Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 8 Oct 1998 08:05:30 +0200
From:      J Wunsch <j@uriah.heep.sax.de>
To:        FreeBSD SCSI list <freebsd-scsi@FreeBSD.ORG>
Subject:   Re: SCSI Differential controllers
Message-ID:  <19981008080530.60854@uriah.heep.sax.de>

next in thread | raw e-mail | index | archive | help

Simon asked me to post this clarification to the mailinglist as well:

-----Forwarded message from Simon Shapiro <shimon@simon-shapiro.org>-----

Message-ID: <XFMail.981007204530.shimon@simon-shapiro.org>
Date: Wed, 07 Oct 1998 20:45:30 -0400 (EDT)
Reply-To: shimon@simon-shapiro.org
Organization: The Simon Shapiro Foundation
From: Simon Shapiro <shimon@simon-shapiro.org>
To: Joerg Wunsch <joerg_wunsch@uriah.heep.sax.de>
Subject: Re: SCSI Differential controllers


J Wunsch, On 07-Oct-98 you wrote:
>  As Simon Shapiro wrote:
>  
> >   Some new controllers use what is known as
> > ``low voltage differential'' with means the controller (correct name is
> > HBA) will ``talk'' differential to differential devices, single-ended
> > to SE
> > devices.
>  
>  I thought a single SE device will degrade the entire bus to SE?  I
>  can't imagine otherwise, since this SE device grounds one half of all
>  the `return' signal lines on the bus.

I think you are right.  I only deal with these things on the HBA-Bay leg. 
All the bays I use are SE inside and diff outside.  There is nothing on the
Diff side except for boxes and DPTs.  The box has this dual capacity.  The
DPTs are simply differential.  I think the i2o (5th generation) DPTs are
dual mode.

>  So far, i've only browsed a little through the SCSI-3 working group
>  documents, they speak that only devices with active (?) negation are
>  guaranteed to co-operate correctly with LVD, while older SE devices
>  might cause damage to the LVD output stages.  If i understood it
>  correctly, all differential output stages are required to use the
>  DIFFSENSE line on the bus in order to detect whether an SE device has
>  accidentally been connected to the bus (so they can turn off their
>  output drivers), i assume that's also the way LVD decides in which
>  mode to operate.

Again, you are correct.  I suggest you post this to the SCSI list, so
others can benefit.

My rule is:  Check each device, never mix modes, check each device again,
then buy the disk bays from reputable source, with the disks.
Solvs lots of problems :-)

Simon


-----End of forwarded message-----

-- 
cheers, J"org

joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19981008080530.60854>