tlug: Yutex

["Stephen J. Turnbull" <turnbull@example.com>]

>>>>> "Craig" == Craig Oda <craig@example.com> writes:

    Craig> I'll probably give the compile a shot in a few days.  I
    Craig> have an old copy of Mootiff with Motif 2.0 from March 1995.
    Craig> I'm not even sure if these are elf binaries or not.

If it's the same Mootiff I have (I think it is), you get both.

    Craig> I'm also looking at lesstif http://www.lesstif.org.  Gaspar
    Craig> mentioned that Motif was needed to input Japanese.

    Craig> I read on fj.os.linux that mule now compiles with Lesstif,
    Craig> so I'm thinking of giving Lesstif a shot.  Gaspar or anyone
    Craig> else give this a try yet?

I wouldn't bet on Lesstif supporting XIM properly.  XIM is the only
reason I can think of that Motif would be useful in supporting
Japanese.

I don't know about That Other Emacs, but XEmacs has compiled with
Motif for years now, and Lesstif works with it as well as Lesstif ever
works.  Many people on the XEmacs beta list swear by the Motif
versions on proprietary, well-supported platforms like Sun.  Until
Lesstif 0.81, the event loop was extremely buggy, making XEmacs almost
unusable for anything but developing XEmacs---make sure you get the
most recent Lesstif.  Menubars and stuff with Motif are a no-no, more
core dumps.  I think the dialogs work ok though.

    Craig> Also, any help on the core dumps would be nice.

bash$ gdb myutex core
... gdb splash screen ...
gdb> where

gives a backtrace.  Probably you will need to go deep into it to find
where in myutex it's happening, typically these things wend their way
through various libraries before actually exploding.  Often it will be
something like a NULL pointer to a string function, or an
uninitialized pointer anywhere.

If it stopped due to a signal, that's useful information.  SIGSEGV
almost surely means a bogus pointer, while SIGBUS means an alignment
problem (I think); this is also typically a bogus pointer.  In garbage 
collecting environments, the low bits of a pointer may be used for
mark bits; this would also generate that signal.

gdb of course gives lots of other information if you know how to use
it, but that often depends on intimate knowledge of implementation
and/or hardware details.

A useful trick for getting a log of your gdb session (gdb may
implement logs internally, but I couldn't find it) is

gdb myutex core | tee gdb.log

This has the disadvantage that gdb does not use its internal pager,
but OTOH you can review the log later and even mail it to TLUG.

---------------------------------------------------------------
TLUG Meeting Dec. 13, 12:30 at Tokyo station Yaesu Chuo ticket gate
13:30 Starbuck's coffee.  13:45 HSBC | info: joem@example.com
At least 3 functional Sparc IPC machines will be raffled out
---------------------------------------------------------------
a word from the sponsor:
TWICS - Japan's First Public-Access Internet System
www.twics.com  info@example.com  Tel:03-3351-5977  Fax:03-3353-6096