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