Dag Brattli wrote:
>
> Yes, you should probably have limit on the Rx queue and tell IrLAP to be
> local busy when the queue builds up ;-) That needs to be done for your 2.2
> impl!
>
> This explains why the flow-control scheme you have made is not
> working. Well, it's working nice for outgoing traffic (skb destructor), but
> this is not working for incomming traffic. We must set the local busy flag
> if queues build up, and that means we _must_ impl. IrLMP exclusive mode :-(
> All other TTP connections would be pretty unhappy when setting local busy!
>
Yup.
And to think I was waiting for someone else to do it..
Now my secret's out.
--
-------------------+--------------------------------------------------------
Thomas Davis | Supernova's are industrial accidents.
[EMAIL PROTECTED] | One of the Linux/IrDA guys.
http://www.jps.net/tadavis
