[Openmcl-devel] Asynchronous callback made from real-time thread

[Gary Byers]

On Mon, 2 Jun 2003, Stephane Letz wrote:

> What this the relation between the OpenMCL projet and the Digitool MCL
> version? will the new natives thread developement be added in the Digitool
> MCL version at some point?
>

Aside from the fact that they still share a fair amount of code, there
isn't any real relationship between MCL and OpenMCL.  Aside from a few
bug fixes, I can't think of any cases where OpenMCL code has made it
back into MCL.

Digitool wanted MCL 5 to run under both OS9/CarbonLib and OSX; that
constraint makes it difficult to incorporate OpenMCL features that
depend on a Unix-like OS.  I did some work on the MCL 5 kernel a few
months ago, and I found that that constraint made it much harder than
it should have been to fix the problems that I was trying to fix.

I have a bit of difficulty imagining how MCL's event system (which is
based on #_WaitNextEvent and closely intertwined with the cooperative
scheduler) would work with native threads.  I don't know how much code
would have to change if MCL switched to a more modern event model
(Carbon Events); I imagine that this change would have had a
significant impact on user code as well as on MCL itself.  Since
OS9/CarbonLib supports Carbon Events, I assume that the reasons that
this change was not made in MCL 5 were not strictly technical.

The last time I heard, a significant majority of Mac users were still
using OS9, and the rate of migration to OSX isn't as high as I would
have expected.  I don't know if that necessarily makes Digitool's
conservatism with respect to MCL correct, but in that context it's
at least understandable.

_______________________________________________
Openmcl-devel mailing list
Openmcl-devel at clozure.com
http://clozure.com/cgi-bin/mailman/listinfo/openmcl-devel