Re: closing heap relation

From: Zhihong Yu <zyu(at)yugabyte(dot)com>
To: PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: closing heap relation
Date: 2021-07-13 22:21:18
Message-ID: CALNJ-vSEVCF6KFhah31waLtAR-j1+cT-BS8ngUE3uGkstGLGHA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi,

On Tue, Jul 13, 2021 at 3:13 PM Zhihong Yu <zyu(at)yugabyte(dot)com> wrote:

> Hi,
> I was looking at index_drop() in PG 11 branch.
> In if (concurrent)block, the heap and index relations are overwritten
> since they were opened a few lines above the concurrent check.
>
> Shouldn't the two relations be closed first ?
>
> thanks
>
> diff --git a/src/backend/catalog/index.c b/src/backend/catalog/index.c
> index 9d8f873944..625b72ae85 100644
> --- a/src/backend/catalog/index.c
> +++ b/src/backend/catalog/index.c
> @@ -1641,6 +1641,9 @@ index_drop(Oid indexId, bool concurrent)
> * conflicts with existing predicate locks, so now is the
> time to move
> * them to the heap relation.
> */
> + heap_close(userHeapRelation, NoLock);
> + index_close(userIndexRelation, NoLock);
> +
> userHeapRelation = heap_open(heapId,
> ShareUpdateExclusiveLock);
> userIndexRelation = index_open(indexId,
> ShareUpdateExclusiveLock);
> TransferPredicateLocksToHeapRelation(userIndexRelation);
>
Please disregard the above.

The relations were closed a bit earlier.

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Stephen Frost 2021-07-13 22:32:17 Re: proposal: possibility to read dumped table's name from file
Previous Message Tom Lane 2021-07-13 22:16:21 Re: psql - factor out echo code