Use maxBound for threadDelay

[srid]

No description provided.

[Ericson2314]

http://hackage.haskell.org/package/fsnotify-0.3.0.1/docs/System-FSNotify.html#v:watchTree It looks like we shouldn't forkIO at all, but instead just be careful to call the action to kill the thread created by watchTree if we ourselves are killed.

[3noch]

performEventAsync still runs in the same thread; it's up to the Performable m action to forkIO. Although I'm not sure what the implications are if you don't.

[Ericson2314]

@3noch I think watchTree doing its own forkIO. performEventAsync is basically performEvent + triggerEvent. It's not a problem if you don't fork IO and even if for some reason watchTree wasn't forkIO-ing (e.g. it was communicating with a preexisting thread) it should be fine.