9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
* Another system running
@ 1996-03-05 18:01 Kevin
  0 siblings, 0 replies; 9+ messages in thread
From: Kevin @ 1996-03-05 18:01 UTC (permalink / raw)


FS using IDE?  do tell

ksh






^ permalink raw reply	[flat|nested] 9+ messages in thread

* Another system running
@ 1996-03-06  1:39 jim
  0 siblings, 0 replies; 9+ messages in thread
From: jim @ 1996-03-06  1:39 UTC (permalink / raw)


	>>Times change.

	not that much; i'd still go scsi.
absolutely. i'm not suggesting IDE as a reasonable alternative, just
that had large drives been available when the port was done i might
have made (a little) effort to make them useable.
even today, the best you could do is 8 drives at < 2Mb each, not to
mention the configuration nightmare of 4 controllers and the performance
problems if they're not busmastering. admittedly the plan9 ide driver
makes no effort to optimise performance, again showing that it may look
like a pc but it's really just a terminal.






^ permalink raw reply	[flat|nested] 9+ messages in thread

* Another system running
@ 1996-03-06  1:00 forsyth
  0 siblings, 0 replies; 9+ messages in thread
From: forsyth @ 1996-03-06  1:00 UTC (permalink / raw)


>>Times change.

not that much; i'd still go scsi.






^ permalink raw reply	[flat|nested] 9+ messages in thread

* Another system running
@ 1996-03-05 18:39 jim
  0 siblings, 0 replies; 9+ messages in thread
From: jim @ 1996-03-05 18:39 UTC (permalink / raw)


Just for the record, the reason there is no support for IDE
drives as the main filesystem is that, at the time the port
was done, there were no large IDE drives or multiple IDE
controllers available and I didn't think it was worth it.

Times change.

------ forwarded message follows ------

>From cse.psu.edu!9fans-outgoing-owner Tue Mar  5 13:26:48 EST 1996
Received: from colossus.cse.psu.edu by plan9; Tue Mar  5 13:26:48 EST 1996
Received: by colossus.cse.psu.edu id <78577>; Tue, 5 Mar 1996 13:02:44 -0500
Received: from envy.ugcs.caltech.edu ([131.215.128.135]) by colossus.cse.psu.edu with SMTP id <78578>; Tue, 5 Mar 1996 13:02:23 -0500
Received: from avarice.ugcs.caltech.edu by envy.ugcs.caltech.edu with ESMTP 
	(8.6.12/UGCS:4.43) id KAA09694; Tue, 5 Mar 1996 10:01:56 -0800
Received: by avarice.ugcs.caltech.edu
	(8.6.12/UGCS:4.43) id KAA12332; Tue, 5 Mar 1996 10:01:27 -0800
Date:	Tue, 5 Mar 1996 13:01:27 -0500
Message-Id: <199603051801.KAA12332@avarice.ugcs.caltech.edu>
From:	"Kevin S. Ho" <ugcs.caltech.edu!ksh>
To:	cse.psu.edu!9fans
In-reply-to: forsyth@plan9.CS.york.ac.UK's message of Mon, 4 Mar 1996 15:57:55
	GMT
Subject: Re: Another system running
Sender: cse.psu.edu!owner-9fans
Precedence: bulk
Reply-To: cse.psu.edu!9fans

FS using IDE?  do tell

ksh







^ permalink raw reply	[flat|nested] 9+ messages in thread

* Another system running
@ 1996-03-04 15:48 forsyth
  0 siblings, 0 replies; 9+ messages in thread
From: forsyth @ 1996-03-04 15:48 UTC (permalink / raw)


the 386 has got the 32 bit instructions (unlike the 286).
the 386sx uses a 16-bit bus, but that doesn't
change the set of 32 bit operations available (it just makes the system
slower).

i used a 4mbyte 386sx16 as a Plan 9 terminal for several years
on the old release.  in that configuration you want a bigger machine
somewhere to act as a CPU server for compilations (and floating-point
work), and to run
acme, or the heavy-duty part of sam.

i've even used an 8mbyte 386sx16 as a file server, to test a version
of the file server kernel that used IDE instead of SCSI.
a 386sx16 is a bit slow for production use, but it works.

it might still be attractive to use small, cheap 386sx configurations
as highly specialised cpu servers controlling small devices on a network.






^ permalink raw reply	[flat|nested] 9+ messages in thread

* Another system running
@ 1996-03-04 15:20 Steve
  0 siblings, 0 replies; 9+ messages in thread
From: Steve @ 1996-03-04 15:20 UTC (permalink / raw)



> FYI, the file server has a Micronics M54Hi PCI/ISA motherboard with a
> pentium100, a 1.2 Gb ide drive (thanks to forsyth) and a 3c509; the
> authentification server is a 386sx16. The terminals are sun3s (more
> thanks to forsyth), sun4s and various brands of PC (most notably:
> 3c590 pci cards work fine).

I thot that plan9 only ran on 32bit intels..  The mention of a 
386sx16, while interesting (considering the low cost of such
hardware), is confusing to me.  

Can some or all of plan9 really run on a 386SX?

--steve






^ permalink raw reply	[flat|nested] 9+ messages in thread

* Another system running
@ 1996-02-27 17:57 Tim
  0 siblings, 0 replies; 9+ messages in thread
From: Tim @ 1996-02-27 17:57 UTC (permalink / raw)





> From: forsyth@plan9.cs.york.ac.uk
> Date: 	Tue, 27 Feb 1996 06:58:30 -0500
> Subject: Re: Another system running

> >>authentification server is a 386sx16. The terminals are sun3s (more
> >>thanks to forsyth) ...

> i should point out that dhog did the work for that sun3 version.
Is this version public? What would I have to do to get it ? (I've got an old
3/50 that would make a nice terminal)

Thanks,
Tim.








^ permalink raw reply	[flat|nested] 9+ messages in thread

* Another system running
@ 1996-02-27 11:58 forsyth
  0 siblings, 0 replies; 9+ messages in thread
From: forsyth @ 1996-02-27 11:58 UTC (permalink / raw)


>>authentification server is a 386sx16. The terminals are sun3s (more
>>thanks to forsyth) ...

i should point out that dhog did the work for that sun3 version.






^ permalink raw reply	[flat|nested] 9+ messages in thread

* Another system running
@ 1996-02-27  9:55 Jean
  0 siblings, 0 replies; 9+ messages in thread
From: Jean @ 1996-02-27  9:55 UTC (permalink / raw)



There is here another plan9 network up and going.

FYI, the file server has a Micronics M54Hi PCI/ISA motherboard with a
pentium100, a 1.2 Gb ide drive (thanks to forsyth) and a 3c509; the
authentification server is a 386sx16. The terminals are sun3s (more
thanks to forsyth), sun4s and various brands of PC (most notably:
3c590 pci cards work fine).

It was nice to compile 8½ in 10 seconds. The constrat with the time
to compile X11 reminded me unix v6 vs. rsx11-m kernel compilations.

I feel pretty silly with a 386sx16 as cpu server, but I preferred use
the other pentium as terminal and compute on it.






^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~1996-03-06  1:39 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1996-03-05 18:01 Another system running Kevin
  -- strict thread matches above, loose matches on Subject: below --
1996-03-06  1:39 jim
1996-03-06  1:00 forsyth
1996-03-05 18:39 jim
1996-03-04 15:48 forsyth
1996-03-04 15:20 Steve
1996-02-27 17:57 Tim
1996-02-27 11:58 forsyth
1996-02-27  9:55 Jean

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).