Postgres Pro
Downloads
Home   >   mailing lists

Re: Unaccounted regression from postgresql 11 in later versions - Mailing list pgsql-performance

From Michael Christofides
Subject Re: Unaccounted regression from postgresql 11 in later versions
Date
Msg-id CAFwT4nB8YBy2UW3mRTNrL7J6z0tji0XquS=K6WBHskAtMce11g@mail.gmail.com
Whole thread Raw
In response to Unaccounted regression from postgresql 11 in later versions  (Bob Jolliffe <bobjolliffe@gmail.com>)
Responses Re: Unaccounted regression from postgresql 11 in later versions
List pgsql-performance
Tree view
Does anyone have a theory of why pg15 should behave so differently to pg11 here?  Better still, any suggestions for configuration that might make pg15 behave more like pg10.  I am really dreading the prospect of stepping our many live implementations back to pg11 :-(.

One major factor here appears to be JIT compilation, which is off by default in pg11, but on by default in pg12+.

You can see at the bottom of your slowest query plan that about 233s of the 240s are JIT related.

There is good info in the docs about tuning, or turning off, JIT: https://www.postgresql.org/docs/current/jit-decision.html 

pgsql-performance by date:

Previous
From: Bob Jolliffe
Date:
Subject: Unaccounted regression from postgresql 11 in later versions
Next
From: Bob Jolliffe
Date:
Subject: Re: Unaccounted regression from postgresql 11 in later versions