Re: Fix performance degradation of contended LWLock on NUMA

On Mon, Jun 5, 2017 at 9:22 AM, Sokolov Yura
<funny(dot)falcon(at)postgrespro(dot)ru> wrote:
> Good day, everyone.
>
> This patch improves performance of contended LWLock.
> It was tested on 4 socket 72 core x86 server (144 HT) Centos 7.1
> gcc 4.8.5
>
> Patch makes lock acquiring in single CAS loop:
> 1. LWLock->state is read, and ability for lock acquiring is detected.
> If there is possibility to take a lock, CAS tried.
> If CAS were successful, lock is aquired (same to original version).
> 2. but if lock is currently held by other backend, we check ability for
> taking WaitList lock. If wait list lock is not help by anyone, CAS
> perfomed for taking WaitList lock and set LW_FLAG_HAS_WAITERS at once.
> If CAS were successful, then LWLock were still held at the moment wait
> list lock were held - this proves correctness of new algorithm. And
> Proc is queued to wait list then.
> 3. Otherwise spin_delay is performed, and loop returns to step 1.

Interesting work. Thanks for posting.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company