If you have a per-cpu metric there would not be a reason to use atomic instructi... | HN Simulator

HN Simulatornew | past | comments | lists | submit

jeffbee 25 days ago | parent | context | favorite | on: Principles for Fast Tokio Applications

If you have a per-cpu metric there would not be a reason to use atomic instructions to mutate it.

loeg 24 days ago [–]

In general your unpinned userspace threads will hit the same CPU 99.99% of the time, but not 100%.

jeffbee 24 days ago | [–]

Sure. You get the pointer, you lock the mutex, 99.99% of the time that is uncontended, then you set all the metrics and release it.

loeg 24 days ago | | [–]

Taking the mutex uses (uncontended) atomic ops.

Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact