[coreboot] unknown board - via chipset

Antoine Martin antoine at nagafix.co.uk
Fri Apr 11 17:47:19 CEST 2008

Hi Carl-Daniel,

I've taken the board out (wasn't actually hard at all - it is not glued)
And here is what I found:
* Via VT1612A
* Via VT1211
* Via VT6103
* ram chips: HY5DU121622CTP
* SST39SF020A
which is soldered onto the board :(

Someone with superhuman soldering skills might be able to fit a socket 
in its place... ain't me.
(board is dual layered, so you would have to re-use those tiny pins...)
If anyone can, I'll give them a unit for free - just email me.


Carl-Daniel Hailfinger wrote:
> Hi Antoine,
> I'll defer most of your questions to the real VIA experts on this list.
> On 11.04.2008 01:16, Antoine Martin wrote:
>> Carl-Daniel Hailfinger wrote:
>> | On 10.04.2008 18:23, Antoine Martin wrote:
>> |> I've got this board I would like to get coreboot flashed on:
>> |> *eBox-3850
>> |> **http://www.icoptech.com/ebox-pc/*
>> |>
>> |> 1) CPU is a via C3 Nemehia
>> |> Northbridge is a VT8623
>> |> Southbridge is a VT82xxxxx?
> To have any reasonable chance of checking whether coreboot supports your
> chipset, we need to know exactly which chipset it is (that includes the
> exact sub-revision). Without that, you'll have virtually zero chance of
> getting the machine to reach early init at all, and even less so if you
> want it to boot a payload.
>> |> [...]
>> |> The box cannot be opened easily as the cpu is glued on to the case...
>> |> Which is why I would rather make sure it is possible to flash before I
>> |> start damaging the box.
>> |>
>> |
>> | Please do not try to flash any BIOS or firmware on a board you can't
>> | access physically. That also applies if you can't open the box.
>> | Flashing can go wrong and if you're porting coreboot to a new board, the
>> | first image you flash is very likely not to boot far enough to reflash
>> | the BIOS chip, so you have to open the case and remove the flash chip
>> | anyway. If that is not possible, the machine is bricked.
>> |
>> If there is a reasonable chance that I can make it boot, I might take
>> the risk of bricking one unit or having to find a creative way of
>> opening and putting it back together (and that's assuming that the bios
>> chip can be swapped out..)
>> I won't blame anyone for it but myself if this goes wrong - and it often
>> does...
> Sorry, without exact chip revisions, there is no reasonable chance to
> make it boot.
>> So does it look like it might work from those specs?
>> A lot of the via boards seem to be very similar, does the hardware and
>> payload vary so much that this would be pointless?
> To give you a perspective on the difficulty of bringing up a board: Even
> if chipset and processor are perfectly identical to another board, you
> will have to go through a few iterations until it works (IRQ routing,
> GPIO configuration and a few other things usually differ). For your
> board, we don't even know the exact chipset, so the only way I see is to
> open the box and write down the numbers on the chips. And please find
> out whether flash is soldered (will need to be changed) or socketed (good).
> Regards,
> Carl-Daniel

More information about the coreboot mailing list