Re: pg_stat_statements and "IN" conditions

From: Dmitry Dolgov <9erthalion6(at)gmail(dot)com>
To: Sutou Kouhei <kou(at)clear-code(dot)com>
Cc: yasuo(dot)honda(at)gmail(dot)com, tgl(at)sss(dot)pgh(dot)pa(dot)us, smithpb2250(at)gmail(dot)com, vignesh21(at)gmail(dot)com, michael(at)paquier(dot)xyz, nathandbossart(at)gmail(dot)com, stark(dot)cfm(at)gmail(dot)com, geidav(dot)pg(at)gmail(dot)com, sk(at)zsrv(dot)org, alvherre(at)alvh(dot)no-ip(dot)org, marcos(at)f10(dot)com(dot)br, robertmhaas(at)gmail(dot)com, david(at)pgmasters(dot)net, pgsql-hackers(at)postgresql(dot)org, pavel(dot)trukhanov(at)gmail(dot)com
Subject: Re: pg_stat_statements and "IN" conditions
Date: 2024-05-12 11:38:55
Message-ID: 20240512113855.2ghd6qhgnddwhfci@erthalion.local
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

> On Tue, Apr 23, 2024 at 10:18:15AM +0200, Dmitry Dolgov wrote:
> > On Mon, Apr 15, 2024 at 06:09:29PM +0900, Sutou Kouhei wrote:
> >
> > Thanks. I'm not familiar with this code base but I've
> > reviewed these patches because I'm interested in this
> > feature too.
>
> Thanks for the review! The commentaries for the first patch make sense
> to me, will apply.

Here is the new version. It turned out you were right about memory for
the normalized query, if the number of constants goes close to INT_MAX,
there were indeed not enough allocated. I've added a fix for this on top
of the applied changes, and also improved readability for
pg_stat_statements part.

Attachment Content-Type Size
v20-0001-Prevent-jumbling-of-every-element-in-ArrayExpr.patch text/x-diff 29.1 KB
v20-0002-Reusable-decimalLength-functions.patch text/x-diff 4.7 KB
v20-0003-Merge-constants-in-ArrayExpr-into-groups.patch text/x-diff 19.4 KB
v20-0004-Introduce-query_id_const_merge_threshold.patch text/x-diff 15.0 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Eisentraut 2024-05-12 11:46:29 Re: Add TAP tests for backtrace functionality (was Re: Add test module for verifying backtrace functionality)
Previous Message Alexander Lakhin 2024-05-12 11:00:00 Re: Why is citext/regress failing on hamerkop?