Subject: Re: UltraDMA problems under 1.6F (probably hardware)
To: gabriel rosenkoetter <gr@eclipsed.net>
From: Stephen Degler <stephen@degler.net>
List: current-users
Date: 08/14/2002 08:34:14
Hi,
I'm using VIA 8233A, Highpoint, and Promise controllers with new Western
Digital drives and good cables. I'm getting UDMA100 with no problems on
each configuration. I do see UDMA133 timings in the set up code for
some controllers, but I don't have a UDMA133 drive, so I haven't tested
this.
skd
On Wed, 2002-08-14 at 08:01, gabriel rosenkoetter wrote:
> On Wed, Aug 14, 2002 at 01:47:02PM +0200, Matthias Drochner wrote:
> > gr@eclipsed.net said:
> > > Maybe I'll try trading the UDMA 66 10 GB drive with the UDMA 133
> > > 160(ish) GB drive to confirm Matthias's experience.
> > My drives don't support UDMA133, this happens with UDMA100.
>
> Huh.
>
> That didn't happen on my old (now dead) motherboard, where I had the
> disks plugged in identically to the way they are now. I believe
> that was a Via UDMA 100 controller (I'm sure the {north,south}bridge
> chips were Via, and that it was UDMA 100; I'm not sure that it
> wasn't an off-chip IDE controller, but I doubt that), but I couldn't
> tell you which one without digging through something like, oh, this
> looks good:
>
> http://www.plasma-online.de/english/identify/picture/via.html
>
> > > do we even support UDMA 133 yet?
> > No, afaict. Definitely not for VIA chipsets.
>
> Well. Then maybe we also don't support 48-bit LBA, which could
> plausibly be part of the problem (though it really shouldn't be; the
> OS addressing the controller with only 28-bit commands is no
> justification for it to completely flake out).
>
> It's probably worth my telling the BIOS to limit itself to UDMA 100.
> Provided it's not too brain-damaged to do that. (Cf, my previously
> stated feelings on IA32...)
>
> --
> gabriel rosenkoetter
> gr@eclipsed.net