Wifi / let's have RF stuff powered with high current source all the time because...

Andy Green andy at openmoko.com
Tue Aug 26 20:19:46 CEST 2008


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Somebody in the thread at some point said:
| On 8/26/08, Joerg Reisenweber <joerg at openmoko.org> wrote:
|> Am Di  26. August 2008 schrieb Uwe Klein:
|>
|>> Hi,
|>  > in this context:
|>  >
|>  > is there any (hardware) watchdog stuff that limits transmit time and
|>  > similar things?
|>
|>
|> There's a watchdog in CPU, but it's unused by all of the recent sw-stacks
|>  AFAIK.
| brilliant.
|
| It is a good idea to have a hardware watchdog pulling the plug on airwaves
| and other resource  hoggers. This is imho more important in a less
controlled
| sw stack like openmoko and similar.

Unfortunately whatever creates this situation with the high current
business does not take down the CPU, so the CPU watchdog would not
trigger.  What might do it is a watchdog in the GSM side firmware that
required to hear from the CPU periodically that it was indeed meant to
still be on a call, otherwise it would choose to drop it.  Dunno if it
exists.

Depending on what causes it, periodic poll of GSM stuff from a daemon on
CPU side confirming it can still talk to GSM side and it is down when it
is meant to be down (or it pulls the power on GSM side) might help.

Actually there's another idea for William if he sees it again, use the
UI to "turn off GSM" see if it changes current consumption.

- -Andy
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAki0SUIACgkQOjLpvpq7dMpT7QCferNwRanC9DljdRI08h8wZnnI
5jcAniaLEsJtMOT6ZLUnfkrwASdFx4mo
=nf+9
-----END PGP SIGNATURE-----



More information about the hardware mailing list