I've already got many answers, thanks to all who responded, they're
too many to list here. The unanimous (!!) answer is to use `kbd_mode
-a' to restore the keyboard, after making sure that Xsun isn't running
any more, of course. I couldn't test it yet, since the problem didn't
show up yet, as Murphy predicts.
The question of why this happens isn't solved, however. Xsun SHOULD
exit cleanly, but for some reason it doesn't, just hangs. There should
be no reason to have to put `kbd_mode -a' at the end of .xinitrc! The
man page of xinit says that when the script is done it will kill the
server and exit. Apparently xinit is sending a signal to Xsun, since
all the windows disappear. In fact, I think just sticking this at the
end of the script may not be enough, since Xsun might still be around.
Conclusion: it seems there's a bug in Xsun. Might be some race
condition, since this doesn't happen every time.
The question about how L1-A works wasn't answered yet. I think the
invocation of the ROM monitor must be done by unix itself, since a key
press is received by the cpu as an event in the port, which is handled
by the kernel directly, unless the kernel runs on top of the monitor,
which then intercepts the key press. If you know how this works and it
isn't a Sun trade secret please tell me and I'll summarize (if it's a
secret just tell me so, anyway).
Carlos
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:07:54 CDT