[LinuxBIOS] Improvements to "Supported Motherboards" wiki page.

[Uwe Hermann]

Hi,

I've modified the wiki page
http://www.linuxbios.org/Supported_Motherboards
quite a bit recently. I'm planning some more changes to make it easier
for people interested in LinuxBIOS to check whether their mainboard is
supported.

I'd like to remove the "Specification" (mostly useless IMHO, info can be
found in the linked vendor pages) and "Chipsets" entries.
They should be replaced with the most important parts someone has to
check to know whether the mainboard is supported.

At a bare minimum I think these entries should be listed:
 * Northbridge
 * Southbridge
 * Super I/O

Any further entries? CPU/Socket maybe?

Also, I'll add an entry for _all_ mainboards in the svn repository, and
list those which are not (yet) functional as "work in progress" or so...

Comments?

Uwe.

[Stefan Reinauer]

* Uwe Hermann  [061020 23:08]:
> I'd like to remove the "Specification" (mostly useless IMHO, info can be
> found in the linked vendor pages) and "Chipsets" entries.
> They should be replaced with the most important parts someone has to
> check to know whether the mainboard is supported.
>
> At a bare minimum I think these entries should be listed:
> Northbridge
> Southbridge
> Super I/O
>
> Any further entries? CPU/Socket maybe?

CPU/Socket might be useful because it makes it easier to estimate the
effort for porting to a new mainboard by just doing a quick check of the
page.

In addition to southbridge we might want to add "other bridges" like the
8131 PCI-X bridge (we call all of them southbridges in LinuxBIOS context
so far). Maybe "special items"? Then we could add HTC connectors, PCIe,
and some such tags.
> Also, I'll add an entry for _all_ mainboards in the svn repository, and
> list those which are not (yet) functional as "work in progress" or so...

great!

[Uwe Hermann]

Hi,

On Sat, Oct 21, 2006 at 11:20:49AM +0200, Stefan Reinauer wrote:
> CPU/Socket might be useful because it makes it easier to estimate the
> effort for porting to a new mainboard by just doing a quick check of the
> page.

Done.

I've now added all boards from v2 with their respective data (where I
was able to determine it), as well as the v1 boards.

I noticed that almost none of the v1 boards were ported to v2.
Volunteers welcome :)

Please everyone let me know if you have more information on any "?" entries
on the http://www.linuxbios.org/Supported_Motherboards page (or update the
wiki directly).

The most important right now is the Status field, it should be clear
for newbies from looking at the page whether they can try to use the
code and expect it to work.

The color codes are:
 * green: complete, supported, everything works.
 * yellow: most hardware works, but some smaller problems aren't fixed yet
 * red: port not yet complete, on hold, or major components aren't
   supported, yet.

Thanks, Uwe.

[da d]

Thats quite a project you have undertaken, im sure now you have raised
awareness, others will not deign to contribute.

Thank You

On 10/22/06, Uwe Hermann  wrote:
> Hi,
>
> On Sat, Oct 21, 2006 at 11:20:49AM +0200, Stefan Reinauer wrote:
> > CPU/Socket might be useful because it makes it easier to estimate the
> > effort for porting to a new mainboard by just doing a quick check of the
> > page.
>
> Done.
>
> I've now added all boards from v2 with their respective data (where I
> was able to determine it), as well as the v1 boards.
>
> I noticed that almost none of the v1 boards were ported to v2.
> Volunteers welcome :)
>
> Please everyone let me know if you have more information on any "?" entries
> on the http://www.linuxbios.org/Supported_Motherboards page (or update the
> wiki directly).
>
> The most important right now is the Status field, it should be clear
> for newbies from looking at the page whether they can try to use the
> code and expect it to work.
>
> The color codes are:
> green: complete, supported, everything works.
> yellow: most hardware works, but some smaller problems aren't fixed yet
> red: port not yet complete, on hold, or major components aren't
> supported, yet.
> Thanks, Uwe.
>
> Uwe Hermann
> http://www.hermann-uwe.de
> http://www.it-services-uh.de  | http://www.crazy-hacks.org
> http://www.holsham-traders.de | http://www.unmaintained-free-software.org
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.5 (GNU/Linux)
>
> iD8DBQFFO4DRXdVoV3jWIbQRAnGKAJ0Yj58ur/yU7ilQ/5Y7n1CynWhNzwCfWPNF
> m7avzwOzOJ9Eri4gRjhd6Jg=
> =i4eN
> -----END PGP SIGNATURE-----

[ron minnich]

OLPC status should all be green
arima/hdama should be green
artecgroup should be green.
ibm 325 and 326 should be green (we did the port years ago and shipped the
systems back to IBM ...)
yinghai is the sun ultra40 in the tree yet? (it still needs some superio
mods for fans though)

ron

[yhlu]

Iwill dk8_htx should be green.

sun ultra40 should be green. ( I will check in some changes about it,
the fans speed reducing worked).

YH

On 10/22/06, ron minnich  wrote:
> OLPC status should all be green
> arima/hdama should be green
> artecgroup should be green.
> ibm 325 and 326 should be green (we did the port years ago and shipped the
> systems back to IBM ...)
> yinghai is the sun ultra40 in the tree yet? (it still needs some superio
> mods for fans though)
>
> ron

[Uwe Hermann]

Hi,

On Sun, Oct 22, 2006 at 09:33:00AM -0600, ron minnich wrote:
> OLPC status should all be green
> arima/hdama should be green
> artecgroup should be green.
> ibm 325 and 326 should be green (we did the port years ago and shipped the
> systems back to IBM ...)
> yinghai is the sun ultra40 in the tree yet? (it still needs some superio
> mods for fans though)

Updated, thanks.

Uwe.

[Richard Smith]

> Please everyone let me know if you have more information on any "?" entries
> on the http://www.linuxbios.org/Supported_Motherboards page (or update the
> wiki directly).

Excellent work.. Consider adding a color legend?  Red and green are
pretty obviousto the lay reader but yellow may not be.

[Uwe Hermann]

Hi,

On Mon, Oct 23, 2006 at 08:36:17AM -0500, Richard Smith wrote:
> Excellent work.. Consider adding a color legend?  Red and green are
> pretty obviousto the lay reader but yellow may not be.

Good point. Added.

Uwe.

[Indrek Kruusa]

Uwe Hermann wrote:
> Hi,
>
> I've modified the wiki page
> http://www.linuxbios.org/Supported_Motherboards
> quite a bit recently. I'm planning some more changes to make it easier
> for people interested in LinuxBIOS to check whether their mainboard is
> supported.
>
> I'd like to remove the "Specification" (mostly useless IMHO, info can be
> found in the linked vendor pages) and "Chipsets" entries.
> They should be replaced with the most important parts someone has to
> check to know whether the mainboard is supported.
>
> At a bare minimum I think these entries should be listed:
> Northbridge
> Southbridge
> Super I/O
>
> Any further entries? CPU/Socket maybe?
>
> Also, I'll add an entry for _all_ mainboards in the svn repository, and
> list those which are not (yet) functional as "work in progress" or so...
>
> Comments?

Nice work! Couple of fixes: ArtecGroup dbe61 doesn't have separate Super 
I/O and CPU is AMD Geode™ LX 700.

thanks,
Indrek

[Uwe Hermann]

Hi,

On Mon, Oct 23, 2006 at 11:06:57AM +0300, Indrek Kruusa wrote:
> Nice work! Couple of fixes: ArtecGroup dbe61 doesn't have separate Super 
> I/O and CPU is AMD Geode™ LX 700.

Done. Does 'Socket' apply for the Geode at all? Shall I put a
"maintainer" for the boardon the wiki page?

Uwe.

[Indrek Kruusa]

Uwe Hermann wrote:
> Hi,
>
> On Mon, Oct 23, 2006 at 11:06:57AM +0300, Indrek Kruusa wrote:
> > Nice work! Couple of fixes: ArtecGroup dbe61 doesn't have separate Super 
> > I/O and CPU is AMD Geode™ LX 700.
>
> Done. Does 'Socket' apply for the Geode at all? Shall I put a
> "maintainer" for the boardon the wiki page?

Geode GX/LX CPU (only BGD/BGU (EBGA/TEPBGA) packages) are always 
soldered onto the board, so socket is obsolete parameter.

About the maintainer - someone from Artec Group but I can't say at 
moment who.

I hope I can send some URLs for dbe61 quite soon.
About the status of dbe61. We have working linuxBIOS but commited 
changes (by Ron) into the mainline needs to be tested. There are some 
differences which are good for linuxBIOS but we can't take those to our 
own tree at this stage. I can work a bit later on this.

Thanks!

Indrek

[Tom Sylla]

A couple of notes: "GX2" was only an internal name for the part, it
has and always will be sold as "GX". Yes it is confusing, GX1 came
before GX, but that is how it is. All the "GX2"s should be "GX"s.
Also, the CPU for both OLPC platforms is:

AMD Geode™ GX 500@1.0W processor

which also is on the Lippert board.

To answer your question about Sockets for Geode, GX and LX are only
available as BGAs, and there are really no commercial sockets for
them. You could just make the "N/A". GX1 was a PGA, and had a socket,
but I don't remember which.

Tom

On 10/20/06, Uwe Hermann  wrote:
> Hi,
>
> I've modified the wiki page
> http://www.linuxbios.org/Supported_Motherboards
> quite a bit recently. I'm planning some more changes to make it easier
> for people interested in LinuxBIOS to check whether their mainboard is
> supported.
>
> I'd like to remove the "Specification" (mostly useless IMHO, info can be
> found in the linked vendor pages) and "Chipsets" entries.
> They should be replaced with the most important parts someone has to
> check to know whether the mainboard is supported.
>
> At a bare minimum I think these entries should be listed:
> Northbridge
> Southbridge
> Super I/O
>
> Any further entries? CPU/Socket maybe?
>
> Also, I'll add an entry for _all_ mainboards in the svn repository, and
> list those which are not (yet) functional as "work in progress" or so...
>
> Comments?
> Uwe.
>
> Uwe Hermann
> http://www.hermann-uwe.de
> http://www.it-services-uh.de  | http://www.crazy-hacks.org
> http://www.holsham-traders.de | http://www.unmaintained-free-software.org
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.5 (GNU/Linux)
>
> iD8DBQFFOTrdXdVoV3jWIbQRAihjAJ4nN5wNxnTIGihBJ5dUmMXsJBni2gCeLJJ6
> J6zyMZ6iy4mmp63cgky5CG4=
> =jAoC
> -----END PGP SIGNATURE-----

[Tom Sylla]

Oops, one other thing, the OLPC platforms have no Super I/O. (they
have an EC, but it does not get programmed in the "LB Super I/O way")

On 10/20/06, Uwe Hermann  wrote:
> Hi,
>
> I've modified the wiki page
> http://www.linuxbios.org/Supported_Motherboards
> quite a bit recently. I'm planning some more changes to make it easier
> for people interested in LinuxBIOS to check whether their mainboard is
> supported.
>
> I'd like to remove the "Specification" (mostly useless IMHO, info can be
> found in the linked vendor pages) and "Chipsets" entries.
> They should be replaced with the most important parts someone has to
> check to know whether the mainboard is supported.
>
> At a bare minimum I think these entries should be listed:
> Northbridge
> Southbridge
> Super I/O
>
> Any further entries? CPU/Socket maybe?
>
> Also, I'll add an entry for _all_ mainboards in the svn repository, and
> list those which are not (yet) functional as "work in progress" or so...
>
> Comments?
> Uwe.
>
> Uwe Hermann
> http://www.hermann-uwe.de
> http://www.it-services-uh.de  | http://www.crazy-hacks.org
> http://www.holsham-traders.de | http://www.unmaintained-free-software.org
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.5 (GNU/Linux)
>
> iD8DBQFFOTrdXdVoV3jWIbQRAihjAJ4nN5wNxnTIGihBJ5dUmMXsJBni2gCeLJJ6
> J6zyMZ6iy4mmp63cgky5CG4=
> =jAoC
> -----END PGP SIGNATURE-----

[Uwe Hermann]

On Mon, Oct 23, 2006 at 02:49:11PM -0400, Tom Sylla wrote:
> Oops, one other thing, the OLPC platforms have no Super I/O. (they
> have an EC, but it does not get programmed in the "LB Super I/O way")

Fixed, thanks!

Uwe.

[Uwe Hermann]

Hi all,

there are a few boards left which don't yet have a status entryhere:
http://www.linuxbios.org/Supported_Motherboards

Advantech 	 SOM GX DB533-C
Agami 	Aruma
AMD 	Quartet
AMD 	Rumba
AMD 	Serenade
Broadcom 	 Blast
Broadcom BCM5780
Dell 	S1850
DigitalLogic 	 smartModule855
DigitalLogic 	msm586seg
DigitalLogic 	msm800sev
EagleLion 	5BCM
Intel 	 Jarrell
Intel 	XE7501devkit
IWILL 	 DK8S2
IWILL 	DK8X
Lippert 	Cool Frontrunner
Supermicro 	 X6DAi-G
Supermicro 	X6DHE-G
Supermicro 	X6DHE-G2
Supermicro 	X6DHR-iG
Supermicro 	X6DHR-iG2
Technologic 	TS5300
Totalimpact 	Briq
VIA 	 EPIA
VIA 	EPIA-M
VIA 	EPIA-MII

What's the status of those? Do they all work?

Thanks, Uwe.

[ron minnich]

On 10/25/06, Uwe Hermann  wrote:
> Dell    S1850

red

DigitalLogic     smartModule855

red

DigitalLogic    msm586seg

yellow

DigitalLogic    msm800sev

red
> Lippert         Cool Frontrunner

red

btw, the web page is absolutely fantastic. I love it.

ron

[Uwe Hermann]

Hi,

On Wed, Oct 25, 2006 at 01:11:12PM -0600, ron minnich wrote:
> > Dell    S1850
>
> red
>
> DigitalLogic     smartModule855
> red
>
> DigitalLogic    msm586seg
> yellow
>
> DigitalLogic    msm800sev
> red
> > Lippert         Cool Frontrunner
>
> red

For the yellow one, which are already working, which not? Are the red
ones still being worked on (or will be at some time), or are they halted?
> btw, the web page is absolutely fantastic. I love it.

Thanks! I'm glad you like it :)

Uwe.

[Daniel Parthey]

Uwe Hermann wrote:
> VIA 	 EPIA
> VIA 	EPIA-M
> VIA 	EPIA-MII

Please add
VIA EPIA-ML  - Status: yellow

Works with EPIA-M target plus a few hacks (disable firewire/cf-slot,
because of they lack on the board), somebody will have to write clean
modifications and maybe create a configuration or subtree for the ML.

VGA works with an old binary
VGA BIOS Image 1.13 available at
http://www.via.com.tw/download/mainboards/1/0/i0100113.bin

I'm not the maintainer, I just did some booting time measurements
using LinuxBIOS on the EPIA-ML6000EA while working on a project.
http://archiv.tu-chemnitz.de/pub/2006/0066/data/studienarbeit.pdf

Just for your information:

In March 2007 the Chemnitzer Linux-Tage (the biggest linux fair in the
eastern part of germany) will take place as every year at the University
of Technology, Chemnitz. Maybe you could meet some developers there to
help improving LinuxBIOS. Register for free if you would like to give a
lecture or present a live demo, to increase the popularity of LinuxBIOS.

Register for a live demo here:
http://chemnitzer.linux-tage.de/2007/live/call_form.html
Register for a lecture or a 3-hour workshop here:
http://chemnitzer.linux-tage.de/2007/vortraege/call_form.html

Daniel

[Uwe Hermann]

Hi,

On Thu, Oct 26, 2006 at 11:12:25AM +0200, Daniel Parthey wrote:
> Please add
> VIA EPIA-ML  - Status: yellow

Added, thanks. I also linked to your post for more information.

Uwe.

[Segher Boessenkool]

> Broadcom BCM5780

That's not a board but an I/O bridge (also known as HT2000).
The "Blast" board is the development board for it.

Segher

[Uwe Hermann]

On Thu, Oct 26, 2006 at 02:42:31PM +0200, Segher Boessenkool wrote:
> > Broadcom BCM5780
>
> That's not a board but an I/O bridge (also known as HT2000).
> The "Blast" board is the development board for it.

Oops. Yeah, that's a copy+paste error in my mail, the content in the
wiki is correct.

Uwe.

[Uwe Hermann]

A few more boards are left... any hints?

On Wed, Oct 25, 2006 at 07:50:32PM +0200, Uwe Hermann wrote:
> there are a few boards left which don't yet have a status entryhere:
> http://www.linuxbios.org/Supported_Motherboards
>
> Advantech 	 SOM GX DB533-C
> Agami 	Aruma
> AMD 	Quartet
> AMD 	Rumba
> AMD 	Serenade
> EagleLion 	5BCM
> Intel 	 Jarrell
> Intel 	XE7501devkit
> IWILL 	 DK8S2
> IWILL 	DK8X
> Supermicro 	 X6DAi-G
> Supermicro 	X6DHE-G
> Supermicro 	X6DHE-G2
> Supermicro 	X6DHR-iG
> Supermicro 	X6DHR-iG2
> Technologic 	TS5300
> Totalimpact 	Briq

Uwe.

[Richard Smith]

Uwe Hermann wrote:
> A few more boards are left... any hints?
>
> On Wed, Oct 25, 2006 at 07:50:32PM +0200, Uwe Hermann wrote:
> > there are a few boards left which don't yet have a status entry?here:
> > http://www.linuxbios.org/Supported_Motherboards
> >
> > Advantech 	 SOM GX DB533-C

You can essentially drop this one.  I started on it to work on the OLPC 
(is a geode) but found out it had a cs5535 rather than a cs5536.   It 
would be marked RED.

[ron minnich]

If we're dropping it let's drop it. All these red lines give a misleading
impression.

ron

[Uwe Hermann]

On Tue, Oct 31, 2006 at 10:14:39AM -0700, ron minnich wrote:
> If we're dropping it let's drop it. All these red lines give a misleading
> impression.

I agree _if_ the board never worked at all and there's no chance that
someone will work on it over the next 1-2 years or so.

But I would _not_ remove it (from svn) if it already worked at some
point (or parts of it).

Currently the wiki lists everything that has a directory in
src/mainboard. Maybe we should list only those boards which actually
work, at least partially...

Uwe.

[Greg Lindahl]

Dumb question: IBM is currently shipping the e326m, not the e326, is
this supported?

I don't know how different it is. I don't have one, we're x-series
bigots :-)

-- greg

[ron minnich]

On 10/31/06, Greg Lindahl  wrote:
> Dumb question: IBM is currently shipping the e326m, not the e326, is
> this supported?

hard to say without testing it :-(

get ibm to send us one :-)

ron

[LinuxBIOS] Supported motherboards: List all boards in the tree?

[Stefan Reinauer]

* Uwe Hermann  [061031 19:22]:
> On Tue, Oct 31, 2006 at 10:14:39AM -0700, ron minnich wrote:
> > If we're dropping it let's drop it. All these red lines give a misleading
> > impression.
>
> I agree _if_ the board never worked at all and there's no chance that
> someone will work on it over the next 1-2 years or so.
>
> But I would _not_ remove it (from svn) if it already worked at some
> point (or parts of it).

It did not. And it's still in the revision history.
> Currently the wiki lists everything that has a directory in
> src/mainboard. Maybe we should list only those boards which actually
> work, at least partially...

abuild tests for $LBROOT/src/mainboard/${VENDOR}/${MAINBOARD}/BROKEN

and ignores boards with that line

Maybe it should read 
$LBROOT/src/mainboard/${VENDOR}/${MAINBOARD}/UNSUPPORTED

and be a basis whether this should appear in the web list or not.

Also,...

would it make sense to generate the web page from the repository,
by adding a boardinfo.xml file in each board subdirectory with the
information?

This information could be used to create a multitude of pages,.. for
example the qa.linuxbios.org things as well...

Stefan

[Stefan Reinauer]

* Richard Smith  [061031 17:54]:
> You can essentially drop this one.  I started on it to work on the OLPC 
> (is a geode) but found out it had a cs5535 rather than a cs5536.   It 
> would be marked RED.

I dropped it on your behalf.

Should we delete the cs5535 component as well?

[Richard Smith]

Stefan Reinauer wrote:
> Richard Smith  [061031 17:54]:
> > You can essentially drop this one.  I started on it to work on the OLPC 
> > (is a geode) but found out it had a cs5535 rather than a cs5536.   It 
> > would be marked RED.
>
> I dropped it on your behalf.

Thanks.
> Should we delete the cs5535 component as well?

If it was the only motherboard to use it then yeah.

[Stefan Reinauer]

* Richard Smith  [061104 17:46]:
> Stefan Reinauer wrote:
> > Richard Smith  [061031 17:54]:
> > > You can essentially drop this one.  I started on it to work on the OLPC 
> > > (is a geode) but found out it had a cs5535 rather than a cs5536.   It 
> > > would be marked RED.
> >
> > I dropped it on your behalf.
>
> Thanks.
> > Should we delete the cs5535 component as well?
>
> If it was the only motherboard to use it then yeah.

I think the GX1 code does not do any memory sizing, does it?
Though the eaglelion/5bcm is using it

lippert/frontrunner seems to be using cs5535

Stefan

[Luis Correia]

Stefan Reinauer wrote:
> Richard Smith  [061104 17:46]:
> > Stefan Reinauer wrote:
> > > Richard Smith  [061031 17:54]:
> > > > You can essentially drop this one.  I started on it to work on the OLPC 
> > > > (is a geode) but found out it had a cs5535 rather than a cs5536.   It 
> > > > would be marked RED.
> > >
> > > I dropped it on your behalf.
> >
> > Thanks.
> > > Should we delete the cs5535 component as well?
> >
> > If it was the only motherboard to use it then yeah.
>
> I think the GX1 code does not do any memory sizing, does it?
> Though the eaglelion/5bcm is using it

The board i'm working on (IEI Nova 4899R) does detect properly
the memory I'm using (124Mb, 128Mb total with 4Mb reserved for the VGA).
> lippert/frontrunner seems to be using cs5535

But it uses the cs5530 :)
> Stefan

I'm going to need some assistance with this board, but
i'll start a new thread about it soon.

Luis Correia