I guess this is because in a multicore scenario, atomic operations will involve inter core transactions which are expensive under virtualization (hypervisor interference).
On Fri, Nov 14, 2025, 11:33 AM gfphoenix78 (via GitHub) <[email protected]> wrote: > > GitHub user gfphoenix78 added a comment to the discussion: [Ideas] > 关于提升Cloudberry高并发查询锁优化的建议 > > Tests 3 spinlock implementations on x86_64 and arm64: > 1. c11 atomic flag (`flag_.test_and_set(std::memory_order_acquire)`) > 2. __atomic_compare_exchange_n > 3. __sync_lock_test_and_set > > The atomic implementations performs worse than the third implementation on > arm64. > But on a small virtual host, the atomic implementations performs better. > It's not clear > how it affects the performance in practice. > > > GitHub link: > https://github.com/apache/cloudberry/discussions/1427#discussioncomment-14964037 > > ---- > This is an automatically sent email for [email protected]. > To unsubscribe, please send an email to: > [email protected] > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
