[Arm-netbook] seatron/si-chang so-dimm, hardware bugs in netbook

Luke Kenneth Casson Leighton luke.leighton at googlemail.com
Mon Mar 29 14:02:12 BST 2010


On Mon, Mar 29, 2010 at 10:30 AM, Ben Dooks <ben at fluff.org> wrote:
>> because we've found that designers, Si-Chang (also known as Seatron)
>> _also_ can provide the exact same hardware, and adam is going to talk
>> to them later this week.
>
> Do they sell the DIMM module seperately,

don't know.  question for adam to be asking, methinks.

> and if so, do they provide a
> pinout if other people want to use it?

 chitech wouldn't answer that, but si-chang might.

 si-chang is a R&D company.  they're involved in a "cooperative
venture" i.e. what that means is that someone in the chinese
government is putting up whopping great sums of money to fund the next
generation of low-cost computing.  they're funding the OS development
(MOS) and they're funding the netbook and mid development.

 so, si-chang have both an S3C6410 ARM SO-DIMM and an SIRF Prima IV
ARM SO-DIMM.  from what chitech told us, i think it's safe to say that
the two SO-DIMMs are pin-compatible.

> I'd be careful, the DM9000 design guides say something huge (iirc, 100uF)
> for the regulator out decoupling, but this may be overkill for the worst
> case. It may be there's some decoupling nearer the magnetics than the
> DM9000. There could be other layout issues involved with the ethernet.

 understood.  well, what we know is that a) when the ethernet comes
up, it crashes (triggering the timeout) b) following a reset, it's
fine.

> WRT to the watchdog, I belive the watchdog should generate an internal
> reset to reboot the device, so it would be interesting to find out why
> it isn't working.

 the working hypothesis is that it isn't working because they haven't
connected the internal reset line to anything!!

 ben, could you confirm for me what actually happens?  watchdog
generates an internal reset, then that pulls a hardware line, is that
right?

 looking at e.g. the S5PC100 datasheet, section 7.4, says that
XnRSTOUT is asserted.  could it be the case that simply the netbook's
design just.... ignores that hardware line?

 l.



More information about the Arm-netbook mailing list