[Openmcl-devel] Slime still doesn't like new flavor

Derrick Baumer derrick at bigspace-project.com
Wed Aug 31 18:13:27 UTC 2005

It appears I was premature in declaring all things good, referring to
the use of the CVS OpenMCL w/ Slime~Emacs under Ubuntu(Debian)
(pre-TLS?) Linux. Slime does, indeed, start up and smile beautifully for
me w/ the semaphore bug work-around reinstated, but when actually trying
to compile (w/ C-c k) or load an application (w/ C-c l), Emacs likes to
lock up solid (gotta use kill from the command line to close the window)
or (not consistent) say "Lisp connection closed unexpectedly: exited
abnormally with code 256" at the bottom.

In the second case, there appear to be no further diagnostic messages to
give you.  When I type "M-x slime" again, though, it asks if I want to
create an additional inferior lisp (because the old one is still
running).  Behavior is the same w/ Slime 1.2.1 and CVS.  I *do* realize
this *really* seems like a problem with Slime, especially considering
that my application runs fine from the terminal REPL, and I'll be happy
to subscribe to their devel list and pass this buck on to them if y'all
would like.  My only hesitation is that the same version(s) of Slime
worked well under 0.14.3.

Additionally, if I avoid using the control key combinations and just 
type the compile and load commands into the Slime REPL, it works well 
for an inconsistent number of times, then tells me that I'm not 
connected and to use M-x slime to start an inferior lisp.  When I do so, 
however, it asks the same - do I want to create an additional 

It would seem to my novice eyes that somewhere along the way Slime loses 
track of OpenMCL and thinks it has shut down; though, OpenMCL seems to 
think nothing's wrong and keeps plugging away waiting for commands.

I am downloading Fedora Core 4 as I write this - I do want to use 
OpenMCL for my tinkering, so it'll be available to me should this issue 
not be resolvable.  BTW, did I forget to say thank you for all your hard 
work?  Thank you.

Derrick Baumer

More information about the Openmcl-devel mailing list