|
88
|
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)
|
|
120
|
6 (updated for 20.1)
|
|
|
7
|
|
|
8 * Watch out for .emacs file
|
|
|
9
|
|
|
10 ~/.emacs is your Emacs init file. If you observe strange problems,
|
|
|
11 invoke XEmacs with the `-q' option and see if you can repeat the
|
|
|
12 problem.
|
|
|
13
|
|
|
14 * "Symbol's value as variable is void: unread-command-char".
|
|
|
15 * "Wrong type argument: arrayp, #<keymap 143 entries>"
|
|
|
16 * "Wrong type argument: stringp, [#<keypress-event return>]"
|
|
|
17
|
|
|
18 There are a few incompatible changes in XEmacs, and these are the
|
|
|
19 symptoms. Some of the emacs-lisp code you are running needs to be
|
|
|
20 updated to be compatible with XEmacs.
|
|
|
21
|
|
|
22 The code should not treat keymaps as arrays (use `define-key', etc.),
|
|
|
23 should not use obsolete variables like `unread-command-char' (use
|
|
|
24 `unread-command-event'). Many (most) of the new ways of doing things
|
|
|
25 are compatible in GNU Emacs and XEmacs.
|
|
|
26
|
|
|
27 Modern Emacs packages (Gnus, VM, etc) are written cleanly, as to
|
|
|
28 support GNU Emacs and XEmacs. We have provided modified versions of
|
|
|
29 several popular emacs packages (dired, etc) which are compatible with
|
|
|
30 this version of emacs. Check to make sure you have not set your
|
|
|
31 load-path so that your private copies of these packages are being
|
|
|
32 found before the versions in the lisp directory.
|
|
|
33
|
|
|
34 Make sure that your load-path and your $EMACSLOADPATH environment
|
|
|
35 variable are not pointing at an Emacs18 lisp directory. This will
|
|
|
36 cripple emacs.
|
|
0
|
37
|
|
|
38 * On Irix, I don't see the toolbar icons and I'm getting lots of
|
|
|
39 entries in the warnings buffer.
|
|
|
40
|
|
|
41 SGI ships a really old Xpm library in /usr/lib which does not work at
|
|
|
42 all well with XEmacs. The solution is to install your own copy of the
|
|
|
43 latest version of Xpm somewhere and then use the --site-includes and
|
|
|
44 --site-libraries flags to tell configure where to find it.
|
|
|
45
|
|
|
46 * On Digital UNIX, the DEC C compiler might have a problem compiling
|
|
|
47 some files.
|
|
|
48
|
|
|
49 In particular, src/extents.c and src/faces.c might cause the DEC C
|
|
|
50 compiler to abort. When this happens: cd src, compile the files by
|
|
|
51 hand, cd .., and redo the "make" command. When recompiling the files by
|
|
|
52 hand, use the old C compiler for the following versions of Digital UNIX:
|
|
|
53 - V3.n: Remove "-migrate" from the compile command.
|
|
|
54 - V4.n: Add "-oldc" to the compile command.
|
|
|
55
|
|
|
56 * On HPUX, the HP C compiler might have a problem compiling some files
|
|
|
57 with optimization.
|
|
|
58
|
|
|
59 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
|
|
|
60
|
|
|
61 Had to drop once again to level 2 optimization, at least to
|
|
|
62 compile lstream.c. Otherwise, I get a "variable is void: \if"
|
|
|
63 problem while dumping (this is a problem I already reported
|
|
|
64 with vanilla hpux 10.01 and 9.07, which went away after
|
|
|
65 applying patches for the C compiler). Trouble is I still
|
|
|
66 haven't found the same patch for hpux 10.10, and I don't
|
|
|
67 remember the patch numbers. I think potential XEmacs builders
|
|
|
68 on HP should be warned about this.
|
|
|
69
|
|
|
70 * On HPUX, you get "poll: Interrupted system call" message in the window
|
|
|
71 where XEmacs was launched.
|
|
|
72
|
|
|
73 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
|
|
|
74
|
|
|
75 I get a very strange problem when linking libc.a
|
|
|
76 dynamically: every event (mouse, keyboard, expose...) results
|
|
|
77 in a "poll: Interrupted system call" message in the window
|
|
|
78 where XEmacs was launched. Forcing a static link of libc.a
|
|
|
79 alone by adding /usr/lib/libc.a at the end of the link line
|
|
|
80 solves this. Note that my 9.07 build of 19.14b17 and my (old)
|
|
|
81 build of 19.13 both exhibit the same behaviour. I've tried
|
|
|
82 various hpux patches to no avail. If this problem cannot be
|
|
|
83 solved before the release date, binary kits for HP *must* be
|
|
|
84 linked statically against libc, otherwise this problem will
|
|
|
85 show up. (This is directed at whoever will volunteer for this
|
|
|
86 kit, as I won't be available to do it, unless 19.14 gets
|
|
|
87 delayed until mid-june ;-). I think this problem will be an FAQ
|
|
|
88 soon after the release otherwise.
|
|
|
89
|
|
|
90 * Native cc on SCO OpenServer 5 is now OK. Icc may still throw you
|
|
|
91 a curve. Here is what Robert Lipe <robertl@arnet.com> says:
|
|
|
92
|
|
|
93 Unlike XEmacs 19.13, building with the native cc on SCO OpenServer 5
|
|
|
94 now produces a functional binary. I will typically build this
|
|
|
95 configuration for COFF with:
|
|
|
96
|
|
|
97 /path_to_XEmacs_source/configure --with-gcc=no \
|
|
|
98 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
|
|
|
99 --with-xpm --with-xface --with-sound=nas
|
|
|
100
|
|
|
101 This version now supports ELF builds. I highly recommend this to
|
|
|
102 reduce the in-core footprint of XEmacs. This is now how I compile
|
|
|
103 all my test releases. Build it like this:
|
|
|
104
|
|
|
105 /path_to_XEmacs_source/configure --with-gcc=no \
|
|
|
106 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
|
|
|
107 --with-xpm --with-xface --with-sound=nas --dynamic
|
|
|
108
|
|
|
109 The compiler known as icc [ supplied with the OpenServer 5 Development
|
|
|
110 System ] generates a working binary, but it takes forever to generate
|
|
|
111 XEmacs. ICC also whines more about the code than /bin/cc does. I do
|
|
|
112 believe all its whining is legitimate, however. Note that you do
|
|
|
113 have to 'cd src ; make LD=icc' to avoid linker errors.
|
|
|
114
|
|
|
115 The way I handle the build procedure is:
|
|
|
116
|
|
|
117 /path_to_XEmacs_source/configure --with-gcc=no \
|
|
|
118 --site-includes=/usr/local/include --site-libraries=/usr/local/lib \
|
|
|
119 --with-xpm --with-xface --with-sound=nas --dynamic --compiler="icc"
|
|
|
120
|
|
|
121 *NOTE* I have the xpm, xface, and audio libraries and includes in
|
|
|
122 /usr/local/lib, /usr/local/include. If you don't have these,
|
|
|
123 don't include the "--with-*" arguments in any of my examples.
|
|
|
124
|
|
|
125 In previous versions of XEmacs, you had to override the defaults while
|
|
|
126 compiling font-lock.o and extents.o when building with icc. This seems
|
|
|
127 to no longer be true, but I'm including this old information in case it
|
|
|
128 resurfaces. The process I used was:
|
|
|
129
|
|
|
130 make -k
|
|
|
131 [ procure pizza, beer, repeat ]
|
|
|
132 cd src
|
|
|
133 make CC="icc -W0,-mP1COPT_max_tree_size=3000" font-lock.o extents.o
|
|
|
134 make LD=icc
|
|
|
135
|
|
|
136 If you want sound support, get the tls566 supplement from
|
|
|
137 ftp.sco.com:/TLS or any of its mirrors. It works just groovy
|
|
|
138 with XEmacs.
|
|
|
139
|
|
|
140 The M-x manual-entry is known not to work. If you know Lisp and would
|
|
|
141 like help in making it work, e-mail me at <robertl@dgii.com>
|
|
|
142
|
|
|
143 In earlier releases, gnuserv/gnuclient/gnudoit would open a frame
|
|
|
144 just fine, but the client would lock up and the server would
|
|
|
145 terminate when you used C-x # to close the frame. This is now
|
|
|
146 fixed in XEmacs.
|
|
|
147
|
|
|
148 In etc/ there are two files of note. emacskeys.sco and emacsstrs.sco.
|
|
|
149 The comments at the top of emacskeys.sco describe its function, and
|
|
|
150 the emacstrs.sco is a suitable candidate for /usr/lib/keyboard/strings
|
|
|
151 to take advantage of the keyboard map in emacskeys.sco.
|
|
|
152
|
|
120
|
153 * Don't use -O2 with gcc 2.7.2 under Linux without also using
|
|
|
154 -fno-strength-reduce.
|
|
0
|
155
|
|
120
|
156 gcc will generate incorrect code otherwise. This bug is present in at
|
|
|
157 least 2.6.x and 2.7.[0-2]. This bug has been fixed in GCC 2.7.2.1 and
|
|
|
158 later.
|
|
0
|
159
|
|
|
160 * Under some versions of OSF XEmacs runs fine if built without
|
|
|
161 optimization but will crash randomly if built with optimization.
|
|
|
162 Using 'cc -g' is not sufficient to eliminate all optimization. Try
|
|
|
163 'cc -g -O0' instead.
|
|
|
164
|
|
|
165 * On HP/UX configure selects gcc even though it isn't actually present.
|
|
|
166
|
|
|
167 Some versions of SoftBench have an executable called 'gcc' that is not
|
|
|
168 actually the GNU C compiler. Use the --with-gcc=no flag when running
|
|
|
169 configure.
|
|
|
170
|
|
|
171 * When Emacs tries to ring the bell, you get an error like
|
|
|
172
|
|
|
173 audio: sst_open: SETQSIZE" Invalid argument
|
|
|
174 audio: sst_close: SETREG MMR2, Invalid argument
|
|
|
175
|
|
|
176 you have probably compiled using an ANSI C compiler, but with non-ANSI include
|
|
|
177 files. In particular, on Suns, the file /usr/include/sun/audioio.h uses the
|
|
|
178 _IOW macro to define the constant AUDIOSETQSIZE. _IOW in turn uses a K&R
|
|
|
179 preprocessor feature that is now explicitly forbidden in ANSI preprocessors,
|
|
|
180 namely substitution inside character constants. All ANSI C compilers must
|
|
|
181 provide a workaround for this problem. Lucid's C compiler is shipped with a
|
|
|
182 new set of system include files. If you are using GCC, there is a script
|
|
|
183 called fixincludes that creates new versions of some system include files that
|
|
|
184 use this obsolete feature.
|
|
|
185
|
|
|
186 * The `Alt' key doesn't behave as `Meta' when running DECwindows.
|
|
|
187
|
|
|
188 The default DEC keyboard mapping has the Alt keys set up to generate the
|
|
|
189 keysym `Multi_key', which has a meaning to xemacs which is distinct from that
|
|
|
190 of the `Meta_L' and `Meta-R' keysyms. A second problem is that certain keys
|
|
|
191 have the Mod2 modifier attached to them for no adequately explored reason.
|
|
|
192 The correct fix is to pass this file to xmodmap upon starting X:
|
|
|
193
|
|
|
194 clear mod2
|
|
|
195 keysym Multi_key = Alt_L
|
|
|
196 add mod1 = Alt_L
|
|
|
197 add mod1 = Alt_R
|
|
|
198
|
|
|
199 * I get complaints about the mapping of my HP keyboard at startup, but I
|
|
|
200 haven't changed anything.
|
|
|
201
|
|
|
202 The default HP keymap is set up to have Mod1 assigned to two different keys:
|
|
|
203 Meta_L and Mode_switch (even though there is not actually a Mode_switch key on
|
|
|
204 the keyboard -- it uses an "imaginary" keycode.) There actually is a reason
|
|
|
205 for this, but it's not a good one. The correct fix is to execute this command
|
|
|
206 upon starting X:
|
|
|
207
|
|
|
208 xmodmap -e 'remove mod1 = Mode_switch'
|
|
|
209
|
|
|
210 * I have focus problems when I use `M-o' to switch to another screen without
|
|
|
211 using the mouse.
|
|
|
212
|
|
|
213 The focus issues with a program like XEmacs, which has multiple homogeneous
|
|
|
214 top-level windows, are very complicated, and as a result, most window managers
|
|
|
215 don't implement them correctly.
|
|
|
216
|
|
|
217 The R4/R5 version of twm (and all of its descendants) had buggy focus
|
|
|
218 handling; there is a patch in .../xemacs/etc/twm-patch which fixes this.
|
|
|
219 Sufficiently recent versions of tvtwm do not need this patch, but most other
|
|
|
220 versions of twm do. If you need to apply this patch, please try to get it
|
|
|
221 integrated by the maintainer of whichever version of twm you're using.
|
|
|
222
|
|
|
223 In addition, if you're using twm, make sure you have not specified
|
|
|
224 "NoTitleFocus" in your .tvtwmrc file. The very nature of this option makes
|
|
|
225 twm do some illegal focus tricks, even with the patch.
|
|
|
226
|
|
|
227 It is known that olwm and olvwm are buggy, and in different ways. If you're
|
|
|
228 using click-to-type mode, try using point-to-type, or vice versa.
|
|
|
229
|
|
|
230 In older versions of NCDwm, one could not even type at XEmacs windows. This
|
|
|
231 has been fixed in newer versions (2.4.3, and possibly earlier).
|
|
|
232
|
|
|
233 (Many people suggest that XEmacs should warp the mouse when focusing on
|
|
|
234 another screen in point-to-type mode. This is not ICCCM-compliant behavior.
|
|
|
235 Implementing such policy is the responsibility of the window manager itself,
|
|
|
236 it is not legal for a client to do this.)
|
|
|
237
|
|
|
238 * My buffers are full of \000 characters or otherwise corrupt.
|
|
|
239
|
|
|
240 Some compilers have trouble with gmalloc.c and ralloc.c; try recompiling
|
|
|
241 without optimization. If that doesn't work, try recompiling with
|
|
|
242 SYSTEM_MALLOC defined, and/or with REL_ALLOC undefined.
|
|
|
243
|
|
|
244 * Some packages that worked before now cause the error
|
|
|
245 Wrong type argument: arrayp, #<face ... >
|
|
|
246
|
|
|
247 Code which uses the `face' accessor functions must be recompiled with xemacs
|
|
|
248 19.9 or later. The functions whose callers must be recompiled are: face-font,
|
|
|
249 face-foreground, face-background, face-background-pixmap, and face-underline-p.
|
|
|
250 The .elc files generated by version 19.9 will work in 19.6 and 19.8, but older
|
|
|
251 .elc files which contain calls to these functions will not work in 19.9.
|
|
|
252
|
|
|
253 * On Solaris 2.* I get undefined symbols from libcurses.a.
|
|
|
254
|
|
|
255 You probably have /usr/ucblib/ on your LD_LIBRARY_PATH. Do the link with
|
|
|
256 LD_LIBRARY_PATH unset.
|
|
|
257
|
|
88
|
258 * On Solaris 2.* I cannot make alloc.o, glyphs.o or process.o.
|
|
|
259
|
|
|
260 The SparcWorks C compiler may have difficulty building those modules
|
|
|
261 with optimization level -xO4. Try using only "-fast" optimization
|
|
|
262 for just those modules. (Or use gcc).
|
|
|
263
|
|
0
|
264 * I don't have `xmkmf' and `imake' on my HP.
|
|
|
265
|
|
|
266 You can get these standard X tools by anonymous FTP to hpcvaaz.cv.hp.com.
|
|
|
267 Essentially all X programs need these.
|
|
|
268
|
|
|
269 * When emacs starts up, I get lots of warnings about unknown keysyms.
|
|
|
270
|
|
|
271 If you are running the prebuilt binaries, the Motif library expects to find
|
|
|
272 certain thing in the XKeysymDB file. This file is normally in /usr/lib/X11/
|
|
|
273 or in /usr/openwin/lib/. If you keep yours in a different place, set the
|
|
|
274 environment variable $XKEYSYMDB to point to it before starting emacs. If
|
|
|
275 you still have the problem after doing that, perhaps your version of X is
|
|
|
276 too old. There is a copy of the MIT X11R5 XKeysymDB file in the emacs `etc'
|
|
|
277 directory. Try using that one.
|
|
|
278
|
|
|
279 * My X resources used to work, and now some of them are being ignored.
|
|
|
280
|
|
|
281 Check the resources in .../etc/Emacs.ad (which is the same as the file
|
|
|
282 sample.Xdefaults). Perhaps some of the default resources built in to
|
|
|
283 emacs are now overriding your existing resources. Copy and edit the
|
|
|
284 resources in Emacs.ad as necessary.
|
|
|
285
|
|
|
286 * Solaris 2.3 /bin/sh coredumps during configuration.
|
|
|
287
|
|
|
288 This only occurs if you have LANG != C. This is a known bug with
|
|
|
289 /bin/sh fixed by installing Patch-ID# 101613-01.
|
|
|
290
|
|
|
291 * "Cannot find callback list" messages from dialog boxes on HPUX, in
|
|
120
|
292 Emacs built with Motif.
|
|
0
|
293
|
|
|
294 This problem resulted from a bug in GCC 2.4.5. Newer GCC versions
|
|
|
295 such as 2.7.0 fix the problem.
|
|
|
296
|
|
|
297 * On Irix 6.0, make tries (and fails) to build a program named unexelfsgi
|
|
|
298
|
|
|
299 A compiler bug inserts spaces into the string "unexelfsgi . o"
|
|
|
300 in src/Makefile. Edit src/Makefile, after configure is run,
|
|
|
301 find that string, and take out the spaces.
|
|
|
302
|
|
|
303 Compiler fixes in Irix 6.0.1 should eliminate this problem.
|
|
|
304
|
|
|
305 * With certain fonts, when the cursor appears on a character, the
|
|
120
|
306 character doesn't appear--you get a solid box instead.
|
|
0
|
307
|
|
88
|
308 One user on a Linux system reported that this problem went away with
|
|
|
309 installation of a new X server. The failing server was XFree86 3.1.1.
|
|
|
310 XFree86 3.1.2 works.
|
|
0
|
311
|
|
|
312 * On SunOS 4.1.3, Emacs unpredictably crashes in _yp_dobind_soft.
|
|
|
313
|
|
|
314 This happens if you configure Emacs specifying just `sparc-sun-sunos4'
|
|
|
315 on a system that is version 4.1.3. You must specify the precise
|
|
|
316 version number (or let configure figure out the configuration, which
|
|
|
317 it can do perfectly well for SunOS).
|
|
|
318
|
|
|
319 * On SunOS 4, Emacs processes keep going after you kill the X server
|
|
120
|
320 (or log out, if you logged in using X).
|
|
0
|
321
|
|
|
322 Someone reported that recompiling with GCC 2.7.0 fixed this problem.
|
|
|
323
|
|
|
324 * On AIX 4, some programs fail when run in a Shell buffer
|
|
120
|
325 with an error message like No terminfo entry for "unknown".
|
|
0
|
326
|
|
|
327 On AIX, many terminal type definitions are not installed by default.
|
|
|
328 `unknown' is one of them. Install the "Special Generic Terminal
|
|
|
329 Definitions" to make them defined.
|
|
|
330
|
|
|
331 * On SunOS, you get linker errors
|
|
|
332 ld: Undefined symbol
|
|
|
333 _get_wmShellWidgetClass
|
|
|
334 _get_applicationShellWidgetClass
|
|
|
335
|
|
|
336 The fix to this is to install patch 100573 for OpenWindows 3.0
|
|
|
337 or link libXmu statically.
|
|
|
338
|
|
|
339 * On AIX 4.1.2, linker error messages such as
|
|
|
340 ld: 0711-212 SEVERE ERROR: Symbol .__quous, found in the global symbol table
|
|
|
341 of archive /usr/lib/libIM.a, was not defined in archive member shr.o.
|
|
|
342
|
|
|
343 This is a problem in libIM.a. You can work around it by executing
|
|
|
344 these shell commands in the src subdirectory of the directory where
|
|
|
345 you build Emacs:
|
|
|
346
|
|
|
347 cp /usr/lib/libIM.a .
|
|
|
348 chmod 664 libIM.a
|
|
|
349 ranlib libIM.a
|
|
|
350
|
|
|
351 Then change -lIM to ./libIM.a in the command to link temacs (in
|
|
|
352 Makefile).
|
|
|
353
|
|
|
354 * Emacs exits with "X protocol error" when run with an X server for
|
|
120
|
355 Windows.
|
|
0
|
356
|
|
|
357 A certain X server for Windows had a bug which caused this.
|
|
|
358 Supposedly the newer 32-bit version of this server doesn't have the
|
|
|
359 problem.
|
|
|
360
|
|
|
361 * A position you specified in .Xdefaults is ignored, using twm.
|
|
|
362
|
|
|
363 twm normally ignores "program-specified" positions.
|
|
|
364 You can tell it to obey them with this command in your `.twmrc' file:
|
|
|
365
|
|
|
366 UsePPosition "on" #allow clents to request a position
|
|
|
367
|
|
|
368 * Compiling lib-src says there is no rule to make test-distrib.c.
|
|
|
369
|
|
|
370 This results from a bug in a VERY old version of GNU Sed. To solve
|
|
|
371 the problem, install the current version of GNU Sed, then rerun
|
|
|
372 Emacs's configure script.
|
|
|
373
|
|
|
374 * On Sunos 4.1.1, there are errors compiling sysdep.c.
|
|
|
375
|
|
|
376 If you get errors such as
|
|
|
377
|
|
|
378 "sysdep.c", line 2017: undefined structure or union
|
|
|
379 "sysdep.c", line 2017: undefined structure or union
|
|
|
380 "sysdep.c", line 2019: nodename undefined
|
|
|
381
|
|
|
382 This can result from defining LD_LIBRARY_PATH. It is very tricky
|
|
|
383 to use that environment variable with Emacs. The Emacs configure
|
|
|
384 script links many test programs with the system libraries; you must
|
|
|
385 make sure that the libraries available to configure are the same
|
|
|
386 ones available when you build Emacs.
|
|
|
387
|
|
|
388 * The right Alt key works wrong on German HP keyboards (and perhaps
|
|
120
|
389 other non-English HP keyboards too).
|
|
0
|
390
|
|
|
391 This is because HPUX defines the modifiers wrong in X. Here is a
|
|
|
392 shell script to fix the problem; be sure that it is run after VUE
|
|
|
393 configures the X server.
|
|
|
394
|
|
|
395 xmodmap 2> /dev/null - << EOF
|
|
|
396 keysym Alt_L = Meta_L
|
|
|
397 keysym Alt_R = Meta_R
|
|
|
398 EOF
|
|
|
399
|
|
|
400 xmodmap - << EOF
|
|
|
401 clear mod1
|
|
|
402 keysym Mode_switch = NoSymbol
|
|
|
403 add mod1 = Meta_L
|
|
|
404 keysym Meta_R = Mode_switch
|
|
|
405 add mod2 = Mode_switch
|
|
|
406 EOF
|
|
|
407
|
|
|
408 * The Emacs window disappears when you type M-q.
|
|
|
409
|
|
|
410 Some versions of the Open Look window manager interpret M-q as a quit
|
|
|
411 command for whatever window you are typing at. If you want to use
|
|
|
412 Emacs with that window manager, you should try to configure the window
|
|
|
413 manager to use some other command. You can disable the
|
|
|
414 shortcut keys entirely by adding this line to ~/.OWdefaults:
|
|
|
415
|
|
|
416 OpenWindows.WindowMenuAccelerators: False
|
|
|
417
|
|
|
418 * Emacs does not notice when you release the mouse.
|
|
|
419
|
|
|
420 There are reports that this happened with (some) Microsoft mice and
|
|
|
421 that replacing the mouse made it stop.
|
|
|
422
|
|
|
423 * Trouble using ptys on IRIX, or running out of ptys.
|
|
|
424
|
|
|
425 The program mkpts (which may be in `/usr/adm' or `/usr/sbin') needs to
|
|
|
426 be set-UID to root, or non-root programs like Emacs will not be able
|
|
|
427 to allocate ptys reliably.
|
|
|
428
|
|
|
429 * On Irix 5.2, unexelfsgi.c can't find cmplrs/stsupport.h.
|
|
|
430
|
|
|
431 The file cmplrs/stsupport.h was included in the wrong file set in the
|
|
|
432 Irix 5.2 distribution. You can find it in the optional fileset
|
|
|
433 compiler_dev, or copy it from some other Irix 5.2 system. A kludgy
|
|
|
434 workaround is to change unexelfsgi.c to include sym.h instead of
|
|
|
435 syms.h.
|
|
|
436
|
|
|
437 * Slow startup on Linux.
|
|
|
438
|
|
|
439 People using systems based on the Linux kernel sometimes report that
|
|
|
440 startup takes 10 to 15 seconds longer than `usual'.
|
|
|
441
|
|
|
442 This is because Emacs looks up the host name when it starts.
|
|
|
443 Normally, this takes negligible time; the extra delay is due to
|
|
|
444 improper system configuration. This problem can occur for both
|
|
|
445 networked and non-networked machines.
|
|
|
446
|
|
|
447 Here is how to fix the configuration. It requires being root.
|
|
|
448
|
|
|
449 ** Networked Case
|
|
|
450
|
|
|
451 First, make sure the files `/etc/hosts' and `/etc/host.conf' both
|
|
|
452 exist. The first line in the `/etc/hosts' file should look like this
|
|
|
453 (replace HOSTNAME with your host name):
|
|
|
454
|
|
88
|
455 127.0.0.1 localhost HOSTNAME
|
|
0
|
456
|
|
|
457 Also make sure that the `/etc/host.conf' files contains the following
|
|
|
458 lines:
|
|
|
459
|
|
|
460 order hosts, bind
|
|
|
461 multi on
|
|
|
462
|
|
|
463 Any changes, permanent and temporary, to the host name should be
|
|
|
464 indicated in the `/etc/hosts' file, since it acts a limited local
|
|
|
465 database of addresses and names (e.g., some SLIP connections
|
|
|
466 dynamically allocate ip addresses).
|
|
|
467
|
|
|
468 ** Non-Networked Case
|
|
|
469
|
|
|
470 The solution described in the networked case applies here as well.
|
|
|
471 However, if you never intend to network your machine, you can use a
|
|
|
472 simpler solution: create an empty `/etc/host.conf' file. The command
|
|
|
473 `touch /etc/host.conf' suffices to create the file. The `/etc/hosts'
|
|
|
474 file is not necessary with this approach.
|
|
|
475
|
|
|
476 * On Solaris 2.4, Dired hangs and C-g does not work. Or Emacs hangs
|
|
120
|
477 forever waiting for termination of a subprocess that is a zombie.
|
|
0
|
478
|
|
|
479 casper@fwi.uva.nl says the problem is in X11R6. Rebuild libX11.so
|
|
|
480 after changing the file xc/config/cf/sunLib.tmpl. Change the lines
|
|
|
481
|
|
|
482 #if ThreadedX
|
|
|
483 #define SharedX11Reqs -lthread
|
|
|
484 #endif
|
|
|
485
|
|
|
486 to:
|
|
|
487
|
|
|
488 #if OSMinorVersion < 4
|
|
|
489 #if ThreadedX
|
|
|
490 #define SharedX11Reqs -lthread
|
|
|
491 #endif
|
|
|
492 #endif
|
|
|
493
|
|
|
494 Be sure also to edit x/config/cf/sun.cf so that OSMinorVersion is 4
|
|
|
495 (as it should be for Solaris 2.4). The file has three definitions for
|
|
|
496 OSMinorVersion: the first is for x86, the second for SPARC under
|
|
|
497 Solaris, and the third for SunOS 4. Make sure to update the
|
|
|
498 definition for your type of machine and system.
|
|
|
499
|
|
|
500 Then do `make Everything' in the top directory of X11R6, to rebuild
|
|
|
501 the makefiles and rebuild X. The X built this way work only on
|
|
|
502 Solaris 2.4, not on 2.3.
|
|
|
503
|
|
|
504 For multithreaded X to work it necessary to install patch
|
|
|
505 101925-02 to fix problems in header files [2.4]. You need
|
|
|
506 to reinstall gcc or re-run just-fixinc after installing that
|
|
|
507 patch.
|
|
|
508
|
|
|
509 However, Frank Rust <frust@iti.cs.tu-bs.de> used a simpler solution:
|
|
|
510 he changed
|
|
|
511 #define ThreadedX YES
|
|
|
512 to
|
|
|
513 #define ThreadedX NO
|
|
|
514 in sun.cf and did `make World' to rebuild X11R6. Removing all
|
|
|
515 `-DXTHREAD*' flags and `-lthread' entries from lib/X11/Makefile and
|
|
|
516 typing 'make install' in that directory also seemed to work.
|
|
|
517
|
|
120
|
518 * With M-x enable-flow-control, you need to type C-\ twice to do
|
|
|
519 incremental search--a single C-\ gets no response.
|
|
0
|
520
|
|
|
521 This has been traced to communicating with your machine via kermit,
|
|
|
522 with C-\ as the kermit escape character. One solution is to use
|
|
|
523 another escape character in kermit. One user did
|
|
|
524
|
|
|
525 set escape-character 17
|
|
|
526
|
|
|
527 in his .kermrc file, to make C-q the kermit escape character.
|
|
|
528
|
|
|
529 * The Motif version of Emacs paints the screen a solid color.
|
|
|
530
|
|
|
531 This has been observed to result from the following X resource:
|
|
|
532
|
|
|
533 Emacs*default.attributeFont: -*-courier-medium-r-*-*-*-140-*-*-*-*-iso8859-*
|
|
|
534
|
|
|
535 That the resource has this effect indicates a bug in something, but we
|
|
|
536 do not yet know what. If it is an Emacs bug, we hope someone can
|
|
|
537 explain what the bug is so we can fix it. In the mean time, removing
|
|
|
538 the resource prevents the problem.
|
|
|
539
|
|
|
540 * Emacs gets hung shortly after startup, on Sunos 4.1.3.
|
|
|
541
|
|
|
542 We think this is due to a bug in Sunos. The word is that
|
|
|
543 one of these Sunos patches fixes the bug:
|
|
|
544
|
|
|
545 100075-11 100224-06 100347-03 100482-05 100557-02 100623-03 100804-03 101080-01
|
|
|
546 100103-12 100249-09 100496-02 100564-07 100630-02 100891-10 101134-01
|
|
|
547 100170-09 100296-04 100377-09 100507-04 100567-04 100650-02 101070-01 101145-01
|
|
|
548 100173-10 100305-15 100383-06 100513-04 100570-05 100689-01 101071-03 101200-02
|
|
|
549 100178-09 100338-05 100421-03 100536-02 100584-05 100784-01 101072-01 101207-01
|
|
|
550
|
|
|
551 We don't know which of these patches really matter. If you find out
|
|
|
552 which ones, please inform bug-gnu-emacs@prep.ai.mit.edu.
|
|
|
553
|
|
|
554 * Emacs aborts while starting up, only when run without X.
|
|
|
555
|
|
|
556 This problem often results from compiling Emacs with GCC when GCC was
|
|
|
557 installed incorrectly. The usual error in installing GCC is to
|
|
|
558 specify --includedir=/usr/include. Installation of GCC makes
|
|
|
559 corrected copies of the system header files. GCC is supposed to use
|
|
|
560 the corrected copies in preference to the original system headers.
|
|
|
561 Specifying --includedir=/usr/include causes the original system header
|
|
|
562 files to be used. On some systems, the definition of ioctl in the
|
|
|
563 original system header files is invalid for ANSI C and causes Emacs
|
|
|
564 not to work.
|
|
|
565
|
|
|
566 The fix is to reinstall GCC, and this time do not specify --includedir
|
|
|
567 when you configure it. Then recompile Emacs. Specifying --includedir
|
|
|
568 is appropriate only in very special cases and it should *never* be the
|
|
|
569 same directory where system header files are kept.
|
|
|
570
|
|
|
571 * The Compose key on a DEC keyboard does not work as Meta key.
|
|
|
572
|
|
|
573 This shell command should fix it:
|
|
|
574
|
|
|
575 xmodmap -e 'keycode 0xb1 = Meta_L'
|
|
|
576
|
|
|
577 * Regular expressions matching bugs on SCO systems.
|
|
|
578
|
|
|
579 On SCO, there are problems in regexp matching when Emacs is compiled
|
|
|
580 with the system compiler. The compiler version is "Microsoft C
|
|
|
581 version 6", SCO 4.2.0h Dev Sys Maintenance Supplement 01/06/93; Quick
|
|
|
582 C Compiler Version 1.00.46 (Beta). The solution is to compile with
|
|
|
583 GCC.
|
|
|
584
|
|
|
585 * On Sunos 4, you get the error ld: Undefined symbol __lib_version.
|
|
|
586
|
|
|
587 This is the result of using cc or gcc with the shared library meant
|
|
|
588 for acc (the Sunpro compiler). Check your LD_LIBRARY_PATH and delete
|
|
|
589 /usr/lang/SC2.0.1 or some similar directory.
|
|
|
590
|
|
|
591 * You can't select from submenus.
|
|
|
592
|
|
|
593 On certain systems, mouse-tracking and selection in top-level menus
|
|
|
594 works properly with the X toolkit, but neither of them works when you
|
|
|
595 bring up a submenu (such as Bookmarks or Compare or Apply Patch, in
|
|
|
596 the Files menu).
|
|
|
597
|
|
|
598 This works on most systems. There is speculation that the failure is
|
|
|
599 due to bugs in old versions of X toolkit libraries, but no one really
|
|
|
600 knows. If someone debugs this and finds the precise cause, perhaps a
|
|
|
601 workaround can be found.
|
|
|
602
|
|
|
603 * Unusable default font on SCO 3.2v4.
|
|
|
604
|
|
|
605 The Open Desktop environment comes with default X resource settings
|
|
|
606 that tell Emacs to use a variable-width font. Emacs cannot use such
|
|
|
607 fonts, so it does not work.
|
|
|
608
|
|
|
609 This is caused by the file /usr/lib/X11/app-defaults/ScoTerm, which is
|
|
|
610 the application-specific resource file for the `scoterm' terminal
|
|
|
611 emulator program. It contains several extremely general X resources
|
|
|
612 that affect other programs besides `scoterm'. In particular, these
|
|
|
613 resources affect Emacs also:
|
|
|
614
|
|
|
615 *Font: -*-helvetica-medium-r-*--12-*-p-*
|
|
|
616 *Background: scoBackground
|
|
|
617 *Foreground: scoForeground
|
|
|
618
|
|
|
619 The best solution is to create an application-specific resource file for
|
|
|
620 Emacs, /usr/lib/X11/app-defaults/Emacs, with the following contents:
|
|
|
621
|
|
|
622 Emacs*Font: -*-courier-medium-r-*-*-*-120-*-*-*-*-iso8859-1
|
|
|
623 Emacs*Background: white
|
|
|
624 Emacs*Foreground: black
|
|
|
625
|
|
|
626 (or whatever other defaults you prefer).
|
|
|
627
|
|
|
628 These resource files are not normally shared across a network of SCO
|
|
|
629 machines; you must create the file on each machine individually.
|
|
|
630
|
|
|
631 * rcs2log gives you the awk error message "too many fields".
|
|
|
632
|
|
|
633 This is due to an arbitrary limit in certain versions of awk.
|
|
|
634 The solution is to use gawk (GNU awk).
|
|
|
635
|
|
|
636 * Emacs is slow using X11R5 on HP/UX.
|
|
|
637
|
|
|
638 This happens if you use the MIT versions of the X libraries--it
|
|
|
639 doesn't run as fast as HP's version. People sometimes use the version
|
|
|
640 because they see the HP version doesn't have the libraries libXaw.a,
|
|
|
641 libXmu.a, libXext.a and others. HP/UX normally doesn't come with
|
|
|
642 those libraries installed. To get good performance, you need to
|
|
|
643 install them and rebuild Emacs.
|
|
|
644
|
|
|
645 * Loading fonts is very slow.
|
|
|
646
|
|
|
647 You might be getting scalable fonts instead of precomputed bitmaps.
|
|
|
648 Known scalable font directories are "Type1" and "Speedo". A font
|
|
|
649 directory contains scalable fonts if it contains the file
|
|
|
650 "fonts.scale".
|
|
|
651
|
|
|
652 If this is so, re-order your X windows font path to put the scalable
|
|
|
653 font directories last. See the documentatoin of `xset' for details.
|
|
|
654
|
|
|
655 With some X servers, it may be necessary to take the scalable font
|
|
|
656 directories out of your path entirely, at least for Emacs 19.26.
|
|
|
657 Changes in the future may make this unnecessary.
|
|
|
658
|
|
|
659 * On AIX 3.2.4, releasing Ctrl/Act key has no effect, if Shift is down.
|
|
|
660
|
|
|
661 Due to a feature of AIX, pressing or releasing the Ctrl/Act key is
|
|
|
662 ignored when the Shift, Alt or AltGr keys are held down. This can
|
|
|
663 lead to the keyboard being "control-locked"--ordinary letters are
|
|
|
664 treated as control characters.
|
|
|
665
|
|
|
666 You can get out of this "control-locked" state by pressing and
|
|
|
667 releasing Ctrl/Act while not pressing or holding any other keys.
|
|
|
668
|
|
|
669 * display-time causes kernel problems on ISC systems.
|
|
|
670
|
|
|
671 Under Interactive Unix versions 3.0.1 and 4.0 (and probably other
|
|
|
672 versions), display-time causes the loss of large numbers of STREVENT
|
|
|
673 cells. Eventually the kernel's supply of these cells is exhausted.
|
|
|
674 This makes emacs and the whole system run slow, and can make other
|
|
|
675 processes die, in particular pcnfsd.
|
|
|
676
|
|
|
677 Other emacs functions that communicate with remote processes may have
|
|
|
678 the same problem. Display-time seems to be far the worst.
|
|
|
679
|
|
|
680 The only known fix: Don't run display-time.
|
|
|
681
|
|
|
682 * On Solaris, C-x doesn't get through to Emacs when you use the console.
|
|
|
683
|
|
|
684 This is a Solaris feature (at least on Intel x86 cpus). Type C-r
|
|
|
685 C-r C-t, to toggle whether C-x gets through to Emacs.
|
|
|
686
|
|
|
687 * Error message `Symbol's value as variable is void: x', followed by
|
|
|
688 segmentation fault and core dump.
|
|
|
689
|
|
|
690 This has been tracked to a bug in tar! People report that tar erroneously
|
|
|
691 added a line like this at the beginning of files of Lisp code:
|
|
|
692
|
|
|
693 x FILENAME, N bytes, B tape blocks
|
|
|
694
|
|
|
695 If your tar has this problem, install GNU tar--if you can manage to
|
|
|
696 untar it :-).
|
|
|
697
|
|
|
698 * Link failure when using acc on a Sun.
|
|
|
699
|
|
|
700 To use acc, you need additional options just before the libraries, such as
|
|
|
701
|
|
|
702 /usr/lang/SC2.0.1/values-Xt.o -L/usr/lang/SC2.0.1/cg87 -L/usr/lang/SC2.0.1
|
|
|
703
|
|
|
704 and you need to add -lansi just before -lc.
|
|
|
705
|
|
|
706 The precise file names depend on the compiler version, so we
|
|
|
707 cannot easily arrange to supply them.
|
|
|
708
|
|
|
709 * Link failure on IBM AIX 1.3 ptf 0013.
|
|
|
710
|
|
|
711 There is a real duplicate definition of the function `_slibc_free' in
|
|
|
712 the library /lib/libc_s.a (just do nm on it to verify). The
|
|
|
713 workaround/fix is:
|
|
|
714
|
|
|
715 cd /lib
|
|
|
716 ar xv libc_s.a NLtmtime.o
|
|
|
717 ar dv libc_s.a NLtmtime.o
|
|
|
718
|
|
|
719 * Undefined symbols _dlopen, _dlsym and/or _dlclose on a Sun.
|
|
|
720
|
|
|
721 If you see undefined symbols _dlopen, _dlsym, or _dlclose when linking
|
|
|
722 with -lX11, compile and link against the file mit/util/misc/dlsym.c in
|
|
|
723 the MIT X11R5 distribution. Alternatively, link temacs using shared
|
|
|
724 libraries with s/sunos4shr.h. (This doesn't work if you use the X
|
|
|
725 toolkit.)
|
|
|
726
|
|
|
727 If you get the additional error that the linker could not find
|
|
|
728 lib_version.o, try extracting it from X11/usr/lib/X11/libvim.a in
|
|
|
729 X11R4, then use it in the link.
|
|
|
730
|
|
|
731 * In Shell mode, you get a ^M at the end of every line.
|
|
|
732
|
|
|
733 This happens to people who use tcsh, because it is trying to be too
|
|
|
734 smart. It sees that the Shell uses terminal type `unknown' and turns
|
|
|
735 on the flag to output ^M at the end of each line. You can fix the
|
|
|
736 problem by adding this to your .cshrc file:
|
|
|
737
|
|
|
738 if ($?EMACS) then
|
|
|
739 if ($EMACS == "t") then
|
|
|
740 unset edit
|
|
|
741 stty -icrnl -onlcr -echo susp ^Z
|
|
|
742 endif
|
|
|
743 endif
|
|
|
744
|
|
|
745 * An error message such as `X protocol error: BadMatch (invalid
|
|
|
746 parameter attributes) on protocol request 93'.
|
|
|
747
|
|
|
748 This comes from having an invalid X resource, such as
|
|
|
749 emacs*Cursor: black
|
|
|
750 (which is invalid because it specifies a color name for something
|
|
|
751 that isn't a color.)
|
|
|
752
|
|
|
753 The fix is to correct your X resources.
|
|
|
754
|
|
|
755 * Undefined symbols when linking on Sunos 4.1.
|
|
|
756
|
|
|
757 If you get the undefined symbols _atowc _wcslen, _iswprint, _iswspace,
|
|
|
758 _iswcntrl, _wcscpy, and _wcsncpy, then you need to add -lXwchar after
|
|
|
759 -lXaw in the command that links temacs.
|
|
|
760
|
|
|
761 This problem seems to arise only when the international language
|
|
|
762 extensions to X11R5 are installed.
|
|
|
763
|
|
|
764 * src/Makefile and lib-src/Makefile are truncated--most of the file missing.
|
|
|
765
|
|
|
766 This can happen if configure uses GNU sed version 2.03. That version
|
|
|
767 had a bug. GNU sed version 2.05 works properly.
|
|
|
768
|
|
|
769 * Slow startup on X11R6 with X windows.
|
|
|
770
|
|
|
771 If Emacs takes two minutes to start up on X11R6, see if your X
|
|
|
772 resources specify any Adobe fonts. That causes the type-1 font
|
|
|
773 renderer to start up, even if the font you asked for is not a type-1
|
|
|
774 font.
|
|
|
775
|
|
|
776 One way to avoid this problem is to eliminate the type-1 fonts from
|
|
|
777 your font path, like this:
|
|
|
778
|
|
|
779 xset -fp /usr/X11R6/lib/X11/fonts/Type1/
|
|
|
780
|
|
|
781 * Pull-down menus appear in the wrong place, in the toolkit version of Emacs.
|
|
|
782
|
|
|
783 An X resource of this form can cause the problem:
|
|
|
784
|
|
|
785 Emacs*geometry: 80x55+0+0
|
|
|
786
|
|
|
787 This resource is supposed to apply, and does apply, to the menus
|
|
|
788 individually as well as to Emacs frames. If that is not what you
|
|
|
789 want, rewrite the resource.
|
|
|
790
|
|
|
791 To check thoroughly for such resource specifications, use `xrdb
|
|
|
792 -query' to see what resources the X server records, and also look at
|
|
|
793 the user's ~/.Xdefaults and ~/.Xdefaults-* files.
|
|
|
794
|
|
|
795 * `make install' fails on install-doc with `Error 141'.
|
|
|
796
|
|
|
797 This happens on Ultrix 4.2 due to failure of a pipeline of tar
|
|
|
798 commands. We don't know why they fail, but the bug seems not to be in
|
|
|
799 Emacs. The workaround is to run the shell command in install-doc by
|
|
|
800 hand.
|
|
|
801
|
|
|
802 * Subprocesses remain, hanging but not zombies, on Sunos 5.3.
|
|
|
803
|
|
|
804 A bug in Sunos 5.3 causes Emacs subprocesses to remain after Emacs
|
|
|
805 exits. Sun patch # 101415-02 is part of the fix for this, but it only
|
|
|
806 applies to ptys, and doesn't fix the problem with subprocesses
|
|
|
807 communicating through pipes.
|
|
|
808
|
|
|
809 * Mail is lost when sent to local aliases.
|
|
|
810
|
|
|
811 Many emacs mail user agents (VM and rmail, for instance) use the
|
|
|
812 sendmail.el library. This library can arrange for mail to be
|
|
|
813 delivered by passing messages to the /usr/lib/sendmail (usually)
|
|
|
814 program . In doing so, it passes the '-t' flag to sendmail, which
|
|
|
815 means that the name of the recipient of the message is not on the
|
|
|
816 command line and, therefore, that sendmail must parse the message to
|
|
|
817 obtain the destination address.
|
|
|
818
|
|
|
819 There is a bug in the SunOS4.1.1 and SunOS4.1.3 versions of sendmail.
|
|
|
820 In short, when given the -t flag, the SunOS sendmail won't recognize
|
|
|
821 non-local (i.e. NIS) aliases. It has been reported that the Solaris
|
|
|
822 2.x versions of sendmail do not have this bug. For those using SunOS
|
|
|
823 4.1, the best fix is to install sendmail V8 or IDA sendmail (which
|
|
|
824 have other advantages over the regular sendmail as well). At the time
|
|
|
825 of this writing, these official versions are available:
|
|
|
826
|
|
|
827 Sendmail V8 on ftp.cs.berkeley.edu in /ucb/sendmail:
|
|
|
828 sendmail.8.6.9.base.tar.Z (the base system source & documentation)
|
|
|
829 sendmail.8.6.9.cf.tar.Z (configuration files)
|
|
|
830 sendmail.8.6.9.misc.tar.Z (miscellaneous support programs)
|
|
|
831 sendmail.8.6.9.xdoc.tar.Z (extended documentation, with postscript)
|
|
|
832
|
|
|
833 IDA sendmail on vixen.cso.uiuc.edu in /pub:
|
|
|
834 sendmail-5.67b+IDA-1.5.tar.gz
|
|
|
835
|
|
|
836 * On AIX, you get this message when running Emacs:
|
|
|
837
|
|
|
838 Could not load program emacs
|
|
|
839 Symbol smtcheckinit in csh is undefined
|
|
|
840 Error was: Exec format error
|
|
|
841
|
|
|
842 or this one:
|
|
|
843
|
|
|
844 Could not load program .emacs
|
|
|
845 Symbol _system_con in csh is undefined
|
|
|
846 Symbol _fp_trapsta in csh is undefined
|
|
|
847 Error was: Exec format error
|
|
|
848
|
|
|
849 These can happen when you try to run on AIX 3.2.5 a program that was
|
|
|
850 compiled with 3.2.4. The fix is to recompile.
|
|
|
851
|
|
|
852 * On AIX, you get this compiler error message:
|
|
|
853
|
|
|
854 Processing include file ./XMenuInt.h
|
|
|
855 1501-106: (S) Include file X11/Xlib.h not found.
|
|
|
856
|
|
|
857 This means your system was installed with only the X11 runtime i.d
|
|
|
858 libraries. You have to find your sipo (bootable tape) and install
|
|
|
859 X11Dev... with smit.
|
|
|
860
|
|
|
861 * You "lose characters" after typing Compose Character key.
|
|
|
862
|
|
|
863 This is because the Compose Character key is defined as the keysym
|
|
|
864 Multi_key, and Emacs (seeing that) does the proper X11
|
|
|
865 character-composition processing. If you don't want your Compose key
|
|
|
866 to do that, you can redefine it with xmodmap.
|
|
|
867
|
|
|
868 For example, here's one way to turn it into a Meta key:
|
|
|
869
|
|
|
870 xmodmap -e "keysym Multi_key = Meta_L"
|
|
|
871
|
|
|
872 If all users at your site of a particular keyboard prefer Meta to
|
|
|
873 Compose, you can make the remapping happen automatically by adding the
|
|
|
874 xmodmap command to the xdm setup script for that display.
|
|
|
875
|
|
|
876 * C-z just refreshes the screen instead of suspending Emacs.
|
|
|
877
|
|
|
878 You are probably using a shell that doesn't support job control, even
|
|
|
879 though the system itself is capable of it. Either use a different shell,
|
|
|
880 or set the variable `cannot-suspend' to a non-nil value.
|
|
|
881
|
|
|
882 * After running emacs once, subsequent invocations crash.
|
|
|
883
|
|
|
884 Some versions of SVR4 have a serious bug in the implementation of the
|
|
|
885 mmap () system call in the kernel; this causes emacs to run correctly
|
|
|
886 the first time, and then crash when run a second time.
|
|
|
887
|
|
|
888 Contact your vendor and ask for the mmap bug fix; in the mean time,
|
|
|
889 you may be able to work around the problem by adding a line to your
|
|
|
890 operating system description file (whose name is reported by the
|
|
|
891 configure script) that reads:
|
|
|
892 #define SYSTEM_MALLOC
|
|
|
893 This makes Emacs use memory less efficiently, but seems to work around
|
|
|
894 the kernel bug.
|
|
|
895
|
|
|
896 * Inability to send an Alt-modified key, when Emacs is communicating
|
|
|
897 directly with an X server.
|
|
|
898
|
|
|
899 If you have tried to bind an Alt-modified key as a command, and it
|
|
|
900 does not work to type the command, the first thing you should check is
|
|
|
901 whether the key is getting through to Emacs. To do this, type C-h c
|
|
|
902 followed by the Alt-modified key. C-h c should say what kind of event
|
|
|
903 it read. If it says it read an Alt-modified key, then make sure you
|
|
|
904 have made the key binding correctly.
|
|
|
905
|
|
|
906 If C-h c reports an event that doesn't have the Alt modifier, it may
|
|
|
907 be because your X server has no key for the Alt modifier. The X
|
|
|
908 server that comes from MIT does not set up the Alt modifier by
|
|
|
909 default.
|
|
|
910
|
|
|
911 If your keyboard has keys named Alt, you can enable them as follows:
|
|
|
912
|
|
|
913 xmodmap -e 'add mod2 = Alt_L'
|
|
|
914 xmodmap -e 'add mod2 = Alt_R'
|
|
|
915
|
|
|
916 If the keyboard has just one key named Alt, then only one of those
|
|
|
917 commands is needed. The modifier `mod2' is a reasonable choice if you
|
|
|
918 are using an unmodified MIT version of X. Otherwise, choose any
|
|
|
919 modifier bit not otherwise used.
|
|
|
920
|
|
|
921 If your keyboard does not have keys named Alt, you can use some other
|
|
|
922 keys. Use the keysym command in xmodmap to turn a function key (or
|
|
|
923 some other 'spare' key) into Alt_L or into Alt_R, and then use the
|
|
|
924 commands show above to make them modifier keys.
|
|
|
925
|
|
|
926 Note that if you have Alt keys but no Meta keys, Emacs translates Alt
|
|
|
927 into Meta. This is because of the great importance of Meta in Emacs.
|
|
|
928
|
|
|
929 * `Pid xxx killed due to text modification or page I/O error'
|
|
|
930
|
|
|
931 On HP/UX, you can get that error when the Emacs executable is on an NFS
|
|
|
932 file system. HP/UX responds this way if it tries to swap in a page and
|
|
|
933 does not get a response from the server within a timeout whose default
|
|
|
934 value is just ten seconds.
|
|
|
935
|
|
|
936 If this happens to you, extend the timeout period.
|
|
|
937
|
|
|
938 * `expand-file-name' fails to work on any but the machine you dumped Emacs on.
|
|
|
939
|
|
|
940 On Ultrix, if you use any of the functions which look up information
|
|
|
941 in the passwd database before dumping Emacs (say, by using
|
|
|
942 expand-file-name in site-init.el), then those functions will not work
|
|
|
943 in the dumped Emacs on any host but the one Emacs was dumped on.
|
|
|
944
|
|
|
945 The solution? Don't use expand-file-name in site-init.el, or in
|
|
|
946 anything it loads. Yuck - some solution.
|
|
|
947
|
|
|
948 I'm not sure why this happens; if you can find out exactly what is
|
|
|
949 going on, and perhaps find a fix or a workaround, please let us know.
|
|
|
950 Perhaps the YP functions cache some information, the cache is included
|
|
|
951 in the dumped Emacs, and is then inaccurate on any other host.
|
|
|
952
|
|
|
953 * On some variants of SVR4, Emacs does not work at all with X.
|
|
|
954
|
|
|
955 Try defining BROKEN_FIONREAD in your config.h file. If this solves
|
|
|
956 the problem, please send a bug report to tell us this is needed; be
|
|
|
957 sure to say exactly what type of machine and system you are using.
|
|
|
958
|
|
|
959 * Emacs fails to understand most Internet host names, even though
|
|
|
960 the names work properly with other programs on the same system.
|
|
|
961 * Emacs won't work with X-windows if the value of DISPLAY is HOSTNAME:0.
|
|
120
|
962 * Gnus can't make contact with the specified host for nntp.
|
|
0
|
963
|
|
|
964 This typically happens on Suns and other systems that use shared
|
|
|
965 libraries. The cause is that the site has installed a version of the
|
|
|
966 shared library which uses a name server--but has not installed a
|
|
|
967 similar version of the unshared library which Emacs uses.
|
|
|
968
|
|
|
969 The result is that most programs, using the shared library, work with
|
|
|
970 the nameserver, but Emacs does not.
|
|
|
971
|
|
|
972 The fix is to install an unshared library that corresponds to what you
|
|
|
973 installed in the shared library, and then relink Emacs.
|
|
|
974
|
|
|
975 On SunOS 4.1, simply define HAVE_RES_INIT.
|
|
|
976
|
|
|
977 If you have already installed the name resolver in the file libresolv.a,
|
|
|
978 then you need to compile Emacs to use that library. The easiest way to
|
|
|
979 do this is to add to config.h a definition of LIBS_SYSTEM, LIBS_MACHINE
|
|
|
980 or LIB_STANDARD which uses -lresolv. Watch out! If you redefine a macro
|
|
|
981 that is already in use in your configuration to supply some other libraries,
|
|
|
982 be careful not to lose the others.
|
|
|
983
|
|
|
984 Thus, you could start by adding this to config.h:
|
|
|
985
|
|
|
986 #define LIBS_SYSTEM -lresolv
|
|
|
987
|
|
|
988 Then if this gives you an error for redefining a macro, and you see that
|
|
|
989 the s- file defines LIBS_SYSTEM as -lfoo -lbar, you could change config.h
|
|
|
990 again to say this:
|
|
|
991
|
|
|
992 #define LIBS_SYSTEM -lresolv -lfoo -lbar
|
|
|
993
|
|
|
994 * On a Sun running SunOS 4.1.1, you get this error message from GNU ld:
|
|
|
995
|
|
|
996 /lib/libc.a(_Q_sub.o): Undefined symbol __Q_get_rp_rd referenced from text segment
|
|
|
997
|
|
|
998 The problem is in the Sun shared C library, not in GNU ld.
|
|
|
999
|
|
|
1000 The solution is to install Patch-ID# 100267-03 from Sun.
|
|
|
1001
|
|
|
1002 * SunOS 4.1.2: undefined symbol _get_wmShellWidgetClass
|
|
|
1003
|
|
|
1004 Apparently the version of libXmu.so.a that Sun ships is hosed: it's missing
|
|
|
1005 some stuff that is in libXmu.a (the static version). Sun has a patch for
|
|
|
1006 this, but a workaround is to use the static version of libXmu, by changing
|
|
|
1007 the link command from "-lXmu" to "-Bstatic -lXmu -Bdynamic". If you have
|
|
|
1008 OpenWindows 3.0, ask Sun for these patches:
|
|
|
1009 100512-02 4.1.x OpenWindows 3.0 libXt Jumbo patch
|
|
|
1010 100573-03 4.1.x OpenWindows 3.0 undefined symbols with shared libXmu
|
|
|
1011
|
|
|
1012 * Random other SunOS 4.1.[12] link errors.
|
|
|
1013
|
|
|
1014 The X headers and libraries that Sun ships in /usr/{include,lib}/X11 are
|
|
|
1015 broken. Use the ones in /usr/openwin/{include,lib} instead.
|
|
|
1016
|
|
|
1017 * Bus errors on startup when compiled with Sun's "acc" (in the routine
|
|
|
1018 make_string_internal() called from initialize_environment_alist())
|
|
|
1019
|
|
|
1020 The Sun ANSI compiler doesn't place uninitialized static variables in BSS
|
|
|
1021 space like other compilers do. This breaks emacs. If you want to use acc,
|
|
|
1022 you need to make the file "lastfile.o" be the *first* file in the link
|
|
|
1023 command. Better yet, use Lucid C or GCC.
|
|
|
1024
|
|
|
1025 * The compiler generates lots and lots of syntax errors.
|
|
|
1026
|
|
|
1027 Are you using an ANSI C compiler, like lcc or gcc? The SunOS 4.1 bundled cc
|
|
|
1028 is not ANSI.
|
|
|
1029
|
|
|
1030 If X has not been configured to compile itself using lcc, gcc, or another ANSI
|
|
|
1031 compiler, then you will have to hack the automatically-generated makefile in
|
|
|
1032 the `lwlib' directory by hand to make it use an ANSI compiler.
|
|
|
1033
|
|
|
1034 * When using gcc, you get the error message "undefined symbol __fixunsdfsi".
|
|
|
1035 * When using gcc, you get the error message "undefined symbol __main".
|
|
|
1036
|
|
|
1037 This means that you need to link with the gcc library. It may be called
|
|
|
1038 "gcc-gnulib" or "libgcc.a"; figure out where it is, and define LIB_GCC in
|
|
|
1039 config.h to point to it.
|
|
|
1040
|
|
|
1041 It may also work to use the GCC version of `ld' instead of the standard one.
|
|
|
1042
|
|
|
1043 * When compiling with X11, you get "undefined symbol _XtStrings".
|
|
|
1044
|
|
|
1045 This means that you are trying to link emacs against the X11r4 version of
|
|
|
1046 libXt.a, but you have compiled either Emacs or the code in the lwlib
|
|
|
1047 subdirectory with the X11r5 header files. That doesn't work.
|
|
|
1048
|
|
|
1049 Remember, you can't compile lwlib for r4 and emacs for r5, or vice versa.
|
|
|
1050 They must be in sync.
|
|
|
1051
|
|
|
1052 * Self documentation messages are garbled.
|
|
|
1053
|
|
|
1054 This means that the file `etc/DOC-...' doesn't properly correspond
|
|
|
1055 with the Emacs executable. Redumping Emacs and then installing the
|
|
|
1056 corresponding pair of files should fix the problem.
|
|
|
1057
|
|
|
1058 * Trouble using ptys on AIX.
|
|
|
1059
|
|
|
1060 People often install the pty devices on AIX incorrectly.
|
|
|
1061 Use `smit pty' to reinstall them properly.
|
|
|
1062
|
|
|
1063 * Shell mode on HP/UX gives the message, "`tty`: Ambiguous".
|
|
|
1064
|
|
|
1065 christos@theory.tn.cornell.edu says:
|
|
|
1066
|
|
|
1067 The problem is that in your .cshrc you have something that tries to
|
|
|
1068 execute `tty`. If you are not running the shell on a real tty then
|
|
|
1069 tty will print "not a tty". Csh expects one word in some places,
|
|
|
1070 but tty is giving it back 3.
|
|
|
1071
|
|
|
1072 The solution is to add a pair of quotes around `tty` to make it a single
|
|
|
1073 word:
|
|
|
1074
|
|
|
1075 if (`tty` == "/dev/console")
|
|
|
1076
|
|
|
1077 should be changed to:
|
|
|
1078
|
|
|
1079 if ("`tty`" == "/dev/console")
|
|
|
1080
|
|
|
1081 Even better, move things that set up terminal sections out of .cshrc
|
|
|
1082 and into .login.
|
|
|
1083
|
|
|
1084 * With process-connection-type set to t, each line of subprocess output is
|
|
|
1085 terminated with a ^M, making ange-ftp and GNUS not work.
|
|
|
1086
|
|
|
1087 On SunOS systems, this problem has been seen to be a result of an incomplete
|
|
|
1088 installation of gcc 2.2 which allowed some non-ANSI compatible include files
|
|
|
1089 into the compilation. In particular this affected virtually all ioctl() calls.
|
|
|
1090
|
|
|
1091 * Once you pull down a menu from the menubar, it won't go away.
|
|
|
1092
|
|
|
1093 It has been claimed that this is caused by a bug in certain very old (1990?)
|
|
|
1094 versions of the twm window manager. It doesn't happen with recent vintages,
|
|
|
1095 or with other window managers.
|
|
|
1096
|
|
|
1097 * Emacs ignores the "help" key when running OLWM.
|
|
|
1098
|
|
|
1099 OLWM grabs the help key, and retransmits it to the appropriate client using
|
|
|
1100 XSendEvent. Allowing emacs to react to synthetic events is a security hole,
|
|
|
1101 so this is turned off by default. You can enable it by setting the variable
|
|
|
1102 x-allow-sendevents to t. You can also cause fix this by telling OLWM to not
|
|
|
1103 grab the help key, with the null binding "OpenWindows.KeyboardCommand.Help:".
|
|
|
1104
|
|
120
|
1105 * Using X11, control-shift-leftbutton makes Emacs hang.
|
|
0
|
1106
|
|
|
1107 Use the shell command `xset bc' to make the old X Menu package work.
|
|
|
1108
|
|
120
|
1109 * Emacs running under X11 does not handle mouse clicks.
|
|
0
|
1110 * `emacs -geometry 80x20' finds a file named `80x20'.
|
|
|
1111
|
|
|
1112 One cause of such problems is having (setq term-file-prefix nil) in
|
|
|
1113 your .emacs file. Another cause is a bad value of EMACSLOADPATH in
|
|
|
1114 the environment.
|
|
|
1115
|
|
|
1116 * Emacs gets error message from linker on Sun.
|
|
|
1117
|
|
|
1118 If the error message says that a symbol such as `f68881_used' or
|
|
|
1119 `ffpa_used' or `start_float' is undefined, this probably indicates
|
|
|
1120 that you have compiled some libraries, such as the X libraries,
|
|
|
1121 with a floating point option other than the default.
|
|
|
1122
|
|
|
1123 It's not terribly hard to make this work with small changes in
|
|
|
1124 crt0.c together with linking with Fcrt1.o, Wcrt1.o or Mcrt1.o.
|
|
|
1125 However, the easiest approach is to build Xlib with the default
|
|
|
1126 floating point option: -fsoft to decide at run time what hardware
|
|
|
1127 is available.
|
|
|
1128
|
|
|
1129 * Keyboard input gets confused after a beep when using a DECserver
|
|
|
1130 as a concentrator.
|
|
|
1131
|
|
|
1132 This problem seems to be a matter of configuring the DECserver to use
|
|
|
1133 7 bit characters rather than 8 bit characters.
|
|
|
1134
|
|
|
1135 * M-x shell persistently reports "Process shell exited abnormally with code 1".
|
|
|
1136
|
|
|
1137 This happened on Suns as a result of what is said to be a bug in Sunos
|
|
|
1138 version 4.0.x. The only fix was to reboot the machine.
|
|
|
1139
|
|
|
1140 * Programs running under terminal emulator do not recognize `emacs'
|
|
|
1141 terminal type.
|
|
|
1142
|
|
|
1143 The cause of this is a shell startup file that sets the TERMCAP
|
|
|
1144 environment variable. The terminal emulator uses that variable to
|
|
|
1145 provide the information on the special terminal type that Emacs
|
|
|
1146 emulates.
|
|
|
1147
|
|
|
1148 Rewrite your shell startup file so that it does not change TERMCAP
|
|
|
1149 in such a case. You could use the following conditional which sets
|
|
|
1150 it only if it is undefined.
|
|
|
1151
|
|
|
1152 if ( ! ${?TERMCAP} ) setenv TERMCAP ~/my-termcap-file
|
|
|
1153
|
|
|
1154 Or you could set TERMCAP only when you set TERM--which should not
|
|
|
1155 happen in a non-login shell.
|
|
|
1156
|
|
|
1157 * Problem with remote X server on Suns.
|
|
|
1158
|
|
|
1159 On a Sun, running Emacs on one machine with the X server on another
|
|
|
1160 may not work if you have used the unshared system libraries. This
|
|
|
1161 is because the unshared libraries fail to use YP for host name lookup.
|
|
|
1162 As a result, the host name you specify may not be recognized.
|
|
|
1163
|
|
|
1164 * Shell mode ignores interrupts on Apollo Domain
|
|
|
1165
|
|
|
1166 You may find that M-x shell prints the following message:
|
|
|
1167
|
|
|
1168 Warning: no access to tty; thus no job control in this shell...
|
|
|
1169
|
|
|
1170 This can happen if there are not enough ptys on your system.
|
|
|
1171 Here is how to make more of them.
|
|
|
1172
|
|
|
1173 % cd /dev
|
|
|
1174 % ls pty*
|
|
|
1175 # shows how many pty's you have. I had 8, named pty0 to pty7)
|
|
|
1176 % /etc/crpty 8
|
|
|
1177 # creates eight new pty's
|
|
|
1178
|
|
|
1179 * Fatal signal in the command temacs -l loadup inc dump
|
|
|
1180
|
|
|
1181 This command is the final stage of building Emacs. It is run by the
|
|
|
1182 Makefile in the src subdirectory, or by build.com on VMS.
|
|
|
1183
|
|
|
1184 It has been known to get fatal errors due to insufficient swapping
|
|
|
1185 space available on the machine.
|
|
|
1186
|
|
|
1187 On 68000's, it has also happened because of bugs in the
|
|
|
1188 subroutine `alloca'. Verify that `alloca' works right, even
|
|
|
1189 for large blocks (many pages).
|
|
|
1190
|
|
|
1191 * test-distrib says that the distribution has been clobbered
|
|
|
1192 * or, temacs prints "Command key out of range 0-127"
|
|
|
1193 * or, temacs runs and dumps xemacs, but xemacs totally fails to work.
|
|
|
1194 * or, temacs gets errors dumping xemacs
|
|
|
1195
|
|
|
1196 This can be because the .elc files have been garbled. Do not be
|
|
|
1197 fooled by the fact that most of a .elc file is text: these are
|
|
|
1198 binary files and can contain all 256 byte values.
|
|
|
1199
|
|
|
1200 In particular `shar' cannot be used for transmitting GNU Emacs.
|
|
|
1201 It typically truncates "lines". What appear to be "lines" in
|
|
|
1202 a binary file can of course be of any length. Even once `shar'
|
|
|
1203 itself is made to work correctly, `sh' discards null characters
|
|
|
1204 when unpacking the shell archive.
|
|
|
1205
|
|
|
1206 I have also seen character \177 changed into \377. I do not know
|
|
|
1207 what transfer means caused this problem. Various network
|
|
|
1208 file transfer programs are suspected of clobbering the high bit.
|
|
|
1209
|
|
|
1210 If you have a copy of Emacs that has been damaged in its
|
|
|
1211 nonprinting characters, you can fix them:
|
|
|
1212
|
|
|
1213 1) Record the names of all the .elc files.
|
|
|
1214 2) Delete all the .elc files.
|
|
|
1215 3) Recompile alloc.c with a value of PURESIZE twice as large.
|
|
|
1216 You might as well save the old alloc.o.
|
|
|
1217 4) Remake xemacs. It should work now.
|
|
|
1218 5) Running xemacs, do Meta-x byte-compile-file repeatedly
|
|
|
1219 to recreate all the .elc files that used to exist.
|
|
|
1220 You may need to increase the value of the variable
|
|
|
1221 max-lisp-eval-depth to succeed in running the compiler interpreted
|
|
|
1222 on certain .el files. 400 was sufficient as of last report.
|
|
|
1223 6) Reinstall the old alloc.o (undoing changes to alloc.c if any)
|
|
|
1224 and remake temacs.
|
|
|
1225 7) Remake xemacs. It should work now, with valid .elc files.
|
|
|
1226
|
|
|
1227 * temacs prints "Pure Lisp storage exhausted"
|
|
|
1228
|
|
|
1229 This means that the Lisp code loaded from the .elc and .el
|
|
|
1230 files during temacs -l loadup inc dump took up more
|
|
|
1231 space than was allocated.
|
|
|
1232
|
|
|
1233 This could be caused by
|
|
|
1234 1) adding code to the preloaded Lisp files
|
|
|
1235 2) adding more preloaded files in loadup.el
|
|
|
1236 3) having a site-init.el or site-load.el which loads files.
|
|
|
1237 Note that ANY site-init.el or site-load.el is nonstandard;
|
|
|
1238 if you have received Emacs from some other site
|
|
|
1239 and it contains a site-init.el or site-load.el file, consider
|
|
|
1240 deleting that file.
|
|
|
1241 4) getting the wrong .el or .elc files
|
|
|
1242 (not from the directory you expected).
|
|
|
1243 5) deleting some .elc files that are supposed to exist.
|
|
|
1244 This would cause the source files (.el files) to be
|
|
|
1245 loaded instead. They take up more room, so you lose.
|
|
|
1246 6) a bug in the Emacs distribution which underestimates
|
|
|
1247 the space required.
|
|
|
1248
|
|
|
1249 If the need for more space is legitimate, use the --puresize option
|
|
|
1250 to `configure' to specify more pure space.
|
|
|
1251
|
|
|
1252 But in some of the cases listed above, this problem is a consequence
|
|
|
1253 of something else that is wrong. Be sure to check and fix the real
|
|
|
1254 problem.
|
|
|
1255
|
|
|
1256 * Changes made to .el files do not take effect.
|
|
|
1257
|
|
|
1258 You may have forgotten to recompile them into .elc files.
|
|
|
1259 Then the old .elc files will be loaded, and your changes
|
|
|
1260 will not be seen. To fix this, do M-x byte-recompile-directory
|
|
|
1261 and specify the directory that contains the Lisp files.
|
|
|
1262
|
|
|
1263 Note that you may get a warning when loading a .elc file that
|
|
|
1264 is older than the corresponding .el file.
|
|
|
1265
|
|
|
1266 * Things which should be bold or italic (such as the initial copyright notice)
|
|
|
1267 are not.
|
|
|
1268
|
|
|
1269 The fonts of the "bold" and "italic" faces are generated from the font of
|
|
|
1270 the "default" face; in this way, your bold and italic fonts will have the
|
|
|
1271 appropriate size and family. However, emacs can only be clever in this
|
|
|
1272 way if you have specified the default font using the XLFD (X Logical Font
|
|
|
1273 Description) format, which looks like
|
|
|
1274
|
|
|
1275 *-courier-medium-r-*-*-*-120-*-*-*-*-*-*
|
|
|
1276
|
|
|
1277 if you use any of the other, less strict font name formats, some of which
|
|
|
1278 look like
|
|
|
1279 lucidasanstypewriter-12
|
|
|
1280 and fixed
|
|
|
1281 and 9x13
|
|
|
1282
|
|
|
1283 then emacs won't be able to guess the names of the "bold" and "italic"
|
|
|
1284 versions. All X fonts can be referred to via XLFD-style names, so you
|
|
|
1285 should use those forms. See the man pages for X(1), xlsfonts(1), and
|
|
|
1286 xfontsel(1).
|
|
|
1287
|
|
120
|
1288 * The dumped Emacs (XEmacs) crashes when run, trying to write pure data.
|
|
0
|
1289
|
|
|
1290 Two causes have been seen for such problems.
|
|
|
1291
|
|
|
1292 1) On a system where getpagesize is not a system call, it is defined
|
|
|
1293 as a macro. If the definition (in both unexec.c and malloc.c) is wrong,
|
|
|
1294 it can cause problems like this. You might be able to find the correct
|
|
|
1295 value in the man page for a.out (5).
|
|
|
1296
|
|
|
1297 2) Some systems allocate variables declared static among the
|
|
|
1298 initialized variables. Emacs makes all initialized variables in most
|
|
|
1299 of its files pure after dumping, but the variables declared static and
|
|
|
1300 not initialized are not supposed to be pure. On these systems you
|
|
|
1301 may need to add "#define static" to the m- or the s- file.
|
|
|
1302
|
|
|
1303 * Reading and writing files is very very slow.
|
|
|
1304
|
|
|
1305 Try evaluating the form (setq lock-directory nil) and see if that helps.
|
|
|
1306 There is a problem with file-locking on some systems (possibly related
|
|
|
1307 to NFS) that I don't understand. Please send mail to the address
|
|
|
1308 xemacs@xemacs.org if you figure this one out.
|
|
|
1309
|
|
|
1310 * Compilation errors on VMS.
|
|
|
1311
|
|
|
1312 Sorry, XEmacs does not work under VMS. You might consider working on
|
|
|
1313 the port if you really want to have XEmacs work under VMS.
|
|
|
1314
|
|
120
|
1315 * Mail agents (VM, Gnus, rmail) cannot get new mail
|
|
0
|
1316
|
|
|
1317 rmail and VM get new mail from /usr/spool/mail/$USER using a program
|
|
120
|
1318 called `movemail'. This program interlocks with /bin/mail using the
|
|
|
1319 protocol defined by /bin/mail.
|
|
0
|
1320
|
|
|
1321 There are two different protocols in general use. One of them uses
|
|
|
1322 the `flock' system call. The other involves creating a lock file;
|
|
|
1323 `movemail' must be able to write in /usr/spool/mail in order to do
|
|
120
|
1324 this. You control which one is used by defining, or not defining, the
|
|
|
1325 macro MAIL_USE_FLOCK in config.h or the m- or s- file it includes. IF
|
|
|
1326 YOU DON'T USE THE FORM OF INTERLOCKING THAT IS NORMAL ON YOUR SYSTEM,
|
|
|
1327 YOU CAN LOSE MAIL!
|
|
0
|
1328
|
|
|
1329 If your system uses the lock file protocol, and fascist restrictions
|
|
|
1330 prevent ordinary users from writing the lock files in /usr/spool/mail,
|
|
|
1331 you may need to make `movemail' setgid to a suitable group such as
|
|
|
1332 `mail'. You can use these commands (as root):
|
|
|
1333
|
|
|
1334 chgrp mail movemail
|
|
|
1335 chmod 2755 movemail
|
|
|
1336
|
|
|
1337 If your system uses the lock file protocol, and fascist restrictions
|
|
|
1338 prevent ordinary users from writing the lock files in /usr/spool/mail,
|
|
|
1339 you may need to make `movemail' setgid to a suitable group such as
|
|
120
|
1340 `mail'. To do this, use the following commands (as root) after doing
|
|
|
1341 the make install.
|
|
0
|
1342
|
|
|
1343 chgrp mail movemail
|
|
|
1344 chmod 2755 movemail
|
|
|
1345
|
|
|
1346 Installation normally copies movemail from the build directory to an
|
|
|
1347 installation directory which is usually under /usr/local/lib. The
|
|
|
1348 installed copy of movemail is usually in the directory
|
|
|
1349 /usr/local/lib/emacs/VERSION/TARGET. You must change the group and
|
|
|
1350 mode of the installed copy; changing the group and mode of the build
|
|
|
1351 directory copy is ineffective.
|
|
|
1352
|
|
|
1353 * Emacs spontaneously displays "I-search: " at the bottom of the screen.
|
|
|
1354
|
|
|
1355 This means that Control-S/Control-Q (XON/XOFF) "flow control" is being
|
|
|
1356 used. C-s/C-q flow control is bad for Emacs editors because it takes
|
|
|
1357 away C-s and C-q as user commands. Since editors do not output long
|
|
|
1358 streams of text without user commands, there is no need for a
|
|
|
1359 user-issuable "stop output" command in an editor; therefore, a
|
|
|
1360 properly designed flow control mechanism would transmit all possible
|
|
|
1361 input characters without interference. Designing such a mechanism is
|
|
|
1362 easy, for a person with at least half a brain.
|
|
|
1363
|
|
|
1364 There are three possible reasons why flow control could be taking place:
|
|
|
1365
|
|
|
1366 1) Terminal has not been told to disable flow control
|
|
|
1367 2) Insufficient padding for the terminal in use
|
|
|
1368 3) Some sort of terminal concentrator or line switch is responsible
|
|
|
1369
|
|
|
1370 First of all, many terminals have a set-up mode which controls whether
|
|
|
1371 they generate XON/XOFF flow control characters. This must be set to
|
|
|
1372 "no XON/XOFF" in order for Emacs to work. Sometimes there is an
|
|
|
1373 escape sequence that the computer can send to turn flow control off
|
|
|
1374 and on. If so, perhaps the termcap `ti' string should turn flow
|
|
|
1375 control off, and the `te' string should turn it on.
|
|
|
1376
|
|
|
1377 Once the terminal has been told "no flow control", you may find it
|
|
|
1378 needs more padding. The amount of padding Emacs sends is controlled
|
|
|
1379 by the termcap entry for the terminal in use, and by the output baud
|
|
|
1380 rate as known by the kernel. The shell command `stty' will print
|
|
|
1381 your output baud rate; `stty' with suitable arguments will set it if
|
|
|
1382 it is wrong. Setting to a higher speed causes increased padding. If
|
|
|
1383 the results are wrong for the correct speed, there is probably a
|
|
|
1384 problem in the termcap entry. You must speak to a local Unix wizard
|
|
|
1385 to fix this. Perhaps you are just using the wrong terminal type.
|
|
|
1386
|
|
|
1387 For terminals that lack a "no flow control" mode, sometimes just
|
|
|
1388 giving lots of padding will prevent actual generation of flow control
|
|
|
1389 codes. You might as well try it.
|
|
|
1390
|
|
|
1391 If you are really unlucky, your terminal is connected to the computer
|
|
|
1392 through a concentrator which sends XON/XOFF flow control to the
|
|
|
1393 computer, or it insists on sending flow control itself no matter how
|
|
|
1394 much padding you give it. Unless you can figure out how to turn flow
|
|
|
1395 control off on this concentrator (again, refer to your local wizard),
|
|
|
1396 you are screwed! You should have the terminal or concentrator
|
|
|
1397 replaced with a properly designed one. In the mean time, some drastic
|
|
|
1398 measures can make Emacs semi-work.
|
|
|
1399
|
|
|
1400 You can make Emacs ignore C-s and C-q and let the operating system
|
|
|
1401 handle them. To do this on a per-session basis, just type M-x
|
|
|
1402 enable-flow-control RET. You will see a message that C-\ and C-^ are
|
|
|
1403 now translated to C-s and C-q. (Use the same command M-x
|
|
|
1404 enable-flow-control to turn *off* this special mode. It toggles flow
|
|
|
1405 control handling.)
|
|
|
1406
|
|
|
1407 If C-\ and C-^ are inconvenient for you (for example, if one of them
|
|
|
1408 is the escape character of your terminal concentrator), you can choose
|
|
|
1409 other characters by setting the variables flow-control-c-s-replacement
|
|
|
1410 and flow-control-c-q-replacement. But choose carefully, since all
|
|
|
1411 other control characters are already used by emacs.
|
|
|
1412
|
|
|
1413 IMPORTANT: if you type C-s by accident while flow control is enabled,
|
|
|
1414 Emacs output will freeze, and you will have to remember to type C-q in
|
|
|
1415 order to continue.
|
|
|
1416
|
|
|
1417 If you work in an environment where a majority of terminals of a
|
|
|
1418 certain type are flow control hobbled, you can use the function
|
|
|
1419 `enable-flow-control-on' to turn on this flow control avoidance scheme
|
|
|
1420 automatically. Here is an example:
|
|
|
1421
|
|
|
1422 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
|
|
|
1423
|
|
|
1424 If this isn't quite correct (e.g. you have a mixture of flow-control hobbled
|
|
|
1425 and good vt200 terminals), you can still run enable-flow-control
|
|
|
1426 manually.
|
|
|
1427
|
|
|
1428 I have no intention of ever redesigning the Emacs command set for the
|
|
|
1429 assumption that terminals use C-s/C-q flow control. XON/XOFF flow
|
|
|
1430 control technique is a bad design, and terminals that need it are bad
|
|
|
1431 merchandise and should not be purchased. Now that X is becoming
|
|
|
1432 widespread, XON/XOFF seems to be on the way out. If you can get some
|
|
|
1433 use out of GNU Emacs on inferior terminals, more power to you, but I
|
|
|
1434 will not make Emacs worse for properly designed systems for the sake
|
|
|
1435 of inferior systems.
|
|
|
1436
|
|
|
1437 * Control-S and Control-Q commands are ignored completely.
|
|
|
1438
|
|
|
1439 For some reason, your system is using brain-damaged C-s/C-q flow
|
|
|
1440 control despite Emacs's attempts to turn it off. Perhaps your
|
|
|
1441 terminal is connected to the computer through a concentrator
|
|
|
1442 that wants to use flow control.
|
|
|
1443
|
|
|
1444 You should first try to tell the concentrator not to use flow control.
|
|
|
1445 If you succeed in this, try making the terminal work without
|
|
|
1446 flow control, as described in the preceding section.
|
|
|
1447
|
|
|
1448 If that line of approach is not successful, map some other characters
|
|
|
1449 into C-s and C-q using keyboard-translate-table. The example above
|
|
|
1450 shows how to do this with C-^ and C-\.
|
|
|
1451
|
|
120
|
1452 * Control-S and Control-Q commands are ignored completely on a net
|
|
|
1453 connection.
|
|
0
|
1454
|
|
|
1455 Some versions of rlogin (and possibly telnet) do not pass flow
|
|
|
1456 control characters to the remote system to which they connect.
|
|
|
1457 On such systems, emacs on the remote system cannot disable flow
|
|
|
1458 control on the local system.
|
|
|
1459
|
|
|
1460 One way to cure this is to disable flow control on the local host
|
|
|
1461 (the one running rlogin, not the one running rlogind) using the
|
|
|
1462 stty command, before starting the rlogin process. On many systems,
|
|
120
|
1463 `stty start u stop u' will do this.
|
|
0
|
1464
|
|
|
1465 Some versions of tcsh will prevent even this from working. One way
|
|
|
1466 around this is to start another shell before starting rlogin, and
|
|
|
1467 issue the stty command to disable flow control from that shell.
|
|
|
1468
|
|
|
1469 If none of these methods work, the best solution is to type
|
|
120
|
1470 `M-x enable-flow-control' at the beginning of your emacs session, or
|
|
0
|
1471 if you expect the problem to continue, add a line such as the
|
|
|
1472 following to your .emacs (on the host running rlogind):
|
|
|
1473
|
|
|
1474 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131")
|
|
|
1475
|
|
|
1476 See the entry about spontaneous display of I-search (above) for more
|
|
|
1477 info.
|
|
|
1478
|
|
|
1479 * Screen is updated wrong, but only on one kind of terminal.
|
|
|
1480
|
|
120
|
1481 This could mean that the termcap entry you are using for that terminal
|
|
|
1482 is wrong, or it could mean that Emacs has a bug handing the
|
|
|
1483 combination of features specified for that terminal.
|
|
0
|
1484
|
|
|
1485 The first step in tracking this down is to record what characters
|
|
|
1486 Emacs is sending to the terminal. Execute the Lisp expression
|
|
120
|
1487 (open-termscript "./emacs-script") to make Emacs write all terminal
|
|
|
1488 output into the file ~/emacs-script as well; then do what makes the
|
|
|
1489 screen update wrong, and look at the file and decode the characters
|
|
|
1490 using the manual for the terminal. There are several possibilities:
|
|
0
|
1491
|
|
|
1492 1) The characters sent are correct, according to the terminal manual.
|
|
|
1493
|
|
|
1494 In this case, there is no obvious bug in Emacs, and most likely you
|
|
|
1495 need more padding, or possibly the terminal manual is wrong.
|
|
|
1496
|
|
120
|
1497 2) The characters sent are incorrect, due to an obscure aspect of the
|
|
|
1498 terminal behavior not described in an obvious way by termcap.
|
|
0
|
1499
|
|
120
|
1500 This case is hard. It will be necessary to think of a way for Emacs
|
|
|
1501 to distinguish between terminals with this kind of behavior and other
|
|
|
1502 terminals that behave subtly differently but are classified the same
|
|
|
1503 by termcap; or else find an algorithm for Emacs to use that avoids the
|
|
|
1504 difference. Such changes must be tested on many kinds of terminals.
|
|
0
|
1505
|
|
|
1506 3) The termcap entry is wrong.
|
|
|
1507
|
|
120
|
1508 See the file etc/TERMS for information on changes that are known to be
|
|
|
1509 needed in commonly used termcap entries for certain terminals.
|
|
0
|
1510
|
|
120
|
1511 4) The characters sent are incorrect, and clearly cannot be right for
|
|
|
1512 any terminal with the termcap entry you were using.
|
|
0
|
1513
|
|
120
|
1514 This is unambiguously an Emacs bug, and can probably be fixed in
|
|
|
1515 termcap.c, tparam.c, term.c, scroll.c, cm.c or dispnew.c.
|
|
0
|
1516
|
|
|
1517 * Output from Control-V is slow.
|
|
|
1518
|
|
|
1519 On many bit-map terminals, scrolling operations are fairly slow.
|
|
|
1520 Often the termcap entry for the type of terminal in use fails
|
|
|
1521 to inform Emacs of this. The two lines at the bottom of the screen
|
|
|
1522 before a Control-V command are supposed to appear at the top after
|
|
|
1523 the Control-V command. If Emacs thinks scrolling the lines is fast,
|
|
|
1524 it will scroll them to the top of the screen.
|
|
|
1525
|
|
|
1526 If scrolling is slow but Emacs thinks it is fast, the usual reason is
|
|
|
1527 that the termcap entry for the terminal you are using does not
|
|
|
1528 specify any padding time for the `al' and `dl' strings. Emacs
|
|
|
1529 concludes that these operations take only as much time as it takes to
|
|
|
1530 send the commands at whatever line speed you are using. You must
|
|
|
1531 fix the termcap entry to specify, for the `al' and `dl', as much
|
|
|
1532 time as the operations really take.
|
|
|
1533
|
|
|
1534 Currently Emacs thinks in terms of serial lines which send characters
|
|
|
1535 at a fixed rate, so that any operation which takes time for the
|
|
|
1536 terminal to execute must also be padded. With bit-map terminals
|
|
|
1537 operated across networks, often the network provides some sort of
|
|
|
1538 flow control so that padding is never needed no matter how slow
|
|
|
1539 an operation is. You must still specify a padding time if you want
|
|
|
1540 Emacs to realize that the operation takes a long time. This will
|
|
|
1541 cause padding characters to be sent unnecessarily, but they do
|
|
|
1542 not really cost much. They will be transmitted while the scrolling
|
|
|
1543 is happening and then discarded quickly by the terminal.
|
|
|
1544
|
|
|
1545 Most bit-map terminals provide commands for inserting or deleting
|
|
|
1546 multiple lines at once. Define the `AL' and `DL' strings in the
|
|
|
1547 termcap entry to say how to do these things, and you will have
|
|
|
1548 fast output without wasted padding characters. These strings should
|
|
|
1549 each contain a single %-spec saying how to send the number of lines
|
|
|
1550 to be scrolled. These %-specs are like those in the termcap
|
|
|
1551 `cm' string.
|
|
|
1552
|
|
|
1553 You should also define the `IC' and `DC' strings if your terminal
|
|
|
1554 has a command to insert or delete multiple characters. These
|
|
|
1555 take the number of positions to insert or delete as an argument.
|
|
|
1556
|
|
|
1557 A `cs' string to set the scrolling region will reduce the amount
|
|
|
1558 of motion you see on the screen when part of the screen is scrolled.
|
|
|
1559
|
|
|
1560 * Your Delete key sends a Backspace to the terminal, using an AIXterm.
|
|
|
1561
|
|
|
1562 The solution is to include in your .Xdefaults the lines:
|
|
|
1563
|
|
|
1564 *aixterm.Translations: #override <Key>BackSpace: string(0x7f)
|
|
|
1565 aixterm*ttyModes: erase ^?
|
|
|
1566
|
|
|
1567 This makes your Backspace key send DEL (ASCII 127).
|
|
|
1568
|
|
|
1569 * You type Control-H (Backspace) expecting to delete characters.
|
|
|
1570
|
|
120
|
1571 Emacs has traditionally used Control-H for help; unfortunately this
|
|
|
1572 interferes with its use as Backspace on TTY's. One way to solve this
|
|
|
1573 problem is to put this in your .emacs:
|
|
0
|
1574
|
|
120
|
1575 (keyboard-translate ?\C-h ?\C-?)
|
|
|
1576 (global-set-key "\M-?" 'help-command)
|
|
0
|
1577
|
|
|
1578 This makes Control-H (Backspace) work sensibly, and moves help to
|
|
120
|
1579 Meta-? (ESC ?).
|
|
0
|
1580
|
|
|
1581 Note that you can probably also access help using F1.
|
|
|
1582
|
|
|
1583 * Editing files through RFS gives spurious "file has changed" warnings.
|
|
|
1584 It is possible that a change in Emacs 18.37 gets around this problem,
|
|
|
1585 but in case not, here is a description of how to fix the RFS bug that
|
|
|
1586 causes it.
|
|
|
1587
|
|
|
1588 There was a serious pair of bugs in the handling of the fsync() system
|
|
|
1589 call in the RFS server.
|
|
|
1590
|
|
|
1591 The first is that the fsync() call is handled as another name for the
|
|
|
1592 close() system call (!!). It appears that fsync() is not used by very
|
|
|
1593 many programs; Emacs version 18 does an fsync() before closing files
|
|
|
1594 to make sure that the bits are on the disk.
|
|
|
1595
|
|
|
1596 This is fixed by the enclosed patch to the RFS server.
|
|
|
1597
|
|
|
1598 The second, more serious problem, is that fsync() is treated as a
|
|
|
1599 non-blocking system call (i.e., it's implemented as a message that
|
|
|
1600 gets sent to the remote system without waiting for a reply). Fsync is
|
|
|
1601 a useful tool for building atomic file transactions. Implementing it
|
|
|
1602 as a non-blocking RPC call (when the local call blocks until the sync
|
|
|
1603 is done) is a bad idea; unfortunately, changing it will break the RFS
|
|
|
1604 protocol. No fix was supplied for this problem.
|
|
|
1605
|
|
|
1606 (as always, your line numbers may vary)
|
|
|
1607
|
|
|
1608 % rcsdiff -c -r1.2 serversyscall.c
|
|
|
1609 RCS file: RCS/serversyscall.c,v
|
|
|
1610 retrieving revision 1.2
|
|
|
1611 diff -c -r1.2 serversyscall.c
|
|
|
1612 *** /tmp/,RCSt1003677 Wed Jan 28 15:15:02 1987
|
|
|
1613 --- serversyscall.c Wed Jan 28 15:14:48 1987
|
|
|
1614 ***************
|
|
|
1615 *** 163,169 ****
|
|
|
1616 /*
|
|
|
1617 * No return sent for close or fsync!
|
|
|
1618 */
|
|
|
1619 ! if (syscall == RSYS_close || syscall == RSYS_fsync)
|
|
|
1620 proc->p_returnval = deallocate_fd(proc, msg->m_args[0]);
|
|
|
1621 else
|
|
|
1622 {
|
|
|
1623 --- 166,172 ----
|
|
|
1624 /*
|
|
|
1625 * No return sent for close or fsync!
|
|
|
1626 */
|
|
|
1627 ! if (syscall == RSYS_close)
|
|
|
1628 proc->p_returnval = deallocate_fd(proc, msg->m_args[0]);
|
|
|
1629 else
|
|
|
1630 {
|
|
|
1631
|
|
|
1632 * Vax C compiler bugs affecting Emacs.
|
|
|
1633
|
|
|
1634 You may get one of these problems compiling Emacs:
|
|
|
1635
|
|
|
1636 foo.c line nnn: compiler error: no table entry for op STASG
|
|
|
1637 foo.c: fatal error in /lib/ccom
|
|
|
1638
|
|
|
1639 These are due to bugs in the C compiler; the code is valid C.
|
|
|
1640 Unfortunately, the bugs are unpredictable: the same construct
|
|
|
1641 may compile properly or trigger one of these bugs, depending
|
|
|
1642 on what else is in the source file being compiled. Even changes
|
|
|
1643 in header files that should not affect the file being compiled
|
|
|
1644 can affect whether the bug happens. In addition, sometimes files
|
|
|
1645 that compile correctly on one machine get this bug on another machine.
|
|
|
1646
|
|
|
1647 As a result, it is hard for me to make sure this bug will not affect
|
|
|
1648 you. I have attempted to find and alter these constructs, but more
|
|
|
1649 can always appear. However, I can tell you how to deal with it if it
|
|
|
1650 should happen. The bug comes from having an indexed reference to an
|
|
|
1651 array of Lisp_Objects, as an argument in a function call:
|
|
|
1652 Lisp_Object *args;
|
|
|
1653 ...
|
|
|
1654 ... foo (5, args[i], ...)...
|
|
|
1655 putting the argument into a temporary variable first, as in
|
|
|
1656 Lisp_Object *args;
|
|
|
1657 Lisp_Object tem;
|
|
|
1658 ...
|
|
|
1659 tem = args[i];
|
|
|
1660 ... foo (r, tem, ...)...
|
|
|
1661 causes the problem to go away.
|
|
|
1662 The `contents' field of a Lisp vector is an array of Lisp_Objects,
|
|
|
1663 so you may see the problem happening with indexed references to that.
|
|
|
1664
|
|
|
1665 * 68000 C compiler problems
|
|
|
1666
|
|
|
1667 Various 68000 compilers have different problems.
|
|
|
1668 These are some that have been observed.
|
|
|
1669
|
|
|
1670 ** Using value of assignment expression on union type loses.
|
|
|
1671 This means that x = y = z; or foo (x = z); does not work
|
|
|
1672 if x is of type Lisp_Object.
|
|
|
1673
|
|
|
1674 ** "cannot reclaim" error.
|
|
|
1675
|
|
|
1676 This means that an expression is too complicated. You get the correct
|
|
|
1677 line number in the error message. The code must be rewritten with
|
|
|
1678 simpler expressions.
|
|
|
1679
|
|
|
1680 ** XCONS, XSTRING, etc macros produce incorrect code.
|
|
|
1681
|
|
|
1682 If temacs fails to run at all, this may be the cause.
|
|
|
1683 Compile this test program and look at the assembler code:
|
|
|
1684
|
|
|
1685 struct foo { char x; unsigned int y : 24; };
|
|
|
1686
|
|
|
1687 lose (arg)
|
|
|
1688 struct foo arg;
|
|
|
1689 {
|
|
|
1690 test ((int *) arg.y);
|
|
|
1691 }
|
|
|
1692
|
|
|
1693 If the code is incorrect, your compiler has this problem.
|
|
|
1694 In the XCONS, etc., macros in lisp.h you must replace (a).u.val with
|
|
|
1695 ((a).u.val + coercedummy) where coercedummy is declared as int.
|
|
|
1696
|
|
|
1697 This problem will not happen if the m-...h file for your type
|
|
|
1698 of machine defines NO_UNION_TYPE. That is the recommended setting now.
|
|
|
1699
|
|
|
1700 * C compilers lose on returning unions
|
|
|
1701
|
|
|
1702 I hear that some C compilers cannot handle returning a union type.
|
|
|
1703 Most of the functions in GNU Emacs return type Lisp_Object, which is
|
|
|
1704 defined as a union on some rare architectures.
|
|
|
1705
|
|
120
|
1706 This problem will not happen if the m-...h file for your type of
|
|
|
1707 machine defines NO_UNION_TYPE. That is the recommended setting now.
|
|
88
|
1708
|
|
|
1709 * `Error: No ExtNode to pop!' on Linux systems with Lesstif.
|
|
|
1710
|
|
|
1711 This error message has been observed with lesstif-0.75a. It does not
|
|
|
1712 appear to cause any harm.
|
|
|
1713
|
|
|
1714 * Sparc Linux -vs- libXmu.
|
|
|
1715
|
|
|
1716 There have been reports of configure not detecting libXmu on
|
|
|
1717 SparcLinux. The fix is to add -lXmu to the link flags.
|
|
|
1718
|
|
|
1719 * Debian Linux and Berkeley db include files.
|
|
|
1720
|
|
|
1721 Debian Linux puts the Berkeley db include files in /usr/include/db
|
|
|
1722 instead of /usr/include. The fix is to use
|
|
|
1723 --site-includes=/usr/include/db with configure.
|
|
|
1724
|
|
|
1725 * Signaling: (error "Byte code stack underflow (byte compiler bug), pc 38")
|
|
|
1726
|
|
120
|
1727 This error is given when XEmacs 20 is compiled without MULE support
|
|
88
|
1728 but is attempting to load a .elc which requires MULE support. The fix
|
|
|
1729 is to rebytecompile the offending file.
|
|
|
1730
|
|
|
1731 * alloc.c will not compile without -P on HP-UX 9.05
|
|
|
1732
|
|
|
1733 Pekka Marjola <pema@iki.fi> writes:
|
|
120
|
1734 Gcc (2.7.2, with cpplib IIRC) required something (-P worked :) to
|
|
|
1735 get it to compile. Otherwise it failed on those DEFUN macros with
|
|
|
1736 comments inside parameter lists (like buffer.c, line 296).
|
|
88
|
1737
|
|
120
|
1738 * Excessive optimization with pgcc can break XEmacs
|
|
88
|
1739
|
|
|
1740 It has been reported on some systems that compiling with -O6 can lead
|
|
|
1741 to XEmacs failures. The workaround is to use a lower optimization
|
|
|
1742 level. -O2 and -O4 have been tested extensively.
|
|
|
1743
|
|
|
1744 * -O2 optimization on Irix 5.3 can cause compiler complaint.
|
|
|
1745
|
|
|
1746 Nick J. Crabtree <nickc@scopic.com> writes:
|
|
120
|
1747 Comes up OK on a tty (all I have available over this slow link). Ill
|
|
|
1748 give it a hammering tomorrow. The -O2 optimisation complained about
|
|
|
1749 sizes exceeding thresholds; I haven't bothered to use the -Olimit
|
|
|
1750 option it recommends.
|
|
88
|
1751
|
|
|
1752 * Excessive optimization on AIX 4.2 can lead to compiler failure.
|
|
|
1753
|
|
|
1754 Valdis.Kletnieks@vt.edu writes:
|
|
120
|
1755 At least at the b34 level, and the latest-and-greatest IBM xlc
|
|
|
1756 (3.1.4.4), there are problems with -O3. I haven't investigated
|
|
|
1757 further.
|
|
88
|
1758
|
|
|
1759 * Sed problems on Solaris 2.5
|
|
|
1760
|
|
|
1761 There have been reports of Sun sed truncating very lines in the
|
|
120
|
1762 Makefile during configuration. The workaround is to use GNU sed or,
|
|
|
1763 even better, think of a better way to generate Makefile, and send us a
|
|
|
1764 patch. :-)
|
|
88
|
1765
|
|
|
1766 * CDE is not autodetected on HP.
|
|
|
1767
|
|
|
1768 Richard Cognot <cognot@ensg.u-nancy.fr> writes:
|
|
120
|
1769 I have to force /usr/dt/{lib,include} into the site include/lib
|
|
|
1770 command line options. I could add these in hpux10.h, but then I
|
|
|
1771 would think these should be pretty standard (to my knowledge, that's
|
|
|
1772 also where Sun puts its CDE stuff), so that wouldn't fix the problem
|
|
|
1773 on other architectures. AAMOF, when these path are given, CDE is
|
|
|
1774 detected, and DragAndDrop works (more or less, see next issue).
|
|
88
|
1775
|
|
|
1776 * Signalling: (wrong-type-argument ...) when loading mail-abbrevs
|
|
|
1777
|
|
|
1778 The is seen when installing the Big Brother Data Base (bbdb) which
|
|
|
1779 includes an outdated copy of mail-abbrevs.el. Remove the copy that
|
|
|
1780 comes with bbdb and use the one that comes with XEmacs.
|
|
116
|
1781
|
|
|
1782 * Linking with -rpath on IRIX.
|
|
|
1783
|
|
|
1784 Darrell Kindred <dkindred@cmu.edu> writes:
|
|
|
1785 There are a couple of problems [with use of -rpath with Irix ld], though:
|
|
|
1786
|
|
|
1787 1. The ld in IRIX 5.3 ignores all but the last -rpath
|
|
|
1788 spec, so the patched configure spits out a warning
|
|
|
1789 if --x-libraries or --site-runtime-libraries are
|
|
|
1790 specified under irix 5.x, and it only adds -rpath
|
|
|
1791 entries for the --site-runtime-libraries. This bug was
|
|
|
1792 fixed sometime between 5.3 and 6.2.
|
|
|
1793
|
|
|
1794 2. IRIX gcc 2.7.2 doesn't accept -rpath directly, so
|
|
|
1795 it would have to be prefixed by -Xlinker or "-Wl,".
|
|
|
1796 This would be fine, except that configure compiles with
|
|
|
1797 ${CC-cc} $CFLAGS $LDFLAGS ...
|
|
|
1798 rather than quoting $LDFLAGS with prefix-args, like
|
|
|
1799 src/Makefile does. So if you specify --x-libraries
|
|
|
1800 or --site-runtime-libraries, you must use --use-gcc=no,
|
|
|
1801 or configure will fail.
|
|
|
1802
|
|
|
1803 * On Irix 5.x and 6.x, the dumped XEmacs (xemacs) core dumps when executed
|
|
|
1804 on another machine, or after newer SGI IRIX patches have been installed.
|
|
|
1805
|
|
120
|
1806 The xemacs binary must be executed with the same "libc.so" file which
|
|
|
1807 was used when the xemacs binary was dumped. Some SGI IRIX patches
|
|
|
1808 update this file. Make sure that all machines using the xemacs binary
|
|
|
1809 are using the same set of IRIX patches. If xemacs core dumps after a
|
|
|
1810 patch upgrade then you will have to redump it from temacs.
|
|
116
|
1811
|
|
|
1812 * xemacs: can't resolve symbol '__malloc_hook'
|
|
|
1813
|
|
|
1814 This is a Linux problem where you've compiled the XEmacs binary on a libc
|
|
|
1815 5.4 with version higher than 5.4.19 and attempted to run the binary against
|
|
|
1816 an earlier version. The solution is to upgrade your old library.
|
|
|
1817
|
|
|
1818 * VM appears to hang in large folders
|
|
|
1819
|
|
120
|
1820 This is normal (trust us) when upgrading to VM-6.22 from earlier
|
|
|
1821 versions. Let VM finish what it is doing and all will be well.
|
