Re: [PATCHv2 13/22] scsi_dh_alua: Use workqueue for RTPG

[Hannes Reinecke <hare@suse.de>]

On 01/13/2016 08:02 AM, Hannes Reinecke wrote:
> On 01/12/2016 06:14 PM, Christoph Hellwig wrote:
>>> -        kref_get(&pg->kref);
>>> +        if (!kref_get_unless_zero(&pg->kref))
>>> +            continue;
>>
>> As pointed out earlier this should be done from the start.
>>
> Yep.
>
>>> +    /* Check for existing port_group references */
>>> +    spin_lock(&h->pg_lock);
>>> +    if (h->pg) {
>>> +        old_pg = pg;
>>> +        /* port_group has changed. Update to new port group */
>>> +        if (h->pg != pg) {
>>> +            old_pg = h->pg;
>>> +            rcu_assign_pointer(h->pg, pg);
>>> +            pg_updated = true;
>>> +        }
>>> +    } else {
>>> +        rcu_assign_pointer(h->pg, pg);
>>> +        pg_updated = true;
>>> +    }
>>> +    alua_rtpg_queue(h->pg, sdev, NULL);
>>> +    spin_unlock(&h->pg_lock);
>>> +
>>> +    if (pg_updated)
>>> +        synchronize_rcu();
>>> +    if (old_pg) {
>>> +        if (old_pg->rtpg_sdev)
>>> +            flush_delayed_work(&old_pg->rtpg_work);
>>> +        kref_put(&old_pg->kref, release_port_group);
>>> +    }
>>
>> The synchronize_rcu() needs to be done in release_port_group, or even
>> better be replaced by doing a kfree_rcu there instead of a kfree.
>>
> Point is that we don't necessarily have an old_pg to call
> release_port_group() on, but we still need to call synchronize_rcu()
> to inform everyong that h->pg now has a new value.
> So while I could do that it would end with a mess of if-clauses here.
>
>> And unless I'm mistaken the flush_delayed_work should probably be
>> done in release_port_group as well.
>>
> _Actually_ we only need to call flush_delayed_work if sdev ==
> rtgp_sdev. Otherwise the workqueue item is running off a different
> device and won't be affected.
>
Hmm. Well, not quite. We run into flush_delayed_work()  only if the 
port group changes or upon bus detach.
For all other callers rtpg_sdev() should already be NULL.
But looking at the call sites we can indeed move the 
flush_delayed_work() into the release function.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		   Teamlead Storage & Networking
hare@suse.de			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html