[rossmohax]

What if you add PGO to the mix, how fast it is going to be?

[gregod]

Weirdly enough, PGO equalizes and increases the runtime, where both methods take approximately 330 us. (I simply applied PGO to the full benchmark harness, no idea if this is proper). Them being equal sounds more reasonable, but the increase in runtime doesn't feel right.

Micro benchmarking remains a fickle beast :)

[renox]

I remember the talk where two different users measured different performance due to the username length..