Re: Debian 12 gcc warning

From: Bruce Momjian <bruce(at)momjian(dot)us>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: John Naylor <john(dot)naylor(at)enterprisedb(dot)com>, David Rowley <dgrowleyml(at)gmail(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Debian 12 gcc warning
Date: 2023-08-30 02:52:06
Message-ID: ZO6u1hySX4lBI/31@momjian.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, Aug 29, 2023 at 10:18:36AM -0400, Tom Lane wrote:
> Bruce Momjian <bruce(at)momjian(dot)us> writes:
> > On Tue, Aug 29, 2023 at 10:26:27AM +0700, John Naylor wrote:
> >> It looks like the former, since I can silence it on gcc 13 / -O1 by doing:
> >> /* keep compiler quiet */
> >> actual_arg_types[0] = InvalidOid;
>
> > Agreed, that fixes it for me too. In fact, assigning to only element 99 or
> > 200 also prevents the warning, and considering the array is defined for
> > 100 elements, the fact is accepts 200 isn't a good thing. Patch attached.
>
> That seems like a pretty clear compiler bug, particularly since it just
> appears in this one version. Rather than contorting our code, I'd
> suggest filing a gcc bug.

I assume I have to create a test case to report this to the gcc team. I
tried to create such a test case on gcc 12 but it doesn't generate the
warning. Attached is my attempt. Any ideas? I assume we can't just
tell them to download our software and compile it.

--
Bruce Momjian <bruce(at)momjian(dot)us> https://momjian.us
EDB https://enterprisedb.com

Only you can decide what is important to you.

Attachment Content-Type Size
test_gcc_12.2.0-O1.zip application/zip 6.4 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2023-08-30 03:30:06 Re: Debian 12 gcc warning
Previous Message Peter Smith 2023-08-30 02:24:43 Re: [PoC] pg_upgrade: allow to upgrade publisher node