Porting to different architectures

Hans wrote:
> There are still a number of people on this list but
most of us are in
> "screen saver" mode 

I've been following this project for about 7 years now, and
it seems
like it regularly goes into "screen saver" mode.
:(

> Porting ELKS to another (embedded) processor will
definitely bring ELKS
> back to live (IMHO) but this is a huge and complicated
task. Just
> changing the compiler from BCC to say Watcom is already
a difficult task

I've been working on getting it to work with GCC for 386+
processors,
and it's not pretty. GCC support would make porting a lot
easier, given
the number of processor targets it supports.

The next trick would be to remove some of the PC-specific
stuff - not
all systems have a BIOS, for example. I keep meaning to dig
out the old
laptop I was given (286 based Toshiba thing) and get a
serial console
working. I'm still keen to get it running on my Psion 3C.


> (I have to admit I am a hardware designer .

May come in useful... 

> I am pretty sure that one of us can email you the
latest source files

The SourceForge CVS server works, it's just the download
instructions on
the site that are wrong.

If you replace "anonymouscvs.sourceforge.net"
with
"anonymouselks.cvs.sourceforge.net" it all
works well.

There doesn't seem to be any of the project admins (the
ones with the
CVS SSH key) awake ATM - I was going to see if I could get a
copy of the
key so that *someone* could commit changes! Failing that, I
would
(reluctantly) have to fork the project just to keep things
moving.

So, project admins - anyone awake?

-- 
Richard Wallman
-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html

Richard Wallman wrote:
[...]
> I've been working on getting it to work with GCC for
386+ processors,
> and it's not pretty. GCC support would make porting a
lot easier, given
> the number of processor targets it supports.

Is anyone interested in hearing that I've nearly managed to
get the ACK
(Minix' native compiler) into a usable state? It supports
ANSI C, K&R C,
Pascal, Modula-2, Occam and Basic for a bunch of different
architectures, which include i80, i86 and i386. The code
isn't as good
as what gcc produces, but it is a tiny fraction of the size,
and it's
also BSD licensed.

I'm still working on the syscall libraries required for the
ACK's own
built-in libc, but if you didn't want to use that, it's
more or less
there now.

(It should also be reasonably easy to retarget --- compared
to gcc, at
least. Does the 65C265 have a compiler?)

-- 
+- David Given --McQ-+ "You cannot truly appreciate
_Atlas Shrugged_
|  dgcowlark.com    | until you have read it in the
original Klingon."
| (dgtao-group.com) | --- Sea Wasp on r.a.sf.w
+- www.cowlark.com --+

David Given wrote:
> Is anyone interested in hearing that I've nearly
managed to get the ACK
> (Minix' native compiler) into a usable state?

Running under ELKS, or to compile ELKS?

If it's running under ELKS, that would be good - I've been
using
OnboardC on my Palm, but the lack of a keyboard makes
coding...interesting. My Psion 3C has a nice keyboard,
though, so coding
on the train would be a possibility. 

-- 
Richard Wallman
-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html

Richard Wallman wrote:
[...]
> Running under ELKS, or to compile ELKS?

Neither --- cross-compiling from some big operating system
to generic
i86 code. It *may* be possible to cross-compile the ACK so
that it runs
on ELKS; after all, Minix i86 did that, and it has the same
code size
restrictions ELKS does. However, this isn't a priority
right now.

There ought to be enough there now to build standalone
images, though.

-- 
+- David Given --McQ-+ "You cannot truly appreciate
_Atlas Shrugged_
|  dgcowlark.com    | until you have read it in the
original Klingon."
| (dgtao-group.com) | --- Sea Wasp on r.a.sf.w
+- www.cowlark.com --+

I'm here.

My main machine (the host) had an Athlon XP barbecue.

I'm using a temporary Linux box but haven't gotten around
to putting 
thttpd on it yet.

I'll see what I can do, but I'll definitely need someone
to mail me the 
ELKS source tarball again so I can put it back up.  I've
got real UNIX 
now, so you won't see CRLF problems anymore 

Also, I have full access to the entire ELKS project.  Mail
me if you 
need stuff done and I'll try to take care of it.

Thanks!

Jody

Richard Wallman wrote:
> Hans wrote:
>> There are still a number of people on this list but
most of us are in
>> "screen saver" mode 
> 
> I've been following this project for about 7 years
now, and it seems
> like it regularly goes into "screen saver"
mode. :(
> 
>> Porting ELKS to another (embedded) processor will
definitely bring ELKS
>> back to live (IMHO) but this is a huge and
complicated task. Just
>> changing the compiler from BCC to say Watcom is
already a difficult task
> 
> I've been working on getting it to work with GCC for
386+ processors,
> and it's not pretty. GCC support would make porting a
lot easier, given
> the number of processor targets it supports.
> 
> The next trick would be to remove some of the
PC-specific stuff - not
> all systems have a BIOS, for example. I keep meaning to
dig out the old
> laptop I was given (286 based Toshiba thing) and get a
serial console
> working. I'm still keen to get it running on my Psion
3C. 
> 
>> (I have to admit I am a hardware designer .
> 
> May come in useful... 
> 
>> I am pretty sure that one of us can email you the
latest source files
> 
> The SourceForge CVS server works, it's just the
download instructions on
> the site that are wrong.
> 
> If you replace "anonymouscvs.sourceforge.net"
with
> "anonymouselks.cvs.sourceforge.net" it all
works well.
> 
> There doesn't seem to be any of the project admins
(the ones with the
> CVS SSH key) awake ATM - I was going to see if I could
get a copy of the
> key so that *someone* could commit changes! Failing
that, I would
> (reluctantly) have to fork the project just to keep
things moving.
> 
> So, project admins - anyone awake?
> 
-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html

Hi,

Richard Wallman wrote:
> 
> There doesn't seem to be any of the project admins
(the ones with the
> CVS SSH key) awake ATM - I was going to see if I could
get a copy of the
> key so that *someone* could commit changes! Failing
that, I would
> (reluctantly) have to fork the project just to keep
things moving.
> 

	I think there is a (slow, painful, bureaucratic) mechanism
to retake 
abandoned projects. If you really want to do it, mail the
sourceforge 
team to look for a solution.


> So, project admins - anyone awake?
> 

	You can also try to mail to their sourceforge accounts 
(usersourceforge.net, or something similar). I think it is
forwarded to 
a real account (at least, it is forwarded in my case).

-- 
Javier Sedano Jarillo: javier.sedanoagora-2000.com
Agora Systems S.A.
C/Velazquez, 46, 6º derecha;    E-28001 Madrid (Spain)
Tel.: +34 676 48 29 55;         Fax.: +34 91 577 76 72
--------
Más vale ser cobarde un minuto que muerto todo el resto de
la vida.
-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html

Ar Llu, 2006-07-31 am 21:47 +0200, ysgrifennodd Javier
Sedano:
> Hi,
> 
> Richard Wallman wrote:
> > 
> > There doesn't seem to be any of the project
admins (the ones with the
> > CVS SSH key) awake ATM - I was going to see if I
could get a copy of the
> > key so that *someone* could commit changes!
Failing that, I would
> > (reluctantly) have to fork the project just to
keep things moving.
> > 
> 
> 	I think there is a (slow, painful, bureaucratic)
mechanism to retake 
> abandoned projects. If you really want to do it, mail
the sourceforge 
> team to look for a solution.

If you plan to use it as the basis for a new OS for
something else I
would fork it. The existing project is as good as dead and
it's probably
easier to just turn it into what you need than worry about
breaking the
existing code.

Alan

-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html

--- Alan Cox <alanlxorguk.ukuu.org.uk> wrote:

> Ar Llu, 2006-07-31 am 21:47 +0200, ysgrifennodd
> Javier Sedano:
> > Hi,
> > 
> > Richard Wallman wrote:
> > > 
> > > There doesn't seem to be any of the project
> admins (the ones with the
> > > CVS SSH key) awake ATM - I was going to see
if I
> could get a copy of the
> > > key so that *someone* could commit changes!
> Failing that, I would
> > > (reluctantly) have to fork the project just
to
> keep things moving.
> > > 
> > 
> > 	I think there is a (slow, painful, bureaucratic)
> mechanism to retake 
> > abandoned projects. If you really want to do it,
> mail the sourceforge 
> > team to look for a solution.
> 
> If you plan to use it as the basis for a new OS for
> something else I
> would fork it. The existing project is as good as
> dead and it's probably
> easier to just turn it into what you need than worry
> about breaking the
> existing code.

Agreed. We can keep the current version for historical
purposes, and if someone wants to build a better OS
for ancient machines, there are more popular and
better designs than the 8088.

> Alan
> 
> -
> To unsubscribe from this list: send the line
> "unsubscribe linux-8086" in
> the body of a message to majordomovger.kernel.org
> More majordomo info at 
> http://vge
r.kernel.org/majordomo-info.html
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection
around 
http://mail.yahoo.com 
-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html

On top of that, the 8088 isn't really even used in embedded
stuff 
anymore AFAIK.  From what I understand, the embedded Intel
CPU of choice 
is an i386EX and that likely can run Linux unmodified.

The biggest problem I see with ELKS is that it has a lot of
features, 
and 90% of said features are either broken, unimplemented,
or 
half-functional.  Furthermore, we're so lacking in
knowledgeable kernel 
developers with TIME available to work on the project that
it's not 
going to go anywhere.

Compounding the problems is the fact that bcc has proven
time and time 
again to be a very volatile compiler package, to the point
that even one 
version revision can mean the difference between successful
compilation 
or errors galore.

On a side note, I'm developing an OS for the
6502/65C02/65816 line of 
CPUs from Western Design Center, if anyone's interested in
that kind of 
stuff.

~Jody

Tom McCabe wrote:
> 
> --- Alan Cox <alanlxorguk.ukuu.org.uk>
wrote:
> 
>> Ar Llu, 2006-07-31 am 21:47 +0200, ysgrifennodd
>> Javier Sedano:
>>> Hi,
>>>
>>> Richard Wallman wrote:
>>>> There doesn't seem to be any of the
project
>> admins (the ones with the
>>>> CVS SSH key) awake ATM - I was going to see
if I
>> could get a copy of the
>>>> key so that *someone* could commit changes!
>> Failing that, I would
>>>> (reluctantly) have to fork the project just
to
>> keep things moving.
>>> 	I think there is a (slow, painful,
bureaucratic)
>> mechanism to retake 
>>> abandoned projects. If you really want to do
it,
>> mail the sourceforge 
>>> team to look for a solution.
>> If you plan to use it as the basis for a new OS for
>> something else I
>> would fork it. The existing project is as good as
>> dead and it's probably
>> easier to just turn it into what you need than
worry
>> about breaking the
>> existing code.
> 
> Agreed. We can keep the current version for historical
> purposes, and if someone wants to build a better OS
> for ancient machines, there are more popular and
> better designs than the 8088.
> 
>> Alan
>>
>> -
>> To unsubscribe from this list: send the line
>> "unsubscribe linux-8086" in
>> the body of a message to majordomovger.kernel.org
>> More majordomo info at 
>> http://vge
r.kernel.org/majordomo-info.html
>>
> 
> 
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam
protection around 
> http://mail.yahoo.com 
> -
> To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
> the body of a message to majordomovger.kernel.org
> More majordomo info at  http://vge
r.kernel.org/majordomo-info.html
> 
-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html

Jody Bruchon wrote:

> My main machine (the host) had an Athlon XP barbecue.

Heh - I had one over the weekend. Sorting out my Dad's
computer, IDE cable got
in the way of the CPU fan and I didn't notice until it
started a reset loop.
It took a couple of minutes before it cooled down enough for
the motherboard
to be willing/able to report the CPU temperature -
apparently it can't report
temperatures over 99 degrees C. Amazing thing it, it still
works!

Cup of tea, anyone? :P

> I'll see what I can do, but I'll definitely need
someone to mail me the
> ELKS source tarball again so I can put it back up.

Why put tarballs on a separate site when the SourceForge
tools work? If CVS is
good to go, that's a lot easier to manage that manually
patching and releasing
tarballs, especially when (and I'm being optimistic and not
saying 'if') there
are multiple developers working on different things.

You could always create tarballs from a CVS export, and if
SF host it then you
don't have to worry about your computer/bandwidth/etc.

> Also, I have full access to the entire ELKS project. 
Mail me if you
> need stuff done and I'll try to take care of it.

Do you have the SSH key needed for CVS commits?
Do you have the rights needed to add/remove project
admins/developers?

I posted a PATCH a while back that adds the ability to
select the root device
at compile time - I'd like that tested and committed, if
possible.

-- 
Richard Wallman

            Where are we going, and why am I in this
handbasket?
-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html

Ar Maw, 2006-08-01 am 01:27 -0400, ysgrifennodd Jody
Bruchon:
> On top of that, the 8088 isn't really even used in
embedded stuff 
> anymore AFAIK.  From what I understand, the embedded
Intel CPU of choice 
> is an i386EX and that likely can run Linux unmodified.

Some of the embedded 386 chips can some can't. Not all of
them have a
paging MMU some have just segmentation. Nowdays the CPU of
choice seems
to be either the VIA fanless processors or the AMD (ex Nat
Semi) Geode
system on a chip, both of which are Pentium class or higher

-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html

Agreed... What about MIPS chips ? I hear they are pretty
popular.
And what about custom FPGA chips ? I am not an Elec...
engineer, but I
am interested as to what kind of operating systems do custom
FPGA
chips run ?

regards,
Vikas
On 8/1/06, Alan Cox <alanlxorguk.ukuu.org.uk>
wrote:
> Ar Maw, 2006-08-01 am 01:27 -0400, ysgrifennodd Jody
Bruchon:
> > On top of that, the 8088 isn't really even used
in embedded stuff
> > anymore AFAIK.  From what I understand, the
embedded Intel CPU of choice
> > is an i386EX and that likely can run Linux
unmodified.
>
> Some of the embedded 386 chips can some can't. Not all
of them have a
> paging MMU some have just segmentation. Nowdays the CPU
of choice seems
> to be either the VIA fanless processors or the AMD (ex
Nat Semi) Geode
> system on a chip, both of which are Pentium class or
higher
>
> -
> To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
> the body of a message to majordomovger.kernel.org
> More majordomo info at  http://vge
r.kernel.org/majordomo-info.html
>
-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html

Richard Wallman wrote:
> Jody Bruchon wrote:
> 
> 
>>My main machine (the host) had an Athlon XP
barbecue.
> 
> 
> Heh - I had one over the weekend. Sorting out my Dad's
computer, IDE cable got
> in the way of the CPU fan and I didn't notice until it
started a reset loop.
> It took a couple of minutes before it cooled down
enough for the motherboard
> to be willing/able to report the CPU temperature -
apparently it can't report
> temperatures over 99 degrees C. Amazing thing it, it
still works!
> 
> Cup of tea, anyone? :P
I had a system do a similar thing cause the fan get it's
power via the 
disk drive power cables, and if the wires are put in a
certain position, 
the fan quits working. At any rate, when I rebooted, i got a
"Power 
Managemenr failure" from the BIOS, checking it out
revealed that the CPU 
was roasting at a good 253 degrees, while running. Luckly it
didn't break.

-- 
The real problem with C++ for kernel modules is: the
language just sucks.
	-- Linus Torvalds
-
To unsubscribe from this list: send the line
"unsubscribe linux-8086" in
the body of a message to majordomovger.kernel.org
More majordomo info at  http://vge
r.kernel.org/majordomo-info.html