On Sun, Dec 20, 2015 at 10:31 AM, Jesse Gross <[email protected]> wrote:
> On Sun, Dec 20, 2015 at 1:21 AM, Pravin B Shelar <[email protected]> wrote:
>> Keep devices in UP state in stt-list. So that devices in UP state
>> would be searched on a packet receive in STT. This way devices in
>> DOWN state would not accept any packets.
>>
>> Signed-off-by: Pravin B Shelar <[email protected]>
>
> I think this works but it seems like it deviates from the other tunnel
> types more than is desirable. The difference between those and STT is
> that STT does not have a hash table used to look up devices in the
> non-OVS case. However, it might be good to keep STT as close to
> upstream best practices as possible even if it isn't immediately
> necessary.
>

fair enough, I will send out a patch to make it more in sync with
other tunnel type.

> One other thing that I noticed is that the STT fragment hash table
> commingles fragments that were addressed to different destination
> ports. This is unlikely to matter with the current implementation but
> I'm not sure that it is right.

STT fragment hash table has ack_seq which should be unique across
different stt-dst-ports. So I am not sure if we need to check dst_port
in the fragment lookup.
_______________________________________________
dev mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/dev

Reply via email to