[Nickle]Priority inversion and smart semaphores

Barton C Massey bart@cs.pdx.edu
Wed, 11 Apr 2001 21:53:02 -0700


One of the things on my "to-look-up" list is the whole issue
of priority inversion in realtime scheduling.  I suspect
that Nickle semaphores want to always allow the
least-recently suspended *highest-priority* process to wake
up (rather than their current plain LRS behavior), but I
can't say I've followed the issues well enough to be sure.
If someone else wants to check into this, that would be great!

	Bart Massey
	bart@cs.pdx.edu