On Sat, 9 Jul 2022 00:04:27 +0200 Morten Brørup <m...@smartsharesystems.com> wrote:
> Always the creative mind, Stephen. :-) > > You might consider adding/updating even more comments. > > Acked-by: Morten Brørup <m...@smartsharesystems.com> The motivation is that our work load is reader/writer lock heavy with small number of threads. Therefore the number of atomic operations per lock matters, but starving is bad. And any compare-exchange on ARM is expensive and should be avoided if possible. The concept here came from this great page. https://locklessinc.com/articles/locks/ Will add link in next version.
