grant@torque.net
Wed, 2 Sep 1998 06:16:45 -0400 (EDT)
> First, I have to make clear that I was unable to find a parameter
> to explicitly tell pd or ppa to use interrupt, so it may turn out
> that their drivers were actually polling the devices. I suspect this
> because the responsiveness of my Linux box wasn't better than with
> the unmodified kernel. It is noticable when the PHd is in action,
> even changing VTs are very slow when I use it. PHd is a very slow,
> nibble mode device, it would help a lot if it was able to use an
> interrupt. I use "options pd nice=1 drive0=0x378,,,,,,0"
> in /etc/conf.modules to make it a bit faster. In fact, this way
> it is ~1.5x faster than with its DOS driver. :)
Oops. I should have responded to Andrea's original request for you to
do this test. None of the PARIDE drivers use an interrupt. Nor does PPA.
In almost every case this is not a choice, the hardware simply does not
support the possibility. The single exception is the Shuttle EPAT adapter
which has a useful interrupt, but after an enormous number of complaints
I removed the interrupt support for that driver too.
--------------------------------------------------------------------------
Grant R. Guenther grant@torque.net
--------------------------------------------------------------------------
-- To unsubscribe, send mail to: linux-parport-request@torque.net --
-- with the single word "unsubscribe" in the body of the message. --
This archive was generated by hypermail 2.0b3 on Wed 30 Dec 1998 - 10:18:13 EST