From: Tom Lane <tgl@sss.pgh.pa.us>
To: Alvaro Herrera <alvherre@alvh.no-ip.org>
cc: Andrew Dunstan <andrew@dunslane.net>,
 Justin Pryzby <pryzby@telsasoft.com>,
        pgsql-hackers@postgresql.org, Andres Freund <andres@anarazel.de>
Subject: Re: make -C libpq check fails obscurely if tap tests are disabled
In-reply-to: <20220722201945.wk22sn2evf4enztn@alvherre.pgsql>
References: <20220722201945.wk22sn2evf4enztn@alvherre.pgsql>
Comments: In-reply-to Alvaro Herrera <alvherre@alvh.no-ip.org>
	message dated "Fri, 22 Jul 2022 22:19:45 +0200"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <748837.1658522443.1@sss.pgh.pa.us>
Date: Fri, 22 Jul 2022 16:40:43 -0400
Message-ID: <748838.1658522443@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/748838.1658522443%40sss.pgh.pa.us>
Precedence: bulk

Alvaro Herrera <alvherre@alvh.no-ip.org> writes:
> On 2022-Jul-22, Alvaro Herrera wrote:
>> It's not very common -- we do have some target-specific variable
>> assignments, but none of them use 'export'.  I saw somewhere that this
>> works from Make 3.77 onwards, and we require 3.80, so it should be okay.
>> The buildfarm will tell us ...

> Hm, so prairiedog didn't like this:
> According to 32568.1536241083@sss.pgh.pa.us, prairiedog is on Make 3.80.

Yeah, it is.  I looked at the gmake manual on that machine, and its
description of "export" seems about the same as what I see in a
modern version.  So it should work ... but we've found bugs in 3.80
before.

Let me poke at it and see if there's a variant that works.
The wording of the message suggests that maybe breaking it into
two separate rules would help.

			regards, tom lane