[bitc-dev] BitC status update

[Jonathan S. Shapiro]

On Sun, 2005-09-04 at 23:50 +0100, David Hopwood wrote:
> Jonathan S. Shapiro wrote:
> > I should perhaps add that my current focus of attention is generating
> > code for a garbage-collected BitC implementation. Basically, I want to
> > get us self-hosted as fast as possible.
> 
> Have you considered C--?

In response to your note, I went over and took a look. I think that we
will not use C-- at this time. I have a couple of concerns with C--
intrinsically as a research back end, but the main issue is that we
probably want to generate C or C++ for non-GC'd mode anyway, and I would
prefer to use the same target language for both back ends. C-- doesn't
appear to address our needs in that area. It is possible that I am being
short-sighted here. I will definitely re-examine C-- when the time comes
to look at a serious BitC back end.

Another issue is that the stability of C-- seems dodgy at the moment
(based on their release notes).

Overall, C-- looks like an EXCELLENT job!

I did send a note to the C-- list enumerating the issues I saw.


shap