[slime-devel] Re: slime-interrupt in openmcl

[Sven Van Caekenberghe]

On 15 Dec 2003, at 15:27, Luke Gorrie wrote:

> Alan Ruttenberg <alanralanr at comcast.net> writes:
>
>> Then add this redefinition
>>
>> (defun ccl::force-break-in-listener (p)
>>    (ccl::process-interrupt p
>
> Is there something extra needed? I get this in "(Alpha: Darwin) 
> 0.14-031018":
>
>> Error in process listener(1): The function 
>> CCL::FORCE-BREAK-IN-LISTENER is predefined in OpenMCL.
>> While executing: CCL::REDEFINE-KERNEL-FUNCTION
>> Type :GO to continue, :POP to abort.
>> If continued: Replace the definition of CCL::FORCE-BREAK-IN-LISTENER.
>   Type :? for other options.

 From a user standpoint, you start some long computation, type c-c c-c, 
and then I would expect to be thrown in the SLIME debugger, with the 
option continue or abort, expect them to behave accordingly.
Now I get

CL-USER> (dotimes (i 10) (print i) (sleep 1))

0
1
 > Break in process listener(1):
 > While executing: #<Anonymous Function #x512AD0E>
 > Type :GO to continue, :POP to abort.
 > If continued: Return from BREAK.
Type :? for other options.
1 > :go

This shows up in the REPL, and everything is broken when I continue..
Maybe I am doing something wrong...

Sven