[coreboot] Issues with Supermicro H8SCM

[Thomas Gstädtner]

On Wed, Oct 5, 2011 at 20:08, Marc Jones <marcj303 at gmail.com> wrote:
> On Tue, Oct 4, 2011 at 5:26 PM, Thomas Gstädtner <thomas at gstaedtner.net> wrote:
>> Hi everyone,
>>
>> I own a Supermicro H8SCM-F-O mainboard (AMDs SP5100 Chipset), one of
>> the boards AMD as of recently claims to officially support.
>> So I decided to try coreboot, but (as expected) it is not really an
>> out-of-the-box experience.
>>
>> First of all, flashrom does not support flashing out-of-the-box, but
>> Frederic Temporelli's patches (
>> http://patchwork.coreboot.org/patch/3326/ and
>> http://patchwork.coreboot.org/patch/3326/ ) work fine for me (I tested
>> multiple reads and multiple writes, always worked), I can report that
>> to the flashrom list separately if you want.
>>
>> I built coreboot from the gerrit-branch (also tried a "stable"
>> revision), coreboot-4.0-1714-gf8adf7a, using the "default"
>> board-config incl. seabios (stable) as payload.
>> After powering on nothing happens (apart from the fans running and the
>> power led flashing) and it stays that way, it is silent on VGA as well
>> as serial.
>> After manually triggering a reset I immediately get the following
>> output on serial:
>>
>> coreboot-4.0-1714-gf8adf7a Wed Oct  5 00:31:05 CEST 2011 starting...
>> BSP Family_Model:
>>
>> Here it hangs, and I don't quite know how to move on and debug the issue.
>>
>> Can you give me any hints where to go on from here? I have external
>> flashing equipment at hand (which also works fine btw.), but no fancy
>> bios-debugging hardware.
>>
>> Regards,
>>
>> thomasg
>>
>> P.S. In case anyone else stumbles upon it: the board has 2 separate
>> flash chips, one Macronix MX25L12845E with 128 Mbit (the SOIC16W
>> package on the left of the board) as well as a SST25VF016B, 16 Mbit,
>> as the SOIC8W on the right. The first one holds the IPMI firmware, the
>> second one the AMI BIOS.
>> Flashing requires a testclip or some really serious soldering skills.
>> Attaching a SO8 socket can only be done with hot air soldering.
>
> Hi Thomas,
>
> Thanks for the report. While many people overlap with the flashrom
> list, please post about those patches on that list. Seems that you
> should be able to ack them.
>
>  It looks like maybe the platform build doesn't support your CPU.
> There have been a number of recent contributions from AMD, but they
> don't all align with the platform contributions. So, the H8SCM should
> support all fam10 revD, but may have issue with Fam10 rev E. To get
> more information, you can turn the console debug output to SPEW in
> menuconfig.
>
> Marc
>
>
>
> --
> http://se-eng.com
>

Thanks for the advice Marc,

I already had the loglevel set to spew (8). I additionally tried the
other "Verbose output" debugging options, unfortunately I don't get
any output at all when they are built in.
Also, while it is hard to find and AMD doesn't have any official
document confirming this, according to some google hits the processor
should be revision D1 not rev E (AMD model number: OS4170OFU6DGOWOF ).

(Sry for sending the mail twice, hit the wrong button)