Date: Wed, 15 Dec 2021 17:50:45 +0900
From: Michael Paquier <michael@paquier.xyz>
To: Thomas Munro <thomas.munro@gmail.com>
Cc: Andres Freund <andres@anarazel.de>,
	Andrew Dunstan <andrew@dunslane.net>,
	Anastasia Lubennikova <lubennikovaav@gmail.com>,
	Tom Lane <tgl@sss.pgh.pa.us>,
	pgsql-hackers <pgsql-hackers@postgresql.org>
Subject: Re: A test for replay of regression tests
Message-ID: <YbmsZbEO1QGMymNF@paquier.xyz>
References: 
 <CA+hUKGKwH-mDeReafALoR5Ui5MVM_HKjyskbkMh+U=6oKk+E3w@mail.gmail.com>
 <CA+hUKG+stUD8qE6s5uJPpFaTW-5ZADp+Bq2Bhiu+-00j3xz7gg@mail.gmail.com>
 <CA+hUKGK=FdtCWjx5eqJ8DVtOoAT2W9os+DYNT3Z9MjX-iaa8zA@mail.gmail.com>
 <1f2ea615-2c26-633d-5018-19e4315f84d3@dunslane.net>
 <e05b213c-1257-84d4-f079-5c4d8c79e3ad@dunslane.net>
 <CA+hUKG+nk_WCqi_y+9TentqwXvy=1dW0DgBB-8yTtgdTTLtP0Q@mail.gmail.com>
 <3e9d1044-ddb2-91a4-917b-8cab3ea93b69@dunslane.net>
 <CA+hUKGLnJoeSqtp0tNyFj2GUYaNgbCQ3KTZqCfFH7QY2+Ojcdg@mail.gmail.com>
 <20211209213605.6lwb7o2kzdla34qk@alap3.anarazel.de>
 <CA+hUKGJd_BSh_6hNCDPfvEcRDk6g7gYxybEvVuGgFWe+D3useA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha512;
	protocol="application/pgp-signature"; boundary="KHPizvp7C3Wo4r3G"
Content-Disposition: inline
In-Reply-To: 
 <CA+hUKGJd_BSh_6hNCDPfvEcRDk6g7gYxybEvVuGgFWe+D3useA@mail.gmail.com>
Archived-At: 
 <https://www.postgresql.org/message-id/YbmsZbEO1QGMymNF%40paquier.xyz>
Precedence: bulk


--KHPizvp7C3Wo4r3G
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Fri, Dec 10, 2021 at 12:58:01PM +1300, Thomas Munro wrote:
> -# required for 017_shm.pl
> +# required for 017_shm.pl and 027_stream_regress.pl
>  REGRESS_SHLIB=$(abs_top_builddir)/src/test/regress/regress$(DLSUFFIX)
>  export REGRESS_SHLIB

Hmm.  FWIW, I am working on doing similar for pg_upgrade to switch to
TAP there, and we share a lot in terms of running pg_regress on an
exising cluster.  Wouldn't it be better to move this definition to
src/Makefile.global.in rather than src/test/recovery/?

My pg_regress command is actually very similar to yours, so I am
wondering if this would be better if somehow centralized, perhaps in
Cluster.pm.
--
Michael

--KHPizvp7C3Wo4r3G
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
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=vACs
-----END PGP SIGNATURE-----

--KHPizvp7C3Wo4r3G--