Re: [RFC PATCH 05/14] sched: account for blocked load waking back up

[Paul Turner]

On Thu, Feb 16, 2012 at 8:02 AM, Peter Zijlstra <a.p.zijlstra@xxxxxxxxx> wrote:
> On Wed, 2012-02-01 at 17:38 -0800, Paul Turner wrote:
>
>>         u64 last_runnable_update, decay_count;
>
>> +       /* we track migrations using entity decay_count == 0 */
>> +       if (unlikely(se->avg.decay_count <= 0)) {
>
> !sleep dequeue isn't migrate only,

Well they mostly are these days; but in the == 0 case we're either a
load-balancer migration or someone doing a dequeue/enqueue pair on an
entity about some sort of update.

The key is that when it is an load-balancer move we'll resync
appropriately on enqueue (which we need to do).  We will essentially
sync with ourselves in the other cases, but they have no bearing on
why we do this.

> also, <= 0 on an unsigned is weird.

Yeah that should be a s64 not u64 (fixed).

>
--
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/