Subject: Re: Preventing abort() and exit() calls in libpq
To: Jacob Champion <pchampion@vmware.com>,
 "tgl@sss.pgh.pa.us" <tgl@sss.pgh.pa.us>
Cc: "daniel@yesql.se" <daniel@yesql.se>,
 "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>,
 "michael@paquier.xyz" <michael@paquier.xyz>
References: <3128896.1624742969@sss.pgh.pa.us>
 <18eda75d-af6d-3fc0-03b6-607eff9f392a@enterprisedb.com>
 <622907.1625084114@sss.pgh.pa.us> <630837.1625092151@sss.pgh.pa.us>
 <67ddcb26f82acaffb26c59b7d8e4cd75399b2cd3.camel@vmware.com>
From: Peter Eisentraut <peter.eisentraut@enterprisedb.com>
Message-ID: <9d35a4ac-2c75-cb44-95ab-61d69deb072e@enterprisedb.com>
Date: Thu, 1 Jul 2021 20:05:59 +0200
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:78.0)
 Gecko/20100101 Thunderbird/78.11.0
MIME-Version: 1.0
In-Reply-To: <67ddcb26f82acaffb26c59b7d8e4cd75399b2cd3.camel@vmware.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Archived-At: 
 <https://www.postgresql.org/message-id/9d35a4ac-2c75-cb44-95ab-61d69deb072e%40enterprisedb.com>
Precedence: bulk

On 01.07.21 00:41, Jacob Champion wrote:
> On Wed, 2021-06-30 at 18:29 -0400, Tom Lane wrote:
>> I wrote:
>>> Peter Eisentraut <peter.eisentraut@enterprisedb.com> writes:
>>>> Could we set this rule up a little bit differently so that it is only
>>>> run when the library is built.
>>>> Right now, make world on a built tree makes 17 calls to this "nm" line,
>>>> and make check-world calls it 81 times.  I think once would be enough. ;-)
>>> Hmm, didn't realize that would happen.  Will see what can be done.
>>
>> Looks like we'd have to make use of a dummy stamp-file, more or less
>> as attached.  Any objections?
> 
> Spitballing -- if you don't like the stamp file, you could add the
> check to the end of the $(shlib) rule, surrounded by an ifeq check.
> Then .DELETE_ON_ERROR should take care of the rest, I think.

Somewhere in the $(shlib) rule would seem most appropriate.  But I don't 
understand the rest: What ifeq, and why .DELETE_ON_ERROR?