Mercurial > emacs
annotate etc/PROBLEMS @ 34001:fc9ba8a24dde
*** empty log message ***
author | Dave Love <fx@gnu.org> |
---|---|
date | Wed, 29 Nov 2000 15:02:30 +0000 |
parents | 686023ece47f |
children | 30fe602d7443 |
rev | line source |
---|---|
25853 | 1 This file describes various problems that have been encountered |
2 in compiling, installing and running GNU Emacs. | |
3 | |
33964 | 4 * On systems with shared libraries you might encounter run-time errors |
5 from the dynamic linker telling you that it is unable to find some | |
6 shared libraries, for instance those for Xaw3d or image support. | |
7 These errors mean Emacs has been linked with a library whose shared | |
8 library is not in the default search path of the dynamic linker. | |
9 | |
10 On many systems, it is possible to set LD_LIBRARY_PATH in your | |
11 environment to specify additional directories where shared libraries | |
12 can be found. | |
13 | |
14 Other systems allow to set LD_RUN_PATH in a similar way, but before | |
15 Emacs is linked. With LD_RUN_PATH set, the linker will include a | |
16 specified run-time search path in the executable. | |
17 | |
18 Please refer to the documentation of your dynamic linker for details. | |
19 | |
33788 | 20 * On Solaris 2.7, building Emacs with WorkShop Compilers 5.0 98/12/15 |
34001 | 21 C 5.0 failed, apparently with non-default CFLAGS, most probably due to |
22 compiler bugs. Using Sun Solaris 2.7 Sun WorkShop 6 update 1 C | |
23 release was reported to work without problems. It worked OK on | |
24 another system with Solaris 8 using apparently the same 5.0 compiler | |
25 and the default CFLAGS. | |
33788 | 26 |
33614
9c067a999061
Added note about subprocesses on Windows 9x.
Jason Rumney <jasonr@gnu.org>
parents:
33455
diff
changeset
|
27 * On Windows 95/98/ME, subprocesses do not terminate properly. |
9c067a999061
Added note about subprocesses on Windows 9x.
Jason Rumney <jasonr@gnu.org>
parents:
33455
diff
changeset
|
28 |
9c067a999061
Added note about subprocesses on Windows 9x.
Jason Rumney <jasonr@gnu.org>
parents:
33455
diff
changeset
|
29 This is a limitation of the Operating System, and can cause problems |
9c067a999061
Added note about subprocesses on Windows 9x.
Jason Rumney <jasonr@gnu.org>
parents:
33455
diff
changeset
|
30 when shutting down Windows. Ensure that all subprocesses are exited |
9c067a999061
Added note about subprocesses on Windows 9x.
Jason Rumney <jasonr@gnu.org>
parents:
33455
diff
changeset
|
31 cleanly before exiting Emacs. For more details, see the FAQ at |
9c067a999061
Added note about subprocesses on Windows 9x.
Jason Rumney <jasonr@gnu.org>
parents:
33455
diff
changeset
|
32 ftp://ftp.gnu.org/gnu/windows/emacs/doc/index.html |
9c067a999061
Added note about subprocesses on Windows 9x.
Jason Rumney <jasonr@gnu.org>
parents:
33455
diff
changeset
|
33 |
33455 | 34 * Mail sent through Microsoft Exchange in some encodings appears to be |
35 mangled and is not seen correctly in Rmail or Gnus. We don't know | |
36 exactly what happens, but it isn't an Emacs problem in cases we've | |
37 seen. | |
38 | |
31514 | 39 * On OSF/Dec Unix/Tru64/<whatever it is this year> under X locally or |
40 remotely, M-SPC acts as a `compose' key with strange results. See | |
41 keyboard(5). | |
42 | |
43 Changing Alt_L to Meta_L fixes it: | |
44 % xmodmap -e 'keysym Alt_L = Meta_L Alt_L' | |
45 % xmodmap -e 'keysym Alt_R = Meta_R Alt_R' | |
46 | |
25853 | 47 * Error "conflicting types for `initstate'" compiling with GCC on Irix 6. |
48 | |
49 Install GCC 2.95 or a newer version, and this problem should go away. | |
50 It is possible that this problem results from upgrading the operating | |
51 system without reinstalling GCC; so you could also try reinstalling | |
52 the same version of GCC, and telling us whether that fixes the problem. | |
53 | |
54 * On Solaris 7, Emacs gets a segmentation fault when starting up using X. | |
55 | |
56 This results from Sun patch 107058-01 (SunOS 5.7: Patch for | |
26664
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
57 assembler) if you use GCC version 2.7 or later. |
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
58 To work around it, either install patch 106950-03 or later, |
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
59 or uninstall patch 107058-01, or install the GNU Binutils. |
25853 | 60 Then recompile Emacs, and it should work. |
61 | |
62 * With X11R6.4, public-patch-3, Emacs crashes at startup. | |
63 | |
64 Reportedly this patch in X fixes the problem. | |
65 | |
66 --- xc/lib/X11/imInt.c~ Wed Jun 30 13:31:56 1999 | |
67 +++ xc/lib/X11/imInt.c Thu Jul 1 15:10:27 1999 | |
68 @@ -1,4 +1,4 @@ | |
69 -/* $TOG: imInt.c /main/5 1998/05/30 21:11:16 kaleb $ */ | |
70 +/* $TOG: imInt.c /main/5 1998/05/30 21:11:16 kaleb $ */ | |
71 /****************************************************************** | |
72 | |
73 Copyright 1992, 1993, 1994 by FUJITSU LIMITED | |
74 @@ -166,8 +166,8 @@ | |
75 _XimMakeImName(lcd) | |
76 XLCd lcd; | |
77 { | |
78 - char* begin; | |
79 - char* end; | |
80 + char* begin = NULL; | |
81 + char* end = NULL; | |
82 char* ret; | |
83 int i = 0; | |
84 char* ximmodifier = XIMMODIFIER; | |
85 @@ -182,7 +182,11 @@ | |
86 } | |
87 ret = Xmalloc(end - begin + 2); | |
88 if (ret != NULL) { | |
89 - (void)strncpy(ret, begin, end - begin + 1); | |
90 + if (begin != NULL) { | |
91 + (void)strncpy(ret, begin, end - begin + 1); | |
92 + } else { | |
93 + ret[0] = '\0'; | |
94 + } | |
95 ret[end - begin + 1] = '\0'; | |
96 } | |
97 return ret; | |
98 | |
99 | |
100 * Emacs crashes on Irix 6.5 on the SGI R10K, when compiled with GCC. | |
101 | |
102 This seems to be fixed in GCC 2.95. | |
103 | |
104 * Emacs crashes in utmpname on Irix 5.3. | |
105 | |
106 This problem is fixed in Patch 3175 for Irix 5.3. | |
107 It is also fixed in Irix versions 6.2 and up. | |
108 | |
109 * On Solaris, CTRL-t is ignored by Emacs when you use | |
110 the fr.ISO-8859-15 locale (and maybe other related locales). | |
111 | |
112 You can fix this by editing the file: | |
113 | |
114 /usr/openwin/lib/locale/iso8859-15/Compose | |
115 | |
116 Near the bottom there is a line that reads: | |
117 | |
118 Ctrl<t> <quotedbl> <Y> : "\276" threequarters | |
119 | |
120 that should read: | |
121 | |
122 Ctrl<T> <quotedbl> <Y> : "\276" threequarters | |
123 | |
124 Note the lower case <t>. Changing this line should make C-t work. | |
125 | |
126 * Emacs on Digital Unix 4.0 fails to build, giving error message | |
127 Invalid dimension for the charset-ID 160 | |
128 | |
129 This is due to a bug or an installation problem in GCC 2.8.0. | |
130 Installing a more recent version of GCC fixes the problem. | |
131 | |
132 * Buffers from `with-output-to-temp-buffer' get set up in Help mode. | |
133 | |
134 Changes in Emacs 20.4 to the hooks used by that function cause | |
135 problems for some packages, specifically BBDB. See the function's | |
136 documentation for the hooks involved. BBDB 2.00.06 fixes the problem. | |
137 | |
138 * Under X, C-v and/or other keys don't work. | |
139 | |
140 These may have been intercepted by your window manager. In | |
141 particular, AfterStep 1.6 is reported to steal C-v in its default | |
142 configuration. Various Meta keys are also likely to be taken by the | |
143 configuration of the `feel'. See the WM's documentation for how to | |
144 change this. | |
145 | |
146 * When using Exceed, fonts sometimes appear too tall. | |
147 | |
148 When the display is set to an Exceed X-server and fonts are specified | |
149 (either explicitly with the -fn option or implicitly with X resources) | |
150 then the fonts may appear "too tall". The actual character sizes are | |
151 correct but there is too much vertical spacing between rows, which | |
152 gives the appearance of "double spacing". | |
153 | |
154 To prevent this, turn off the Exceed's "automatic font substitution" | |
155 feature (in the font part of the configuration window). | |
156 | |
157 * Failure in unexec while dumping emacs on Digital Unix 4.0 | |
158 | |
159 This problem manifests itself as an error message | |
160 | |
161 unexec: Bad address, writing data section to ... | |
162 | |
163 The user suspects that this happened because his X libraries | |
164 were built for an older system version, | |
165 | |
166 ./configure --x-includes=/usr/include --x-libraries=/usr/shlib | |
167 | |
168 made the problem go away. | |
169 | |
170 * No visible display on mips-sgi-irix6.2 when compiling with GCC 2.8.1. | |
171 | |
172 This problem went away after installing the latest IRIX patches | |
173 as of 8 Dec 1998. | |
174 | |
175 The same problem has been reported on Irix 6.3. | |
176 | |
177 * As of version 20.4, Emacs doesn't work properly if configured for | |
178 the Motif toolkit and linked against the free LessTif library. The | |
179 next Emacs release is expected to work with LessTif. | |
180 | |
181 * Emacs gives the error, Couldn't find per display information. | |
182 | |
183 This can result if the X server runs out of memory because Emacs uses | |
184 a large number of fonts. On systems where this happens, C-h h is | |
185 likely to cause it. | |
186 | |
187 We do not know of a way to prevent the problem. | |
188 | |
189 * Emacs makes HPUX 11.0 crash. | |
190 | |
191 This is a bug in HPUX; HPUX patch PHKL_16260 is said to fix it. | |
192 | |
193 * Emacs crashes during dumping on the HPPA machine (HPUX 10.20). | |
194 | |
195 This seems to be due to a GCC bug; it is fixed in GCC 2.8.1. | |
196 | |
197 * The Hyperbole package causes *Help* buffers not to be displayed in | |
198 Help mode due to setting `temp-buffer-show-hook' rather than using | |
199 `add-hook'. Using `(add-hook 'temp-buffer-show-hook | |
200 'help-mode-maybe)' after loading Hyperbole should fix this. | |
201 | |
202 * Versions of the PSGML package earlier than 1.0.3 (stable) or 1.1.2 | |
203 (alpha) fail to parse DTD files correctly in Emacs 20.3 and later. | |
204 Here is a patch for psgml-parse.el from PSGML 1.0.1 and, probably, | |
205 earlier versions. | |
206 | |
207 --- psgml-parse.el 1998/08/21 19:18:18 1.1 | |
208 +++ psgml-parse.el 1998/08/21 19:20:00 | |
209 @@ -2383,7 +2383,7 @@ (defun sgml-push-to-entity (entity &opti | |
210 (setq sgml-buffer-parse-state nil)) | |
211 (cond | |
212 ((stringp entity) ; a file name | |
213 - (save-excursion (insert-file-contents entity)) | |
214 + (insert-file-contents entity) | |
215 (setq default-directory (file-name-directory entity))) | |
216 ((consp (sgml-entity-text entity)) ; external id? | |
217 (let* ((extid (sgml-entity-text entity)) | |
218 | |
219 * Running TeX from AUXTeX package with Emacs 20.3 gives a Lisp error | |
220 about a read-only tex output buffer. | |
221 | |
222 This problem appeared for AUC TeX version 9.9j and some earlier | |
223 versions. Here is a patch for the file tex-buf.el in the AUC TeX | |
224 package. | |
225 | |
226 diff -c auctex/tex-buf.el~ auctex/tex-buf.el | |
227 *** auctex/tex-buf.el~ Wed Jul 29 18:35:32 1998 | |
228 --- auctex/tex-buf.el Sat Sep 5 15:20:38 1998 | |
229 *************** | |
230 *** 545,551 **** | |
231 (dir (TeX-master-directory))) | |
232 (TeX-process-check file) ; Check that no process is running | |
233 (setq TeX-command-buffer (current-buffer)) | |
234 ! (with-output-to-temp-buffer buffer) | |
235 (set-buffer buffer) | |
236 (if dir (cd dir)) | |
237 (insert "Running `" name "' on `" file "' with ``" command "''\n") | |
238 - --- 545,552 ---- | |
239 (dir (TeX-master-directory))) | |
240 (TeX-process-check file) ; Check that no process is running | |
241 (setq TeX-command-buffer (current-buffer)) | |
242 ! (let (temp-buffer-show-function temp-buffer-show-hook) | |
243 ! (with-output-to-temp-buffer buffer)) | |
244 (set-buffer buffer) | |
245 (if dir (cd dir)) | |
246 (insert "Running `" name "' on `" file "' with ``" command "''\n") | |
247 | |
248 * On Irix 6.3, substituting environment variables in file names | |
249 in the minibuffer gives peculiar error messages such as | |
250 | |
251 Substituting nonexistent environment variable "" | |
252 | |
253 This is not an Emacs bug; it is caused by something in SGI patch | |
254 003082 August 11, 1998. | |
255 | |
256 * After a while, Emacs slips into unibyte mode. | |
257 | |
258 The VM mail package, which is not part of Emacs, sometimes does | |
259 (standard-display-european t) | |
260 That should be changed to | |
261 (standard-display-european 1 t) | |
262 | |
263 * Installing Emacs gets an error running `install-info'. | |
264 | |
265 You need to install a recent version of Texinfo; that package | |
266 supplies the `install-info' command. | |
267 | |
268 * Emacs does not recognize the AltGr key, on HPUX. | |
269 | |
270 To fix this, set up a file ~/.dt/sessions/sessionetc with executable | |
271 rights, containing this text: | |
272 | |
273 -------------------------------- | |
274 xmodmap 2> /dev/null - << EOF | |
275 keysym Alt_L = Meta_L | |
276 keysym Alt_R = Meta_R | |
277 EOF | |
278 | |
279 xmodmap - << EOF | |
280 clear mod1 | |
281 keysym Mode_switch = NoSymbol | |
282 add mod1 = Meta_L | |
283 keysym Meta_R = Mode_switch | |
284 add mod2 = Mode_switch | |
285 EOF | |
286 -------------------------------- | |
287 | |
288 * Emacs compiled with DJGPP for MS-DOS/MS-Windows cannot access files | |
289 in the directory with the special name `dev' under the root of any | |
290 drive, e.g. `c:/dev'. | |
291 | |
292 This is an unfortunate side-effect of the support for Unix-style | |
293 device names such as /dev/null in the DJGPP runtime library. A | |
294 work-around is to rename the problem directory to another name. | |
295 | |
296 * M-SPC seems to be ignored as input. | |
297 | |
298 See if your X server is set up to use this as a command | |
299 for character composition. | |
300 | |
301 * Emacs startup on GNU/Linux systems (and possibly other systems) is slow. | |
302 | |
303 This can happen if the system is misconfigured and Emacs can't get the | |
304 full qualified domain name, FQDN. You should have your FQDN in the | |
305 /etc/hosts file, something like this: | |
306 | |
307 127.0.0.1 localhost | |
308 129.187.137.82 nuc04.t30.physik.tu-muenchen.de nuc04 | |
309 | |
310 The way to set this up may vary on non-GNU systems. | |
311 | |
312 * Garbled display on non-X terminals when Emacs runs on Digital Unix 4.0. | |
313 | |
314 So far it appears that running `tset' triggers this problem (when TERM | |
315 is vt100, at least). If you do not run `tset', then Emacs displays | |
316 properly. If someone can tell us precisely which effect of running | |
317 `tset' actually causes the problem, we may be able to implement a fix | |
318 in Emacs. | |
319 | |
320 * When you run Ispell from Emacs, it reports a "misalignment" error. | |
321 | |
322 This can happen if you compiled Ispell to use ASCII characters only | |
323 and then try to use it from Emacs with non-ASCII characters, | |
324 specifically Latin-1. The solution is to recompile Ispell with | |
325 Latin-1 support. | |
326 | |
327 This can also happen if the version of Ispell installed on your | |
328 machine is old. | |
329 | |
330 * On Linux-based GNU systems using libc versions 5.4.19 through | |
331 5.4.22, Emacs crashes at startup with a segmentation fault. | |
332 | |
333 This problem happens if libc defines the symbol __malloc_initialized. | |
334 One known solution is to upgrade to a newer libc version. 5.4.33 is | |
335 known to work. | |
336 | |
337 * On Windows, you cannot use the right-hand ALT key and the left-hand | |
338 CTRL key together to type a Control-Meta character. | |
339 | |
340 This is a consequence of a misfeature beyond Emacs's control. | |
341 | |
342 Under Windows, the AltGr key on international keyboards generates key | |
343 events with the modifiers Right-Alt and Left-Ctrl. Since Emacs cannot | |
344 distinguish AltGr from an explicit Right-Alt and Left-Ctrl | |
345 combination, whenever it sees Right-Alt and Left-Ctrl it assumes that | |
346 AltGr has been pressed. | |
347 | |
348 * Under some Windows X-servers, Emacs' display is incorrect | |
349 | |
350 The symptoms are that Emacs does not completely erase blank areas of the | |
351 screen during scrolling or some other screen operations (e.g., selective | |
352 display or when killing a region). M-x recenter will cause the screen | |
353 to be completely redisplayed and the "extra" characters will disappear. | |
354 | |
355 This is known to occur under Exceed 6, and possibly earlier versions as | |
356 well. The problem lies in the X-server settings. | |
357 | |
358 There are reports that you can solve the problem with Exceed by | |
359 running `Xconfig' from within NT, choosing "X selection", then | |
360 un-checking the boxes "auto-copy X selection" and "auto-paste to X | |
361 selection". | |
362 | |
363 Of this does not work, please inform bug-gnu-emacs@gnu.org. Then | |
364 please call support for your X-server and see if you can get a fix. | |
365 If you do, please send it to bug-gnu-emacs@gnu.org so we can list it | |
366 here. | |
367 | |
368 * On Solaris 2, Emacs dumps core when built with Motif. | |
369 | |
370 The Solaris Motif libraries are buggy, at least up through Solaris 2.5.1. | |
371 Install the current Motif runtime library patch appropriate for your host. | |
372 (Make sure the patch is current; some older patch versions still have the bug.) | |
373 You should install the other patches recommended by Sun for your host, too. | |
374 You can obtain Sun patches from ftp://sunsolve.sun.com/pub/patches/; | |
375 look for files with names ending in `.PatchReport' to see which patches | |
376 are currently recommended for your host. | |
377 | |
378 On Solaris 2.6, Emacs is said to work with Motif when Solaris patch | |
379 105284-12 is installed, but fail when 105284-15 is installed. | |
380 105284-18 might fix it again. | |
381 | |
26664
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
382 * On Solaris 2.6 and 7, the Compose key does not work. |
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
383 |
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
384 This is a bug in Motif in Solaris. Supposedly it has been fixed for |
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
385 the next major release of Solaris. However, if someone with Sun |
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
386 support complains to Sun about the bug, they may release a patch. |
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
387 If you do this, mention Sun bug #4188711. |
25853 | 388 |
389 One workaround is to use a locale that allows non-ASCII characters. | |
390 For example, before invoking emacs, set the LC_ALL environment | |
391 variable to "en_US" (American English). The directory /usr/lib/locale | |
392 lists the supported locales; any locale other than "C" or "POSIX" | |
393 should do. | |
394 | |
26664
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
395 pen@lysator.liu.se says (Feb 1998) that the Compose key does work |
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
396 if you link with the MIT X11 libraries instead of the Solaris X11 |
a7963e66c555
Update Solaris 2.6 and 7 problems.
Paul Eggert <eggert@twinsun.com>
parents:
25853
diff
changeset
|
397 libraries. |
25853 | 398 |
399 * Emacs does not know your host's fully-qualified domain name. | |
400 | |
401 You need to configure your machine with a fully qualified domain name, | |
402 either in /etc/hosts, /etc/hostname, the NIS, or wherever your system | |
403 calls for specifying this. | |
404 | |
405 If you cannot fix the configuration, you can set the Lisp variable | |
406 mail-host-address to the value you want. | |
407 | |
408 * Error 12 (virtual memory exceeded) when dumping Emacs, on UnixWare 2.1 | |
409 | |
410 Paul Abrahams (abrahams@acm.org) reports that with the installed | |
411 virtual memory settings for UnixWare 2.1.2, an Error 12 occurs during | |
412 the "make" that builds Emacs, when running temacs to dump emacs. That | |
413 error indicates that the per-process virtual memory limit has been | |
414 exceeded. The default limit is probably 32MB. Raising the virtual | |
415 memory limit to 40MB should make it possible to finish building Emacs. | |
416 | |
417 You can do this with the command `ulimit' (sh) or `limit' (csh). | |
418 But you have to be root to do it. | |
419 | |
420 According to Martin Sohnius, you can also retune this in the kernel: | |
421 | |
422 # /etc/conf/bin/idtune SDATLIM 33554432 ## soft data size limit | |
423 # /etc/conf/bin/idtune HDATLIM 33554432 ## hard " | |
424 # /etc/conf/bin/idtune SVMMSIZE unlimited ## soft process size limit | |
425 # /etc/conf/bin/idtune HVMMSIZE unlimited ## hard " | |
426 # /etc/conf/bin/idbuild -B | |
427 | |
428 (He recommends you not change the stack limit, though.) | |
429 These changes take effect when you reboot. | |
430 | |
431 * Redisplay using X11 is much slower than previous Emacs versions. | |
432 | |
433 We've noticed that certain X servers draw the text much slower when | |
434 scroll bars are on the left. We don't know why this happens. If this | |
435 happens to you, you can work around it by putting the scroll bars | |
436 on the right (as they were in Emacs 19). | |
437 | |
438 Here's how to do this: | |
439 | |
440 (set-scroll-bar-mode 'right) | |
441 | |
442 If you're not sure whether (or how much) this problem affects you, | |
443 try that and see how much difference it makes. To set things back | |
444 to normal, do | |
445 | |
446 (set-scroll-bar-mode 'left) | |
447 | |
448 * Under X11, some characters appear as hollow boxes. | |
449 | |
450 Each X11 font covers just a fraction of the characters that Emacs | |
451 supports. To display the whole range of Emacs characters requires | |
452 many different fonts, collected into a fontset. | |
453 | |
454 If some of the fonts called for in your fontset do not exist on your X | |
455 server, then the characters that have no font appear as hollow boxes. | |
456 You can remedy the problem by installing additional fonts. | |
457 | |
458 The intlfonts distribution includes a full spectrum of fonts that can | |
459 display all the characters Emacs supports. | |
460 | |
461 * Under X11, some characters appear improperly aligned in their lines. | |
462 | |
463 You may have bad X11 fonts; try installing the intlfonts distribution. | |
464 | |
465 * Certain fonts make each line take one pixel more than it "should". | |
466 | |
467 This is because these fonts contain characters a little taller | |
468 than the font's nominal height. Emacs needs to make sure that | |
469 lines do not overlap. | |
470 | |
471 * You request inverse video, and the first Emacs frame is in inverse | |
472 video, but later frames are not in inverse video. | |
473 | |
474 This can happen if you have an old version of the custom library in | |
475 your search path for Lisp packages. Use M-x list-load-path-shadows to | |
476 check whether this is true. If it is, delete the old custom library. | |
477 | |
478 * In FreeBSD 2.1.5, useless symbolic links remain in /tmp or other | |
479 directories that have the +t bit. | |
480 | |
481 This is because of a kernel bug in FreeBSD 2.1.5 (fixed in 2.2). | |
482 Emacs uses symbolic links to implement file locks. In a directory | |
483 with +t bit, the directory owner becomes the owner of the symbolic | |
484 link, so that it cannot be removed by anyone else. | |
485 | |
486 If you don't like those useless links, you can let Emacs not to using | |
487 file lock by adding #undef CLASH_DETECTION to config.h. | |
488 | |
489 * When using M-x dbx with the SparcWorks debugger, the `up' and `down' | |
490 commands do not move the arrow in Emacs. | |
491 | |
492 You can fix this by adding the following line to `~/.dbxinit': | |
493 | |
494 dbxenv output_short_file_name off | |
495 | |
496 * Emacs says it has saved a file, but the file does not actually | |
497 appear on disk. | |
498 | |
499 This can happen on certain systems when you are using NFS, if the | |
500 remote disk is full. It is due to a bug in NFS (or certain NFS | |
501 implementations), and there is apparently nothing Emacs can do to | |
502 detect the problem. Emacs checks the failure codes of all the system | |
503 calls involved in writing a file, including `close'; but in the case | |
504 where the problem occurs, none of those system calls fails. | |
505 | |
506 * "Compose Character" key does strange things when used as a Meta key. | |
507 | |
508 If you define one key to serve as both Meta and Compose Character, you | |
509 will get strange results. In previous Emacs versions, this "worked" | |
510 in that the key acted as Meta--that's because the older Emacs versions | |
511 did not try to support Compose Character. Now Emacs tries to do | |
512 character composition in the standard X way. This means that you | |
513 must pick one meaning or the other for any given key. | |
514 | |
515 You can use both functions (Meta, and Compose Character) if you assign | |
516 them to two different keys. | |
517 | |
518 * Emacs gets a segmentation fault at startup, on AIX4.2. | |
519 | |
520 If you are using IBM's xlc compiler, compile emacs.c | |
521 without optimization; that should avoid the problem. | |
522 | |
523 * movemail compiled with POP support can't connect to the POP server. | |
524 | |
525 Make sure that the `pop' entry in /etc/services, or in the services | |
526 NIS map if your machine uses NIS, has the same port number as the | |
527 entry on the POP server. A common error is for the POP server to be | |
528 listening on port 110, the assigned port for the POP3 protocol, while | |
529 the client is trying to connect on port 109, the assigned port for the | |
530 old POP protocol. | |
531 | |
532 * Emacs crashes in x-popup-dialog. | |
533 | |
534 This can happen if the dialog widget cannot find the font it wants to | |
535 use. You can work around the problem by specifying another font with | |
536 an X resource--for example, `Emacs.dialog*.font: 9x15' (or any font that | |
537 happens to exist on your X server). | |
538 | |
539 * Emacs crashes when you use Bibtex mode. | |
540 | |
541 This happens if your system puts a small limit on stack size. You can | |
542 prevent the problem by using a suitable shell command (often `ulimit') | |
543 to raise the stack size limit before you run Emacs. | |
544 | |
545 Patches to raise the stack size limit automatically in `main' | |
546 (src/emacs.c) on various systems would be greatly appreciated. | |
547 | |
548 * Emacs crashes with SIGBUS or SIGSEGV on HPUX 9 after you delete a frame. | |
549 | |
550 We think this is due to a bug in the X libraries provided by HP. With | |
551 the alternative X libraries in /usr/contrib/mitX11R5/lib, the problem | |
552 does not happen. | |
553 | |
554 * Emacs crashes with SIGBUS or SIGSEGV on Solaris after you delete a frame. | |
555 | |
556 We suspect that this is a similar bug in the X libraries provided by | |
557 Sun. There is a report that one of these patches fixes the bug and | |
558 makes the problem stop: | |
559 | |
560 105216-01 105393-01 105518-01 105621-01 105665-01 105615-02 105216-02 | |
561 105667-01 105401-08 105615-03 105621-02 105686-02 105736-01 105755-03 | |
562 106033-01 105379-01 105786-01 105181-04 105379-03 105786-04 105845-01 | |
563 105284-05 105669-02 105837-01 105837-02 105558-01 106125-02 105407-01 | |
564 | |
565 Another person using a newer system (kernel patch level Generic_105181-06) | |
566 suspects that the bug was fixed by one of these more recent patches: | |
567 | |
568 106040-07 SunOS 5.6: X Input & Output Method patch | |
569 106222-01 OpenWindows 3.6: filemgr (ff.core) fixes | |
570 105284-12 Motif 1.2.7: sparc Runtime library patch | |
571 | |
572 * Problems running Perl under Emacs on Windows NT/95. | |
573 | |
574 `perl -de 0' just hangs when executed in an Emacs subshell. | |
575 The fault lies with Perl (indirectly with Windows NT/95). | |
576 | |
577 The problem is that the Perl debugger explicitly opens a connection to | |
578 "CON", which is the DOS/NT equivalent of "/dev/tty", for interacting | |
579 with the user. | |
580 | |
581 On Unix, this is okay, because Emacs (or the shell?) creates a | |
582 pseudo-tty so that /dev/tty is really the pipe Emacs is using to | |
583 communicate with the subprocess. | |
584 | |
585 On NT, this fails because CON always refers to the handle for the | |
586 relevant console (approximately equivalent to a tty), and cannot be | |
587 redirected to refer to the pipe Emacs assigned to the subprocess as | |
588 stdin. | |
589 | |
590 A workaround is to modify perldb.pl to use STDIN/STDOUT instead of CON. | |
591 | |
592 For Perl 4: | |
593 | |
594 *** PERL/LIB/PERLDB.PL.orig Wed May 26 08:24:18 1993 | |
595 --- PERL/LIB/PERLDB.PL Mon Jul 01 15:28:16 1996 | |
596 *************** | |
597 *** 68,74 **** | |
598 $rcfile=".perldb"; | |
599 } | |
600 else { | |
601 ! $console = "con"; | |
602 $rcfile="perldb.ini"; | |
603 } | |
604 | |
605 --- 68,74 ---- | |
606 $rcfile=".perldb"; | |
607 } | |
608 else { | |
609 ! $console = ""; | |
610 $rcfile="perldb.ini"; | |
611 } | |
612 | |
613 | |
614 For Perl 5: | |
615 *** perl/5.001/lib/perl5db.pl.orig Sun Jun 04 21:13:40 1995 | |
616 --- perl/5.001/lib/perl5db.pl Mon Jul 01 17:00:08 1996 | |
617 *************** | |
618 *** 22,28 **** | |
619 $rcfile=".perldb"; | |
620 } | |
621 elsif (-e "con") { | |
622 ! $console = "con"; | |
623 $rcfile="perldb.ini"; | |
624 } | |
625 else { | |
626 --- 22,28 ---- | |
627 $rcfile=".perldb"; | |
628 } | |
629 elsif (-e "con") { | |
630 ! $console = ""; | |
631 $rcfile="perldb.ini"; | |
632 } | |
633 else { | |
634 | |
635 * Problems running DOS programs on Windows NT versions earlier than 3.51. | |
636 | |
637 Some DOS programs, such as pkzip/pkunzip will not work at all, while | |
638 others will only work if their stdin is redirected from a file or NUL. | |
639 | |
640 When a DOS program does not work, a new process is actually created, but | |
641 hangs. It cannot be interrupted from Emacs, and might need to be killed | |
642 by an external program if Emacs is hung waiting for the process to | |
643 finish. If Emacs is not waiting for it, you should be able to kill the | |
644 instance of ntvdm that is running the hung process from Emacs, if you | |
645 can find out the process id. | |
646 | |
647 It is safe to run most DOS programs using call-process (eg. M-! and | |
648 M-|) since stdin is then redirected from a file, but not with | |
649 start-process since that redirects stdin to a pipe. Also, running DOS | |
650 programs in a shell buffer prompt without redirecting stdin does not | |
651 work. | |
652 | |
653 * Problems on MS-DOG if DJGPP v2.0 is used to compile Emacs: | |
654 | |
655 There are two DJGPP library bugs which cause problems: | |
656 | |
657 * Running `shell-command' (or `compile', or `grep') you get | |
658 `Searching for program: permission denied (EACCES), c:/command.com'; | |
659 * After you shell to DOS, Ctrl-Break kills Emacs. | |
660 | |
661 To work around these bugs, you can use two files in the msdos | |
662 subdirectory: `is_exec.c' and `sigaction.c'. Compile them and link | |
663 them into the Emacs executable `temacs'; then they will replace the | |
664 incorrect library functions. | |
665 | |
666 * When compiling with DJGPP on Windows 95, Make fails for some targets | |
667 like make-docfile. | |
668 | |
669 This can happen if long file name support (the setting of environment | |
670 variable LFN) when Emacs distribution was unpacked and during | |
671 compilation are not the same. See the MSDOG section of INSTALL for | |
672 the explanation of how to avoid this problem. | |
673 | |
674 * Emacs compiled for MSDOS cannot find some Lisp files, or other | |
675 run-time support files, when long filename support is enabled. | |
676 (Usually, this problem will manifest itself when Emacs exits | |
677 immediately after flashing the startup screen, because it cannot find | |
678 the Lisp files it needs to load at startup. Redirect Emacs stdout | |
679 and stderr to a file to see the error message printed by Emacs.) | |
680 | |
681 This can happen if the Emacs distribution was unzipped without LFN | |
682 support, thus causing long filenames to be truncated to the first 6 | |
683 characters and a numeric tail that Windows 95 normally attaches to it. | |
684 You should unzip the files again with a utility that supports long | |
685 filenames (such as djtar from DJGPP or InfoZip's UnZip program | |
686 compiled with DJGPP v2). The MSDOG section of the file INSTALL | |
687 explains this issue in more detail. | |
688 | |
689 * Emacs compiled with DJGPP complains at startup: | |
690 | |
691 "Wrong type of argument: internal-facep, msdos-menu-active-face" | |
692 | |
693 This can happen if you define an environment variable `TERM'. Emacs | |
694 on MSDOS uses an internal terminal emulator which is disabled if the | |
695 value of `TERM' is anything but the string "internal". Emacs then | |
696 works as if its terminal were a dumb glass teletype that doesn't | |
697 support faces. To work around this, arrange for `TERM' to be | |
698 undefined when Emacs runs. The best way to do that is to add an | |
699 [emacs] section to the DJGPP.ENV file which defines an empty value for | |
700 `TERM'; this way, only Emacs gets the empty value, while the rest of | |
701 your system works as before. | |
702 | |
703 * On Windows 95, Alt-f6 does not get through to Emacs. | |
704 | |
705 This character seems to be trapped by the kernel in Windows 95. | |
706 You can enter M-f6 by typing ESC f6. | |
707 | |
708 * Typing Alt-Shift has strange effects on Windows 95. | |
709 | |
710 This combination of keys is a command to change keyboard layout. If | |
711 you proceed to type another non-modifier key before you let go of Alt | |
712 and Shift, the Alt and Shift act as modifiers in the usual way. | |
713 | |
714 * `tparam' reported as a multiply-defined symbol when linking with ncurses. | |
715 | |
716 This problem results from an incompatible change in ncurses, in | |
717 version 1.9.9e approximately. This version is unable to provide a | |
718 definition of tparm without also defining tparam. This is also | |
719 incompatible with Terminfo; as a result, the Emacs Terminfo support | |
720 does not work with this version of ncurses. | |
721 | |
722 The fix is to install a newer version of ncurses, such as version 4.2. | |
723 | |
724 * Strange results from format %d in a few cases, on a Sun. | |
725 | |
726 Sun compiler version SC3.0 has been found to miscompile part of | |
727 editfns.c. The workaround is to compile with some other compiler such | |
728 as GCC. | |
729 | |
730 * Output from subprocess (such as man or diff) is randomly truncated | |
731 on GNU/Linux systems. | |
732 | |
733 This is due to a kernel bug which seems to be fixed in Linux version | |
734 1.3.75. | |
735 | |
736 * Error messages `internal facep []' happen on GNU/Linux systems. | |
737 | |
738 There is a report that replacing libc.so.5.0.9 with libc.so.5.2.16 | |
739 caused this to start happening. People are not sure why, but the | |
740 problem seems unlikely to be in Emacs itself. Some suspect that it | |
741 is actually Xlib which won't work with libc.so.5.2.16. | |
742 | |
743 Using the old library version is a workaround. | |
744 | |
745 * On Solaris, Emacs crashes if you use (display-time). | |
746 | |
747 This can happen if you configure Emacs without specifying the precise | |
748 version of Solaris that you are using. | |
749 | |
750 * Emacs dumps core on startup, on Solaris. | |
751 | |
752 Bill Sebok says that the cause of this is Solaris 2.4 vendor patch | |
753 102303-05, which extends the Solaris linker to deal with the Solaris | |
754 Common Desktop Environment's linking needs. You can fix the problem | |
755 by removing this patch and installing patch 102049-02 instead. | |
756 However, that linker version won't work with CDE. | |
757 | |
758 Solaris 2.5 comes with a linker that has this bug. It is reported that if | |
759 you install all the latest patches (as of June 1996), the bug is fixed. | |
760 We suspect the crucial patch is one of these, but we don't know | |
761 for certain. | |
762 | |
763 103093-03: [README] SunOS 5.5: kernel patch (2140557 bytes) | |
764 102832-01: [README] OpenWindows 3.5: Xview Jumbo Patch (4181613 bytes) | |
765 103242-04: [README] SunOS 5.5: linker patch (595363 bytes) | |
766 | |
767 (One user reports that the bug was fixed by those patches together | |
768 with patches 102980-04, 103279-01, 103300-02, and 103468-01.) | |
769 | |
770 If you can determine which patch does fix the bug, please tell | |
771 bug-gnu-emacs@gnu.org. | |
772 | |
773 Meanwhile, the GNU linker links Emacs properly on both Solaris 2.4 and | |
774 Solaris 2.5. | |
775 | |
776 * Emacs dumps core if lisp-complete-symbol is called, on Solaris. | |
777 | |
778 If you compile Emacs with the -fast or -xO4 option with version 3.0.2 | |
779 of the Sun C compiler, Emacs dumps core when lisp-complete-symbol is | |
780 called. The problem does not happen if you compile with GCC. | |
781 | |
782 * "Cannot find callback list" messages from dialog boxes on HPUX, in | |
783 Emacs built with Motif. | |
784 | |
785 This problem resulted from a bug in GCC 2.4.5. Newer GCC versions | |
786 such as 2.7.0 fix the problem. | |
787 | |
788 * On Irix 6.0, make tries (and fails) to build a program named unexelfsgi | |
789 | |
790 A compiler bug inserts spaces into the string "unexelfsgi . o" | |
791 in src/Makefile. Edit src/Makefile, after configure is run, | |
792 find that string, and take out the spaces. | |
793 | |
794 Compiler fixes in Irix 6.0.1 should eliminate this problem. | |
795 | |
796 * "out of virtual swap space" on Irix 5.3 | |
797 | |
798 This message occurs when the system runs out of swap space due to too | |
799 many large programs running. The solution is either to provide more | |
800 swap space or to reduce the number of large programs being run. You | |
801 can check the current status of the swap space by executing the | |
802 command `swap -l'. | |
803 | |
804 You can increase swap space by changing the file /etc/fstab. Adding a | |
805 line like this: | |
806 | |
807 /usr/swap/swap.more swap swap pri=3 0 0 | |
808 | |
809 where /usr/swap/swap.more is a file previously created (for instance | |
810 by using /etc/mkfile), will increase the swap space by the size of | |
811 that file. Execute `swap -m' or reboot the machine to activate the | |
812 new swap area. See the manpages for `swap' and `fstab' for further | |
813 information. | |
814 | |
815 The objectserver daemon can use up lots of memory because it can be | |
816 swamped with NIS information. It collects information about all users | |
817 on the network that can log on to the host. | |
818 | |
819 If you want to disable the objectserver completely, you can execute | |
820 the command `chkconfig objectserver off' and reboot. That may disable | |
821 some of the window system functionality, such as responding CDROM | |
822 icons. | |
823 | |
824 You can also remove NIS support from the objectserver. The SGI `admin' | |
825 FAQ has a detailed description on how to do that; see question 35 | |
826 ("Why isn't the objectserver working?"). The admin FAQ can be found at | |
827 ftp://viz.tamu.edu/pub/sgi/faq/. | |
828 | |
829 * With certain fonts, when the cursor appears on a character, the | |
830 character doesn't appear--you get a solid box instead. | |
831 | |
832 One user on a Linux-based GNU system reported that this problem went | |
833 away with installation of a new X server. The failing server was | |
834 XFree86 3.1.1. XFree86 3.1.2 works. | |
835 | |
836 * On SunOS 4.1.3, Emacs unpredictably crashes in _yp_dobind_soft. | |
837 | |
838 This happens if you configure Emacs specifying just `sparc-sun-sunos4' | |
839 on a system that is version 4.1.3. You must specify the precise | |
840 version number (or let configure figure out the configuration, which | |
841 it can do perfectly well for SunOS). | |
842 | |
843 * On SunOS 4, Emacs processes keep going after you kill the X server | |
844 (or log out, if you logged in using X). | |
845 | |
846 Someone reported that recompiling with GCC 2.7.0 fixed this problem. | |
847 | |
848 * On AIX 4, some programs fail when run in a Shell buffer | |
849 with an error message like No terminfo entry for "unknown". | |
850 | |
851 On AIX, many terminal type definitions are not installed by default. | |
852 `unknown' is one of them. Install the "Special Generic Terminal | |
853 Definitions" to make them defined. | |
854 | |
855 * On SunOS, you get linker errors | |
856 ld: Undefined symbol | |
857 _get_wmShellWidgetClass | |
858 _get_applicationShellWidgetClass | |
859 | |
860 The fix to this is to install patch 100573 for OpenWindows 3.0 | |
861 or link libXmu statically. | |
862 | |
863 * On AIX 4.1.2, linker error messages such as | |
864 ld: 0711-212 SEVERE ERROR: Symbol .__quous, found in the global symbol table | |
865 of archive /usr/lib/libIM.a, was not defined in archive member shr.o. | |
866 | |
867 This is a problem in libIM.a. You can work around it by executing | |
868 these shell commands in the src subdirectory of the directory where | |
869 you build Emacs: | |
870 | |
871 cp /usr/lib/libIM.a . | |
872 chmod 664 libIM.a | |
873 ranlib libIM.a | |
874 | |
875 Then change -lIM to ./libIM.a in the command to link temacs (in | |
876 Makefile). | |
877 | |
878 * Unpredictable segmentation faults on Solaris 2.3 and 2.4. | |
879 | |
880 A user reported that this happened in 19.29 when it was compiled with | |
881 the Sun compiler, but not when he recompiled with GCC 2.7.0. | |
882 | |
883 We do not know whether something in Emacs is partly to blame for this. | |
884 | |
885 * Emacs exits with "X protocol error" when run with an X server for | |
886 Windows. | |
887 | |
888 A certain X server for Windows had a bug which caused this. | |
889 Supposedly the newer 32-bit version of this server doesn't have the | |
890 problem. | |
891 | |
892 * Emacs crashes at startup on MSDOS. | |
893 | |
894 Some users report that Emacs 19.29 requires dpmi memory management, | |
895 and crashes on startup if the system does not have it. We don't yet | |
896 know why this happens--perhaps these machines don't have enough real | |
897 memory, or perhaps something is wrong in Emacs or the compiler. | |
898 However, arranging to use dpmi support is a workaround. | |
899 | |
900 You can find out if you have a dpmi host by running go32 without | |
901 arguments; it will tell you if it uses dpmi memory. For more | |
902 information about dpmi memory, consult the djgpp FAQ. (djgpp | |
903 is the GNU C compiler as packaged for MSDOS.) | |
904 | |
905 Compiling Emacs under MSDOS is extremely sensitive for proper memory | |
906 configuration. If you experience problems during compilation, consider | |
907 removing some or all memory resident programs (notably disk caches) | |
908 and make sure that your memory managers are properly configured. See | |
909 the djgpp faq for configuration hints. | |
910 | |
911 * A position you specified in .Xdefaults is ignored, using twm. | |
912 | |
913 twm normally ignores "program-specified" positions. | |
914 You can tell it to obey them with this command in your `.twmrc' file: | |
915 | |
916 UsePPosition "on" #allow clients to request a position | |
917 | |
918 * Compiling lib-src says there is no rule to make test-distrib.c. | |
919 | |
920 This results from a bug in a VERY old version of GNU Sed. To solve | |
921 the problem, install the current version of GNU Sed, then rerun | |
922 Emacs's configure script. | |
923 | |
924 * Compiling wakeup, in lib-src, says it can't make wakeup.c. | |
925 | |
926 This results from a bug in GNU Sed version 2.03. To solve the | |
927 problem, install the current version of GNU Sed, then rerun Emacs's | |
928 configure script. | |
929 | |
930 * On Sunos 4.1.1, there are errors compiling sysdep.c. | |
931 | |
932 If you get errors such as | |
933 | |
934 "sysdep.c", line 2017: undefined structure or union | |
935 "sysdep.c", line 2017: undefined structure or union | |
936 "sysdep.c", line 2019: nodename undefined | |
937 | |
938 This can result from defining LD_LIBRARY_PATH. It is very tricky | |
939 to use that environment variable with Emacs. The Emacs configure | |
940 script links many test programs with the system libraries; you must | |
941 make sure that the libraries available to configure are the same | |
942 ones available when you build Emacs. | |
943 | |
944 * The right Alt key works wrong on German HP keyboards (and perhaps | |
945 other non-English HP keyboards too). | |
946 | |
947 This is because HPUX defines the modifiers wrong in X. Here is a | |
948 shell script to fix the problem; be sure that it is run after VUE | |
949 configures the X server. | |
950 | |
951 xmodmap 2> /dev/null - << EOF | |
952 keysym Alt_L = Meta_L | |
953 keysym Alt_R = Meta_R | |
954 EOF | |
955 | |
956 xmodmap - << EOF | |
957 clear mod1 | |
958 keysym Mode_switch = NoSymbol | |
959 add mod1 = Meta_L | |
960 keysym Meta_R = Mode_switch | |
961 add mod2 = Mode_switch | |
962 EOF | |
963 | |
964 * The Emacs window disappears when you type M-q. | |
965 | |
966 Some versions of the Open Look window manager interpret M-q as a quit | |
967 command for whatever window you are typing at. If you want to use | |
968 Emacs with that window manager, you should try to configure the window | |
969 manager to use some other command. You can disable the | |
970 shortcut keys entirely by adding this line to ~/.OWdefaults: | |
971 | |
972 OpenWindows.WindowMenuAccelerators: False | |
973 | |
974 * Emacs does not notice when you release the mouse. | |
975 | |
976 There are reports that this happened with (some) Microsoft mice and | |
977 that replacing the mouse made it stop. | |
978 | |
979 * Trouble using ptys on IRIX, or running out of ptys. | |
980 | |
981 The program mkpts (which may be in `/usr/adm' or `/usr/sbin') needs to | |
982 be set-UID to root, or non-root programs like Emacs will not be able | |
983 to allocate ptys reliably. | |
984 | |
985 * On Irix 5.2, unexelfsgi.c can't find cmplrs/stsupport.h. | |
986 | |
987 The file cmplrs/stsupport.h was included in the wrong file set in the | |
988 Irix 5.2 distribution. You can find it in the optional fileset | |
989 compiler_dev, or copy it from some other Irix 5.2 system. A kludgy | |
990 workaround is to change unexelfsgi.c to include sym.h instead of | |
991 syms.h. | |
992 | |
993 * Slow startup on Linux-based GNU systems. | |
994 | |
995 People using systems based on the Linux kernel sometimes report that | |
996 startup takes 10 to 15 seconds longer than `usual'. | |
997 | |
998 This is because Emacs looks up the host name when it starts. | |
999 Normally, this takes negligible time; the extra delay is due to | |
1000 improper system configuration. This problem can occur for both | |
1001 networked and non-networked machines. | |
1002 | |
1003 Here is how to fix the configuration. It requires being root. | |
1004 | |
1005 ** Networked Case | |
1006 | |
1007 First, make sure the files `/etc/hosts' and `/etc/host.conf' both | |
1008 exist. The first line in the `/etc/hosts' file should look like this | |
1009 (replace HOSTNAME with your host name): | |
1010 | |
1011 127.0.0.1 HOSTNAME | |
1012 | |
1013 Also make sure that the `/etc/host.conf' files contains the following | |
1014 lines: | |
1015 | |
1016 order hosts, bind | |
1017 multi on | |
1018 | |
1019 Any changes, permanent and temporary, to the host name should be | |
1020 indicated in the `/etc/hosts' file, since it acts a limited local | |
1021 database of addresses and names (e.g., some SLIP connections | |
1022 dynamically allocate ip addresses). | |
1023 | |
1024 ** Non-Networked Case | |
1025 | |
1026 The solution described in the networked case applies here as well. | |
1027 However, if you never intend to network your machine, you can use a | |
1028 simpler solution: create an empty `/etc/host.conf' file. The command | |
1029 `touch /etc/host.conf' suffices to create the file. The `/etc/hosts' | |
1030 file is not necessary with this approach. | |
1031 | |
1032 * On Solaris 2.4, Dired hangs and C-g does not work. Or Emacs hangs | |
1033 forever waiting for termination of a subprocess that is a zombie. | |
1034 | |
1035 casper@fwi.uva.nl says the problem is in X11R6. Rebuild libX11.so | |
1036 after changing the file xc/config/cf/sunLib.tmpl. Change the lines | |
1037 | |
1038 #if ThreadedX | |
1039 #define SharedX11Reqs -lthread | |
1040 #endif | |
1041 | |
1042 to: | |
1043 | |
1044 #if OSMinorVersion < 4 | |
1045 #if ThreadedX | |
1046 #define SharedX11Reqs -lthread | |
1047 #endif | |
1048 #endif | |
1049 | |
1050 Be sure also to edit x/config/cf/sun.cf so that OSMinorVersion is 4 | |
1051 (as it should be for Solaris 2.4). The file has three definitions for | |
1052 OSMinorVersion: the first is for x86, the second for SPARC under | |
1053 Solaris, and the third for SunOS 4. Make sure to update the | |
1054 definition for your type of machine and system. | |
1055 | |
1056 Then do `make Everything' in the top directory of X11R6, to rebuild | |
1057 the makefiles and rebuild X. The X built this way work only on | |
1058 Solaris 2.4, not on 2.3. | |
1059 | |
1060 For multithreaded X to work it is necessary to install patch | |
1061 101925-02 to fix problems in header files [2.4]. You need | |
1062 to reinstall gcc or re-run just-fixinc after installing that | |
1063 patch. | |
1064 | |
1065 However, Frank Rust <frust@iti.cs.tu-bs.de> used a simpler solution: | |
1066 he changed | |
1067 #define ThreadedX YES | |
1068 to | |
1069 #define ThreadedX NO | |
1070 in sun.cf and did `make World' to rebuild X11R6. Removing all | |
1071 `-DXTHREAD*' flags and `-lthread' entries from lib/X11/Makefile and | |
1072 typing 'make install' in that directory also seemed to work. | |
1073 | |
1074 * With M-x enable-flow-control, you need to type C-\ twice | |
1075 to do incremental search--a single C-\ gets no response. | |
1076 | |
1077 This has been traced to communicating with your machine via kermit, | |
1078 with C-\ as the kermit escape character. One solution is to use | |
1079 another escape character in kermit. One user did | |
1080 | |
1081 set escape-character 17 | |
1082 | |
1083 in his .kermrc file, to make C-q the kermit escape character. | |
1084 | |
1085 * The Motif version of Emacs paints the screen a solid color. | |
1086 | |
1087 This has been observed to result from the following X resource: | |
1088 | |
1089 Emacs*default.attributeFont: -*-courier-medium-r-*-*-*-140-*-*-*-*-iso8859-* | |
1090 | |
1091 That the resource has this effect indicates a bug in something, but we | |
1092 do not yet know what. If it is an Emacs bug, we hope someone can | |
1093 explain what the bug is so we can fix it. In the mean time, removing | |
1094 the resource prevents the problem. | |
1095 | |
1096 * Emacs gets hung shortly after startup, on Sunos 4.1.3. | |
1097 | |
1098 We think this is due to a bug in Sunos. The word is that | |
1099 one of these Sunos patches fixes the bug: | |
1100 | |
1101 100075-11 100224-06 100347-03 100482-05 100557-02 100623-03 100804-03 101080-01 | |
1102 100103-12 100249-09 100496-02 100564-07 100630-02 100891-10 101134-01 | |
1103 100170-09 100296-04 100377-09 100507-04 100567-04 100650-02 101070-01 101145-01 | |
1104 100173-10 100305-15 100383-06 100513-04 100570-05 100689-01 101071-03 101200-02 | |
1105 100178-09 100338-05 100421-03 100536-02 100584-05 100784-01 101072-01 101207-01 | |
1106 | |
1107 We don't know which of these patches really matter. If you find out | |
1108 which ones, please inform bug-gnu-emacs@gnu.org. | |
1109 | |
1110 * Emacs aborts while starting up, only when run without X. | |
1111 | |
1112 This problem often results from compiling Emacs with GCC when GCC was | |
1113 installed incorrectly. The usual error in installing GCC is to | |
1114 specify --includedir=/usr/include. Installation of GCC makes | |
1115 corrected copies of the system header files. GCC is supposed to use | |
1116 the corrected copies in preference to the original system headers. | |
1117 Specifying --includedir=/usr/include causes the original system header | |
1118 files to be used. On some systems, the definition of ioctl in the | |
1119 original system header files is invalid for ANSI C and causes Emacs | |
1120 not to work. | |
1121 | |
1122 The fix is to reinstall GCC, and this time do not specify --includedir | |
1123 when you configure it. Then recompile Emacs. Specifying --includedir | |
1124 is appropriate only in very special cases and it should *never* be the | |
1125 same directory where system header files are kept. | |
1126 | |
1127 * On Solaris 2.x, GCC complains "64 bit integer types not supported" | |
1128 | |
1129 This suggests that GCC is not installed correctly. Most likely you | |
1130 are using GCC 2.7.2.3 (or earlier) on Solaris 2.6 (or later); this | |
1131 does not work without patching. To run GCC 2.7.2.3 on Solaris 2.6 or | |
1132 later, you must patch fixinc.svr4 and reinstall GCC from scratch as | |
1133 described in the Solaris FAQ | |
1134 <http://www.wins.uva.nl/pub/solaris/solaris2.html>. A better fix is | |
1135 to upgrade to GCC 2.8.1 or later. | |
1136 | |
1137 * The Compose key on a DEC keyboard does not work as Meta key. | |
1138 | |
1139 This shell command should fix it: | |
1140 | |
1141 xmodmap -e 'keycode 0xb1 = Meta_L' | |
1142 | |
1143 * Regular expressions matching bugs on SCO systems. | |
1144 | |
1145 On SCO, there are problems in regexp matching when Emacs is compiled | |
1146 with the system compiler. The compiler version is "Microsoft C | |
1147 version 6", SCO 4.2.0h Dev Sys Maintenance Supplement 01/06/93; Quick | |
1148 C Compiler Version 1.00.46 (Beta). The solution is to compile with | |
1149 GCC. | |
1150 | |
1151 * On Sunos 4, you get the error ld: Undefined symbol __lib_version. | |
1152 | |
1153 This is the result of using cc or gcc with the shared library meant | |
1154 for acc (the Sunpro compiler). Check your LD_LIBRARY_PATH and delete | |
1155 /usr/lang/SC2.0.1 or some similar directory. | |
1156 | |
1157 * You can't select from submenus (in the X toolkit version). | |
1158 | |
1159 On certain systems, mouse-tracking and selection in top-level menus | |
1160 works properly with the X toolkit, but neither of them works when you | |
1161 bring up a submenu (such as Bookmarks or Compare or Apply Patch, in | |
1162 the Files menu). | |
1163 | |
1164 This works on most systems. There is speculation that the failure is | |
1165 due to bugs in old versions of X toolkit libraries, but no one really | |
1166 knows. If someone debugs this and finds the precise cause, perhaps a | |
1167 workaround can be found. | |
1168 | |
1169 * Unusable default font on SCO 3.2v4. | |
1170 | |
1171 The Open Desktop environment comes with default X resource settings | |
1172 that tell Emacs to use a variable-width font. Emacs cannot use such | |
1173 fonts, so it does not work. | |
1174 | |
1175 This is caused by the file /usr/lib/X11/app-defaults/ScoTerm, which is | |
1176 the application-specific resource file for the `scoterm' terminal | |
1177 emulator program. It contains several extremely general X resources | |
1178 that affect other programs besides `scoterm'. In particular, these | |
1179 resources affect Emacs also: | |
1180 | |
1181 *Font: -*-helvetica-medium-r-*--12-*-p-* | |
1182 *Background: scoBackground | |
1183 *Foreground: scoForeground | |
1184 | |
1185 The best solution is to create an application-specific resource file for | |
1186 Emacs, /usr/lib/X11/sco/startup/Emacs, with the following contents: | |
1187 | |
1188 Emacs*Font: -*-courier-medium-r-*-*-*-120-*-*-*-*-iso8859-1 | |
1189 Emacs*Background: white | |
1190 Emacs*Foreground: black | |
1191 | |
1192 (These settings mimic the Emacs defaults, but you can change them to | |
1193 suit your needs.) This resource file is only read when the X server | |
1194 starts up, so you should restart it by logging out of the Open Desktop | |
1195 environment or by running `scologin stop; scologin start` from the shell | |
1196 as root. Alternatively, you can put these settings in the | |
1197 /usr/lib/X11/app-defaults/Emacs resource file and simply restart Emacs, | |
1198 but then they will not affect remote invocations of Emacs that use the | |
1199 Open Desktop display. | |
1200 | |
1201 These resource files are not normally shared across a network of SCO | |
1202 machines; you must create the file on each machine individually. | |
1203 | |
1204 * rcs2log gives you the awk error message "too many fields". | |
1205 | |
1206 This is due to an arbitrary limit in certain versions of awk. | |
1207 The solution is to use gawk (GNU awk). | |
1208 | |
1209 * Emacs is slow using X11R5 on HP/UX. | |
1210 | |
1211 This happens if you use the MIT versions of the X libraries--it | |
1212 doesn't run as fast as HP's version. People sometimes use the version | |
1213 because they see the HP version doesn't have the libraries libXaw.a, | |
1214 libXmu.a, libXext.a and others. HP/UX normally doesn't come with | |
1215 those libraries installed. To get good performance, you need to | |
1216 install them and rebuild Emacs. | |
1217 | |
1218 * Loading fonts is very slow. | |
1219 | |
1220 You might be getting scalable fonts instead of precomputed bitmaps. | |
1221 Known scalable font directories are "Type1" and "Speedo". A font | |
1222 directory contains scalable fonts if it contains the file | |
1223 "fonts.scale". | |
1224 | |
1225 If this is so, re-order your X windows font path to put the scalable | |
1226 font directories last. See the documentation of `xset' for details. | |
1227 | |
1228 With some X servers, it may be necessary to take the scalable font | |
1229 directories out of your path entirely, at least for Emacs 19.26. | |
1230 Changes in the future may make this unnecessary. | |
1231 | |
1232 * On AIX 3.2.4, releasing Ctrl/Act key has no effect, if Shift is down. | |
1233 | |
1234 Due to a feature of AIX, pressing or releasing the Ctrl/Act key is | |
1235 ignored when the Shift, Alt or AltGr keys are held down. This can | |
1236 lead to the keyboard being "control-locked"--ordinary letters are | |
1237 treated as control characters. | |
1238 | |
1239 You can get out of this "control-locked" state by pressing and | |
1240 releasing Ctrl/Act while not pressing or holding any other keys. | |
1241 | |
1242 * display-time causes kernel problems on ISC systems. | |
1243 | |
1244 Under Interactive Unix versions 3.0.1 and 4.0 (and probably other | |
1245 versions), display-time causes the loss of large numbers of STREVENT | |
1246 cells. Eventually the kernel's supply of these cells is exhausted. | |
1247 This makes emacs and the whole system run slow, and can make other | |
1248 processes die, in particular pcnfsd. | |
1249 | |
1250 Other emacs functions that communicate with remote processes may have | |
1251 the same problem. Display-time seems to be far the worst. | |
1252 | |
1253 The only known fix: Don't run display-time. | |
1254 | |
1255 * On Solaris, C-x doesn't get through to Emacs when you use the console. | |
1256 | |
1257 This is a Solaris feature (at least on Intel x86 cpus). Type C-r | |
1258 C-r C-t, to toggle whether C-x gets through to Emacs. | |
1259 | |
1260 * Error message `Symbol's value as variable is void: x', followed by | |
1261 segmentation fault and core dump. | |
1262 | |
1263 This has been tracked to a bug in tar! People report that tar erroneously | |
1264 added a line like this at the beginning of files of Lisp code: | |
1265 | |
1266 x FILENAME, N bytes, B tape blocks | |
1267 | |
1268 If your tar has this problem, install GNU tar--if you can manage to | |
1269 untar it :-). | |
1270 | |
1271 * Link failure when using acc on a Sun. | |
1272 | |
1273 To use acc, you need additional options just before the libraries, such as | |
1274 | |
1275 /usr/lang/SC2.0.1/values-Xt.o -L/usr/lang/SC2.0.1/cg87 -L/usr/lang/SC2.0.1 | |
1276 | |
1277 and you need to add -lansi just before -lc. | |
1278 | |
1279 The precise file names depend on the compiler version, so we | |
1280 cannot easily arrange to supply them. | |
1281 | |
1282 * Link failure on IBM AIX 1.3 ptf 0013. | |
1283 | |
1284 There is a real duplicate definition of the function `_slibc_free' in | |
1285 the library /lib/libc_s.a (just do nm on it to verify). The | |
1286 workaround/fix is: | |
1287 | |
1288 cd /lib | |
1289 ar xv libc_s.a NLtmtime.o | |
1290 ar dv libc_s.a NLtmtime.o | |
1291 | |
1292 * Undefined symbols _dlopen, _dlsym and/or _dlclose on a Sun. | |
1293 | |
1294 If you see undefined symbols _dlopen, _dlsym, or _dlclose when linking | |
1295 with -lX11, compile and link against the file mit/util/misc/dlsym.c in | |
1296 the MIT X11R5 distribution. Alternatively, link temacs using shared | |
1297 libraries with s/sunos4shr.h. (This doesn't work if you use the X | |
1298 toolkit.) | |
1299 | |
1300 If you get the additional error that the linker could not find | |
1301 lib_version.o, try extracting it from X11/usr/lib/X11/libvim.a in | |
1302 X11R4, then use it in the link. | |
1303 | |
1304 * Error messages `Wrong number of arguments: #<subr where-is-internal>, 5' | |
1305 | |
1306 This typically results from having the powerkey library loaded. | |
1307 Powerkey was designed for Emacs 19.22. It is obsolete now because | |
1308 Emacs 19 now has this feature built in; and powerkey also calls | |
1309 where-is-internal in an obsolete way. | |
1310 | |
1311 So the fix is to arrange not to load powerkey. | |
1312 | |
1313 * In Shell mode, you get a ^M at the end of every line. | |
1314 | |
1315 This happens to people who use tcsh, because it is trying to be too | |
1316 smart. It sees that the Shell uses terminal type `unknown' and turns | |
1317 on the flag to output ^M at the end of each line. You can fix the | |
1318 problem by adding this to your .cshrc file: | |
1319 | |
1320 if ($?EMACS) then | |
1321 if ($EMACS == "t") then | |
1322 unset edit | |
1323 stty -icrnl -onlcr -echo susp ^Z | |
1324 endif | |
1325 endif | |
1326 | |
1327 * An error message such as `X protocol error: BadMatch (invalid | |
1328 parameter attributes) on protocol request 93'. | |
1329 | |
1330 This comes from having an invalid X resource, such as | |
1331 emacs*Cursor: black | |
1332 (which is invalid because it specifies a color name for something | |
1333 that isn't a color.) | |
1334 | |
1335 The fix is to correct your X resources. | |
1336 | |
1337 * Undefined symbols when linking on Sunos 4.1 using --with-x-toolkit. | |
1338 | |
1339 If you get the undefined symbols _atowc _wcslen, _iswprint, _iswspace, | |
1340 _iswcntrl, _wcscpy, and _wcsncpy, then you need to add -lXwchar after | |
1341 -lXaw in the command that links temacs. | |
1342 | |
1343 This problem seems to arise only when the international language | |
1344 extensions to X11R5 are installed. | |
1345 | |
1346 * Typing C-c C-c in Shell mode kills your X server. | |
1347 | |
1348 This happens with Linux kernel 1.0 thru 1.04, approximately. The workaround is | |
1349 to define SIGNALS_VIA_CHARACTERS in config.h and recompile Emacs. | |
1350 Newer Linux kernel versions don't have this problem. | |
1351 | |
1352 * src/Makefile and lib-src/Makefile are truncated--most of the file missing. | |
1353 | |
1354 This can happen if configure uses GNU sed version 2.03. That version | |
1355 had a bug. GNU sed version 2.05 works properly. | |
1356 | |
1357 * Slow startup on X11R6 with X windows. | |
1358 | |
1359 If Emacs takes two minutes to start up on X11R6, see if your X | |
1360 resources specify any Adobe fonts. That causes the type-1 font | |
1361 renderer to start up, even if the font you asked for is not a type-1 | |
1362 font. | |
1363 | |
1364 One way to avoid this problem is to eliminate the type-1 fonts from | |
1365 your font path, like this: | |
1366 | |
1367 xset -fp /usr/X11R6/lib/X11/fonts/Type1/ | |
1368 | |
1369 * Pull-down menus appear in the wrong place, in the toolkit version of Emacs. | |
1370 | |
1371 An X resource of this form can cause the problem: | |
1372 | |
1373 Emacs*geometry: 80x55+0+0 | |
1374 | |
1375 This resource is supposed to apply, and does apply, to the menus | |
1376 individually as well as to Emacs frames. If that is not what you | |
1377 want, rewrite the resource. | |
1378 | |
1379 To check thoroughly for such resource specifications, use `xrdb | |
1380 -query' to see what resources the X server records, and also look at | |
1381 the user's ~/.Xdefaults and ~/.Xdefaults-* files. | |
1382 | |
1383 * --with-x-toolkit version crashes when used with shared libraries. | |
1384 | |
1385 On some systems, including Sunos 4 and DGUX 5.4.2 and perhaps others, | |
1386 unexec doesn't work properly with the shared library for the X | |
1387 toolkit. You might be able to work around this by using a nonshared | |
1388 libXt.a library. The real fix is to upgrade the various versions of | |
1389 unexec and/or ralloc. We think this has been fixed on Sunos 4 | |
1390 and Solaris in version 19.29. | |
1391 | |
1392 * `make install' fails on install-doc with `Error 141'. | |
1393 | |
1394 This happens on Ultrix 4.2 due to failure of a pipeline of tar | |
1395 commands. We don't know why they fail, but the bug seems not to be in | |
1396 Emacs. The workaround is to run the shell command in install-doc by | |
1397 hand. | |
1398 | |
1399 * --with-x-toolkit option configures wrong on BSD/386. | |
1400 | |
1401 This problem is due to bugs in the shell in version 1.0 of BSD/386. | |
1402 The workaround is to edit the configure file to use some other shell, | |
1403 such as bash. | |
1404 | |
1405 * Subprocesses remain, hanging but not zombies, on Sunos 5.3. | |
1406 | |
1407 A bug in Sunos 5.3 causes Emacs subprocesses to remain after Emacs | |
1408 exits. Sun patch # 101415-02 is part of the fix for this, but it only | |
1409 applies to ptys, and doesn't fix the problem with subprocesses | |
1410 communicating through pipes. | |
1411 | |
1412 * Mail is lost when sent to local aliases. | |
1413 | |
1414 Many emacs mail user agents (VM and rmail, for instance) use the | |
1415 sendmail.el library. This library can arrange for mail to be | |
1416 delivered by passing messages to the /usr/lib/sendmail (usually) | |
1417 program . In doing so, it passes the '-t' flag to sendmail, which | |
1418 means that the name of the recipient of the message is not on the | |
1419 command line and, therefore, that sendmail must parse the message to | |
1420 obtain the destination address. | |
1421 | |
1422 There is a bug in the SunOS4.1.1 and SunOS4.1.3 versions of sendmail. | |
1423 In short, when given the -t flag, the SunOS sendmail won't recognize | |
1424 non-local (i.e. NIS) aliases. It has been reported that the Solaris | |
1425 2.x versions of sendmail do not have this bug. For those using SunOS | |
1426 4.1, the best fix is to install sendmail V8 or IDA sendmail (which | |
1427 have other advantages over the regular sendmail as well). At the time | |
1428 of this writing, these official versions are available: | |
1429 | |
1430 Sendmail V8 on ftp.cs.berkeley.edu in /ucb/sendmail: | |
1431 sendmail.8.6.9.base.tar.Z (the base system source & documentation) | |
1432 sendmail.8.6.9.cf.tar.Z (configuration files) | |
1433 sendmail.8.6.9.misc.tar.Z (miscellaneous support programs) | |
1434 sendmail.8.6.9.xdoc.tar.Z (extended documentation, with postscript) | |
1435 | |
1436 IDA sendmail on vixen.cso.uiuc.edu in /pub: | |
1437 sendmail-5.67b+IDA-1.5.tar.gz | |
1438 | |
1439 * On AIX, you get this message when running Emacs: | |
1440 | |
1441 Could not load program emacs | |
1442 Symbol smtcheckinit in csh is undefined | |
1443 Error was: Exec format error | |
1444 | |
1445 or this one: | |
1446 | |
1447 Could not load program .emacs | |
1448 Symbol _system_con in csh is undefined | |
1449 Symbol _fp_trapsta in csh is undefined | |
1450 Error was: Exec format error | |
1451 | |
1452 These can happen when you try to run on AIX 3.2.5 a program that was | |
1453 compiled with 3.2.4. The fix is to recompile. | |
1454 | |
1455 * On AIX, you get this compiler error message: | |
1456 | |
1457 Processing include file ./XMenuInt.h | |
1458 1501-106: (S) Include file X11/Xlib.h not found. | |
1459 | |
1460 This means your system was installed with only the X11 runtime i.d | |
1461 libraries. You have to find your sipo (bootable tape) and install | |
1462 X11Dev... with smit. | |
1463 | |
1464 * You "lose characters" after typing Compose Character key. | |
1465 | |
1466 This is because the Compose Character key is defined as the keysym | |
1467 Multi_key, and Emacs (seeing that) does the proper X11 | |
1468 character-composition processing. If you don't want your Compose key | |
1469 to do that, you can redefine it with xmodmap. | |
1470 | |
1471 For example, here's one way to turn it into a Meta key: | |
1472 | |
1473 xmodmap -e "keysym Multi_key = Meta_L" | |
1474 | |
1475 If all users at your site of a particular keyboard prefer Meta to | |
1476 Compose, you can make the remapping happen automatically by adding the | |
1477 xmodmap command to the xdm setup script for that display. | |
1478 | |
1479 * C-z just refreshes the screen instead of suspending Emacs. | |
1480 | |
1481 You are probably using a shell that doesn't support job control, even | |
1482 though the system itself is capable of it. Either use a different shell, | |
1483 or set the variable `cannot-suspend' to a non-nil value. | |
1484 | |
1485 * Watch out for .emacs files and EMACSLOADPATH environment vars | |
1486 | |
1487 These control the actions of Emacs. | |
1488 ~/.emacs is your Emacs init file. | |
1489 EMACSLOADPATH overrides which directories the function | |
1490 "load" will search. | |
1491 | |
1492 If you observe strange problems, check for these and get rid | |
1493 of them, then try again. | |
1494 | |
1495 * After running emacs once, subsequent invocations crash. | |
1496 | |
1497 Some versions of SVR4 have a serious bug in the implementation of the | |
1498 mmap () system call in the kernel; this causes emacs to run correctly | |
1499 the first time, and then crash when run a second time. | |
1500 | |
1501 Contact your vendor and ask for the mmap bug fix; in the mean time, | |
1502 you may be able to work around the problem by adding a line to your | |
1503 operating system description file (whose name is reported by the | |
1504 configure script) that reads: | |
1505 #define SYSTEM_MALLOC | |
1506 This makes Emacs use memory less efficiently, but seems to work around | |
1507 the kernel bug. | |
1508 | |
1509 * Inability to send an Alt-modified key, when Emacs is communicating | |
1510 directly with an X server. | |
1511 | |
1512 If you have tried to bind an Alt-modified key as a command, and it | |
1513 does not work to type the command, the first thing you should check is | |
1514 whether the key is getting through to Emacs. To do this, type C-h c | |
1515 followed by the Alt-modified key. C-h c should say what kind of event | |
1516 it read. If it says it read an Alt-modified key, then make sure you | |
1517 have made the key binding correctly. | |
1518 | |
1519 If C-h c reports an event that doesn't have the Alt modifier, it may | |
1520 be because your X server has no key for the Alt modifier. The X | |
1521 server that comes from MIT does not set up the Alt modifier by | |
1522 default. | |
1523 | |
1524 If your keyboard has keys named Alt, you can enable them as follows: | |
1525 | |
1526 xmodmap -e 'add mod2 = Alt_L' | |
1527 xmodmap -e 'add mod2 = Alt_R' | |
1528 | |
1529 If the keyboard has just one key named Alt, then only one of those | |
1530 commands is needed. The modifier `mod2' is a reasonable choice if you | |
1531 are using an unmodified MIT version of X. Otherwise, choose any | |
1532 modifier bit not otherwise used. | |
1533 | |
1534 If your keyboard does not have keys named Alt, you can use some other | |
1535 keys. Use the keysym command in xmodmap to turn a function key (or | |
1536 some other 'spare' key) into Alt_L or into Alt_R, and then use the | |
1537 commands show above to make them modifier keys. | |
1538 | |
1539 Note that if you have Alt keys but no Meta keys, Emacs translates Alt | |
1540 into Meta. This is because of the great importance of Meta in Emacs. | |
1541 | |
1542 * `Pid xxx killed due to text modification or page I/O error' | |
1543 | |
1544 On HP/UX, you can get that error when the Emacs executable is on an NFS | |
1545 file system. HP/UX responds this way if it tries to swap in a page and | |
1546 does not get a response from the server within a timeout whose default | |
1547 value is just ten seconds. | |
1548 | |
1549 If this happens to you, extend the timeout period. | |
1550 | |
1551 * `expand-file-name' fails to work on any but the machine you dumped Emacs on. | |
1552 | |
1553 On Ultrix, if you use any of the functions which look up information | |
1554 in the passwd database before dumping Emacs (say, by using | |
1555 expand-file-name in site-init.el), then those functions will not work | |
1556 in the dumped Emacs on any host but the one Emacs was dumped on. | |
1557 | |
1558 The solution? Don't use expand-file-name in site-init.el, or in | |
1559 anything it loads. Yuck - some solution. | |
1560 | |
1561 I'm not sure why this happens; if you can find out exactly what is | |
1562 going on, and perhaps find a fix or a workaround, please let us know. | |
1563 Perhaps the YP functions cache some information, the cache is included | |
1564 in the dumped Emacs, and is then inaccurate on any other host. | |
1565 | |
1566 * On some variants of SVR4, Emacs does not work at all with X. | |
1567 | |
1568 Try defining BROKEN_FIONREAD in your config.h file. If this solves | |
1569 the problem, please send a bug report to tell us this is needed; be | |
1570 sure to say exactly what type of machine and system you are using. | |
1571 | |
1572 * Linking says that the functions insque and remque are undefined. | |
1573 | |
1574 Change oldXMenu/Makefile by adding insque.o to the variable OBJS. | |
1575 | |
1576 * Emacs fails to understand most Internet host names, even though | |
1577 the names work properly with other programs on the same system. | |
1578 * Emacs won't work with X-windows if the value of DISPLAY is HOSTNAME:0. | |
1579 * GNUs can't make contact with the specified host for nntp. | |
1580 | |
1581 This typically happens on Suns and other systems that use shared | |
1582 libraries. The cause is that the site has installed a version of the | |
1583 shared library which uses a name server--but has not installed a | |
1584 similar version of the unshared library which Emacs uses. | |
1585 | |
1586 The result is that most programs, using the shared library, work with | |
1587 the nameserver, but Emacs does not. | |
1588 | |
1589 The fix is to install an unshared library that corresponds to what you | |
1590 installed in the shared library, and then relink Emacs. | |
1591 | |
1592 On SunOS 4.1, simply define HAVE_RES_INIT. | |
1593 | |
1594 If you have already installed the name resolver in the file libresolv.a, | |
1595 then you need to compile Emacs to use that library. The easiest way to | |
1596 do this is to add to config.h a definition of LIBS_SYSTEM, LIBS_MACHINE | |
1597 or LIB_STANDARD which uses -lresolv. Watch out! If you redefine a macro | |
1598 that is already in use in your configuration to supply some other libraries, | |
1599 be careful not to lose the others. | |
1600 | |
1601 Thus, you could start by adding this to config.h: | |
1602 | |
1603 #define LIBS_SYSTEM -lresolv | |
1604 | |
1605 Then if this gives you an error for redefining a macro, and you see that | |
1606 the s- file defines LIBS_SYSTEM as -lfoo -lbar, you could change config.h | |
1607 again to say this: | |
1608 | |
1609 #define LIBS_SYSTEM -lresolv -lfoo -lbar | |
1610 | |
1611 * On a Sun running SunOS 4.1.1, you get this error message from GNU ld: | |
1612 | |
1613 /lib/libc.a(_Q_sub.o): Undefined symbol __Q_get_rp_rd referenced from text segment | |
1614 | |
1615 The problem is in the Sun shared C library, not in GNU ld. | |
1616 | |
1617 The solution is to install Patch-ID# 100267-03 from Sun. | |
1618 | |
1619 * Self documentation messages are garbled. | |
1620 | |
1621 This means that the file `etc/DOC-...' doesn't properly correspond | |
1622 with the Emacs executable. Redumping Emacs and then installing the | |
1623 corresponding pair of files should fix the problem. | |
1624 | |
1625 * Trouble using ptys on AIX. | |
1626 | |
1627 People often install the pty devices on AIX incorrectly. | |
1628 Use `smit pty' to reinstall them properly. | |
1629 | |
1630 * Shell mode on HP/UX gives the message, "`tty`: Ambiguous". | |
1631 | |
1632 christos@theory.tn.cornell.edu says: | |
1633 | |
1634 The problem is that in your .cshrc you have something that tries to | |
1635 execute `tty`. If you are not running the shell on a real tty then | |
1636 tty will print "not a tty". Csh expects one word in some places, | |
1637 but tty is giving it back 3. | |
1638 | |
1639 The solution is to add a pair of quotes around `tty` to make it a single | |
1640 word: | |
1641 | |
1642 if (`tty` == "/dev/console") | |
1643 | |
1644 should be changed to: | |
1645 | |
1646 if ("`tty`" == "/dev/console") | |
1647 | |
1648 Even better, move things that set up terminal sections out of .cshrc | |
1649 and into .login. | |
1650 | |
1651 * Using X Windows, control-shift-leftbutton makes Emacs hang. | |
1652 | |
1653 Use the shell command `xset bc' to make the old X Menu package work. | |
1654 | |
1655 * Emacs running under X Windows does not handle mouse clicks. | |
1656 * `emacs -geometry 80x20' finds a file named `80x20'. | |
1657 | |
1658 One cause of such problems is having (setq term-file-prefix nil) in | |
1659 your .emacs file. Another cause is a bad value of EMACSLOADPATH in | |
1660 the environment. | |
1661 | |
1662 * Emacs gets error message from linker on Sun. | |
1663 | |
1664 If the error message says that a symbol such as `f68881_used' or | |
1665 `ffpa_used' or `start_float' is undefined, this probably indicates | |
1666 that you have compiled some libraries, such as the X libraries, | |
1667 with a floating point option other than the default. | |
1668 | |
1669 It's not terribly hard to make this work with small changes in | |
1670 crt0.c together with linking with Fcrt1.o, Wcrt1.o or Mcrt1.o. | |
1671 However, the easiest approach is to build Xlib with the default | |
1672 floating point option: -fsoft. | |
1673 | |
1674 * Emacs fails to get default settings from X Windows server. | |
1675 | |
1676 The X library in X11R4 has a bug; it interchanges the 2nd and 3rd | |
1677 arguments to XGetDefaults. Define the macro XBACKWARDS in config.h to | |
1678 tell Emacs to compensate for this. | |
1679 | |
1680 I don't believe there is any way Emacs can determine for itself | |
1681 whether this problem is present on a given system. | |
1682 | |
1683 * Keyboard input gets confused after a beep when using a DECserver | |
1684 as a concentrator. | |
1685 | |
1686 This problem seems to be a matter of configuring the DECserver to use | |
1687 7 bit characters rather than 8 bit characters. | |
1688 | |
1689 * M-x shell persistently reports "Process shell exited abnormally with code 1". | |
1690 | |
1691 This happened on Suns as a result of what is said to be a bug in Sunos | |
1692 version 4.0.x. The only fix was to reboot the machine. | |
1693 | |
1694 * Programs running under terminal emulator do not recognize `emacs' | |
1695 terminal type. | |
1696 | |
1697 The cause of this is a shell startup file that sets the TERMCAP | |
1698 environment variable. The terminal emulator uses that variable to | |
1699 provide the information on the special terminal type that Emacs | |
1700 emulates. | |
1701 | |
1702 Rewrite your shell startup file so that it does not change TERMCAP | |
1703 in such a case. You could use the following conditional which sets | |
1704 it only if it is undefined. | |
1705 | |
1706 if ( ! ${?TERMCAP} ) setenv TERMCAP ~/my-termcap-file | |
1707 | |
1708 Or you could set TERMCAP only when you set TERM--which should not | |
1709 happen in a non-login shell. | |
1710 | |
1711 * X Windows doesn't work if DISPLAY uses a hostname. | |
1712 | |
1713 People have reported kernel bugs in certain systems that cause Emacs | |
1714 not to work with X Windows if DISPLAY is set using a host name. But | |
1715 the problem does not occur if DISPLAY is set to `unix:0.0'. I think | |
1716 the bug has to do with SIGIO or FIONREAD. | |
1717 | |
1718 You may be able to compensate for the bug by doing (set-input-mode nil nil). | |
1719 However, that has the disadvantage of turning off interrupts, so that | |
1720 you are unable to quit out of a Lisp program by typing C-g. | |
1721 | |
1722 The easy way to do this is to put | |
1723 | |
1724 (setq x-sigio-bug t) | |
1725 | |
1726 in your site-init.el file. | |
1727 | |
1728 * Problem with remote X server on Suns. | |
1729 | |
1730 On a Sun, running Emacs on one machine with the X server on another | |
1731 may not work if you have used the unshared system libraries. This | |
1732 is because the unshared libraries fail to use YP for host name lookup. | |
1733 As a result, the host name you specify may not be recognized. | |
1734 | |
1735 * Shell mode ignores interrupts on Apollo Domain | |
1736 | |
1737 You may find that M-x shell prints the following message: | |
1738 | |
1739 Warning: no access to tty; thus no job control in this shell... | |
1740 | |
1741 This can happen if there are not enough ptys on your system. | |
1742 Here is how to make more of them. | |
1743 | |
1744 % cd /dev | |
1745 % ls pty* | |
1746 # shows how many pty's you have. I had 8, named pty0 to pty7) | |
1747 % /etc/crpty 8 | |
1748 # creates eight new pty's | |
1749 | |
1750 * Fatal signal in the command temacs -l loadup inc dump | |
1751 | |
1752 This command is the final stage of building Emacs. It is run by the | |
1753 Makefile in the src subdirectory, or by build.com on VMS. | |
1754 | |
1755 It has been known to get fatal errors due to insufficient swapping | |
1756 space available on the machine. | |
1757 | |
1758 On 68000's, it has also happened because of bugs in the | |
1759 subroutine `alloca'. Verify that `alloca' works right, even | |
1760 for large blocks (many pages). | |
1761 | |
1762 * test-distrib says that the distribution has been clobbered | |
1763 * or, temacs prints "Command key out of range 0-127" | |
1764 * or, temacs runs and dumps emacs, but emacs totally fails to work. | |
1765 * or, temacs gets errors dumping emacs | |
1766 | |
1767 This can be because the .elc files have been garbled. Do not be | |
1768 fooled by the fact that most of a .elc file is text: these are | |
1769 binary files and can contain all 256 byte values. | |
1770 | |
1771 In particular `shar' cannot be used for transmitting GNU Emacs. | |
1772 It typically truncates "lines". What appear to be "lines" in | |
1773 a binary file can of course be of any length. Even once `shar' | |
1774 itself is made to work correctly, `sh' discards null characters | |
1775 when unpacking the shell archive. | |
1776 | |
1777 I have also seen character \177 changed into \377. I do not know | |
1778 what transfer means caused this problem. Various network | |
1779 file transfer programs are suspected of clobbering the high bit. | |
1780 | |
1781 If you have a copy of Emacs that has been damaged in its | |
1782 nonprinting characters, you can fix them: | |
1783 | |
1784 1) Record the names of all the .elc files. | |
1785 2) Delete all the .elc files. | |
1786 3) Recompile alloc.c with a value of PURESIZE twice as large. | |
1787 (See puresize.h.) You might as well save the old alloc.o. | |
1788 4) Remake emacs. It should work now. | |
1789 5) Running emacs, do Meta-x byte-compile-file repeatedly | |
1790 to recreate all the .elc files that used to exist. | |
1791 You may need to increase the value of the variable | |
1792 max-lisp-eval-depth to succeed in running the compiler interpreted | |
1793 on certain .el files. 400 was sufficient as of last report. | |
1794 6) Reinstall the old alloc.o (undoing changes to alloc.c if any) | |
1795 and remake temacs. | |
1796 7) Remake emacs. It should work now, with valid .elc files. | |
1797 | |
1798 * temacs prints "Pure Lisp storage exhausted" | |
1799 | |
1800 This means that the Lisp code loaded from the .elc and .el | |
1801 files during temacs -l loadup inc dump took up more | |
1802 space than was allocated. | |
1803 | |
1804 This could be caused by | |
1805 1) adding code to the preloaded Lisp files | |
1806 2) adding more preloaded files in loadup.el | |
1807 3) having a site-init.el or site-load.el which loads files. | |
1808 Note that ANY site-init.el or site-load.el is nonstandard; | |
1809 if you have received Emacs from some other site | |
1810 and it contains a site-init.el or site-load.el file, consider | |
1811 deleting that file. | |
1812 4) getting the wrong .el or .elc files | |
1813 (not from the directory you expected). | |
1814 5) deleting some .elc files that are supposed to exist. | |
1815 This would cause the source files (.el files) to be | |
1816 loaded instead. They take up more room, so you lose. | |
1817 6) a bug in the Emacs distribution which underestimates | |
1818 the space required. | |
1819 | |
1820 If the need for more space is legitimate, change the definition | |
1821 of PURESIZE in puresize.h. | |
1822 | |
1823 But in some of the cases listed above, this problem is a consequence | |
1824 of something else that is wrong. Be sure to check and fix the real | |
1825 problem. | |
1826 | |
1827 * Changes made to .el files do not take effect. | |
1828 | |
1829 You may have forgotten to recompile them into .elc files. | |
1830 Then the old .elc files will be loaded, and your changes | |
1831 will not be seen. To fix this, do M-x byte-recompile-directory | |
1832 and specify the directory that contains the Lisp files. | |
1833 | |
1834 Emacs should print a warning when loading a .elc file which is older | |
1835 than the corresponding .el file. | |
1836 | |
1837 * The dumped Emacs crashes when run, trying to write pure data. | |
1838 | |
1839 Two causes have been seen for such problems. | |
1840 | |
1841 1) On a system where getpagesize is not a system call, it is defined | |
1842 as a macro. If the definition (in both unexec.c and malloc.c) is wrong, | |
1843 it can cause problems like this. You might be able to find the correct | |
1844 value in the man page for a.out (5). | |
1845 | |
1846 2) Some systems allocate variables declared static among the | |
1847 initialized variables. Emacs makes all initialized variables in most | |
1848 of its files pure after dumping, but the variables declared static and | |
1849 not initialized are not supposed to be pure. On these systems you | |
1850 may need to add "#define static" to the m- or the s- file. | |
1851 | |
1852 * Compilation errors on VMS. | |
1853 | |
1854 You will get warnings when compiling on VMS because there are | |
1855 variable names longer than 32 (or whatever it is) characters. | |
1856 This is not an error. Ignore it. | |
1857 | |
1858 VAX C does not support #if defined(foo). Uses of this construct | |
1859 were removed, but some may have crept back in. They must be rewritten. | |
1860 | |
1861 There is a bug in the C compiler which fails to sign extend characters | |
1862 in conditional expressions. The bug is: | |
1863 char c = -1, d = 1; | |
1864 int i; | |
1865 | |
1866 i = d ? c : d; | |
1867 The result is i == 255; the fix is to typecast the char in the | |
1868 conditional expression as an (int). Known occurrences of such | |
1869 constructs in Emacs have been fixed. | |
1870 | |
1871 * rmail gets error getting new mail | |
1872 | |
1873 rmail gets new mail from /usr/spool/mail/$USER using a program | |
1874 called `movemail'. This program interlocks with /bin/mail using | |
1875 the protocol defined by /bin/mail. | |
1876 | |
1877 There are two different protocols in general use. One of them uses | |
1878 the `flock' system call. The other involves creating a lock file; | |
1879 `movemail' must be able to write in /usr/spool/mail in order to do | |
1880 this. You control which one is used by defining, or not defining, | |
1881 the macro MAIL_USE_FLOCK in config.h or the m- or s- file it includes. | |
1882 IF YOU DON'T USE THE FORM OF INTERLOCKING THAT IS NORMAL ON YOUR | |
1883 SYSTEM, YOU CAN LOSE MAIL! | |
1884 | |
1885 If your system uses the lock file protocol, and fascist restrictions | |
1886 prevent ordinary users from writing the lock files in /usr/spool/mail, | |
1887 you may need to make `movemail' setgid to a suitable group such as | |
1888 `mail'. You can use these commands (as root): | |
1889 | |
1890 chgrp mail movemail | |
1891 chmod 2755 movemail | |
1892 | |
1893 If your system uses the lock file protocol, and fascist restrictions | |
1894 prevent ordinary users from writing the lock files in /usr/spool/mail, | |
1895 you may need to make `movemail' setgid to a suitable group such as | |
1896 `mail'. To do this, use the following commands (as root) after doing the | |
1897 make install. | |
1898 | |
1899 chgrp mail movemail | |
1900 chmod 2755 movemail | |
1901 | |
1902 Installation normally copies movemail from the build directory to an | |
1903 installation directory which is usually under /usr/local/lib. The | |
1904 installed copy of movemail is usually in the directory | |
1905 /usr/local/lib/emacs/VERSION/TARGET. You must change the group and | |
1906 mode of the installed copy; changing the group and mode of the build | |
1907 directory copy is ineffective. | |
1908 | |
1909 * Emacs spontaneously displays "I-search: " at the bottom of the screen. | |
1910 | |
1911 This means that Control-S/Control-Q (XON/XOFF) "flow control" is being | |
1912 used. C-s/C-q flow control is bad for Emacs editors because it takes | |
1913 away C-s and C-q as user commands. Since editors do not output long | |
1914 streams of text without user commands, there is no need for a | |
1915 user-issuable "stop output" command in an editor; therefore, a | |
1916 properly designed flow control mechanism would transmit all possible | |
1917 input characters without interference. Designing such a mechanism is | |
1918 easy, for a person with at least half a brain. | |
1919 | |
1920 There are three possible reasons why flow control could be taking place: | |
1921 | |
1922 1) Terminal has not been told to disable flow control | |
1923 2) Insufficient padding for the terminal in use | |
1924 3) Some sort of terminal concentrator or line switch is responsible | |
1925 | |
1926 First of all, many terminals have a set-up mode which controls whether | |
1927 they generate XON/XOFF flow control characters. This must be set to | |
1928 "no XON/XOFF" in order for Emacs to work. Sometimes there is an | |
1929 escape sequence that the computer can send to turn flow control off | |
1930 and on. If so, perhaps the termcap `ti' string should turn flow | |
1931 control off, and the `te' string should turn it on. | |
1932 | |
1933 Once the terminal has been told "no flow control", you may find it | |
1934 needs more padding. The amount of padding Emacs sends is controlled | |
1935 by the termcap entry for the terminal in use, and by the output baud | |
1936 rate as known by the kernel. The shell command `stty' will print | |
1937 your output baud rate; `stty' with suitable arguments will set it if | |
1938 it is wrong. Setting to a higher speed causes increased padding. If | |
1939 the results are wrong for the correct speed, there is probably a | |
1940 problem in the termcap entry. You must speak to a local Unix wizard | |
1941 to fix this. Perhaps you are just using the wrong terminal type. | |
1942 | |
1943 For terminals that lack a "no flow control" mode, sometimes just | |
1944 giving lots of padding will prevent actual generation of flow control | |
1945 codes. You might as well try it. | |
1946 | |
1947 If you are really unlucky, your terminal is connected to the computer | |
1948 through a concentrator which sends XON/XOFF flow control to the | |
1949 computer, or it insists on sending flow control itself no matter how | |
1950 much padding you give it. Unless you can figure out how to turn flow | |
1951 control off on this concentrator (again, refer to your local wizard), | |
1952 you are screwed! You should have the terminal or concentrator | |
1953 replaced with a properly designed one. In the mean time, some drastic | |
1954 measures can make Emacs semi-work. | |
1955 | |
1956 You can make Emacs ignore C-s and C-q and let the operating system | |
1957 handle them. To do this on a per-session basis, just type M-x | |
1958 enable-flow-control RET. You will see a message that C-\ and C-^ are | |
1959 now translated to C-s and C-q. (Use the same command M-x | |
1960 enable-flow-control to turn *off* this special mode. It toggles flow | |
1961 control handling.) | |
1962 | |
1963 If C-\ and C-^ are inconvenient for you (for example, if one of them | |
1964 is the escape character of your terminal concentrator), you can choose | |
1965 other characters by setting the variables flow-control-c-s-replacement | |
1966 and flow-control-c-q-replacement. But choose carefully, since all | |
1967 other control characters are already used by emacs. | |
1968 | |
1969 IMPORTANT: if you type C-s by accident while flow control is enabled, | |
1970 Emacs output will freeze, and you will have to remember to type C-q in | |
1971 order to continue. | |
1972 | |
1973 If you work in an environment where a majority of terminals of a | |
1974 certain type are flow control hobbled, you can use the function | |
1975 `enable-flow-control-on' to turn on this flow control avoidance scheme | |
1976 automatically. Here is an example: | |
1977 | |
1978 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131") | |
1979 | |
1980 If this isn't quite correct (e.g. you have a mixture of flow-control hobbled | |
1981 and good vt200 terminals), you can still run enable-flow-control | |
1982 manually. | |
1983 | |
1984 I have no intention of ever redesigning the Emacs command set for the | |
1985 assumption that terminals use C-s/C-q flow control. XON/XOFF flow | |
1986 control technique is a bad design, and terminals that need it are bad | |
1987 merchandise and should not be purchased. Now that X is becoming | |
1988 widespread, XON/XOFF seems to be on the way out. If you can get some | |
1989 use out of GNU Emacs on inferior terminals, more power to you, but I | |
1990 will not make Emacs worse for properly designed systems for the sake | |
1991 of inferior systems. | |
1992 | |
1993 * Control-S and Control-Q commands are ignored completely. | |
1994 | |
1995 For some reason, your system is using brain-damaged C-s/C-q flow | |
1996 control despite Emacs's attempts to turn it off. Perhaps your | |
1997 terminal is connected to the computer through a concentrator | |
1998 that wants to use flow control. | |
1999 | |
2000 You should first try to tell the concentrator not to use flow control. | |
2001 If you succeed in this, try making the terminal work without | |
2002 flow control, as described in the preceding section. | |
2003 | |
2004 If that line of approach is not successful, map some other characters | |
2005 into C-s and C-q using keyboard-translate-table. The example above | |
2006 shows how to do this with C-^ and C-\. | |
2007 | |
2008 * Control-S and Control-Q commands are ignored completely on a net connection. | |
2009 | |
2010 Some versions of rlogin (and possibly telnet) do not pass flow | |
2011 control characters to the remote system to which they connect. | |
2012 On such systems, emacs on the remote system cannot disable flow | |
2013 control on the local system. | |
2014 | |
2015 One way to cure this is to disable flow control on the local host | |
2016 (the one running rlogin, not the one running rlogind) using the | |
2017 stty command, before starting the rlogin process. On many systems, | |
2018 "stty start u stop u" will do this. | |
2019 | |
2020 Some versions of tcsh will prevent even this from working. One way | |
2021 around this is to start another shell before starting rlogin, and | |
2022 issue the stty command to disable flow control from that shell. | |
2023 | |
2024 If none of these methods work, the best solution is to type | |
2025 M-x enable-flow-control at the beginning of your emacs session, or | |
2026 if you expect the problem to continue, add a line such as the | |
2027 following to your .emacs (on the host running rlogind): | |
2028 | |
2029 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131") | |
2030 | |
2031 See the entry about spontaneous display of I-search (above) for more | |
2032 info. | |
2033 | |
2034 * Screen is updated wrong, but only on one kind of terminal. | |
2035 | |
2036 This could mean that the termcap entry you are using for that | |
2037 terminal is wrong, or it could mean that Emacs has a bug handing | |
2038 the combination of features specified for that terminal. | |
2039 | |
2040 The first step in tracking this down is to record what characters | |
2041 Emacs is sending to the terminal. Execute the Lisp expression | |
2042 (open-termscript "./emacs-script") to make Emacs write all | |
2043 terminal output into the file ~/emacs-script as well; then do | |
2044 what makes the screen update wrong, and look at the file | |
2045 and decode the characters using the manual for the terminal. | |
2046 There are several possibilities: | |
2047 | |
2048 1) The characters sent are correct, according to the terminal manual. | |
2049 | |
2050 In this case, there is no obvious bug in Emacs, and most likely you | |
2051 need more padding, or possibly the terminal manual is wrong. | |
2052 | |
2053 2) The characters sent are incorrect, due to an obscure aspect | |
2054 of the terminal behavior not described in an obvious way | |
2055 by termcap. | |
2056 | |
2057 This case is hard. It will be necessary to think of a way for | |
2058 Emacs to distinguish between terminals with this kind of behavior | |
2059 and other terminals that behave subtly differently but are | |
2060 classified the same by termcap; or else find an algorithm for | |
2061 Emacs to use that avoids the difference. Such changes must be | |
2062 tested on many kinds of terminals. | |
2063 | |
2064 3) The termcap entry is wrong. | |
2065 | |
2066 See the file etc/TERMS for information on changes | |
2067 that are known to be needed in commonly used termcap entries | |
2068 for certain terminals. | |
2069 | |
2070 4) The characters sent are incorrect, and clearly cannot be | |
2071 right for any terminal with the termcap entry you were using. | |
2072 | |
2073 This is unambiguously an Emacs bug, and can probably be fixed | |
2074 in termcap.c, tparam.c, term.c, scroll.c, cm.c or dispnew.c. | |
2075 | |
2076 * Output from Control-V is slow. | |
2077 | |
2078 On many bit-map terminals, scrolling operations are fairly slow. | |
2079 Often the termcap entry for the type of terminal in use fails | |
2080 to inform Emacs of this. The two lines at the bottom of the screen | |
2081 before a Control-V command are supposed to appear at the top after | |
2082 the Control-V command. If Emacs thinks scrolling the lines is fast, | |
2083 it will scroll them to the top of the screen. | |
2084 | |
2085 If scrolling is slow but Emacs thinks it is fast, the usual reason is | |
2086 that the termcap entry for the terminal you are using does not | |
2087 specify any padding time for the `al' and `dl' strings. Emacs | |
2088 concludes that these operations take only as much time as it takes to | |
2089 send the commands at whatever line speed you are using. You must | |
2090 fix the termcap entry to specify, for the `al' and `dl', as much | |
2091 time as the operations really take. | |
2092 | |
2093 Currently Emacs thinks in terms of serial lines which send characters | |
2094 at a fixed rate, so that any operation which takes time for the | |
2095 terminal to execute must also be padded. With bit-map terminals | |
2096 operated across networks, often the network provides some sort of | |
2097 flow control so that padding is never needed no matter how slow | |
2098 an operation is. You must still specify a padding time if you want | |
2099 Emacs to realize that the operation takes a long time. This will | |
2100 cause padding characters to be sent unnecessarily, but they do | |
2101 not really cost much. They will be transmitted while the scrolling | |
2102 is happening and then discarded quickly by the terminal. | |
2103 | |
2104 Most bit-map terminals provide commands for inserting or deleting | |
2105 multiple lines at once. Define the `AL' and `DL' strings in the | |
2106 termcap entry to say how to do these things, and you will have | |
2107 fast output without wasted padding characters. These strings should | |
2108 each contain a single %-spec saying how to send the number of lines | |
2109 to be scrolled. These %-specs are like those in the termcap | |
2110 `cm' string. | |
2111 | |
2112 You should also define the `IC' and `DC' strings if your terminal | |
2113 has a command to insert or delete multiple characters. These | |
2114 take the number of positions to insert or delete as an argument. | |
2115 | |
2116 A `cs' string to set the scrolling region will reduce the amount | |
2117 of motion you see on the screen when part of the screen is scrolled. | |
2118 | |
2119 * Your Delete key sends a Backspace to the terminal, using an AIXterm. | |
2120 | |
2121 The solution is to include in your .Xdefaults the lines: | |
2122 | |
2123 *aixterm.Translations: #override <Key>BackSpace: string(0x7f) | |
2124 aixterm*ttyModes: erase ^? | |
2125 | |
2126 This makes your Backspace key send DEL (ASCII 127). | |
2127 | |
2128 * You type Control-H (Backspace) expecting to delete characters. | |
2129 | |
2130 Put `stty dec' in your .login file and your problems will disappear | |
2131 after a day or two. | |
2132 | |
2133 The choice of Backspace for erasure was based on confusion, caused by | |
2134 the fact that backspacing causes erasure (later, when you type another | |
2135 character) on most display terminals. But it is a mistake. Deletion | |
2136 of text is not the same thing as backspacing followed by failure to | |
2137 overprint. I do not wish to propagate this confusion by conforming | |
2138 to it. | |
2139 | |
2140 For this reason, I believe `stty dec' is the right mode to use, | |
2141 and I have designed Emacs to go with that. If there were a thousand | |
2142 other control characters, I would define Control-h to delete as well; | |
2143 but there are not very many other control characters, and I think | |
2144 that providing the most mnemonic possible Help character is more | |
2145 important than adapting to people who don't use `stty dec'. | |
2146 | |
2147 If you are obstinate about confusing buggy overprinting with deletion, | |
2148 you can redefine Backspace in your .emacs file: | |
2149 (global-set-key "\b" 'delete-backward-char) | |
2150 You can probably access help-command via f1. | |
2151 | |
2152 * Editing files through RFS gives spurious "file has changed" warnings. | |
2153 It is possible that a change in Emacs 18.37 gets around this problem, | |
2154 but in case not, here is a description of how to fix the RFS bug that | |
2155 causes it. | |
2156 | |
2157 There was a serious pair of bugs in the handling of the fsync() system | |
2158 call in the RFS server. | |
2159 | |
2160 The first is that the fsync() call is handled as another name for the | |
2161 close() system call (!!). It appears that fsync() is not used by very | |
2162 many programs; Emacs version 18 does an fsync() before closing files | |
2163 to make sure that the bits are on the disk. | |
2164 | |
2165 This is fixed by the enclosed patch to the RFS server. | |
2166 | |
2167 The second, more serious problem, is that fsync() is treated as a | |
2168 non-blocking system call (i.e., it's implemented as a message that | |
2169 gets sent to the remote system without waiting for a reply). Fsync is | |
2170 a useful tool for building atomic file transactions. Implementing it | |
2171 as a non-blocking RPC call (when the local call blocks until the sync | |
2172 is done) is a bad idea; unfortunately, changing it will break the RFS | |
2173 protocol. No fix was supplied for this problem. | |
2174 | |
2175 (as always, your line numbers may vary) | |
2176 | |
2177 % rcsdiff -c -r1.2 serversyscall.c | |
2178 RCS file: RCS/serversyscall.c,v | |
2179 retrieving revision 1.2 | |
2180 diff -c -r1.2 serversyscall.c | |
2181 *** /tmp/,RCSt1003677 Wed Jan 28 15:15:02 1987 | |
2182 --- serversyscall.c Wed Jan 28 15:14:48 1987 | |
2183 *************** | |
2184 *** 163,169 **** | |
2185 /* | |
2186 * No return sent for close or fsync! | |
2187 */ | |
2188 ! if (syscall == RSYS_close || syscall == RSYS_fsync) | |
2189 proc->p_returnval = deallocate_fd(proc, msg->m_args[0]); | |
2190 else | |
2191 { | |
2192 --- 166,172 ---- | |
2193 /* | |
2194 * No return sent for close or fsync! | |
2195 */ | |
2196 ! if (syscall == RSYS_close) | |
2197 proc->p_returnval = deallocate_fd(proc, msg->m_args[0]); | |
2198 else | |
2199 { | |
2200 | |
2201 * Vax C compiler bugs affecting Emacs. | |
2202 | |
2203 You may get one of these problems compiling Emacs: | |
2204 | |
2205 foo.c line nnn: compiler error: no table entry for op STASG | |
2206 foo.c: fatal error in /lib/ccom | |
2207 | |
2208 These are due to bugs in the C compiler; the code is valid C. | |
2209 Unfortunately, the bugs are unpredictable: the same construct | |
2210 may compile properly or trigger one of these bugs, depending | |
2211 on what else is in the source file being compiled. Even changes | |
2212 in header files that should not affect the file being compiled | |
2213 can affect whether the bug happens. In addition, sometimes files | |
2214 that compile correctly on one machine get this bug on another machine. | |
2215 | |
2216 As a result, it is hard for me to make sure this bug will not affect | |
2217 you. I have attempted to find and alter these constructs, but more | |
2218 can always appear. However, I can tell you how to deal with it if it | |
2219 should happen. The bug comes from having an indexed reference to an | |
2220 array of Lisp_Objects, as an argument in a function call: | |
2221 Lisp_Object *args; | |
2222 ... | |
2223 ... foo (5, args[i], ...)... | |
2224 putting the argument into a temporary variable first, as in | |
2225 Lisp_Object *args; | |
2226 Lisp_Object tem; | |
2227 ... | |
2228 tem = args[i]; | |
2229 ... foo (r, tem, ...)... | |
2230 causes the problem to go away. | |
2231 The `contents' field of a Lisp vector is an array of Lisp_Objects, | |
2232 so you may see the problem happening with indexed references to that. | |
2233 | |
2234 * 68000 C compiler problems | |
2235 | |
2236 Various 68000 compilers have different problems. | |
2237 These are some that have been observed. | |
2238 | |
2239 ** Using value of assignment expression on union type loses. | |
2240 This means that x = y = z; or foo (x = z); does not work | |
2241 if x is of type Lisp_Object. | |
2242 | |
2243 ** "cannot reclaim" error. | |
2244 | |
2245 This means that an expression is too complicated. You get the correct | |
2246 line number in the error message. The code must be rewritten with | |
2247 simpler expressions. | |
2248 | |
2249 ** XCONS, XSTRING, etc macros produce incorrect code. | |
2250 | |
2251 If temacs fails to run at all, this may be the cause. | |
2252 Compile this test program and look at the assembler code: | |
2253 | |
2254 struct foo { char x; unsigned int y : 24; }; | |
2255 | |
2256 lose (arg) | |
2257 struct foo arg; | |
2258 { | |
2259 test ((int *) arg.y); | |
2260 } | |
2261 | |
2262 If the code is incorrect, your compiler has this problem. | |
2263 In the XCONS, etc., macros in lisp.h you must replace (a).u.val with | |
2264 ((a).u.val + coercedummy) where coercedummy is declared as int. | |
2265 | |
2266 This problem will not happen if the m-...h file for your type | |
2267 of machine defines NO_UNION_TYPE. That is the recommended setting now. | |
2268 | |
2269 * C compilers lose on returning unions | |
2270 | |
2271 I hear that some C compilers cannot handle returning a union type. | |
2272 Most of the functions in GNU Emacs return type Lisp_Object, which is | |
2273 defined as a union on some rare architectures. | |
2274 | |
2275 This problem will not happen if the m-...h file for your type | |
2276 of machine defines NO_UNION_TYPE. | |
2277 |