Dynamic Code Optimization and the Nvidia Denver Processor [2015]

With the earlier discussion on VLIW archs[0], I was reminded of a great, frank talk about Denver's internal architecture.

I don't agree with everything he says (particularly that it wouldn't make any sense to expose the underlying VLIW ISA, when JITs are so prevalent these days for code that gets run on Android phones), but it was really great to see a lot more on both the design of the processors, and the thought processes that led to that design.

[0] - https://news.ycombinator.com/item?id=24525372