[openib-general] Re: ipoib_mcast_send.patch

[Michael S. Tsirkin]

Quoting r. Roland Dreier <rdreier at cisco.com>:
> Subject: Re: ipoib_mcast_send.patch
> 
> I think we might want to be even more paranoid and wait until the
> broadcast join succeeds before allowing send-only joins.  Otherwise
> we could create a send-only MCG with the wrong Q_Key, SL, etc.
> 
> something like this maybe?

Right, but I thought atomic test_and_set_bit implied smp_wmb already?

-- 
Michael S. Tsirkin
Staff Engineer, Mellanox Technologies