(This branch may should leave from this thread..)
At Fri, 15 Oct 2021 15:00:57 +0900 (JST), Kyotaro Horiguchi <horikyota.ntt@gmail.com> wrote in
> At Thu, 14 Oct 2021 10:53:06 -0700, Andres Freund <andres@anarazel.de> wrote in
> > This'd get rid of the need of density *and* make SIInsertDataEntries()
> > cheaper.
>
> Yes. So.. I tried that. The only part where memory-flush timing is
> crucial seems to be between writing messages and setting maxMsgNum.
> By placing memory barrier between them it seems *to me* we can read
> maxMsgNum safely without locks.
Maybe we need another memory barrier here and the patch was broken
about the rechecking on the members in GetSIGetDataEntries..
regards.
--
Kyotaro Horiguchi
NTT Open Source Software Center
