What is the purpose of itineraries and stages in tablegen?

As far as I understand itineraries instruction stages and instruction itinerary data classes are used to describe how instruction is moving through the pipeline. The LLVM for OR1K (https://github.com/openrisc/llvm-or1k) does not have these kind of descriptions. What am I missing? How come it doesn’t have it? Is there another way to provide this information?

Any help is appreciated.