9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
From: Rob Pike <robpike@gmail.com>
To: 9fans <9fans@9fans.net>
Subject: Re: [9fans] Plan9 multi-core support
Date: Tue, 29 Aug 2023 17:36:11 +1000	[thread overview]
Message-ID: <CAKzdPgzgdPsc4GhwV392YTrQ_HmuKnNvdQN7zGqD-Uey3g-7wQ@mail.gmail.com> (raw)
In-Reply-To: <432BC782-2860-4931-B865-E3FC94ED8BEC@quintile.net>

[-- Attachment #1: Type: text/plain, Size: 2492 bytes --]

The dual VAX was the first machine we tried to make work, but for various
reasons including the machine's peculiarities and our own embryonic
knowledge, we abandoned it. The first working Plan 9 kernel was for a 4-CPU
(one MIPS chip per board) IRIS machine, with custom locking hardware (on
another board) because the MIPS 2000 had no synchronization instructions.
That was started in early 1989.

If there is any evidence of the VAX attempt around, I would disavow it.

-rob


On Tue, Aug 29, 2023 at 4:22 PM Steve Simon <steve@quintile.net> wrote:

>
> there was a vax compiler and i think a vax kenfs implementation, i don’t
> know if there was a vax cpu/auth kernel. quite possibly not.
>
> currently i can only find my own post on tuhs confirming the vax was a
> dead end. but i am sure jmk told me he found a vax compiler binary in the
> labs dump.
>
> i think vaxes where becoming rather passé by the time plan9 was born.
>
> -Steve
>
>
> > On 28 Aug 2023, at 7:21 pm, Kurt H Maier via 9fans <9fans@9fans.net>
> wrote:
> >
> > On Mon, Aug 28, 2023 at 12:32:55PM +0000, G B via 9fans wrote:
> >> Windows and Linux began on single-core single processor machines.
> Multiprocessor had been around for some time--IBM's System 360 began using
> multi-processors in 1968--but not for x86. Plan 9 first edition came out in
> 1992, at a time when multicore didn't exist, and multicore was released
> with IBM's Power 4 in 2001.
> >> I can see why someone would ask if Plan 9 supports multicore. Plan 9
> 3rd edition was released in 2000 and 4th edition was released in 2002. In
> each case, going from single core-single processor to multiprocessor and
> then from multiprocessor to multicore would require changes in the
> operating system to recognize the extra processors and then the cores.
> >
> > Symmetric multiprocessing was available in 1992, even on x86
> > machines.  Multics, tops-10, and various unixes all supported it by then.
> > Once you have shared-memory SMP there's little difference between
> > multiprocessor and multicore.  Plan 9's implementation is imo cleaner
> > than most of what came before, but by 1992 there was a lot of
> > multiprocessing going on in the world.
> >
> > khm

------------------------------------------
9fans: 9fans
Permalink: https://9fans.topicbox.com/groups/9fans/T912e4838cb1a371f-M14e2d9d6e4f5fb1a58a1e4f0
Delivery options: https://9fans.topicbox.com/groups/9fans/subscription

[-- Attachment #2: Type: text/html, Size: 4331 bytes --]

  reply	other threads:[~2023-08-29  7:36 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-26 19:33 dusan3sic
2023-08-26 22:40 ` ori
2023-08-26 23:52   ` Rob Pike
2023-08-27  1:27     ` Don Bailey
2023-08-27  9:49       ` dusan3sic
2023-08-27 10:56         ` dusan3sic
2023-08-27 11:32           ` Michael Grunditz
2023-08-27 21:49         ` ori
2023-08-28 12:32           ` G B via 9fans
2023-08-28 15:25             ` Steve Simon
2023-08-28 17:10               ` mkf
2023-08-28 18:20             ` Kurt H Maier via 9fans
2023-08-29  6:21               ` Steve Simon
2023-08-29  7:36                 ` Rob Pike [this message]
2023-08-27  9:50       ` dusan3sic
2023-08-27  8:54         ` mkf
2023-08-27 12:23       ` Dan Cross
2023-08-27  0:13 ` Bakul Shah

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAKzdPgzgdPsc4GhwV392YTrQ_HmuKnNvdQN7zGqD-Uey3g-7wQ@mail.gmail.com \
    --to=robpike@gmail.com \
    --cc=9fans@9fans.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).