On Mon, Dec 11, 2000 at 04:30:59PM -0700, Nancy Davis wrote:
> Hi all,
> 
> I am in need of a driver for AHA-1542CP.  I receive the
> following fault on boot from boot[1,2].fs downloaded 
> last week from the i386 port directory.
> 
> probe(aha0:0:0): timed out
> probe(aha0:0:0): timed out AGAIN

The provided driver is fine. If you get a message about a device in the
kernel boot messages and it doesn't say "not configured" anywhere on the
line then you probably have the driver in the kernel already. From the
above messages we can see that the driver for the aha is already in the
boot floppy.

I have found that board to be a real pain in the neck. It only took me
a couple of weeks to finally beat the daylights out of my P-133 until it 
started working (yadda yadda yadda, IRQ conflicts, DMA conflicts, jumperless
configuration of cards with no PnP and no MSDOS driver config disks, etc
etc.)

My suggestions:
*) only use excellent SCSI cables
*) Terminate the card with an external terminator, and use a terminator
   in the last drive used internally. 

   Then play with the termination settings on the card. Don't be afraid to
   enable the termination on the card despite already having two terminators
   installed already.
*) Disable sync transfers on the card in all cases. I've had problems where
   the card would leave drives in a weird state causing them to hang the
   machine on boot. 
*) only use excellent SCSI cables. Did I mention only using top-of-the-line
   cables? Make sure there are no breaks in the insulation ANYWHERE on the
   cables you use. Make sure there are no crimps in ANY of the wires. 
*) Curse and swear. A lot. You'll feel better. 

Yes, I'm fully aware of the rules for SCSI termination. Frankly, given the
choice between a non-working machine that is "correct" and a working 
machine that is "incorrect" I'll take the working machine any day. 
-- 
Kevin P. Neal                                http://www.pobox.com/~kpn/

"Nonbelievers found it difficult to defend their position in \ 
    the presense of a working computer." -- a DEC Jensen paper