Re: periods and deadlines in SCHED_DEADLINE

[Gregory Haskins]

Hello Peter, Steven, Thomas, et al. !

Its been a while.

>>> On 8/3/2010 at 04:16 AM, in message <1280823360.1923.419.camel@laptop>, Peter
Zijlstra <peterz@xxxxxxxxxxxxx> wrote: 

[snip]

>
> So we replace the global lock with cmpxchg like loops using 2 per-cpu
> locks. Our current SCHED_FIFO balancer does just this and is found to be
> a very good approximation of global-fifo (iirc there's one funny case,
> but I can't remember, maybe Steve or Gregory can remember the details).

The only thing I remember at the moment is that Steven and I found one last remaining bug in an extreme corner
case where priority order may be violated.  Sadly, I can no longer recall the exact specifics and would have to dig
into the code to remember.  The good news is I believe the issue is fixable.  Its just a matter of impl+test, which
neither of us ever seem to have found the time to properly dedicate.  Perhaps this will be the catalyst ;)

-Greg


--
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/