On Sat, Dec 10, 2011 at 10:58 PM, Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
> On Sat, 2011-12-10 at 22:07 +0800, Ming Lei wrote:
>> > While true, it fails to show why this is a problem. I say it is not a
>>
>> IMO, the added two barriers are pairs of the implicit barrier in kref_put, so
>> that we can order between kref_init/kref_get and kref_put.
>
> Yeah so?

I think so, see below:


CPU0			CPU1

atomic_set(v)
smp_mb()
				smp_mb()
				atomic_dec_and_test(v)

Without the barrier after atomic_set, CPU1 may see a stale
value of v first, then decrease it, so may miss a release operation.

The pair of smp_mb can make order between atomic_set
and atomic_dec_and_test, can't it?

> If there's a destruction race with kref_put() the barrier won't

Sorry, could you say what the destruction race is?

> solve it. Other than that the actual order of get/put is irrelevant.


thanks,
-- 
Ming Lei