In those days it was called Direct Data Intervention.

It’s still cool!

And, like all aspects of cache design, it takes massive brainpower to figure out what you want to do, and then a trivial amount of logic to actually implement it!

Thanks for a great review of the architecture.

It is certainly one of the “logical” successors–to a certain extent the matrix math units on the Power10 are too. But it may not be a literal successor.