darcs patch: Fix semaphore implementation to *not* sh... (and 1 more)

[Tony Garnock-Jones]

Brian Rice wrote:
> I have some local patches that conflict with this, so I haven't
> successfully applied them yet. I'll push them out as soon as I can.

Cool.

> Thanks a lot for focussing on this!

I'm both pleased to have found the fault, and embarrassed there was a
fault there to begin with. Yuck!

Anyway, the road is open now to getting a timeout-heap integrated with
the select() call and then integrating SDL's events.

Any comments on whether CML's events design is a good one to lift for
slate? It seems clean, elegant and powerful, and perhaps also a bit
slow... I do like the composability of events. (Almost as nice as STMs!)

Tony
-- 
 [][][] Tony Garnock-Jones     | Mob: +44 (0)7905 974 211
   [][] LShift Ltd             | Tel: +44 (0)20 7729 7060
 []  [] http://www.lshift.net/ | Email: tonyg at lshift.net