On Thu, Jul 14, 2005 at 09:36:54PM -0500, Alan Grimes wrote:
...
> There was a discussion about preemptive tasking in the image (that would
> permit the image to run on a threaded VM) a few weeks ago but the few
> (and I really mean few as in there are only a dozen people on this
> planet who are qualified to actually do significant changes to the VM)
> people who could make the improvements refused to 

Well, they explained why they did not regard it as a priority,
particularly given the substantial work that would be required.
squeak is in significant part a volunteer effort, and to the extent
people are paid to work on it they are paid to work on specific
things.

While the ability to use multiple processors does strike me as
increasingly important, some in the discussion did point out you can
use multiple processors by running multiple images.  While that may
not be an ideal solution, it may be quite serviceable.  And the
additional complexity of really handling native threads may not be an
"ideal" solution either.

...
> 
> As for the VM code itself, it has one of the weirdest source layouts
> I've ever seen. I tried to even look at it with Kdevelop and couldn't
> make much progress. The source itself is undocumented (or only
> documented in books sufficiently old as to have major inacuracies) --
> Really, the documentation should be with (but not in!) the source itself!

Are you aware that the source to squeak is mostly in the squeak image
itself (or perhaps some add-on packages these days?)?  Your reference
to Kdevelop suggests you are looking at the output C files; they were
never intended for human consumption.  While I haven't looked at the
code for the VM in great depth, I did see some fairly extensive
comments in parts of it (e.g., the class comment in ObjectMemory).

Ross Boylan