On Mon, Feb 04, 2002 at 11:19:22PM +0100, Martin Husemann wrote:
> Someone with the proper hardware for testing available needs to write a
> driver (or port the OpenBSD one). Often there are architectural differences
> at this point - OpenBSD puts many such drivers in arch/sparc64/dev - which
> is from the POV of most NetBSD developers just plain wrong and needs fixing
> before importing those drivers. The necessary work sometimes makes it easier
> to start from scratch :-(
> 
Oh please.  autri (driver for the chip on blade100) is in /sys/dev/pci.
It's a generic AC97 interface.  As such, it worked right out of the box
for this machine on OpenBSD.

The drivers in sys/arch/sparc64/dev on OpenBSD are ones that:
	a> were imported from NetBSD but not needed (and/or wanted) on other arch's
	b> ones that are truely sparc64 specific
	c> ones that was cause havok in other ports if put in the "right" place

Location of a file doesn't mean anything when it comes to how difficult
it is to port a driver.

--Jason L. Wright