I thought TPUs couldn't reasonably run LINPACK at all because TPUs do not acknowledge that FP64 exists.

I know Google wants to compare their stuff to El Capitan or whatever but the comparison does not seem valid to me.

Even before AI training clusters became important, Google has had an outstanding custom fabric (there's papers about it) together with the ability to tune NICs for their own cases, and "their own cases" meant nearly everything engineered within Google. Ethernet hardware has had low kernel latency and DMA for a long time; it's the rest of the stack that hurts. But as far back as the early 2010s (if not further back, that goes beyond my knowledge horizon), you could just make it not hurt, if you had the software engineers to do it.