Wifi / off-state leakage

[Werner Almesberger]

Andy Green wrote:
> Have a look at the circuit and give some examples of signals that can be
> in this condition in "off" state, then we can talk about it.  Otherwise
> it's just handwaving.

I don't have any specific problem I suspect to happen. I just wanted
to point out the general problem (which seems particularly bad with
those Samsung SoCs), which seems to be important since Joerg recently
suggested to run all sorts of GPIOs to subsystems that may be powered
when the CPU isn't.

> AFAICT only the GSM PA has the situation it is powered outside PMU and
> discrete MOSFET switch, so he is the only guy that can leak "from
> outside".

Yes, TX_ENABLE is the potential weak link. As long as we Hi-Z all
modem lines and the infamout IRDA interface, I guess we should be
safe. (We may be in PMU.ACTIVE in "fly mode", so we can't count on
just the IO_3V3 rail being down anyway.)

If I'm in a situation where I absolutely want to make sure we don't
emit any power, I'd remove the battery anyway.

- Werner