[slime-devel] Re: C-M-x kills emacs

[Helmut Eller]

Jeffrey Cunningham <jeffrey at cunningham.net> writes:

> WRITE: (:return (:ok (15086 "CMU Common Lisp" "cmucl" (:asdf :mk-defsystem :common-lisp-controller :gerds-pcl :pcl-structures :portable-commonloops :pcl :cmu19 :cmu19a :python :conservative-float-type :modular-arith :mp :x86 :linkage-table :relative-package-names :linux :glibc2 :unix :random-mt19937 :gencgc :pentium :i486 :hash-new :heap-overflow-check :stack-checking :common :common-lisp :ansi-cl :ieee-floating-point :cmu))) 1)
>
> ;; Connection to Emacs lost.
> ;; [End-of-File on #<Stream for descriptor 7>]
>
>
>
> Not very illuminating (to me). The last two lines are what happened
> when I pressed C-M-x in emacs after connecting it to the server. 

For C-M-x, Emacs usually sends a request to the server and the server
sends the answer back.  But apparently Emacs crashes before sending
the request.

Just a stupid question: C-M-x isn't perchance hijacked by your window
manager?

> Yes, this seems to be working properly. I can use the interpeter in
> the *slime-repl[1]* buffer just fine. And stuff happens over in the
> xterm with the server. If I hit RET by itself, I get:
>
> DISPATCHING: (:EMACS-REX
>               (SWANK:LISTENER-EVAL "
> ")
>               "COMMON-LISP-USER" :REPL-THREAD 5)
> DISPATCHING: (:RETURN #<Process Initial {580087CD}> (:OK "; No value") 5)
> WRITE: (:return (:ok "; No value") 5)

Very strange.  C-M-x does basically the same.

I'm running out of ideas.  Perhaps you can find out something if you
try to debug the Emacs Lisp code yourself.  Get the SLIME CVS version
and step through the `slime-eval-defun' function (that's the function
for the C-M-x key).  Type `M-x debug-on-entry RET slime-eval-defun'
before pressing C-M-x.  In the debugger buffer, you can then step with
the `d' and `c' keys.

Sorry for not being more helpful.

Helmut.