[TUHS] Calgary buffer modifications

[TUHS] Calgary buffer modifications

[Paul Ruizendaal]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 902 bytes --]


For a while I had been wondering about the origins of a modification to the V6 kernel that places the kernel buffers in a separately mapped area, freeing up space for other things. It is used in some of the early networking code, i.e. both the UoI code and the V6 BBN code (as available on the TUHS Unix Tree).

I came across the following reference in https://www.osti.gov/scitech/servlets/purl/12130675:

“One widely distributed (though undocumented) solution to this hardware limit on the model 40 was a version of Unix by Robert Sidebotham, Faculty of Environmental Design, University of Calgary. His solution was to move the I/O buffers out of kernel space.”

This would explain the modifications being bracketed with "#ifdef UCBUFMOD”.

Some questions for the old hands:

- Was this patch indeed widely known / distributed?

- Widely distributed is not the same as widely used: was it?

[TUHS] Calgary buffer modifications

[Clem Cole]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2576 bytes --]

On Tue, Jan 30, 2018 at 5:28 AM, Paul Ruizendaal <pnr at planet.nl> wrote:

>
> For a while I had been wondering about the origins of a modification to
> the V6 kernel that places the kernel buffers in a separately mapped area,
> freeing up space for other things. It is used in some of the early
> networking code, i.e. both the UoI code and the V6 BBN code (as available
> on the TUHS Unix Tree).
>
> I came across the following reference in https://www.osti.gov/scitech/
> servlets/purl/12130675:
>
> “One widely distributed (though undocumented) solution to this hardware
> limit on the model 40 was a version of Unix by Robert Sidebotham, Faculty
> of Environmental Design, University of Calgary. His solution was to move
> the I/O buffers out of kernel space.”
>
> This would explain the modifications being bracketed with "#ifdef
> UCBUFMOD”.
>
> Some questions for the old hands:
>
> - Was this patch indeed widely known / distributed?
>
​Hmm - a little like the halting problem I fear - how can you tell?

I can say we knew about them because Gosling brought them to CMU when he
was a grad student; but by that time, we had started to switched to Unix/TS
and Seventh edition that Ted had brought from BTL.   As I recall, it was
more hacking/surgery then I personally wanted to do in the EE 11/34 based
systems.  I don't think they got added to the SUS or IUS @ CS which were
the other heavy 11/40 machines at the time.




>
> - Widely distributed is not the same as widely used: was it?
>
​Not that I remember, but others of course may remember otherwise.

I think many of us knew that Calgary (and Toronto to an extent) was a hot
bed of Unix hacking because of Gosling, but the places outside the US that
were probably having the most effect on the community at the time were in
OZ and the UK.  ​


​Remember Net.noise had not really started as UUCP was part Seventh Edition
so the news systems are in the future.  Also the ArpaNet was limited in
reach, and not that many UNIX sites were apart (it was mostly PDP-10s).
Thus, unless someone came to a USENIX and talked about something, or a
student somehow switch schools (like to be a grad student) folks did not
know about the changes.​   For instance, I knew a little about what was
going on at MIT and UMICH because of friends that left CMU and went to
there or vise versa; and I would bring CMU things to UCB etc...

Clem


ᐧ
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20180130/3fdab537/attachment.html>

[TUHS] Calgary buffer modifications

[Dave Horsfall]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 759 bytes --]

On Tue, 30 Jan 2018, Paul Ruizendaal wrote:

> “One widely distributed (though undocumented) solution to this hardware 
> limit on the model 40 was a version of Unix by Robert Sidebotham, 
> Faculty of Environmental Design, University of Calgary. His solution was 
> to move the I/O buffers out of kernel space.”

I wonder if that inspired the AUSAM buffer management?  The current buffer 
header was "b" (like "u") which got mapped by KISA5.

For the first time I was able to avoid our 11/40 deadlocking; I can't 
remember all the devices that were on it, but there were a *lot* 
(including the printer driver etc which used the buffer pool, not the 
character queue).

-- 
Dave Horsfall DTM (VK2KFU)  "Those who don't understand security will suffer."