Feedback-ID: id4a34324:Fastmail
Date: Thu, 15 Sep 2022 14:09:38 -0700
From: Andres Freund <andres@anarazel.de>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Noah Misch <noah@leadboat.com>, pgsql-committers@lists.postgresql.org
Subject: Re: pgsql: aix: No need to use mkldexport when we want to export all
 symbol
Message-ID: <20220915210938.xxo4ysqin45h5oni@awork3.anarazel.de>
References: <20220910043749.koidc5k4j5dxo3if@awork3.anarazel.de>
 <20220910052741.t7w7csyrggwus2ze@awork3.anarazel.de>
 <338810.1662787972@sss.pgh.pa.us>
 <20220910055736.htsr5juo3ldlay2l@awork3.anarazel.de>
 <20220910081944.oiayp5lky3kv2o43@awork3.anarazel.de>
 <20220911181925.ygxbync3uk5ezcic@awork3.anarazel.de>
 <20220913182530.4yq35ralujjsv2k3@awork3.anarazel.de>
 <2490844.1663123546@sss.pgh.pa.us>
 <20220915194422.bbdzwo7gmcvwzv4b@awork3.anarazel.de>
 <3442427.1663274884@sss.pgh.pa.us>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3442427.1663274884@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/20220915210938.xxo4ysqin45h5oni%40awork3.anarazel.de>
Precedence: bulk

Hi,

On 2022-09-15 16:48:04 -0400, Tom Lane wrote:
> Andres Freund <andres@anarazel.de> writes:
> > On 2022-09-13 22:45:46 -0400, Tom Lane wrote:
> >> A three-line patch doesn't seem like an unreasonable thing to carry,
> >> at least till these systems go out of support.  We've jumped through
> >> much higher hoops in the past to support niche platforms.
> 
> > xlc 12.1 *is* out of support and there's a newer compiler installed.
> 
> AIX 7.1 will go EOL in May.  It seems to me to be reasonable to keep
> these animals running till then and then shut them down [1].  In the
> meantime, messing with their configurations seems like an unreasonable
> amount of work compared to applying this patch.

I guess I think updating the path to the compiler in the config isn't all that
heavy a lift, compared to making others understand this stuff, but...  I'll
push a revised patch.


> > The amount of lines itself doesn't bother me. What does is that it makes an
> > already complicated platform more complicated: All symbols from extension
> > .so's are exported, but only for old and unsupported versions of the
> > proprietary compiler.
> 
> I have no objection to not supporting the case in the meson scripts,
> if that'd make you any happier.

Meson doesn't, so far, support xlc, just gcc on AIX. It'd probably not too
hard to teach it, but I'm not planning to do so - doing the work to make aix
w/ gcc work was already plenty effort on my end.

Greetings,

Andres Freund