibuffer and then processed by Isearch afterwards.
2064 Since egg is now supported in the minibuffer using string mode you can
2065 now use egg to input your Japanese, Korean or Chinese string, then hit
2066 return to send that to Isearch and then use standard Isearch commands
2069 ** Using egg and mousing around while in 'fence' mode screws up my
2072 Don't do this. The fence modes of egg and canna are currently very
2073 modal, and messing with where they expect point to be and what they
2074 think is the current buffer is just asking for trouble. If you're
2075 lucky they will realize that something is awry, and simply delete the
2076 fence, but worst case can trash other buffers too. We've tried to
2077 protect against this where we can, but there still are many ways to
2078 shoot yourself in the foot. So just finish what you are typing into
2079 the fence before reaching for the mouse.
2081 ** Not all languages in Quail are supported like Devanagari and Indian
2082 languages, Lao and Tibetan.
2084 Quail requires more work and testing. Although it has been ported to
2085 XEmacs, it works really well for Japanese and for the European
2088 ** Right-to-left mode is not yet implemented, so languages like
2089 Arabic, Hebrew and Thai don't work.
2091 Getting this right requires more work. It may be implemented in a
2092 future XEmacs version, but don't hold your breath. If you know
2093 someone who is ready to implement this, please let us know.
2095 ** We need more developers and native language testers. It's extremely
2096 difficult (and not particularly productive) to address languages that
2097 nobody is using and testing.
2099 ** The kWnn and cWnn support for Chinese and Korean needs developers
2100 and testers. It probably doesn't work.
2102 ** There are no `native XEmacs' TUTORIALs for any Asian languages,
2103 including Japanese. FSF Emacs and XEmacs tutorials are quite similar,
2104 so it should be sufficient to skim through the differences and apply
2105 them to the Japanese version.
2107 ** We only have localized menus translated for Japanese, and the
2108 Japanese menus are developing bitrot (the Mule menu appears in
2111 ** XIM is untested for any language other than Japanese.