I thought that the mark-sweep isn't even necessary, if you don't have cycles. Reference counting can be a sufficient GC strategy but when cycles are introduced,

I thought that the mark-sweep isn't even necessary, if you don't have cycles. Reference counting can be a sufficient GC strategy but when cycles are introduced, I think that's the problem that mark-sweep solves. That's awesome. Which papers do you find the most useful?