Re: tulip driver in 2.1.11* - 2.1.21 is broken - new driver

kuznet@ms2.inr.ac.ru
Sat, 19 Sep 1998 19:18:14 +0400 (MSK DST)


Hello!

> So I think dev->interrupt should disappear, and dev->tbusy should stay
> as is.

It is not so clear question.

dev->interrupt is critical for fastroute.

In any case driver should have some lock visible from outside.
Now normal mode has no lock (because they are not necessary),
fastroute has two locks: dev->tx_semaphore and dev->interrupt.
It would be good to merge them, but I see no direct way to make it.

If you remember, I proposed to make one sample driver with
really correct interface to upper and lower layers
and to maintain it permanently, keeping it in sync both
with networking core and with irq/smp, and, maybe, ignoring news
on hardware market. This driver would serve as sample for driver writers.

Becker's tulip 0.83 was very good starting point (probably, it is 0.88 now)
It would not handle the newest chips, but the latest Becker's driver also
exists.

Alexey

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/