Feedback-ID: ia2694551:Fastmail
Date: Thu, 14 Jul 2022 16:46:52 +0200
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: Thomas Munro <thomas.munro@gmail.com>
Cc: Andres Freund <andres@anarazel.de>,
	Pg Hackers <pgsql-hackers@lists.postgresql.org>
Subject: Re: EINTR in ftruncate()
Message-ID: <20220714144652.jph7exiywushg4gc@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: 
 <CA+hUKGKx6Biq7_UuV0kn9DW+8QWcpJC1qwhizdtD9tN-fn0H0g@mail.gmail.com>
Archived-At: 
 <https://www.postgresql.org/message-id/20220714144652.jph7exiywushg4gc%40alvherre.pgsql>
Precedence: bulk

On 2022-Jul-15, Thomas Munro wrote:

> I checked that this throw-away assertion doesn't fail currently:
> 
>         if (IsUnderPostmaster)
> +       {
> +               sigset_t old;
> +               sigprocmask(SIG_SETMASK, NULL, &old);
> +               Assert(memcmp(&old, &UnBlockSig, sizeof(UnBlockSig)) == 0);
>                 PG_SETMASK(&BlockSig);
> +       }
> 
> ... but now I'm wondering if we should be more defensive and possibly
> even save/restore the mask.

Yeah, that sounds better to me.

> Originally I discounted that because I thought I had to go through
> PG_SETMASK for portability reasons, but on closer inspection, I don't
> see any reason not to use sigprocmask directly in Unix-only code.

ISTM it would be cleaner to patch PG_SETMASK to have a second argument
and to return the original mask if that's not NULL.  This is more
invasive, but there aren't that many callsites of that macro.

-- 
Álvaro Herrera               48°01'N 7°57'E  —  https://www.EnterpriseDB.com/