VM-level scheduling (was "Flow 2 alpha 1 released")

Craig Latta craig.latta at netjam.org
Sat Aug 17 02:12:39 UTC 2002


Hi Tim--

> It ain't gonna work on Acorn without a lot of work that may not be
> worth it.

	Hmm. I just found a RiscOS pthreads library that's part of Rozilla
(http://www.wellytop.com/Rozilla/pthread.html). Should I not bother with
it?

> Even I have to admit that not many people actually use Squeak on an
> Acorn :-(

	Despite my wisecracks... I'd say the number is still sufficient. Of
course, if you drop the platform for some reason, then... :)

> I wonder if there is any possible way to avoid them?

	There surely is, but in my opinion it's a lot more work than just using
an available threads library (which there seems to be, on all Squeak
platforms).

	But even in considering Squeak on metal, I still think it makes sense
to implement similiar threading behavior at the VM level. E.g.,
incorporate select parts of a freely-available Unix into the VM (as has
been discussed here recently). I think it's a big win to implement some
of the scheduling details below the object level, to keep the object
level simple.

> Somewhere in your code and comments I spotted a musing about whether
> one can have plugins that depend upon others as a way to split
> things up more. Absolutely we can. They can load each other,
> unload or simply do cleanup when a companion is unloaded, all that
> good stuff. It would make ports to other machines easier if unneeded
> parts could simply be ignored rather than being stubbed out.

	I agree; after Flow gets more transport coverage (1394, etc.) I'm going
to see how smoothly the whole installation/usage process goes with some
separation.


	thanks for the feedback!

-C

--
Craig Latta
improvisational musical informaticist
craig at netjam.org
www.netjam.org/resume
Smalltalkers do: [:it | All with: Class, (And love: it)]



More information about the Squeak-dev mailing list