Horton

Re: Incorrect result of bitmap heap scan.

Core Studios Inc. <corestudiosinc@gmail.com>

View thread View on postgr.es
From: "Core Studios Inc." <corestudiosinc@gmail.com>
To: Andres Freund <andres@anarazel.de>, Matthias van de Meent <boekewurm+postgres@gmail.com>, Peter Geoghegan <pg@bowt.ie>, Tom Lane <tgl@sss.pgh.pa.us>
Cc: Andrew Dunstan <andrew@dunslane.net>, Melanie Plageman <melanieplageman@gmail.com>, Thomas Munro <thomas.munro@gmail.com>, Konstantin Knizhnik <knizhnik@garret.ru>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>, Alexander Kuzmenkov <akuzmenkov@timescale.com>
Date: 2025-09-16T12:55:49Z
Lists: pgsql-hackers 
Following up to correct my previous email: the query plan did not really 
change - we see Bitmap Heap Scans even before the upgrade. Therefore, 
the query plan doesn't seem to have changed after the upgrade.

However after the upgrade it seems that those queries are doing more 
disk IO (increased index_blocks_read in that table).

Thanks in advance


On 9/16/25 3:25 PM, Andres Freund wrote:
> Hi,
>
> On September 16, 2025 7:57:54 AM EDT, "Core Studios Inc." <corestudiosinc@gmail.com> wrote:
>> Hello,
>>
>> We noticed a sustained increased in IO Wait of read queries after upgrading from 13.13 to 13.21. Eventually, we narrowed it down to a spike in index_blocks_read of a certain table where Bitmap Heap Scans do happen.
>>
>> Do you think that this change (i.e. removing the optimization) could be what caused this regression?
> You're not providing enough details for us to answer that question. We'd need an explain verbose for the query.
>
> Greetings,
>
> Andres

Commits

  1. Add test for HeapBitmapScan's broken skip_fetch optimization

  2. Remove HeapBitmapScan's skip_fetch optimization

  3. Allow bitmap scans to operate as index-only scans when possible.