|
44
|
1 -*- mode:outline; minor-mode:outl-mouse -*-
|
|
0
|
2 This file describes various problems that have been encountered
|
|
|
3 in compiling, installing and running XEmacs.
|
|
|
4
|
|
|
5 (synched up with: 19.30)
|
|
|
6
|
|
|
7 * On Irix, I don't see the toolbar icons and I'm getting lots of
|
|
|
8 entries in the warnings buffer.
|
|
|
9
|
|
|
10 SGI ships a really old Xpm library in /usr/lib which does not work at
|
|
|
11 all well with XEmacs. The solution is to install your own copy of the
|
|
|
12 latest version of Xpm somewhere and then use the --site-includes and
|
|
|
13 --site-libraries flags to tell configure where to find it.
|
|
|
14
|
|
|
15 * On Digital UNIX, the DEC C compiler might have a problem compiling
|
|
|
16 some files.
|
|
|
17
|
|
|
18 In particular, src/extents.c and src/faces.c might cause the DEC C
|
|
|
19 compiler to abort. When this happens: cd src, compile the files by
|
|
|
20 hand, cd .., and redo the "make" command. When recompiling the files by
|
|
|
21 hand, use the old C compiler for the following versions of Digital UNIX:
|
|
|
22 - V3.n: Remove "-migrate" from the compile command.
|
|
|
23 - V4.n: Add "-oldc" to the compile command.
|
|
|
24
|
|
|
25 * On HPUX, the HP C compiler might have a problem compiling some files
|
|
|
26 with optimization.
|
|
|
27
|
|
|
28 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
|
|
|
29
|
|
|
30 Had to drop once again to level 2 optimization, at least to
|
|
|
31 compile lstream.c. Otherwise, I get a "variable is void: \if"
|
|
|
32 problem while dumping (this is a problem I already reported
|
|
|
33 with vanilla hpux 10.01 and 9.07, which went away after
|
|
|
34 applying patches for the C compiler). Trouble is I still
|
|
|
35 haven't found the same patch for hpux 10.10, and I don't
|
|
|
36 remember the patch numbers. I think potential XEmacs builders
|
|
|
37 on HP should be warned about this.
|
|
|
38
|
|
|
39 * On HPUX, you get "poll: Interrupted system call" message in the window
|
|
|
40 where XEmacs was launched.
|
|
|
41
|
|
|
42 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
|
|
|
43
|
|
|
44 I get a very strange problem when linking libc.a
|
|
|
45 dynamically: every event (mouse, keyboard, expose...) results
|
|
|
46 in a "poll: Interrupted system call" message in the window
|
|
|
47 where XEmacs was launched. Forcing a static link of libc.a
|
|
|
48 alone by adding /usr/lib/libc.a at the end of the link line
|
|
|
49 solves this. Note that my 9.07 build of 19.14b17 and my (old)
|
|
|
50 build of 19.13 both exhibit the same behaviour. I've tried
|
|
|
51 various hpux patches to no avail. If this problem cannot be
|
|
|
52 solved before the release date, binary kits for HP *must* be
|
|
|
53 linked statically against libc, otherwise this problem will
|
|
|
54 show up. (This is directed at whoever will volunteer for this
|
|
|
55 kit, as I won't be available to do it, unless 19.14 gets
|
|
|
56 delayed until mid-june ;-). I think this problem will be an FAQ
|
|
|
57 soon after the release otherwise.
|
|
|
58
|
|
|
59 * Native cc on SCO OpenServer 5 is now OK. Icc may still throw you
|
|
|
60 a curve. Here is what Robert Lipe <robertl@arnet.com> says:
|
|
|
61
|
|
|
62 Unlike XEmacs 19.13, building with the native cc on SCO OpenServer 5
|
|
|
63 now produces a functional binary. I will typically build this
|
|
|
64 configuration for COFF with:
|
|
|
65
|
|
|
66 /path_to_XEmacs_source/configure --with-gcc=no \
|
|
|
67 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
|
|
|
68 --with-xpm --with-xface --with-sound=nas
|
|
|
69
|
|
|
70 This version now supports ELF builds. I highly recommend this to
|
|
|
71 reduce the in-core footprint of XEmacs. This is now how I compile
|
|
|
72 all my test releases. Build it like this:
|
|
|
73
|
|
|
74 /path_to_XEmacs_source/configure --with-gcc=no \
|
|
|
75 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
|
|
|
76 --with-xpm --with-xface --with-sound=nas --dynamic
|
|
|
77
|
|
|
78 The compiler known as icc [ supplied with the OpenServer 5 Development
|
|
|
79 System ] generates a working binary, but it takes forever to generate
|
|
|
80 XEmacs. ICC also whines more about the code than /bin/cc does. I do
|
|
|
81 believe all its whining is legitimate, however. Note that you do
|
|
|
82 have to 'cd src ; make LD=icc' to avoid linker errors.
|
|
|
83
|
|
|
84 The way I handle the build procedure is:
|
|
|
85
|
|
|
86 /path_to_XEmacs_source/configure --with-gcc=no \
|
|
|
87 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
|
|
|
88 --with-xpm --with-xface --with-sound=nas --dynamic --compiler="icc"
|
|
|
89
|
|
|
90 *NOTE* I have the xpm, xface, and audio libraries and includes in
|
|
|
91 /usr/local/lib, /usr/local/include. If you don't have these,
|
|
|
92 don't include the "--with-*" arguments in any of my examples.
|
|
|
93
|
|
|
94 In previous versions of XEmacs, you had to override the defaults while
|
|
|
95 compiling font-lock.o and extents.o when building with icc. This seems
|
|
|
96 to no longer be true, but I'm including this old information in case it
|
|
|
97 resurfaces. The process I used was:
|
|
|
98
|
|
|
99 make -k
|
|
|
100 [ procure pizza, beer, repeat ]
|
|
|
101 cd src
|
|
|
102 make CC="icc -W0,-mP1COPT_max_tree_size=3000" font-lock.o extents.o
|
|
|
103 make LD=icc
|
|
|
104
|
|
|
105 If you want sound support, get the tls566 supplement from
|
|
|
106 ftp.sco.com:/TLS or any of its mirrors. It works just groovy
|
|
|
107 with XEmacs.
|
|
|
108
|
|
|
109 The M-x manual-entry is known not to work. If you know Lisp and would
|
|
|
110 like help in making it work, e-mail me at <robertl@dgii.com>
|
|
|
111
|
|
|
112 In earlier releases, gnuserv/gnuclient/gnudoit would open a frame
|
|
|
113 just fine, but the client would lock up and the server would
|
|
|
114 terminate when you used C-x # to close the frame. This is now
|
|
|
115 fixed in XEmacs.
|
|
|
116
|
|
|
117 In etc/ there are two files of note. emacskeys.sco and emacsstrs.sco.
|
|
|
118 The comments at the top of emacskeys.sco describe its function, and
|
|
|
119 the emacstrs.sco is a suitable candidate for /usr/lib/keyboard/strings
|
|
|
120 to take advantage of the keyboard map in emacskeys.sco.
|
|
|
121
|
|
|
122 * Don't use -O2 with gcc under Linux without also using
|
|
|
123 -fno-strength-reduce. gcc will generate incorrect code otherwise.
|
|
|
124 This bug is present in at least 2.6.x and 2.7.[0-2]. A patched
|
|
|
125 binary for 2.7.2 is available in
|
|
|
126
|
|
|
127 ftp://tsx-11.mit.edu/pub/linux/packages/GCC/gcc272-no-sr-bug.lbin.tgz
|
|
|
128
|
|
|
129 Or wait for GCC 2.7.3.
|
|
|
130
|
|
|
131 * Under some versions of OSF XEmacs runs fine if built without
|
|
|
132 optimization but will crash randomly if built with optimization.
|
|
|
133 Using 'cc -g' is not sufficient to eliminate all optimization. Try
|
|
|
134 'cc -g -O0' instead.
|
|
|
135
|
|
|
136 * On HP/UX configure selects gcc even though it isn't actually present.
|
|
|
137
|
|
|
138 Some versions of SoftBench have an executable called 'gcc' that is not
|
|
|
139 actually the GNU C compiler. Use the --with-gcc=no flag when running
|
|
|
140 configure.
|
|
|
141
|
|
|
142
|
|
|
143 * When Emacs tries to ring the bell, you get an error like
|
|
|
144
|
|
|
145 audio: sst_open: SETQSIZE" Invalid argument
|
|
|
146 audio: sst_close: SETREG MMR2, Invalid argument
|
|
|
147
|
|
|
148 you have probably compiled using an ANSI C compiler, but with non-ANSI include
|
|
|
149 files. In particular, on Suns, the file /usr/include/sun/audioio.h uses the
|
|
|
150 _IOW macro to define the constant AUDIOSETQSIZE. _IOW in turn uses a K&R
|
|
|
151 preprocessor feature that is now explicitly forbidden in ANSI preprocessors,
|
|
|
152 namely substitution inside character constants. All ANSI C compilers must
|
|
|
153 provide a workaround for this problem. Lucid's C compiler is shipped with a
|
|
|
154 new set of system include files. If you are using GCC, there is a script
|
|
|
155 called fixincludes that creates new versions of some system include files that
|
|
|
156 use this obsolete feature.
|
|
|
157
|
|
|
158 * The `Alt' key doesn't behave as `Meta' when running DECwindows.
|
|
|
159
|
|
|
160 The default DEC keyboard mapping has the Alt keys set up to generate the
|
|
|
161 keysym `Multi_key', which has a meaning to xemacs which is distinct from that
|
|
|
162 of the `Meta_L' and `Meta-R' keysyms. A second problem is that certain keys
|
|
|
163 have the Mod2 modifier attached to them for no adequately explored reason.
|
|
|
164 The correct fix is to pass this file to xmodmap upon starting X:
|
|
|
165
|
|
|
166 clear mod2
|
|
|
167 keysym Multi_key = Alt_L
|
|
|
168 add mod1 = Alt_L
|
|
|
169 add mod1 = Alt_R
|
|
|
170
|
|
|
171 * I get complaints about the mapping of my HP keyboard at startup, but I
|
|
|
172 haven't changed anything.
|
|
|
173
|
|
|
174 The default HP keymap is set up to have Mod1 assigned to two different keys:
|
|
|
175 Meta_L and Mode_switch (even though there is not actually a Mode_switch key on
|
|
|
176 the keyboard -- it uses an "imaginary" keycode.) There actually is a reason
|
|
|
177 for this, but it's not a good one. The correct fix is to execute this command
|
|
|
178 upon starting X:
|
|
|
179
|
|
|
180 xmodmap -e 'remove mod1 = Mode_switch'
|
|
|
181
|
|
|
182 * I have focus problems when I use `M-o' to switch to another screen without
|
|
|
183 using the mouse.
|
|
|
184
|
|
|
185 The focus issues with a program like XEmacs, which has multiple homogeneous
|
|
|
186 top-level windows, are very complicated, and as a result, most window managers
|
|
|
187 don't implement them correctly.
|
|
|
188
|
|
|
189 The R4/R5 version of twm (and all of its descendants) had buggy focus
|
|
|
190 handling; there is a patch in .../xemacs/etc/twm-patch which fixes this.
|
|
|
191 Sufficiently recent versions of tvtwm do not need this patch, but most other
|
|
|
192 versions of twm do. If you need to apply this patch, please try to get it
|
|
|
193 integrated by the maintainer of whichever version of twm you're using.
|
|
|
194
|
|
|
195 In addition, if you're using twm, make sure you have not specified
|
|
|
196 "NoTitleFocus" in your .tvtwmrc file. The very nature of this option makes
|
|
|
197 twm do some illegal focus tricks, even with the patch.
|
|
|
198
|
|
|
199 It is known that olwm and olvwm are buggy, and in different ways. If you're
|
|
|
200 using click-to-type mode, try using point-to-type, or vice versa.
|
|
|
201
|
|
|
202 In older versions of NCDwm, one could not even type at XEmacs windows. This
|
|
|
203 has been fixed in newer versions (2.4.3, and possibly earlier).
|
|
|
204
|
|
|
205 (Many people suggest that XEmacs should warp the mouse when focusing on
|
|
|
206 another screen in point-to-type mode. This is not ICCCM-compliant behavior.
|
|
|
207 Implementing such policy is the responsibility of the window manager itself,
|
|
|
208 it is not legal for a client to do this.)
|
|
|
209
|
|
|
210 * My buffers are full of \000 characters or otherwise corrupt.
|
|
|
211
|
|
|
212 Some compilers have trouble with gmalloc.c and ralloc.c; try recompiling
|
|
|
213 without optimization. If that doesn't work, try recompiling with
|
|
|
214 SYSTEM_MALLOC defined, and/or with REL_ALLOC undefined.
|
|
|
215
|
|
|
216 * Some packages that worked before now cause the error
|
|
|
217 Wrong type argument: arrayp, #<face ... >
|
|
|
218
|
|
|
219 Code which uses the `face' accessor functions must be recompiled with xemacs
|
|
|
220 19.9 or later. The functions whose callers must be recompiled are: face-font,
|
|
|
221 face-foreground, face-background, face-background-pixmap, and face-underline-p.
|
|
|
222 The .elc files generated by version 19.9 will work in 19.6 and 19.8, but older
|
|
|
223 .elc files which contain calls to these functions will not work in 19.9.
|
|
|
224
|
|
|
225 * On Solaris 2.* I get undefined symbols from libcurses.a.
|
|
|
226
|
|
|
227 You probably have /usr/ucblib/ on your LD_LIBRARY_PATH. Do the link with
|
|
|
228 LD_LIBRARY_PATH unset.
|
|
|
229
|
|
|
230 * I don't have `xmkmf' and `imake' on my HP.
|
|
|
231
|
|
|
232 You can get these standard X tools by anonymous FTP to hpcvaaz.cv.hp.com.
|
|
|
233 Essentially all X programs need these.
|
|
|
234
|
|
|
235 * When emacs starts up, I get lots of warnings about unknown keysyms.
|
|
|
236
|
|
|
237 If you are running the prebuilt binaries, the Motif library expects to find
|
|
|
238 certain thing in the XKeysymDB file. This file is normally in /usr/lib/X11/
|
|
|
239 or in /usr/openwin/lib/. If you keep yours in a different place, set the
|
|
|
240 environment variable $XKEYSYMDB to point to it before starting emacs. If
|
|
|
241 you still have the problem after doing that, perhaps your version of X is
|
|
|
242 too old. There is a copy of the MIT X11R5 XKeysymDB file in the emacs `etc'
|
|
|
243 directory. Try using that one.
|
|
|
244
|
|
|
245 * My X resources used to work, and now some of them are being ignored.
|
|
|
246
|
|
|
247 Check the resources in .../etc/Emacs.ad (which is the same as the file
|
|
|
248 sample.Xdefaults). Perhaps some of the default resources built in to
|
|
|
249 emacs are now overriding your existing resources. Copy and edit the
|
|
|
250 resources in Emacs.ad as necessary.
|
|
|
251
|
|
|
252 * Solaris 2.3 /bin/sh coredumps during configuration.
|
|
|
253
|
|
|
254 This only occurs if you have LANG != C. This is a known bug with
|
|
|
255 /bin/sh fixed by installing Patch-ID# 101613-01.
|
|
|
256
|
|
|
257 * "Cannot find callback list" messages from dialog boxes on HPUX, in
|
|
|
258 Emacs built with Motif.
|
|
|
259
|
|
|
260 This problem resulted from a bug in GCC 2.4.5. Newer GCC versions
|
|
|
261 such as 2.7.0 fix the problem.
|
|
|
262
|
|
|
263 * On Irix 6.0, make tries (and fails) to build a program named unexelfsgi
|
|
|
264
|
|
|
265 A compiler bug inserts spaces into the string "unexelfsgi . o"
|
|
|
266 in src/Makefile. Edit src/Makefile, after configure is run,
|
|
|
267 find that string, and take out the spaces.
|
|
|
268
|
|
|
269 Compiler fixes in Irix 6.0.1 should eliminate this problem.
|
|
|
270
|
|
|
271 * With certain fonts, when the cursor appears on a character, the
|
|
|
272 character doesn't appear--you get a solid box instead.
|
|
|
273
|
|
|
274 One user on a Linux-based GNU system reported that this problem went
|
|
|
275 away with installation of a new X server. The failing server was
|
|
|
276 XFree86 3.1.1. XFree86 3.1.2 works.
|
|
|
277
|
|
|
278 * On SunOS 4.1.3, Emacs unpredictably crashes in _yp_dobind_soft.
|
|
|
279
|
|
|
280 This happens if you configure Emacs specifying just `sparc-sun-sunos4'
|
|
|
281 on a system that is version 4.1.3. You must specify the precise
|
|
|
282 version number (or let configure figure out the configuration, which
|
|
|
283 it can do perfectly well for SunOS).
|
|
|
284
|
|
|
285 * On SunOS 4, Emacs processes keep going after you kill the X server
|
|
|
286 (or log out, if you logged in using X).
|
|
|
287
|
|
|
288 Someone reported that recompiling with GCC 2.7.0 fixed this problem.
|
|
|
289
|
|
|
290 * On AIX 4, some programs fail when run in a Shell buffer
|
|
|
291 with an error message like No terminfo entry for "unknown".
|
|
|
292
|
|
|
293 On AIX, many terminal type definitions are not installed by default.
|
|
|
294 `unknown' is one of them. Install the "Special Generic Terminal
|
|
|
295 Definitions" to make them defined.
|
|
|
296
|
|
|
297 * On SunOS, you get linker errors
|
|
|
298 ld: Undefined symbol
|
|
|
299 _get_wmShellWidgetClass
|
|
|
300 _get_applicationShellWidgetClass
|
|
|
301
|
|
|
302 The fix to this is to install patch 100573 for OpenWindows 3.0
|
|
|
303 or link libXmu statically.
|
|
|
304
|
|
|
305 * On AIX 4.1.2, linker error messages such as
|
|
|
306 ld: 0711-212 SEVERE ERROR: Symbol .__quous, found in the global symbol table
|
|
|
307 of archive /usr/lib/libIM.a, was not defined in archive member shr.o.
|
|
|
308
|
|
|
309 This is a problem in libIM.a. You can work around it by executing
|
|
|
310 these shell commands in the src subdirectory of the directory where
|
|
|
311 you build Emacs:
|
|
|
312
|
|
|
313 cp /usr/lib/libIM.a .
|
|
|
314 chmod 664 libIM.a
|
|
|
315 ranlib libIM.a
|
|
|
316
|
|
|
317 Then change -lIM to ./libIM.a in the command to link temacs (in
|
|
|
318 Makefile).
|
|
|
319
|
|
|
320 * Unpredictable segmentation faults on Solaris 2.3 and 2.4.
|
|
|
321
|
|
|
322 A user reported that this happened in 19.29 when it was compiled with
|
|
|
323 the Sun compiler, but not when he recompiled with GCC 2.7.0.
|
|
|
324
|
|
|
325 We do not know whether something in Emacs is partly to blame for this.
|
|
|
326
|
|
|
327 * Emacs exits with "X protocol error" when run with an X server for
|
|
|
328 Windows.
|
|
|
329
|
|
|
330 A certain X server for Windows had a bug which caused this.
|
|
|
331 Supposedly the newer 32-bit version of this server doesn't have the
|
|
|
332 problem.
|
|
|
333
|
|
|
334 * A position you specified in .Xdefaults is ignored, using twm.
|
|
|
335
|
|
|
336 twm normally ignores "program-specified" positions.
|
|
|
337 You can tell it to obey them with this command in your `.twmrc' file:
|
|
|
338
|
|
|
339 UsePPosition "on" #allow clents to request a position
|
|
|
340
|
|
|
341 * Compiling lib-src says there is no rule to make test-distrib.c.
|
|
|
342
|
|
|
343 This results from a bug in a VERY old version of GNU Sed. To solve
|
|
|
344 the problem, install the current version of GNU Sed, then rerun
|
|
|
345 Emacs's configure script.
|
|
|
346
|
|
|
347 * On Sunos 4.1.1, there are errors compiling sysdep.c.
|
|
|
348
|
|
|
349 If you get errors such as
|
|
|
350
|
|
|
351 "sysdep.c", line 2017: undefined structure or union
|
|
|
352 "sysdep.c", line 2017: undefined structure or union
|
|
|
353 "sysdep.c", line 2019: nodename undefined
|
|
|
354
|
|
|
355 This can result from defining LD_LIBRARY_PATH. It is very tricky
|
|
|
356 to use that environment variable with Emacs. The Emacs configure
|
|
|
357 script links many test programs with the system libraries; you must
|
|
|
358 make sure that the libraries available to configure are the same
|
|
|
359 ones available when you build Emacs.
|
|
|
360
|
|
|
361 * The right Alt key works wrong on German HP keyboards (and perhaps
|
|
|
362 other non-English HP keyboards too).
|
|
|
363
|
|
|
364 This is because HPUX defines the modifiers wrong in X. Here is a
|
|
|
365 shell script to fix the problem; be sure that it is run after VUE
|
|
|
366 configures the X server.
|
|
|
367
|
|
|
368 xmodmap 2> /dev/null - << EOF
|
|
|
369 keysym Alt_L = Meta_L
|
|
|
370 keysym Alt_R = Meta_R
|
|
|
371 EOF
|
|
|
372
|
|
|
373 xmodmap - << EOF
|
|
|
374 clear mod1
|
|
|
375 keysym Mode_switch = NoSymbol
|
|
|
376 add mod1 = Meta_L
|
|
|
377 keysym Meta_R = Mode_switch
|
|
|
378 add mod2 = Mode_switch
|
|
|
379 EOF
|
|
|
380
|
|
|
381 * The Emacs window disappears when you type M-q.
|
|
|
382
|
|
|
383 Some versions of the Open Look window manager interpret M-q as a quit
|
|
|
384 command for whatever window you are typing at. If you want to use
|
|
|
385 Emacs with that window manager, you should try to configure the window
|
|
|
386 manager to use some other command. You can disable the
|
|
|
387 shortcut keys entirely by adding this line to ~/.OWdefaults:
|
|
|
388
|
|
|
389 OpenWindows.WindowMenuAccelerators: False
|
|
|
390
|
|
|
391 * Emacs does not notice when you release the mouse.
|
|
|
392
|
|
|
393 There are reports that this happened with (some) Microsoft mice and
|
|
|
394 that replacing the mouse made it stop.
|
|
|
395
|
|
|
396 * Trouble using ptys on IRIX, or running out of ptys.
|
|
|
397
|
|
|
398 The program mkpts (which may be in `/usr/adm' or `/usr/sbin') needs to
|
|
|
399 be set-UID to root, or non-root programs like Emacs will not be able
|
|
|
400 to allocate ptys reliably.
|
|
|
401
|
|
|
402 * On Irix 5.2, unexelfsgi.c can't find cmplrs/stsupport.h.
|
|
|
403
|
|
|
404 The file cmplrs/stsupport.h was included in the wrong file set in the
|
|
|
405 Irix 5.2 distribution. You can find it in the optional fileset
|
|
|
406 compiler_dev, or copy it from some other Irix 5.2 system. A kludgy
|
|
|
407 workaround is to change unexelfsgi.c to include sym.h instead of
|
|
|
408 syms.h.
|
|
|
409
|
|
|
410 * Slow startup on Linux.
|
|
|
411
|
|
|
412 People using systems based on the Linux kernel sometimes report that
|
|
|
413 startup takes 10 to 15 seconds longer than `usual'.
|
|
|
414
|
|
|
415 This is because Emacs looks up the host name when it starts.
|
|
|
416 Normally, this takes negligible time; the extra delay is due to
|
|
|
417 improper system configuration. This problem can occur for both
|
|
|
418 networked and non-networked machines.
|
|
|
419
|
|
|
420 Here is how to fix the configuration. It requires being root.
|
|
|
421
|
|
|
422 ** Networked Case
|
|
|
423
|
|
|
424 First, make sure the files `/etc/hosts' and `/etc/host.conf' both
|
|
|
425 exist. The first line in the `/etc/hosts' file should look like this
|
|
|
426 (replace HOSTNAME with your host name):
|
|
|
427
|
|
|
428 127.0.0.1 HOSTNAME
|
|
|
429
|
|
|
430 Also make sure that the `/etc/host.conf' files contains the following
|
|
|
431 lines:
|
|
|
432
|
|
|
433 order hosts, bind
|
|
|
434 multi on
|
|
|
435
|
|
|
436 Any changes, permanent and temporary, to the host name should be
|
|
|
437 indicated in the `/etc/hosts' file, since it acts a limited local
|
|
|
438 database of addresses and names (e.g., some SLIP connections
|
|
|
439 dynamically allocate ip addresses).
|
|
|
440
|
|
|
441 ** Non-Networked Case
|
|
|
442
|
|
|
443 The solution described in the networked case applies here as well.
|
|
|
444 However, if you never intend to network your machine, you can use a
|
|
|
445 simpler solution: create an empty `/etc/host.conf' file. The command
|
|
|
446 `touch /etc/host.conf' suffices to create the file. The `/etc/hosts'
|
|
|
447 file is not necessary with this approach.
|
|
|
448
|
|
|
449 * On Solaris 2.4, Dired hangs and C-g does not work. Or Emacs hangs
|
|
|
450 forever waiting for termination of a subprocess that is a zombie.
|
|
|
451
|
|
|
452 casper@fwi.uva.nl says the problem is in X11R6. Rebuild libX11.so
|
|
|
453 after changing the file xc/config/cf/sunLib.tmpl. Change the lines
|
|
|
454
|
|
|
455 #if ThreadedX
|
|
|
456 #define SharedX11Reqs -lthread
|
|
|
457 #endif
|
|
|
458
|
|
|
459 to:
|
|
|
460
|
|
|
461 #if OSMinorVersion < 4
|
|
|
462 #if ThreadedX
|
|
|
463 #define SharedX11Reqs -lthread
|
|
|
464 #endif
|
|
|
465 #endif
|
|
|
466
|
|
|
467 Be sure also to edit x/config/cf/sun.cf so that OSMinorVersion is 4
|
|
|
468 (as it should be for Solaris 2.4). The file has three definitions for
|
|
|
469 OSMinorVersion: the first is for x86, the second for SPARC under
|
|
|
470 Solaris, and the third for SunOS 4. Make sure to update the
|
|
|
471 definition for your type of machine and system.
|
|
|
472
|
|
|
473 Then do `make Everything' in the top directory of X11R6, to rebuild
|
|
|
474 the makefiles and rebuild X. The X built this way work only on
|
|
|
475 Solaris 2.4, not on 2.3.
|
|
|
476
|
|
|
477 For multithreaded X to work it necessary to install patch
|
|
|
478 101925-02 to fix problems in header files [2.4]. You need
|
|
|
479 to reinstall gcc or re-run just-fixinc after installing that
|
|
|
480 patch.
|
|
|
481
|
|
|
482 However, Frank Rust <frust@iti.cs.tu-bs.de> used a simpler solution:
|
|
|
483 he changed
|
|
|
484 #define ThreadedX YES
|
|
|
485 to
|
|
|
486 #define ThreadedX NO
|
|
|
487 in sun.cf and did `make World' to rebuild X11R6. Removing all
|
|
|
488 `-DXTHREAD*' flags and `-lthread' entries from lib/X11/Makefile and
|
|
|
489 typing 'make install' in that directory also seemed to work.
|
|
|
490
|
|
|
491 * With M-x enable-flow-control, you need to type C-\ twice
|
|
|
492 to do incremental search--a single C-\ gets no response.
|
|
|
493
|
|
|
494 This has been traced to communicating with your machine via kermit,
|
|
|
495 with C-\ as the kermit escape character. One solution is to use
|
|
|
496 another escape character in kermit. One user did
|
|
|
497
|
|
|
498 set escape-character 17
|
|
|
499
|
|
|
500 in his .kermrc file, to make C-q the kermit escape character.
|
|
|
501
|
|
|
502 * The Motif version of Emacs paints the screen a solid color.
|
|
|
503
|
|
|
504 This has been observed to result from the following X resource:
|
|
|
505
|
|
|
506 Emacs*default.attributeFont: -*-courier-medium-r-*-*-*-140-*-*-*-*-iso8859-*
|
|
|
507
|
|
|
508 That the resource has this effect indicates a bug in something, but we
|
|
|
509 do not yet know what. If it is an Emacs bug, we hope someone can
|
|
|
510 explain what the bug is so we can fix it. In the mean time, removing
|
|
|
511 the resource prevents the problem.
|
|
|
512
|
|
|
513 * Emacs gets hung shortly after startup, on Sunos 4.1.3.
|
|
|
514
|
|
|
515 We think this is due to a bug in Sunos. The word is that
|
|
|
516 one of these Sunos patches fixes the bug:
|
|
|
517
|
|
|
518 100075-11 100224-06 100347-03 100482-05 100557-02 100623-03 100804-03 101080-01
|
|
|
519 100103-12 100249-09 100496-02 100564-07 100630-02 100891-10 101134-01
|
|
|
520 100170-09 100296-04 100377-09 100507-04 100567-04 100650-02 101070-01 101145-01
|
|
|
521 100173-10 100305-15 100383-06 100513-04 100570-05 100689-01 101071-03 101200-02
|
|
|
522 100178-09 100338-05 100421-03 100536-02 100584-05 100784-01 101072-01 101207-01
|
|
|
523
|
|
|
524 We don't know which of these patches really matter. If you find out
|
|
|
525 which ones, please inform bug-gnu-emacs@prep.ai.mit.edu.
|
|
|
526
|
|
|
527 * Emacs aborts while starting up, only when run without X.
|
|
|
528
|
|
|
529 This problem often results from compiling Emacs with GCC when GCC was
|
|
|
530 installed incorrectly. The usual error in installing GCC is to
|
|
|
531 specify --includedir=/usr/include. Installation of GCC makes
|
|
|
532 corrected copies of the system header files. GCC is supposed to use
|
|
|
533 the corrected copies in preference to the original system headers.
|
|
|
534 Specifying --includedir=/usr/include causes the original system header
|
|
|
535 files to be used. On some systems, the definition of ioctl in the
|
|
|
536 original system header files is invalid for ANSI C and causes Emacs
|
|
|
537 not to work.
|
|
|
538
|
|
|
539 The fix is to reinstall GCC, and this time do not specify --includedir
|
|
|
540 when you configure it. Then recompile Emacs. Specifying --includedir
|
|
|
541 is appropriate only in very special cases and it should *never* be the
|
|
|
542 same directory where system header files are kept.
|
|
|
543
|
|
|
544 * The Compose key on a DEC keyboard does not work as Meta key.
|
|
|
545
|
|
|
546 This shell command should fix it:
|
|
|
547
|
|
|
548 xmodmap -e 'keycode 0xb1 = Meta_L'
|
|
|
549
|
|
|
550 * Regular expressions matching bugs on SCO systems.
|
|
|
551
|
|
|
552 On SCO, there are problems in regexp matching when Emacs is compiled
|
|
|
553 with the system compiler. The compiler version is "Microsoft C
|
|
|
554 version 6", SCO 4.2.0h Dev Sys Maintenance Supplement 01/06/93; Quick
|
|
|
555 C Compiler Version 1.00.46 (Beta). The solution is to compile with
|
|
|
556 GCC.
|
|
|
557
|
|
|
558 * On Sunos 4, you get the error ld: Undefined symbol __lib_version.
|
|
|
559
|
|
|
560 This is the result of using cc or gcc with the shared library meant
|
|
|
561 for acc (the Sunpro compiler). Check your LD_LIBRARY_PATH and delete
|
|
|
562 /usr/lang/SC2.0.1 or some similar directory.
|
|
|
563
|
|
|
564 * You can't select from submenus.
|
|
|
565
|
|
|
566 On certain systems, mouse-tracking and selection in top-level menus
|
|
|
567 works properly with the X toolkit, but neither of them works when you
|
|
|
568 bring up a submenu (such as Bookmarks or Compare or Apply Patch, in
|
|
|
569 the Files menu).
|
|
|
570
|
|
|
571 This works on most systems. There is speculation that the failure is
|
|
|
572 due to bugs in old versions of X toolkit libraries, but no one really
|
|
|
573 knows. If someone debugs this and finds the precise cause, perhaps a
|
|
|
574 workaround can be found.
|
|
|
575
|
|
|
576 * Unusable default font on SCO 3.2v4.
|
|
|
577
|
|
|
578 The Open Desktop environment comes with default X resource settings
|
|
|
579 that tell Emacs to use a variable-width font. Emacs cannot use such
|
|
|
580 fonts, so it does not work.
|
|
|
581
|
|
|
582 This is caused by the file /usr/lib/X11/app-defaults/ScoTerm, which is
|
|
|
583 the application-specific resource file for the `scoterm' terminal
|
|
|
584 emulator program. It contains several extremely general X resources
|
|
|
585 that affect other programs besides `scoterm'. In particular, these
|
|
|
586 resources affect Emacs also:
|
|
|
587
|
|
|
588 *Font: -*-helvetica-medium-r-*--12-*-p-*
|
|
|
589 *Background: scoBackground
|
|
|
590 *Foreground: scoForeground
|
|
|
591
|
|
|
592 The best solution is to create an application-specific resource file for
|
|
|
593 Emacs, /usr/lib/X11/app-defaults/Emacs, with the following contents:
|
|
|
594
|
|
|
595 Emacs*Font: -*-courier-medium-r-*-*-*-120-*-*-*-*-iso8859-1
|
|
|
596 Emacs*Background: white
|
|
|
597 Emacs*Foreground: black
|
|
|
598
|
|
|
599 (or whatever other defaults you prefer).
|
|
|
600
|
|
|
601 These resource files are not normally shared across a network of SCO
|
|
|
602 machines; you must create the file on each machine individually.
|
|
|
603
|
|
|
604 * rcs2log gives you the awk error message "too many fields".
|
|
|
605
|
|
|
606 This is due to an arbitrary limit in certain versions of awk.
|
|
|
607 The solution is to use gawk (GNU awk).
|
|
|
608
|
|
|
609 * Emacs is slow using X11R5 on HP/UX.
|
|
|
610
|
|
|
611 This happens if you use the MIT versions of the X libraries--it
|
|
|
612 doesn't run as fast as HP's version. People sometimes use the version
|
|
|
613 because they see the HP version doesn't have the libraries libXaw.a,
|
|
|
614 libXmu.a, libXext.a and others. HP/UX normally doesn't come with
|
|
|
615 those libraries installed. To get good performance, you need to
|
|
|
616 install them and rebuild Emacs.
|
|
|
617
|
|
|
618 * Loading fonts is very slow.
|
|
|
619
|
|
|
620 You might be getting scalable fonts instead of precomputed bitmaps.
|
|
|
621 Known scalable font directories are "Type1" and "Speedo". A font
|
|
|
622 directory contains scalable fonts if it contains the file
|
|
|
623 "fonts.scale".
|
|
|
624
|
|
|
625 If this is so, re-order your X windows font path to put the scalable
|
|
|
626 font directories last. See the documentatoin of `xset' for details.
|
|
|
627
|
|
|
628 With some X servers, it may be necessary to take the scalable font
|
|
|
629 directories out of your path entirely, at least for Emacs 19.26.
|
|
|
630 Changes in the future may make this unnecessary.
|
|
|
631
|
|
|
632 * On AIX 3.2.4, releasing Ctrl/Act key has no effect, if Shift is down.
|
|
|
633
|
|
|
634 Due to a feature of AIX, pressing or releasing the Ctrl/Act key is
|
|
|
635 ignored when the Shift, Alt or AltGr keys are held down. This can
|
|
|
636 lead to the keyboard being "control-locked"--ordinary letters are
|
|
|
637 treated as control characters.
|
|
|
638
|
|
|
639 You can get out of this "control-locked" state by pressing and
|
|
|
640 releasing Ctrl/Act while not pressing or holding any other keys.
|
|
|
641
|
|
|
642 * display-time causes kernel problems on ISC systems.
|
|
|
643
|
|
|
644 Under Interactive Unix versions 3.0.1 and 4.0 (and probably other
|
|
|
645 versions), display-time causes the loss of large numbers of STREVENT
|
|
|
646 cells. Eventually the kernel's supply of these cells is exhausted.
|
|
|
647 This makes emacs and the whole system run slow, and can make other
|
|
|
648 processes die, in particular pcnfsd.
|
|
|
649
|
|
|
650 Other emacs functions that communicate with remote processes may have
|
|
|
651 the same problem. Display-time seems to be far the worst.
|
|
|
652
|
|
|
653 The only known fix: Don't run display-time.
|
|
|
654
|
|
|
655 * On Solaris, C-x doesn't get through to Emacs when you use the console.
|
|
|
656
|
|
|
657 This is a Solaris feature (at least on Intel x86 cpus). Type C-r
|
|
|
658 C-r C-t, to toggle whether C-x gets through to Emacs.
|
|
|
659
|
|
|
660 * Error message `Symbol's value as variable is void: x', followed by
|
|
|
661 segmentation fault and core dump.
|
|
|
662
|
|
|
663 This has been tracked to a bug in tar! People report that tar erroneously
|
|
|
664 added a line like this at the beginning of files of Lisp code:
|
|
|
665
|
|
|
666 x FILENAME, N bytes, B tape blocks
|
|
|
667
|
|
|
668 If your tar has this problem, install GNU tar--if you can manage to
|
|
|
669 untar it :-).
|
|
|
670
|
|
|
671 * Link failure when using acc on a Sun.
|
|
|
672
|
|
|
673 To use acc, you need additional options just before the libraries, such as
|
|
|
674
|
|
|
675 /usr/lang/SC2.0.1/values-Xt.o -L/usr/lang/SC2.0.1/cg87 -L/usr/lang/SC2.0.1
|
|
|
676
|
|
|
677 and you need to add -lansi just before -lc.
|
|
|
678
|
|
|
679 The precise file names depend on the compiler version, so we
|
|
|
680 cannot easily arrange to supply them.
|
|
|
681
|
|
|
682 * Link failure on IBM AIX 1.3 ptf 0013.
|
|
|
683
|
|
|
684 There is a real duplicate definition of the function `_slibc_free' in
|
|
|
685 the library /lib/libc_s.a (just do nm on it to verify). The
|
|
|
686 workaround/fix is:
|
|
|
687
|
|
|
688 cd /lib
|
|
|
689 ar xv libc_s.a NLtmtime.o
|
|
|
690 ar dv libc_s.a NLtmtime.o
|
|
|
691
|
|
|
692 * Undefined symbols _dlopen, _dlsym and/or _dlclose on a Sun.
|
|
|
693
|
|
|
694 If you see undefined symbols _dlopen, _dlsym, or _dlclose when linking
|
|
|
695 with -lX11, compile and link against the file mit/util/misc/dlsym.c in
|
|
|
696 the MIT X11R5 distribution. Alternatively, link temacs using shared
|
|
|
697 libraries with s/sunos4shr.h. (This doesn't work if you use the X
|
|
|
698 toolkit.)
|
|
|
699
|
|
|
700 If you get the additional error that the linker could not find
|
|
|
701 lib_version.o, try extracting it from X11/usr/lib/X11/libvim.a in
|
|
|
702 X11R4, then use it in the link.
|
|
|
703
|
|
|
704 * In Shell mode, you get a ^M at the end of every line.
|
|
|
705
|
|
|
706 This happens to people who use tcsh, because it is trying to be too
|
|
|
707 smart. It sees that the Shell uses terminal type `unknown' and turns
|
|
|
708 on the flag to output ^M at the end of each line. You can fix the
|
|
|
709 problem by adding this to your .cshrc file:
|
|
|
710
|
|
|
711 if ($?EMACS) then
|
|
|
712 if ($EMACS == "t") then
|
|
|
713 unset edit
|
|
|
714 stty -icrnl -onlcr -echo susp ^Z
|
|
|
715 endif
|
|
|
716 endif
|
|
|
717
|
|
|
718 * An error message such as `X protocol error: BadMatch (invalid
|
|
|
719 parameter attributes) on protocol request 93'.
|
|
|
720
|
|
|
721 This comes from having an invalid X resource, such as
|
|
|
722 emacs*Cursor: black
|
|
|
723 (which is invalid because it specifies a color name for something
|
|
|
724 that isn't a color.)
|
|
|
725
|
|
|
726 The fix is to correct your X resources.
|
|
|
727
|
|
|
728 * Undefined symbols when linking on Sunos 4.1.
|
|
|
729
|
|
|
730 If you get the undefined symbols _atowc _wcslen, _iswprint, _iswspace,
|
|
|
731 _iswcntrl, _wcscpy, and _wcsncpy, then you need to add -lXwchar after
|
|
|
732 -lXaw in the command that links temacs.
|
|
|
733
|
|
|
734 This problem seems to arise only when the international language
|
|
|
735 extensions to X11R5 are installed.
|
|
|
736
|
|
|
737 * Typing C-c C-c in Shell mode kills your X server.
|
|
|
738
|
|
|
739 This happens with Linux kernel 1.0 thru 1.04, approximately. The workaround is
|
|
|
740 to define SIGNALS_VIA_CHARACTERS in config.h and recompile Emacs.
|
|
|
741 Newer Linux kernel versions don't have this problem.
|
|
|
742
|
|
|
743 * src/Makefile and lib-src/Makefile are truncated--most of the file missing.
|
|
|
744
|
|
|
745 This can happen if configure uses GNU sed version 2.03. That version
|
|
|
746 had a bug. GNU sed version 2.05 works properly.
|
|
|
747
|
|
|
748 * Slow startup on X11R6 with X windows.
|
|
|
749
|
|
|
750 If Emacs takes two minutes to start up on X11R6, see if your X
|
|
|
751 resources specify any Adobe fonts. That causes the type-1 font
|
|
|
752 renderer to start up, even if the font you asked for is not a type-1
|
|
|
753 font.
|
|
|
754
|
|
|
755 One way to avoid this problem is to eliminate the type-1 fonts from
|
|
|
756 your font path, like this:
|
|
|
757
|
|
|
758 xset -fp /usr/X11R6/lib/X11/fonts/Type1/
|
|
|
759
|
|
|
760 * Pull-down menus appear in the wrong place, in the toolkit version of Emacs.
|
|
|
761
|
|
|
762 An X resource of this form can cause the problem:
|
|
|
763
|
|
|
764 Emacs*geometry: 80x55+0+0
|
|
|
765
|
|
|
766 This resource is supposed to apply, and does apply, to the menus
|
|
|
767 individually as well as to Emacs frames. If that is not what you
|
|
|
768 want, rewrite the resource.
|
|
|
769
|
|
|
770 To check thoroughly for such resource specifications, use `xrdb
|
|
|
771 -query' to see what resources the X server records, and also look at
|
|
|
772 the user's ~/.Xdefaults and ~/.Xdefaults-* files.
|
|
|
773
|
|
|
774 * `make install' fails on install-doc with `Error 141'.
|
|
|
775
|
|
|
776 This happens on Ultrix 4.2 due to failure of a pipeline of tar
|
|
|
777 commands. We don't know why they fail, but the bug seems not to be in
|
|
|
778 Emacs. The workaround is to run the shell command in install-doc by
|
|
|
779 hand.
|
|
|
780
|
|
|
781 * Subprocesses remain, hanging but not zombies, on Sunos 5.3.
|
|
|
782
|
|
|
783 A bug in Sunos 5.3 causes Emacs subprocesses to remain after Emacs
|
|
|
784 exits. Sun patch # 101415-02 is part of the fix for this, but it only
|
|
|
785 applies to ptys, and doesn't fix the problem with subprocesses
|
|
|
786 communicating through pipes.
|
|
|
787
|
|
|
788 * Mail is lost when sent to local aliases.
|
|
|
789
|
|
|
790 Many emacs mail user agents (VM and rmail, for instance) use the
|
|
|
791 sendmail.el library. This library can arrange for mail to be
|
|
|
792 delivered by passing messages to the /usr/lib/sendmail (usually)
|
|
|
793 program . In doing so, it passes the '-t' flag to sendmail, which
|
|
|
794 means that the name of the recipient of the message is not on the
|
|
|
795 command line and, therefore, that sendmail must parse the message to
|
|
|
796 obtain the destination address.
|
|
|
797
|
|
|
798 There is a bug in the SunOS4.1.1 and SunOS4.1.3 versions of sendmail.
|
|
|
799 In short, when given the -t flag, the SunOS sendmail won't recognize
|
|
|
800 non-local (i.e. NIS) aliases. It has been reported that the Solaris
|
|
|
801 2.x versions of sendmail do not have this bug. For those using SunOS
|
|
|
802 4.1, the best fix is to install sendmail V8 or IDA sendmail (which
|
|
|
803 have other advantages over the regular sendmail as well). At the time
|
|
|
804 of this writing, these official versions are available:
|
|
|
805
|
|
|
806 Sendmail V8 on ftp.cs.berkeley.edu in /ucb/sendmail:
|
|
|
807 sendmail.8.6.9.base.tar.Z (the base system source & documentation)
|
|
|
808 sendmail.8.6.9.cf.tar.Z (configuration files)
|
|
|
809 sendmail.8.6.9.misc.tar.Z (miscellaneous support programs)
|
|
|
810 sendmail.8.6.9.xdoc.tar.Z (extended documentation, with postscript)
|
|
|
811
|
|
|
812 IDA sendmail on vixen.cso.uiuc.edu in /pub:
|
|
|
813 sendmail-5.67b+IDA-1.5.tar.gz
|
|
|
814
|
|
|
815 * On AIX, you get this message when running Emacs:
|
|
|
816
|
|
|
817 Could not load program emacs
|
|
|
818 Symbol smtcheckinit in csh is undefined
|
|
|
819 Error was: Exec format error
|
|
|
820
|
|
|
821 or this one:
|
|
|
822
|
|
|
823 Could not load program .emacs
|
|
|
824 Symbol _system_con in csh is undefined
|
|
|
825 Symbol _fp_trapsta in csh is undefined
|
|
|
826 Error was: Exec format error
|
|
|
827
|
|
|
828 These can happen when you try to run on AIX 3.2.5 a program that was
|
|
|
829 compiled with 3.2.4. The fix is to recompile.
|
|
|
830
|
|
|
831 * On AIX, you get this compiler error message:
|
|
|
832
|
|
|
833 Processing include file ./XMenuInt.h
|
|
|
834 1501-106: (S) Include file X11/Xlib.h not found.
|
|
|
835
|
|
|
836 This means your system was installed with only the X11 runtime i.d
|
|
|
837 libraries. You have to find your sipo (bootable tape) and install
|
|
|
838 X11Dev... with smit.
|
|
|
839
|
|
|
840 * You "lose characters" after typing Compose Character key.
|
|
|
841
|
|
|
842 This is because the Compose Character key is defined as the keysym
|
|
|
843 Multi_key, and Emacs (seeing that) does the proper X11
|
|
|
844 character-composition processing. If you don't want your Compose key
|
|
|
845 to do that, you can redefine it with xmodmap.
|
|
|
846
|
|
|
847 For example, here's one way to turn it into a Meta key:
|
|
|
848
|
|
|
849 xmodmap -e "keysym Multi_key = Meta_L"
|
|
|
850
|
|
|
851 If all users at your site of a particular keyboard prefer Meta to
|
|
|
852 Compose, you can make the remapping happen automatically by adding the
|
|
|
853 xmodmap command to the xdm setup script for that display.
|
|
|
854
|
|
|
855 * C-z just refreshes the screen instead of suspending Emacs.
|
|
|
856
|
|
|
857 You are probably using a shell that doesn't support job control, even
|
|
|
858 though the system itself is capable of it. Either use a different shell,
|
|
|
859 or set the variable `cannot-suspend' to a non-nil value.
|
|
|
860
|
|
|
861 * Watch out for .emacs files and EMACSLOADPATH environment vars
|
|
|
862
|
|
|
863 These control the actions of Emacs.
|
|
|
864 ~/.emacs is your Emacs init file.
|
|
|
865 EMACSLOADPATH overrides which directories the function
|
|
|
866 "load" will search.
|
|
|
867
|
|
|
868 If you observe strange problems, check for these and get rid
|
|
|
869 of them, then try again.
|
|
|
870
|
|
|
871 * After running emacs once, subsequent invocations crash.
|
|
|
872
|
|
|
873 Some versions of SVR4 have a serious bug in the implementation of the
|
|
|
874 mmap () system call in the kernel; this causes emacs to run correctly
|
|
|
875 the first time, and then crash when run a second time.
|
|
|
876
|
|
|
877 Contact your vendor and ask for the mmap bug fix; in the mean time,
|
|
|
878 you may be able to work around the problem by adding a line to your
|
|
|
879 operating system description file (whose name is reported by the
|
|
|
880 configure script) that reads:
|
|
|
881 #define SYSTEM_MALLOC
|
|
|
882 This makes Emacs use memory less efficiently, but seems to work around
|
|
|
883 the kernel bug.
|
|
|
884
|
|
|
885 * Inability to send an Alt-modified key, when Emacs is communicating
|
|
|
886 directly with an X server.
|
|
|
887
|
|
|
888 If you have tried to bind an Alt-modified key as a command, and it
|
|
|
889 does not work to type the command, the first thing you should check is
|
|
|
890 whether the key is getting through to Emacs. To do this, type C-h c
|
|
|
891 followed by the Alt-modified key. C-h c should say what kind of event
|
|
|
892 it read. If it says it read an Alt-modified key, then make sure you
|
|
|
893 have made the key binding correctly.
|
|
|
894
|
|
|
895 If C-h c reports an event that doesn't have the Alt modifier, it may
|
|
|
896 be because your X server has no key for the Alt modifier. The X
|
|
|
897 server that comes from MIT does not set up the Alt modifier by
|
|
|
898 default.
|
|
|
899
|
|
|
900 If your keyboard has keys named Alt, you can enable them as follows:
|
|
|
901
|
|
|
902 xmodmap -e 'add mod2 = Alt_L'
|
|
|
903 xmodmap -e 'add mod2 = Alt_R'
|
|
|
904
|
|
|
905 If the keyboard has just one key named Alt, then only one of those
|
|
|
906 commands is needed. The modifier `mod2' is a reasonable choice if you
|
|
|
907 are using an unmodified MIT version of X. Otherwise, choose any
|
|
|
908 modifier bit not otherwise used.
|
|
|
909
|
|
|
910 If your keyboard does not have keys named Alt, you can use some other
|
|
|
911 keys. Use the keysym command in xmodmap to turn a function key (or
|
|
|
912 some other 'spare' key) into Alt_L or into Alt_R, and then use the
|
|
|
913 commands show above to make them modifier keys.
|
|
|
914
|
|
|
915 Note that if you have Alt keys but no Meta keys, Emacs translates Alt
|
|
|
916 into Meta. This is because of the great importance of Meta in Emacs.
|
|
|
917
|
|
|
918 * `Pid xxx killed due to text modification or page I/O error'
|
|
|
919
|
|
|
920 On HP/UX, you can get that error when the Emacs executable is on an NFS
|
|
|
921 file system. HP/UX responds this way if it tries to swap in a page and
|
|
|
922 does not get a response from the server within a timeout whose default
|
|
|
923 value is just ten seconds.
|
|
|
924
|
|
|
925 If this happens to you, extend the timeout period.
|
|
|
926
|
|
|
927 * `expand-file-name' fails to work on any but the machine you dumped Emacs on.
|
|
|
928
|
|
|
929 On Ultrix, if you use any of the functions which look up information
|
|
|
930 in the passwd database before dumping Emacs (say, by using
|
|
|
931 expand-file-name in site-init.el), then those functions will not work
|
|
|
932 in the dumped Emacs on any host but the one Emacs was dumped on.
|
|
|
933
|
|
|
934 The solution? Don't use expand-file-name in site-init.el, or in
|
|
|
935 anything it loads. Yuck - some solution.
|
|
|
936
|
|
|
937 I'm not sure why this happens; if you can find out exactly what is
|
|
|
938 going on, and perhaps find a fix or a workaround, please let us know.
|
|
|
939 Perhaps the YP functions cache some information, the cache is included
|
|
|
940 in the dumped Emacs, and is then inaccurate on any other host.
|
|
|
941
|
|
|
942 * On some variants of SVR4, Emacs does not work at all with X.
|
|
|
943
|
|
|
944 Try defining BROKEN_FIONREAD in your config.h file. If this solves
|
|
|
945 the problem, please send a bug report to tell us this is needed; be
|
|
|
946 sure to say exactly what type of machine and system you are using.
|
|
|
947
|
|
|
948 * Emacs fails to understand most Internet host names, even though
|
|
|
949 the names work properly with other programs on the same system.
|
|
|
950 * Emacs won't work with X-windows if the value of DISPLAY is HOSTNAME:0.
|
|
|
951 * GNUs can't make contact with the specified host for nntp.
|
|
|
952
|
|
|
953 This typically happens on Suns and other systems that use shared
|
|
|
954 libraries. The cause is that the site has installed a version of the
|
|
|
955 shared library which uses a name server--but has not installed a
|
|
|
956 similar version of the unshared library which Emacs uses.
|
|
|
957
|
|
|
958 The result is that most programs, using the shared library, work with
|
|
|
959 the nameserver, but Emacs does not.
|
|
|
960
|
|
|
961 The fix is to install an unshared library that corresponds to what you
|
|
|
962 installed in the shared library, and then relink Emacs.
|
|
|
963
|
|
|
964 On SunOS 4.1, simply define HAVE_RES_INIT.
|
|
|
965
|
|
|
966 If you have already installed the name resolver in the file libresolv.a,
|
|
|
967 then you need to compile Emacs to use that library. The easiest way to
|
|
|
968 do this is to add to config.h a definition of LIBS_SYSTEM, LIBS_MACHINE
|
|
|
969 or LIB_STANDARD which uses -lresolv. Watch out! If you redefine a macro
|
|
|
970 that is already in use in your configuration to supply some other libraries,
|
|
|
971 be careful not to lose the others.
|
|
|
972
|
|
|
973 Thus, you could start by adding this to config.h:
|
|
|
974
|
|
|
975 #define LIBS_SYSTEM -lresolv
|
|
|
976
|
|
|
977 Then if this gives you an error for redefining a macro, and you see that
|
|
|
978 the s- file defines LIBS_SYSTEM as -lfoo -lbar, you could change config.h
|
|
|
979 again to say this:
|
|
|
980
|
|
|
981 #define LIBS_SYSTEM -lresolv -lfoo -lbar
|
|
|
982
|
|
|
983 * On a Sun running SunOS 4.1.1, you get this error message from GNU ld:
|
|
|
984
|
|
|
985 /lib/libc.a(_Q_sub.o): Undefined symbol __Q_get_rp_rd referenced from text segment
|
|
|
986
|
|
|
987 The problem is in the Sun shared C library, not in GNU ld.
|
|
|
988
|
|
|
989 The solution is to install Patch-ID# 100267-03 from Sun.
|
|
|
990
|
|
|
991 * SunOS 4.1.2: undefined symbol _get_wmShellWidgetClass
|
|
|
992
|
|
|
993 Apparently the version of libXmu.so.a that Sun ships is hosed: it's missing
|
|
|
994 some stuff that is in libXmu.a (the static version). Sun has a patch for
|
|
|
995 this, but a workaround is to use the static version of libXmu, by changing
|
|
|
996 the link command from "-lXmu" to "-Bstatic -lXmu -Bdynamic". If you have
|
|
|
997 OpenWindows 3.0, ask Sun for these patches:
|
|
|
998 100512-02 4.1.x OpenWindows 3.0 libXt Jumbo patch
|
|
|
999 100573-03 4.1.x OpenWindows 3.0 undefined symbols with shared libXmu
|
|
|
1000
|
|
|
1001 * Random other SunOS 4.1.[12] link errors.
|
|
|
1002
|
|
|
1003 The X headers and libraries that Sun ships in /usr/{include,lib}/X11 are
|
|
|
1004 broken. Use the ones in /usr/openwin/{include,lib} instead.
|
|
|
1005
|
|
|
1006 * Bus errors on startup when compiled with Sun's "acc" (in the routine
|
|
|
1007 make_string_internal() called from initialize_environment_alist())
|
|
|
1008
|
|
|
1009 The Sun ANSI compiler doesn't place uninitialized static variables in BSS
|
|
|
1010 space like other compilers do. This breaks emacs. If you want to use acc,
|
|
|
1011 you need to make the file "lastfile.o" be the *first* file in the link
|
|
|
1012 command. Better yet, use Lucid C or GCC.
|
|
|
1013
|
|
|
1014 * The compiler generates lots and lots of syntax errors.
|
|
|
1015
|
|
|
1016 Are you using an ANSI C compiler, like lcc or gcc? The SunOS 4.1 bundled cc
|
|
|
1017 is not ANSI.
|
|
|
1018
|
|
|
1019 If X has not been configured to compile itself using lcc, gcc, or another ANSI
|
|
|
1020 compiler, then you will have to hack the automatically-generated makefile in
|
|
|
1021 the `lwlib' directory by hand to make it use an ANSI compiler.
|
|
|
1022
|
|
|
1023 * When using gcc, you get the error message "undefined symbol __fixunsdfsi".
|
|
|
1024 * When using gcc, you get the error message "undefined symbol __main".
|
|
|
1025
|
|
|
1026 This means that you need to link with the gcc library. It may be called
|
|
|
1027 "gcc-gnulib" or "libgcc.a"; figure out where it is, and define LIB_GCC in
|
|
|
1028 config.h to point to it.
|
|
|
1029
|
|
|
1030 It may also work to use the GCC version of `ld' instead of the standard one.
|
|
|
1031
|
|
|
1032 * When compiling with X11, you get "undefined symbol _XtStrings".
|
|
|
1033
|
|
|
1034 This means that you are trying to link emacs against the X11r4 version of
|
|
|
1035 libXt.a, but you have compiled either Emacs or the code in the lwlib
|
|
|
1036 subdirectory with the X11r5 header files. That doesn't work.
|
|
|
1037
|
|
|
1038 Remember, you can't compile lwlib for r4 and emacs for r5, or vice versa.
|
|
|
1039 They must be in sync.
|
|
|
1040
|
|
|
1041 * Self documentation messages are garbled.
|
|
|
1042
|
|
|
1043 This means that the file `etc/DOC-...' doesn't properly correspond
|
|
|
1044 with the Emacs executable. Redumping Emacs and then installing the
|
|
|
1045 corresponding pair of files should fix the problem.
|
|
|
1046
|
|
|
1047 * Trouble using ptys on AIX.
|
|
|
1048
|
|
|
1049 People often install the pty devices on AIX incorrectly.
|
|
|
1050 Use `smit pty' to reinstall them properly.
|
|
|
1051
|
|
|
1052 * Shell mode on HP/UX gives the message, "`tty`: Ambiguous".
|
|
|
1053
|
|
|
1054 christos@theory.tn.cornell.edu says:
|
|
|
1055
|
|
|
1056 The problem is that in your .cshrc you have something that tries to
|
|
|
1057 execute `tty`. If you are not running the shell on a real tty then
|
|
|
1058 tty will print "not a tty". Csh expects one word in some places,
|
|
|
1059 but tty is giving it back 3.
|
|
|
1060
|
|
|
1061 The solution is to add a pair of quotes around `tty` to make it a single
|
|
|
1062 word:
|
|
|
1063
|
|
|
1064 if (`tty` == "/dev/console")
|
|
|
1065
|
|
|
1066 should be changed to:
|
|
|
1067
|
|
|
1068 if ("`tty`" == "/dev/console")
|
|
|
1069
|
|
|
1070 Even better, move things that set up terminal sections out of .cshrc
|
|
|
1071 and into .login.
|
|
|
1072
|
|
|
1073 * With process-connection-type set to t, each line of subprocess output is
|
|
|
1074 terminated with a ^M, making ange-ftp and GNUS not work.
|
|
|
1075
|
|
|
1076 On SunOS systems, this problem has been seen to be a result of an incomplete
|
|
|
1077 installation of gcc 2.2 which allowed some non-ANSI compatible include files
|
|
|
1078 into the compilation. In particular this affected virtually all ioctl() calls.
|
|
|
1079
|
|
|
1080 * Once you pull down a menu from the menubar, it won't go away.
|
|
|
1081
|
|
|
1082 It has been claimed that this is caused by a bug in certain very old (1990?)
|
|
|
1083 versions of the twm window manager. It doesn't happen with recent vintages,
|
|
|
1084 or with other window managers.
|
|
|
1085
|
|
|
1086 * Emacs ignores the "help" key when running OLWM.
|
|
|
1087
|
|
|
1088 OLWM grabs the help key, and retransmits it to the appropriate client using
|
|
|
1089 XSendEvent. Allowing emacs to react to synthetic events is a security hole,
|
|
|
1090 so this is turned off by default. You can enable it by setting the variable
|
|
|
1091 x-allow-sendevents to t. You can also cause fix this by telling OLWM to not
|
|
|
1092 grab the help key, with the null binding "OpenWindows.KeyboardCommand.Help:".
|
|
|
1093
|
|
|
1094 * Something awful happens when I type M-ESC, instead of `eval-expression'.
|
|
|
1095
|
|
|
1096 MWM intercepts this and several other keys. Turn this off by adding this to
|
|
|
1097 your resources: "mwm*keyBindings: NoKeyBindings".
|
|
|
1098
|
|
|
1099 * Using X Windows, control-shift-leftbutton makes Emacs hang.
|
|
|
1100
|
|
|
1101 Use the shell command `xset bc' to make the old X Menu package work.
|
|
|
1102
|
|
|
1103 * Emacs running under X Windows does not handle mouse clicks.
|
|
|
1104 * `emacs -geometry 80x20' finds a file named `80x20'.
|
|
|
1105
|
|
|
1106 One cause of such problems is having (setq term-file-prefix nil) in
|
|
|
1107 your .emacs file. Another cause is a bad value of EMACSLOADPATH in
|
|
|
1108 the environment.
|
|
|
1109
|
|
|
1110 * Emacs gets error message from linker on Sun.
|
|
|
1111
|
|
|
1112 If the error message says that a symbol such as `f68881_used' or
|
|
|
1113 `ffpa_used' or `start_float' is undefined, this probably indicates
|
|
|
1114 that you have compiled some libraries, such as the X libraries,
|
|
|
1115 with a floating point option other than the default.
|
|
|
1116
|
|
|
1117 It's not terribly hard to make this work with small changes in
|
|
|
1118 crt0.c together with linking with Fcrt1.o, Wcrt1.o or Mcrt1.o.
|
|
|
1119 However, the easiest approach is to build Xlib with the default
|
|
|
1120 floating point option: -fsoft to decide at run time what hardware
|
|
|
1121 is available.
|
|
|
1122
|
|
|
1123 * Keyboard input gets confused after a beep when using a DECserver
|
|
|
1124 as a concentrator.
|
|
|
1125
|
|
|
1126 This problem seems to be a matter of configuring the DECserver to use
|
|
|
1127 7 bit characters rather than 8 bit characters.
|
|
|
1128
|
|
|
1129 * M-x shell persistently reports "Process shell exited abnormally with code 1".
|
|
|
1130
|
|
|
1131 This happened on Suns as a result of what is said to be a bug in Sunos
|
|
|
1132 version 4.0.x. The only fix was to reboot the machine.
|
|
|
1133
|
|
|
1134 * Programs running under terminal emulator do not recognize `emacs'
|
|
|
1135 terminal type.
|
|
|
1136
|
|
|
1137 The cause of this is a shell startup file that sets the TERMCAP
|
|
|
1138 environment variable. The terminal emulator uses that variable to
|
|
|
1139 provide the information on the special terminal type that Emacs
|
|
|
1140 emulates.
|
|
|
1141
|
|
|
1142 Rewrite your shell startup file so that it does not change TERMCAP
|
|
|
1143 in such a case. You could use the following conditional which sets
|
|
|
1144 it only if it is undefined.
|
|
|
1145
|
|
|
1146 if ( ! ${?TERMCAP} ) setenv TERMCAP ~/my-termcap-file
|
|
|
1147
|
|
|
1148 Or you could set TERMCAP only when you set TERM--which should not
|
|
|
1149 happen in a non-login shell.
|
|
|
1150
|
|
|
1151 * Problem with remote X server on Suns.
|
|
|
1152
|
|
|
1153 On a Sun, running Emacs on one machine with the X server on another
|
|
|
1154 may not work if you have used the unshared system libraries. This
|
|
|
1155 is because the unshared libraries fail to use YP for host name lookup.
|
|
|
1156 As a result, the host name you specify may not be recognized.
|
|
|
1157
|
|
|
1158 * Shell mode ignores interrupts on Apollo Domain
|
|
|
1159
|
|
|
1160 You may find that M-x shell prints the following message:
|
|
|
1161
|
|
|
1162 Warning: no access to tty; thus no job control in this shell...
|
|
|
1163
|
|
|
1164 This can happen if there are not enough ptys on your system.
|
|
|
1165 Here is how to make more of them.
|
|
|
1166
|
|
|
1167 % cd /dev
|
|
|
1168 % ls pty*
|
|
|
1169 # shows how many pty's you have. I had 8, named pty0 to pty7)
|
|
|
1170 % /etc/crpty 8
|
|
|
1171 # creates eight new pty's
|
|
|
1172
|
|
|
1173 * Fatal signal in the command temacs -l loadup inc dump
|
|
|
1174
|
|
|
1175 This command is the final stage of building Emacs. It is run by the
|
|
|
1176 Makefile in the src subdirectory, or by build.com on VMS.
|
|
|
1177
|
|
|
1178 It has been known to get fatal errors due to insufficient swapping
|
|
|
1179 space available on the machine.
|
|
|
1180
|
|
|
1181 On 68000's, it has also happened because of bugs in the
|
|
|
1182 subroutine `alloca'. Verify that `alloca' works right, even
|
|
|
1183 for large blocks (many pages).
|
|
|
1184
|
|
|
1185 * test-distrib says that the distribution has been clobbered
|
|
|
1186 * or, temacs prints "Command key out of range 0-127"
|
|
|
1187 * or, temacs runs and dumps xemacs, but xemacs totally fails to work.
|
|
|
1188 * or, temacs gets errors dumping xemacs
|
|
|
1189
|
|
|
1190 This can be because the .elc files have been garbled. Do not be
|
|
|
1191 fooled by the fact that most of a .elc file is text: these are
|
|
|
1192 binary files and can contain all 256 byte values.
|
|
|
1193
|
|
|
1194 In particular `shar' cannot be used for transmitting GNU Emacs.
|
|
|
1195 It typically truncates "lines". What appear to be "lines" in
|
|
|
1196 a binary file can of course be of any length. Even once `shar'
|
|
|
1197 itself is made to work correctly, `sh' discards null characters
|
|
|
1198 when unpacking the shell archive.
|
|
|
1199
|
|
|
1200 I have also seen character \177 changed into \377. I do not know
|
|
|
1201 what transfer means caused this problem. Various network
|
|
|
1202 file transfer programs are suspected of clobbering the high bit.
|
|
|
1203
|
|
|
1204 If you have a copy of Emacs that has been damaged in its
|
|
|
1205 nonprinting characters, you can fix them:
|
|
|
1206
|
|
|
1207 1) Record the names of all the .elc files.
|
|
|
1208 2) Delete all the .elc files.
|
|
|
1209 3) Recompile alloc.c with a value of PURESIZE twice as large.
|
|
|
1210 You might as well save the old alloc.o.
|
|
|
1211 4) Remake xemacs. It should work now.
|
|
|
1212 5) Running xemacs, do Meta-x byte-compile-file repeatedly
|
|
|
1213 to recreate all the .elc files that used to exist.
|
|
|
1214 You may need to increase the value of the variable
|
|
|
1215 max-lisp-eval-depth to succeed in running the compiler interpreted
|
|
|
1216 on certain .el files. 400 was sufficient as of last report.
|
|
|
1217 6) Reinstall the old alloc.o (undoing changes to alloc.c if any)
|
|
|
1218 and remake temacs.
|
|
|
1219 7) Remake xemacs. It should work now, with valid .elc files.
|
|
|
1220
|
|
|
1221 * temacs prints "Pure Lisp storage exhausted"
|
|
|
1222
|
|
|
1223 This means that the Lisp code loaded from the .elc and .el
|
|
|
1224 files during temacs -l loadup inc dump took up more
|
|
|
1225 space than was allocated.
|
|
|
1226
|
|
|
1227 This could be caused by
|
|
|
1228 1) adding code to the preloaded Lisp files
|
|
|
1229 2) adding more preloaded files in loadup.el
|
|
|
1230 3) having a site-init.el or site-load.el which loads files.
|
|
|
1231 Note that ANY site-init.el or site-load.el is nonstandard;
|
|
|
1232 if you have received Emacs from some other site
|
|
|
1233 and it contains a site-init.el or site-load.el file, consider
|
|
|
1234 deleting that file.
|
|
|
1235 4) getting the wrong .el or .elc files
|
|
|
1236 (not from the directory you expected).
|
|
|
1237 5) deleting some .elc files that are supposed to exist.
|
|
|
1238 This would cause the source files (.el files) to be
|
|
|
1239 loaded instead. They take up more room, so you lose.
|
|
|
1240 6) a bug in the Emacs distribution which underestimates
|
|
|
1241 the space required.
|
|
|
1242
|
|
|
1243 If the need for more space is legitimate, use the --puresize option
|
|
|
1244 to `configure' to specify more pure space.
|
|
|
1245
|
|
|
1246 But in some of the cases listed above, this problem is a consequence
|
|
|
1247 of something else that is wrong. Be sure to check and fix the real
|
|
|
1248 problem.
|
|
|
1249
|
|
|
1250 * Changes made to .el files do not take effect.
|
|
|
1251
|
|
|
1252 You may have forgotten to recompile them into .elc files.
|
|
|
1253 Then the old .elc files will be loaded, and your changes
|
|
|
1254 will not be seen. To fix this, do M-x byte-recompile-directory
|
|
|
1255 and specify the directory that contains the Lisp files.
|
|
|
1256
|
|
|
1257 Note that you may get a warning when loading a .elc file that
|
|
|
1258 is older than the corresponding .el file.
|
|
|
1259
|
|
|
1260 * Things which should be bold or italic (such as the initial copyright notice)
|
|
|
1261 are not.
|
|
|
1262
|
|
|
1263 The fonts of the "bold" and "italic" faces are generated from the font of
|
|
|
1264 the "default" face; in this way, your bold and italic fonts will have the
|
|
|
1265 appropriate size and family. However, emacs can only be clever in this
|
|
|
1266 way if you have specified the default font using the XLFD (X Logical Font
|
|
|
1267 Description) format, which looks like
|
|
|
1268
|
|
|
1269 *-courier-medium-r-*-*-*-120-*-*-*-*-*-*
|
|
|
1270
|
|
|
1271 if you use any of the other, less strict font name formats, some of which
|
|
|
1272 look like
|
|
|
1273 lucidasanstypewriter-12
|
|
|
1274 and fixed
|
|
|
1275 and 9x13
|
|
|
1276
|
|
|
1277 then emacs won't be able to guess the names of the "bold" and "italic"
|
|
|
1278 versions. All X fonts can be referred to via XLFD-style names, so you
|
|
|
1279 should use those forms. See the man pages for X(1), xlsfonts(1), and
|
|
|
1280 xfontsel(1).
|
|
|
1281
|
|
|
1282 * The dumped Emacs (xemacs) crashes when run, trying to write pure data.
|
|
|
1283
|
|
|
1284 Two causes have been seen for such problems.
|
|
|
1285
|
|
|
1286 1) On a system where getpagesize is not a system call, it is defined
|
|
|
1287 as a macro. If the definition (in both unexec.c and malloc.c) is wrong,
|
|
|
1288 it can cause problems like this. You might be able to find the correct
|
|
|
1289 value in the man page for a.out (5).
|
|
|
1290
|
|
|
1291 2) Some systems allocate variables declared static among the
|
|
|
1292 initialized variables. Emacs makes all initialized variables in most
|
|
|
1293 of its files pure after dumping, but the variables declared static and
|
|
|
1294 not initialized are not supposed to be pure. On these systems you
|
|
|
1295 may need to add "#define static" to the m- or the s- file.
|
|
|
1296
|
|
|
1297 * Reading and writing files is very very slow.
|
|
|
1298
|
|
|
1299 Try evaluating the form (setq lock-directory nil) and see if that helps.
|
|
|
1300 There is a problem with file-locking on some systems (possibly related
|
|
|
1301 to NFS) that I don't understand. Please send mail to the address
|
|
|
1302 xemacs@xemacs.org if you figure this one out.
|
|
|
1303
|
|
|
1304 * Compilation errors on VMS.
|
|
|
1305
|
|
|
1306 Sorry, XEmacs does not work under VMS. You might consider working on
|
|
|
1307 the port if you really want to have XEmacs work under VMS.
|
|
|
1308
|
|
|
1309 * "Symbol's value as variable is void: unread-command-char".
|
|
|
1310 * "Wrong type argument: arrayp, #<keymap 143 entries>"
|
|
|
1311 * "Wrong type argument: stringp, [#<keypress-event return>]"
|
|
|
1312
|
|
|
1313 There are a few incompatible changes in XEmacs, and these are the
|
|
|
1314 symptoms. Some of the emacs-lisp code you are running needs to be
|
|
|
1315 updated to be compatible with XEmacs.
|
|
|
1316
|
|
|
1317 We have provided modified versions of several popular emacs packages (GNUS,
|
|
|
1318 VM, etc) which are compatible with this version of emacs. Check to make
|
|
|
1319 sure you have not set your load-path so that your private copies of these
|
|
|
1320 packages are being found before the versions in the lisp directory.
|
|
|
1321
|
|
|
1322 Make sure that your load-path and your $EMACSLOADPATH environment variable
|
|
|
1323 are not pointing at an Emacs18 lisp directory. This will cripple emacs.
|
|
|
1324
|
|
|
1325 * rmail or VM gets error getting new mail
|
|
|
1326
|
|
|
1327 rmail and VM get new mail from /usr/spool/mail/$USER using a program
|
|
|
1328 called `movemail'. This program interlocks with /bin/mail using
|
|
|
1329 the protocol defined by /bin/mail.
|
|
|
1330
|
|
|
1331 There are two different protocols in general use. One of them uses
|
|
|
1332 the `flock' system call. The other involves creating a lock file;
|
|
|
1333 `movemail' must be able to write in /usr/spool/mail in order to do
|
|
|
1334 this. You control which one is used by defining, or not defining,
|
|
|
1335 the macro MAIL_USE_FLOCK in config.h or the m- or s- file it includes.
|
|
|
1336 IF YOU DON'T USE THE FORM OF INTERLOCKING THAT IS NORMAL ON YOUR
|
|
|
1337 SYSTEM, YOU CAN LOSE MAIL!
|
|
|
1338
|
|
|
1339 If your system uses the lock file protocol, and fascist restrictions
|
|
|
1340 prevent ordinary users from writing the lock files in /usr/spool/mail,
|
|
|
1341 you may need to make `movemail' setgid to a suitable group such as
|
|
|
1342 `mail'. You can use these commands (as root):
|
|
|
1343
|
|
|
1344 chgrp mail movemail
|
|
|
1345 chmod 2755 movemail
|
|
|
1346
|
|
|
1347 If your system uses the lock file protocol, and fascist restrictions
|
|
|
1348 prevent ordinary users from writing the lock files in /usr/spool/mail,
|
|
|
1349 you may need to make `movemail' setgid to a suitable group such as
|
|
|
1350 `mail'. To do this, use the following commands (as root) after doing the
|
|
|
1351 make install.
|
|
|
1352
|
|
|
1353 chgrp mail movemail
|
|
|
1354 chmod 2755 movemail
|
|
|
1355
|
|
|
1356 Installation normally copies movemail from the build directory to an
|
|
|
1357 installation directory which is usually under /usr/local/lib. The
|
|
|
1358 installed copy of movemail is usually in the directory
|
|
|
1359 /usr/local/lib/emacs/VERSION/TARGET. You must change the group and
|
|
|
1360 mode of the installed copy; changing the group and mode of the build
|
|
|
1361 directory copy is ineffective.
|
|
|
1362
|
|
|
1363 * Emacs spontaneously displays "I-search: " at the bottom of the screen.
|
|
|
1364
|
|
|
1365 This means that Control-S/Control-Q (XON/XOFF) "flow control" is being
|
|
|
1366 used. C-s/C-q flow control is bad for Emacs editors because it takes
|
|
|
1367 away C-s and C-q as user commands. Since editors do not output long
|
|
|
1368 streams of text without user commands, there is no need for a
|
|
|
1369 user-issuable "stop output" command in an editor; therefore, a
|
|
|
1370 properly designed flow control mechanism would transmit all possible
|
|
|
1371 input characters without interference. Designing such a mechanism is
|
|
|
1372 easy, for a person with at least half a brain.
|
|
|
1373
|
|
|
1374 There are three possible reasons why flow control could be taking place:
|
|
|
1375
|
|
|
1376 1) Terminal has not been told to disable flow control
|
|
|
1377 2) Insufficient padding for the terminal in use
|
|
|
1378 3) Some sort of terminal concentrator or line switch is responsible
|
|
|
1379
|
|
|
1380 First of all, many terminals have a set-up mode which controls whether
|
|
|
1381 they generate XON/XOFF flow control characters. This must be set to
|
|
|
1382 "no XON/XOFF" in order for Emacs to work. Sometimes there is an
|
|
|
1383 escape sequence that the computer can send to turn flow control off
|
|
|
1384 and on. If so, perhaps the termcap `ti' string should turn flow
|
|
|
1385 control off, and the `te' string should turn it on.
|
|
|
1386
|
|
|
1387 Once the terminal has been told "no flow control", you may find it
|
|
|
1388 needs more padding. The amount of padding Emacs sends is controlled
|
|
|
1389 by the termcap entry for the terminal in use, and by the output baud
|
|
|
1390 rate as known by the kernel. The shell command `stty' will print
|
|
|
1391 your output baud rate; `stty' with suitable arguments will set it if
|
|
|
1392 it is wrong. Setting to a higher speed causes increased padding. If
|
|
|
1393 the results are wrong for the correct speed, there is probably a
|
|
|
1394 problem in the termcap entry. You must speak to a local Unix wizard
|
|
|
1395 to fix this. Perhaps you are just using the wrong terminal type.
|
|
|
1396
|
|
|
1397 For terminals that lack a "no flow control" mode, sometimes just
|
|
|
1398 giving lots of padding will prevent actual generation of flow control
|
|
|
1399 codes. You might as well try it.
|
|
|
1400
|
|
|
1401 If you are really unlucky, your terminal is connected to the computer
|
|
|
1402 through a concentrator which sends XON/XOFF flow control to the
|
|
|
1403 computer, or it insists on sending flow control itself no matter how
|
|
|
1404 much padding you give it. Unless you can figure out how to turn flow
|
|
|
1405 control off on this concentrator (again, refer to your local wizard),
|
|
|
1406 you are screwed! You should have the terminal or concentrator
|
|
|
1407 replaced with a properly designed one. In the mean time, some drastic
|
|
|
1408 measures can make Emacs semi-work.
|
|
|
1409
|
|
|
1410 You can make Emacs ignore C-s and C-q and let the operating system
|
|
|
1411 handle them. To do this on a per-session basis, just type M-x
|
|
|
1412 enable-flow-control RET. You will see a message that C-\ and C-^ are
|
|
|
1413 now translated to C-s and C-q. (Use the same command M-x
|
|
|
1414 enable-flow-control to turn *off* this special mode. It toggles flow
|
|
|
1415 control handling.)
|
|
|
1416
|
|
|
1417 If C-\ and C-^ are inconvenient for you (for example, if one of them
|
|
|
1418 is the escape character of your terminal concentrator), you can choose
|
|
|
1419 other characters by setting the variables flow-control-c-s-replacement
|
|
|
1420 and flow-control-c-q-replacement. But choose carefully, since all
|
|
|
1421 other control characters are already used by emacs.
|
|
|
1422
|
|
|
1423 IMPORTANT: if you type C-s by accident while flow control is enabled,
|
|
|
1424 Emacs output will freeze, and you will have to remember to type C-q in
|
|
|
1425 order to continue.
|
|
|
1426
|
|
|
1427 If you work in an environment where a majority of terminals of a
|
|
|
1428 certain type are flow control hobbled, you can use the function
|
|
|
1429 `enable-flow-control-on' to turn on this flow control avoidance scheme
|
|
|
1430 automatically. Here is an example:
|
|
|
1431
|
|
|
1432 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
|
|
|
1433
|
|
|
1434 If this isn't quite correct (e.g. you have a mixture of flow-control hobbled
|
|
|
1435 and good vt200 terminals), you can still run enable-flow-control
|
|
|
1436 manually.
|
|
|
1437
|
|
|
1438 I have no intention of ever redesigning the Emacs command set for the
|
|
|
1439 assumption that terminals use C-s/C-q flow control. XON/XOFF flow
|
|
|
1440 control technique is a bad design, and terminals that need it are bad
|
|
|
1441 merchandise and should not be purchased. Now that X is becoming
|
|
|
1442 widespread, XON/XOFF seems to be on the way out. If you can get some
|
|
|
1443 use out of GNU Emacs on inferior terminals, more power to you, but I
|
|
|
1444 will not make Emacs worse for properly designed systems for the sake
|
|
|
1445 of inferior systems.
|
|
|
1446
|
|
|
1447 * Control-S and Control-Q commands are ignored completely.
|
|
|
1448
|
|
|
1449 For some reason, your system is using brain-damaged C-s/C-q flow
|
|
|
1450 control despite Emacs's attempts to turn it off. Perhaps your
|
|
|
1451 terminal is connected to the computer through a concentrator
|
|
|
1452 that wants to use flow control.
|
|
|
1453
|
|
|
1454 You should first try to tell the concentrator not to use flow control.
|
|
|
1455 If you succeed in this, try making the terminal work without
|
|
|
1456 flow control, as described in the preceding section.
|
|
|
1457
|
|
|
1458 If that line of approach is not successful, map some other characters
|
|
|
1459 into C-s and C-q using keyboard-translate-table. The example above
|
|
|
1460 shows how to do this with C-^ and C-\.
|
|
|
1461
|
|
|
1462 * Control-S and Control-Q commands are ignored completely on a net connection.
|
|
|
1463
|
|
|
1464 Some versions of rlogin (and possibly telnet) do not pass flow
|
|
|
1465 control characters to the remote system to which they connect.
|
|
|
1466 On such systems, emacs on the remote system cannot disable flow
|
|
|
1467 control on the local system.
|
|
|
1468
|
|
|
1469 One way to cure this is to disable flow control on the local host
|
|
|
1470 (the one running rlogin, not the one running rlogind) using the
|
|
|
1471 stty command, before starting the rlogin process. On many systems,
|
|
|
1472 "stty start u stop u" will do this.
|
|
|
1473
|
|
|
1474 Some versions of tcsh will prevent even this from working. One way
|
|
|
1475 around this is to start another shell before starting rlogin, and
|
|
|
1476 issue the stty command to disable flow control from that shell.
|
|
|
1477
|
|
|
1478 If none of these methods work, the best solution is to type
|
|
|
1479 M-x enable-flow-control at the beginning of your emacs session, or
|
|
|
1480 if you expect the problem to continue, add a line such as the
|
|
|
1481 following to your .emacs (on the host running rlogind):
|
|
|
1482
|
|
|
1483 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
|
|
|
1484
|
|
|
1485 See the entry about spontaneous display of I-search (above) for more
|
|
|
1486 info.
|
|
|
1487
|
|
|
1488 * Screen is updated wrong, but only on one kind of terminal.
|
|
|
1489
|
|
|
1490 This could mean that the termcap entry you are using for that
|
|
|
1491 terminal is wrong, or it could mean that Emacs has a bug handing
|
|
|
1492 the combination of features specified for that terminal.
|
|
|
1493
|
|
|
1494 The first step in tracking this down is to record what characters
|
|
|
1495 Emacs is sending to the terminal. Execute the Lisp expression
|
|
|
1496 (open-termscript "./emacs-script") to make Emacs write all
|
|
|
1497 terminal output into the file ~/emacs-script as well; then do
|
|
|
1498 what makes the screen update wrong, and look at the file
|
|
|
1499 and decode the characters using the manual for the terminal.
|
|
|
1500 There are several possibilities:
|
|
|
1501
|
|
|
1502 1) The characters sent are correct, according to the terminal manual.
|
|
|
1503
|
|
|
1504 In this case, there is no obvious bug in Emacs, and most likely you
|
|
|
1505 need more padding, or possibly the terminal manual is wrong.
|
|
|
1506
|
|
|
1507 2) The characters sent are incorrect, due to an obscure aspect
|
|
|
1508 of the terminal behavior not described in an obvious way
|
|
|
1509 by termcap.
|
|
|
1510
|
|
|
1511 This case is hard. It will be necessary to think of a way for
|
|
|
1512 Emacs to distinguish between terminals with this kind of behavior
|
|
|
1513 and other terminals that behave subtly differently but are
|
|
|
1514 classified the same by termcap; or else find an algorithm for
|
|
|
1515 Emacs to use that avoids the difference. Such changes must be
|
|
|
1516 tested on many kinds of terminals.
|
|
|
1517
|
|
|
1518 3) The termcap entry is wrong.
|
|
|
1519
|
|
|
1520 See the file etc/TERMS for information on changes
|
|
|
1521 that are known to be needed in commonly used termcap entries
|
|
|
1522 for certain terminals.
|
|
|
1523
|
|
|
1524 4) The characters sent are incorrect, and clearly cannot be
|
|
|
1525 right for any terminal with the termcap entry you were using.
|
|
|
1526
|
|
|
1527 This is unambiguously an Emacs bug, and can probably be fixed
|
|
|
1528 in termcap.c, tparam.c, term.c, scroll.c, cm.c or dispnew.c.
|
|
|
1529
|
|
|
1530 * Output from Control-V is slow.
|
|
|
1531
|
|
|
1532 On many bit-map terminals, scrolling operations are fairly slow.
|
|
|
1533 Often the termcap entry for the type of terminal in use fails
|
|
|
1534 to inform Emacs of this. The two lines at the bottom of the screen
|
|
|
1535 before a Control-V command are supposed to appear at the top after
|
|
|
1536 the Control-V command. If Emacs thinks scrolling the lines is fast,
|
|
|
1537 it will scroll them to the top of the screen.
|
|
|
1538
|
|
|
1539 If scrolling is slow but Emacs thinks it is fast, the usual reason is
|
|
|
1540 that the termcap entry for the terminal you are using does not
|
|
|
1541 specify any padding time for the `al' and `dl' strings. Emacs
|
|
|
1542 concludes that these operations take only as much time as it takes to
|
|
|
1543 send the commands at whatever line speed you are using. You must
|
|
|
1544 fix the termcap entry to specify, for the `al' and `dl', as much
|
|
|
1545 time as the operations really take.
|
|
|
1546
|
|
|
1547 Currently Emacs thinks in terms of serial lines which send characters
|
|
|
1548 at a fixed rate, so that any operation which takes time for the
|
|
|
1549 terminal to execute must also be padded. With bit-map terminals
|
|
|
1550 operated across networks, often the network provides some sort of
|
|
|
1551 flow control so that padding is never needed no matter how slow
|
|
|
1552 an operation is. You must still specify a padding time if you want
|
|
|
1553 Emacs to realize that the operation takes a long time. This will
|
|
|
1554 cause padding characters to be sent unnecessarily, but they do
|
|
|
1555 not really cost much. They will be transmitted while the scrolling
|
|
|
1556 is happening and then discarded quickly by the terminal.
|
|
|
1557
|
|
|
1558 Most bit-map terminals provide commands for inserting or deleting
|
|
|
1559 multiple lines at once. Define the `AL' and `DL' strings in the
|
|
|
1560 termcap entry to say how to do these things, and you will have
|
|
|
1561 fast output without wasted padding characters. These strings should
|
|
|
1562 each contain a single %-spec saying how to send the number of lines
|
|
|
1563 to be scrolled. These %-specs are like those in the termcap
|
|
|
1564 `cm' string.
|
|
|
1565
|
|
|
1566 You should also define the `IC' and `DC' strings if your terminal
|
|
|
1567 has a command to insert or delete multiple characters. These
|
|
|
1568 take the number of positions to insert or delete as an argument.
|
|
|
1569
|
|
|
1570 A `cs' string to set the scrolling region will reduce the amount
|
|
|
1571 of motion you see on the screen when part of the screen is scrolled.
|
|
|
1572
|
|
|
1573 * Your Delete key sends a Backspace to the terminal, using an AIXterm.
|
|
|
1574
|
|
|
1575 The solution is to include in your .Xdefaults the lines:
|
|
|
1576
|
|
|
1577 *aixterm.Translations: #override <Key>BackSpace: string(0x7f)
|
|
|
1578 aixterm*ttyModes: erase ^?
|
|
|
1579
|
|
|
1580 This makes your Backspace key send DEL (ASCII 127).
|
|
|
1581
|
|
|
1582 * You type Control-H (Backspace) expecting to delete characters.
|
|
|
1583
|
|
|
1584 Emacs has traditionally used Control-H for help; unfortunately
|
|
|
1585 this interferes with its use as Backspace on TTY's. This has not
|
|
|
1586 been fixed due to an incredible arrogance on RMS's part. One way
|
|
|
1587 to solve this problem is to put this in your .emacs:
|
|
|
1588
|
|
|
1589 (global-set-key "\b" 'delete-backward-char)
|
|
|
1590 (global-set-key "\M-h" 'help-command)
|
|
|
1591
|
|
|
1592 This makes Control-H (Backspace) work sensibly, and moves help to
|
|
|
1593 Meta-H (ESC H).
|
|
|
1594
|
|
|
1595 Note that you can probably also access help using F1.
|
|
|
1596
|
|
|
1597 * Editing files through RFS gives spurious "file has changed" warnings.
|
|
|
1598 It is possible that a change in Emacs 18.37 gets around this problem,
|
|
|
1599 but in case not, here is a description of how to fix the RFS bug that
|
|
|
1600 causes it.
|
|
|
1601
|
|
|
1602 There was a serious pair of bugs in the handling of the fsync() system
|
|
|
1603 call in the RFS server.
|
|
|
1604
|
|
|
1605 The first is that the fsync() call is handled as another name for the
|
|
|
1606 close() system call (!!). It appears that fsync() is not used by very
|
|
|
1607 many programs; Emacs version 18 does an fsync() before closing files
|
|
|
1608 to make sure that the bits are on the disk.
|
|
|
1609
|
|
|
1610 This is fixed by the enclosed patch to the RFS server.
|
|
|
1611
|
|
|
1612 The second, more serious problem, is that fsync() is treated as a
|
|
|
1613 non-blocking system call (i.e., it's implemented as a message that
|
|
|
1614 gets sent to the remote system without waiting for a reply). Fsync is
|
|
|
1615 a useful tool for building atomic file transactions. Implementing it
|
|
|
1616 as a non-blocking RPC call (when the local call blocks until the sync
|
|
|
1617 is done) is a bad idea; unfortunately, changing it will break the RFS
|
|
|
1618 protocol. No fix was supplied for this problem.
|
|
|
1619
|
|
|
1620 (as always, your line numbers may vary)
|
|
|
1621
|
|
|
1622 % rcsdiff -c -r1.2 serversyscall.c
|
|
|
1623 RCS file: RCS/serversyscall.c,v
|
|
|
1624 retrieving revision 1.2
|
|
|
1625 diff -c -r1.2 serversyscall.c
|
|
|
1626 *** /tmp/,RCSt1003677 Wed Jan 28 15:15:02 1987
|
|
|
1627 --- serversyscall.c Wed Jan 28 15:14:48 1987
|
|
|
1628 ***************
|
|
|
1629 *** 163,169 ****
|
|
|
1630 /*
|
|
|
1631 * No return sent for close or fsync!
|
|
|
1632 */
|
|
|
1633 ! if (syscall == RSYS_close || syscall == RSYS_fsync)
|
|
|
1634 proc->p_returnval = deallocate_fd(proc, msg->m_args[0]);
|
|
|
1635 else
|
|
|
1636 {
|
|
|
1637 --- 166,172 ----
|
|
|
1638 /*
|
|
|
1639 * No return sent for close or fsync!
|
|
|
1640 */
|
|
|
1641 ! if (syscall == RSYS_close)
|
|
|
1642 proc->p_returnval = deallocate_fd(proc, msg->m_args[0]);
|
|
|
1643 else
|
|
|
1644 {
|
|
|
1645
|
|
|
1646 * Vax C compiler bugs affecting Emacs.
|
|
|
1647
|
|
|
1648 You may get one of these problems compiling Emacs:
|
|
|
1649
|
|
|
1650 foo.c line nnn: compiler error: no table entry for op STASG
|
|
|
1651 foo.c: fatal error in /lib/ccom
|
|
|
1652
|
|
|
1653 These are due to bugs in the C compiler; the code is valid C.
|
|
|
1654 Unfortunately, the bugs are unpredictable: the same construct
|
|
|
1655 may compile properly or trigger one of these bugs, depending
|
|
|
1656 on what else is in the source file being compiled. Even changes
|
|
|
1657 in header files that should not affect the file being compiled
|
|
|
1658 can affect whether the bug happens. In addition, sometimes files
|
|
|
1659 that compile correctly on one machine get this bug on another machine.
|
|
|
1660
|
|
|
1661 As a result, it is hard for me to make sure this bug will not affect
|
|
|
1662 you. I have attempted to find and alter these constructs, but more
|
|
|
1663 can always appear. However, I can tell you how to deal with it if it
|
|
|
1664 should happen. The bug comes from having an indexed reference to an
|
|
|
1665 array of Lisp_Objects, as an argument in a function call:
|
|
|
1666 Lisp_Object *args;
|
|
|
1667 ...
|
|
|
1668 ... foo (5, args[i], ...)...
|
|
|
1669 putting the argument into a temporary variable first, as in
|
|
|
1670 Lisp_Object *args;
|
|
|
1671 Lisp_Object tem;
|
|
|
1672 ...
|
|
|
1673 tem = args[i];
|
|
|
1674 ... foo (r, tem, ...)...
|
|
|
1675 causes the problem to go away.
|
|
|
1676 The `contents' field of a Lisp vector is an array of Lisp_Objects,
|
|
|
1677 so you may see the problem happening with indexed references to that.
|
|
|
1678
|
|
|
1679 * 68000 C compiler problems
|
|
|
1680
|
|
|
1681 Various 68000 compilers have different problems.
|
|
|
1682 These are some that have been observed.
|
|
|
1683
|
|
|
1684 ** Using value of assignment expression on union type loses.
|
|
|
1685 This means that x = y = z; or foo (x = z); does not work
|
|
|
1686 if x is of type Lisp_Object.
|
|
|
1687
|
|
|
1688 ** "cannot reclaim" error.
|
|
|
1689
|
|
|
1690 This means that an expression is too complicated. You get the correct
|
|
|
1691 line number in the error message. The code must be rewritten with
|
|
|
1692 simpler expressions.
|
|
|
1693
|
|
|
1694 ** XCONS, XSTRING, etc macros produce incorrect code.
|
|
|
1695
|
|
|
1696 If temacs fails to run at all, this may be the cause.
|
|
|
1697 Compile this test program and look at the assembler code:
|
|
|
1698
|
|
|
1699 struct foo { char x; unsigned int y : 24; };
|
|
|
1700
|
|
|
1701 lose (arg)
|
|
|
1702 struct foo arg;
|
|
|
1703 {
|
|
|
1704 test ((int *) arg.y);
|
|
|
1705 }
|
|
|
1706
|
|
|
1707 If the code is incorrect, your compiler has this problem.
|
|
|
1708 In the XCONS, etc., macros in lisp.h you must replace (a).u.val with
|
|
|
1709 ((a).u.val + coercedummy) where coercedummy is declared as int.
|
|
|
1710
|
|
|
1711 This problem will not happen if the m-...h file for your type
|
|
|
1712 of machine defines NO_UNION_TYPE. That is the recommended setting now.
|
|
|
1713
|
|
|
1714 * C compilers lose on returning unions
|
|
|
1715
|
|
|
1716 I hear that some C compilers cannot handle returning a union type.
|
|
|
1717 Most of the functions in GNU Emacs return type Lisp_Object, which is
|
|
|
1718 defined as a union on some rare architectures.
|
|
|
1719
|
|
|
1720 This problem will not happen if the m-...h file for your type
|
|
|
1721 of machine defines NO_UNION_TYPE. That is the recommended setting now.
|
|
44
|
1722
|
|
|
1723 * On Irix 5.x and 6.x, the dumped XEmacs (xemacs) core dumps when executed
|
|
|
1724 on another machine, or after newer SGI IRIX patches have been installed.
|
|
|
1725
|
|
|
1726 The xemacs binary must be executed with the same "libc.so" file which was used
|
|
|
1727 when the xemacs binary was dumped. Some SGI IRIX patches updates this file.
|
|
|
1728 Make sure that all machines using the xemacs binary are using the same
|
|
|
1729 set of IRIX patches. If xemacs core dumps after a patch upgrade then you
|
|
|
1730 will have to redump it from temacs.
|
