Re: Upgrade from Fedora 40 to Fedora 42, or from PostgreSQL 16.3 to PostgreSQL 16.9

From: "Michael J(dot) Baars" <mjbaars1977(dot)pgsql(at)gmail(dot)com>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: Upgrade from Fedora 40 to Fedora 42, or from PostgreSQL 16.3 to PostgreSQL 16.9
Date: 2025-07-19 09:44:46
Message-ID: CALudE+73-SNoybAfNo1V95XjaVr=_EC5LvwT2LTmYLFcxRb1=Q@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Fri, Jul 18, 2025 at 4:09 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>
> "Michael J. Baars" <mjbaars1977(dot)pgsql(at)gmail(dot)com> writes:
> > Somewhere in between release 16.3 and
> > release 16.9, changes must have been implemented that make the execution
> > engine about two times slower than it was.
>
> That seems quite unlikely, and you've surely provided no details that
> would help identify any such problem. My own first guess is that the
> newer install was somehow built with debugging assertions enabled
> (configure's --enable-cassert option). Those are pretty heavyweight
> and could possibly account for this. A second guess is that you were
> relying on JIT compilation and the newer install somehow omitted that
> (failed to use --with-llvm). Anyway, I'd start by comparing the
> output of pg_config from both installs.

Thanks for the tip!

These are the difference I find:

16.3 is built with

--sbindir=/usr/sbin

16.9 is built with

--sbindir=/usr/bin
-mtls-dialect=gnu2
-DOPENSSL_NO_ENGINE
-std=c17

Could these have the same effect to some degree?

Neither was built with --enable-cassert or --with-llvm.

> regards, tom lane

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Shlok Kyal 2025-07-19 10:44:03 Re: Skipping schema changes in publication
Previous Message Michael J. Baars 2025-07-19 09:38:48 Re: Lossless transmission of double precision floating point