16 Apr
2025
16 Apr
'25
12:27 p.m.
The "RK11-D and RK11-E moving head disk drive controller manual"
(EK-RK11D-MM-002) states in par. 2.5 (Installation Procedure) that the register and vector
addresses of a RK11-D can be configured by "seven interrupt vector address jumpers
and eight address selection jumpers of the M7257". The manual however doesn't
state the address range that can be configured by these jumpers and refers to engineering
drawing D-CS-M7257-0-1 for that information. I found these drawings in the RK11-D
Engineering Drawings but cannot find information in them on the valid address ranges.
Therefore my question: what is the address range a RK11 controller can be configured for?
16 Apr
16 Apr
11:50 p.m.
Hello!
Den ons 16 apr. 2025 kl 21:36 skrev Jos Fries via cctalk <
cctalk(a)classiccmp.org>gt;:
The "RK11-D and RK11-E moving head disk drive
controller manual"
(EK-RK11D-MM-002) states in par. 2.5 (Installation Procedure) that the
register and vector addresses of a RK11-D can be configured by "seven
interrupt vector address jumpers and eight address selection jumpers of the
M7257". The manual however doesn't state the address range that can be
configured by these jumpers and refers to engineering drawing
D-CS-M7257-0-1 for that information. I found these drawings in the RK11-D
Engineering Drawings but cannot find information in them on the valid
address ranges. Therefore my question: what is the address range a RK11
controller can be configured for?
Most (All?) unibus devices can be set anywhere in the PDP-11 IO space. This
means the last 8kbyte of the 256k Unibus space. Addresses 760000 - 777777
octal. A quick glance at the schematic shows that this is true for the
RK11-D as well. What is interesting though is that the doc mentions 8
address jumpers while according to the schematic it has 10. Jumpers W4-W13.
/Mattis
17 Apr
17 Apr
5:36 a.m.
Mattis Lind wrote:
Most (All?) unibus devices can be set anywhere in the
PDP-11 IO space. This
means the last 8kbyte of the 256k Unibus space. Addresses 760000 - 777777
octal. A quick glance at the schematic shows that this is true for the
RK11-D as well. What is interesting though is that the doc mentions 8
address jumpers while according to the schematic it has 10. Jumpers W4-W13.
/Mattis
Aha, thanks, that explains the situation! I found the jumpers in the drawings and found
the address bits that can be configured.
Jos
6:12 a.m.
On Apr 17, 2025, at 7:36 AM, Jos Fries via cctalk
<cctalk(a)classiccmp.org> wrote:
Mattis Lind wrote:
Check the PDP-11 Architecture Handbook (1983-84) Appendix A — Assignment of Addresses and
Vectors, for DEC’s preferred addresses for devices. The default address for an RK11 is
777400 with an interrupt vector of 220. You can put it anywhere you want that doesn’t
conflict with another device, but the benefit of using the DEC default is that things like
bootstrap code and pre-built s/w images will find and use it at the default address.
—Milo
Most (All?) unibus devices can be set anywhere in
the PDP-11 IO space. This
means the last 8kbyte of the 256k Unibus space. Addresses 760000 - 777777
octal. A quick glance at the schematic shows that this is true for the
RK11-D as well. What is interesting though is that the doc mentions 8
address jumpers while according to the schematic it has 10. Jumpers W4-W13.
/Mattis
Aha, thanks, that explains the situation! I found the jumpers in the drawings and found
the address bits that can be configured.
Jos
10:21 a.m.
Milo Velimirović wrote:
Check the PDP-11 Architecture Handbook (1983-84)
Appendix A — Assignment of Addresses and
Vectors, for DEC’s preferred addresses for devices. The default address for an RK11 is
777400 with an interrupt vector of 220. You can put it anywhere you want that doesn’t
conflict with another device, but the benefit of using the DEC default is that things
like
bootstrap code and pre-built s/w images will find and use it at the default address.
—Milo
Yes, i do realize that. The reason I was asking this is because I want to code the RK11-D
device in my PDP-11 simulator (https://github.com/friesga/sim-11) as faithfully as
possible, including the device configuration.
Jos
7:48 p.m.
Jos, I am curious why you are developing a new PDP-11 simulator? There are
a few other existing simulators or emulators.
Thanks
Tom
On Fri, 18 Apr 2025, 1:31 am Jos Fries via cctalk, <cctalk(a)classiccmp.org>
wrote:
Milo Velimirović wrote:
Check the PDP-11 Architecture Handbook (1983-84)
Appendix A — Assignment
of Addresses and
Vectors, for DEC’s preferred addresses for
devices. The default address
for an RK11 is
777400 with an interrupt vector of 220. You can
put it anywhere you want
that doesn’t
conflict with another device, but the benefit of
using the DEC default
is that things like
bootstrap code and pre-built s/w images will find
and use it at the
default address.
—Milo
Yes, i do realize that. The reason I was asking this is because I want to
code the RK11-D device in my PDP-11 simulator (
https://github.com/friesga/sim-11) as faithfully as possible, including
the device configuration.
Jos
18 Apr
18 Apr
1:05 a.m.
Tom Hunter wrote:
Jos, I am curious why you are developing a new PDP-11
simulator? There are
a few other existing simulators or emulators.
Thanks
Tom
Tom, glad you asked :-). There are several reasons for wanting to build another simulator.
The first reason is that none of the other simulators provides what I'm looking for.
With simh e.g. I'm missing the feeling of really being at the controls of the system,
using the original ODT commands and seeing the lights of a disk drive flashing. Then, my
simulator should be able to use the original boot rom code and run the original XXDP
diagnostics, all as realistically as achievable. Ultimately one should have the feeling of
owning his/hers own PDP-11, albeit in the form of a simulator.
The second reason is that - as a once professional software engineer - I want the
simulator to have a maintainable architecture and maintainable code, written in a modern
programming language. And lastly, it's just awesome to delve in the PDP-11
architecture and very satisfying to write software.
Jos
199
days inactive
201
days old
6 comments
4 participants
participants (4)
-
Jos Fries -
Mattis Lind -
Milo Velimirović -
Tom Hunter