RE: Thread starvation with mutex

> From: Luke Elliott
> 
> Forgot to mention that this is on a UP machine with HT disabled.
> 
> Can we call it "sensible" instead of fair..?! I think I'll go and read
> up on some standards. Is there a reason why NPTL does not use this
> "fair" method? Surely this is pretty normal, expected behaviour of a mutex?

Would you please give it a shot under RTNPTL 
(http://developer.osdl.org/dev/robustmutexes) and tell me what happens?

Still you program will fail unless you are in a deterministic scheduling
policy. In SCHED_OTHER (or SCHED_NORMAL), the priority is moved up and
down in a certain range of your nice level to accommodate for your CPU
consumption and improve system responsiveness.

Iñaky Pérez-González -- Not speaking for Intel -- all opinions are my own (and my fault)