[blibble]

> The CPU will randomly speed up or slow down

constant_tsc has been a thing for more than a decade

> execute instructions out of order, and even speculatively execute

rdtscp serialises

> Not to mention that it'll have a dozen different clocks - which are not guaranteed to be in sync.

synchronised_tsc has been a think for about 6 years now

[lazide]

None of these are performant, no?

Generally, you can have consistency, speed, or low cost - but not more than two at the same time.

[kprotty]

Invariant (Constant) TSC is detectable via `cpuid` and applies to `rdtsc/rdtscp` by default. In that aspect, there's no tradeoff being made there (observable to software) AFAICK.

[lazide]

interesting results! i guess it depends in if you consider 35ish cycles expensive or not.

[https://community.intel.com/t5/Software-Tuning-Performance/H...]

[kprotty]

Are there cheaper ways of getting elapsed time with sub microsecond precision? Interested as I've only ever heard of rdtsc at the lowest level in userspace for x86.

[lazide]

I ran across a random stackoverload thread with benchmarks claiming it was about 2x the cost of doing a naive gettime(). But frankly, hard to figure out when you factor in all the various caches, OO execution pipelines, etc,