From: Tom Lane <tgl@sss.pgh.pa.us>
To: "Tristan Partin" <tristan@partin.io>
cc: "pgsql-hackers" <pgsql-hackers@postgresql.org>
Subject: Re: Interest in hosting a buildfarm animal
In-reply-to: <DI6P3PWLOCXS.3AX299MX6N8CN@partin.io>
References: <DI6P3PWLOCXS.3AX299MX6N8CN@partin.io>
Comments: In-reply-to "Tristan Partin" <tristan@partin.io>
	message dated "Thu, 30 Apr 2026 18:19:40 -0000"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <768565.1777574155.1@sss.pgh.pa.us>
Date: Thu, 30 Apr 2026 14:35:55 -0400
Message-ID: <768566.1777574155@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/768566.1777574155%40sss.pgh.pa.us>
Precedence: bulk

"Tristan Partin" <tristan@partin.io> writes:
> Due to a recent change in position, I am now in the position to host 
> buildfarm animals. Are there any configurations that don't have any 
> testing or enough testing? Perhaps something valgrind related? Let me 
> know!

One silly thing I think is uncovered is whether a production
build (*without* cassert or debug) gets any compiler warnings.
We've so often written code that will generate "unused variable"
or similar complaints if assertions are off, yet nobody notices
that till well after commit.  Developers are unlikely to test
that way, and there are next to no BF animals that do either.

			regards, tom lane