On Fri, May 19, 2023 at 08:34:16PM +0530, Bharath Rupireddy wrote:
> I get it. How about the following similar to what
> ProcessProcSignalBarrier() has?
>
> + * Note that pg_atomic_exchange_u64 is a full barrier, so we're guaranteed
> + * that the variable is updated before waking up waiters.
> + */
>
> + * Note that pg_atomic_exchange_u64 is a full barrier, so we're guaranteed
> + * that the variable is updated before releasing the lock.
> */
>
> Please find the attached v8 patch with the above change.
Simpler and consistent, nice. I don't have much more to add, so I
have switched the patch as RfC.
--
Michael