Kickstarting a revolution ...

[Stefan Reinauer]

Dear OpenBIOS folks,

The past few weeks have been silent on this mailing list, but fortunately that
was not a bad thing. In fact, if you had a look at the web page, you might have
noticed it already. We finally have some code now that aims toward the projects
goal, creating a free implementation of IEEE-1275.

administrative
--------------
The OpenBIOS website has been updated and restructured as it didn't really lead
to the right direction for working on OpenBIOS. Have a look at 
    http://www.openbios.net/ or http://www.freiburg.linux.de/OpenBIOS/
For ease of updates the website meta language (wml) is now used for creating
the pages. Thanks a lot to Patrick Mauritz for his work.
Please check the documentation links on
    http://www.freiburg.linux.de/OpenBIOS/docs/
for information on Open Firmware, OpenBIOS, Forth, ...
If you miss information on the OpenBIOS web page, please drop a note to the
list for an update.

Second, we have a cvs tree for the OpenBIOS project now. Access is granted read only
for anonymous users and read/write for core developers. Read the information on
http://www.freiburg.linux.de/OpenBIOS/dev/cvs.html
For web access to the cvs tree go to
http://openbios.ph-freiburg.de/cgi-dom/viewcvs.cgi/
Thanks to Armin JoLo Herbert for his efforts of finding a machine and setting
up the cvs server and viewcvs.

detok
-----
Detok is a an fcode detokenizer. That's a small utility that "disassembles" fcode
bytecode files. It's useful for debugging purposes and it's the first part of
the OpenBIOS development suite. Detok follows IEEE-1275-1994 and should work
pretty stable.  This component should be heavily tested as it is about feature
complete. The latest release of detok can be found at
http://www.freiburg.linux.de/OpenBIOS/bin/detok-0.2.3.tar.gz

feval
-----
Feval is the the fcode evaluator needed to execute the cpu independant hardware
initialization part of OpenBIOS. It is able to execute simple fcode bytecode
files right now, but there are many fcode words missing and it is quite a big
step left before this is complete. But it is a clear base to discuss and work
on. Feval is tested on IA32, IA64, Alpha, Sparc and m68k and should be endian
and 64bit clean. Preliminary support for packages and properties is implemented
but not yet ready for prime time. The latest release of feval can be found at
http://www.freiburg.linux.de/OpenBIOS/bin/feval-0.2.6.tar.gz

Best regards,
  Stefan Reinauer

[Mads Martin Jørgensen]

* Stefan Reinauer  [Dec 19. 2001 01:56]:
> The OpenBIOS website has been updated and restructured as it didn't really lead
> to the right direction for working on OpenBIOS. Have a look at 
>    http://www.openbios.net/ or http://www.freiburg.linux.de/OpenBIOS/
> For ease of updates the website meta language (wml) is now used for creating
> the pages. Thanks a lot to Patrick Mauritz for his work.
> Please check the documentation links on
>    http://www.freiburg.linux.de/OpenBIOS/docs/
> for information on Open Firmware, OpenBIOS, Forth, ...
> If you miss information on the OpenBIOS web page, please drop a note to the
> list for an update.

People who do IRC should also stop by #openbios on OPN,
irc.openprojects.net. We've got the channel registered.
> Best regards,
>   Stefan Reinauer

:-) Maybe our project founder would elaborate on that one? Not all of us
speak Forth that good (yet :)

[Stefan Reinauer]

* Mads Martin Jørgensen  [011219 12:50]:
> Stefan Reinauer  [Dec 19. 2001 01:56]:
> >
>
> :-) Maybe our project founder would elaborate on that one? Not all of us
> speak Forth that good (yet :)

:-) Actually it should have been a little forth word definition that
calculates faculty of the given number recursively. But there were 2 mistakes
in it.
The correct version would be:
    Ok : fc recursive dup 1 > if dup 1 - fc * then ; 42 fc .
(with Ok being the Open Firmware prompt)

The program can be written in a more readable way:

: FC RECURSIVE ( n -- n! | 1 )	\ This is the word definition including
    			\ the stack diagram.
  DUP 1 >			\ safe n and compare it to 1
  IF				\ if (n>1)
    DUP 1 - FC *		\ fc(n) = fc(n-1) * n
  ELSE
    DROP 1			\ If n=1 or below (n! not defined for n<1)
    			\ bail out of recursion.
  THEN
;

"42 fc ." then pushes 42 to the stack, calculates faculty and prints it.

Best regards,
  Stefan Reinauer

[Ronald G Minnich]

let me know if your fcode could be loaded under linuxbios instead of
loading linux. I think we can save each other a lot of work.

ron

-
To unsubscribe: send mail to [e-mail]
with 'unsubscribe openbios' in the body of the message
http://www.freiburg.linux.de/OpenBIOS/ - free your system..

[James Logan]

Hello,

I was wondering if anyone would be putting together
a task list and requesting developers to work on
certain aspects? Basically, what can I do to help?

Regards,
James

On Wed, 19 Dec 2001, Ronald G Minnich wrote:
> Date: Wed, 19 Dec 2001 07:43:17 -0700 (MST)
> From: Ronald G Minnich 
> Reply-To: [e-mail]
> To: [e-mail]
> Subject: Re: [OpenBIOS] Kickstarting a revolution ...
>
> let me know if your fcode could be loaded under linuxbios instead of
> loading linux. I think we can save each other a lot of work.
>
> ron
>
> To unsubscribe: send mail to [e-mail]
> with 'unsubscribe openbios' in the body of the message
> http://www.freiburg.linux.de/OpenBIOS/ - free your system..

-
To unsubscribe: send mail to [e-mail]
with 'unsubscribe openbios' in the body of the message
http://www.freiburg.linux.de/OpenBIOS/ - free your system..

[Armin Herbert]

Hi James,

on Wed, 19 Dec 2001 16:36 GMT James Logan wrote:
> I was wondering if anyone would be putting together
> a task list and requesting developers to work on
> certain aspects? Basically, what can I do to help?

I suggest you join us on irc - openproject servers, #openbios channel, at 
daytime (gmt). And after xmas :) I think it's quite good to have an online 
chat about  it, so people can get the big picture by asking questions and 
discussing etc.

[Stefan Reinauer]

* Ronald G Minnich  [011219 15:43]:
> let me know if your fcode could be loaded under linuxbios instead of
> loading linux. I think we can save each other a lot of work.

We're quite some time away from this approach. What we definitely have
to have is an almost complete fcode evaluator, so we can think about using
the existing fcode capable pci hardware. There are some Adaptec SCSI host 
adaptors and some ATI cards that are definitely supported. A very useful thing
to have would be a list of supported hardware.
We also need at least some code on top of the evaluator, written in Forth.
This does not have to be complete to boot an OS, but it should be able to 
search for expansion cards, initialize them and load the kernel or a
bootloader.
I'd really appreciate cooperation between LinuxBIOS and OpenBIOS. It does
not make sense to invent the wheel several times. Both the lowlevel
init code and the abstraction levels we are implementing now are pretty big
tasks, so having more people work on both makes perfect sense.

Best regards,
  Stefan Reinauer