Re: scheduler interactivity: timeslice calculation seem wrong

[Eric St-Laurent]

> You mean this the other way round, no? +nice means more nice.

sure you're right. and i know that timeslices get asssigned based on
static priority (which is nice value rescaled).

> For the most part, most tasks start at nice 0 so they pretty much all get the 
> same size timslices unless they get preempted.  The rest of the discussion 

i've read that tasks should start at higher dynamic priority with a
small timeslice (a priority boost for a starting task) then immediatly
drop to a lower priority if it use all it's timeslice.

> implemented theory. Changing it up and down by dynamic priority one way and 
> then the other wasn't helpful when I've tried it previously.

maybe it's because the timeslice calculation is reversed?


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/