Re: yield() in i2c non-happy paths hits BUG under -rt patch

[Peter Zijlstra]

On Thu, 2009-11-19 at 12:59 +0000, Alan Cox wrote:
> > Well, I guess only people monitoring system latency would notice, as
> > this is the only thing yield() was supposed to help with in the first
> > place.
> 
> 	if (need_resched())
> 		schedule();

aka.

	cond_resched();

> will make non-rt tasks act politely at the right moments. RT tasks will
> likely immediately get to take the CPU again depending upon the
> scheduling parameters in use.

Right, FIFO will simply NOP it, since if it was the highest running
task, it will still be. RR could possibly run out of its slice and
schedule to another RR of the same prio.


--
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/