PF> It is my understanding (from talking to someone who's been
 PF> in the code before) that a blocked thread is removed from
 PF> the scheduler's list altogether.  Only "ready to run"
 PF> threads are seen by the scheduler.

 MB> That's clearly impossible.  What would be the mechanism, then, for 
 MB> making the thread wake up?  Something, somewhere, must

A thread blocked in a device driver will only be unblocked by the dd,so it
can be forgotten about.  Perhaps this is the only type of block he meant.

 MB> repeatedly check the semaphore on each scheduler 
 MB> transition in order to determine whether it has come 
 MB> ready.

Why couldn't the Reset/Post sem function cause it to be reinserted into the
ready to run list.

As far as sems expiring,  you're right -- something must be looking after
them -- but this could be boiled down to only the closest sem to
expiration.

(I don't have Kogan's book anymore -- does he talk about this?)



--- Maximus/2 3.00