Thanks Tim,
This is a known bug which we haven't been able to track down properly yet.
I think it appeared in redhat 5.0 (glibc problem??)
I stumbled across a fix a month or so back but didn't
have time to test it properly at the time and can't
remember it too clearly now. I think it was either:
o change all uses of setjmp to sigsetjmp; or
o #define HANDLERS_CANT_LONGJMP 1 in prelude.h
(The new Hugs-GHC system handles interrupts in a different
way which avoids this problem - which is why I didn't pursue this
any further.)
Alastair
> hugs release: hugs-June98-source.tgz
> environment: RH 5.0 Linux 2.0.31 i686, gcc 2.7.2.3, libc-2.0.5
>
> Second use of keyboard interrupt ^C won't stop interpreter.
>
> This behaviour seems same whether readline, editline, or neither, was
> configured. (I remember seeing reports where same effect was apparently
> attributed to readline)
>
> Am I the only user ever to try non-terminating expressions twice in one
> session, or is there something broken in this environment?
>
> TWP
