Re: [PATCH RFC]: napi_struct V4

[David Miller <davem@davemloft.net>]

From: "Michael Chan" <mchan@broadcom.com>
Date: Thu, 26 Jul 2007 14:38:09 -0700

> If the driver wants a simple solution, it can do what you did in the
> patch:  wrap the tx cleanup code with netif_tx_lock() and
> netif_tx_unlock().
> 
> If a NAPI driver wants to be more clever, it can do something such as
> this in tg3's poll_controller:
> 
> if (netif_rx_schedule_prep(dev, &tp->napi)) {
> 	tg3_tx(tp);
> 	netif_poll_enable(tp->napi);
> }

Thanks Michael, that's a good suggestion and would work.

In thinking about this some more over the weekend I've decided
that my plan to rip out RX support from netpoll is a bit too
ambitious.

Therefore, for the time being, I'm going to add a special driver
function for netpoll that will allow it to ask the driver to invoke
->poll() over all the NAPI structs assosciated with the netdev.