Subject: Re: Core's, panic's, and hard lockups
To: Leonard Chung <leonard@ssl.berkeley.edu>
From: Henry B. Hotz <hotz@jpl.nasa.gov>
List: port-mac68k
Date: 10/30/2000 11:47:55
At 6:43 PM -0700 10/27/00, Leonard Chung wrote:
>I recently installed NetBSD 1.4.2 on an old Mac IIsi. The IIsi has 
>an FPU and 8 megs of RAM. There are also 7 drives attached to it, 5 
>are CDROM drives

Seven devices!!!!

They had better only be in max two boxes and you had better have an 
active terminator to that will never work reliably.  I know the spec 
allows that, but if you want to push the spec to the limit you really 
need to make sure every bit of the cable chain and the terminators is 
correct.

You should allow 1 foot between devices on internal cables, although 
if you can get devices less than a couple of inches it may be OK. 
All external cables and cable/box transitions are suspect.  As I said 
active terminators can make up for problems elsewhere.  Some people 
have had luck putting separate terminators at the end of internal 
cable runs instead of using the terminators built into most disk 
drives.

In theory the active terminators on disk drives that you jumper 
on/off should work better than the passive ones where you 
remove/install resister SIP's, but my personal, limited experience is 
the opposite.

There are known to be some problems with the SCSI drivers on 
port-mac68k which the developers have been unable to debug.  That's 
why we provide two different SCSI drivers since they usually trip on 
completely different problems.  You have the privelage of having a 
system which has problems with both so if you can get the attention 
of e.g. Allen Briggs, your system may be a good test case for them.


Signature held pending an ISO 9000 compliant
signature design and approval process.
h.b.hotz@jpl.nasa.gov, or hbhotz@oxy.edu