Re: DEC Keyboard Question

From: FredK (fred.nospam_at_nospam.dec.com)
Date: 01/26/05 

Date: Wed, 26 Jan 2005 14:34:01 GMT

The "default" is a PC-style KB. To get the driver to handle the KB as a
LK411
there is an obscure handshake where the driver sends certain commands that
will return "error" characters on a standard KB, and return something else
on
an LK411 (the LK450 has a subtle variation on this as well).

Plugging in a KB (and it's powerup) causes the KB to send 0xAA to the
system,
which causes the driver to start this handshake. The generic problem with
pass-thru switches is that the power up only gets sent to one system (or
none,
if the KVM is supplying power all the time). VMS will poll the KB when the
driver
is reset to force a powerup sequence as well. But if no KB is there, the
sequence
may never complete.

There is a solution, I suppose, which is to build a driver that can decode
mode 2
keycodes, and that always expects a LK411. But this driver would have to be
forced to be loaded - since there would be no way to auto-detect it. When
the
drivers were written, the expectation was that a KB would always be
attached.

"Christoph Gartmann" <gartmann@non.immunbio.mpg.de.sens> wrote in message
news:ct7u9f$62t$1@news.BelWue.DE...
> In article <5nxJd.6532$mv4.3353@news.cpqcorp.net>, "FredK"
<fred.nospam@nospam.dec.com> writes:
> >
> >There are two types of KVM switches - dumb pass-thru and "smart"
switches.
> >The dumb ones simply pass the data through with little intellegence. The
> >only thing that they need to do at least for VMS is to cause a power up
> >sequence to be generated when the KB is connected. This way we will
> >reprogram the KB automatically. The smart ones actually have a uProc
that
> >handles the KB and then emulates a KB on the other side. These keyboards
> >need to either know about the extended codes - or at least be smart
enough
> >to pass them through. In the worst case, these KVMs handle USB as well,
and
> >then they need to know how to turn the extendd PS2 codes into the right
USB
> >codes.
>
> Interesting. So I would like to get an explanation for the phenomenon I
> experience with our Rose switches:
> 1) Extended keyboards like my LK411-AG connected to the switch work as
expected
> in "mode 3" with any Alpha so far. The same is true for a 3-button
mouse.
> 2) But after a reboot of the Alpha the keyboard is recognized by OpenVMS
as
> one of these PC keyboards (only 15 function keys. As soon as I plug a
real
> keyboard into the Alpha, type a compose character and reconnect the KVM
> switch all is working fine again.
>
> Regards,
> Christoph Gartmann
>
> --
> Max-Planck-Institut fuer Phone : +49-761-5108-464 Fax: -452
> Immunbiologie
> Postfach 1169 Internet: gartmann@immunbio dot mpg dot de
> D-79011 Freiburg, Germany
> http://www.immunbio.mpg.de/home/menue.html

Quantcast