[slime-devel] Re: A possible bug in slime

[Luke Gorrie]

Ignas Mikalajunas <voblia at centras.lt> writes:

>   If you start fiddling with *standard-input* in slime
> like (read *standard-input*) slime becomes unstable.
> 
> I just tried reading from *standard-input* then my program 
> started computations and i tried to do Ctrl+C in Eterm
> where cmucl was running. You can see the result.
> 
> I have attached the log.

Thanks. It appears the problem is that ^C'ing CMUCL put it into the
debugger, but our protocol forbids Lisp to enter the debugger while
it's supposed to be waiting while reading input. One of those
"asynchronous errors" that we don't handle very well right now.

You might have to live with it for a little while :-). I'm
protocol-hacking currently, but first of all so that Emacs can connect
to multiple Lisps at once. If this works well we can use it as Gary
Byers described for multithreaded operation, which would also give us
the nice "use two debuggers at once" feature Peter Siebel suggested.

Cheers,
Luke