Hi! I saw some mails on the community list (which I don't subscribe to) discussing setting a threshold for accelerometer [1] data. I just wanted to say that I wrote a patch just recently that does this (and from some quick tests seem to work).
The "problem" is that it's based on my earlier patch which used a workqueue for the interrupt handling. It was a lot easier to start hacking on it that way since it allows just reinserting the module instead of rebooting the freerunner if you change mach-gta02.c with the current approach. The modification needed is basically to read the HP_FILTER_RESET register on interrupts to reset the base from which the acceleration should be measured. Incidently, I also found the debug patch I posted earlier very helpful when trying this out - with that I could "implement" and test the functionality in a shell script before changing the C code. :-) // Simon [1] Hey, I'm starting to learn to spell that correct (I think!)
