From: | Andrey Borodin <x4mmm(at)yandex-team(dot)ru> |
---|---|
To: | Michail Nikolaev <michail(dot)nikolaev(at)gmail(dot)com> |
Cc: | Alexander Korotkov <aekorotkov(at)gmail(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, Kirill Reshke <reshkekirill(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Slow standby snapshot |
Date: | 2022-02-20 17:56:08 |
Message-ID: | FC9685E6-85FF-4092-9E52-C0F8A2C0D625@yandex-team.ru |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
> On 22 Nov 2021, at 14:05, Michail Nikolaev <michail(dot)nikolaev(at)gmail(dot)com> wrote:
>
>> Write barrier must be issued after write, not before.
>> Don't we need to issue read barrier too?
>
> The write barrier is issued after the changes to KnownAssignedXidsNext
> and KnownAssignedXidsValid arrays and before the update of
> headKnownAssignedXids.
> So, it seems to be correct. We make sure once the CPU sees changes of
> headKnownAssignedXids - underlying arrays contain all the required
> data.
Patch on barrier seems too complicated to me right now. I’d propose to focus on KnowAssignedXidsNext patch: it’s clean, simple and effective.
I’ve rebased the patch so that it does not depend on previous step. Please check out it’s current state, if you are OK with it - let’s mark the patch Ready for Committer. Just maybe slightly better commit message would make the patch easier to understand.
Thanks! Best regards, Andrey Borodin.
Attachment | Content-Type | Size |
---|---|---|
v4-0001-Use-linked-list-to-improve-KnownAssignedXids-perf.patch | application/octet-stream | 6.1 KB |
From | Date | Subject | |
---|---|---|---|
Next Message | Justin Pryzby | 2022-02-20 19:36:55 | Re: Adding CI to our tree (ccache) |
Previous Message | Robert Haas | 2022-02-20 15:30:21 | Re: Removing more vacuumlazy.c special cases, relfrozenxid optimizations |