Why does the @golang garbage collector pause all threads to scan the stack? Why not pin stack objects and go 100% nonblocking?
If callers were responsible for pinning parameters, then they could do a simple lifetime analysis and avoid most duplicate pinning.
-
-
My timing indicates a throughput of ~15 cycles per lock [inc] for uncontended cache lines on Haswell, similar to an L1 cache miss.
-
Uncontended LOCK aren't bad themselves, but they also stall the core. No ILP/MLP for you.
- 1 more reply
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.