Raymond Toy <[EMAIL PROTECTED]> writes:

>     Greg>   30: (SWANK-BACKEND::SIGIO-HANDLER #<#1=unused-arg> #<#1#> #<#1#>)
>
> I don't think CMUCL is interrupt-safe, so using sigio-handler with
> slime can cause weird failures.  At least it happens once in a while
> on Sparc; I think x86 is a bit better here.

I'd like to point out that a SIGIO handler is no less robust than a
SIGINT handler.  So, if you experience weird failures with Slime you
should also experience them if you run CMUCL in a xterm and press ^C.

Helmut.


Reply via email to