[9fans] SSH v2

[9fans] SSH v2

[Lyndon Nerenberg]

The TODO list says SSH2 is stalled.  Who was working on this?  And is  
there a snapshot of the work in progress?  I need SSH2 if I'm going  
to run native at the office, so I'll pick this up.  (But I'm stalled  
until we have a bge driver so's I can get my laptop on the wire.)

--lyndon

Re: [9fans] SSH v2

[jmk]

On Fri Feb 10 21:46:59 EST 2006, lyndon@orthanc.ca wrote:
> The TODO list says SSH2 is stalled.  Who was working on this?  And is  
> there a snapshot of the work in progress?  I need SSH2 if I'm going  
> to run native at the office, so I'll pick this up.  (But I'm stalled  
> until we have a bge driver so's I can get my laptop on the wire.)
> 
> --lyndon

Which bge, the 44xx or the 57xx?

[9fans] bge

[Lyndon Nerenberg]

On Feb 10, 2006, at 7:03 PM, jmk@plan9.bell-labs.com wrote:

> Which bge, the 44xx or the 57xx?

57xx.  (Dell Latitude D610)

Re: [9fans] bge

[jmk]

As mentioned last week, Christopher Nielsen <cnielsen@pobox.com>
has volunteered some time to work on the 57xx driver.
I'm expecting a laptop with a 44xx chip to arrive while
I am away all next week. Whether I'll decide to piss myself off
with the Atheros driver or the Broadcom driver first is up for
grabs, send your votes in.

On Fri Feb 10 22:38:41 EST 2006, lyndon@orthanc.ca wrote:
> 
> On Feb 10, 2006, at 7:03 PM, jmk@plan9.bell-labs.com wrote:
> 
> > Which bge, the 44xx or the 57xx?
> 
> 57xx.  (Dell Latitude D610)

Re: [9fans] bge

[jmk]

Actually, better than votes, send in your drivers...

On Fri Feb 10 23:14:09 EST 2006, jmk@plan9.bell-labs.com wrote:
> As mentioned last week, Christopher Nielsen <cnielsen@pobox.com>
> has volunteered some time to work on the 57xx driver.
> I'm expecting a laptop with a 44xx chip to arrive while
> I am away all next week. Whether I'll decide to piss myself off
> with the Atheros driver or the Broadcom driver first is up for
> grabs, send your votes in.
> 
> On Fri Feb 10 22:38:41 EST 2006, lyndon@orthanc.ca wrote:
> > 
> > On Feb 10, 2006, at 7:03 PM, jmk@plan9.bell-labs.com wrote:
> > 
> > > Which bge, the 44xx or the 57xx?
> > 
> > 57xx.  (Dell Latitude D610)

Re: [9fans] bge

[Lyndon Nerenberg]

On Feb 10, 2006, at 8:12 PM, jmk@plan9.bell-labs.com wrote:

> I'm expecting a laptop with a 44xx chip to arrive while
> I am away all next week. Whether I'll decide to piss myself off
> with the Atheros driver or the Broadcom driver first is up for
> grabs, send your votes in.

You have the wrong Broadcom chipset, so I'll throw Scotch at the  
Atheros driver.  But only for the next week ;-P

Re: [9fans] bge

[jmk]

>From the Fount of All Knowledge:

uriel     (from what I recall, there are some issues with something
          reading one byte at a time or somesuch, but I'm personally
          clueless about USB stack internals, but the people
          I have heard talk about it do have a clue)
          didn't jmk ask for testers of the damn broadcom drivers
          like months ago? /me has yet to see a single line of
          code of them... *Sigh*

I think you are referring to this:

On Fri Feb 10 23:14:07 EST 2006, jmk@plan9.bell-labs.com wrote:
> As mentioned last week, Christopher Nielsen <cnielsen@pobox.com>
> has volunteered some time to work on the 57xx driver.
> I'm expecting a laptop with a 44xx chip to arrive while
> I am away all next week. Whether I'll decide to piss myself off
> with the Atheros driver or the Broadcom driver first is up for
> grabs, send your votes in.
>
> On Fri Feb 10 22:38:41 EST 2006, lyndon@orthanc.ca wrote:
> >
> > On Feb 10, 2006, at 7:03 PM, jmk@plan9.bell-labs.com wrote:
> >
> > > Which bge, the 44xx or the 57xx?
> >
> > 57xx.  (Dell Latitude D610)

which does not ask for testers, only votes, and was less than a
month ago. Anyway, a BCM4401 driver has been written and
is being tested by others. However, in light of the following:

uriel     the live CD needs a bit of polishing, but I plan to
          merge it with the installer
          but before that comes killing 9load

I'll just hang on to it and save myself the time and trouble of
polishing it off until I can integrate it with your replacement
for 9load (if necessary). Let me know when it's ready.

--jim

Re: [9fans] bge

[uriel]

> uriel     the live CD needs a bit of polishing, but I plan to
>           merge it with the installer
>           but before that comes killing 9load
>
> I'll just hang on to it and save myself the time and trouble of
> polishing it off until I can integrate it with your replacement
> for 9load (if necessary). Let me know when it's ready.

As russ explained, the replacement for 9load is a standard Plan 9
kernel.  I am still thinking about what the best way design the user
space bits that parse the plan9.ini invoque /dev/reboot. I'm happy to
hear suggestions.

uriel

Re: [9fans] bge

[Russ Cox]

> As russ explained, the replacement for 9load is a standard Plan 9
> kernel.  I am still thinking about what the best way design the user
> space bits that parse the plan9.ini invoque /dev/reboot. I'm happy to
> hear suggestions.

it's great that you can just think through the design
without coding anything.  i wish i could do that.

russ

Re: [9fans] bge

[andrey mirtchovski]

> it's great that you can just think through the design
> without coding anything.  i wish i could do that.

the rise of the philosopher programmer, surely! programmers who write
code are dead like the dinosaurs.

Re: [9fans] bge

[Bruce Ellis]

it's  a matter of balance, i find.  if you write the code without
thinking through the design you are doomed, except for
very trivial tasks.  if you think through the design, and think,
and think, and never code then you may have had fun but
don't end up with a program.

brucee

On 3/10/06, andrey mirtchovski <mirtchovski@gmail.com> wrote:
> > it's great that you can just think through the design
> > without coding anything.  i wish i could do that.
>
> the rise of the philosopher programmer, surely! programmers who write
> code are dead like the dinosaurs.

Re: [9fans] bge

[jmk]

On Thu Mar  9 16:05:53 EST 2006, bruce.ellis@gmail.com wrote:
> it's  a matter of balance, i find.  if you write the code without
> thinking through the design you are doomed, except for
> very trivial tasks.  if you think through the design, and think,
> and think, and never code then you may have had fun but
> don't end up with a program.
>
> brucee
> ...

Yes. The following seems to hit all the bases - there's a lot
of think and think and think about the trivial part with no code
forthcoming and no thought whatsoever given to the hard part.
And he doesn't sound like he's having any fun, either. *SIGH*

--jim

On Thu Mar  9 14:45:34 EST 2006, uriel@cat-v.org wrote:
> > uriel     the live CD needs a bit of polishing, but I plan to
> >           merge it with the installer
> >           but before that comes killing 9load
> >
> > I'll just hang on to it and save myself the time and trouble of
> > polishing it off until I can integrate it with your replacement
> > for 9load (if necessary). Let me know when it's ready.
>
> As russ explained, the replacement for 9load is a standard Plan 9
> kernel.  I am still thinking about what the best way design the user
> space bits that parse the plan9.ini invoque /dev/reboot. I'm happy to
> hear suggestions.
>
> uriel

Re: [9fans] bge

[Dan Cross]

On Thu, Mar 09, 2006 at 08:45:03PM +0100, uriel@cat-v.org wrote:
> As russ explained, the replacement for 9load is a standard Plan 9
> kernel.  I am still thinking about what the best way design the user
> space bits that parse the plan9.ini invoque /dev/reboot. I'm happy to
> hear suggestions.

No need for suggestions, surely.  It sounds like you've got the situation
well in hand.  We'll defer to your superior knowledge and experience and
just wait for your solution.

	- Dan C.

Re: [9fans] bge

[Bruce Ellis]

Dan, some people have unusual skills.  Let's play pool again.

brucee

On 3/10/06, Dan Cross <cross@math.psu.edu> wrote:
> On Thu, Mar 09, 2006 at 08:45:03PM +0100, uriel@cat-v.org wrote:
> > As russ explained, the replacement for 9load is a standard Plan 9
> > kernel.  I am still thinking about what the best way design the user
> > space bits that parse the plan9.ini invoque /dev/reboot. I'm happy to
> > hear suggestions.
>
> No need for suggestions, surely.  It sounds like you've got the situation
> well in hand.  We'll defer to your superior knowledge and experience and
> just wait for your solution.
>
>        - Dan C.
>
>

Re: [9fans] bge

[Dan Cross]

On Fri, Mar 10, 2006 at 05:44:54PM +1100, Bruce Ellis wrote:
> Dan, some people have unusual skills.  Let's play pool again.

Oh, burn.

Okay, it's on.  Brucee just took it to 11.

	- Dan C.

Re: [9fans] SSH v2

[Russ Cox]

> The TODO list says SSH2 is stalled.  Who was working on this?  And is
> there a snapshot of the work in progress?  I need SSH2 if I'm going
> to run native at the office, so I'll pick this up.  (But I'm stalled
> until we have a bge driver so's I can get my laptop on the wire.)

/n/sources/contrib/rsc/ssh2.tar.gz

The ssh2 directory has my recent attempt at starting over.
The ssh2/old directory has wkj's earlier attempt.  I think his
version is unnecessarily complex as far as its use of channels
and I was trying to do something simpler based on a
libmux-ish approach.  But I implemented very little.

Russ

Re: [9fans] SSH v2

[William Josephson]

On Fri, Feb 10, 2006 at 10:10:02PM -0500, Russ Cox wrote:
> The ssh2 directory has my recent attempt at starting over.
> The ssh2/old directory has wkj's earlier attempt.  I think his
> version is unnecessarily complex as far as its use of channels
> and I was trying to do something simpler based on a
> libmux-ish approach.  But I implemented very little.

I implemented channels?  Who knew.
All that I recall doing was the
transport layer and crypto.  The
channel multiplexor never got
implemented, as far as I know.