Hi all,
After hours spent yesterday, some Chipquik, one heatgun pass, and a dead board scare later, I managed to solder a PLCC socket on my TUSI-M board.
SIS630ET is still on my radar, but I'm going to need some help.
In the spirit of the "coreboot challenge" below, let me propose another one: Port coreboot v4 to SIS630ET, then to Asus TUSI-M. This is a challenge because most if not all of the SIS chipsets have no public documentation, and so far from reversing my factory BIOS I don't recognize any of the registers the TUSI-M factory BIOS programs, after reading what's available from coreboot v1. To be legally safe, I would like one person/team to reverse the factory BIOS of a number of 630ET boards, then produce a documentation. Then another person/team (can be myself) do the port to that documentation.
And one interesting bit: I noticed a special check for "CentaurHauls" CPUID in the factory BIOS, which also uses CAR.
Cheers Keith
On Tue, Apr 27, 2010 at 10:17 AM, Joseph Smith joe@settoplinux.org wrote:
On Tue, 27 Apr 2010 10:01:28 -0400, Keith Hui buurin@gmail.com wrote:
Hello, I want to propose a "coreboot challenge" to first port coreboot to the i840 chpset, second port coreboot to support a very rare i840 dual cpu board, third instead of RAMBUS (which i840 is designed for) this board uses SDRAM. I think this would be a very interesting port for coreboot.
The board is a AMI Megatrends MegaDual. The user manual can be found here: http://www.ami.com/support/doc/MegaDual_mb_man.PDF
I have four of these rare boards and am willing to donate three of them (one person each, less shipping costs of course) to anyone who is willing to take on this challenge. Let me know if your interested.
Sounds like some sort of reference board for workstations/servers.
i840+SDRAM means this board has one or two MTHs. This is going to be
epic.
Yes it should be interesting :-)
(No I won't be able to attempt it - I still have 440BX and SIS630ET in my pipeline.)
That's OK Keith. I actually have allot of other i8xx boards that are a higher priority than this one. I knew I was not going to be able to get to it for a while that is why I am offering up my 3 spare boards for free so others could hack along if they want.
I am actually suprised no one has jumped for the oppritunity to get a free motherboard.
I do have a few SIS 630's kicking around so I will be looking forward to your work on that :-)
-- Thanks, Joseph Smith Set-Top-Linux www.settoplinux.org
Am 24.05.2010 20:38, schrieb Keith Hui:
In the spirit of the "coreboot challenge" below, let me propose another one: Port coreboot v4 to SIS630ET, then to Asus TUSI-M. This is a challenge because most if not all of the SIS chipsets have no public documentation, and so far from reversing my factory BIOS I don't recognize any of the registers the TUSI-M factory BIOS programs, after reading what's available from coreboot v1. To be legally safe, I would like one person/team to reverse the factory BIOS of a number of 630ET boards, then produce a documentation. Then another person/team (can be myself) do the port to that documentation.
Are you aware of SerialICE? It might give you the necessary information without disassembling other people's code. With only working from SerialICE logs, you should be safe, as they only record what happens (but I'm not a lawyer, etc)
As an additional feature, once you got serialice to work for your board (which requires serial enable, nothing else), you can run vendor bios or coreboot at wish, without touching the flash again (as the bios is loaded in the qemu host that controls the target)
This still requires you to make sense of what happens, and in some ways, it's harder as you only see hardware accesses, not logic decisions (that is, lots of "inb 0x40 == 0x80", followed by a single "inb 0x40 == 0x81", instead of "while (inb(0x40)&&1) udelay(1);"), but it's questionable how much knowledge you _really_ get out of disassembly, too.
Regards, Patrick
On 5/24/10 8:38 PM, Keith Hui wrote:
To be legally safe, I would like one person/team to reverse the factory BIOS of a number of 630ET boards, then produce a documentation.
It might be easier to contact SiS for documentation. Did you try?
Stefan
SiS is mostly dead nowadays. Might be hard, no?
Tiago
On Mon, May 24, 2010 at 9:04 PM, Stefan Reinauer < stefan.reinauer@coresystems.de> wrote:
On 5/24/10 8:38 PM, Keith Hui wrote:
To be legally safe, I would like one person/team to reverse the factory BIOS of a number of 630ET boards, then produce a documentation.
It might be easier to contact SiS for documentation. Did you try?
Stefan
-- coresystems GmbH • Brahmsstr. 16 • D-79104 Freiburg i. Br. Tel.: +49 761 7668825 • Fax: +49 761 7664613 Email: info@coresystems.de • http://www.coresystems.de/ Registergericht: Amtsgericht Freiburg • HRB 7656 Geschäftsführer: Stefan Reinauer • Ust-IdNr.: DE245674866
-- coreboot mailing list: coreboot@coreboot.org http://www.coreboot.org/mailman/listinfo/coreboot
On 5/31/10 2:38 AM, Tiago Marques wrote:
SiS is mostly dead nowadays. Might be hard, no?
Looks very much alive ...?
http://www.sis.com/contact/contact.htm
Stefan
Tiago
On Mon, May 24, 2010 at 9:04 PM, Stefan Reinauer <stefan.reinauer@coresystems.de mailto:stefan.reinauer@coresystems.de> wrote:
On 5/24/10 8:38 PM, Keith Hui wrote: > To be legally safe, I > would like one person/team to reverse the factory BIOS of a number of > 630ET boards, then produce a documentation. > It might be easier to contact SiS for documentation. Did you try? Stefan -- coresystems GmbH • Brahmsstr. 16 • D-79104 Freiburg i. Br. Tel.: +49 761 7668825 • Fax: +49 761 7664613 Email: info@coresystems.de <mailto:info@coresystems.de> • http://www.coresystems.de/ Registergericht: Amtsgericht Freiburg • HRB 7656 Geschäftsführer: Stefan Reinauer • Ust-IdNr.: DE245674866 -- coreboot mailing list: coreboot@coreboot.org <mailto:coreboot@coreboot.org> http://www.coreboot.org/mailman/listinfo/coreboot
Stefan Reinauer wrote:
SiS is mostly dead nowadays. Might be hard, no?
Looks very much alive ...?
They're still alive, but they are not in the chipset business anymore. I expect that all the chipset experts will have moved on to other companies by now, but sure - give it a go!
//Peter
See the attachment, doesn't look like something from a company that's alive and kicking. I know them since the SiS 730. Last time I heard they had some design win for a Core 2 chipset that I haven't seen in any product and started selling SiS branded SO-DIMMS :|
Best regards, Tiago
On Mon, May 31, 2010 at 2:10 AM, Stefan Reinauer < stefan.reinauer@coresystems.de> wrote:
On 5/31/10 2:38 AM, Tiago Marques wrote:
SiS is mostly dead nowadays. Might be hard, no?
Looks very much alive ...?
http://www.sis.com/contact/contact.htm
Stefan
Tiago
On Mon, May 24, 2010 at 9:04 PM, Stefan Reinauer < stefan.reinauer@coresystems.de> wrote:
On 5/24/10 8:38 PM, Keith Hui wrote:
To be legally safe, I would like one person/team to reverse the factory BIOS of a number of 630ET boards, then produce a documentation.
It might be easier to contact SiS for documentation. Did you try?
Stefan
-- coresystems GmbH • Brahmsstr. 16 • D-79104 Freiburg i. Br. Tel.: +49 761 7668825 • Fax: +49 761 7664613 Email: info@coresystems.de • http://www.coresystems.de/ Registergericht: Amtsgericht Freiburg • HRB 7656 Geschäftsführer: Stefan Reinauer • Ust-IdNr.: DE245674866
-- coreboot mailing list: coreboot@coreboot.org http://www.coreboot.org/mailman/listinfo/coreboot
-- coresystems GmbH • Brahmsstr. 16 • D-79104 Freiburg i. Br. Tel.: +49 761 7668825 • Fax: +49 761 7664613 Email: info@coresystems.de • http://www.coresystems.de/
Registergericht: Amtsgericht Freiburg • HRB 7656 Geschäftsführer: Stefan Reinauer • Ust-IdNr.: DE245674866
Tiago Marques wrote:
doesn't look like something from a company that's alive and kicking.
They were doing fine last year, but no longer have focus on the chipset market.
//Peter
On 6/5/10 5:15 AM, Tiago Marques wrote:
See the attachment, doesn't look like something from a company that's alive and kicking. I know them since the SiS 730. Last time I heard they had some design win for a Core 2 chipset that I haven't seen in any product and started selling SiS branded SO-DIMMS :|
So you did contact them for data sheets?
Me? I wasn't after that.
Best regards
On Sat, Jun 5, 2010 at 9:32 AM, Stefan Reinauer < stefan.reinauer@coresystems.de> wrote:
On 6/5/10 5:15 AM, Tiago Marques wrote:
See the attachment, doesn't look like something from a company that's alive and kicking. I know them since the SiS 730. Last time I heard they had some design win for a Core 2 chipset that I haven't seen in any product and started selling SiS branded SO-DIMMS :|
So you did contact them for data sheets?
-- coreboot mailing list: coreboot@coreboot.org http://www.coreboot.org/mailman/listinfo/coreboot
Hi,
http://dev.gentoo.org/~dberkholz/articles/specifications/sis/
There is a docu for sis630 maybe it is enough for you?
Thanks, Rudolf
This looks more like it. Thanks!
But pages 88-120 was corrupted... There goes coverage for much of the RAM init registers.
Keith
On Tue, Jun 1, 2010 at 7:45 AM, Rudolf Marek r.marek@assembler.cz wrote:
Hi,
http://dev.gentoo.org/~dberkholz/articles/specifications/sis/
There is a docu for sis630 maybe it is enough for you?
Thanks, Rudolf