Re: MP kernels are broken (RE: upgrading to 5.0.1)

[Flavio Donadio <flavio%donadio.com.br@localhost>]

Tim,


Mac OS X and/or the hardware implementation on most Macs (I don't know  
for sure) use to be very picky about the jumper configurations on IDE/ 
ATA devices. I don't know if this stands true for NetBSD, but...

Usually, the best practice is to configure the devices as master or  
slave (as appropriate), never as "cable select".

Also, from what I've heard, when using cable select, the master should  
be connected at the end of the flat cable, with the slave in the  
middle connector. This should work only on 80-conductor cables. When  
using 40-conductor cables, the drives have to be configured as master  
or slave, because the cable select system will not work.

Being paranoid as I am, I always configure them as master or slave AND  
connect them on the right positions in the cable, just to be sure.

Be sure and double-check if everything is configured and connected at  
the right places. If it doesn't work, it's got to be a software bug or  
hardware malfunction.

Maybe I just said a lot of nonsense, but there's a lot of people here  
to correct me if I am wrong.


Cheers,
Flavio

On 19/08/2009, at 00:12, Larson, Timothy E. wrote:

> > > Hmm, works here ( GigE with 2x 500MHz G4 ).
> > > No clue what could be causing it - I have a similar setup as you do,
> > 2
> > > disks at the ATA66 channel, a burner and a ZIP at the 2nd channel.
> > > What you could do is to disconnect the burner and the ZIP to see  
> > > from
> > > which channel the panic comes. I also remember weird problems
> > > depending on how the devices are jumpered - do you have then set to
> > > cable select or master/slave? It might be worth trying the other way
> > > around.
> >
> > I've disconnected the optical and zip drives, but still get similar
> > error.  The wdc_iobio_dma (etc) stuff makes me think it's the HD
> > driver, since it's a Western Digital drive.  But what do I know?
> >
> > The HD was jumpered cable select.  I'm working my way through other
> > possible options now, but I've already tried most of them.....
>
> I've now tried every jumper position the vendor says is valid for  
> the drive.  Except for slave (which won't boot), they all result in  
> similar errors with the MP kernel.
>
> Tim