[bitc-dev] bitc and coroutines

[Jonathan S. Shapiro]

On Sat, 2006-12-02 at 02:04 +0000, Sam Mason wrote:
> The reason I'm mentioning them here is that it sounds as though Coyotos
> is going to be using scheduler activations.  Coroutines appear to me to
> be quite a natural way of expressing what (I expect) to be happening in
> the user land part of the scheduler.  I don't know how much you mind the
> user level scheduler existing as an opaque blob of assembly code though.

Portions of the activation handler clearly need to be in assembler
because they deal with register-level interactions.

The coroutines question is indeed interesting, and at the moment I don't
have any answer for you except to say that we should add things to BitC
very cautiously. I'm not convinced that the language should provide
support for threading or co-routines. Speaking from my perspective as a
programmer, most language support for these features is badly flawed.

shap