Inefficient memory order in SpinLock
Open
- Dominant language
- C++
- Stars
- 30.5k
- Forks
- 5.9k
- PR merge metrics
- No merged PRs in 30d
Description
Probably, `std::memory_order_acquire` was intended [here](https://github.com/facebook/folly/blob/b028517fc58f3cec70905f05504e39e270a4457b/folly/synchronization/MicroSpinLock.h#L103), in `folly/folly/synchronization/MicroSpinLock.h`, as `xchg` is only used when obtaining the lock.
Contributor guide
Assessment
This issue has not been assessed yet.