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 ===============================
44 Much general information is in INSTALL. If it's covered in
45 INSTALL, we don't repeat it here.
47 *** X11/bitmaps/gray (or other X11-related file) not found.
49 The X11R6 distribution was monolithic, but the X11R7 distribution is
50 much more modular. Many OS distributions omit these bitmaps (assuming
51 nobody uses them, evidently). Your OS distribution should have a
52 developer's package containing these files, probably with a name
53 containing the string "bitmap". Known package names (you may need to
54 add an extension such as .deb or .rpm) include x11/xbitmaps (Ubuntu)
55 and xorg-x11-xbitmaps (Fedora Core 5).
57 *** How do I configure to get the buffer tabs/progress bars?
59 These features depend on support for "native widgets". Use the
60 --with-widgets option to configure. Configuration of widgets is
61 automatic for "modern" toolkits (MS Windows, GTK, and Motif), but if
62 you are using Xt and the Athena widgets, you will probably want to
63 specify a "3d" widget set. See configure --usage, and don't forget to
64 install the corresponding development libraries.
66 *** I know I have libfoo installed, but configure doesn't find it.
68 Typical of Linux systems with package managers. To link with a shared
69 library, you only need the shared library. To compile objects that
70 link with it, you need the headers---and distros don't provide them with
71 the libraries. You need the additional "development" package, too.
73 *** When using gcc, you get the error message "undefined symbol __fixunsdfsi".
74 When using gcc, you get the error message "undefined symbol __main".
76 This means that you need to link with the gcc library. It may be called
77 "gcc-gnulib" or "libgcc.a"; figure out where it is, and define LIB_GCC in
78 config.h to point to it.
80 It may also work to use the GCC version of `ld' instead of the standard one.
82 *** Excessive optimization with pgcc can break XEmacs
84 It has been reported on some systems that compiling with -O6 can lead
85 to XEmacs failures. The workaround is to use a lower optimization
86 level. -O2 and -O4 have been tested extensively.
88 All of this depends heavily on the version of pgcc and the version
89 of libc. Snapshots near the release of pgcc-1.0 have been tested
90 extensively and no sign of breakage has been seen on systems using
93 *** src/Makefile and lib-src/Makefile are truncated--most of the file missing.
95 This can happen if configure uses GNU sed version 2.03. That version
96 had a bug. GNU sed version 2.05 works properly.
98 *** When compiling with X11, you get "undefined symbol _XtStrings".
100 This means that you are trying to link emacs against the X11r4 version of
101 libXt.a, but you have compiled either Emacs or the code in the lwlib
102 subdirectory with the X11r5 header files. That doesn't work.
104 Remember, you can't compile lwlib for r4 and emacs for r5, or vice versa.
105 They must be in sync.
107 *** test-distrib says that the distribution has been clobbered
108 or, temacs prints "Command key out of range 0-127"
109 or, temacs runs and dumps xemacs, but xemacs totally fails to work.
110 or, temacs gets errors dumping xemacs
112 This can be because the .elc files have been garbled. Do not be
113 fooled by the fact that most of a .elc file is text: these are binary
114 files and can contain all 256 byte values.
116 In particular `shar' cannot be used for transmitting GNU Emacs. It
117 typically truncates "lines". (this does not apply to GNU shar, which
118 uses uuencode to encode binary files.)
120 If you have a copy of Emacs that has been damaged in its nonprinting
121 characters, you can fix them by running:
125 This will rebuild all the needed .elc files.
127 ** Intel Architecture General
129 *** Don't use -O2 or -O3 with Cygwin 1.0, CodeFusion-99070 or gcc 2.7.2 on x86
130 without also using `-fno-strength-reduce'.
132 gcc will generate incorrect code otherwise. This bug is present in at
133 least 2.6.x and 2.7.[0-2]. This bug has been fixed in GCC 2.7.2.1 and
134 later. This bug is O/S independent, but is limited to x86 architectures.
136 This problem is known to be fixed in egcs (or pgcc) 1.0 or later.
138 Unfortunately, later releases of Cygnus-released compilers (not the
139 Net-released ones) have a bug with the same `problem signature'.
141 If you're lucky, you'll get an error while compiling that looks like:
143 event-stream.c:3189: internal error--unrecognizable insn:
144 (insn 256 14 15 (set (reg/v:SI 24)
145 (minus:SI (reg/v:SI 25)
146 (const_int 2))) -1 (insn_list 11 (nil))
150 If you're unlucky, your code will simply execute incorrectly.
152 *** Don't use -O2 with gcc 2.7.2 under Intel architectures without also
153 using `-fno-caller-saves'.
155 gcc will generate incorrect code otherwise. This bug is still
156 present in gcc 2.7.2.3. There have been no reports to indicate the
157 bug is present in egcs 1.0 (or pgcc 1.0) or later. This bug is O/S
158 independent, but limited to x86 architectures.
160 This problem is known to be fixed in egcs (or pgcc) 1.0 or later.
162 *** `compress' and `uncompress' not found and XFree86
164 XFree86 installs a very old version of libz.a by default ahead of where
165 more modern version of libz might be installed. This will cause problems
166 when attempting to link against libMagick. The fix is to remove the old
167 libz.a in the X11 binary directory.
172 Motif is the X11 version of the Gnus torture test: if there's a way to
173 crash, Motif will find it. With the open source release of Motif, it
174 seems like a good idea to collect all Motif-related issues in one
177 You should also look in your OS's section, as it may not be Motif's
180 *** XEmacs visibly repaints itty-bitty rectangles very slowly.
182 This should only be visible on a slow X connection (ISDN, maybe T1).
184 At least some versions of Motif apparently do not implement
185 XtExposeCompressMaximal properly, so it is disabled. If you wish to
186 experiment, you can remove the #ifdef LWLIB_NEEDS_MOTIF at line 238
187 (or so) of src/EmacsFrame.c, leaving only the line
189 /* compress_exposure */ XtExposeCompressMaximal | XtExposeNoRegion,
191 and recompile. This enables exposure compression, giving a 10:1 or
192 better speedup for some users. However, on some Motif platforms (Red
193 Hat Linux 9.0 and Solaris 2.8, at least), this causes XEmacs to hang
194 while displaying the progress bar (eg, in font-lock). A workaround
195 for that problem is to setq `progress-feedback-use-echo-area' to `t'.
197 *** XEmacs crashes on exit (#1).
199 The backtrace is something like:
202 #0 0xfeb9a480 in _libc_kill () from /usr/lib/libc.so.1
203 #1 0x000b0388 in fatal_error_signal ()
204 #2 <signal handler called>
205 #3 YowIter (ht=0xb, id=0x0, v=0x74682074, client=0x47e3c0)
207 #4 0xff26cc5c in _LTHashTableForEachItem (ht=0x4725e8,
208 iter=0xff26dda0 <YowIter>, ClientData=0x47e3c0) at Hash.c:671
209 #5 0xff2a4664 in destroy (w=0x496550) at Screen.c:352
210 #6 0xfef92118 in Phase2Destroy () from /usr/openwin/lib/libXt.so.4
211 #7 0xfef91940 in Recursive () from /usr/openwin/lib/libXt.so.4
212 #8 0xfef91e44 in XtPhase2Destroy () from /usr/openwin/lib/libXt.so.4
213 #9 0xfef91ae8 in _XtDoPhase2Destroy () from /usr/openwin/lib/libXt.so.4
214 #10 0xfef918cc in XtDestroyWidget () from /usr/openwin/lib/libXt.so.4
215 #11 0xfef91438 in CloseDisplay () from /usr/openwin/lib/libXt.so.4
216 #12 0xfef91394 in XtCloseDisplay () from /usr/openwin/lib/libXt.so.4
217 #13 0x0025b8b0 in x_delete_device ()
218 #14 0x000940b0 in delete_device_internal ()
219 #15 0x000806a0 in delete_console_internal ()
221 This is known to happen with Lesstif version 0.93.36. Similar
222 backtraces have also been observed on HP/UX and Solaris. There is a
223 patch for Lesstif. (This is not a solution; it just stops the crash.
224 It may or may not be harmless, but "it works for the author".)
226 Note that this backtrace looks a lot like the one in the next item.
227 However, this one is invulnerable to the Solaris patches mentioned there.
229 Frank McIngvale <frankm@hiwaay.net> says:
231 Ok, 0.93.34 works, and I tracked down the crash to a section
232 marked "experimental" in 0.93.36. Patch attached, "works for me".
234 diff -u -r lesstif-0.93.36/lib/Xm/ImageCache.c lesstif-0.93.36-mod/lib/Xm/ImageCache.c
235 --- lesstif-0.93.36/lib/Xm/ImageCache.c 2002-08-05 14:53:24.000000000 -0500
236 +++ lesstif-0.93.36-mod/lib/Xm/ImageCache.c 2002-11-11 11:13:12.000000000 -0600
237 @@ -1166,5 +1166,4 @@
238 DEBUGOUT(_LtDebug0(__FILE__, NULL, "_LtImageCacheScreenDestroy (XmGetPixmapByDepth) %p\n",
241 - (void) _LTHashTableForEachItem(PixmapCache, YowIter, (XtPointer)s);
244 *** XEmacs crashes on exit (#2)
246 Especially frequent with multiple frames. Crashes that produce C
247 backtraces like this:
249 #0 0xfec9a118 in _libc_kill () from /usr/lib/libc.so.1
250 #1 0x77f48 in fatal_error_signal (sig=11)
251 at /codes/rpluim/xemacs-21.4/src/emacs.c:539
252 #2 <signal handler called>
253 #3 0xfee929f4 in XFindContext () from /usr/openwin/lib/libX11.so.4
254 #4 0xfee92930 in XFindContext () from /usr/openwin/lib/libX11.so.4
255 #5 0xff297e54 in DisplayDestroy () from /usr/dt/lib/libXm.so.4
256 #6 0xfefbece0 in XtCallCallbackList () from /usr/openwin/lib/libXt.so.4
257 #7 0xfefc486c in XtPhase2Destroy () from /usr/openwin/lib/libXt.so.4
258 #8 0xfefc45d0 in _XtDoPhase2Destroy () from /usr/openwin/lib/libXt.so.4
259 #9 0xfefc43b4 in XtDestroyWidget () from /usr/openwin/lib/libXt.so.4
260 #10 0x15cf9c in x_delete_device (d=0x523f00)
262 are caused by buggy Motif libraries. Installing the following patches
263 has been reported to solve the problem on Solaris 2.7:
267 For information (although they have not been confirmed to work), the
268 equivalent patches for Solaris 2.8 are:
272 *** On HP-UX 11.0 XEmacs causes excessive X11 errors when running.
273 (also appears on AIX as reported in comp.emacs.xemacs)
275 Marcus Thiessel <marcus@xemacs.org>
277 Unfortunately, XEmacs releases prior to 21.0 don't work with
278 Motif2.1. It will compile but you will get excessive X11 errors like
280 xemacs: X Error of failed request: BadGC (invalid GC parameter)
282 and finally XEmacs gets killed. A workaround is to use the
283 Motif1.2_R6 libraries. You can the following line to your call to
286 --x-libraries="/usr/lib/Motif1.2_R6 -L/usr/lib/X11R6"
288 Make sure /usr/lib/Motif1.2_R6/libXm.sl is a link to
289 /usr/lib/Motif1.2_R6/libXm.3.
291 *** On HP-UX 11.0: Object "" does not have windowed ancestor
293 Marcus Thiessel <marcus@xemacs.org>
295 XEmacs dies without core file and reports:
297 Error: Object "" does not have windowed ancestor.
299 This is a bug. Please apply the patch PHSS_19964 (check if
300 superseded). The other alternative is to link with Motif1.2_R6 (see
303 *** Motif dialog boxes lose on Irix.
305 Larry Auton <lda@control.att.com> writes:
306 Beware of not specifying
308 --with-dialogs=athena
310 if it builds with the motif dialogs [boom!] you're a dead man.
314 *** IBM compiler fails: "The character # is not a valid C source character."
316 Most recently observed in 21.5.9, due to USE_KKCC ifdefs (they just
317 happen to tickle the implementation).
319 Valdis Kletnieks says:
321 The problem is that IBM defines a *MACRO* called 'memcpy', and we
322 have stuck a #ifdef/#endif inside the macro call. As a workaround,
323 try adding '-U__STR__' to your CFLAGS - this will cause string.h to
324 not do a #define for strcpy() to __strcpy() - it uses this for
325 automatic inlining support.
327 (For the record, the same issue affects a number of other functions
328 defined in string.h - basically anything the compiler knows how to
331 *** On AIX 4.3, you must specify --with-dialogs=athena with configure
333 *** The libXt shipped with AIX 4.3 up to 4.3.2 is broken. This causes
334 xemacs -nw to fail in various ways. The official APAR is this:
336 APAR NUMBER: <IX89470> RESOLVED AS: PROGRAM ERROR
339 <IX89470>: LIBXT.A INCORRECT HANDLING OF EXCEPTIONS IN XTAPPADDINPUT
341 The solution is to install X11.base.lib at version >=4.3.2.5.
343 *** On AIX, you get this compiler error message:
345 Processing include file ./XMenuInt.h
346 1501-106: (S) Include file X11/Xlib.h not found.
348 This means your system was installed with only the X11 runtime i.d
349 libraries. You have to find your sipo (bootable tape) and install
352 *** On AIX 4.1.2, linker error messages such as
353 ld: 0711-212 SEVERE ERROR: Symbol .__quous, found in the global symbol table
354 of archive /usr/lib/libIM.a, was not defined in archive member shr.o.
356 This is a problem in libIM.a. You can work around it by executing
357 these shell commands in the src subdirectory of the directory where
360 cp /usr/lib/libIM.a .
364 Then change -lIM to ./libIM.a in the command to link temacs (in
367 *** Excessive optimization on AIX 4.2 can lead to compiler failure.
369 Valdis.Kletnieks@vt.edu writes:
370 At least at the b34 level, and the latest-and-greatest IBM xlc
371 (3.1.4.4), there are problems with -O3. I haven't investigated
376 *** Don't use -O2 with gcc 2.8.1 and egcs 1.0 under SPARC architectures
377 without also using `-fno-schedule-insns'.
379 gcc will generate incorrect code otherwise, typically resulting in
380 crashes in the function skip-syntax-backward.
382 *** Don't use gcc-2.95.2 with -mcpu=ultrasparc on Solaris 2.6.
384 gcc will assume a 64-bit operating system, even though you've
385 merely told it to assume a 64-bit instruction set.
387 *** Dumping error when using GNU binutils / GNU ld on a Sun.
389 Errors similar to the following:
391 Dumping under the name xemacs unexec():
392 dldump(/space/rpluim/xemacs-obj/src/xemacs): ld.so.1: ./temacs:
393 fatal: /space/rpluim/xemacs-obj/src/xemacs: unknown dynamic entry:
396 are caused by using GNU ld. There are several workarounds available:
398 In XEmacs 21.2 or later, configure using the new portable dumper
401 Alternatively, you can link using the Sun version of ld, which is
402 normally held in /usr/ccs/bin. This can be done by one of:
404 - building gcc with these configure flags:
405 configure --with-ld=/usr/ccs/bin/ld --with-as=/usr/ccs/bin/as
407 - adding -B/usr/ccs/bin/ to CFLAGS used to configure XEmacs
408 (Note: The trailing '/' there is significant.)
410 - uninstalling GNU ld.
412 The Solaris2 FAQ claims:
414 When you install gcc, don't make the mistake of installing
415 GNU binutils or GNU libc, they are not as capable as their
416 counterparts you get with Solaris 2.x.
418 *** Link failure when using acc on a Sun.
420 To use acc, you need additional options just before the libraries, such as
422 /usr/lang/SC2.0.1/values-Xt.o -L/usr/lang/SC2.0.1/cg87 -L/usr/lang/SC2.0.1
424 and you need to add -lansi just before -lc.
426 The precise file names depend on the compiler version, so we
427 cannot easily arrange to supply them.
429 *** Problems finding X11 libraries on Solaris with Openwindows
431 Some users have reported problems in this area. The reported solution
432 is to define the environment variable OPENWINHOME, even if you must set
433 it to `/usr/openwin'.
435 *** Sed problems on Solaris 2.5
437 There have been reports of Sun sed truncating very lines in the
438 Makefile during configuration. The workaround is to use GNU sed or,
439 even better, think of a better way to generate Makefile, and send us a
442 *** On Solaris 2 I get undefined symbols from libcurses.a.
444 You probably have /usr/ucblib/ on your LD_LIBRARY_PATH. Do the link with
445 LD_LIBRARY_PATH unset. Generally, avoid using any ucb* stuff when
448 *** On Solaris 2 I cannot make alloc.o, glyphs.o or process.o.
450 The SparcWorks C compiler may have difficulty building those modules
451 with optimization level -xO4. Try using only "-fast" optimization
452 for just those modules. (Or use gcc).
454 *** Solaris 2.3 /bin/sh coredumps during configuration.
456 This only occurs if you have LANG != C. This is a known bug with
457 /bin/sh fixed by installing Patch-ID# 101613-01. Or, you can use
458 bash by setting the environment variable CONFIG_SHELL to /bin/bash
460 *** Solaris 2.x configure fails: ./config.status: test: argument expected
462 This is a known bug with /bin/sh and /bin/test, i.e. they do not
463 support the XPG4 standard. You can use bash as a workaround or an
464 XPG4-compliant Bourne shell such as the Sun-supplied /usr/xpg4/bin/sh
465 by setting the environment variable CONFIG_SHELL to /usr/xpg4/bin/sh
467 *** On SunOS, you get linker errors
469 _get_wmShellWidgetClass
470 _get_applicationShellWidgetClass
472 The fix to this is to install patch 100573 for OpenWindows 3.0
473 or link libXmu statically.
475 *** On Sunos 4, you get the error ld: Undefined symbol __lib_version.
477 This is the result of using cc or gcc with the shared library meant
478 for acc (the Sunpro compiler). Check your LD_LIBRARY_PATH and delete
479 /usr/lang/SC2.0.1 or some similar directory.
481 *** Undefined symbols when linking on Sunos 4.1.
483 If you get the undefined symbols _atowc _wcslen, _iswprint, _iswspace,
484 _iswcntrl, _wcscpy, and _wcsncpy, then you need to add -lXwchar after
485 -lXaw in the command that links temacs.
487 This problem seems to arise only when the international language
488 extensions to X11R5 are installed.
490 *** On a Sun running SunOS 4.1.1, you get this error message from GNU ld:
492 /lib/libc.a(_Q_sub.o): Undefined symbol __Q_get_rp_rd referenced from text segment
494 The problem is in the Sun shared C library, not in GNU ld.
496 The solution is to install Patch-ID# 100267-03 from Sun.
498 *** SunOS 4.1.2: undefined symbol _get_wmShellWidgetClass
500 Apparently the version of libXmu.so.a that Sun ships is hosed: it's missing
501 some stuff that is in libXmu.a (the static version). Sun has a patch for
502 this, but a workaround is to use the static version of libXmu, by changing
503 the link command from "-lXmu" to "-Bstatic -lXmu -Bdynamic". If you have
504 OpenWindows 3.0, ask Sun for these patches:
505 100512-02 4.1.x OpenWindows 3.0 libXt Jumbo patch
506 100573-03 4.1.x OpenWindows 3.0 undefined symbols with shared libXmu
508 *** Random other SunOS 4.1.[12] link errors.
510 The X headers and libraries that Sun ships in /usr/{include,lib}/X11 are
511 broken. Use the ones in /usr/openwin/{include,lib} instead.
515 See also Intel Architecture General, above.
517 *** egcs-1.1 on Alpha Linux
519 There have been reports of egcs-1.1 not compiling XEmacs correctly on
520 Alpha Linux. There have also been reports that egcs-1.0.3a is O.K.
522 *** Under Linux, you get "too many arguments to function `getpgrp'".
524 You have probably installed LessTiff under `/usr/local' and `libXm.so'
525 could not be found when linking `getpgrp()' test program, making XEmacs
526 think that `getpgrp()' takes an argument. Try adding `/usr/local/lib'
527 in `/etc/ld.so.conf' and run `ldconfig'. Then run XEmacs's `configure'
528 again. As with all problems of this type, reading the config.log file
529 generated from configure and seeing the log of how the test failed can
532 *** `Error: No ExtNode to pop!' on Linux systems with Lesstif.
534 This error message has been observed with lesstif-0.75a. It does not
535 appear to cause any harm.
537 *** xemacs: can't resolve symbol '__malloc_hook'
539 This is a Linux problem where you've compiled the XEmacs binary on a libc
540 5.4 with version higher than 5.4.19 and attempted to run the binary against
541 an earlier version. The solution is to upgrade your old library.
545 *** More coredumping in Irix (6.5 known to be vulnerable)
547 No fix is known yet. Here's the best information we have:
549 Valdis Kletnieks <Valdis.Kletnieks@vt.edu> writes:
551 Were xemacs and [any 3rd party, locally-compiled] libraries [you use]
552 all compiled with the same ABI ( -o32, -n32, -64) and
553 mips2/mips3/mips4 flags, and are they appropriate for the machine in
554 question? I know the IP30 implies an Octane, so it should be an R10K
555 chipset and above such nonsense, but I've seen the most astoundingly
556 bizzare crashes when somebody managed to compile with -mips4 and get
557 it to run on an R4400 or R5K system. ;)
559 Also, since you're using gcc, try re-running fixincludes and *then*
560 rebuilding xemacs and [any] libraries - mismatched headers can do that
561 sort of thing to you with little or no clue what's wrong (often you
562 get screwed when one routine does an malloc(sizeof(foo_struct)) and
563 passes the result to something that things foo_struct is a bit bigger,
566 Here's typical crash backtrace. With --pdump, this occurs usually at
567 startup under X windows and xemacs -nw at least starts, while without
568 --pdump a similar crash is observed during build.
570 #0 0x0fa460b8 in kill () at regcomp.c:637
571 637 regcomp.c: No such file or directory.
574 #0 0x0fa460b8 in kill () at regcomp.c:637
575 #1 0x10087f34 in fatal_error_signal ()
578 This is confusing because there is no such file in the XEmacs
579 distribution. This is seen on (at least) the following configurations:
581 uname -a: IRIX64 oct202 6.5 01091821 IP30
582 XEmacs 21.4.9 "Informed Management" configured for `mips-sgi-irix6.5'.
583 XEmacs 21.5-b9 "brussels sprouts" configured for `mips-sgi-irix6.5'.
585 *** On Irix 6.5, the MIPSpro compiler gets an internal compiler error
587 The MIPSpro Compiler (at least version 7.2.1) can't seem to handle the
588 union type properly, and fails to compile src/glyphs.c. To avoid this
589 problem, always build ---use-union-type=no (but that's the default, so
590 you should only see this problem if you're an XEmacs maintainer).
592 *** Linking with -rpath on IRIX.
594 Darrell Kindred <dkindred@cmu.edu> writes:
595 There are a couple of problems [with use of -rpath with Irix ld], though:
597 1. The ld in IRIX 5.3 ignores all but the last -rpath
598 spec, so the patched configure spits out a warning
599 if --x-libraries or --site-runtime-libraries are
600 specified under irix 5.x, and it only adds -rpath
601 entries for the --site-runtime-libraries. This bug was
602 fixed sometime between 5.3 and 6.2.
604 2. IRIX gcc 2.7.2 doesn't accept -rpath directly, so
605 it would have to be prefixed by -Xlinker or "-Wl,".
606 This would be fine, except that configure compiles with
607 ${CC-cc} $CFLAGS $LDFLAGS ...
608 rather than quoting $LDFLAGS with prefix-args, like
609 src/Makefile does. So if you specify --x-libraries
610 or --site-runtime-libraries, you must use --use-gcc=no,
611 or configure will fail.
613 *** On Irix 6.3, the SGI ld quits with segmentation fault when linking temacs
615 This occurs if you use the SGI linker version 7.1. Installing the
616 patch SG0001872 fixes this problem.
618 *** On Irix 6.0, make tries (and fails) to build a program named unexelfsgi
620 A compiler bug inserts spaces into the string "unexelfsgi . o"
621 in src/Makefile. Edit src/Makefile, after configure is run,
622 find that string, and take out the spaces.
624 Compiler fixes in Irix 6.0.1 should eliminate this problem.
626 *** On Irix 5.2, unexelfsgi.c can't find cmplrs/stsupport.h.
628 The file cmplrs/stsupport.h was included in the wrong file set in the
629 Irix 5.2 distribution. You can find it in the optional fileset
630 compiler_dev, or copy it from some other Irix 5.2 system. A kludgy
631 workaround is to change unexelfsgi.c to include sym.h instead of
634 *** Coredumping in Irix 6.2
636 Pete Forman <gsez020@compo.bedford.waii.com> writes:
637 A problem noted by myself and others (I've lost the references) was
638 that XEmacs coredumped when the cut or copy toolbar buttons were
639 pressed. This has been fixed by loading the SGI patchset (Feb 98)
640 without having to recompile XEmacs.
642 My versions are XEmacs 20.3 (problem first noted in 19.15) and IRIX
643 6.2, compiled using -n32. I'd guess that the relevant individual
644 patch was "SG0002580: multiple fixes for X libraries". SGI recommends
645 that the complete patch set be installed rather than parts of it.
647 ** Digital UNIX/OSF/VMS
648 *** On Digital UNIX, the DEC C compiler might have a problem compiling
651 In particular, src/extents.c and src/faces.c might cause the DEC C
652 compiler to abort. When this happens: cd src, compile the files by
653 hand, cd .., and redo the "make" command. When recompiling the files by
654 hand, use the old C compiler for the following versions of Digital UNIX:
655 - V3.n: Remove "-migrate" from the compile command.
656 - V4.n: Add "-oldc" to the compile command.
658 A related compiler bug has been fixed by the DEC compiler team. The
659 new versions of the compiler should run fine.
661 *** Under some versions of OSF XEmacs runs fine if built without
662 optimization but will crash randomly if built with optimization.
664 Using 'cc -g' is not sufficient to eliminate all optimization. Try
667 *** Compilation errors on VMS.
669 Sorry, XEmacs does not work under VMS. You might consider working on
670 the port if you really want to have XEmacs work under VMS.
673 *** On HPUX, the HP C compiler might have a problem compiling some files
676 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
678 Had to drop once again to level 2 optimization, at least to
679 compile lstream.c. Otherwise, I get a "variable is void: \if"
680 problem while dumping (this is a problem I already reported
681 with vanilla hpux 10.01 and 9.07, which went away after
682 applying patches for the C compiler). Trouble is I still
683 haven't found the same patch for hpux 10.10, and I don't
684 remember the patch numbers. I think potential XEmacs builders
685 on HP should be warned about this.
687 *** I don't have `xmkmf' and `imake' on my HP.
689 You can get these standard X tools by anonymous FTP to
690 hpcvaaz.cv.hp.com. Essentially all X programs need these.
692 *** On HP-UX, problems with make
694 Marcus Thiessel <marcus@xemacs.org>
696 Some releases of XEmacs (e.g. 20.4) require GNU make to build
697 successfully. You don't need GNU make when building 21.x.
699 *** On HP-UX 9.05 XEmacs won't compile or coredump during the build.
701 Marcus Thiessel <marcus@xemacs.org>
703 This might be a sed problem. For your own safety make sure to use
704 GNU sed while dumping XEmacs.
708 *** Native cc on SCO OpenServer 5 is now OK. Icc may still throw you
709 a curve. Here is what Robert Lipe <robertl@arnet.com> says:
711 Unlike XEmacs 19.13, building with the native cc on SCO OpenServer 5
712 now produces a functional binary. I will typically build this
713 configuration for COFF with:
715 /path_to_xemacs_source/configure --with-gcc=no \
716 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
717 --with-xpm --with-xface --with-sound=nas
719 This version now supports ELF builds. I highly recommend this to
720 reduce the in-core footprint of XEmacs. This is now how I compile
721 all my test releases. Build it like this:
723 /path_to_XEmacs_source/configure --with-gcc=no \
724 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
725 --with-xpm --with-xface --with-sound=nas --dynamic
727 The compiler known as icc [ supplied with the OpenServer 5 Development
728 System ] generates a working binary, but it takes forever to generate
729 XEmacs. ICC also whines more about the code than /bin/cc does. I do
730 believe all its whining is legitimate, however. Note that you do
731 have to 'cd src ; make LD=icc' to avoid linker errors.
733 The way I handle the build procedure is:
735 /path_to_XEmacs_source/configure --with-gcc=no \
736 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
737 --with-xpm --with-xface --with-sound=nas --dynamic --compiler="icc"
739 NOTE I have the xpm, xface, and audio libraries and includes in
740 /usr/local/lib, /usr/local/include. If you don't have these,
741 don't include the "--with-*" arguments in any of my examples.
743 In previous versions of XEmacs, you had to override the defaults while
744 compiling font-lock.o and extents.o when building with icc. This seems
745 to no longer be true, but I'm including this old information in case it
746 resurfaces. The process I used was:
749 [ procure pizza, beer, repeat ]
751 make CC="icc -W0,-mP1COPT_max_tree_size=3000" font-lock.o extents.o
754 If you want sound support, get the tls566 supplement from
755 ftp.sco.com:/TLS or any of its mirrors. It works just groovy
758 The M-x manual-entry is known not to work. If you know Lisp and would
759 like help in making it work, e-mail me at <robertl@dgii.com>.
760 (UNCHECKED for 19.15 -- it might work).
762 In earlier releases, gnuserv/gnuclient/gnudoit would open a frame
763 just fine, but the client would lock up and the server would
764 terminate when you used C-x # to close the frame. This is now
767 In etc/ there are two files of note. emacskeys.sco and emacsstrs.sco.
768 The comments at the top of emacskeys.sco describe its function, and
769 the emacstrs.sco is a suitable candidate for /usr/lib/keyboard/strings
770 to take advantage of the keyboard map in emacskeys.sco.
772 Note: Much of the above entry is probably not valid for XEmacs 21.0
777 *** XEmacs complains "No such file or directory, diff"
779 or "ispell" or other commands that seem related to whatever you just
782 There are a large number of common (in the sense that "everyone has
783 these, really") Unix utilities that are not provided with XEmacs. The
784 GNU Project's implementations are available for Windows in the the
785 Cygwin distribution (http://www.cygwin.com/), which also provides a
786 complete Unix emulation environment (and thus makes ports of Unix
787 utilities nearly trivial). Another implementation is that from MinGW
788 (http://www.mingw.org/msys.shtml).
790 *** Weird crashes in pdump load or shortly after pdump load.
792 This can happen with incremental linking. Check if you have set
793 SUPPORT_EDIT_AND_CONTINUE to non-zero in config.inc, which must allow
794 incremental linking to be enabled (otherwise it's disabled). Either turn
795 this off, execute `nmake -f xemacs.mak clean', or manually remove
796 `temacs.exe' and `xemacs.exe'.
800 See also Intel Architecture General, above.
802 *** Signal 11 when building or running a dumped XEmacs.
804 This appears to happen when using the traditional dumping mechanism and
805 the system malloc. Andy Piper writes:
807 Traditional dumping on Cygwin relies on using gmalloc (there are specific
808 hacks in our version of gmalloc to support this), I suspect using sysmalloc
811 Try configuring with pdump or without system malloc.
813 *** In general use etc/check_cygwin_setup.sh to trap environment problems.
815 The script etc/check_cygwin_setup.sh will attempt to detect whether
816 you have a suitable environment for building. This script may not work
817 correctly if you are using ash instead of bash (see below).
819 *** Syntax errors running configure scripts, make failing with exit code 127
820 in inexplicable situations, etc.
822 [[ This may be because you are using the default Cygwin shell, under old
823 versions of Cygwin. The default Cygwin shell (/bin/sh.exe) is ash, which
824 appears to work in most circumstances but has some weird failure modes.
825 You may need to replace the symlink with bash.exe. ]] This doesn't appear
826 to affect Cygwin any longer, and /bin/sh.exe is no longer a symlink in
829 *** Lots of compile errors, esp. on lines containing macro definitions
830 terminated by backslashes.
832 Your partition holding the source files is mounted binary. It needs
833 to be mounted text. (This will not screw up any binary files because
834 the Cygwin utilities specify explicitly whether they want binary or
835 text mode when working with source vs. binary files, which overrides
836 the mount type.) To fix this, you just need to run the appropriate
837 mount command once -- afterwards, the settings are remembered in the
840 *** Errors from make like /c:not found.
842 Make sure you set the environment variable MAKE_MODE to UNIX in your
843 .bashrc, Control Panel (Windows 2000/NT), or AUTOEXEC.BAT (Windows
846 *** The info files will not build.
848 makeinfo that ships with old versions of Cygwin doesn't work.
849 Upgrade to the latest Cygwin version.
851 *** XEmacs hangs while attempting to rebuild the .elc files.
853 Check to make sure you're not configuring with rel-alloc. The relocating
854 allocator does not currently work under Cygwin due to bugs in Cygwin's
857 *** Trying to build with X, but X11 not detected.
859 This is usually because xmkmf is not in your path or because you are
860 using the default Cygwin shell. (See above.)
863 * Problems with running XEmacs
864 ==============================
867 *** XEmacs consistently crashes in a particular strange place.
869 One known case is on Red Hat Linux, compiled with GCC, attempting to
870 render PNG images. The problem is that XEmacs code is not compliant
871 with ANSI rules about aliasing. Adding -fno-strict-aliasing to CFLAGS
872 may help (or the equivalent for your compiler). (Some versions of
873 XEmacs may already do this automatically, but if you specify CFLAGS or
874 --cflags yourself, you will have to add this flag by hand.)
876 If you diagnose this bug for some other symptoms or systems, please
877 let us know (if you can send mail from the affected system, use M-x
878 report-xemacs-bug) so we can update this entry.
880 *** Changes made to .el files do not take effect.
882 You may have forgotten to recompile them into .elc files. Then the
883 old .elc files will be loaded, and your changes will not be seen. To
884 fix this, do `M-x byte-recompile-directory' and specify the directory
885 that contains the Lisp files.
887 Note that you will get a warning when loading a .elc file that is
888 older than the corresponding .el file.
890 *** VM appears to hang in large folders.
892 This is normal (trust us) when upgrading to VM-6.22 from earlier
893 versions. Let VM finish what it is doing and all will be well.
895 *** Starting with 21.4.x, killing text is absurdly slow.
897 See FAQ Q3.10.6. Should be available on the web near
898 http://www.xemacs.org/faq/xemacs-faq.html#SEC160.
900 *** Whenever I try to retrieve a remote file, I have problems.
902 A typical error: FTP Error: USER request failed; 500 AUTH not understood.
903 Thanks to giacomo boffi <giacomo.boffi@polimi.it> on comp.emacs.xemacs:
905 tell your ftp client to not attempt AUTH authentication (or do not
906 use FTP servers that don't understand AUTH)
908 and notes that you need to add an element (often "-u") to
909 `efs-ftp-program-args'. Use M-x customize-variable, and verify the
910 needed flag with `man ftp' or other local documentation.
912 *** gnuserv is running, some clients can connect, but others cannot.
914 The code in gnuslib.c respects the value of TMPDIR. If the server and
915 the client have different values in their environment, you lose.
916 One program known to set TMPDIR and manifest this problem is exmh.
917 You can defeat the use of TMPDIR by unsetting USE_TMPDIR at the top of
918 gnuserv.h at build time.
922 *** You type Control-H (Backspace) expecting to delete characters.
924 Emacs has traditionally used Control-H for help; unfortunately this
925 interferes with its use as Backspace on TTY's. As of XEmacs 21,
926 XEmacs looks at the "erase" setting of TTY structures and maps C-h to
927 backspace when erase is set to C-h. This is sort of a special hack,
928 but it makes it possible for you to use the standard:
932 to get your backspace key to erase characters. The erase setting is
933 recorded in the Lisp variable `tty-erase-char', which you can use to
934 tune the settings in your .emacs.
936 A major drawback of this is that when C-h becomes backspace, it no
937 longer invokes help. In that case, you need to use f1 for help, or
938 bind another key. An example of the latter is the following code,
939 which moves help to Meta-? (ESC ?):
941 (global-set-key "\M-?" 'help-command)
943 *** At startup I get a warning on stderr about missing charsets:
945 Warning: Missing charsets in String to FontSet conversion
947 You need to specify appropriate charsets for your locale (usually the
948 value of the LANG environment variable) in .Xresources. See
949 etc/Emacs.ad for the relevant resources (mostly menubar fonts and
950 fontsets). Do not edit this file, it's purely informative.
952 If you have no satisfactory fonts for iso-8859-1, XEmacs will crash.
954 It looks like XFree86 4.x (the usual server on Linux and *BSD) has
955 some braindamage where .UTF-8 locales will always generate this
956 message, because the XFree86 (font)server doesn't know that UTF-8 will
957 use the ISO10646-1 font registry (or a Cmap or something).
959 If you are not using a .UTF-8 locale and see this warning for a
960 character set not listed in the default in Emacs.ad, please let
961 xemacs-beta@xemacs.org know about it, so we can add fonts to the
962 appropriate fontsets and stifle this warning. (Unfortunately it's
963 buried in Xlib, so we can't easily get rid of it otherwise.)
965 *** Mail agents (VM, Gnus, rmail) cannot get new mail
967 rmail and VM get new mail from /usr/spool/mail/$USER using a program
968 called `movemail'. This program interlocks with /bin/mail using the
969 protocol defined by /bin/mail.
971 There are two different protocols in general use. One of them uses
972 the `flock' system call. The other involves creating a lock file;
973 `movemail' must be able to write in /usr/spool/mail in order to do
974 this. You control which one is used by defining, or not defining, the
975 macro MAIL_USE_FLOCK in config.h or the m- or s- file it includes. IF
976 YOU DON'T USE THE FORM OF INTERLOCKING THAT IS NORMAL ON YOUR SYSTEM,
979 If your system uses the lock file protocol, and fascist restrictions
980 prevent ordinary users from writing the lock files in /usr/spool/mail,
981 you may need to make `movemail' setgid to a suitable group such as
982 `mail'. To do this, use the following commands (as root) after doing
988 Installation normally copies movemail from the build directory to an
989 installation directory which is usually under /usr/local/lib. The
990 installed copy of movemail is usually in the directory
991 /usr/local/lib/emacs/VERSION/TARGET. You must change the group and
992 mode of the installed copy; changing the group and mode of the build
993 directory copy is ineffective.
995 *** Things which should be bold or italic (such as the initial
996 copyright notice) are not.
998 The fonts of the "bold" and "italic" faces are generated from the font
999 of the "default" face; in this way, your bold and italic fonts will
1000 have the appropriate size and family. However, emacs can only be
1001 clever in this way if you have specified the default font using the
1002 XLFD (X Logical Font Description) format, which looks like
1004 *-courier-medium-r-*-*-*-120-*-*-*-*-*-*
1006 if you use any of the other, less strict font name formats, some of
1009 lucidasanstypewriter-12
1013 then emacs won't be able to guess the names of the "bold" and "italic"
1014 versions. All X fonts can be referred to via XLFD-style names, so you
1015 should use those forms. See the man pages for X(1), xlsfonts(1), and
1018 *** The dumped Emacs crashes when run, trying to write pure data.
1020 Two causes have been seen for such problems.
1022 1) On a system where getpagesize is not a system call, it is defined
1023 as a macro. If the definition (in both unexec.c and malloc.c) is wrong,
1024 it can cause problems like this. You might be able to find the correct
1025 value in the man page for a.out (5).
1027 2) Some systems allocate variables declared static among the
1028 initialized variables. Emacs makes all initialized variables in most
1029 of its files pure after dumping, but the variables declared static and
1030 not initialized are not supposed to be pure. On these systems you
1031 may need to add "#define static" to the m- or the s- file.
1033 *** Reading and writing files is very very slow.
1035 Try evaluating the form (setq lock-directory nil) and see if that helps.
1036 There is a problem with file-locking on some systems (possibly related
1037 to NFS) that I don't understand. Please send mail to the address
1038 xemacs-beta@xemacs.org if you figure this one out.
1040 *** When emacs starts up, I get lots of warnings about unknown keysyms.
1042 If you are running the prebuilt binaries, the Motif library expects to find
1043 certain thing in the XKeysymDB file. This file is normally in /usr/lib/X11/
1044 or in /usr/openwin/lib/. If you keep yours in a different place, set the
1045 environment variable $XKEYSYMDB to point to it before starting emacs. If
1046 you still have the problem after doing that, perhaps your version of X is
1047 too old. There is a copy of the MIT X11R5 XKeysymDB file in the emacs `etc'
1048 directory. Try using that one.
1050 *** My X resources used to work, and now some of them are being ignored.
1052 Check the resources in .../etc/Emacs.ad (which is the same as the file
1053 sample.Xresources). Perhaps some of the default resources built in to
1054 emacs are now overriding your existing resources. Copy and edit the
1055 resources in Emacs.ad as necessary.
1057 *** I have focus problems when I use `M-o' to switch to another screen
1058 without using the mouse.
1060 The focus issues with a program like XEmacs, which has multiple
1061 homogeneous top-level windows, are very complicated, and as a result,
1062 most window managers don't implement them correctly.
1064 The R4/R5 version of twm (and all of its descendants) had buggy focus
1065 handling. Sufficiently recent versions of tvtwm have been fixed. In
1066 addition, if you're using twm, make sure you have not specified
1067 "NoTitleFocus" in your .tvtwmrc file. The very nature of this option
1068 makes twm do some illegal focus tricks, even with the patch.
1070 It is known that olwm and olvwm are buggy, and in different ways. If
1071 you're using click-to-type mode, try using point-to-type, or vice
1074 In older versions of NCDwm, one could not even type at XEmacs windows.
1075 This has been fixed in newer versions (2.4.3, and possibly earlier).
1077 (Many people suggest that XEmacs should warp the mouse when focusing
1078 on another screen in point-to-type mode. This is not ICCCM-compliant
1079 behavior. Implementing such policy is the responsibility of the
1080 window manager itself, it is not legal for a client to do this.)
1082 *** Emacs spontaneously displays "I-search: " at the bottom of the screen.
1084 This means that Control-S/Control-Q (XON/XOFF) "flow control" is being
1085 used. C-s/C-q flow control is bad for Emacs editors because it takes
1086 away C-s and C-q as user commands. Since editors do not output long
1087 streams of text without user commands, there is no need for a
1088 user-issuable "stop output" command in an editor; therefore, a
1089 properly designed flow control mechanism would transmit all possible
1090 input characters without interference. Designing such a mechanism is
1091 easy, for a person with at least half a brain.
1093 There are three possible reasons why flow control could be taking place:
1095 1) Terminal has not been told to disable flow control
1096 2) Insufficient padding for the terminal in use
1097 3) Some sort of terminal concentrator or line switch is responsible
1099 First of all, many terminals have a set-up mode which controls whether
1100 they generate XON/XOFF flow control characters. This must be set to
1101 "no XON/XOFF" in order for Emacs to work. Sometimes there is an
1102 escape sequence that the computer can send to turn flow control off
1103 and on. If so, perhaps the termcap `ti' string should turn flow
1104 control off, and the `te' string should turn it on.
1106 Once the terminal has been told "no flow control", you may find it
1107 needs more padding. The amount of padding Emacs sends is controlled
1108 by the termcap entry for the terminal in use, and by the output baud
1109 rate as known by the kernel. The shell command `stty' will print
1110 your output baud rate; `stty' with suitable arguments will set it if
1111 it is wrong. Setting to a higher speed causes increased padding. If
1112 the results are wrong for the correct speed, there is probably a
1113 problem in the termcap entry. You must speak to a local Unix wizard
1114 to fix this. Perhaps you are just using the wrong terminal type.
1116 For terminals that lack a "no flow control" mode, sometimes just
1117 giving lots of padding will prevent actual generation of flow control
1118 codes. You might as well try it.
1120 If you are really unlucky, your terminal is connected to the computer
1121 through a concentrator which sends XON/XOFF flow control to the
1122 computer, or it insists on sending flow control itself no matter how
1123 much padding you give it. Unless you can figure out how to turn flow
1124 control off on this concentrator (again, refer to your local wizard),
1125 you are screwed! You should have the terminal or concentrator
1126 replaced with a properly designed one. In the mean time, some drastic
1127 measures can make Emacs semi-work.
1129 You can make Emacs ignore C-s and C-q and let the operating system
1130 handle them. To do this on a per-session basis, just type M-x
1131 enable-flow-control RET. You will see a message that C-\ and C-^ are
1132 now translated to C-s and C-q. (Use the same command M-x
1133 enable-flow-control to turn *off* this special mode. It toggles flow
1136 If C-\ and C-^ are inconvenient for you (for example, if one of them
1137 is the escape character of your terminal concentrator), you can choose
1138 other characters by setting the variables flow-control-c-s-replacement
1139 and flow-control-c-q-replacement. But choose carefully, since all
1140 other control characters are already used by emacs.
1142 IMPORTANT: if you type C-s by accident while flow control is enabled,
1143 Emacs output will freeze, and you will have to remember to type C-q in
1146 If you work in an environment where a majority of terminals of a
1147 certain type are flow control hobbled, you can use the function
1148 `enable-flow-control-on' to turn on this flow control avoidance scheme
1149 automatically. Here is an example:
1151 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
1153 If this isn't quite correct (e.g. you have a mixture of flow-control hobbled
1154 and good vt200 terminals), you can still run enable-flow-control
1157 I have no intention of ever redesigning the Emacs command set for the
1158 assumption that terminals use C-s/C-q flow control. XON/XOFF flow
1159 control technique is a bad design, and terminals that need it are bad
1160 merchandise and should not be purchased. Now that X is becoming
1161 widespread, XON/XOFF seems to be on the way out. If you can get some
1162 use out of GNU Emacs on inferior terminals, more power to you, but I
1163 will not make Emacs worse for properly designed systems for the sake
1164 of inferior systems.
1166 *** Control-S and Control-Q commands are ignored completely.
1168 For some reason, your system is using brain-damaged C-s/C-q flow
1169 control despite Emacs's attempts to turn it off. Perhaps your
1170 terminal is connected to the computer through a concentrator
1171 that wants to use flow control.
1173 You should first try to tell the concentrator not to use flow control.
1174 If you succeed in this, try making the terminal work without
1175 flow control, as described in the preceding section.
1177 If that line of approach is not successful, map some other characters
1178 into C-s and C-q using keyboard-translate-table. The example above
1179 shows how to do this with C-^ and C-\.
1181 *** Control-S and Control-Q commands are ignored completely on a net
1184 Some versions of rlogin (and possibly telnet) do not pass flow
1185 control characters to the remote system to which they connect.
1186 On such systems, emacs on the remote system cannot disable flow
1187 control on the local system.
1189 One way to cure this is to disable flow control on the local host
1190 (the one running rlogin, not the one running rlogind) using the
1191 stty command, before starting the rlogin process. On many systems,
1192 `stty start u stop u' will do this.
1194 Some versions of tcsh will prevent even this from working. One way
1195 around this is to start another shell before starting rlogin, and
1196 issue the stty command to disable flow control from that shell.
1198 If none of these methods work, the best solution is to type
1199 `M-x enable-flow-control' at the beginning of your emacs session, or
1200 if you expect the problem to continue, add a line such as the
1201 following to your .emacs (on the host running rlogind):
1203 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
1205 See the entry about spontaneous display of I-search (above) for more
1208 *** TTY redisplay is slow.
1210 XEmacs has fairly new TTY redisplay support (beginning from 19.12),
1211 which doesn't include some basic TTY optimizations -- like using
1212 scrolling regions to move around blocks of text. This is why
1213 redisplay on the traditional terminals, or over slow lines can be very
1216 If you are interested in fixing this, please let us know at
1217 <xemacs-beta@xemacs.org>.
1219 *** Screen is updated wrong, but only on one kind of terminal.
1221 This could mean that the termcap entry you are using for that terminal
1222 is wrong, or it could mean that Emacs has a bug handing the
1223 combination of features specified for that terminal.
1225 The first step in tracking this down is to record what characters
1226 Emacs is sending to the terminal. Execute the Lisp expression
1227 (open-termscript "./emacs-script") to make Emacs write all terminal
1228 output into the file ~/emacs-script as well; then do what makes the
1229 screen update wrong, and look at the file and decode the characters
1230 using the manual for the terminal. There are several possibilities:
1232 1) The characters sent are correct, according to the terminal manual.
1234 In this case, there is no obvious bug in Emacs, and most likely you
1235 need more padding, or possibly the terminal manual is wrong.
1237 2) The characters sent are incorrect, due to an obscure aspect of the
1238 terminal behavior not described in an obvious way by termcap.
1240 This case is hard. It will be necessary to think of a way for Emacs
1241 to distinguish between terminals with this kind of behavior and other
1242 terminals that behave subtly differently but are classified the same
1243 by termcap; or else find an algorithm for Emacs to use that avoids the
1244 difference. Such changes must be tested on many kinds of terminals.
1246 3) The termcap entry is wrong.
1248 See the file etc/TERMS for information on changes that are known to be
1249 needed in commonly used termcap entries for certain terminals.
1251 4) The characters sent are incorrect, and clearly cannot be right for
1252 any terminal with the termcap entry you were using.
1254 This is unambiguously an Emacs bug, and can probably be fixed in
1255 termcap.c, terminfo.c, tparam.c, cm.c, redisplay-tty.c,
1256 redisplay-output.c, or redisplay.c.
1258 *** My buffers are full of \000 characters or otherwise corrupt.
1260 Some compilers have trouble with gmalloc.c and ralloc.c; try recompiling
1261 without optimization. If that doesn't work, try recompiling with
1262 SYSTEM_MALLOC defined, and/or with REL_ALLOC undefined.
1264 *** A position you specified in .Xresources is ignored, using twm.
1266 twm normally ignores "program-specified" positions.
1267 You can tell it to obey them with this command in your `.twmrc' file:
1269 UsePPosition "on" #allow clents to request a position
1271 *** With M-x enable-flow-control, you need to type C-\ twice to do
1272 incremental search--a single C-\ gets no response.
1274 This has been traced to communicating with your machine via kermit,
1275 with C-\ as the kermit escape character. One solution is to use
1276 another escape character in kermit. One user did
1278 set escape-character 17
1280 in his .kermrc file, to make C-q the kermit escape character.
1282 *** The Motif version of Emacs paints the screen a solid color.
1284 This has been observed to result from the following X resource:
1286 Emacs*default.attributeFont: -*-courier-medium-r-*-*-*-140-*-*-*-*-iso8859-*
1288 That the resource has this effect indicates a bug in something, but we
1289 do not yet know what. If it is an Emacs bug, we hope someone can
1290 explain what the bug is so we can fix it. In the mean time, removing
1291 the resource prevents the problem.
1293 *** After running emacs once, subsequent invocations crash.
1295 Some versions of SVR4 have a serious bug in the implementation of the
1296 mmap () system call in the kernel; this causes emacs to run correctly
1297 the first time, and then crash when run a second time.
1299 Contact your vendor and ask for the mmap bug fix; in the mean time,
1300 you may be able to work around the problem by adding a line to your
1301 operating system description file (whose name is reported by the
1302 configure script) that reads:
1303 #define SYSTEM_MALLOC
1304 This makes Emacs use memory less efficiently, but seems to work around
1307 *** Inability to send an Alt-modified key, when Emacs is communicating
1308 directly with an X server.
1310 If you have tried to bind an Alt-modified key as a command, and it
1311 does not work to type the command, the first thing you should check is
1312 whether the key is getting through to Emacs. To do this, type C-h c
1313 followed by the Alt-modified key. C-h c should say what kind of event
1314 it read. If it says it read an Alt-modified key, then make sure you
1315 have made the key binding correctly.
1317 If C-h c reports an event that doesn't have the Alt modifier, it may
1318 be because your X server has no key for the Alt modifier. The X
1319 server that comes from MIT does not set up the Alt modifier by
1322 If your keyboard has keys named Alt, you can enable them as follows:
1324 xmodmap -e 'add mod2 = Alt_L'
1325 xmodmap -e 'add mod2 = Alt_R'
1327 If the keyboard has just one key named Alt, then only one of those
1328 commands is needed. The modifier `mod2' is a reasonable choice if you
1329 are using an unmodified MIT version of X. Otherwise, choose any
1330 modifier bit not otherwise used.
1332 If your keyboard does not have keys named Alt, you can use some other
1333 keys. Use the keysym command in xmodmap to turn a function key (or
1334 some other 'spare' key) into Alt_L or into Alt_R, and then use the
1335 commands show above to make them modifier keys.
1337 Note that if you have Alt keys but no Meta keys, Emacs translates Alt
1338 into Meta. This is because of the great importance of Meta in Emacs.
1340 *** In Shell mode, you get a ^M at the end of every line.
1342 This happens to people who use tcsh, because it is trying to be too
1343 smart. It sees that the Shell uses terminal type `unknown' and turns
1344 on the flag to output ^M at the end of each line. You can fix the
1345 problem by adding this to your .cshrc file:
1348 if ($EMACS == "t") then
1350 stty -icrnl -onlcr -echo susp ^Z
1354 *** An error message such as `X protocol error: BadMatch (invalid
1355 parameter attributes) on protocol request 93'.
1357 This comes from having an invalid X resource, such as
1359 (which is invalid because it specifies a color name for something
1360 that isn't a color.)
1362 The fix is to correct your X resources.
1364 *** Once you pull down a menu from the menubar, it won't go away.
1366 It has been claimed that this is caused by a bug in certain very old
1367 (1990?) versions of the twm window manager. It doesn't happen with
1368 recent vintages, or with other window managers.
1370 *** Emacs ignores the "help" key when running OLWM.
1372 OLWM grabs the help key, and retransmits it to the appropriate client
1373 using XSendEvent. Allowing emacs to react to synthetic events is a
1374 security hole, so this is turned off by default. You can enable it by
1375 setting the variable x-allow-sendevents to t. You can also cause fix
1376 this by telling OLWM to not grab the help key, with the null binding
1377 "OpenWindows.KeyboardCommand.Help:".
1379 *** Programs running under terminal emulator do not recognize `emacs'
1382 The cause of this is a shell startup file that sets the TERMCAP
1383 environment variable. The terminal emulator uses that variable to
1384 provide the information on the special terminal type that Emacs
1387 Rewrite your shell startup file so that it does not change TERMCAP
1388 in such a case. You could use the following conditional which sets
1389 it only if it is undefined.
1391 if ( ! ${?TERMCAP} ) setenv TERMCAP ~/my-termcap-file
1393 Or you could set TERMCAP only when you set TERM--which should not
1394 happen in a non-login shell.
1396 *** The popup menu appears at the bottom/right of my screen.
1398 You probably have something like the following in your ~/.Xresources
1400 Emacs.geometry: 81x56--9--1
1402 Use the following instead
1404 Emacs*EmacsFrame.geometry: 81x56--9--1
1406 *** When I try to use the PostgreSQL functions, I get a message about
1409 The only known case in which this happens is if you are using gcc, you
1410 configured with --error-checking=all and --with-modules, and you
1411 compiled with no optimization. If you encounter this problem in any
1412 other situation, please inform xemacs-beta@xemacs.org.
1414 This problem stems from a gcc bug. With no optimization, functions
1415 declared `extern inline' sometimes are not completely compiled away. An
1416 undefined symbol with the function's name is put into the resulting
1417 object file. In this case, when the postgresql module is loaded, the
1418 linker is unable to resolve that symbol, so the module load fails. The
1419 workaround is to recompile the module with optimization turned on. Any
1420 optimization level, including -Os, appears to work.
1422 *** C-z just refreshes the screen instead of suspending Emacs.
1424 You are probably using a shell that doesn't support job control, even
1425 though the system itself is capable of it. Try using a different
1429 *** XEmacs crashes on MacOS within font-lock, or when dealing
1430 with large compilation buffers, or in other regex applications.
1432 The default stack size under MacOS/X is rather small (512k as opposed
1433 to Solaris 8M), hosing the regexp code, which uses alloca()
1434 extensively, overflowing the stack when complex regexps are used.
1437 1) Increase your stack size, using `ulimit -s 8192' or a (t)csh
1440 2) Recompile regex.c with REGEX_MALLOC defined.
1443 *** Your Delete key sends a Backspace to the terminal, using an AIXterm.
1445 The solution is to include in your .Xresources the lines:
1447 *aixterm.Translations: #override <Key>BackSpace: string(0x7f)
1448 aixterm*ttyModes: erase ^?
1450 This makes your Backspace key send DEL (ASCII 127).
1452 *** On AIX 4, some programs fail when run in a Shell buffer
1453 with an error message like No terminfo entry for "unknown".
1455 On AIX, many terminal type definitions are not installed by default.
1456 `unknown' is one of them. Install the "Special Generic Terminal
1457 Definitions" to make them defined.
1459 *** On AIX, you get this message when running Emacs:
1461 Could not load program emacs
1462 Symbol smtcheckinit in csh is undefined
1463 Error was: Exec format error
1467 Could not load program .emacs
1468 Symbol _system_con in csh is undefined
1469 Symbol _fp_trapsta in csh is undefined
1470 Error was: Exec format error
1472 These can happen when you try to run on AIX 3.2.5 a program that was
1473 compiled with 3.2.4. The fix is to recompile.
1475 *** Trouble using ptys on AIX.
1477 People often install the pty devices on AIX incorrectly.
1478 Use `smit pty' to reinstall them properly.
1482 *** The Emacs window disappears when you type M-q.
1484 Some versions of the Open Look window manager interpret M-q as a quit
1485 command for whatever window you are typing at. If you want to use
1486 Emacs with that window manager, you should try to configure the window
1487 manager to use some other command. You can disable the
1488 shortcut keys entirely by adding this line to ~/.OWdefaults:
1490 OpenWindows.WindowMenuAccelerators: False
1492 *** When Emacs tries to ring the bell, you get an error like
1494 audio: sst_open: SETQSIZE" Invalid argument
1495 audio: sst_close: SETREG MMR2, Invalid argument
1497 you have probably compiled using an ANSI C compiler, but with non-ANSI
1498 include files. In particular, on Suns, the file
1499 /usr/include/sun/audioio.h uses the _IOW macro to define the constant
1500 AUDIOSETQSIZE. _IOW in turn uses a K&R preprocessor feature that is
1501 now explicitly forbidden in ANSI preprocessors, namely substitution
1502 inside character constants. All ANSI C compilers must provide a
1503 workaround for this problem. Lucid's C compiler is shipped with a new
1504 set of system include files. If you are using GCC, there is a script
1505 called fixincludes that creates new versions of some system include
1506 files that use this obsolete feature.
1508 *** On Solaris 2.6, XEmacs dumps core when exiting.
1510 This happens if you're XEmacs is running on the same machine as the X
1511 server, and the optimized memory transport has been turned on by
1512 setting the environment variable XSUNTRANSPORT. The crash occurs
1513 during the call to XCloseDisplay.
1515 If this describes your situation, you need to undefine the
1516 XSUNTRANSPORT environment variable.
1518 *** On Solaris, C-x doesn't get through to Emacs when you use the console.
1520 This is a Solaris feature (at least on Intel x86 cpus). Type C-r
1521 C-r C-t, to toggle whether C-x gets through to Emacs.
1523 *** On Solaris 2.4, Dired hangs and C-g does not work. Or Emacs hangs
1524 forever waiting for termination of a subprocess that is a zombie.
1526 casper@fwi.uva.nl says the problem is in X11R6. Rebuild libX11.so
1527 after changing the file xc/config/cf/sunLib.tmpl. Change the lines
1530 #define SharedX11Reqs -lthread
1535 #if OSMinorVersion < 4
1537 #define SharedX11Reqs -lthread
1541 Be sure also to edit x/config/cf/sun.cf so that OSMinorVersion is 4
1542 (as it should be for Solaris 2.4). The file has three definitions for
1543 OSMinorVersion: the first is for x86, the second for SPARC under
1544 Solaris, and the third for SunOS 4. Make sure to update the
1545 definition for your type of machine and system.
1547 Then do `make Everything' in the top directory of X11R6, to rebuild
1548 the makefiles and rebuild X. The X built this way work only on
1549 Solaris 2.4, not on 2.3.
1551 For multithreaded X to work it necessary to install patch
1552 101925-02 to fix problems in header files [2.4]. You need
1553 to reinstall gcc or re-run just-fixinc after installing that
1556 However, Frank Rust <frust@iti.cs.tu-bs.de> used a simpler solution:
1558 #define ThreadedX YES
1560 #define ThreadedX NO
1561 in sun.cf and did `make World' to rebuild X11R6. Removing all
1562 `-DXTHREAD*' flags and `-lthread' entries from lib/X11/Makefile and
1563 typing 'make install' in that directory also seemed to work.
1565 *** On SunOS 4.1.3, Emacs unpredictably crashes in _yp_dobind_soft.
1567 This happens if you configure Emacs specifying just `sparc-sun-sunos4'
1568 on a system that is version 4.1.3. You must specify the precise
1569 version number (or let configure figure out the configuration, which
1570 it can do perfectly well for SunOS).
1572 *** Mail is lost when sent to local aliases.
1574 Many emacs mail user agents (VM and rmail, for instance) use the
1575 sendmail.el library. This library can arrange for mail to be
1576 delivered by passing messages to the /usr/lib/sendmail (usually)
1577 program . In doing so, it passes the '-t' flag to sendmail, which
1578 means that the name of the recipient of the message is not on the
1579 command line and, therefore, that sendmail must parse the message to
1580 obtain the destination address.
1582 There is a bug in the SunOS4.1.1 and SunOS4.1.3 versions of sendmail.
1583 In short, when given the -t flag, the SunOS sendmail won't recognize
1584 non-local (i.e. NIS) aliases. It has been reported that the Solaris
1585 2.x versions of sendmail do not have this bug. For those using SunOS
1586 4.1, the best fix is to install sendmail V8 or IDA sendmail (which
1587 have other advantages over the regular sendmail as well). At the time
1588 of this writing, these official versions are available:
1590 Sendmail V8 on ftp.cs.berkeley.edu in /ucb/sendmail:
1591 sendmail.8.6.9.base.tar.Z (the base system source & documentation)
1592 sendmail.8.6.9.cf.tar.Z (configuration files)
1593 sendmail.8.6.9.misc.tar.Z (miscellaneous support programs)
1594 sendmail.8.6.9.xdoc.tar.Z (extended documentation, with postscript)
1596 IDA sendmail on vixen.cso.uiuc.edu in /pub:
1597 sendmail-5.67b+IDA-1.5.tar.gz
1599 *** Emacs fails to understand most Internet host names, even though
1600 the names work properly with other programs on the same system.
1601 Emacs won't work with X-windows if the value of DISPLAY is HOSTNAME:0.
1602 Gnus can't make contact with the specified host for nntp.
1604 This typically happens on Suns and other systems that use shared
1605 libraries. The cause is that the site has installed a version of the
1606 shared library which uses a name server--but has not installed a
1607 similar version of the unshared library which Emacs uses.
1609 The result is that most programs, using the shared library, work with
1610 the nameserver, but Emacs does not.
1612 The fix is to install an unshared library that corresponds to what you
1613 installed in the shared library, and then relink Emacs.
1615 On SunOS 4.1, simply define HAVE_RES_INIT.
1617 If you have already installed the name resolver in the file libresolv.a,
1618 then you need to compile Emacs to use that library. The easiest way to
1619 do this is to add to config.h a definition of LIBS_SYSTEM, LIBS_MACHINE
1620 or LIB_STANDARD which uses -lresolv. Watch out! If you redefine a macro
1621 that is already in use in your configuration to supply some other libraries,
1622 be careful not to lose the others.
1624 Thus, you could start by adding this to config.h:
1626 #define LIBS_SYSTEM -lresolv
1628 Then if this gives you an error for redefining a macro, and you see that
1629 the s- file defines LIBS_SYSTEM as -lfoo -lbar, you could change config.h
1632 #define LIBS_SYSTEM -lresolv -lfoo -lbar
1634 *** With process-connection-type set to t, each line of subprocess
1635 output is terminated with a ^M, making ange-ftp and GNUS not work.
1637 On SunOS systems, this problem has been seen to be a result of an
1638 incomplete installation of gcc 2.2 which allowed some non-ANSI
1639 compatible include files into the compilation. In particular this
1640 affected virtually all ioctl() calls.
1644 *** XEmacs crashes on startup, in make-frame.
1646 Typically the Lisp backtrace includes
1648 make-frame(nil #<x-device on ":0.0" 0x2558>)
1650 somewhere near the top. The problem is due to an improvement in GNU
1651 ld that sorts the ELF reloc sections in the executable, giving
1652 dramatic speedups in startup for large executables. It also confuses
1653 the traditional unexec code in XEmacs, leading to the core dump. The
1654 solution is to use the --pdump or --ldflags='-z nocombreloc' options
1655 to configure. Recent 21.4 and 12.5 autodetect this in configure.
1657 Red Hat and SuSE (at least) distributed a prerelease version of ld
1658 (versions around 2.11.90.x.y) where autodetection is impossible. The
1659 recommended procedure is to upgrade to binutils >= 2.12 and rerun
1660 configure. Otherwise you must apply the flags by hand. --pdump is
1663 *** I want XEmacs to use the Alt key, not the XXX key, for Meta commands
1665 For historical reasons, XEmacs looks for a Meta key, then an Alt key.
1666 It binds Meta commands to the X11 modifier bit attached to the first
1667 of these it finds. On PCs, the Windows key is often assigned the Meta
1668 bit, but many desktop environments go to great lengths to get all apps
1669 to use the Alt key, and reserve the Windows key to (sensibly enough)
1672 One correct way to implement this was suggested on comp.emacs.xemacs
1673 (by Kilian Foth and in more detail by Michael Piotrowski): unmap the
1674 Meta modifier using xmodmap or xkb, and then map the Meta/Windows key
1675 to the Super or Hyper keysym and an appropriate mod bit. XEmacs will
1676 not find the Meta keysym, and default to using the Alt key for Meta
1677 keybindings. Typically few applications use the (X11) Meta modifier;
1678 it is tedious but not too much so to teach the ones you need to use
1679 Super instead of Meta. There may be further useful hints in the
1680 discussion of keymapping on non-Linux platforms.
1682 *** The color-gcc wrapper
1684 This wrapper colorizes the error messages from gcc. By default XEmacs
1685 does not interpret the escape sequences used to generate colors,
1686 resulting in a cluttered, hard-to-read buffer. You can remove the
1687 wrapper, or defeat the wrapper colorization in Emacs process buffers
1688 by editing the "nocolor" attribute in /etc/colorgccrc:
1690 $ diff -u /etc/colorgccrc.old /etc/colorgccrc
1691 --- /etc/colorgccrc.old Tue Dec 26 02:17:46 2000
1692 +++ /etc/colorgccrc Tue Dec 26 02:15:48 2000
1695 +nocolor: dumb emacs
1697 If you want colorization in your Emacs buffers, you may get good
1698 results from the ansi-color.el library:
1700 http://www.geocities.com/kensanata/color-emacs.html#ansicolors
1702 This is written for the mainline GNU Emacs but the author has made
1703 efforts to adapt it to XEmacs. YMMV.
1705 *** Slow startup on Linux.
1707 People using systems based on the Linux kernel sometimes report that
1708 startup takes 10 to 15 seconds longer than `usual'. There are two
1709 problems, one older, one newer.
1711 **** Old problem: IPv4 host lookup
1713 On older systems, this is because Emacs looks up the host name when it
1714 starts. Normally, this takes negligible time; the extra delay is due
1715 to improper system configuration. (Recent Linux distros usually have
1716 this configuration correct "out of the box".) This problem can occur
1717 for both networked and non-networked machines.
1719 Here is how to fix the configuration. It requires being root.
1721 ***** Networked Case
1723 First, make sure the files `/etc/hosts' and `/etc/host.conf' both
1724 exist. The first line in the `/etc/hosts' file should look like this
1725 (replace HOSTNAME with your host name):
1727 127.0.0.1 localhost HOSTNAME
1729 Also make sure that the `/etc/host.conf' files contains the following
1735 Any changes, permanent and temporary, to the host name should be
1736 indicated in the `/etc/hosts' file, since it acts a limited local
1737 database of addresses and names (e.g., some SLIP connections
1738 dynamically allocate ip addresses).
1740 ***** Non-Networked Case
1742 The solution described in the networked case applies here as well.
1743 However, if you never intend to network your machine, you can use a
1744 simpler solution: create an empty `/etc/host.conf' file. The command
1745 `touch /etc/host.conf' suffices to create the file. The `/etc/hosts'
1746 file is not necessary with this approach.
1748 **** New problem: IPv6 CNAME lookup
1750 A newer problem is due to XEmacs changing to use the modern
1751 getaddrinfo() interface from the older gethostbyname() interface. The
1752 solution above is insufficient, because getaddrinfo() by default tries
1753 to get IPv6 information for localhost. This always involves a dns
1754 lookup to get the CNAME, and the strategies above don't work. It then
1755 falls back to IPv4 behavior. This is good[tm] according the people at
1756 WIDE who know about IPv6.
1758 ***** Robust network case
1760 Configure your network so that there are no nameservers configured
1761 until the network is actually running. getaddrinfo() will not try to
1762 access a nameserver that isn't configured.
1764 ***** Flaky network case
1766 If you have a flaky modem or DSL connection that can be relied on only
1767 to go down whenever you want to bring XEmacs up, you need to force
1768 IPv4 behavior. Explicitly setting DISPLAY=127.0.0.1:0.0 (or whatever
1769 is appropriate) works in most cases.
1771 If you cannot or do not want to do that, you can hard code IPv4
1772 behavior in src/process-unix.c. This is bad[tm], on your own head be
1773 it. Use the configure option `--with-ipv6-cname=no'.
1777 The Mandrake Linux distribution is attempting to comprehensively
1778 update the user interface, and make it consistent across
1779 applications. This is very difficult, and will occasionally cause
1780 conflicts with applications like Emacs with their own long-established
1781 interfaces. Known issues specific to Mandrake or especially common:
1783 Some versions of XEmacs (21.1.9 is known) distributed with Mandrake
1784 were patched to make the Meta and Alt keysyms synonymous. These
1785 normally work as expected in the Mandrake environment. However,
1786 custom-built XEmacsen (including all 21.2 betas) will "inexplicably"
1787 not respect the "Alt-invokes-Meta-commands" convention. See "I want
1788 XEmacs to use the Alt key" below.
1790 The color-gcc wrapper (see below) is in common use on the Mandrake
1793 *** You get crashes in a non-C locale with Linux GNU Libc 2.0.
1795 Internationalization was not the top priority for GNU Libc 2.0.
1796 As of this writing (1998-12-28) you may get crashes while running
1797 XEmacs in a non-C locale. For example, `LC_ALL=en_US xemacs' crashes
1798 while `LC_ALL=C xemacs' runs fine. This happens for example with GNU
1799 libc 2.0.7. Installing libintl.a and libintl.h built from gettext
1800 0.10.35 and re-building XEmacs solves the crashes. Presumably soon
1801 everyone will upgrade to GNU Libc 2.1 and this problem will go away.
1803 *** `C-z', or `M-x suspend-emacs' hangs instead of suspending.
1805 If you build with `gpm' support on Linux, you cannot suspend XEmacs
1806 because gpm installs a buggy SIGTSTP handler. Either compile with
1807 `--with-gpm=no', or don't suspend XEmacs on the Linux console until
1810 *** With certain fonts, when the cursor appears on a character, the
1811 character doesn't appear--you get a solid box instead.
1813 One user on a Linux system reported that this problem went away with
1814 installation of a new X server. The failing server was XFree86 3.1.1.
1815 XFree86 3.1.2 works.
1818 *** On Irix, I don't see the toolbar icons and I'm getting lots of
1819 entries in the warnings buffer.
1821 SGI ships a really old Xpm library in /usr/lib which does not work at
1822 all well with XEmacs. The solution is to install your own copy of the
1823 latest version of Xpm somewhere and then use the --site-includes and
1824 --site-libraries flags to tell configure where to find it.
1826 *** Trouble using ptys on IRIX, or running out of ptys.
1828 The program mkpts (which may be in `/usr/adm' or `/usr/sbin') needs to
1829 be set-UID to root, or non-root programs like Emacs will not be able
1830 to allocate ptys reliably.
1832 *** Beware of the default image & graphics library on Irix
1834 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
1836 You *have* to compile your own jpeg lib. The one delivered with SGI
1837 systems is a C++ lib, which apparently XEmacs cannot cope with.
1840 ** Digital UNIX/OSF/VMS/Ultrix
1841 *** XEmacs crashes on Digital Unix within font-lock, or when dealing
1842 with large compilation buffers, or in other regex applications.
1844 The default stack size under Digital Unix is rather small (2M as
1845 opposed to Solaris 8M), hosing the regexp code, which uses alloca()
1846 extensively, overflowing the stack when complex regexps are used.
1849 1) Increase your stack size, using `ulimit -s 8192' or a (t)csh
1852 2) Recompile regex.c with REGEX_MALLOC defined.
1854 *** The `Alt' key doesn't behave as `Meta' when running DECwindows.
1856 The default DEC keyboard mapping has the Alt keys set up to generate the
1857 keysym `Multi_key', which has a meaning to xemacs which is distinct from that
1858 of the `Meta_L' and `Meta-R' keysyms. A second problem is that certain keys
1859 have the Mod2 modifier attached to them for no adequately explored reason.
1860 The correct fix is to pass this file to xmodmap upon starting X:
1863 keysym Multi_key = Alt_L
1867 *** The Compose key on a DEC keyboard does not work as Meta key.
1869 This shell command should fix it:
1871 xmodmap -e 'keycode 0xb1 = Meta_L'
1873 *** `expand-file-name' fails to work on any but the machine you dumped
1876 On Ultrix, if you use any of the functions which look up information
1877 in the passwd database before dumping Emacs (say, by using
1878 expand-file-name in site-init.el), then those functions will not work
1879 in the dumped Emacs on any host but the one Emacs was dumped on.
1881 The solution? Don't use expand-file-name in site-init.el, or in
1882 anything it loads. Yuck - some solution.
1884 I'm not sure why this happens; if you can find out exactly what is
1885 going on, and perhaps find a fix or a workaround, please let us know.
1886 Perhaps the YP functions cache some information, the cache is included
1887 in the dumped Emacs, and is then inaccurate on any other host.
1891 *** I get complaints about the mapping of my HP keyboard at startup,
1892 but I haven't changed anything.
1894 The default HP keymap is set up to have Mod1 assigned to two different keys:
1895 Meta_L and Mode_switch (even though there is not actually a Mode_switch key on
1896 the keyboard -- it uses an "imaginary" keycode.) There actually is a reason
1897 for this, but it's not a good one. The correct fix is to execute this command
1900 xmodmap -e 'remove mod1 = Mode_switch'
1902 *** On HP-UX, you get "poll: Interrupted system call" message in the
1903 window where XEmacs was launched.
1905 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
1907 I get a very strange problem when linking libc.a dynamically: every
1908 event (mouse, keyboard, expose...) results in a "poll: Interrupted
1909 system call" message in the window where XEmacs was
1910 launched. Forcing a static link of libc.a alone by adding
1911 /usr/lib/libc.a at the end of the link line solves this. Note that
1912 my 9.07 build of 19.14b17 and my (old) build of 19.13 both exhibit
1913 the same behavior. I've tried various hpux patches to no avail. If
1914 this problem cannot be solved before the release date, binary kits
1915 for HP *must* be linked statically against libc, otherwise this
1916 problem will show up. (This is directed at whoever will volunteer
1917 for this kit, as I won't be available to do it, unless 19.14 gets
1918 delayed until mid-june ;-). I think this problem will be an FAQ soon
1919 after the release otherwise.
1921 Note: The above entry is probably not valid for XEmacs 21.0 and
1924 *** The right Alt key works wrong on German HP keyboards (and perhaps
1925 other non-English HP keyboards too).
1927 This is because HP-UX defines the modifiers wrong in X. Here is a
1928 shell script to fix the problem; be sure that it is run after VUE
1929 configures the X server.
1931 xmodmap 2> /dev/null - << EOF
1932 keysym Alt_L = Meta_L
1933 keysym Alt_R = Meta_R
1938 keysym Mode_switch = NoSymbol
1940 keysym Meta_R = Mode_switch
1941 add mod2 = Mode_switch
1945 *** XEmacs dumps core at startup when native audio is used. Native
1946 audio does not work with recent versions of HP-UX.
1948 Under HP-UX 10.20 and later (e.g., HP-UX 11.XX), with native audio
1949 enabled, the dumped XEmacs binary ("xemacs") core dumps at startup if
1950 recent versions of the libAlib.sl audio shared library is used. Note
1951 that "temacs" will run, but "xemacs" will dump core. This, of course,
1952 causes the XEmacs build to fail. If GNU malloc is enabled, a stack
1953 trace will show XEmacs to have crashed in the "first" call to malloc().
1955 This bug currently exists in all versions of XEmacs, when the undump
1956 mechanism is used. It is not known if using the experimental portable
1957 dumper will allow native audio to work.
1961 Recent versions of the HP-UX 10.20 (and later) audio shared library (in
1962 /opt/audio/lib), pulls in the libdce shared library, which pulls in a
1963 thread (libcma) library. This prevents the HP-UX undump() routine (in
1964 unexhp9k800.c) from properly working. What's happening is that some
1965 initialization routines are being called in the libcma library, *BEFORE*
1966 main() is called, and these initialization routines are calling
1967 malloc(). Unfortunately, in order for the undumper to work, XEmacs must
1968 adjust (move upwards) the sbrk() value *BEFORE* the first call to
1969 malloc(); if malloc() is called before XEmacs has properly adjusted sbrk
1970 (which is what is happening), dumped memory that is being used by
1971 XEmacs, is improperly re-allocated for use by malloc() and the dumped
1972 memory is corrupted. This causes XEmacs to die an horrible death.
1974 It is believed that versions of the audio library past December 1998
1975 will trigger this problem. Under HP-UX 10.20, you probably have to
1976 install audio library patches to encounter this. It's probable that
1977 recent "fresh, out-of-the-box" HP-UX 11.XX workstations also have this
1978 problem. For HP-UX 10.20, it's believed that audio patch PHSS_17121 (or
1979 a superceeding one, like PHSS_17554, PHSS_17971, PHSS_18777, PHSS_21481,
1980 or PHSS_21662, etc.) will trigger this.
1982 To check if your audio library will cause problems for XEmacs, run
1983 "chatr /opt/audio/lib/libAlib.sl". If "libdce" appears in the displayed
1984 shared library list, XEmacs will probably encounter problems if audio is
1989 Don't enable native audio. Re-run configure without native audio
1992 If your site supports it, try using NAS (Network Audio Support).
1994 Try using the experimental portable dumper. It may work, or it may
1998 *** `Pid xxx killed due to text modification or page I/O error'
2000 On HP-UX, you can get that error when the Emacs executable is on an NFS
2001 file system. HP-UX responds this way if it tries to swap in a page and
2002 does not get a response from the server within a timeout whose default
2003 value is just ten seconds.
2005 If this happens to you, extend the timeout period.
2007 *** Shell mode on HP-UX gives the message, "`tty`: Ambiguous".
2009 christos@theory.tn.cornell.edu says:
2011 The problem is that in your .cshrc you have something that tries to
2012 execute `tty`. If you are not running the shell on a real tty then tty
2013 will print "not a tty". Csh expects one word in some places, but tty
2014 is giving it back 3.
2016 The solution is to add a pair of quotes around `tty` to make it a
2019 if (`tty` == "/dev/console")
2021 should be changed to:
2023 if ("`tty`" == "/dev/console")
2025 Even better, move things that set up terminal sections out of .cshrc
2030 *** Regular expressions matching bugs on SCO systems.
2032 On SCO, there are problems in regexp matching when Emacs is compiled
2033 with the system compiler. The compiler version is "Microsoft C
2034 version 6", SCO 4.2.0h Dev Sys Maintenance Supplement 01/06/93; Quick
2035 C Compiler Version 1.00.46 (Beta). The solution is to compile with
2040 *** Conflicts with FSF NTEmacs
2042 Depending on how it is installed, FSF NTEmacs may setup various EMACS*
2043 variables in your environment. The presence of these variables may
2044 cause XEmacs to fail at startup, cause you to see corrupted
2045 doc-strings, or cause other random problems.
2047 You should remove these variables from your environment. These
2048 variables are not required to run FSF NTEmacs if you start it by
2051 *** XEmacs can't find my init file
2053 XEmacs looks for your init in your "home" directory -- either in
2054 `~/.xemacs/init.el' or `~/.emacs'. XEmacs decides that your "home"
2055 directory is, in order of preference:
2057 - The value of the HOME environment variable, if the variable exists.
2058 - The value of the registry entry SOFTWARE\XEmacs\XEmacs\HOME,
2060 - The value of the HOMEDRIVE and HOMEPATH environment variables, if
2061 these variables both exist.
2064 To determine what XEmacs thinks your home directory is, try opening
2065 a file in the `~' directory, and you should see its expansion in the
2066 modeline. If this doesn't work, type ESC : (user-home-directory).
2068 *** XEmacs can't find any packages
2070 XEmacs looks for your packages in subdirectories of a directory which
2071 is set at compile-time (see `config.inc'), and whose default is
2072 `C:\Program Files\XEmacs'. XEmacs also looks in `~/.xemacs', where
2073 `~' refers to your home directory (see previous entry). The variable
2074 `configure-package-path' holds the actual path that was compiled into
2075 your copy of XEmacs.
2077 The compile-time default location can be overridden by the EMACSPACKAGEPATH
2078 environment variable or by the SOFTWARE\XEmacs\XEmacs\EMACSPACKAGEPATH
2079 registry entry. You should check that these variables, if they exist,
2080 point to the actual location of your package tree.
2082 *** XEmacs doesn't die when shutting down Windows 95 or 98
2084 When shutting down Windows 95 or 98 you may see a dialog that says
2085 "xemacs / You must quit this program before you quit Windows".
2087 "Click OK to quit the program and Windows",
2088 but you won't be offered a chance to save any modified XEmacs buffers.
2092 The C-z, C-x, C-c, and C-v keystrokes have traditional uses in both
2093 emacs and Windows programs. XEmacs binds these keys to their
2094 traditional emacs uses, and provides Windows 3.x style bindings for
2095 the Cut, Copy and Paste functions.
2097 Function XEmacs binding
2098 -------- --------------
2104 You can rebind keys to make XEmacs more Windows-compatible; for
2105 example, to bind C-z to undo:
2107 (global-set-key [(control z)] 'undo)
2109 Rebindind C-x and C-c is trickier because by default these are prefix
2110 keys in XEmacs. See the "Key Bindings" node in the XEmacs manual.
2112 *** Behavior of selected regions
2114 Use the pending-del package to enable the standard Windows behavior of
2115 self-inserting deletes region.
2117 *** Limitations on the use of the AltGr key.
2119 In some locale and OS combinations you can't generate M-AltGr-key or
2120 C-M-AltGr-key sequences at all.
2122 To generate C-AltGr-key or C-M-AltGr-key sequences you must use the
2123 right-hand Control key and you must press it *after* AltGr.
2125 These limitations arise from fundamental problems in the way that the
2126 win32 API reports AltGr key events. There isn't anything that XEmacs
2127 can do to work round these problems that it isn't already doing.
2129 You may want to create alternative bindings if any of the standard
2130 XEmacs bindings require you to use some combination of Control or Meta
2133 *** Limited support for subprocesses under Windows 9x
2135 Attempting to use call-process to run a 16bit program gives a
2136 "Spawning child process: Exec format error". For example shell-command
2137 fails under Windows 95 and 98 if you use command.com or any other
2138 16bit program as your shell.
2140 XEmacs may incorrectly quote your call-process command if it contains
2141 double quotes, backslashes or spaces.
2143 start-process and functions that rely on it are supported under Windows 95,
2144 98 and NT. However, starting a 16bit program that requires keyboard input
2145 may cause XEmacs to hang or crash under Windows 95 and 98, and will leave
2146 the orphaned 16bit program consuming all available CPU time.
2148 Sending signals to subprocesses started by call-process or by
2149 start-process fails with a "Cannot send signal to process" error under
2150 Windows 95 and 98. As a side effect of this, quitting XEmacs while it
2151 is still running subprocesses causes it to crash under Windows 95 and
2156 *** Signal 11 when building or running a dumped XEmacs.
2158 See the section on Cygwin above, under building.
2160 *** XEmacs fails to start because cygXpm-noX4.dll was not found.
2162 Andy Piper <andy@xemacs.org> sez:
2164 cygXpm-noX4 is part of the cygwin distribution under libraries or
2165 graphics, but is not installed by default. You need to run the
2166 cygwin setup again and select this package.
2168 *** Subprocesses do not work.
2170 You do not have "tty" in your CYGWIN environment variable. This must
2171 be set in your autoexec.bat (win95) or the system properties (winnt)
2172 as it must be read before the cygwin DLL initializes.
2174 *** ^G does not work on hung subprocesses.
2176 This is a known problem. It can be remedied by defining BROKEN_SIGIO
2177 in src/s/cygwin.h, however this currently leads to instability in XEmacs.
2178 (#### is this still true?)
2180 *** Errors from make like `/c:not found' when running `M-x compile'.
2182 Make sure you set the environment variable MAKE_MODE to UNIX in your
2183 init file (.xemacs/init.el), Control Panel (Windows 2000/NT), or
2184 AUTOEXEC.BAT (Windows 98/95).
2186 *** There are no images in the toolbar buttons.
2188 You need version 4.71 of commctrl.dll which does not ship with windows
2189 95. You can get this by installing IE 4.0 or downloading it from the
2193 * Compatibility problems (with Emacs 18, GNU Emacs, or previous XEmacs/lemacs)
2194 ==============================================================================
2196 *** "Symbol's value as variable is void: unread-command-char".
2197 "Wrong type argument: arrayp, #<keymap 143 entries>"
2198 "Wrong type argument: stringp, [#<keypress-event return>]"
2200 There are a few incompatible changes in XEmacs, and these are the
2201 symptoms. Some of the emacs-lisp code you are running needs to be
2202 updated to be compatible with XEmacs.
2204 The code should not treat keymaps as arrays (use `define-key', etc.),
2205 should not use obsolete variables like `unread-command-char' (use
2206 `unread-command-events'). Many (most) of the new ways of doing things
2207 are compatible in GNU Emacs and XEmacs.
2209 Modern Emacs packages (Gnus, VM, W3, efs, etc) are written to support
2210 GNU Emacs and XEmacs. We have provided modified versions of several
2211 popular emacs packages (dired, etc) which are compatible with this
2212 version of emacs. Check to make sure you have not set your load-path
2213 so that your private copies of these packages are being found before
2214 the versions in the lisp directory.
2216 Make sure that your load-path and your $EMACSLOADPATH environment
2217 variable are not pointing at an Emacs18 lisp directory. This will
2220 ** Some packages that worked before now cause the error
2221 Wrong type argument: arrayp, #<face ... >
2223 Code which uses the `face' accessor functions must be recompiled with
2224 xemacs 19.9 or later. The functions whose callers must be recompiled
2225 are: face-font, face-foreground, face-background,
2226 face-background-pixmap, and face-underline-p. The .elc files
2227 generated by version 19.9 will work in 19.6 and 19.8, but older .elc
2228 files which contain calls to these functions will not work in 19.9.
2230 ** Signaling: (error "Byte code stack underflow (byte compiler bug), pc 38")
2232 This error is given when XEmacs 20 is compiled without MULE support
2233 but is attempting to load a .elc which requires MULE support. The fix
2234 is to rebytecompile the offending file.
2236 ** Signaling: (wrong-type-argument ...) when loading mail-abbrevs
2238 The is seen when installing the Insidious Big Brother Data Base (bbdb)
2239 which includes an outdated copy of mail-abbrevs.el. Remove the copy
2240 that comes with bbdb and use the one that comes with XEmacs.
2246 ** A reminder: XEmacs/Mule work does not currently receive *any*
2247 funding, and all work is done by volunteers. If you think you can
2248 help, please contact the XEmacs maintainers.
2250 ** XEmacs/Mule doesn't support TTY's satisfactorily.
2252 This is a major problem, which we plan to address in a future release
2253 of XEmacs. Basically, XEmacs should have primitives to be told
2254 whether the terminal can handle international output, and which
2255 locale. Also, it should be able to do approximations of characters to
2256 the nearest supported by the locale.
2258 ** Internationalized (Asian) Isearch doesn't work.
2260 Currently, Isearch doesn't directly support any of the input methods
2261 that are not XIM based (like egg, canna and quail) (and there are
2262 potential problems with XIM version too...). If you're using egg
2263 there is a workaround. Hitting <RET> right after C-s to invoke
2264 Isearch will put Isearch in string mode, where a complete string can
2265 be typed into the minibuffer and then processed by Isearch afterwards.
2266 Since egg is now supported in the minibuffer using string mode you can
2267 now use egg to input your Japanese, Korean or Chinese string, then hit
2268 return to send that to Isearch and then use standard Isearch commands
2271 ** Using egg and mousing around while in 'fence' mode screws up my
2274 Don't do this. The fence modes of egg and canna are currently very
2275 modal, and messing with where they expect point to be and what they
2276 think is the current buffer is just asking for trouble. If you're
2277 lucky they will realize that something is awry, and simply delete the
2278 fence, but worst case can trash other buffers too. We've tried to
2279 protect against this where we can, but there still are many ways to
2280 shoot yourself in the foot. So just finish what you are typing into
2281 the fence before reaching for the mouse.
2283 ** Not all languages in Quail are supported like Devanagari and Indian
2284 languages, Lao and Tibetan.
2286 Quail requires more work and testing. Although it has been ported to
2287 XEmacs, it works really well for Japanese and for the European
2290 ** Right-to-left mode is not yet implemented, so languages like
2291 Arabic, Hebrew and Thai don't work.
2293 Getting this right requires more work. It may be implemented in a
2294 future XEmacs version, but don't hold your breath. If you know
2295 someone who is ready to implement this, please let us know.
2297 ** We need more developers and native language testers. It's extremely
2298 difficult (and not particularly productive) to address languages that
2299 nobody is using and testing.
2301 ** The kWnn and cWnn support for Chinese and Korean needs developers
2302 and testers. It probably doesn't work.
2304 ** There are no `native XEmacs' TUTORIALs for any Asian languages,
2305 including Japanese. FSF Emacs and XEmacs tutorials are quite similar,
2306 so it should be sufficient to skim through the differences and apply
2307 them to the Japanese version.
2309 ** We only have localized menus translated for Japanese, and the
2310 Japanese menus are developing bitrot (the Mule menu appears in
2313 ** XIM is untested for any language other than Japanese.