9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
* Re: [9fans] General question about hosted interfaces
@ 2001-07-10 18:22 rob pike
  2001-07-10 19:08 ` Mike Haertel
  2001-07-11  8:34 ` Douglas A. Gwyn
  0 siblings, 2 replies; 30+ messages in thread
From: rob pike @ 2001-07-10 18:22 UTC (permalink / raw)
  To: 9fans

> Isn't that largely what drawterm does?
No, because drawterm needs a CPU server to connect to.
In fact, the new 9P2000 version of drawterm that rsc built
builds from kernel sources directly.

I agree with the sentiment that DAGwyn's proposal isn't
very compelling.  With so many alternatives for hosting
or simulating one operating system over another, I don't
see the need for another, its advantages notwithstanding.
Inferno's emu gives a different language and a variety
of new things, but running a Plan 9 simulator in a Unix
window doesn't add nearly as much.  Moreover, it requires
a huge (by my standards, anyway) file system to support it:
the kernel is a teeny part of the whole, in great distinction
with Inferno.

-rob



^ permalink raw reply	[flat|nested] 30+ messages in thread
* Re: [9fans] General question about hosted interfaces
@ 2001-07-12 21:58 jmk
  0 siblings, 0 replies; 30+ messages in thread
From: jmk @ 2001-07-12 21:58 UTC (permalink / raw)
  To: 9fans

On Thu Jul 12 17:13:24 EDT 2001, chris@cjl1.demon.co.uk wrote:
> > So? What makes you think that someone else can?
>
> As an individual (not representing a corporation)
> I find it extremely difficult to get hold of data sheets.
>
> I have filled in countless web-forms to register
> for the ability to access the 'developer' pages
> on a manufacturers site.
>
> I have yet to gain access to a single one!
>
> This make it hard for the hobbyist to get involved in
> writing drivers.
>

You are missing 2 things:
1) it's not any easier for someone working for a corporation - in many
   ways it's worse as a) they know you have a lawyer, b) they think they
   may be able to make money out of it, both of which colour the discussion,
   and 3) they may consider you a competitor in some sense.
2) the datasheet may not exist. The lifetime of interface cards is so short
   and the margins of the manufacturer so thin they often don't bother with
   a technical writer.

Like Boyd I've used high-quality documentation from Digital. That level can
still be found from some of the established component manufacturers (e.g. Intel,
National Semiconductor) but even that doesn't always help when the components
are combined with others to make a system.


^ permalink raw reply	[flat|nested] 30+ messages in thread
* Re: [9fans] General question about hosted interfaces
@ 2001-07-12 21:48 bwc
  2001-07-13 14:53 ` Douglas A. Gwyn
  0 siblings, 1 reply; 30+ messages in thread
From: bwc @ 2001-07-12 21:48 UTC (permalink / raw)
  To: 9fans

[-- Attachment #1: Type: text/plain, Size: 860 bytes --]

I've found both the 3com site and the Intel developer's site to be useful.
It still takes digging, but I've been lucky.  The smaller companies are
a problem.  Many of their products are `clones' and they don't even bother
to write a manual.

Also, the PCMCIA, PCI, USB, and other consortia provide the baseline
description of much of this stuff.  The vendors often just say what
parts they really do.  ANSI for the ATA spec, for example.  Since
there are many vendors playing in the same space this makes sense.
Back when only DEC defined what a VAX was, each peripheral's documentation
would fully describe that peripheral.  Now they say `vesa video controller'
and just describe what makes them different.

In 1996 I had top secret copies of one of Intel's chipset I was using.
Today you can the equivilant document off the web.

  Brantley

[-- Attachment #2: Type: message/rfc822, Size: 2013 bytes --]

From: "Chris Locke" <chris@cjl1.demon.co.uk>
To: <9fans@cse.psu.edu>
Subject: Re: [9fans] General question about hosted interfaces
Date: Thu, 12 Jul 2001 22:09:10 +0100
Message-ID: <003d01c10b16$e8128b40$2248dec2@falken>

> So? What makes you think that someone else can?

As an individual (not representing a corporation)
I find it extremely difficult to get hold of data sheets.

I have filled in countless web-forms to register
for the ability to access the 'developer' pages
on a manufacturers site.

I have yet to gain access to a single one!

This make it hard for the hobbyist to get involved in
writing drivers.


^ permalink raw reply	[flat|nested] 30+ messages in thread
* Re: [9fans] General question about hosted interfaces
@ 2001-07-12 19:51 geoff
  0 siblings, 0 replies; 30+ messages in thread
From: geoff @ 2001-07-12 19:51 UTC (permalink / raw)
  To: 9fans

jmk has a point.  I'll try to spend some time applying Jedi mind
tricks to hardware vendors.  I'm told that Nvidia will sell anybody
their interface documentation for $1,000, so they may be more
susceptible than most.

Figuring out which cards (or even chip sets) are going to have the
longest shelf lives, so that the effort is worthwhile, would help
to set priorities.


^ permalink raw reply	[flat|nested] 30+ messages in thread
* Re: [9fans] General question about hosted interfaces
@ 2001-07-12 19:45 geoff
  0 siblings, 0 replies; 30+ messages in thread
From: geoff @ 2001-07-12 19:45 UTC (permalink / raw)
  To: 9fans

Doug asks how much of the 26,300-line difference in .c files
between the PC and the Bitsy is due to the Bitsy having had
less time to evolve.

The Bitsy family could well sprout new gadgets over time,
but it's somewhat physically constrained by its size and
I hope that Compaq will have the good sense to keep using
the same components unless it has really good reason
(not just the price on the spot market for components this
morning).  Meanwhile, in the wild chaos of the PC I count
12,181 lines of pc/ether*.c in 14 files for roughly that
many different kinds of Ethernet cards, and 6,583 lines
of pc/vga*.c in 21 files, again for roughly that many kinds
of VGA cards.  So 18,764 lines of the difference (all but
roughly 7,500 lines [my earlier line counts were rounded])
are accounted for by the lack of interface standards and
the needless complexity and pointless diversity of many of
those interfaces.  And there are lots of other incompatible
devices available.  I give jmk a lot of credit for keeping
those line counts as low as they are.

I know, let's make a version of Plan 9 that runs on Mach;
that'll solve all our problems!  ☺


^ permalink raw reply	[flat|nested] 30+ messages in thread
* Re: [9fans] General question about hosted interfaces
@ 2001-07-12 19:31 jmk
  2001-07-12 21:09 ` Chris Locke
  0 siblings, 1 reply; 30+ messages in thread
From: jmk @ 2001-07-12 19:31 UTC (permalink / raw)
  To: 9fans

On Thu Jul 12 14:19:22 EDT 2001, chris@cjl1.demon.co.uk wrote:
>
> > Why don't you write the Plan 9 driver for the hardware you want to use?
>
> 'cos not all of us can get the data sheets required.
>

So? What makes you think that someone else can?


^ permalink raw reply	[flat|nested] 30+ messages in thread
* Re: [9fans] General question about hosted interfaces
@ 2001-07-12 18:38 nemo
  0 siblings, 0 replies; 30+ messages in thread
From: nemo @ 2001-07-12 18:38 UTC (permalink / raw)
  To: 9fans

: 'cos not all of us can get the data sheets required.

Now that you say that...

	anyone has information about the toshiba
2G pcmcia disk?

Just got one and have been looking for information
to write a driver ☺

Although linux seems to support it, I have found
no specific driver for the beast, perhaps it's a
generic pcmcia disk driver the one which works...



^ permalink raw reply	[flat|nested] 30+ messages in thread
* Re: [9fans] General question about hosted interfaces
@ 2001-07-12 15:26 jmk
  2001-07-12 18:14 ` Chris Locke
  2001-07-13 14:53 ` Douglas A. Gwyn
  0 siblings, 2 replies; 30+ messages in thread
From: jmk @ 2001-07-12 15:26 UTC (permalink / raw)
  To: 9fans

On Thu Jul 12 04:38:24 EDT 2001, DAGwyn@null.net wrote:
> ...
> Yeah, if we want to run Plan 9, currently we have to do that.
> I am not satisfied at having to dig up junk hardware.  Ideally,
> I should have one workstation that can be used for all workstation
> tasks, be it interactive 3D games or software development.  That
> workstation ought to be as nice as possible within my budget.
> ...

Why don't you write the Plan 9 driver for the hardware you want to use?


^ permalink raw reply	[flat|nested] 30+ messages in thread
* Re: [9fans] General question about hosted interfaces
@ 2001-07-11 10:54 forsyth
  2001-07-11 10:58 ` Lucio De Re
  2001-07-11 13:26 ` Boyd Roberts
  0 siblings, 2 replies; 30+ messages in thread
From: forsyth @ 2001-07-11 10:54 UTC (permalink / raw)
  To: 9fans

>>All of this is in part why I'd like to get an Ipaq (bitsy) when (if)
>>they stop being so scarce.  It's got a processor that Intel didn't

you can get them off the shelf at several shops in York.



^ permalink raw reply	[flat|nested] 30+ messages in thread
* Re: [9fans] General question about hosted interfaces
@ 2001-07-11 10:32 geoff
  2001-07-12  8:31 ` Douglas A. Gwyn
  0 siblings, 1 reply; 30+ messages in thread
From: geoff @ 2001-07-11 10:32 UTC (permalink / raw)
  To: 9fans

I'm no fan of USB, but it seems closer to what you want than SCSI.  In
fact, though, SCSI disks (for example) from different vendors do just
seem to plug in and work.  Admittedly adding support for a new class
of SCSI device is more work (unless you just punt the whole problem to
scuzz, and thus largely to the end user).

But the peripherals themselves (or classes of devices) aren't the big
problem, at least for SCSI or USB, it's the interfaces (controllers,
host adapters, whatever) themselves, at least in the wonderful world
of unstandardised PCs.  Plan 9 has drivers for the PCI Buslogic/Mylex
and NCR/Symbios/Tekram SCSI host adapters, but not the PCI Adaptec or
any of the lesser known ones.  And as far as I know, there are at
least two USB interfaces already (logically different and using
incompatible chip sets: UCHI vs OCHI and Intel's chips vs Via's).  Our
biggest headache is VGA cards, which seem to have shelf lives of a few
months, and Ethernet cards, which are beginning to have shorter shelf
lives (can you still get FA310s?) but which fortunately mostly just
reuse a few Ethernet chips (or emulately them inaccurately).

So to have a relatively stable peripheral environment one would need
to (somehow) standardise buses (internal and external), controllers
and their interface protocols, and the protocols between the
controllers and the actual peripherals.  But then the world would
change around us; people want greater speed or utility (Shrug and Pray
on live systems seems to be the only compelling reason for USB over
SCSI).  In the decade or so, the PC has been through ISA, EISA, PCI,
AGP, PCMCIA and Cardbus buses (among others), and PCI is certainly an
improvement over ISA. We've seen IDE/ATA get less stupid (real DMA,
what a concept!), SCSI get faster, USB get invented and then get
faster (to the point of being more than a toy).  Ethernet has gone
from 10Mb/s to 100Mb/s to 1GB/s and I believe the 10GB/s committee is
already at work.  And it's all gotten smaller and cheaper.  Microsoft
try to influence the design of PCs with their periodic standards
(e.g., PC98), but they don't have very good taste in hardware.  I'm
not sure who else has the influence or power to enforce hardware
standards, since nobody is in charge of the overall PC architecture.

I cope by stock-piling cheap supported controllers and trying to
simplify hardware configurations.  For CPU servers and terminals,
booting off floppy lets you avoid local disk entirely if you have
enough RAM, so Ethernet should be the only real pain for CPU servers.
There are several cheap supported cards out now; the NetGear FA310 is
a 2114x, but watch out that the 311 and 312 are not.  For terminals,
there's VGA Hell to be suffered.  The ATI Xpert 98 cards were good for
a while, but now it's hard to find any of them that don't have the
Rage XL engine, and we don't have support for those currently.  It
appears that the Nvidia TNT2 M64 situation is getting sorted out.

All of this is in part why I'd like to get an Ipaq (bitsy) when (if)
they stop being so scarce.  It's got a processor that Intel didn't
design and it looks like the only Intel PC goo in it is a PCMCIA
controller and the flash memory.  No wretched Intel interrupt
controllers, no red-hot glowing CISC processor, no IDE cretinism, 001
USB controller, 001 video interface.  38,000 lines in
/sys/src/9/pc/*.c vs. 11,700 in /sys/src/9/bitsy/*.c.  Amazing what
you can do when there's someone in charge to tell the sleazier
hardware designers, `no, you can't save 5 cents per unit by omitting
DMA, making the software swap the bytes, and leaving out a boot ROM so
that the user has to type in the boot program in hex on the serial
port at every boot'.  I'm skeptical about the bitsy's suitability as a
terminal, but it might make a fine CPU server since it's got what
matters: a reasonable CPU, a lump of RAM and an optional PCMCIA
network card (Wavelan currently).


^ permalink raw reply	[flat|nested] 30+ messages in thread
* Re: [9fans] General question about hosted interfaces
@ 2001-07-11  0:37 presotto
  0 siblings, 0 replies; 30+ messages in thread
From: presotto @ 2001-07-11  0:37 UTC (permalink / raw)
  To: 9fans

[-- Attachment #1: Type: text/plain, Size: 129 bytes --]

We already support TLS(& SSL2) encrypted connections.  That stuff
will go out with the next release.  Sean Dorward did it all.

[-- Attachment #2: Type: message/rfc822, Size: 1932 bytes --]

From: "William K. Josephson" <wkj@eecs.harvard.edu>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] General question about hosted interfaces
Date: Tue, 10 Jul 2001 19:27:02 -0400
Message-ID: <20010710192702.D14363@honk.eecs.harvard.edu>

On Tue, Jul 10, 2001 at 12:08:16PM -0700, Mike Haertel wrote:
> >In fact, the new 9P2000 version of drawterm that rsc built
> >builds from kernel sources directly.
>
> Does this mean the new version of drawterm will support SSL
> encrypted connections?

I don't know the current status, but Russ and I were discussing adding
SSL support in late February or so.

 -WJ

^ permalink raw reply	[flat|nested] 30+ messages in thread
* [9fans] General question about hosted interfaces
@ 2001-07-10 15:17 Douglas A. Gwyn
  2001-07-10 18:09 ` Lucio De Re
  0 siblings, 1 reply; 30+ messages in thread
From: Douglas A. Gwyn @ 2001-07-10 15:17 UTC (permalink / raw)
  To: 9fans

What is it, apart from someone having to do the work, that would prevent
the above-device-level portion of Plan 9, perhaps Inferno's emu, running
on top of a UNIX system (or even Windows, gasp), with the hosted-Plan-9
first importing the UNIX /-rooted file system globally then overlay-
mounting the hosted-Plan-9 specific parts of the name space?  I.e. true
rio, acme, etc. for people who are not in a position to control the
choice of operating system.

I know that would introduce inefficiencies in the data paths, but that
is not necessarily a fatal problem.


^ permalink raw reply	[flat|nested] 30+ messages in thread

end of thread, other threads:[~2001-07-14  0:40 UTC | newest]

Thread overview: 30+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2001-07-10 18:22 [9fans] General question about hosted interfaces rob pike
2001-07-10 19:08 ` Mike Haertel
2001-07-10 23:27   ` William K. Josephson
2001-07-11  8:34 ` Douglas A. Gwyn
  -- strict thread matches above, loose matches on Subject: below --
2001-07-12 21:58 jmk
2001-07-12 21:48 bwc
2001-07-13 14:53 ` Douglas A. Gwyn
2001-07-13 15:32   ` Boyd Roberts
2001-07-12 19:51 geoff
2001-07-12 19:45 geoff
2001-07-12 19:31 jmk
2001-07-12 21:09 ` Chris Locke
2001-07-12 21:24   ` Boyd Roberts
2001-07-12 22:01   ` Jim Choate
2001-07-12 18:38 nemo
2001-07-12 15:26 jmk
2001-07-12 18:14 ` Chris Locke
2001-07-13 14:53 ` Douglas A. Gwyn
2001-07-13 15:28   ` Boyd Roberts
2001-07-13 16:46     ` Douglas A. Gwyn
2001-07-14  0:40       ` Boyd Roberts
2001-07-11 10:54 forsyth
2001-07-11 10:58 ` Lucio De Re
2001-07-11 13:26 ` Boyd Roberts
2001-07-11 10:32 geoff
2001-07-12  8:31 ` Douglas A. Gwyn
2001-07-11  0:37 presotto
2001-07-10 15:17 Douglas A. Gwyn
2001-07-10 18:09 ` Lucio De Re
2001-07-11  8:34   ` Douglas A. Gwyn

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).