Subject: Re: uVAXII with Emulex 2nd
To: Holm Tiffe <holm@magnet.geophysik.tu-freiberg.de>
From: Anders Magnusson <ragge@ludd.luth.se>
List: port-vax
Date: 09/28/1995 15:00:57
>
> unpopulated 00760000 - 00760076
> populated 00760100 - 00760106
> unpopulated 00760110 - 00766136
> populated 00766140 - 00766154
> unpopulated 00766156 - 00772146
> populated 00772150 - 00772152 /* emulex ? */
> unpopulated 00772154 - 00774436
> populated 00774440 - 00774456 /* delqa ? */
> unpopulated 00774460 - 00774476
> populated 00774500 - 00774502 /* tmscp0 ? */
> unpopulated 00774504 - 00777476
> populated 00777500 - 00777500
> unpopulated 00777502 - 00777776
>
> The NetBSD kernel prints out:
>
> uda0 at uba0 csr 172150 vec 177, ipl 17
> qe0 at uba0 csr 174400 vec 175, ipl 17
> tmscp0 at uba0 csr 174500 vec 176, ipl 17
>
> what to say about " csr " ? is it an IO-Adress ?
>
Yes, it is.
> why the difference between the table and the kernel ?
>
:-) This is of historical reasons. Actually the I/O space is only
a 8K window, and this uses to be referenced as 760000 - 777777 octal.
(It is the highest 8K in the 18 bit Unibus). Stripping the highest
2 bits and writing 172150 does mean the same as 772150.
-- Ragge