The future of 9x...

The future of 9x...

[James]

Russ wrote (among other things):
> I also wouldn't mind seeing 9x revived.  I would do it
> myself except I don't use X at all anymore.
> I expect to be using X a lot in a couple of months, and
> I might get sufficiently annoyed to do it then.  

On at least two occasions in the last 6-8 months I have handed out
the 9x source to people who `desperately needed it' asking them to
get back to me when (if) they got it working so I could make some
kind of release (I've got a serious time shortage at the moment while
I finish my degree).  As of now I've not heard back from them.

The good news is, I want it too and in a month or two will have
finished (ha! ;-) and I will have time to do all the things that
have been buggin me for the last year (like clean up 9term, work on
9x, &c.).

					Matty.
-- 
James Matthew Farrow                    | "For in that moment I beheld the ruin
matty@cs.usyd.edu.au                    | of my existence.  My world fell dark
Basser Department of Computer Science   | and my life became a shallow dream.
Sydney University - Fax: +61 2 9351 3838| `Odi et amo. Excrucior.'" - Tlindah

The future of 9x...

[howard]

OK, I'll finally fess up.  I did the Plan 9 X server here at Bell Labs,
but it never fully worked.  Later I made another attempt, and got
it working just fine, but by that time we were using Brazil, which
has a different graphics model.  Basically the server works by
doing all screen operations into a memory pixmap, keeping track
of changed bounding boxes, and transferring those changed
rectangles to the screen at opportune moments.  It works surprisingly
well.  It wouldn't be too hard to do for Plan 9, but I've always been
to busy.  Also, there are three problems I can think of:

- the Plan 9 graphics model only allows you to write whole lines
 of bitmaps, so transferring narrow vertical rectangles to the screen has
 to be done carefully---my abortive Plan 9 port used a series of
 off screen temporaries of varying widths that could be used as
 midway transfer bitmaps.
- the whole make procedure is rather dodgy, and some of what I did
 is so far in the past that I can't quite remember what I did either.
 I used a make, the korn shell (ape/sh), and ape/cc.  I used imake to
 make the initial makefiles, but there were difficulties in what imake
 expected cpp to do versus what our cpp does (ours followed the
 ansi standard), so I ended up using a gnu cpp for that part.  Various
 parts of these have sources that I don't think I can distribute.  And
 even with the makefiles that are there, the situation with building for
 multiple architectures is awful -- you have to clean out completely,
 and I think build on the architecture you're building for.  All or much
 of this could be corrected but I don't have the time or inclination.
 This is the reason I haven't even released the client libraries.
- I needed to add another fork flag to fork the rendezvous group
 in brazil, and that never made it back to Plan 9, as far as I know.
 It was needed to make select() work in the presence of forked
 subprocesses that also wanted to select, but I forget the details.
 Perhaps there is a workaround.

Anyway, with all this said, I'm sure what I've done is more use as a start
than starting from scratch, and I'm willing to put it on the net for
people to have a go with (with the understanding that I'm too busy
with Inferno at the moment to help or answer many questions).
Any suggestions for how to do it?  My stuff is intimately mixed in
with the whole X11/R6 release (with patches up to patch #04),
and I would have to put all of that up  -- about 210 Megabytes.

Howard Trickey
howard@plan9.bell-labs.com

The future of 9x...

[Digby]

>On at least two occasions in the last 6-8 months I have handed out
>the 9x source to people who `desperately needed it' asking them to
>get back to me when (if) they got it working so I could make some
>kind of release (I've got a serious time shortage at the moment while
>I finish my degree).  As of now I've not heard back from them.
>
>The good news is, I want it too and in a month or two will have
>finished (ha! ;-) and I will have time to do all the things that
>have been buggin me for the last year (like clean up 9term, work on
>9x, &c.).
>
That is good news.

I havn't got lots of time either at the moment, but getting
X going on it would be usefull to me, and probably a good way
to learn some of the techniques of programming on Plan 9.
I just didn't want to completely re-invent the proverbial
wheel..

Let me know if I can help out.

Regards,
DigbyT
-- 
Digby R. S. Tarvin                                              digbyt@acm.org
http://www.users.dircon.co.uk/~cthulhu/