PWM traps all signals, even those normally ignored about rpio

[festlv]

I helped to debug a similar issue for a friend and arrived at similar conclusion.

I'm not that deep into writing C modules for python, but there has got to be a way to register C function which is being called when python interpreter quits.

Otherwise this behaviour at least has to be documented and workaround (see @mlv's comment on #19) should be provided to ignore legitimate signals- WINCH is just one of the signals whose default action is not to terminate.

[maroc81]

I've also been bitten by this issue. My python application would randomly exit with no output or any indication why. I was finally able to consistently reproduce and it and traced it down to this issue.

My application has a number of threads so SIGCHILD would get sent by the children threads whenever they were interrupted. This seems to be another signal that by default should be ignored.

I'm really surprised that the author didn't catch this or even did it in the first place. The rest of his code works great. I wouldn't be able to do my project on the pi without since the pi only has one pwm.