matt_hsu wrote:
Just be curious, why you need to turn off Ar6k completely?
Matt
Because being waterboarded in Gitmo doesn't sound like fun.
Is there a spec that *guarantees* that the ar6k will _not_ transmit
under any circumstances when it is suspended?
BTW, as a general observation, this concern is not limited to the wifi
stuff; there has been no real attention paid to airplane mode with
either the GTA01 or GTA02. Specifically, we need someone to catalog the
hardware behavior required of the GSM, bluetooth, wifi, and GPS chipsets
during airplane mode (and any other modes as required; e.g. blast
zones). Once we know that, we can then begin to understand which
firmware images comply, and what user actions are required to ensure
compliance.
At the moment, the only way to absolutely ensure compliance in any area
where RF emissions are restricted is to remove the battery[1]!
Mike (mwester)
[1] Some firmwares (FSO at least) will power off the GSM at shutdown,
but its noteworthy that on the GTA01 there is no possible way to power
off the GSM if serial communications with the GSM is lost by a
user-space process. I am led to understand that the GTA02 offers a GPIO
that permits the kernel to forcibly remove power from the GSM, but
nobody has confirmed that this action is controllable from userspace, or
if the kernel actually performs this power removal upon normal shutdown
-- so at this point I think the safe thing to do is consider the GTA01
and GTA02 to both be entirely reliant upon user-space code to power off
the GSM. I don't trust user-space code - especially the fragile stuff
that is currently managing the GSM, so it's pretty clear that right now
airplane mode is battery removal.
___
hardware mailing list
hardware@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/hardware
