![]() |
2010-04-07
, 06:49
|
Posts: 1 |
Thanked: 0 times |
Joined on Apr 2010
|
#2
|
![]() |
2010-04-07
, 07:22
|
Posts: 999 |
Thanked: 1,117 times |
Joined on Dec 2009
@ earth?
|
#3
|
I've read through http://talk.maemo.org/showthread.php?t=21182 and tried it, which doesn't work. I've also used dbus-monitor, and no events are generated when pressing the button.
Digging a bit deeper, watching /dev/input/event4 using evtest also showed nothing, however, when a call was made an event was indeed registered by evtest. Rather interesting.
It seems as if the phone somehow enables the headset button. Looking through /sys/devices/platform/nokia-av and nokia-av.c it appears that "madc" holds some sort of voltage value. Anyhow, when the button is not "magically activated" by the phone app, this value hovers around 2 - 4. It stays like this regardless of music playing or button pressing. When a call is made however, this value jumps to ~800 or so, and pressing down the button drops it to ~22.
Kernel hacking isn't exactly my strength, so the solution to activate the button is probably sitting there in nokia-av.c glaring at me. I've also tried strace'ing a few processes, but none of them seem to reveal anything interesting...