Re: COPY FREEZE and setting PD_ALL_VISIBLE/visibility map bits - Mailing list pgsql-hackers

From Ibrar Ahmed
Subject Re: COPY FREEZE and setting PD_ALL_VISIBLE/visibility map bits
Date
Msg-id CALtqXTdwh06yNn_DzQDOKaGFyY750Vw5WdipZOF=XHGrWKdxOw@mail.gmail.com
Whole thread Raw
In response to Re: COPY FREEZE and setting PD_ALL_VISIBLE/visibility map bits  (Ibrar Ahmed <ibrar.ahmad@gmail.com>)
Responses Re: COPY FREEZE and setting PD_ALL_VISIBLE/visibility map bits  (Daniel Gustafsson <daniel@yesql.se>)
List pgsql-hackers


On Tue, Mar 24, 2020 at 10:06 PM Ibrar Ahmed <ibrar.ahmad@gmail.com> wrote:


On Fri, Mar 13, 2020 at 6:58 AM Justin Pryzby <pryzby@telsasoft.com> wrote:

Thanks for picking up this patch.  There's a minor typo:

+                        * readable outside of this sessoin. Therefore doing IO here isn't

=> session

--
Justin

Thanks, please see the updated and rebased patch. (master 17a28b03645e27d73bf69a95d7569b61e58f06eb)

--
Ibrar Ahmed

Andres while fixing the one FIXME in the patch

"            visibilitymap_pin(relation, BufferGetBlockNumber(buffer), &vmbuffer);


            /*

             * FIXME: setting recptr here is a dirty dirty hack, to prevent

             * visibilitymap_set() from WAL logging.

             *
"
I am not able to see any scenario where recptr is not set before reaching to that statement. Can you clarify why you think recptr will not be set at that statement?


--
Ibrar Ahmed

pgsql-hackers by date:

Previous
From: Pengzhou Tang
Date:
Subject: Re: Parallel grouping sets
Next
From: Laurenz Albe
Date:
Subject: Re: Berserk Autovacuum (let's save next Mandrill)