From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andres Freund <andres@anarazel.de>
cc: Peter Eisentraut <peter.eisentraut@enterprisedb.com>,
        pgsql-hackers@postgresql.org, David Rowley <dgrowleyml@gmail.com>,
        Dagfinn Ilmari =?utf-8?Q?Manns=C3=A5ker?= <ilmari@ilmari.org>
Subject: Re: perl 5.36, C99,
 -Wdeclaration-after-statement -Wshadow=compatible-local
In-reply-to: <20221229184236.3v7bvsumefpslq7h@awork3.anarazel.de>
References: <20221101180158.l4zg6ljah5ow2vvo@awork3.anarazel.de>
 <e5e13d55-8c8b-30a6-0b13-0304a3e562dd@enterprisedb.com>
 <20221102234346.zcupk5mqysavci4t@awork3.anarazel.de>
 <1881935.1667433465@sss.pgh.pa.us>
 <20221103000334.sbh5jqw37i3y7pc3@awork3.anarazel.de>
 <20221228182455.hfdwd22zztvkojy2@awork3.anarazel.de>
 <356473.1672253007@sss.pgh.pa.us>
 <20221229000223.2pdwxexfrgoximul@awork3.anarazel.de>
 <438139.1672272335@sss.pgh.pa.us>
 <20221229184236.3v7bvsumefpslq7h@awork3.anarazel.de>
Comments: In-reply-to Andres Freund <andres@anarazel.de>
	message dated "Thu, 29 Dec 2022 10:42:36 -0800"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <586728.1672339897.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Thu, 29 Dec 2022 13:51:37 -0500
Message-ID: <586729.1672339897@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/586729.1672339897%40sss.pgh.pa.us>
Precedence: bulk

Andres Freund <andres@anarazel.de> writes:
> It turns out to not work terribly well. gcc, quite reasonably, warns abo=
ut the
> pragma used in .c files, and there's no easy way that I found to have au=
toconf
> name its test .h. We could include a test header in the compile test, bu=
t that
> also adds some complication. As gcc has supported the pragma since 2000,=
 I
> think a simple
>   #ifdef __GNUC__
>   #define HAVE_PRAGMA_SYSTEM_HEADER	1
>   #endif
> should suffice.

We might find that some GCC-impostor compilers have trouble with it,
but if so we can adjust the #ifdef here.

Getting nitpicky, I suggest calling it "HAVE_PRAGMA_GCC_SYSTEM_HEADER"
to align better with what you actually have to write.  Also:

+ * Newer versions the perl headers trigger a lot of warnings with our com=
piler

"Newer versions of ..." please.  Otherwise LGTM.

> Should we backpatch this? Given the volume of warnings it's probably a g=
ood
> idea. But I'd let it step in HEAD for a few days of buildfarm coverage f=
irst.

+1 to both points.

			regards, tom lane