Subject: Re: pciide performance on alpha
To: Manuel Bouyer <email@example.com>
From: Andreas Johansson <firstname.lastname@example.org>
Date: 08/15/1999 13:55:14
On Fri, 13 Aug 1999, Manuel Bouyer wrote:
> If adresses above 0xffff are ok on the alpha (Jason, is this true ? You know
> alpha hardware better than me :) then NetBSD certainly does the rigth thing
> (otherwise a lot of other drivers would fail as well).
> My guess is that the promise chip, which was specifically designed for
> PCs, doesn't support it.
Ok, I have mapped the NetBSD driver to 0x8000 and then set the chip up at
0x18000. This doesn't work and therefore I believe the chip does not
wrap. There was no machine check in this configuration like I had when
NetBSD was told that the chip was at 0x10200. To me this looks like the
machine is set up with an I/O area that is only 0x10000 large but
configures devices at addresses above this area.
> Manuel Bouyer, LIP6, Universite Paris VI. Manuel.Bouyer@lip6.fr