It’s important to note that most of the things mentioned here are just “tricks”: they’re fun to discover and talk about, but they really only end up being minor wins in practice. TSO is great…if you are trying to make a simpler Rosetta (it’s not even necessary on the M1, although I think Apple is still using it for convenience; I’m still trying to find out where). The JavaScript instruction speeds up…one specific rounding in JavaScript. Fast atomics help…if you’re that one specific part of Apple’s runtime that has taken advantage of it yet. They are all cool, important little optimizations, and I’m sure that few other companies could do what Apple is doing in this space.
However, it’s very much not true that Apple is “cheating” their way to performance with a bunch of these tricks for their own software (as much as you can call this cheating). Their processors are optimized for general-purpose workloads, and blow everything else out of the water even if you’re doing something profoundly “un-specialized” like run Photoshop or a Linux VM.
It’s easy to come up with takes like these that point at buzzwords that you’ve seen being thrown around in the last couple weeks in an attempt to explain this chips. The reality is that the reasons these chips are fast are either unknown or boring. I suspect that these will one out as we play around with them more, but we don’t have the details right now.