Re: Improving spin-lock implementation on ARM.

On Sat, 5 Dec 2020 at 02:55, Alexander Korotkov <aekorotkov(at)gmail(dot)com> wrote:
>
> On Wed, Dec 2, 2020 at 6:58 AM Krunal Bauskar <krunalbauskar(at)gmail(dot)com> wrote:
> > Let me know what do you think about this analysis and any specific direction that we should consider to help move forward.
>
> BTW, it would be also nice to benchmark my lwlock patch on the
> Kunpeng. I'm very optimistic about this patch, but it wouldn't be
> fair to completely throw it away. It still might be useful for
> LSE-disabled builds.

Coincidentally I was also looking at some hotspot locations around
LWLockAcquire() and LWLockAttempt() for read-only work-loads on both
arm64 and x86, and had narrowed down to the place where
pg_atomic_compare_exchange_u32() is called. So it's likely we are
working on the same hotspot area. When I get a chance, I do plan to
look at your patch while I myself am trying to see if we can do some
optimizations. Although, this is unrelated to the optimization of this
mail thread, so this will need a different mail thread.

--
Thanks,
-Amit Khandekar
Huawei Technologies