Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
From: Daniel Gustafsson <daniel@yesql.se>
Mime-Version: 1.0 (1.0)
Subject: Re: Prevent an error on attaching/creating a DSM/DSA from an interrupt handler.
Date: Thu, 1 May 2025 12:08:09 +0200
Message-Id: <18F24077-BCF2-4892-A9F6-72DE9AD254BC@yesql.se>
References: <CA+TgmoYWbjZ9k1kuia2CXfJ3+b6G+=AayzU-hsXdtvGLgstFcw@mail.gmail.com>
Cc: Rahila Syed <rahilasyed90@gmail.com>,
 PostgreSQL Hackers <pgsql-hackers@postgresql.org>,
 Fujii Masao <masao.fujii@oss.nttdata.com>, Tomas Vondra <tomas@vondra.me>
In-Reply-To: <CA+TgmoYWbjZ9k1kuia2CXfJ3+b6G+=AayzU-hsXdtvGLgstFcw@mail.gmail.com>
To: Robert Haas <robertmhaas@gmail.com>
Archived-At: <https://www.postgresql.org/message-id/18F24077-BCF2-4892-A9F6-72DE9AD254BC%40yesql.se>
Precedence: bulk

> On 1 May 2025, at 00:04, Robert Haas <robertmhaas@gmail.com> wrote:
>=20
> =EF=BB=BFOn Wed, Apr 30, 2025 at 5:24=E2=80=AFPM Daniel Gustafsson <daniel=
@yesql.se> wrote:
>> Attached is a current v4 with a few small tweaks.
>=20
> Sorry to turn up late here, but I strongly disagree with the notion
> that this is a bug in the DSM or DSA code. It seems to me that it is
> the caller's responsibility to provide a valid resource owner, not the
> job of the called code to ignore the resource owner when it's
> unusable. I suspect that there are many other parts of the system that
> rely on the ResourceOwner machinery which likewise assume that the
> ResourceOwner that they are passed is valid.

Thanks for review, I=E2=80=99m currently travelling over the extended weeken=
d but will rework the approach shortly when back at the office.

./daniel=