Subject: Re: Continuing sun3 nbsd142 funzies.....SUCCESS!!!!!
To: NetBSD Bob <nbsdbob@weedcon1.cropsci.ncsu.edu>
From: Greg A. Woods <woods@weird.com>
List: port-sun3
Date: 05/31/2000 01:48:48
[ On Tuesday, May 30, 2000 at 14:54:05 (-0400), NetBSD Bob wrote: ]
> Subject: Re: Continuing sun3 nbsd142 funzies.....SUCCESS!!!!!
>
> That is not one of the VME crates.  I think it is different.  I once ran
> a 3/80 on NetBSD, but I don't remember how the drives were set on it.
> On the VME scsi things and on my 3/60, it has to have the motor on at
> power on jumpers set, in my hands for NetBSD 1.4.2 to work.  Conversely,
> sunos doesn't need them.  If there is something I am  missing or some other
> explanation for the stalling attempts to read the scsi at bootup, do holler.

There are two different ways I've seen some disks work.  First is the
mode where they'll spin up on the first command of any sort (or on a bus
reset), and second are those drives that require an explicit SCSI "START
UNIT" (1B) command before they'll spin up.

It would appear that NetBSD will send a "START UNIT" command *unless*
the disks have a "quirk" table entry that sets the SDEV_NOSTARTUNIT
flag.  From what I can see though the only listed fixed drives are two
specific revisions of the Hitachi DK515C (CP15 & CP16).

With drives any older than about 4-5 years it's always safest to jumper
them 100% correctly in all cases, and that includes having them spin up
and go ready on power up.  I've seen drives which SunOS-4.1.4 couldn't
command to spin up, so you just never know....

-- 
							Greg A. Woods

+1 416 218-0098      VE3TCP      <gwoods@acm.org>      <robohack!woods>
Planix, Inc. <woods@planix.com>; Secrets of the Weird <woods@weird.com>