Skip to contentSkip to conversation

Powered state in BlueZ 5

DISCUSSION Emilio Pozuelo Monfort 1 participant 1 message May 27, 2013 Bluetooth drivers

lore
Share & tools
lore
#0 lore raw

Hi,

I'm porting the GNOME stack to BlueZ 5 and I'm hitting one issue. It looks like the Powered state is not persistent anymore:

commit 0f575386a54382b63a1aa8e90589917747ee1a79

Author: Johan Hedberg <[email protected]>
Date:   Thu Dec 13 16:35:30 2012 +0200

core: Remove persistent storing of the powered state

commit 0bd31205e667de7a809a683e5c42c58a8d547a6e

Author: Marcel Holtmann <[email protected]>
Date:   Sat Dec 29 16:28:03 2012 -0800

doc: Mention that the adapter powered property is not persistent

The problem with this is that if the adapter is switched off with rfkill (that's what the toggles to turn bluetooth on and off do), then when it's switched back on it won't be powered. Should those toggle buttons also take care of powering the adapter? What if the user uses a hardware switch to turn it off? Or is there another way this should be done in BlueZ 5? I couldn't find any documentation about this in [1], but [2] suggests this needs to be handled elsewhere, such as in NM.

Thanks,
Emilio

[1] http://www.bluez.org/bluez-5-api-introduction-and-porting-guide/
[2] http://article.gmane.org/gmane.linux.bluez.kernel/35532