Date: Mon, 15 Nov 2021 11:23:40 -0800
From: Andres Freund <andres@anarazel.de>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-hackers@postgresql.org
Subject: Re: [RFC] building postgres with meson
Message-ID: <20211115192340.xoaddx5nwiu2ftud@alap3.anarazel.de>
References: <20211012083721.hvixq4pnh2pixr3j@alap3.anarazel.de>
 <20211115183415.gkcx4akz7vmsvgik@alap3.anarazel.de>
 <2145728.1637003485@sss.pgh.pa.us>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <2145728.1637003485@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/20211115192340.xoaddx5nwiu2ftud%40alap3.anarazel.de>
Precedence: bulk

Hi,

On 2021-11-15 14:11:25 -0500, Tom Lane wrote:
> Andres Freund <andres@anarazel.de> writes:
> >   One important thing to call out: Meson has support for the AIX linker, but
> >   *not* the xlc compiler. I.e. one has to use gcc (or clang, but I didn't
> >   try). I don't know if we'd require adding support for xlc to meson - xlc is
> >   pretty buggy and it doesn't seem particularly crucial to support such an old
> >   crufty compiler on a platform that's not used to a significant degree?
>
> While I have no particular interest in AIX or xlc specifically, I do
> worry about us becoming a builds-on-gcc-or-workalikes-only project.
> I suppose MSVC provides a little bit of a cross-check, but I don't
> really like giving up on other compilers.  Discounting gcc+clang+MSVC
> leaves just a few buildfarm animals, and the xlc ones are a significant
> part of that population.

Yea, that's a reasonable concern. I wonder if there's some non-mainstream
compiler that actually works on, um, more easily available platforms that we
could utilize.


> (In fact, unless somebody renews fossa/husky's
> icc license, the three xlc animals will be an outright majority of
> them, because wrasse and anole are the only other active animals with
> non-mainstream compilers.)

It should probably be doable to get somebody to run another icc animal. Icc is
supported by meson, fwiw.


> Having said that, I don't plan to be the one trying to get meson
> to add xlc support ...

It'd probably not be too hard. But given that it's quite hard to get access to
AIX + xlc, I'm not sure it's something I want to propose. There's no resources
to run halfway regular tests on that I found...


It's good to make sure we're not growing too reliant on some compiler(s), but
imo only really makes sense if the alternative compilers are meaningfully
available and maintained.

Greetings,

Andres Freund