Mercurial > hg > xemacs-beta
annotate PROBLEMS @ 5391:f9dc75bdbdc4
Implement #'load-time-value less hackishly, by modifying the byte compiler.
2011-04-02 Aidan Kehoe <kehoea@parhasard.net>
* bytecomp.el (byte-compile-output-preface): New.
* bytecomp.el (byte-compile-output-file-form):
* bytecomp.el (byte-compile-output-docform):
* bytecomp.el (byte-compile-file-form):
* bytecomp.el (byte-compile-file-form-defmumble):
* bytecomp.el (symbol-value):
* bytecomp.el (byte-compile-symbol-value): New.
* cl-macs.el (load-time-value):
No longer implement load-time-value by very hackishly redefining
#'byte-compile-file-form-defmumble, instead make the appropriate
changes in #'byte-compile-file-form-defmumble and
#'byte-compile-file-form instead. We also add a specific byte-compile
method for #'symbol-value, using the add-properties-to-a-gensym
approach that worked for #'block and #'return-from.
| author | Aidan Kehoe <kehoea@parhasard.net> |
|---|---|
| date | Sat, 02 Apr 2011 16:13:20 +0100 |
| parents | 1a9c94ba117c |
| children | 388762703a21 |
| rev | line source |
|---|---|
| 278 | 1 -*- mode:outline -*- |
| 2 | |
| 0 | 3 This file describes various problems that have been encountered |
| 197 | 4 in compiling, installing and running XEmacs. It has been updated for |
|
4634
c064d7197712
Add historical deprecation to PROBLEMS.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
3863
diff
changeset
|
5 XEmacs 21.5. Note that the issues are by now mainly historic; XEmacs |
|
c064d7197712
Add historical deprecation to PROBLEMS.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
3863
diff
changeset
|
6 no longer depends on bleeding edge features of operating systems, but |
|
c064d7197712
Add historical deprecation to PROBLEMS.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
3863
diff
changeset
|
7 rather is quite conservative. Operational issues (common user |
|
c064d7197712
Add historical deprecation to PROBLEMS.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
3863
diff
changeset
|
8 misunderstandings and such) are described in the FAQ, not here. |
| 0 | 9 |
| 278 | 10 This file is rather large, but we have tried to sort the entries by |
| 11 their respective relevance for XEmacs, but may have not succeeded | |
| 12 completely in that task. The file is divided into four parts: | |
| 124 | 13 |
| 197 | 14 - Problems with building XEmacs |
| 15 - Problems with running XEmacs | |
| 16 - Compatibility problems | |
| 17 - Mule issues | |
| 120 | 18 |
| 197 | 19 Use `C-c C-f' to move to the next equal level of outline, and |
| 223 | 20 `C-c C-b' to move to previous equal level. `C-h m' will give more |
| 21 info about the Outline mode. | |
| 120 | 22 |
| 197 | 23 Also, Try finding the things you need using one of the search commands |
| 24 XEmacs provides (e.g. `C-s'). | |
| 25 | |
| 524 | 26 General advice: |
| 957 | 27 |
| 524 | 28 WATCH OUT for your init file! (~/.xemacs/init.el or ~/.emacs) If |
| 29 you observe strange problems, invoke XEmacs with the `-vanilla' | |
| 30 option and see if you can repeat the problem. | |
| 197 | 31 |
| 957 | 32 Note that most of the problems described here manifest at RUN |
| 33 time, even those described as BUILD problems. It is quite unusual | |
| 34 for a released XEmacs to fail to build. So a "build problem" | |
| 35 requires you to tweak the build environment, then rebuild XEmacs. | |
| 36 A "runtime problem" is one that can be fixed by proper | |
| 37 configuration of the existing build. Compatibility problems and | |
| 38 Mule issues are generally runtime problems, but are treated | |
| 39 separately for convenience. | |
| 40 | |
| 120 | 41 |
| 124 | 42 * Problems with building XEmacs |
| 197 | 43 =============================== |
| 0 | 44 |
| 373 | 45 ** General |
| 1245 | 46 |
| 915 | 47 Much general information is in INSTALL. If it's covered in |
| 48 INSTALL, we don't repeat it here. | |
| 49 | |
| 3404 | 50 *** X11/bitmaps/gray (or other X11-related file) not found. |
| 51 | |
| 52 The X11R6 distribution was monolithic, but the X11R7 distribution is | |
| 53 much more modular. Many OS distributions omit these bitmaps (assuming | |
| 54 nobody uses them, evidently). Your OS distribution should have a | |
| 55 developer's package containing these files, probably with a name | |
| 56 containing the string "bitmap". Known package names (you may need to | |
| 57 add an extension such as .deb or .rpm) include x11/xbitmaps (Ubuntu) | |
| 58 and xorg-x11-xbitmaps (Fedora Core 5). | |
| 59 | |
| 1098 | 60 *** How do I configure to get the buffer tabs/progress bars? |
| 915 | 61 |
| 62 These features depend on support for "native widgets". Use the | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
63 --with-widgets option to configure. Configuration of widgets is |
| 915 | 64 automatic for "modern" toolkits (MS Windows, GTK, and Motif), but if |
| 65 you are using Xt and the Athena widgets, you will probably want to | |
| 66 specify a "3d" widget set. See configure --usage, and don't forget to | |
| 67 install the corresponding development libraries. | |
| 68 | |
| 69 *** I know I have libfoo installed, but configure doesn't find it. | |
| 70 | |
| 71 Typical of Linux systems with package managers. To link with a shared | |
| 72 library, you only need the shared library. To compile objects that | |
| 73 link with it, you need the headers---and distros don't provide them with | |
| 74 the libraries. You need the additional "development" package, too. | |
| 75 | |
| 373 | 76 *** When using gcc, you get the error message "undefined symbol __fixunsdfsi". |
| 77 When using gcc, you get the error message "undefined symbol __main". | |
| 78 | |
| 79 This means that you need to link with the gcc library. It may be called | |
| 80 "gcc-gnulib" or "libgcc.a"; figure out where it is, and define LIB_GCC in | |
| 81 config.h to point to it. | |
| 82 | |
| 83 It may also work to use the GCC version of `ld' instead of the standard one. | |
| 84 | |
| 85 *** Excessive optimization with pgcc can break XEmacs | |
| 124 | 86 |
| 87 It has been reported on some systems that compiling with -O6 can lead | |
| 88 to XEmacs failures. The workaround is to use a lower optimization | |
| 89 level. -O2 and -O4 have been tested extensively. | |
| 90 | |
| 229 | 91 All of this depends heavily on the version of pgcc and the version |
| 92 of libc. Snapshots near the release of pgcc-1.0 have been tested | |
| 93 extensively and no sign of breakage has been seen on systems using | |
| 94 glibc-2. | |
| 95 | |
| 373 | 96 *** src/Makefile and lib-src/Makefile are truncated--most of the file missing. |
| 229 | 97 |
| 373 | 98 This can happen if configure uses GNU sed version 2.03. That version |
| 99 had a bug. GNU sed version 2.05 works properly. | |
| 124 | 100 |
| 373 | 101 *** When compiling with X11, you get "undefined symbol _XtStrings". |
| 124 | 102 |
| 373 | 103 This means that you are trying to link emacs against the X11r4 version of |
| 104 libXt.a, but you have compiled either Emacs or the code in the lwlib | |
| 105 subdirectory with the X11r5 header files. That doesn't work. | |
| 124 | 106 |
| 373 | 107 Remember, you can't compile lwlib for r4 and emacs for r5, or vice versa. |
| 108 They must be in sync. | |
| 124 | 109 |
| 373 | 110 *** test-distrib says that the distribution has been clobbered |
| 197 | 111 or, temacs prints "Command key out of range 0-127" |
| 112 or, temacs runs and dumps xemacs, but xemacs totally fails to work. | |
| 113 or, temacs gets errors dumping xemacs | |
| 114 | |
| 115 This can be because the .elc files have been garbled. Do not be | |
| 116 fooled by the fact that most of a .elc file is text: these are binary | |
| 117 files and can contain all 256 byte values. | |
| 118 | |
| 119 In particular `shar' cannot be used for transmitting GNU Emacs. It | |
| 120 typically truncates "lines". (this does not apply to GNU shar, which | |
| 121 uses uuencode to encode binary files.) | |
| 122 | |
| 123 If you have a copy of Emacs that has been damaged in its nonprinting | |
| 124 characters, you can fix them by running: | |
| 125 | |
| 126 make all-elc | |
| 127 | |
| 128 This will rebuild all the needed .elc files. | |
| 129 | |
| 1318 | 130 ** Intel Architecture General |
| 131 | |
| 132 *** Don't use -O2 or -O3 with Cygwin 1.0, CodeFusion-99070 or gcc 2.7.2 on x86 | |
| 133 without also using `-fno-strength-reduce'. | |
| 134 | |
| 135 gcc will generate incorrect code otherwise. This bug is present in at | |
| 136 least 2.6.x and 2.7.[0-2]. This bug has been fixed in GCC 2.7.2.1 and | |
| 137 later. This bug is O/S independent, but is limited to x86 architectures. | |
| 138 | |
| 139 This problem is known to be fixed in egcs (or pgcc) 1.0 or later. | |
| 140 | |
| 141 Unfortunately, later releases of Cygnus-released compilers (not the | |
| 142 Net-released ones) have a bug with the same `problem signature'. | |
| 143 | |
| 144 If you're lucky, you'll get an error while compiling that looks like: | |
| 145 | |
| 146 event-stream.c:3189: internal error--unrecognizable insn: | |
| 147 (insn 256 14 15 (set (reg/v:SI 24) | |
| 148 (minus:SI (reg/v:SI 25) | |
| 149 (const_int 2))) -1 (insn_list 11 (nil)) | |
| 150 (nil)) | |
| 151 0 0 [main] | |
| 152 | |
| 153 If you're unlucky, your code will simply execute incorrectly. | |
| 154 | |
| 155 *** Don't use -O2 with gcc 2.7.2 under Intel architectures without also | |
| 156 using `-fno-caller-saves'. | |
| 157 | |
| 158 gcc will generate incorrect code otherwise. This bug is still | |
| 159 present in gcc 2.7.2.3. There have been no reports to indicate the | |
| 160 bug is present in egcs 1.0 (or pgcc 1.0) or later. This bug is O/S | |
| 161 independent, but limited to x86 architectures. | |
| 162 | |
| 163 This problem is known to be fixed in egcs (or pgcc) 1.0 or later. | |
| 164 | |
| 373 | 165 *** `compress' and `uncompress' not found and XFree86 |
| 166 | |
| 167 XFree86 installs a very old version of libz.a by default ahead of where | |
| 168 more modern version of libz might be installed. This will cause problems | |
| 169 when attempting to link against libMagick. The fix is to remove the old | |
| 170 libz.a in the X11 binary directory. | |
| 171 | |
| 172 | |
| 3863 | 173 ** X11 and Motif |
| 1245 | 174 |
| 175 Motif is the X11 version of the Gnus torture test: if there's a way to | |
| 176 crash, Motif will find it. With the open source release of Motif, it | |
| 177 seems like a good idea to collect all Motif-related issues in one | |
| 3863 | 178 place. X11 itself is not all that safe, either. |
| 179 | |
| 180 You should also look in your OS's section, as it may not be the window | |
| 181 system (toolkit's) fault. | |
| 182 | |
| 183 *** XEmacs and the X server crash when inserting or displaying a TAB character. | |
| 184 | |
| 185 If you are using the XFree86 distribution, you need an X server with | |
| 186 this patch applied: | |
| 187 | |
| 188 https://bugs.freedesktop.org/attachment.cgi?id=2016 | |
| 189 | |
| 190 Versions of XFree86 previous to that crashed when an app tried to draw a | |
| 191 literal tab character using many fonts. | |
| 1245 | 192 |
| 193 *** XEmacs crashes on exit (#1). | |
| 194 | |
| 195 The backtrace is something like: | |
| 196 | |
| 197 (gdb) where | |
| 198 #0 0xfeb9a480 in _libc_kill () from /usr/lib/libc.so.1 | |
| 199 #1 0x000b0388 in fatal_error_signal () | |
| 200 #2 <signal handler called> | |
| 201 #3 YowIter (ht=0xb, id=0x0, v=0x74682074, client=0x47e3c0) | |
| 202 at ImageCache.c:1159 | |
| 203 #4 0xff26cc5c in _LTHashTableForEachItem (ht=0x4725e8, | |
| 204 iter=0xff26dda0 <YowIter>, ClientData=0x47e3c0) at Hash.c:671 | |
| 205 #5 0xff2a4664 in destroy (w=0x496550) at Screen.c:352 | |
| 206 #6 0xfef92118 in Phase2Destroy () from /usr/openwin/lib/libXt.so.4 | |
| 207 #7 0xfef91940 in Recursive () from /usr/openwin/lib/libXt.so.4 | |
| 208 #8 0xfef91e44 in XtPhase2Destroy () from /usr/openwin/lib/libXt.so.4 | |
| 209 #9 0xfef91ae8 in _XtDoPhase2Destroy () from /usr/openwin/lib/libXt.so.4 | |
| 210 #10 0xfef918cc in XtDestroyWidget () from /usr/openwin/lib/libXt.so.4 | |
| 211 #11 0xfef91438 in CloseDisplay () from /usr/openwin/lib/libXt.so.4 | |
| 212 #12 0xfef91394 in XtCloseDisplay () from /usr/openwin/lib/libXt.so.4 | |
| 213 #13 0x0025b8b0 in x_delete_device () | |
| 214 #14 0x000940b0 in delete_device_internal () | |
| 215 #15 0x000806a0 in delete_console_internal () | |
| 216 | |
| 217 This is known to happen with Lesstif version 0.93.36. Similar | |
| 218 backtraces have also been observed on HP/UX and Solaris. There is a | |
| 219 patch for Lesstif. (This is not a solution; it just stops the crash. | |
| 220 It may or may not be harmless, but "it works for the author".) | |
| 221 | |
| 222 Note that this backtrace looks a lot like the one in the next item. | |
| 223 However, this one is invulnerable to the Solaris patches mentioned there. | |
| 224 | |
| 225 Frank McIngvale <frankm@hiwaay.net> says: | |
| 226 | |
| 227 Ok, 0.93.34 works, and I tracked down the crash to a section | |
| 228 marked "experimental" in 0.93.36. Patch attached, "works for me". | |
| 229 | |
| 230 diff -u -r lesstif-0.93.36/lib/Xm/ImageCache.c lesstif-0.93.36-mod/lib/Xm/ImageCache.c | |
| 231 --- lesstif-0.93.36/lib/Xm/ImageCache.c 2002-08-05 14:53:24.000000000 -0500 | |
| 232 +++ lesstif-0.93.36-mod/lib/Xm/ImageCache.c 2002-11-11 11:13:12.000000000 -0600 | |
| 233 @@ -1166,5 +1166,4 @@ | |
| 234 DEBUGOUT(_LtDebug0(__FILE__, NULL, "_LtImageCacheScreenDestroy (XmGetPixmapByDepth) %p\n", | |
| 235 s)); | |
| 236 | |
| 237 - (void) _LTHashTableForEachItem(PixmapCache, YowIter, (XtPointer)s); | |
| 238 } | |
| 239 | |
| 240 *** XEmacs crashes on exit (#2) | |
| 241 | |
| 242 Especially frequent with multiple frames. Crashes that produce C | |
| 243 backtraces like this: | |
| 244 | |
| 245 #0 0xfec9a118 in _libc_kill () from /usr/lib/libc.so.1 | |
| 246 #1 0x77f48 in fatal_error_signal (sig=11) | |
| 247 at /codes/rpluim/xemacs-21.4/src/emacs.c:539 | |
| 248 #2 <signal handler called> | |
| 249 #3 0xfee929f4 in XFindContext () from /usr/openwin/lib/libX11.so.4 | |
| 250 #4 0xfee92930 in XFindContext () from /usr/openwin/lib/libX11.so.4 | |
| 251 #5 0xff297e54 in DisplayDestroy () from /usr/dt/lib/libXm.so.4 | |
| 252 #6 0xfefbece0 in XtCallCallbackList () from /usr/openwin/lib/libXt.so.4 | |
| 253 #7 0xfefc486c in XtPhase2Destroy () from /usr/openwin/lib/libXt.so.4 | |
| 254 #8 0xfefc45d0 in _XtDoPhase2Destroy () from /usr/openwin/lib/libXt.so.4 | |
| 255 #9 0xfefc43b4 in XtDestroyWidget () from /usr/openwin/lib/libXt.so.4 | |
| 256 #10 0x15cf9c in x_delete_device (d=0x523f00) | |
| 257 | |
| 258 are caused by buggy Motif libraries. Installing the following patches | |
| 259 has been reported to solve the problem on Solaris 2.7: | |
| 260 | |
| 261 107081-40 107656-07 | |
| 262 | |
| 263 For information (although they have not been confirmed to work), the | |
| 264 equivalent patches for Solaris 2.8 are: | |
| 265 | |
| 266 108940-33 108652-25 | |
| 267 | |
| 268 *** On HP-UX 11.0 XEmacs causes excessive X11 errors when running. | |
| 269 (also appears on AIX as reported in comp.emacs.xemacs) | |
| 270 | |
| 271 Marcus Thiessel <marcus@xemacs.org> | |
| 272 | |
| 273 Unfortunately, XEmacs releases prior to 21.0 don't work with | |
| 274 Motif2.1. It will compile but you will get excessive X11 errors like | |
| 275 | |
| 276 xemacs: X Error of failed request: BadGC (invalid GC parameter) | |
| 277 | |
| 278 and finally XEmacs gets killed. A workaround is to use the | |
| 279 Motif1.2_R6 libraries. You can the following line to your call to | |
| 280 configure: | |
| 281 | |
| 282 --x-libraries="/usr/lib/Motif1.2_R6 -L/usr/lib/X11R6" | |
| 283 | |
| 284 Make sure /usr/lib/Motif1.2_R6/libXm.sl is a link to | |
| 285 /usr/lib/Motif1.2_R6/libXm.3. | |
| 286 | |
| 287 *** On HP-UX 11.0: Object "" does not have windowed ancestor | |
| 288 | |
| 289 Marcus Thiessel <marcus@xemacs.org> | |
| 290 | |
| 291 XEmacs dies without core file and reports: | |
| 292 | |
| 293 Error: Object "" does not have windowed ancestor. | |
| 294 | |
| 295 This is a bug. Please apply the patch PHSS_19964 (check if | |
| 296 superseded). The other alternative is to link with Motif1.2_R6 (see | |
| 297 previous item). | |
| 298 | |
| 299 *** Motif dialog boxes lose on Irix. | |
| 300 | |
| 301 Larry Auton <lda@control.att.com> writes: | |
| 302 Beware of not specifying | |
| 303 | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
304 --with-dialogs=athena |
| 1245 | 305 |
| 306 if it builds with the motif dialogs [boom!] you're a dead man. | |
| 307 | |
| 308 | |
| 373 | 309 ** AIX |
| 1009 | 310 *** IBM compiler fails: "The character # is not a valid C source character." |
| 311 | |
| 312 Most recently observed in 21.5.9, due to USE_KKCC ifdefs (they just | |
| 313 happen to tickle the implementation). | |
| 314 | |
| 315 Valdis Kletnieks says: | |
| 316 | |
| 317 The problem is that IBM defines a *MACRO* called 'memcpy', and we | |
| 318 have stuck a #ifdef/#endif inside the macro call. As a workaround, | |
| 319 try adding '-U__STR__' to your CFLAGS - this will cause string.h to | |
| 320 not do a #define for strcpy() to __strcpy() - it uses this for | |
| 321 automatic inlining support. | |
| 322 | |
| 323 (For the record, the same issue affects a number of other functions | |
| 324 defined in string.h - basically anything the compiler knows how to | |
| 325 inline.) | |
| 326 | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
327 *** On AIX 4.3, you must specify --with-dialogs=athena with configure |
| 373 | 328 |
| 442 | 329 *** The libXt shipped with AIX 4.3 up to 4.3.2 is broken. This causes |
| 330 xemacs -nw to fail in various ways. The official APAR is this: | |
| 331 | |
| 332 APAR NUMBER: <IX89470> RESOLVED AS: PROGRAM ERROR | |
| 333 | |
| 334 ABSTRACT: | |
| 335 <IX89470>: LIBXT.A INCORRECT HANDLING OF EXCEPTIONS IN XTAPPADDINPUT | |
| 336 | |
| 337 The solution is to install X11.base.lib at version >=4.3.2.5. | |
| 392 | 338 |
| 373 | 339 *** On AIX, you get this compiler error message: |
| 340 | |
| 341 Processing include file ./XMenuInt.h | |
| 342 1501-106: (S) Include file X11/Xlib.h not found. | |
| 343 | |
| 344 This means your system was installed with only the X11 runtime i.d | |
| 345 libraries. You have to find your sipo (bootable tape) and install | |
| 346 X11Dev... with smit. | |
| 347 | |
| 348 *** On AIX 4.1.2, linker error messages such as | |
| 349 ld: 0711-212 SEVERE ERROR: Symbol .__quous, found in the global symbol table | |
| 350 of archive /usr/lib/libIM.a, was not defined in archive member shr.o. | |
| 351 | |
| 352 This is a problem in libIM.a. You can work around it by executing | |
| 353 these shell commands in the src subdirectory of the directory where | |
| 354 you build Emacs: | |
| 355 | |
| 356 cp /usr/lib/libIM.a . | |
| 357 chmod 664 libIM.a | |
| 358 ranlib libIM.a | |
| 359 | |
| 360 Then change -lIM to ./libIM.a in the command to link temacs (in | |
| 361 Makefile). | |
| 362 | |
| 363 *** Excessive optimization on AIX 4.2 can lead to compiler failure. | |
| 364 | |
| 365 Valdis.Kletnieks@vt.edu writes: | |
| 366 At least at the b34 level, and the latest-and-greatest IBM xlc | |
| 367 (3.1.4.4), there are problems with -O3. I haven't investigated | |
| 368 further. | |
| 369 | |
| 370 | |
| 371 ** SunOS/Solaris | |
| 1318 | 372 *** Don't use -O2 with gcc 2.8.1 and egcs 1.0 under SPARC architectures |
| 373 without also using `-fno-schedule-insns'. | |
| 374 | |
| 375 gcc will generate incorrect code otherwise, typically resulting in | |
| 376 crashes in the function skip-syntax-backward. | |
| 377 | |
| 378 *** Don't use gcc-2.95.2 with -mcpu=ultrasparc on Solaris 2.6. | |
| 379 | |
| 380 gcc will assume a 64-bit operating system, even though you've | |
| 381 merely told it to assume a 64-bit instruction set. | |
| 382 | |
| 454 | 383 *** Dumping error when using GNU binutils / GNU ld on a Sun. |
| 384 | |
| 385 Errors similar to the following: | |
| 386 | |
| 387 Dumping under the name xemacs unexec(): | |
| 388 dldump(/space/rpluim/xemacs-obj/src/xemacs): ld.so.1: ./temacs: | |
| 389 fatal: /space/rpluim/xemacs-obj/src/xemacs: unknown dynamic entry: | |
| 390 1879048176 | |
| 391 | |
| 392 are caused by using GNU ld. There are several workarounds available: | |
| 393 | |
| 394 In XEmacs 21.2 or later, configure using the new portable dumper | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
395 (--with-pdump). |
| 454 | 396 |
| 397 Alternatively, you can link using the Sun version of ld, which is | |
| 398 normally held in /usr/ccs/bin. This can be done by one of: | |
| 399 | |
| 400 - building gcc with these configure flags: | |
| 401 configure --with-ld=/usr/ccs/bin/ld --with-as=/usr/ccs/bin/as | |
| 402 | |
| 403 - adding -B/usr/ccs/bin/ to CFLAGS used to configure XEmacs | |
| 404 (Note: The trailing '/' there is significant.) | |
| 405 | |
| 406 - uninstalling GNU ld. | |
| 407 | |
| 408 The Solaris2 FAQ claims: | |
| 409 | |
| 410 When you install gcc, don't make the mistake of installing | |
| 411 GNU binutils or GNU libc, they are not as capable as their | |
| 412 counterparts you get with Solaris 2.x. | |
| 413 | |
| 373 | 414 *** Link failure when using acc on a Sun. |
| 415 | |
| 416 To use acc, you need additional options just before the libraries, such as | |
| 417 | |
| 418 /usr/lang/SC2.0.1/values-Xt.o -L/usr/lang/SC2.0.1/cg87 -L/usr/lang/SC2.0.1 | |
| 419 | |
| 420 and you need to add -lansi just before -lc. | |
| 421 | |
| 422 The precise file names depend on the compiler version, so we | |
| 423 cannot easily arrange to supply them. | |
| 424 | |
| 425 *** Problems finding X11 libraries on Solaris with Openwindows | |
| 426 | |
| 427 Some users have reported problems in this area. The reported solution | |
| 428 is to define the environment variable OPENWINHOME, even if you must set | |
| 429 it to `/usr/openwin'. | |
| 430 | |
| 431 *** Sed problems on Solaris 2.5 | |
| 432 | |
| 433 There have been reports of Sun sed truncating very lines in the | |
| 434 Makefile during configuration. The workaround is to use GNU sed or, | |
| 454 | 435 even better, think of a better way to generate Makefile, and send us a |
| 373 | 436 patch. :-) |
| 437 | |
| 438 *** On Solaris 2 I get undefined symbols from libcurses.a. | |
| 439 | |
| 440 You probably have /usr/ucblib/ on your LD_LIBRARY_PATH. Do the link with | |
| 441 LD_LIBRARY_PATH unset. Generally, avoid using any ucb* stuff when | |
| 442 building XEmacs. | |
| 443 | |
| 444 *** On Solaris 2 I cannot make alloc.o, glyphs.o or process.o. | |
| 445 | |
| 446 The SparcWorks C compiler may have difficulty building those modules | |
| 447 with optimization level -xO4. Try using only "-fast" optimization | |
| 448 for just those modules. (Or use gcc). | |
| 449 | |
| 450 *** Solaris 2.3 /bin/sh coredumps during configuration. | |
| 451 | |
| 452 This only occurs if you have LANG != C. This is a known bug with | |
| 453 /bin/sh fixed by installing Patch-ID# 101613-01. Or, you can use | |
| 1697 | 454 bash by setting the environment variable CONFIG_SHELL to /bin/bash |
| 455 | |
| 456 *** Solaris 2.x configure fails: ./config.status: test: argument expected | |
| 457 | |
| 458 This is a known bug with /bin/sh and /bin/test, i.e. they do not | |
| 459 support the XPG4 standard. You can use bash as a workaround or an | |
| 460 XPG4-compliant Bourne shell such as the Sun-supplied /usr/xpg4/bin/sh | |
| 461 by setting the environment variable CONFIG_SHELL to /usr/xpg4/bin/sh | |
| 373 | 462 |
| 463 *** On SunOS, you get linker errors | |
| 454 | 464 ld: Undefined symbol |
| 373 | 465 _get_wmShellWidgetClass |
| 466 _get_applicationShellWidgetClass | |
| 467 | |
| 468 The fix to this is to install patch 100573 for OpenWindows 3.0 | |
| 469 or link libXmu statically. | |
| 470 | |
| 471 *** On Sunos 4, you get the error ld: Undefined symbol __lib_version. | |
| 472 | |
| 473 This is the result of using cc or gcc with the shared library meant | |
| 474 for acc (the Sunpro compiler). Check your LD_LIBRARY_PATH and delete | |
| 475 /usr/lang/SC2.0.1 or some similar directory. | |
| 476 | |
| 477 *** Undefined symbols when linking on Sunos 4.1. | |
| 478 | |
| 479 If you get the undefined symbols _atowc _wcslen, _iswprint, _iswspace, | |
| 480 _iswcntrl, _wcscpy, and _wcsncpy, then you need to add -lXwchar after | |
| 481 -lXaw in the command that links temacs. | |
| 482 | |
| 483 This problem seems to arise only when the international language | |
| 484 extensions to X11R5 are installed. | |
| 485 | |
| 486 *** On a Sun running SunOS 4.1.1, you get this error message from GNU ld: | |
| 487 | |
| 454 | 488 /lib/libc.a(_Q_sub.o): Undefined symbol __Q_get_rp_rd referenced from text segment |
| 373 | 489 |
| 490 The problem is in the Sun shared C library, not in GNU ld. | |
| 491 | |
| 492 The solution is to install Patch-ID# 100267-03 from Sun. | |
| 493 | |
| 494 *** SunOS 4.1.2: undefined symbol _get_wmShellWidgetClass | |
| 495 | |
| 496 Apparently the version of libXmu.so.a that Sun ships is hosed: it's missing | |
| 454 | 497 some stuff that is in libXmu.a (the static version). Sun has a patch for |
| 373 | 498 this, but a workaround is to use the static version of libXmu, by changing |
| 499 the link command from "-lXmu" to "-Bstatic -lXmu -Bdynamic". If you have | |
| 500 OpenWindows 3.0, ask Sun for these patches: | |
| 501 100512-02 4.1.x OpenWindows 3.0 libXt Jumbo patch | |
| 502 100573-03 4.1.x OpenWindows 3.0 undefined symbols with shared libXmu | |
| 503 | |
| 504 *** Random other SunOS 4.1.[12] link errors. | |
| 505 | |
| 506 The X headers and libraries that Sun ships in /usr/{include,lib}/X11 are | |
| 507 broken. Use the ones in /usr/openwin/{include,lib} instead. | |
| 508 | |
| 509 ** Linux | |
| 1318 | 510 |
| 511 See also Intel Architecture General, above. | |
| 512 | |
| 513 *** egcs-1.1 on Alpha Linux | |
| 514 | |
| 515 There have been reports of egcs-1.1 not compiling XEmacs correctly on | |
| 516 Alpha Linux. There have also been reports that egcs-1.0.3a is O.K. | |
| 517 | |
| 373 | 518 *** Under Linux, you get "too many arguments to function `getpgrp'". |
| 519 | |
| 520 You have probably installed LessTiff under `/usr/local' and `libXm.so' | |
| 521 could not be found when linking `getpgrp()' test program, making XEmacs | |
| 522 think that `getpgrp()' takes an argument. Try adding `/usr/local/lib' | |
| 523 in `/etc/ld.so.conf' and run `ldconfig'. Then run XEmacs's `configure' | |
| 524 again. As with all problems of this type, reading the config.log file | |
| 525 generated from configure and seeing the log of how the test failed can | |
| 526 prove enlightening. | |
| 527 | |
| 528 *** `Error: No ExtNode to pop!' on Linux systems with Lesstif. | |
| 197 | 529 |
| 530 This error message has been observed with lesstif-0.75a. It does not | |
| 531 appear to cause any harm. | |
| 532 | |
| 373 | 533 *** xemacs: can't resolve symbol '__malloc_hook' |
| 534 | |
| 535 This is a Linux problem where you've compiled the XEmacs binary on a libc | |
| 536 5.4 with version higher than 5.4.19 and attempted to run the binary against | |
| 537 an earlier version. The solution is to upgrade your old library. | |
| 538 | |
| 539 ** IRIX | |
| 452 | 540 |
| 1098 | 541 *** More coredumping in Irix (6.5 known to be vulnerable) |
| 542 | |
| 543 No fix is known yet. Here's the best information we have: | |
| 544 | |
| 545 Valdis Kletnieks <Valdis.Kletnieks@vt.edu> writes: | |
| 546 | |
| 547 Were xemacs and [any 3rd party, locally-compiled] libraries [you use] | |
| 548 all compiled with the same ABI ( -o32, -n32, -64) and | |
| 549 mips2/mips3/mips4 flags, and are they appropriate for the machine in | |
| 550 question? I know the IP30 implies an Octane, so it should be an R10K | |
| 551 chipset and above such nonsense, but I've seen the most astoundingly | |
| 552 bizzare crashes when somebody managed to compile with -mips4 and get | |
| 553 it to run on an R4400 or R5K system. ;) | |
| 554 | |
| 555 Also, since you're using gcc, try re-running fixincludes and *then* | |
| 556 rebuilding xemacs and [any] libraries - mismatched headers can do that | |
| 557 sort of thing to you with little or no clue what's wrong (often you | |
| 558 get screwed when one routine does an malloc(sizeof(foo_struct)) and | |
| 559 passes the result to something that things foo_struct is a bit bigger, | |
| 560 trashing memory.... | |
| 561 | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
562 Here's typical crash backtrace. With --with-pdump, this occurs |
| 2648 | 563 usually at startup under X windows and xemacs -nw at least starts, while |
| 564 without --pdump a similar crash is observed during build. | |
| 1098 | 565 |
| 566 #0 0x0fa460b8 in kill () at regcomp.c:637 | |
| 567 637 regcomp.c: No such file or directory. | |
| 568 in regcomp.c | |
| 569 (gdb) where | |
| 570 #0 0x0fa460b8 in kill () at regcomp.c:637 | |
| 571 #1 0x10087f34 in fatal_error_signal () | |
| 572 (gdb) quit | |
| 573 | |
| 574 This is confusing because there is no such file in the XEmacs | |
| 575 distribution. This is seen on (at least) the following configurations: | |
| 576 | |
| 577 uname -a: IRIX64 oct202 6.5 01091821 IP30 | |
| 578 XEmacs 21.4.9 "Informed Management" configured for `mips-sgi-irix6.5'. | |
| 579 XEmacs 21.5-b9 "brussels sprouts" configured for `mips-sgi-irix6.5'. | |
| 580 | |
| 452 | 581 *** On Irix 6.5, the MIPSpro compiler gets an internal compiler error |
| 582 | |
| 583 The MIPSpro Compiler (at least version 7.2.1) can't seem to handle the | |
| 584 union type properly, and fails to compile src/glyphs.c. To avoid this | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
585 problem, always build --with-union-type=no (but that's the default, so |
| 452 | 586 you should only see this problem if you're an XEmacs maintainer). |
| 587 | |
| 373 | 588 *** Linking with -rpath on IRIX. |
| 124 | 589 |
| 590 Darrell Kindred <dkindred@cmu.edu> writes: | |
| 591 There are a couple of problems [with use of -rpath with Irix ld], though: | |
| 592 | |
| 593 1. The ld in IRIX 5.3 ignores all but the last -rpath | |
| 594 spec, so the patched configure spits out a warning | |
| 2648 | 595 if --x-libraries or --with-site-runtime-libraries are |
| 454 | 596 specified under irix 5.x, and it only adds -rpath |
| 2648 | 597 entries for the --with-site-runtime-libraries. This bug was |
| 124 | 598 fixed sometime between 5.3 and 6.2. |
| 599 | |
| 600 2. IRIX gcc 2.7.2 doesn't accept -rpath directly, so | |
| 601 it would have to be prefixed by -Xlinker or "-Wl,". | |
| 602 This would be fine, except that configure compiles with | |
| 603 ${CC-cc} $CFLAGS $LDFLAGS ... | |
| 604 rather than quoting $LDFLAGS with prefix-args, like | |
| 605 src/Makefile does. So if you specify --x-libraries | |
| 2648 | 606 or --with-site-runtime-libraries, you must use |
| 607 --with--gcc=no, or configure will fail. | |
| 124 | 608 |
| 373 | 609 *** On Irix 6.3, the SGI ld quits with segmentation fault when linking temacs |
| 207 | 610 |
| 611 This occurs if you use the SGI linker version 7.1. Installing the | |
| 612 patch SG0001872 fixes this problem. | |
| 197 | 613 |
| 373 | 614 *** On Irix 6.0, make tries (and fails) to build a program named unexelfsgi |
| 615 | |
| 616 A compiler bug inserts spaces into the string "unexelfsgi . o" | |
| 617 in src/Makefile. Edit src/Makefile, after configure is run, | |
| 618 find that string, and take out the spaces. | |
| 619 | |
| 620 Compiler fixes in Irix 6.0.1 should eliminate this problem. | |
| 124 | 621 |
| 373 | 622 *** On Irix 5.2, unexelfsgi.c can't find cmplrs/stsupport.h. |
| 124 | 623 |
| 373 | 624 The file cmplrs/stsupport.h was included in the wrong file set in the |
| 625 Irix 5.2 distribution. You can find it in the optional fileset | |
| 626 compiler_dev, or copy it from some other Irix 5.2 system. A kludgy | |
| 627 workaround is to change unexelfsgi.c to include sym.h instead of | |
| 628 syms.h. | |
| 124 | 629 |
| 373 | 630 *** Coredumping in Irix 6.2 |
| 124 | 631 |
| 373 | 632 Pete Forman <gsez020@compo.bedford.waii.com> writes: |
| 633 A problem noted by myself and others (I've lost the references) was | |
| 634 that XEmacs coredumped when the cut or copy toolbar buttons were | |
| 635 pressed. This has been fixed by loading the SGI patchset (Feb 98) | |
| 636 without having to recompile XEmacs. | |
| 124 | 637 |
| 373 | 638 My versions are XEmacs 20.3 (problem first noted in 19.15) and IRIX |
| 639 6.2, compiled using -n32. I'd guess that the relevant individual | |
| 640 patch was "SG0002580: multiple fixes for X libraries". SGI recommends | |
| 641 that the complete patch set be installed rather than parts of it. | |
| 124 | 642 |
| 373 | 643 ** Digital UNIX/OSF/VMS |
| 644 *** On Digital UNIX, the DEC C compiler might have a problem compiling | |
| 197 | 645 some files. |
| 124 | 646 |
| 647 In particular, src/extents.c and src/faces.c might cause the DEC C | |
| 648 compiler to abort. When this happens: cd src, compile the files by | |
| 649 hand, cd .., and redo the "make" command. When recompiling the files by | |
| 650 hand, use the old C compiler for the following versions of Digital UNIX: | |
| 651 - V3.n: Remove "-migrate" from the compile command. | |
| 652 - V4.n: Add "-oldc" to the compile command. | |
| 653 | |
| 197 | 654 A related compiler bug has been fixed by the DEC compiler team. The |
| 655 new versions of the compiler should run fine. | |
| 126 | 656 |
| 373 | 657 *** Under some versions of OSF XEmacs runs fine if built without |
| 658 optimization but will crash randomly if built with optimization. | |
| 659 | |
| 660 Using 'cc -g' is not sufficient to eliminate all optimization. Try | |
| 661 'cc -g -O0' instead. | |
| 662 | |
| 663 *** Compilation errors on VMS. | |
| 664 | |
| 665 Sorry, XEmacs does not work under VMS. You might consider working on | |
| 666 the port if you really want to have XEmacs work under VMS. | |
| 667 | |
| 668 ** HP-UX | |
| 669 *** On HPUX, the HP C compiler might have a problem compiling some files | |
| 278 | 670 with optimization. |
| 124 | 671 |
| 672 Richard Cognot <cognot@ensg.u-nancy.fr> writes: | |
| 673 | |
| 674 Had to drop once again to level 2 optimization, at least to | |
| 675 compile lstream.c. Otherwise, I get a "variable is void: \if" | |
| 676 problem while dumping (this is a problem I already reported | |
| 677 with vanilla hpux 10.01 and 9.07, which went away after | |
| 678 applying patches for the C compiler). Trouble is I still | |
| 679 haven't found the same patch for hpux 10.10, and I don't | |
| 680 remember the patch numbers. I think potential XEmacs builders | |
| 681 on HP should be warned about this. | |
| 682 | |
| 373 | 683 *** I don't have `xmkmf' and `imake' on my HP. |
| 124 | 684 |
| 304 | 685 You can get these standard X tools by anonymous FTP to |
| 686 hpcvaaz.cv.hp.com. Essentially all X programs need these. | |
| 124 | 687 |
| 373 | 688 *** On HP-UX, problems with make |
| 278 | 689 |
| 442 | 690 Marcus Thiessel <marcus@xemacs.org> |
| 278 | 691 |
| 304 | 692 Some releases of XEmacs (e.g. 20.4) require GNU make to build |
| 693 successfully. You don't need GNU make when building 21.x. | |
| 278 | 694 |
| 373 | 695 *** On HP-UX 9.05 XEmacs won't compile or coredump during the build. |
| 278 | 696 |
| 442 | 697 Marcus Thiessel <marcus@xemacs.org> |
| 278 | 698 |
| 699 This might be a sed problem. For your own safety make sure to use | |
| 700 GNU sed while dumping XEmacs. | |
| 701 | |
| 442 | 702 |
| 373 | 703 ** SCO OpenServer |
| 704 *** Native cc on SCO OpenServer 5 is now OK. Icc may still throw you | |
| 197 | 705 a curve. Here is what Robert Lipe <robertl@arnet.com> says: |
| 124 | 706 |
| 454 | 707 Unlike XEmacs 19.13, building with the native cc on SCO OpenServer 5 |
| 124 | 708 now produces a functional binary. I will typically build this |
| 709 configuration for COFF with: | |
| 710 | |
| 197 | 711 /path_to_xemacs_source/configure --with-gcc=no \ |
| 2648 | 712 --with-site-includes=/usr/local/include \ |
| 713 --with-site-libraries=/usr/local/lib \ | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
714 --with-xpm --with-xface --with-sound=nas |
| 124 | 715 |
| 454 | 716 This version now supports ELF builds. I highly recommend this to |
| 717 reduce the in-core footprint of XEmacs. This is now how I compile | |
| 124 | 718 all my test releases. Build it like this: |
| 719 | |
| 720 /path_to_XEmacs_source/configure --with-gcc=no \ | |
| 2648 | 721 --with-site-includes=/usr/local/include |
| 722 --with-site-libraries=/usr/local/lib \ | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
723 --with-xpm --with-xface --with-sound=nas --with-dynamic |
| 124 | 724 |
| 454 | 725 The compiler known as icc [ supplied with the OpenServer 5 Development |
| 124 | 726 System ] generates a working binary, but it takes forever to generate |
| 727 XEmacs. ICC also whines more about the code than /bin/cc does. I do | |
| 728 believe all its whining is legitimate, however. Note that you do | |
| 729 have to 'cd src ; make LD=icc' to avoid linker errors. | |
| 730 | |
| 731 The way I handle the build procedure is: | |
| 732 | |
| 733 /path_to_XEmacs_source/configure --with-gcc=no \ | |
| 2648 | 734 --with-site-includes=/usr/local/include \ |
| 735 --with-site-libraries=/usr/local/lib \ | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
736 --with-xpm --with-xface --with-sound=nas --with-dynamic \ |
| 2648 | 737 --with-compiler="icc" |
| 124 | 738 |
| 454 | 739 NOTE I have the xpm, xface, and audio libraries and includes in |
| 124 | 740 /usr/local/lib, /usr/local/include. If you don't have these, |
| 741 don't include the "--with-*" arguments in any of my examples. | |
| 742 | |
| 454 | 743 In previous versions of XEmacs, you had to override the defaults while |
| 124 | 744 compiling font-lock.o and extents.o when building with icc. This seems |
| 745 to no longer be true, but I'm including this old information in case it | |
| 746 resurfaces. The process I used was: | |
| 747 | |
| 454 | 748 make -k |
| 749 [ procure pizza, beer, repeat ] | |
| 124 | 750 cd src |
| 751 make CC="icc -W0,-mP1COPT_max_tree_size=3000" font-lock.o extents.o | |
| 752 make LD=icc | |
| 753 | |
| 454 | 754 If you want sound support, get the tls566 supplement from |
| 755 ftp.sco.com:/TLS or any of its mirrors. It works just groovy | |
| 124 | 756 with XEmacs. |
| 757 | |
| 758 The M-x manual-entry is known not to work. If you know Lisp and would | |
| 759 like help in making it work, e-mail me at <robertl@dgii.com>. | |
| 760 (UNCHECKED for 19.15 -- it might work). | |
| 761 | |
| 454 | 762 In earlier releases, gnuserv/gnuclient/gnudoit would open a frame |
| 124 | 763 just fine, but the client would lock up and the server would |
| 454 | 764 terminate when you used C-x # to close the frame. This is now |
| 124 | 765 fixed in XEmacs. |
| 766 | |
| 767 In etc/ there are two files of note. emacskeys.sco and emacsstrs.sco. | |
| 768 The comments at the top of emacskeys.sco describe its function, and | |
| 769 the emacstrs.sco is a suitable candidate for /usr/lib/keyboard/strings | |
| 770 to take advantage of the keyboard map in emacskeys.sco. | |
| 771 | |
| 373 | 772 Note: Much of the above entry is probably not valid for XEmacs 21.0 |
| 207 | 773 and later. |
| 197 | 774 |
| 1332 | 775 ** Windows |
| 776 | |
| 1441 | 777 *** XEmacs complains "No such file or directory, diff" |
| 778 | |
| 779 or "ispell" or other commands that seem related to whatever you just | |
| 780 tried to do. | |
| 781 | |
| 782 There are a large number of common (in the sense that "everyone has | |
| 783 these, really") Unix utilities that are not provided with XEmacs. The | |
| 784 GNU Project's implementations are available for Windows in the the | |
| 785 Cygwin distribution (http://www.cygwin.com/), which also provides a | |
| 786 complete Unix emulation environment (and thus makes ports of Unix | |
| 787 utilities nearly trivial). Another implementation is that from MinGW | |
| 788 (http://www.mingw.org/msys.shtml). | |
| 789 | |
| 1332 | 790 *** Weird crashes in pdump load or shortly after pdump load. |
| 791 | |
| 792 This can happen with incremental linking. Check if you have set | |
| 793 SUPPORT_EDIT_AND_CONTINUE to non-zero in config.inc, which must allow | |
| 794 incremental linking to be enabled (otherwise it's disabled). Either turn | |
| 795 this off, execute `nmake -f xemacs.mak clean', or manually remove | |
| 796 `temacs.exe' and `xemacs.exe'. | |
| 797 | |
| 392 | 798 ** Cygwin |
| 524 | 799 |
| 1318 | 800 See also Intel Architecture General, above. |
| 801 | |
| 802 *** Signal 11 when building or running a dumped XEmacs. | |
| 803 | |
|
4877
1a9c94ba117c
add issue about building on Cygwin with C++, NEWGC and optimization
Ben Wing <ben@xemacs.org>
parents:
4842
diff
changeset
|
804 Possibility #1: |
|
1a9c94ba117c
add issue about building on Cygwin with C++, NEWGC and optimization
Ben Wing <ben@xemacs.org>
parents:
4842
diff
changeset
|
805 |
| 1318 | 806 This appears to happen when using the traditional dumping mechanism and |
| 807 the system malloc. Andy Piper writes: | |
| 808 | |
| 1332 | 809 Traditional dumping on Cygwin relies on using gmalloc (there are specific |
| 1318 | 810 hacks in our version of gmalloc to support this), I suspect using sysmalloc |
| 811 is the problem. | |
| 812 | |
| 813 Try configuring with pdump or without system malloc. | |
| 814 | |
|
4877
1a9c94ba117c
add issue about building on Cygwin with C++, NEWGC and optimization
Ben Wing <ben@xemacs.org>
parents:
4842
diff
changeset
|
815 Possibility #2: |
|
1a9c94ba117c
add issue about building on Cygwin with C++, NEWGC and optimization
Ben Wing <ben@xemacs.org>
parents:
4842
diff
changeset
|
816 |
|
1a9c94ba117c
add issue about building on Cygwin with C++, NEWGC and optimization
Ben Wing <ben@xemacs.org>
parents:
4842
diff
changeset
|
817 Crashes when running a dumped XEmacs 21.5.29 have been observed circa |
|
1a9c94ba117c
add issue about building on Cygwin with C++, NEWGC and optimization
Ben Wing <ben@xemacs.org>
parents:
4842
diff
changeset
|
818 January, 2010 in Cygwin 1.7 when configuring with C++, NEWGC and optimization |
|
1a9c94ba117c
add issue about building on Cygwin with C++, NEWGC and optimization
Ben Wing <ben@xemacs.org>
parents:
4842
diff
changeset
|
819 (i.e. `--with-compiler=gcc --with-xemacs-compiler=g++ --with-mc-alloc |
|
1a9c94ba117c
add issue about building on Cygwin with C++, NEWGC and optimization
Ben Wing <ben@xemacs.org>
parents:
4842
diff
changeset
|
820 --with-optimization'). Turning any one of these off produces a working build. |
|
1a9c94ba117c
add issue about building on Cygwin with C++, NEWGC and optimization
Ben Wing <ben@xemacs.org>
parents:
4842
diff
changeset
|
821 |
| 524 | 822 *** Syntax errors running configure scripts, make failing with exit code 127 |
| 823 in inexplicable situations, etc. | |
| 392 | 824 |
| 1332 | 825 [[ This may be because you are using the default Cygwin shell, under old |
| 826 versions of Cygwin. The default Cygwin shell (/bin/sh.exe) is ash, which | |
| 827 appears to work in most circumstances but has some weird failure modes. | |
| 828 You may need to replace the symlink with bash.exe. ]] This doesn't appear | |
| 829 to affect Cygwin any longer, and /bin/sh.exe is no longer a symlink in | |
| 830 any case. | |
| 392 | 831 |
| 524 | 832 *** Lots of compile errors, esp. on lines containing macro definitions |
| 833 terminated by backslashes. | |
| 392 | 834 |
| 524 | 835 Your partition holding the source files is mounted binary. It needs |
| 836 to be mounted text. (This will not screw up any binary files because | |
| 837 the Cygwin utilities specify explicitly whether they want binary or | |
| 838 text mode when working with source vs. binary files, which overrides | |
| 839 the mount type.) To fix this, you just need to run the appropriate | |
| 840 mount command once -- afterwards, the settings are remembered in the | |
| 841 registry. | |
| 392 | 842 |
| 524 | 843 *** Errors from make like /c:not found. |
| 392 | 844 |
| 524 | 845 Make sure you set the environment variable MAKE_MODE to UNIX in your |
| 846 .bashrc, Control Panel (Windows 2000/NT), or AUTOEXEC.BAT (Windows | |
| 847 98/95). | |
| 392 | 848 |
| 849 *** The info files will not build. | |
| 850 | |
| 1332 | 851 makeinfo that ships with old versions of Cygwin doesn't work. |
| 852 Upgrade to the latest Cygwin version. | |
| 392 | 853 |
| 524 | 854 *** XEmacs hangs while attempting to rebuild the .elc files. |
| 392 | 855 |
| 524 | 856 Check to make sure you're not configuring with rel-alloc. The relocating |
| 857 allocator does not currently work under Cygwin due to bugs in Cygwin's | |
| 858 mmap(). | |
| 392 | 859 |
| 524 | 860 *** Trying to build with X, but X11 not detected. |
| 861 | |
| 862 This is usually because xmkmf is not in your path or because you are | |
| 1332 | 863 using the default Cygwin shell. (See above.) |
| 333 | 864 |
| 865 | |
| 373 | 866 * Problems with running XEmacs |
| 867 ============================== | |
| 868 ** General | |
| 1332 | 869 |
| 870 *** Changes made to .el files do not take effect. | |
| 871 | |
| 872 You may have forgotten to recompile them into .elc files. Then the | |
| 873 old .elc files will be loaded, and your changes will not be seen. To | |
| 874 fix this, do `M-x byte-recompile-directory' and specify the directory | |
| 875 that contains the Lisp files. | |
| 876 | |
| 877 Note that you will get a warning when loading a .elc file that is | |
| 878 older than the corresponding .el file. | |
| 879 | |
| 880 *** VM appears to hang in large folders. | |
| 881 | |
| 882 This is normal (trust us) when upgrading to VM-6.22 from earlier | |
| 883 versions. Let VM finish what it is doing and all will be well. | |
| 1042 | 884 |
| 892 | 885 *** Starting with 21.4.x, killing text is absurdly slow. |
| 886 | |
| 887 See FAQ Q3.10.6. Should be available on the web near | |
| 888 http://www.xemacs.org/faq/xemacs-faq.html#SEC160. | |
| 889 | |
| 835 | 890 *** Whenever I try to retrieve a remote file, I have problems. |
| 891 | |
| 892 A typical error: FTP Error: USER request failed; 500 AUTH not understood. | |
| 893 Thanks to giacomo boffi <giacomo.boffi@polimi.it> on comp.emacs.xemacs: | |
| 894 | |
| 895 tell your ftp client to not attempt AUTH authentication (or do not | |
| 896 use FTP servers that don't understand AUTH) | |
| 897 | |
| 898 and notes that you need to add an element (often "-u") to | |
| 899 `efs-ftp-program-args'. Use M-x customize-variable, and verify the | |
| 900 needed flag with `man ftp' or other local documentation. | |
| 901 | |
| 464 | 902 *** gnuserv is running, some clients can connect, but others cannot. |
| 903 | |
| 904 The code in gnuslib.c respects the value of TMPDIR. If the server and | |
| 905 the client have different values in their environment, you lose. | |
| 906 One program known to set TMPDIR and manifest this problem is exmh. | |
| 907 You can defeat the use of TMPDIR by unsetting USE_TMPDIR at the top of | |
| 908 gnuserv.h at build time. | |
| 909 | |
| 1332 | 910 ** General Unix |
| 124 | 911 |
| 373 | 912 *** You type Control-H (Backspace) expecting to delete characters. |
| 124 | 913 |
| 914 Emacs has traditionally used Control-H for help; unfortunately this | |
| 424 | 915 interferes with its use as Backspace on TTY's. As of XEmacs 21, |
| 916 XEmacs looks at the "erase" setting of TTY structures and maps C-h to | |
| 917 backspace when erase is set to C-h. This is sort of a special hack, | |
| 918 but it makes it possible for you to use the standard: | |
| 919 | |
| 920 stty erase ^H | |
| 355 | 921 |
| 424 | 922 to get your backspace key to erase characters. The erase setting is |
| 923 recorded in the Lisp variable `tty-erase-char', which you can use to | |
| 924 tune the settings in your .emacs. | |
| 124 | 925 |
| 424 | 926 A major drawback of this is that when C-h becomes backspace, it no |
| 927 longer invokes help. In that case, you need to use f1 for help, or | |
| 928 bind another key. An example of the latter is the following code, | |
| 929 which moves help to Meta-? (ESC ?): | |
| 124 | 930 |
| 424 | 931 (global-set-key "\M-?" 'help-command) |
| 124 | 932 |
| 1332 | 933 *** At startup I get a warning on stderr about missing charsets: |
| 934 | |
| 935 Warning: Missing charsets in String to FontSet conversion | |
| 936 | |
| 937 You need to specify appropriate charsets for your locale (usually the | |
| 938 value of the LANG environment variable) in .Xresources. See | |
| 939 etc/Emacs.ad for the relevant resources (mostly menubar fonts and | |
| 940 fontsets). Do not edit this file, it's purely informative. | |
| 941 | |
| 942 If you have no satisfactory fonts for iso-8859-1, XEmacs will crash. | |
| 943 | |
| 944 It looks like XFree86 4.x (the usual server on Linux and *BSD) has | |
| 945 some braindamage where .UTF-8 locales will always generate this | |
| 946 message, because the XFree86 (font)server doesn't know that UTF-8 will | |
| 947 use the ISO10646-1 font registry (or a Cmap or something). | |
| 948 | |
| 949 If you are not using a .UTF-8 locale and see this warning for a | |
| 950 character set not listed in the default in Emacs.ad, please let | |
| 951 xemacs-beta@xemacs.org know about it, so we can add fonts to the | |
| 952 appropriate fontsets and stifle this warning. (Unfortunately it's | |
| 953 buried in Xlib, so we can't easily get rid of it otherwise.) | |
| 954 | |
| 373 | 955 *** Mail agents (VM, Gnus, rmail) cannot get new mail |
| 197 | 956 |
| 957 rmail and VM get new mail from /usr/spool/mail/$USER using a program | |
| 958 called `movemail'. This program interlocks with /bin/mail using the | |
| 959 protocol defined by /bin/mail. | |
| 960 | |
| 961 There are two different protocols in general use. One of them uses | |
| 962 the `flock' system call. The other involves creating a lock file; | |
| 963 `movemail' must be able to write in /usr/spool/mail in order to do | |
| 964 this. You control which one is used by defining, or not defining, the | |
| 965 macro MAIL_USE_FLOCK in config.h or the m- or s- file it includes. IF | |
| 966 YOU DON'T USE THE FORM OF INTERLOCKING THAT IS NORMAL ON YOUR SYSTEM, | |
| 967 YOU CAN LOSE MAIL! | |
| 968 | |
| 969 If your system uses the lock file protocol, and fascist restrictions | |
| 970 prevent ordinary users from writing the lock files in /usr/spool/mail, | |
| 971 you may need to make `movemail' setgid to a suitable group such as | |
| 972 `mail'. To do this, use the following commands (as root) after doing | |
| 973 the make install. | |
| 974 | |
| 975 chgrp mail movemail | |
| 976 chmod 2755 movemail | |
| 977 | |
| 978 Installation normally copies movemail from the build directory to an | |
| 979 installation directory which is usually under /usr/local/lib. The | |
| 980 installed copy of movemail is usually in the directory | |
| 981 /usr/local/lib/emacs/VERSION/TARGET. You must change the group and | |
| 982 mode of the installed copy; changing the group and mode of the build | |
| 983 directory copy is ineffective. | |
| 984 | |
| 373 | 985 *** Things which should be bold or italic (such as the initial |
| 197 | 986 copyright notice) are not. |
| 0 | 987 |
| 197 | 988 The fonts of the "bold" and "italic" faces are generated from the font |
| 989 of the "default" face; in this way, your bold and italic fonts will | |
| 990 have the appropriate size and family. However, emacs can only be | |
| 991 clever in this way if you have specified the default font using the | |
| 992 XLFD (X Logical Font Description) format, which looks like | |
| 0 | 993 |
| 994 *-courier-medium-r-*-*-*-120-*-*-*-*-*-* | |
| 995 | |
| 197 | 996 if you use any of the other, less strict font name formats, some of |
| 997 which look like: | |
| 998 | |
| 0 | 999 lucidasanstypewriter-12 |
| 1000 and fixed | |
| 1001 and 9x13 | |
| 1002 | |
| 1003 then emacs won't be able to guess the names of the "bold" and "italic" | |
| 1004 versions. All X fonts can be referred to via XLFD-style names, so you | |
| 1005 should use those forms. See the man pages for X(1), xlsfonts(1), and | |
| 1006 xfontsel(1). | |
| 1007 | |
| 373 | 1008 *** The dumped Emacs crashes when run, trying to write pure data. |
| 0 | 1009 |
| 1010 Two causes have been seen for such problems. | |
| 1011 | |
| 1012 1) On a system where getpagesize is not a system call, it is defined | |
| 1013 as a macro. If the definition (in both unexec.c and malloc.c) is wrong, | |
| 1014 it can cause problems like this. You might be able to find the correct | |
| 1015 value in the man page for a.out (5). | |
| 1016 | |
| 1017 2) Some systems allocate variables declared static among the | |
| 1018 initialized variables. Emacs makes all initialized variables in most | |
| 1019 of its files pure after dumping, but the variables declared static and | |
| 1020 not initialized are not supposed to be pure. On these systems you | |
| 1021 may need to add "#define static" to the m- or the s- file. | |
| 1022 | |
| 373 | 1023 *** Reading and writing files is very very slow. |
| 0 | 1024 |
| 1025 Try evaluating the form (setq lock-directory nil) and see if that helps. | |
| 1026 There is a problem with file-locking on some systems (possibly related | |
| 454 | 1027 to NFS) that I don't understand. Please send mail to the address |
| 1332 | 1028 xemacs-beta@xemacs.org if you figure this one out. |
| 0 | 1029 |
| 373 | 1030 *** When emacs starts up, I get lots of warnings about unknown keysyms. |
| 124 | 1031 |
| 1032 If you are running the prebuilt binaries, the Motif library expects to find | |
| 1033 certain thing in the XKeysymDB file. This file is normally in /usr/lib/X11/ | |
| 1034 or in /usr/openwin/lib/. If you keep yours in a different place, set the | |
| 454 | 1035 environment variable $XKEYSYMDB to point to it before starting emacs. If |
| 1036 you still have the problem after doing that, perhaps your version of X is | |
| 2536 | 1037 too old. There is a copy of the MIT X11R6 XKeysymDB file in the emacs `etc' |
| 124 | 1038 directory. Try using that one. |
| 1039 | |
| 3406 | 1040 *** Lots of warnings generated when displaying via ssh X forwarding. |
| 1041 | |
| 1042 If you are seeing a significant number of X11 warnings (in particular | |
| 1043 BadWindow errors) when using XEmacs via ssh X forwarding try using a | |
| 1044 trusted x11 connection instead (for openssh, use -Y instead of -X). | |
| 1045 | |
| 373 | 1046 *** My X resources used to work, and now some of them are being ignored. |
| 0 | 1047 |
| 124 | 1048 Check the resources in .../etc/Emacs.ad (which is the same as the file |
| 1389 | 1049 sample.Xresources). Perhaps some of the default resources built in to |
| 124 | 1050 emacs are now overriding your existing resources. Copy and edit the |
| 1051 resources in Emacs.ad as necessary. | |
| 1052 | |
| 373 | 1053 *** I have focus problems when I use `M-o' to switch to another screen |
| 197 | 1054 without using the mouse. |
| 124 | 1055 |
| 197 | 1056 The focus issues with a program like XEmacs, which has multiple |
| 1057 homogeneous top-level windows, are very complicated, and as a result, | |
| 1058 most window managers don't implement them correctly. | |
| 0 | 1059 |
| 124 | 1060 The R4/R5 version of twm (and all of its descendants) had buggy focus |
| 197 | 1061 handling. Sufficiently recent versions of tvtwm have been fixed. In |
| 1062 addition, if you're using twm, make sure you have not specified | |
| 1063 "NoTitleFocus" in your .tvtwmrc file. The very nature of this option | |
| 1064 makes twm do some illegal focus tricks, even with the patch. | |
| 0 | 1065 |
| 197 | 1066 It is known that olwm and olvwm are buggy, and in different ways. If |
| 1067 you're using click-to-type mode, try using point-to-type, or vice | |
| 1068 versa. | |
| 0 | 1069 |
| 197 | 1070 In older versions of NCDwm, one could not even type at XEmacs windows. |
| 1071 This has been fixed in newer versions (2.4.3, and possibly earlier). | |
| 0 | 1072 |
| 197 | 1073 (Many people suggest that XEmacs should warp the mouse when focusing |
| 1074 on another screen in point-to-type mode. This is not ICCCM-compliant | |
| 1075 behavior. Implementing such policy is the responsibility of the | |
| 1076 window manager itself, it is not legal for a client to do this.) | |
| 0 | 1077 |
| 373 | 1078 *** Emacs spontaneously displays "I-search: " at the bottom of the screen. |
| 0 | 1079 |
| 1080 This means that Control-S/Control-Q (XON/XOFF) "flow control" is being | |
| 1081 used. C-s/C-q flow control is bad for Emacs editors because it takes | |
| 1082 away C-s and C-q as user commands. Since editors do not output long | |
| 1083 streams of text without user commands, there is no need for a | |
| 1084 user-issuable "stop output" command in an editor; therefore, a | |
| 1085 properly designed flow control mechanism would transmit all possible | |
| 1086 input characters without interference. Designing such a mechanism is | |
| 1087 easy, for a person with at least half a brain. | |
| 1088 | |
| 1089 There are three possible reasons why flow control could be taking place: | |
| 1090 | |
| 1091 1) Terminal has not been told to disable flow control | |
| 1092 2) Insufficient padding for the terminal in use | |
| 1093 3) Some sort of terminal concentrator or line switch is responsible | |
| 1094 | |
| 1095 First of all, many terminals have a set-up mode which controls whether | |
| 1096 they generate XON/XOFF flow control characters. This must be set to | |
| 1097 "no XON/XOFF" in order for Emacs to work. Sometimes there is an | |
| 1098 escape sequence that the computer can send to turn flow control off | |
| 1099 and on. If so, perhaps the termcap `ti' string should turn flow | |
| 1100 control off, and the `te' string should turn it on. | |
| 1101 | |
| 1102 Once the terminal has been told "no flow control", you may find it | |
| 1103 needs more padding. The amount of padding Emacs sends is controlled | |
| 1104 by the termcap entry for the terminal in use, and by the output baud | |
| 1105 rate as known by the kernel. The shell command `stty' will print | |
| 1106 your output baud rate; `stty' with suitable arguments will set it if | |
| 1107 it is wrong. Setting to a higher speed causes increased padding. If | |
| 1108 the results are wrong for the correct speed, there is probably a | |
| 1109 problem in the termcap entry. You must speak to a local Unix wizard | |
| 1110 to fix this. Perhaps you are just using the wrong terminal type. | |
| 1111 | |
| 1112 For terminals that lack a "no flow control" mode, sometimes just | |
| 1113 giving lots of padding will prevent actual generation of flow control | |
| 1114 codes. You might as well try it. | |
| 1115 | |
| 1116 If you are really unlucky, your terminal is connected to the computer | |
| 1117 through a concentrator which sends XON/XOFF flow control to the | |
| 1118 computer, or it insists on sending flow control itself no matter how | |
| 1119 much padding you give it. Unless you can figure out how to turn flow | |
| 1120 control off on this concentrator (again, refer to your local wizard), | |
| 1121 you are screwed! You should have the terminal or concentrator | |
| 1122 replaced with a properly designed one. In the mean time, some drastic | |
| 1123 measures can make Emacs semi-work. | |
| 1124 | |
| 1125 You can make Emacs ignore C-s and C-q and let the operating system | |
| 1126 handle them. To do this on a per-session basis, just type M-x | |
| 1127 enable-flow-control RET. You will see a message that C-\ and C-^ are | |
| 1128 now translated to C-s and C-q. (Use the same command M-x | |
| 1129 enable-flow-control to turn *off* this special mode. It toggles flow | |
| 1130 control handling.) | |
| 1131 | |
| 1132 If C-\ and C-^ are inconvenient for you (for example, if one of them | |
| 1133 is the escape character of your terminal concentrator), you can choose | |
| 1134 other characters by setting the variables flow-control-c-s-replacement | |
| 1135 and flow-control-c-q-replacement. But choose carefully, since all | |
| 1136 other control characters are already used by emacs. | |
| 1137 | |
| 1138 IMPORTANT: if you type C-s by accident while flow control is enabled, | |
| 1139 Emacs output will freeze, and you will have to remember to type C-q in | |
| 1140 order to continue. | |
| 1141 | |
| 1142 If you work in an environment where a majority of terminals of a | |
| 1143 certain type are flow control hobbled, you can use the function | |
| 1144 `enable-flow-control-on' to turn on this flow control avoidance scheme | |
| 1145 automatically. Here is an example: | |
| 1146 | |
| 1147 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131") | |
| 1148 | |
| 1149 If this isn't quite correct (e.g. you have a mixture of flow-control hobbled | |
| 1150 and good vt200 terminals), you can still run enable-flow-control | |
| 1151 manually. | |
| 1152 | |
| 1153 I have no intention of ever redesigning the Emacs command set for the | |
| 1154 assumption that terminals use C-s/C-q flow control. XON/XOFF flow | |
| 1155 control technique is a bad design, and terminals that need it are bad | |
| 1156 merchandise and should not be purchased. Now that X is becoming | |
| 1157 widespread, XON/XOFF seems to be on the way out. If you can get some | |
| 1158 use out of GNU Emacs on inferior terminals, more power to you, but I | |
| 1159 will not make Emacs worse for properly designed systems for the sake | |
| 1160 of inferior systems. | |
| 1161 | |
| 373 | 1162 *** Control-S and Control-Q commands are ignored completely. |
| 0 | 1163 |
| 1164 For some reason, your system is using brain-damaged C-s/C-q flow | |
| 1165 control despite Emacs's attempts to turn it off. Perhaps your | |
| 1166 terminal is connected to the computer through a concentrator | |
| 1167 that wants to use flow control. | |
| 1168 | |
| 1169 You should first try to tell the concentrator not to use flow control. | |
| 1170 If you succeed in this, try making the terminal work without | |
| 1171 flow control, as described in the preceding section. | |
| 1172 | |
| 1173 If that line of approach is not successful, map some other characters | |
| 1174 into C-s and C-q using keyboard-translate-table. The example above | |
| 1175 shows how to do this with C-^ and C-\. | |
| 1176 | |
| 373 | 1177 *** Control-S and Control-Q commands are ignored completely on a net |
| 197 | 1178 connection. |
| 0 | 1179 |
| 1180 Some versions of rlogin (and possibly telnet) do not pass flow | |
| 1181 control characters to the remote system to which they connect. | |
| 1182 On such systems, emacs on the remote system cannot disable flow | |
| 1183 control on the local system. | |
| 1184 | |
| 1185 One way to cure this is to disable flow control on the local host | |
| 1186 (the one running rlogin, not the one running rlogind) using the | |
| 1187 stty command, before starting the rlogin process. On many systems, | |
| 120 | 1188 `stty start u stop u' will do this. |
| 0 | 1189 |
| 1190 Some versions of tcsh will prevent even this from working. One way | |
| 1191 around this is to start another shell before starting rlogin, and | |
| 1192 issue the stty command to disable flow control from that shell. | |
| 1193 | |
| 1194 If none of these methods work, the best solution is to type | |
| 120 | 1195 `M-x enable-flow-control' at the beginning of your emacs session, or |
| 0 | 1196 if you expect the problem to continue, add a line such as the |
| 1197 following to your .emacs (on the host running rlogind): | |
| 1198 | |
| 1199 (enable-flow-control-on "vt200" "vt300" "vt101" "vt131") | |
| 1200 | |
| 1201 See the entry about spontaneous display of I-search (above) for more | |
| 1202 info. | |
| 1203 | |
| 373 | 1204 *** TTY redisplay is slow. |
| 197 | 1205 |
| 1206 XEmacs has fairly new TTY redisplay support (beginning from 19.12), | |
| 1207 which doesn't include some basic TTY optimizations -- like using | |
| 1208 scrolling regions to move around blocks of text. This is why | |
| 454 | 1209 redisplay on the traditional terminals, or over slow lines can be very |
| 197 | 1210 slow. |
| 1211 | |
| 1212 If you are interested in fixing this, please let us know at | |
| 1332 | 1213 <xemacs-beta@xemacs.org>. |
| 197 | 1214 |
| 373 | 1215 *** Screen is updated wrong, but only on one kind of terminal. |
| 0 | 1216 |
| 120 | 1217 This could mean that the termcap entry you are using for that terminal |
| 1218 is wrong, or it could mean that Emacs has a bug handing the | |
| 1219 combination of features specified for that terminal. | |
| 0 | 1220 |
| 1221 The first step in tracking this down is to record what characters | |
| 1222 Emacs is sending to the terminal. Execute the Lisp expression | |
| 120 | 1223 (open-termscript "./emacs-script") to make Emacs write all terminal |
| 1224 output into the file ~/emacs-script as well; then do what makes the | |
| 1225 screen update wrong, and look at the file and decode the characters | |
| 1226 using the manual for the terminal. There are several possibilities: | |
| 0 | 1227 |
| 1228 1) The characters sent are correct, according to the terminal manual. | |
| 1229 | |
| 1230 In this case, there is no obvious bug in Emacs, and most likely you | |
| 1231 need more padding, or possibly the terminal manual is wrong. | |
| 1232 | |
| 120 | 1233 2) The characters sent are incorrect, due to an obscure aspect of the |
| 1234 terminal behavior not described in an obvious way by termcap. | |
| 0 | 1235 |
| 120 | 1236 This case is hard. It will be necessary to think of a way for Emacs |
| 1237 to distinguish between terminals with this kind of behavior and other | |
| 1238 terminals that behave subtly differently but are classified the same | |
| 1239 by termcap; or else find an algorithm for Emacs to use that avoids the | |
| 1240 difference. Such changes must be tested on many kinds of terminals. | |
| 0 | 1241 |
| 1242 3) The termcap entry is wrong. | |
| 1243 | |
| 120 | 1244 See the file etc/TERMS for information on changes that are known to be |
| 1245 needed in commonly used termcap entries for certain terminals. | |
| 0 | 1246 |
| 120 | 1247 4) The characters sent are incorrect, and clearly cannot be right for |
| 1248 any terminal with the termcap entry you were using. | |
| 0 | 1249 |
| 120 | 1250 This is unambiguously an Emacs bug, and can probably be fixed in |
| 197 | 1251 termcap.c, terminfo.c, tparam.c, cm.c, redisplay-tty.c, |
| 1252 redisplay-output.c, or redisplay.c. | |
| 0 | 1253 |
| 373 | 1254 *** My buffers are full of \000 characters or otherwise corrupt. |
| 1255 | |
| 1256 Some compilers have trouble with gmalloc.c and ralloc.c; try recompiling | |
| 1257 without optimization. If that doesn't work, try recompiling with | |
| 1258 SYSTEM_MALLOC defined, and/or with REL_ALLOC undefined. | |
| 1259 | |
| 1389 | 1260 *** A position you specified in .Xresources is ignored, using twm. |
| 373 | 1261 |
| 1262 twm normally ignores "program-specified" positions. | |
| 1263 You can tell it to obey them with this command in your `.twmrc' file: | |
| 1264 | |
| 1265 UsePPosition "on" #allow clents to request a position | |
| 1266 | |
| 1267 *** With M-x enable-flow-control, you need to type C-\ twice to do | |
| 1268 incremental search--a single C-\ gets no response. | |
| 1269 | |
| 1270 This has been traced to communicating with your machine via kermit, | |
| 1271 with C-\ as the kermit escape character. One solution is to use | |
| 1272 another escape character in kermit. One user did | |
| 1273 | |
| 1274 set escape-character 17 | |
| 1275 | |
| 1276 in his .kermrc file, to make C-q the kermit escape character. | |
| 1277 | |
| 1278 *** The Motif version of Emacs paints the screen a solid color. | |
| 1279 | |
| 1280 This has been observed to result from the following X resource: | |
| 1281 | |
| 1282 Emacs*default.attributeFont: -*-courier-medium-r-*-*-*-140-*-*-*-*-iso8859-* | |
| 1283 | |
| 1284 That the resource has this effect indicates a bug in something, but we | |
| 1285 do not yet know what. If it is an Emacs bug, we hope someone can | |
| 1286 explain what the bug is so we can fix it. In the mean time, removing | |
| 1287 the resource prevents the problem. | |
| 1288 | |
| 1289 *** After running emacs once, subsequent invocations crash. | |
| 1290 | |
| 1291 Some versions of SVR4 have a serious bug in the implementation of the | |
| 1292 mmap () system call in the kernel; this causes emacs to run correctly | |
| 1293 the first time, and then crash when run a second time. | |
| 1294 | |
| 1295 Contact your vendor and ask for the mmap bug fix; in the mean time, | |
| 1296 you may be able to work around the problem by adding a line to your | |
| 1297 operating system description file (whose name is reported by the | |
| 1298 configure script) that reads: | |
| 1299 #define SYSTEM_MALLOC | |
| 1300 This makes Emacs use memory less efficiently, but seems to work around | |
| 1301 the kernel bug. | |
| 1302 | |
| 1303 *** Inability to send an Alt-modified key, when Emacs is communicating | |
| 1304 directly with an X server. | |
| 1305 | |
| 1306 If you have tried to bind an Alt-modified key as a command, and it | |
| 1307 does not work to type the command, the first thing you should check is | |
| 1308 whether the key is getting through to Emacs. To do this, type C-h c | |
| 1309 followed by the Alt-modified key. C-h c should say what kind of event | |
| 1310 it read. If it says it read an Alt-modified key, then make sure you | |
| 1311 have made the key binding correctly. | |
| 1312 | |
| 1313 If C-h c reports an event that doesn't have the Alt modifier, it may | |
| 1314 be because your X server has no key for the Alt modifier. The X | |
| 1315 server that comes from MIT does not set up the Alt modifier by | |
| 1316 default. | |
| 1317 | |
| 1318 If your keyboard has keys named Alt, you can enable them as follows: | |
| 1319 | |
| 1320 xmodmap -e 'add mod2 = Alt_L' | |
| 1321 xmodmap -e 'add mod2 = Alt_R' | |
| 1322 | |
| 1323 If the keyboard has just one key named Alt, then only one of those | |
| 1324 commands is needed. The modifier `mod2' is a reasonable choice if you | |
| 1325 are using an unmodified MIT version of X. Otherwise, choose any | |
| 1326 modifier bit not otherwise used. | |
| 1327 | |
| 1328 If your keyboard does not have keys named Alt, you can use some other | |
| 1329 keys. Use the keysym command in xmodmap to turn a function key (or | |
| 1330 some other 'spare' key) into Alt_L or into Alt_R, and then use the | |
| 1331 commands show above to make them modifier keys. | |
| 1332 | |
| 1333 Note that if you have Alt keys but no Meta keys, Emacs translates Alt | |
| 1334 into Meta. This is because of the great importance of Meta in Emacs. | |
| 1335 | |
| 1336 *** In Shell mode, you get a ^M at the end of every line. | |
| 1337 | |
| 1338 This happens to people who use tcsh, because it is trying to be too | |
| 1339 smart. It sees that the Shell uses terminal type `unknown' and turns | |
| 1340 on the flag to output ^M at the end of each line. You can fix the | |
| 1341 problem by adding this to your .cshrc file: | |
| 1342 | |
| 1343 if ($?EMACS) then | |
| 1344 if ($EMACS == "t") then | |
| 454 | 1345 unset edit |
| 373 | 1346 stty -icrnl -onlcr -echo susp ^Z |
| 1347 endif | |
| 1348 endif | |
| 1349 | |
| 1350 *** An error message such as `X protocol error: BadMatch (invalid | |
| 1351 parameter attributes) on protocol request 93'. | |
| 1352 | |
| 1353 This comes from having an invalid X resource, such as | |
| 1354 emacs*Cursor: black | |
| 1355 (which is invalid because it specifies a color name for something | |
| 1356 that isn't a color.) | |
| 1357 | |
| 1358 The fix is to correct your X resources. | |
| 1359 | |
| 1360 *** Once you pull down a menu from the menubar, it won't go away. | |
| 1361 | |
| 1362 It has been claimed that this is caused by a bug in certain very old | |
| 1363 (1990?) versions of the twm window manager. It doesn't happen with | |
| 1364 recent vintages, or with other window managers. | |
| 1365 | |
| 1366 *** Emacs ignores the "help" key when running OLWM. | |
| 1367 | |
| 1368 OLWM grabs the help key, and retransmits it to the appropriate client | |
| 1369 using XSendEvent. Allowing emacs to react to synthetic events is a | |
| 1370 security hole, so this is turned off by default. You can enable it by | |
| 1371 setting the variable x-allow-sendevents to t. You can also cause fix | |
| 1372 this by telling OLWM to not grab the help key, with the null binding | |
| 1373 "OpenWindows.KeyboardCommand.Help:". | |
| 1374 | |
| 1375 *** Programs running under terminal emulator do not recognize `emacs' | |
| 1376 terminal type. | |
| 1377 | |
| 1378 The cause of this is a shell startup file that sets the TERMCAP | |
| 1379 environment variable. The terminal emulator uses that variable to | |
| 1380 provide the information on the special terminal type that Emacs | |
| 1381 emulates. | |
| 1382 | |
| 1383 Rewrite your shell startup file so that it does not change TERMCAP | |
| 1384 in such a case. You could use the following conditional which sets | |
| 1385 it only if it is undefined. | |
| 1386 | |
| 1387 if ( ! ${?TERMCAP} ) setenv TERMCAP ~/my-termcap-file | |
| 1388 | |
| 1389 Or you could set TERMCAP only when you set TERM--which should not | |
| 1390 happen in a non-login shell. | |
| 1391 | |
| 442 | 1392 *** The popup menu appears at the bottom/right of my screen. |
| 373 | 1393 |
| 1389 | 1394 You probably have something like the following in your ~/.Xresources |
| 373 | 1395 |
| 1396 Emacs.geometry: 81x56--9--1 | |
| 1397 | |
| 1398 Use the following instead | |
| 1399 | |
| 1400 Emacs*EmacsFrame.geometry: 81x56--9--1 | |
| 1401 | |
| 1222 | 1402 *** When I try to use the PostgreSQL functions, I get a message about |
| 1403 undefined symbols. | |
| 1404 | |
| 1405 The only known case in which this happens is if you are using gcc, you | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
1406 configured with --with-error-checking=all and --with-modules, and |
| 2648 | 1407 you compiled with no optimization. If you encounter this problem in any |
| 1222 | 1408 other situation, please inform xemacs-beta@xemacs.org. |
| 1409 | |
| 1410 This problem stems from a gcc bug. With no optimization, functions | |
| 1411 declared `extern inline' sometimes are not completely compiled away. An | |
| 1412 undefined symbol with the function's name is put into the resulting | |
| 1413 object file. In this case, when the postgresql module is loaded, the | |
| 1414 linker is unable to resolve that symbol, so the module load fails. The | |
| 1415 workaround is to recompile the module with optimization turned on. Any | |
| 1416 optimization level, including -Os, appears to work. | |
| 1417 | |
| 1332 | 1418 *** C-z just refreshes the screen instead of suspending Emacs. |
| 1419 | |
| 1420 You are probably using a shell that doesn't support job control, even | |
| 1421 though the system itself is capable of it. Try using a different | |
| 1422 shell. | |
| 373 | 1423 |
| 1036 | 1424 ** MacOS/X, Darwin |
|
4753
dd933a82720c
Describe duplicate symbols warning.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
4634
diff
changeset
|
1425 *** The linker warns about duplicate symbols. |
|
dd933a82720c
Describe duplicate symbols warning.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
4634
diff
changeset
|
1426 |
|
dd933a82720c
Describe duplicate symbols warning.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
4634
diff
changeset
|
1427 This occurs in the function alarm(), which we deliberately override, |
|
dd933a82720c
Describe duplicate symbols warning.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
4634
diff
changeset
|
1428 and in vendorShellRec when using Xaw3d for the widgets. |
|
dd933a82720c
Describe duplicate symbols warning.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
4634
diff
changeset
|
1429 |
|
dd933a82720c
Describe duplicate symbols warning.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
4634
diff
changeset
|
1430 For alarm(), the linker chooses our version, as desired. |
|
dd933a82720c
Describe duplicate symbols warning.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
4634
diff
changeset
|
1431 Unfortunately, for vendorShellRec, the Xt version is used instead of |
|
dd933a82720c
Describe duplicate symbols warning.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
4634
diff
changeset
|
1432 Xaw3d's. This does not seem to cause problems. |
|
dd933a82720c
Describe duplicate symbols warning.
Stephen J. Turnbull <stephen@xemacs.org>
parents:
4634
diff
changeset
|
1433 |
| 1036 | 1434 *** XEmacs crashes on MacOS within font-lock, or when dealing |
| 1435 with large compilation buffers, or in other regex applications. | |
| 1436 | |
| 3074 | 1437 The default stack size under MacOS/X prior to 10.3 (Panther) is rather |
| 1438 small (512k as opposed to Solaris 8M), hosing the regexp code, which | |
| 1439 uses alloca() extensively, overflowing the stack when complex regexps | |
| 1440 are used. Workarounds: | |
| 1036 | 1441 |
| 1442 1) Increase your stack size, using `ulimit -s 8192' or a (t)csh | |
| 1443 equivalent; | |
| 1444 | |
| 1445 2) Recompile regex.c with REGEX_MALLOC defined. | |
| 1446 | |
| 373 | 1447 ** AIX |
| 1448 *** Your Delete key sends a Backspace to the terminal, using an AIXterm. | |
| 0 | 1449 |
| 1389 | 1450 The solution is to include in your .Xresources the lines: |
| 0 | 1451 |
| 1452 *aixterm.Translations: #override <Key>BackSpace: string(0x7f) | |
| 1453 aixterm*ttyModes: erase ^? | |
| 1454 | |
| 1455 This makes your Backspace key send DEL (ASCII 127). | |
| 1456 | |
| 373 | 1457 *** On AIX 4, some programs fail when run in a Shell buffer |
| 1458 with an error message like No terminfo entry for "unknown". | |
| 124 | 1459 |
| 373 | 1460 On AIX, many terminal type definitions are not installed by default. |
| 1461 `unknown' is one of them. Install the "Special Generic Terminal | |
| 1462 Definitions" to make them defined. | |
| 124 | 1463 |
| 373 | 1464 *** On AIX, you get this message when running Emacs: |
| 0 | 1465 |
| 373 | 1466 Could not load program emacs |
| 1467 Symbol smtcheckinit in csh is undefined | |
| 1468 Error was: Exec format error | |
| 124 | 1469 |
| 373 | 1470 or this one: |
| 0 | 1471 |
| 373 | 1472 Could not load program .emacs |
| 1473 Symbol _system_con in csh is undefined | |
| 1474 Symbol _fp_trapsta in csh is undefined | |
| 1475 Error was: Exec format error | |
| 124 | 1476 |
| 373 | 1477 These can happen when you try to run on AIX 3.2.5 a program that was |
| 1478 compiled with 3.2.4. The fix is to recompile. | |
| 124 | 1479 |
| 373 | 1480 *** Trouble using ptys on AIX. |
| 1481 | |
| 1482 People often install the pty devices on AIX incorrectly. | |
| 1483 Use `smit pty' to reinstall them properly. | |
| 1484 | |
| 0 | 1485 |
| 373 | 1486 ** SunOS/Solaris |
| 1487 *** The Emacs window disappears when you type M-q. | |
| 197 | 1488 |
| 373 | 1489 Some versions of the Open Look window manager interpret M-q as a quit |
| 1490 command for whatever window you are typing at. If you want to use | |
| 1491 Emacs with that window manager, you should try to configure the window | |
| 1492 manager to use some other command. You can disable the | |
| 1493 shortcut keys entirely by adding this line to ~/.OWdefaults: | |
| 0 | 1494 |
| 373 | 1495 OpenWindows.WindowMenuAccelerators: False |
| 1496 | |
| 1497 *** When Emacs tries to ring the bell, you get an error like | |
| 124 | 1498 |
| 1499 audio: sst_open: SETQSIZE" Invalid argument | |
| 1500 audio: sst_close: SETREG MMR2, Invalid argument | |
| 1501 | |
| 197 | 1502 you have probably compiled using an ANSI C compiler, but with non-ANSI |
| 1503 include files. In particular, on Suns, the file | |
| 1504 /usr/include/sun/audioio.h uses the _IOW macro to define the constant | |
| 1505 AUDIOSETQSIZE. _IOW in turn uses a K&R preprocessor feature that is | |
| 1506 now explicitly forbidden in ANSI preprocessors, namely substitution | |
| 1507 inside character constants. All ANSI C compilers must provide a | |
| 1508 workaround for this problem. Lucid's C compiler is shipped with a new | |
| 1509 set of system include files. If you are using GCC, there is a script | |
| 1510 called fixincludes that creates new versions of some system include | |
| 1511 files that use this obsolete feature. | |
| 124 | 1512 |
| 373 | 1513 *** On Solaris 2.6, XEmacs dumps core when exiting. |
| 0 | 1514 |
| 373 | 1515 This happens if you're XEmacs is running on the same machine as the X |
| 1516 server, and the optimized memory transport has been turned on by | |
| 1517 setting the environment variable XSUNTRANSPORT. The crash occurs | |
| 1518 during the call to XCloseDisplay. | |
| 124 | 1519 |
| 373 | 1520 If this describes your situation, you need to undefine the |
| 1521 XSUNTRANSPORT environment variable. | |
| 126 | 1522 |
| 373 | 1523 *** On Solaris, C-x doesn't get through to Emacs when you use the console. |
| 124 | 1524 |
| 373 | 1525 This is a Solaris feature (at least on Intel x86 cpus). Type C-r |
| 1526 C-r C-t, to toggle whether C-x gets through to Emacs. | |
| 124 | 1527 |
| 373 | 1528 *** On Solaris 2.4, Dired hangs and C-g does not work. Or Emacs hangs |
| 197 | 1529 forever waiting for termination of a subprocess that is a zombie. |
| 124 | 1530 |
| 1531 casper@fwi.uva.nl says the problem is in X11R6. Rebuild libX11.so | |
| 1532 after changing the file xc/config/cf/sunLib.tmpl. Change the lines | |
| 1533 | |
| 1534 #if ThreadedX | |
| 1535 #define SharedX11Reqs -lthread | |
| 1536 #endif | |
| 1537 | |
| 1538 to: | |
| 1539 | |
| 1540 #if OSMinorVersion < 4 | |
| 1541 #if ThreadedX | |
| 1542 #define SharedX11Reqs -lthread | |
| 1543 #endif | |
| 1544 #endif | |
| 1545 | |
| 1546 Be sure also to edit x/config/cf/sun.cf so that OSMinorVersion is 4 | |
| 1547 (as it should be for Solaris 2.4). The file has three definitions for | |
| 1548 OSMinorVersion: the first is for x86, the second for SPARC under | |
| 1549 Solaris, and the third for SunOS 4. Make sure to update the | |
| 1550 definition for your type of machine and system. | |
| 1551 | |
| 1552 Then do `make Everything' in the top directory of X11R6, to rebuild | |
| 1553 the makefiles and rebuild X. The X built this way work only on | |
| 1554 Solaris 2.4, not on 2.3. | |
| 1555 | |
| 1556 For multithreaded X to work it necessary to install patch | |
| 1557 101925-02 to fix problems in header files [2.4]. You need | |
| 1558 to reinstall gcc or re-run just-fixinc after installing that | |
| 1559 patch. | |
| 0 | 1560 |
| 124 | 1561 However, Frank Rust <frust@iti.cs.tu-bs.de> used a simpler solution: |
| 1562 he changed | |
| 1563 #define ThreadedX YES | |
| 1564 to | |
| 1565 #define ThreadedX NO | |
| 1566 in sun.cf and did `make World' to rebuild X11R6. Removing all | |
| 1567 `-DXTHREAD*' flags and `-lthread' entries from lib/X11/Makefile and | |
| 1568 typing 'make install' in that directory also seemed to work. | |
| 1569 | |
| 373 | 1570 *** On SunOS 4.1.3, Emacs unpredictably crashes in _yp_dobind_soft. |
| 124 | 1571 |
| 373 | 1572 This happens if you configure Emacs specifying just `sparc-sun-sunos4' |
| 1573 on a system that is version 4.1.3. You must specify the precise | |
| 1574 version number (or let configure figure out the configuration, which | |
| 1575 it can do perfectly well for SunOS). | |
| 124 | 1576 |
| 373 | 1577 *** Mail is lost when sent to local aliases. |
| 0 | 1578 |
| 124 | 1579 Many emacs mail user agents (VM and rmail, for instance) use the |
| 1580 sendmail.el library. This library can arrange for mail to be | |
| 1581 delivered by passing messages to the /usr/lib/sendmail (usually) | |
| 1582 program . In doing so, it passes the '-t' flag to sendmail, which | |
| 1583 means that the name of the recipient of the message is not on the | |
| 1584 command line and, therefore, that sendmail must parse the message to | |
| 1585 obtain the destination address. | |
| 1586 | |
| 1587 There is a bug in the SunOS4.1.1 and SunOS4.1.3 versions of sendmail. | |
| 1588 In short, when given the -t flag, the SunOS sendmail won't recognize | |
| 1589 non-local (i.e. NIS) aliases. It has been reported that the Solaris | |
| 1590 2.x versions of sendmail do not have this bug. For those using SunOS | |
| 1591 4.1, the best fix is to install sendmail V8 or IDA sendmail (which | |
| 1592 have other advantages over the regular sendmail as well). At the time | |
| 1593 of this writing, these official versions are available: | |
| 1594 | |
| 1595 Sendmail V8 on ftp.cs.berkeley.edu in /ucb/sendmail: | |
| 1596 sendmail.8.6.9.base.tar.Z (the base system source & documentation) | |
| 1597 sendmail.8.6.9.cf.tar.Z (configuration files) | |
| 1598 sendmail.8.6.9.misc.tar.Z (miscellaneous support programs) | |
| 1599 sendmail.8.6.9.xdoc.tar.Z (extended documentation, with postscript) | |
| 1600 | |
| 1601 IDA sendmail on vixen.cso.uiuc.edu in /pub: | |
| 1602 sendmail-5.67b+IDA-1.5.tar.gz | |
| 1603 | |
| 373 | 1604 *** Emacs fails to understand most Internet host names, even though |
| 124 | 1605 the names work properly with other programs on the same system. |
| 197 | 1606 Emacs won't work with X-windows if the value of DISPLAY is HOSTNAME:0. |
| 1607 Gnus can't make contact with the specified host for nntp. | |
| 0 | 1608 |
| 124 | 1609 This typically happens on Suns and other systems that use shared |
| 1610 libraries. The cause is that the site has installed a version of the | |
| 1611 shared library which uses a name server--but has not installed a | |
| 1612 similar version of the unshared library which Emacs uses. | |
| 0 | 1613 |
| 124 | 1614 The result is that most programs, using the shared library, work with |
| 1615 the nameserver, but Emacs does not. | |
| 1616 | |
| 1617 The fix is to install an unshared library that corresponds to what you | |
| 1618 installed in the shared library, and then relink Emacs. | |
| 0 | 1619 |
| 124 | 1620 On SunOS 4.1, simply define HAVE_RES_INIT. |
| 1621 | |
| 1622 If you have already installed the name resolver in the file libresolv.a, | |
| 1623 then you need to compile Emacs to use that library. The easiest way to | |
| 1624 do this is to add to config.h a definition of LIBS_SYSTEM, LIBS_MACHINE | |
| 1625 or LIB_STANDARD which uses -lresolv. Watch out! If you redefine a macro | |
| 1626 that is already in use in your configuration to supply some other libraries, | |
| 1627 be careful not to lose the others. | |
| 1628 | |
| 1629 Thus, you could start by adding this to config.h: | |
| 1630 | |
| 1631 #define LIBS_SYSTEM -lresolv | |
| 1632 | |
| 1633 Then if this gives you an error for redefining a macro, and you see that | |
| 1634 the s- file defines LIBS_SYSTEM as -lfoo -lbar, you could change config.h | |
| 1635 again to say this: | |
| 1636 | |
| 1637 #define LIBS_SYSTEM -lresolv -lfoo -lbar | |
| 1638 | |
| 373 | 1639 *** With process-connection-type set to t, each line of subprocess |
| 1640 output is terminated with a ^M, making ange-ftp and GNUS not work. | |
| 1641 | |
| 1642 On SunOS systems, this problem has been seen to be a result of an | |
| 1643 incomplete installation of gcc 2.2 which allowed some non-ANSI | |
| 1644 compatible include files into the compilation. In particular this | |
| 1645 affected virtually all ioctl() calls. | |
| 1646 | |
| 1647 | |
| 1648 ** Linux | |
| 845 | 1649 *** XEmacs crashes on startup, in make-frame. |
| 1650 | |
| 1651 Typically the Lisp backtrace includes | |
| 1652 | |
| 1653 make-frame(nil #<x-device on ":0.0" 0x2558>) | |
| 1654 | |
| 2648 | 1655 somewhere near the top. The problem is due to an improvement in GNU ld |
| 1656 that sorts the ELF reloc sections in the executable, giving dramatic | |
| 1657 speedups in startup for large executables. It also confuses the | |
| 1658 traditional unexec code in XEmacs, leading to the core dump. The | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
1659 solution is to use the --with-pdump or --with-ldflags='-z nocombreloc' |
| 2648 | 1660 options to configure. Recent 21.4 and 12.5 autodetect this in |
| 1661 configure. | |
| 845 | 1662 |
| 1663 Red Hat and SuSE (at least) distributed a prerelease version of ld | |
| 1664 (versions around 2.11.90.x.y) where autodetection is impossible. The | |
| 1665 recommended procedure is to upgrade to binutils >= 2.12 and rerun | |
|
4842
1d775c6304d1
Use with-FOO consistently instead of enable-FOO
Ben Wing <ben@xemacs.org>
parents:
4753
diff
changeset
|
1666 configure. Otherwise you must apply the flags by hand. --with-pdump |
| 2648 | 1667 is recommended. |
| 448 | 1668 |
| 1669 *** I want XEmacs to use the Alt key, not the XXX key, for Meta commands | |
| 1670 | |
| 1671 For historical reasons, XEmacs looks for a Meta key, then an Alt key. | |
| 1672 It binds Meta commands to the X11 modifier bit attached to the first | |
| 1673 of these it finds. On PCs, the Windows key is often assigned the Meta | |
| 1674 bit, but many desktop environments go to great lengths to get all apps | |
| 1675 to use the Alt key, and reserve the Windows key to (sensibly enough) | |
| 1676 the window manager. | |
| 1677 | |
| 1678 One correct way to implement this was suggested on comp.emacs.xemacs | |
| 1679 (by Kilian Foth and in more detail by Michael Piotrowski): unmap the | |
| 1680 Meta modifier using xmodmap or xkb, and then map the Meta/Windows key | |
| 450 | 1681 to the Super or Hyper keysym and an appropriate mod bit. XEmacs will |
| 1682 not find the Meta keysym, and default to using the Alt key for Meta | |
| 1683 keybindings. Typically few applications use the (X11) Meta modifier; | |
| 1684 it is tedious but not too much so to teach the ones you need to use | |
| 1685 Super instead of Meta. There may be further useful hints in the | |
| 1686 discussion of keymapping on non-Linux platforms. | |
| 1687 | |
| 1688 *** The color-gcc wrapper | |
| 1689 | |
| 1690 This wrapper colorizes the error messages from gcc. By default XEmacs | |
| 1691 does not interpret the escape sequences used to generate colors, | |
| 1692 resulting in a cluttered, hard-to-read buffer. You can remove the | |
| 1693 wrapper, or defeat the wrapper colorization in Emacs process buffers | |
| 1694 by editing the "nocolor" attribute in /etc/colorgccrc: | |
| 1695 | |
| 1696 $ diff -u /etc/colorgccrc.old /etc/colorgccrc | |
| 1697 --- /etc/colorgccrc.old Tue Dec 26 02:17:46 2000 | |
| 1698 +++ /etc/colorgccrc Tue Dec 26 02:15:48 2000 | |
| 1699 @@ -34,1 +34,1 @@ | |
| 1700 -nocolor: dumb | |
| 1701 +nocolor: dumb emacs | |
| 1702 | |
| 1703 If you want colorization in your Emacs buffers, you may get good | |
| 1704 results from the ansi-color.el library: | |
| 1705 | |
| 1706 http://www.geocities.com/kensanata/color-emacs.html#ansicolors | |
| 1707 | |
| 1708 This is written for the mainline GNU Emacs but the author has made | |
| 1709 efforts to adapt it to XEmacs. YMMV. | |
| 448 | 1710 |
| 373 | 1711 *** Slow startup on Linux. |
| 1712 | |
| 1713 People using systems based on the Linux kernel sometimes report that | |
| 448 | 1714 startup takes 10 to 15 seconds longer than `usual'. There are two |
| 1715 problems, one older, one newer. | |
| 1716 | |
| 1717 **** Old problem: IPv4 host lookup | |
| 373 | 1718 |
| 448 | 1719 On older systems, this is because Emacs looks up the host name when it |
| 1720 starts. Normally, this takes negligible time; the extra delay is due | |
| 1721 to improper system configuration. (Recent Linux distros usually have | |
| 1722 this configuration correct "out of the box".) This problem can occur | |
| 1723 for both networked and non-networked machines. | |
| 373 | 1724 |
| 1725 Here is how to fix the configuration. It requires being root. | |
| 1726 | |
| 448 | 1727 ***** Networked Case |
| 373 | 1728 |
| 1729 First, make sure the files `/etc/hosts' and `/etc/host.conf' both | |
| 1730 exist. The first line in the `/etc/hosts' file should look like this | |
| 1731 (replace HOSTNAME with your host name): | |
| 1732 | |
| 1733 127.0.0.1 localhost HOSTNAME | |
| 1734 | |
| 1735 Also make sure that the `/etc/host.conf' files contains the following | |
| 1736 lines: | |
| 1737 | |
| 454 | 1738 order hosts, bind |
| 373 | 1739 multi on |
| 1740 | |
| 1741 Any changes, permanent and temporary, to the host name should be | |
| 1742 indicated in the `/etc/hosts' file, since it acts a limited local | |
| 1743 database of addresses and names (e.g., some SLIP connections | |
| 1744 dynamically allocate ip addresses). | |
| 1745 | |
| 448 | 1746 ***** Non-Networked Case |
| 373 | 1747 |
| 1748 The solution described in the networked case applies here as well. | |
| 1749 However, if you never intend to network your machine, you can use a | |
| 1750 simpler solution: create an empty `/etc/host.conf' file. The command | |
| 1751 `touch /etc/host.conf' suffices to create the file. The `/etc/hosts' | |
| 1752 file is not necessary with this approach. | |
| 1753 | |
| 448 | 1754 **** New problem: IPv6 CNAME lookup |
| 1755 | |
| 1756 A newer problem is due to XEmacs changing to use the modern | |
| 1757 getaddrinfo() interface from the older gethostbyname() interface. The | |
| 1758 solution above is insufficient, because getaddrinfo() by default tries | |
| 1759 to get IPv6 information for localhost. This always involves a dns | |
| 1760 lookup to get the CNAME, and the strategies above don't work. It then | |
| 724 | 1761 falls back to IPv4 behavior. This is good[tm] according the people at |
| 1762 WIDE who know about IPv6. | |
| 448 | 1763 |
| 1764 ***** Robust network case | |
| 1765 | |
| 1766 Configure your network so that there are no nameservers configured | |
| 1767 until the network is actually running. getaddrinfo() will not try to | |
| 1768 access a nameserver that isn't configured. | |
| 1769 | |
| 1770 ***** Flaky network case | |
| 1771 | |
| 1772 If you have a flaky modem or DSL connection that can be relied on only | |
| 1773 to go down whenever you want to bring XEmacs up, you need to force | |
| 1774 IPv4 behavior. Explicitly setting DISPLAY=127.0.0.1:0.0 (or whatever | |
| 1775 is appropriate) works in most cases. | |
| 1776 | |
| 1777 If you cannot or do not want to do that, you can hard code IPv4 | |
| 1778 behavior in src/process-unix.c. This is bad[tm], on your own head be | |
| 724 | 1779 it. Use the configure option `--with-ipv6-cname=no'. |
| 373 | 1780 |
| 845 | 1781 *** Mandrake |
| 1782 | |
| 1783 The Mandrake Linux distribution is attempting to comprehensively | |
| 1784 update the user interface, and make it consistent across | |
| 1785 applications. This is very difficult, and will occasionally cause | |
| 1786 conflicts with applications like Emacs with their own long-established | |
| 1787 interfaces. Known issues specific to Mandrake or especially common: | |
| 1788 | |
| 1789 Some versions of XEmacs (21.1.9 is known) distributed with Mandrake | |
| 1790 were patched to make the Meta and Alt keysyms synonymous. These | |
| 1791 normally work as expected in the Mandrake environment. However, | |
| 1792 custom-built XEmacsen (including all 21.2 betas) will "inexplicably" | |
| 1793 not respect the "Alt-invokes-Meta-commands" convention. See "I want | |
| 1794 XEmacs to use the Alt key" below. | |
| 1795 | |
| 1796 The color-gcc wrapper (see below) is in common use on the Mandrake | |
| 1797 platform. | |
| 1798 | |
| 1799 *** You get crashes in a non-C locale with Linux GNU Libc 2.0. | |
| 1800 | |
| 1801 Internationalization was not the top priority for GNU Libc 2.0. | |
| 1802 As of this writing (1998-12-28) you may get crashes while running | |
| 1803 XEmacs in a non-C locale. For example, `LC_ALL=en_US xemacs' crashes | |
| 1804 while `LC_ALL=C xemacs' runs fine. This happens for example with GNU | |
| 1805 libc 2.0.7. Installing libintl.a and libintl.h built from gettext | |
| 1806 0.10.35 and re-building XEmacs solves the crashes. Presumably soon | |
| 1807 everyone will upgrade to GNU Libc 2.1 and this problem will go away. | |
| 1808 | |
| 1809 *** `C-z', or `M-x suspend-emacs' hangs instead of suspending. | |
| 1810 | |
| 1811 If you build with `gpm' support on Linux, you cannot suspend XEmacs | |
| 1812 because gpm installs a buggy SIGTSTP handler. Either compile with | |
| 1813 `--with-gpm=no', or don't suspend XEmacs on the Linux console until | |
| 1814 this bug is fixed. | |
| 1815 | |
| 1816 *** With certain fonts, when the cursor appears on a character, the | |
| 1817 character doesn't appear--you get a solid box instead. | |
| 1818 | |
| 1819 One user on a Linux system reported that this problem went away with | |
| 1820 installation of a new X server. The failing server was XFree86 3.1.1. | |
| 1821 XFree86 3.1.2 works. | |
| 1822 | |
| 373 | 1823 ** IRIX |
| 1824 *** On Irix, I don't see the toolbar icons and I'm getting lots of | |
| 1825 entries in the warnings buffer. | |
| 1826 | |
| 1827 SGI ships a really old Xpm library in /usr/lib which does not work at | |
| 1828 all well with XEmacs. The solution is to install your own copy of the | |
| 2648 | 1829 latest version of Xpm somewhere and then use the --with-site-includes |
| 1830 and --with-site-libraries flags to tell configure where to find it. | |
| 373 | 1831 |
| 1832 *** Trouble using ptys on IRIX, or running out of ptys. | |
| 1833 | |
| 1834 The program mkpts (which may be in `/usr/adm' or `/usr/sbin') needs to | |
| 1835 be set-UID to root, or non-root programs like Emacs will not be able | |
| 1836 to allocate ptys reliably. | |
| 1837 | |
| 1838 *** Beware of the default image & graphics library on Irix | |
| 1839 | |
| 1840 Richard Cognot <cognot@ensg.u-nancy.fr> writes: | |
| 1841 | |
| 1842 You *have* to compile your own jpeg lib. The one delivered with SGI | |
| 1843 systems is a C++ lib, which apparently XEmacs cannot cope with. | |
| 1844 | |
| 1845 | |
| 1846 ** Digital UNIX/OSF/VMS/Ultrix | |
| 1847 *** XEmacs crashes on Digital Unix within font-lock, or when dealing | |
| 1036 | 1848 with large compilation buffers, or in other regex applications. |
| 373 | 1849 |
| 1850 The default stack size under Digital Unix is rather small (2M as | |
| 1851 opposed to Solaris 8M), hosing the regexp code, which uses alloca() | |
| 1852 extensively, overflowing the stack when complex regexps are used. | |
| 1853 Workarounds: | |
| 312 | 1854 |
| 373 | 1855 1) Increase your stack size, using `ulimit -s 8192' or a (t)csh |
| 1856 equivalent; | |
| 1857 | |
| 1858 2) Recompile regex.c with REGEX_MALLOC defined. | |
| 1859 | |
| 1860 *** The `Alt' key doesn't behave as `Meta' when running DECwindows. | |
| 1861 | |
| 1862 The default DEC keyboard mapping has the Alt keys set up to generate the | |
| 1863 keysym `Multi_key', which has a meaning to xemacs which is distinct from that | |
| 1864 of the `Meta_L' and `Meta-R' keysyms. A second problem is that certain keys | |
| 1865 have the Mod2 modifier attached to them for no adequately explored reason. | |
| 1866 The correct fix is to pass this file to xmodmap upon starting X: | |
| 1867 | |
| 1868 clear mod2 | |
| 1869 keysym Multi_key = Alt_L | |
| 1870 add mod1 = Alt_L | |
| 1871 add mod1 = Alt_R | |
| 1872 | |
| 1873 *** The Compose key on a DEC keyboard does not work as Meta key. | |
| 1874 | |
| 1875 This shell command should fix it: | |
| 1876 | |
| 1877 xmodmap -e 'keycode 0xb1 = Meta_L' | |
| 1878 | |
| 1879 *** `expand-file-name' fails to work on any but the machine you dumped | |
| 1880 Emacs on. | |
| 1881 | |
| 1882 On Ultrix, if you use any of the functions which look up information | |
| 1883 in the passwd database before dumping Emacs (say, by using | |
| 1884 expand-file-name in site-init.el), then those functions will not work | |
| 1885 in the dumped Emacs on any host but the one Emacs was dumped on. | |
| 1886 | |
| 1887 The solution? Don't use expand-file-name in site-init.el, or in | |
| 1888 anything it loads. Yuck - some solution. | |
| 1889 | |
| 1890 I'm not sure why this happens; if you can find out exactly what is | |
| 1891 going on, and perhaps find a fix or a workaround, please let us know. | |
| 1892 Perhaps the YP functions cache some information, the cache is included | |
| 1893 in the dumped Emacs, and is then inaccurate on any other host. | |
| 1894 | |
| 1895 | |
| 1896 ** HP-UX | |
| 1897 *** I get complaints about the mapping of my HP keyboard at startup, | |
| 1898 but I haven't changed anything. | |
| 1899 | |
| 1900 The default HP keymap is set up to have Mod1 assigned to two different keys: | |
| 1901 Meta_L and Mode_switch (even though there is not actually a Mode_switch key on | |
| 1902 the keyboard -- it uses an "imaginary" keycode.) There actually is a reason | |
| 1903 for this, but it's not a good one. The correct fix is to execute this command | |
| 1904 upon starting X: | |
| 1905 | |
| 1906 xmodmap -e 'remove mod1 = Mode_switch' | |
| 312 | 1907 |
| 373 | 1908 *** On HP-UX, you get "poll: Interrupted system call" message in the |
| 1909 window where XEmacs was launched. | |
| 1910 | |
| 1911 Richard Cognot <cognot@ensg.u-nancy.fr> writes: | |
| 1912 | |
| 1913 I get a very strange problem when linking libc.a dynamically: every | |
| 1914 event (mouse, keyboard, expose...) results in a "poll: Interrupted | |
| 1915 system call" message in the window where XEmacs was | |
| 1916 launched. Forcing a static link of libc.a alone by adding | |
| 1917 /usr/lib/libc.a at the end of the link line solves this. Note that | |
| 1918 my 9.07 build of 19.14b17 and my (old) build of 19.13 both exhibit | |
| 442 | 1919 the same behavior. I've tried various hpux patches to no avail. If |
| 373 | 1920 this problem cannot be solved before the release date, binary kits |
| 1921 for HP *must* be linked statically against libc, otherwise this | |
| 1922 problem will show up. (This is directed at whoever will volunteer | |
| 1923 for this kit, as I won't be available to do it, unless 19.14 gets | |
| 1924 delayed until mid-june ;-). I think this problem will be an FAQ soon | |
| 1925 after the release otherwise. | |
| 1926 | |
| 1927 Note: The above entry is probably not valid for XEmacs 21.0 and | |
| 1928 later. | |
| 1929 | |
| 1930 *** The right Alt key works wrong on German HP keyboards (and perhaps | |
| 1931 other non-English HP keyboards too). | |
| 1932 | |
| 1933 This is because HP-UX defines the modifiers wrong in X. Here is a | |
| 1934 shell script to fix the problem; be sure that it is run after VUE | |
| 1935 configures the X server. | |
| 1936 | |
| 1937 xmodmap 2> /dev/null - << EOF | |
| 1938 keysym Alt_L = Meta_L | |
| 1939 keysym Alt_R = Meta_R | |
| 1940 EOF | |
| 1941 | |
| 1942 xmodmap - << EOF | |
| 1943 clear mod1 | |
| 1944 keysym Mode_switch = NoSymbol | |
| 1945 add mod1 = Meta_L | |
| 1946 keysym Meta_R = Mode_switch | |
| 1947 add mod2 = Mode_switch | |
| 1948 EOF | |
| 1949 | |
| 442 | 1950 |
| 1951 *** XEmacs dumps core at startup when native audio is used. Native | |
| 1952 audio does not work with recent versions of HP-UX. | |
| 1953 | |
| 1954 Under HP-UX 10.20 and later (e.g., HP-UX 11.XX), with native audio | |
| 1955 enabled, the dumped XEmacs binary ("xemacs") core dumps at startup if | |
| 1956 recent versions of the libAlib.sl audio shared library is used. Note | |
| 1957 that "temacs" will run, but "xemacs" will dump core. This, of course, | |
| 1958 causes the XEmacs build to fail. If GNU malloc is enabled, a stack | |
| 1959 trace will show XEmacs to have crashed in the "first" call to malloc(). | |
| 1960 | |
| 1961 This bug currently exists in all versions of XEmacs, when the undump | |
| 1962 mechanism is used. It is not known if using the experimental portable | |
| 1963 dumper will allow native audio to work. | |
| 1964 | |
| 1965 **** Cause: | |
| 1966 | |
| 1967 Recent versions of the HP-UX 10.20 (and later) audio shared library (in | |
| 1968 /opt/audio/lib), pulls in the libdce shared library, which pulls in a | |
| 1969 thread (libcma) library. This prevents the HP-UX undump() routine (in | |
| 1970 unexhp9k800.c) from properly working. What's happening is that some | |
| 1971 initialization routines are being called in the libcma library, *BEFORE* | |
| 1972 main() is called, and these initialization routines are calling | |
| 1973 malloc(). Unfortunately, in order for the undumper to work, XEmacs must | |
| 1974 adjust (move upwards) the sbrk() value *BEFORE* the first call to | |
| 1975 malloc(); if malloc() is called before XEmacs has properly adjusted sbrk | |
| 1976 (which is what is happening), dumped memory that is being used by | |
| 1977 XEmacs, is improperly re-allocated for use by malloc() and the dumped | |
| 1978 memory is corrupted. This causes XEmacs to die an horrible death. | |
| 1979 | |
| 1980 It is believed that versions of the audio library past December 1998 | |
| 1981 will trigger this problem. Under HP-UX 10.20, you probably have to | |
| 1982 install audio library patches to encounter this. It's probable that | |
| 1983 recent "fresh, out-of-the-box" HP-UX 11.XX workstations also have this | |
| 1984 problem. For HP-UX 10.20, it's believed that audio patch PHSS_17121 (or | |
| 1985 a superceeding one, like PHSS_17554, PHSS_17971, PHSS_18777, PHSS_21481, | |
| 1986 or PHSS_21662, etc.) will trigger this. | |
| 1987 | |
| 1988 To check if your audio library will cause problems for XEmacs, run | |
| 1989 "chatr /opt/audio/lib/libAlib.sl". If "libdce" appears in the displayed | |
| 1990 shared library list, XEmacs will probably encounter problems if audio is | |
| 1991 enabled. | |
| 1992 | |
| 1993 **** Workaround: | |
| 1994 | |
| 1995 Don't enable native audio. Re-run configure without native audio | |
| 1996 support. | |
| 1997 | |
| 1998 If your site supports it, try using NAS (Network Audio Support). | |
| 1999 | |
| 2000 Try using the experimental portable dumper. It may work, or it may | |
| 2001 not. | |
| 2002 | |
| 2003 | |
| 373 | 2004 *** `Pid xxx killed due to text modification or page I/O error' |
| 2005 | |
| 2006 On HP-UX, you can get that error when the Emacs executable is on an NFS | |
| 2007 file system. HP-UX responds this way if it tries to swap in a page and | |
| 2008 does not get a response from the server within a timeout whose default | |
| 2009 value is just ten seconds. | |
| 2010 | |
| 2011 If this happens to you, extend the timeout period. | |
| 2012 | |
| 2013 *** Shell mode on HP-UX gives the message, "`tty`: Ambiguous". | |
| 124 | 2014 |
| 2015 christos@theory.tn.cornell.edu says: | |
| 2016 | |
| 2017 The problem is that in your .cshrc you have something that tries to | |
| 197 | 2018 execute `tty`. If you are not running the shell on a real tty then tty |
| 2019 will print "not a tty". Csh expects one word in some places, but tty | |
| 2020 is giving it back 3. | |
| 124 | 2021 |
| 197 | 2022 The solution is to add a pair of quotes around `tty` to make it a |
| 2023 single word: | |
| 0 | 2024 |
| 454 | 2025 if (`tty` == "/dev/console") |
| 124 | 2026 |
| 2027 should be changed to: | |
| 2028 | |
| 454 | 2029 if ("`tty`" == "/dev/console") |
| 124 | 2030 |
| 2031 Even better, move things that set up terminal sections out of .cshrc | |
| 2032 and into .login. | |
| 0 | 2033 |
| 2034 | |
| 373 | 2035 ** SCO |
| 2036 *** Regular expressions matching bugs on SCO systems. | |
| 0 | 2037 |
| 373 | 2038 On SCO, there are problems in regexp matching when Emacs is compiled |
| 2039 with the system compiler. The compiler version is "Microsoft C | |
| 2040 version 6", SCO 4.2.0h Dev Sys Maintenance Supplement 01/06/93; Quick | |
| 2041 C Compiler Version 1.00.46 (Beta). The solution is to compile with | |
| 2042 GCC. | |
| 124 | 2043 |
| 88 | 2044 |
| 373 | 2045 ** Windows |
| 1332 | 2046 *** Conflicts with FSF NTEmacs |
| 2047 | |
| 2048 Depending on how it is installed, FSF NTEmacs may setup various EMACS* | |
| 2049 variables in your environment. The presence of these variables may | |
| 2050 cause XEmacs to fail at startup, cause you to see corrupted | |
| 2051 doc-strings, or cause other random problems. | |
| 2052 | |
| 2053 You should remove these variables from your environment. These | |
| 2054 variables are not required to run FSF NTEmacs if you start it by | |
| 2055 running emacs.bat. | |
| 2056 | |
| 2057 *** XEmacs can't find my init file | |
| 2058 | |
| 2059 XEmacs looks for your init in your "home" directory -- either in | |
| 2060 `~/.xemacs/init.el' or `~/.emacs'. XEmacs decides that your "home" | |
| 2061 directory is, in order of preference: | |
| 2062 | |
| 2063 - The value of the HOME environment variable, if the variable exists. | |
| 2064 - The value of the registry entry SOFTWARE\XEmacs\XEmacs\HOME, | |
| 2065 if it exists. | |
| 2066 - The value of the HOMEDRIVE and HOMEPATH environment variables, if | |
| 2067 these variables both exist. | |
| 2068 - C:\. | |
| 2069 | |
| 2070 To determine what XEmacs thinks your home directory is, try opening | |
| 2071 a file in the `~' directory, and you should see its expansion in the | |
| 2072 modeline. If this doesn't work, type ESC : (user-home-directory). | |
| 2073 | |
| 2074 *** XEmacs can't find any packages | |
| 2075 | |
| 2076 XEmacs looks for your packages in subdirectories of a directory which | |
| 2077 is set at compile-time (see `config.inc'), and whose default is | |
| 2078 `C:\Program Files\XEmacs'. XEmacs also looks in `~/.xemacs', where | |
| 2079 `~' refers to your home directory (see previous entry). The variable | |
| 2080 `configure-package-path' holds the actual path that was compiled into | |
| 2081 your copy of XEmacs. | |
| 2082 | |
| 2083 The compile-time default location can be overridden by the EMACSPACKAGEPATH | |
| 2084 environment variable or by the SOFTWARE\XEmacs\XEmacs\EMACSPACKAGEPATH | |
| 2085 registry entry. You should check that these variables, if they exist, | |
| 2086 point to the actual location of your package tree. | |
| 2087 | |
| 2088 *** XEmacs doesn't die when shutting down Windows 95 or 98 | |
| 2089 | |
| 2090 When shutting down Windows 95 or 98 you may see a dialog that says | |
| 2091 "xemacs / You must quit this program before you quit Windows". | |
| 2092 It is safe to | |
| 2093 "Click OK to quit the program and Windows", | |
| 2094 but you won't be offered a chance to save any modified XEmacs buffers. | |
| 2095 | |
| 2096 *** Key bindings | |
| 2097 | |
| 2098 The C-z, C-x, C-c, and C-v keystrokes have traditional uses in both | |
| 2099 emacs and Windows programs. XEmacs binds these keys to their | |
| 2100 traditional emacs uses, and provides Windows 3.x style bindings for | |
| 2101 the Cut, Copy and Paste functions. | |
| 2102 | |
| 2103 Function XEmacs binding | |
| 2104 -------- -------------- | |
| 2105 Undo C-_ | |
| 2106 Cut Sh-Del | |
| 2107 Copy C-Insert | |
| 2108 Paste Sh-Insert | |
| 2109 | |
| 2110 You can rebind keys to make XEmacs more Windows-compatible; for | |
| 2111 example, to bind C-z to undo: | |
| 2112 | |
| 2113 (global-set-key [(control z)] 'undo) | |
| 2114 | |
| 2115 Rebindind C-x and C-c is trickier because by default these are prefix | |
| 2116 keys in XEmacs. See the "Key Bindings" node in the XEmacs manual. | |
| 2117 | |
| 2118 *** Behavior of selected regions | |
| 2119 | |
| 2120 Use the pending-del package to enable the standard Windows behavior of | |
| 2121 self-inserting deletes region. | |
| 2122 | |
| 2123 *** Limitations on the use of the AltGr key. | |
| 2124 | |
| 2125 In some locale and OS combinations you can't generate M-AltGr-key or | |
| 2126 C-M-AltGr-key sequences at all. | |
| 2127 | |
| 2128 To generate C-AltGr-key or C-M-AltGr-key sequences you must use the | |
| 2129 right-hand Control key and you must press it *after* AltGr. | |
| 2130 | |
| 2131 These limitations arise from fundamental problems in the way that the | |
| 2132 win32 API reports AltGr key events. There isn't anything that XEmacs | |
| 2133 can do to work round these problems that it isn't already doing. | |
| 2134 | |
| 2135 You may want to create alternative bindings if any of the standard | |
| 2136 XEmacs bindings require you to use some combination of Control or Meta | |
| 2137 and AltGr. | |
| 2138 | |
| 2139 *** Limited support for subprocesses under Windows 9x | |
| 2140 | |
| 2141 Attempting to use call-process to run a 16bit program gives a | |
| 2142 "Spawning child process: Exec format error". For example shell-command | |
| 2143 fails under Windows 95 and 98 if you use command.com or any other | |
| 2144 16bit program as your shell. | |
| 2145 | |
| 2146 XEmacs may incorrectly quote your call-process command if it contains | |
| 2147 double quotes, backslashes or spaces. | |
| 2148 | |
| 2149 start-process and functions that rely on it are supported under Windows 95, | |
| 2150 98 and NT. However, starting a 16bit program that requires keyboard input | |
| 2151 may cause XEmacs to hang or crash under Windows 95 and 98, and will leave | |
| 2152 the orphaned 16bit program consuming all available CPU time. | |
| 2153 | |
| 2154 Sending signals to subprocesses started by call-process or by | |
| 2155 start-process fails with a "Cannot send signal to process" error under | |
| 2156 Windows 95 and 98. As a side effect of this, quitting XEmacs while it | |
| 2157 is still running subprocesses causes it to crash under Windows 95 and | |
| 2158 98. | |
| 524 | 2159 |
| 2160 | |
| 2161 ** Cygwin | |
| 1318 | 2162 *** Signal 11 when building or running a dumped XEmacs. |
| 2163 | |
| 2164 See the section on Cygwin above, under building. | |
| 2165 | |
| 1058 | 2166 *** XEmacs fails to start because cygXpm-noX4.dll was not found. |
| 2167 | |
| 2168 Andy Piper <andy@xemacs.org> sez: | |
| 2169 | |
| 2170 cygXpm-noX4 is part of the cygwin distribution under libraries or | |
| 2171 graphics, but is not installed by default. You need to run the | |
| 2172 cygwin setup again and select this package. | |
| 2173 | |
| 524 | 2174 *** Subprocesses do not work. |
| 2175 | |
| 2176 You do not have "tty" in your CYGWIN environment variable. This must | |
| 2177 be set in your autoexec.bat (win95) or the system properties (winnt) | |
| 2178 as it must be read before the cygwin DLL initializes. | |
| 2179 | |
| 2180 *** ^G does not work on hung subprocesses. | |
| 124 | 2181 |
| 524 | 2182 This is a known problem. It can be remedied by defining BROKEN_SIGIO |
| 2183 in src/s/cygwin.h, however this currently leads to instability in XEmacs. | |
| 2184 (#### is this still true?) | |
| 2185 | |
| 2186 *** Errors from make like `/c:not found' when running `M-x compile'. | |
| 308 | 2187 |
| 524 | 2188 Make sure you set the environment variable MAKE_MODE to UNIX in your |
| 2189 init file (.xemacs/init.el), Control Panel (Windows 2000/NT), or | |
| 2190 AUTOEXEC.BAT (Windows 98/95). | |
| 2191 | |
| 2192 *** There are no images in the toolbar buttons. | |
| 2193 | |
| 2194 You need version 4.71 of commctrl.dll which does not ship with windows | |
| 2195 95. You can get this by installing IE 4.0 or downloading it from the | |
| 2196 microsoft website. | |
| 308 | 2197 |
| 197 | 2198 |
| 124 | 2199 * Compatibility problems (with Emacs 18, GNU Emacs, or previous XEmacs/lemacs) |
| 197 | 2200 ============================================================================== |
| 88 | 2201 |
| 373 | 2202 *** "Symbol's value as variable is void: unread-command-char". |
| 197 | 2203 "Wrong type argument: arrayp, #<keymap 143 entries>" |
| 2204 "Wrong type argument: stringp, [#<keypress-event return>]" | |
| 88 | 2205 |
| 124 | 2206 There are a few incompatible changes in XEmacs, and these are the |
| 2207 symptoms. Some of the emacs-lisp code you are running needs to be | |
| 2208 updated to be compatible with XEmacs. | |
| 2209 | |
| 2210 The code should not treat keymaps as arrays (use `define-key', etc.), | |
| 2211 should not use obsolete variables like `unread-command-char' (use | |
| 197 | 2212 `unread-command-events'). Many (most) of the new ways of doing things |
| 124 | 2213 are compatible in GNU Emacs and XEmacs. |
| 88 | 2214 |
| 197 | 2215 Modern Emacs packages (Gnus, VM, W3, efs, etc) are written to support |
| 2216 GNU Emacs and XEmacs. We have provided modified versions of several | |
| 2217 popular emacs packages (dired, etc) which are compatible with this | |
| 2218 version of emacs. Check to make sure you have not set your load-path | |
| 2219 so that your private copies of these packages are being found before | |
| 2220 the versions in the lisp directory. | |
| 124 | 2221 |
| 2222 Make sure that your load-path and your $EMACSLOADPATH environment | |
| 2223 variable are not pointing at an Emacs18 lisp directory. This will | |
| 2224 cripple emacs. | |
| 88 | 2225 |
| 124 | 2226 ** Some packages that worked before now cause the error |
| 223 | 2227 Wrong type argument: arrayp, #<face ... > |
| 124 | 2228 |
| 197 | 2229 Code which uses the `face' accessor functions must be recompiled with |
| 2230 xemacs 19.9 or later. The functions whose callers must be recompiled | |
| 2231 are: face-font, face-foreground, face-background, | |
| 2232 face-background-pixmap, and face-underline-p. The .elc files | |
| 2233 generated by version 19.9 will work in 19.6 and 19.8, but older .elc | |
| 2234 files which contain calls to these functions will not work in 19.9. | |
| 124 | 2235 |
| 2236 ** Signaling: (error "Byte code stack underflow (byte compiler bug), pc 38") | |
| 88 | 2237 |
| 120 | 2238 This error is given when XEmacs 20 is compiled without MULE support |
| 88 | 2239 but is attempting to load a .elc which requires MULE support. The fix |
| 2240 is to rebytecompile the offending file. | |
| 2241 | |
| 124 | 2242 ** Signaling: (wrong-type-argument ...) when loading mail-abbrevs |
| 88 | 2243 |
| 197 | 2244 The is seen when installing the Insidious Big Brother Data Base (bbdb) |
| 2245 which includes an outdated copy of mail-abbrevs.el. Remove the copy | |
| 2246 that comes with bbdb and use the one that comes with XEmacs. | |
| 2247 | |
| 144 | 2248 |
| 2249 * MULE issues | |
| 197 | 2250 ============= |
| 144 | 2251 |
| 223 | 2252 ** A reminder: XEmacs/Mule work does not currently receive *any* |
| 2253 funding, and all work is done by volunteers. If you think you can | |
| 2254 help, please contact the XEmacs maintainers. | |
| 2255 | |
| 278 | 2256 ** XEmacs/Mule doesn't support TTY's satisfactorily. |
| 223 | 2257 |
| 2258 This is a major problem, which we plan to address in a future release | |
| 2259 of XEmacs. Basically, XEmacs should have primitives to be told | |
| 2260 whether the terminal can handle international output, and which | |
| 2261 locale. Also, it should be able to do approximations of characters to | |
| 2262 the nearest supported by the locale. | |
| 2263 | |
| 197 | 2264 ** Internationalized (Asian) Isearch doesn't work. |
| 144 | 2265 |
| 2266 Currently, Isearch doesn't directly support any of the input methods | |
| 2267 that are not XIM based (like egg, canna and quail) (and there are | |
| 223 | 2268 potential problems with XIM version too...). If you're using egg |
| 2269 there is a workaround. Hitting <RET> right after C-s to invoke | |
| 2270 Isearch will put Isearch in string mode, where a complete string can | |
| 2271 be typed into the minibuffer and then processed by Isearch afterwards. | |
| 2272 Since egg is now supported in the minibuffer using string mode you can | |
| 2273 now use egg to input your Japanese, Korean or Chinese string, then hit | |
| 2274 return to send that to Isearch and then use standard Isearch commands | |
| 2275 from there. | |
| 144 | 2276 |
| 223 | 2277 ** Using egg and mousing around while in 'fence' mode screws up my |
| 2278 buffer. | |
| 144 | 2279 |
| 2280 Don't do this. The fence modes of egg and canna are currently very | |
| 2281 modal, and messing with where they expect point to be and what they | |
| 2282 think is the current buffer is just asking for trouble. If you're | |
| 2283 lucky they will realize that something is awry, and simply delete the | |
| 2284 fence, but worst case can trash other buffers too. We've tried to | |
| 2285 protect against this where we can, but there still are many ways to | |
| 2286 shoot yourself in the foot. So just finish what you are typing into | |
| 2287 the fence before reaching for the mouse. | |
| 223 | 2288 |
| 2289 ** Not all languages in Quail are supported like Devanagari and Indian | |
| 2290 languages, Lao and Tibetan. | |
| 2291 | |
| 2292 Quail requires more work and testing. Although it has been ported to | |
| 2293 XEmacs, it works really well for Japanese and for the European | |
| 2294 languages. | |
| 2295 | |
| 2296 ** Right-to-left mode is not yet implemented, so languages like | |
| 2297 Arabic, Hebrew and Thai don't work. | |
| 2298 | |
| 2299 Getting this right requires more work. It may be implemented in a | |
| 2300 future XEmacs version, but don't hold your breath. If you know | |
| 2301 someone who is ready to implement this, please let us know. | |
| 2302 | |
| 2303 ** We need more developers and native language testers. It's extremely | |
| 2304 difficult (and not particularly productive) to address languages that | |
| 2305 nobody is using and testing. | |
| 2306 | |
| 2307 ** The kWnn and cWnn support for Chinese and Korean needs developers | |
| 2308 and testers. It probably doesn't work. | |
| 2309 | |
| 2310 ** There are no `native XEmacs' TUTORIALs for any Asian languages, | |
| 454 | 2311 including Japanese. FSF Emacs and XEmacs tutorials are quite similar, |
| 223 | 2312 so it should be sufficient to skim through the differences and apply |
| 2313 them to the Japanese version. | |
| 2314 | |
| 2315 ** We only have localized menus translated for Japanese, and the | |
| 2316 Japanese menus are developing bitrot (the Mule menu appears in | |
| 2317 English). | |
| 2318 | |
| 2319 ** XIM is untested for any language other than Japanese. |