Matthias Horn <[EMAIL PROTECTED]> writes:
> >
> >the update continuation has both a return vector and a direct return
> >address, so any kind of return can be handled by the update frame.
> >
> >
> But wouldn't this imply that each binding has to be compiled in two
> versions. One versio
Hello Simon, hello Kevin,
At first thanks for your replies. But I still have some questions.
On 19 Mar 1998 10:06:17 +, Simon Marlow wrote:
>
>
>the update continuation has both a return vector and a direct return
>address, so any kind of return can be handled by the update frame.
>
Hi Matthias,
At 2:26 pm 18/3/98, Haskell List Moderator wrote:
>After this long introduction my question:
>Suppose a polymorphic closure is evaluated in parallel by another
>processing element. If the resulting type is a type which uses a
>vectored return, the remote processor needs to know this
Haskell List Moderator <[EMAIL PROTECTED]> writes:
[ long question about polymorphic update frames in conjunction with
parallel execution deleted... ]
> Can anybody explain how this problem is solved in the GPH
> implementation?
I don't know how it currently works, but in the new runtime-system
[Forwarded from Haskell list]
I have got a rather special question about the implementation of
Glasgow Parallel Haskell (GPH). GPH uses the `par` operator to
specify expressions which can be evaluated in parallel by another
processing element. Closures representing such an expression are
shipped
