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