> > As I wrote previous mail, this overhead could be workarounded by
> > thread buffering. So this is not problem.
> 
> Only partially.  It does *not* solve the memory usage issues
> (currently 8.5k or so per blocked `thread', minimum).  Nor does it
> solve more subtle issues (such as bad cache behavior from switching
> kernel stacks all the time) that Mach solves.  Solaris does this
> partially via `turnstiles'.

Yes. I agree that this should be addressed in long run.

But I think this problem can be ignored in short term,
since this is not paticular problem of scheduler activations,
other kernel thread system which supports multiple blocked systemcalls/
paging activities has same problem. (e.g. current LinuxThreads, 
legacy kernel threads of commercial UNIXes, ...)
--
soda