> > Not the same CI if I remember right. I'm not sure why DEC produced
> > another
> > CI controller for these machines, but they did. I'll have to assume=20
> > they
> > had a good reason. The same controller is used in a DEC=20
> 7000, but you=20
> > have
> > to change the microcode.
>=20
>    Really??  That is *weird*.  I can understand there being=20
> differences=20
> between VAX and Alpha I/O controllers for the same bus, but between=20
> different VAX implementations??  This doesn't sound very=20
> DECish to me. =20
> Sad. :-(

Were there different controllers for the XMI bus on the VAX 6000
and the VAX 7000? News to me (but I've been wrong before). Anyone
have any controller names?

Typically the VAX 6000 used BI-bus controllers for CI, although
it could use the XMI-based ones for disk, tape, network and CI.
AFAIK, the same ones worked everywhere.

That doesn't mean all controllers would work in all systems. As
an obscure example, the DSB32 VAXBI synch controller would not
work in a VAX 7000 system. The microcode locked the VAXBI bus
in a way that would not work when the VAXBI was hung off an
XMI bus.

Similarly, the XMI CI controller had carnal knowledge of the
system page tables, so it needed different firmware depending
on whether it was used on a VAX or an Alpha. But the only
difference was the firmware.

The XMI SCSI controller was Alpha-only: I expect that was mostly
a firmware issue (not worth expending effort on a VAX variant at
that point in the lifecycle), but I do not know for sure.

So, names and numbers please, someone. I'd be interested to know -
one day I'm bound to trip over a VAX 7000 and lug it home: it=20
would be a terrible waste of effort to pick up the wrong=20
controller :-)

Antonio
=20
--=20

---------------
Antonio Carlini             arcarlini@iee.org
=20