Even just to have a database of "gotchas" would be nice. Not necessarily bugs but places you stub your toe before realizing that the device does not function like you thought it would.

My most recent one was that if you stop the timers (CCU) in the LPC chip and the interrupts are still enabled for it then you have a 1 in 65536 change of stopping the timer at the actual OC compare match value. This causes LOTS of interrupts ;-)

Details are a little fuzzy but I can check my notes if anyone needs more info.

GB