Subject: RE: Vaxstation Turbochannel
To: None <arcarlini@iee.org>
From: Roger Ivie <rivie@ridgenet.net>
List: port-vax
Date: 11/15/2004 17:04:40
On Mon, 15 Nov 2004, Antonio Carlini wrote:
> Not relevant to the current discussion, but just to put
> the record straight for anyone reading this in the future,
> the DMB32 is the VAXBI version of the DMF32 (it does asynch
> and synch and parallel I/O). It has more addressing modes
> than you can shake a stick at (four) and it can indeed read
> VAX page tables (at least until the XVA/XPA project changed
> the format).
>
> FWIW: the VAXBI analogue to the DR780 would probably be
> a DRB32. I think it existed, but I know nothing about it.
You are correct, sir. I once used the DR780 and like it very much.
Although some cow-orkers used the DRB32, I was not involved in that
particular project.
Going back on topic, I thought of a reason that you might not be able to
do DMA beyond the 128MB address space in which memory is expected to
live. TURBOchannel DMA addresses are shifted lift by five bits.
Straightening it out to give it to the VAX would require rotating the
address if all bits were to be visible to the VAX. Since I drew the
first generation of schematics by hand, I probably just grounded the
bits beyond the 128MB address space because it would mean less pencil
time.
--
Roger Ivie
rivie@ridgenet.net
http://anachronda.webhop.org/
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GCS/P d- s:+++ a+ C++ UB--(++++) !P L- !E W++ N++ o-- K w O- M+ V+++ PS+
PE++ Y+ PGP t+ 5+ X-- R tv++ b++ DI+++ D+ G e++ h--- r+++ z+++
------END GEEK CODE BLOCK------