This has varied over time depending on kernel implementations, CPU core count, number of threads, etc. Some relevant reading: https://matklad.github.io/2020/01

This has varied over time depending on kernel implementations, CPU core count, number of threads, etc. Some relevant reading: https://matklad.github.io/2020/01/04/mutexes-are-faster-than-spinlocks.html https://matklad.github.io/2020/01/04/mutexes-are-faster-than...