[slime-devel] Re: serious xemacs/slime bug

[Helmut Eller]

Surendra Singhi <efuzzyone at netscape.net> writes:

> I am using slime-1.0 and Xemacs 21.13.
> At times while running a lisp program through slime due to some error
> in the code the slime connection terminates. And then when I try to
> execute some command which if accidently depends upon slime then my
> xemacs hangs waiting for slime to respond, which on turn is waiting on
> clisp or someone.
> 
> This causes the whole xemacs to hang, and I am not able to execute any
> command, though I am able to maximiaze/minimize, look at the menu bar
> etc. I feel there is some serious multi-threading issue which needs to
> be resolved(if it hasn't already been).

C-g should also work it this situation.  Emacs will then pop up the
Emacs Lisp debugger.

Surendra Singhi <efuzzyone at netscape.net> writes:

> I am using clisp system, and sometimes due to some error in the
> program, when the clisp goes into an infinite loop, slime terminates
> the connection and shows "evaluation aborted" error message.
> And this is when the error in the orignial post occurs. The way out of
> it for me is to restart the slime connection. and luckily, when I
> restart the slime conncetion the memory image which was present before
> the command which caused connection disruption occurred is reloaded,
> i.e., I don't need to redefine everything.

This sounds like a bug in swank-clisp.

> Another problem which I am facing is that I am not able to abort
> evaluation on clisp from slime, I don't know whether it is a bug in
> clisp or slime.

This sounds like a Windows issue.  It works on Linux.

If you want to see these bugs fixed, you have to give some more
details.  Provide enough information so that we can reproduce the bug.
Describe what you are actually doing.  Include the contents of the
*slime-events* buffer, the error message in the *inferior-lisp*
buffer, the exact version of CLISP, and backtraces if available.

Helmut.