Yes, 4MB is plenty of memory. But normal memory is often not the big issue
in RSX,
Yep, that's what killed the PDP-11. Huge physical address space; tiny
virtual address space. We used to have to go thru all kinds of contortions
(overlays and crap) to fit programs on the -11, even when there were loads
of physical RAM available.
Thanks for the tip on pool, though.
Just out of curiosity, does RSX use supervisor mode on the processors that
have it (typically that goes together with I&D space). 2bsd uses super mode
just to get extra address space for the networking code.
Also, don't forget to have the options included to allow for networking in
the kernel. :-)
Umm, Sure.... I trust that those will be obvious when I see them :-)
your requirements on "user friendly" are pretty high.
Well, you can boot the same VMS distro on any VAX ever made and it pretty
much just figures it out :-) Just kidding, though - I'm not looking to
start an argument about which OS is better.
DECnet on 11M also means that you need to understand partitions,
Partitions? You mean memory partitions? On a processor with an MMU?? I
thought that was pretty much all dynamic on M and M+, and only unmapped
systems (like 11S) had to worry about that. Ok, you recently pointed out
that M can run on unmapped systems too, but the 11/23+ has a perfectly nice
22bit MMU and that's not an issue here.
You know that an RQDX and RD32 are bog slow?
Yes, but there are not a lot of other options on the 11/23+. There's an
RL02 drive, but it's not clear that's actually faster; it's also lots
smaller (in Mb, that is, not in cubic feet!) and probably not as reliable
(although that last one is arguable). Besides, I'd rather keep the RL02 as
removable media anyway.
Bob
On 2012-07-08 15:39, Bob Armstrong wrote:
Thanks, guys - I didn't know that M+ would run on a 11/23. I have kits for
M+ v4.3 and DECnet11-M+ v4.3 so that's what I'll try.
As I remember the 11/23+ is pretty slow (like half or 1/3rd the speed of a
11/73) and an RD32 is not a huge amount of space (40Mb, although that's
probably plenty for RSX). Any hints for what I should or should not include
in the SYSGEN?
As the 11/23+ don't have split I/D space, you are going to be running low on pool. Try to keep the number of devices, tasks, and other stuff down. Yes, 4MB is plenty of memory. But normal memory is often not the big issue in RSX, it is pool. Pool is the memory that the kernel uses for all kind of internal stuff, and it is restricted to what is addressable within 64K, minus what the kernel use for other stuff. On split I/D-space machines, all the kernel code are in I space, while pool is in D space. Without split I/D-space, both lots of the kernel code, as well as pool, must coexist in the same memory space.
Also, don't forget to have the options included to allow for networking in the kernel. :-)
Johnny wrote:
SYSGEN, NETGEN and everything in between is rather more cryptic in 11M,
just to warn you... :-)
I guess I'm making the right choice :-) FWIW, I've done M+ SYSGENs
before (years and years ago) and I don't remember those being exceptionally
user friendly!
Either you have never done an 11M SYSGEN, or else your requirements on "user friendly" are pretty high. The M+ SYSGEN gives you help and information at all times, gives sane questions that are understandable, and can be answered pretty straight forward.
11M SYSGEN on the other hand does not give much information, asks that you provide whole sequences of magic as responses sometimes, and is pretty much arcane. A manual beside you is recommended. :-)
Never did DECnet on a 11, however. Actually I'd be tempted
to use M because of the smaller footprint, but I don't have kits for that.
I might be able to provide something. The smaller footprint is indeed a good thing here.
DECnet on 11M also means that you need to understand partitions, since you need to set them up specifically, and understand how large they should be, and so on.
M+ is pretty much just using the default memory partition for everything, and allocates stuff dynamically as needed.
I'll probably just build the system on simh and, if it works, I can
transfer the RD32 image over to the real hardware.
You know that an RQDX and RD32 are bog slow? The actual CPU isn't the only reason why those systems are slow... The disk subsystems often is a big chunk of it.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Just to "point on the nitty gritty details"
To my knowledge; M+ will run on 11/23+ but not upon 11/23 ! (22 bit adressing required). Please also give a thought which serial adapter that is required for terminal access. Old , traditional without buffers etc, or modern 8-port?
One nice thing to include during sysgen would might be a TU58! - It connects to just about every system through one "regular" RS-232 port (properly adressed, though)
There are fine PC-programs available emulating the old, true 1/4" tape hardware running over COM-port.
This gives a path of transfering data "easilly" from SIMH to real HW until the network is up and running.
I've heard that the "max block number" is never checed in the TU58 driver, so if the "tape" happens to be longer than an original tape doesn't matter!
Happy experimenting!
/G ran
On 8 jul 2012 15:39 "Bob Armstrong" <bob at jfcl.com> wrote:
Thanks, guys - I didn't know that M+ would run on a 11/23. I have kits for
M+ v4.3 and DECnet11-M+ v4.3 so that's what I'll try.
As I remember the 11/23+ is pretty slow (like half or 1/3rd the speed of a
11/73) and an RD32 is not a huge amount of space (40Mb, although that's
probably plenty for RSX). Any hints for what I should or should not include
in the SYSGEN?
Johnny wrote:
SYSGEN, NETGEN and everything in between is rather more cryptic in 11M,
just to warn you... :-)
I guess I'm making the right choice :-) FWIW, I've done M+ SYSGENs
before (years and years ago) and I don't remember those being exceptionally
user friendly! Never did DECnet on a 11, however. Actually I'd be tempted
to use M because of the smaller footprint, but I don't have kits for that.
I'll probably just build the system on simh and, if it works, I can
transfer the RD32 image over to the real hardware.
Bob
Thanks, guys - I didn't know that M+ would run on a 11/23. I have kits for
M+ v4.3 and DECnet11-M+ v4.3 so that's what I'll try.
As I remember the 11/23+ is pretty slow (like half or 1/3rd the speed of a
11/73) and an RD32 is not a huge amount of space (40Mb, although that's
probably plenty for RSX). Any hints for what I should or should not include
in the SYSGEN?
Johnny wrote:
SYSGEN, NETGEN and everything in between is rather more cryptic in 11M,
just to warn you... :-)
I guess I'm making the right choice :-) FWIW, I've done M+ SYSGENs
before (years and years ago) and I don't remember those being exceptionally
user friendly! Never did DECnet on a 11, however. Actually I'd be tempted
to use M because of the smaller footprint, but I don't have kits for that.
I'll probably just build the system on simh and, if it works, I can
transfer the RD32 image over to the real hardware.
Bob
As for the rest of the question, yes, 11M supports DECnet with DEQNA.
SYSGEN, NETGEN and everything in between is rather more cryptic in 11M, just to warn you... :-)
I've been able to generate successfully a 11M system, but I've not found any DECNET-11 kit for it...
Johnny
On 2012-07-08 05:28, Bob Armstrong wrote:
I have a 11/23+ system with 2xRD32 disks and a DEQNA (as well as
various peripherals RX50, RL02, RX02, Kennedy 9 track). RSX-11M+
needs I&D space so it can t run on a 11/23+, right? Does RSX-11M
support DECnet and a DEQNA? FWIW, the system as 2048K of RAM quite a
bit by 11/23 standards. I could look it up if I had the SPDs for
DECnet-11, but I haven t seen those online and I figure there are lots
of HECnet people who would know.
Actually I have a KDJ11 (11/73) card and I could just swap the CPUs
easily enough, but I kind of like the system the way it is.
If it is possible to run DECnet on the 11/23+, my next trick will be
to locate kits and figure out how to install it.
Hi, Bob.
Actually, RSX-11M-PLUS does not require I/D-space. It does, however, require 22-bit addressing. The 11/23+, as well as the 11/24 are supported by M+.
But some things get rather tight without I/D-space. But it should work.
As for the rest of the question, yes, 11M supports DECnet with DEQNA.
SYSGEN, NETGEN and everything in between is rather more cryptic in 11M, just to warn you... :-)
Johnny
Bob Armstrong wrote:
I have a 11/23+ system with 2xRD32 disks and a DEQNA (as well as various peripherals RX50, RL02, RX02, Kennedy 9 track). RSX-11M+ needs I&D space so it can t run on a 11/23+, right? Does RSX-11M support DECnet and a DEQNA? FWIW, the system as 2048K of RAM quite a
M+ SYSGEN:
>; RSX-11M-PLUS may be run on the following PDP-11 processors:
>;
>; 11/23-PLUS (also called MicroPDP-11/23 and 11/23-B)
>; 11/24
>; 11/44
>; 11/53
>; 11/70
>; 11/73 (also called MicroPDP-11/73)
>; 11/83 (also called MicroPDP-11/83)
>; 11/93 (also called MicroPDP-11/93)
>; 11/84
>; 11/94
>;
>; M70,80,90,100 MENTEC modules (treated as an 11/83)
>; M11 processor (MENTEC)
>;
>; LSI-11/73
RSX-11M supports DECnet and DEQNA.
I have a 11/23+ system with 2xRD32 disks and a DEQNA (as well as various peripherals RX50, RL02, RX02, Kennedy 9 track). RSX-11M+ needs I&D space so it can t run on a 11/23+, right? Does RSX-11M support DECnet and a DEQNA? FWIW, the system as 2048K of RAM quite a bit by 11/23 standards. I could look it up if I had the SPDs for DECnet-11, but I haven t seen those online and I figure there are lots of HECnet people who would know.
Actually I have a KDJ11 (11/73) card and I could just swap the CPUs easily enough, but I kind of like the system the way it is.
If it is possible to run DECnet on the 11/23+, my next trick will be to locate kits and figure out how to install it.
Bob
Sorry for stupid question, but i can't understand why i get the "error activating transport SMTP" message when i try to send a mail to some host in HECnet,
using DECnet MAIL utility and address in HOST::USER format?
You're probably getting that message from the remote system. The USER on that system may have their mail forwarded to SMTP%"x at y" and it's not working properly there.
--Marc