Re: [PATCH 1/2] d3d9/tests: Extend the FPU setup test.

[Henri Verbeet]

Not that I would terribly mind if this got comitted, but I have to
wonder if this really has a lot of value as a regression test?

[PATCH 1/2] d3d9/tests: Extend the FPU setup test.

[Stefan Dösinger]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Am 2015-02-27 um 12:34 schrieb Henri Verbeet:
> Not that I would terribly mind if this got comitted, but I have to 
> wonder if this really has a lot of value as a regression test?

As a regression test not really, except for cases where someone wants
to add FPU CW set / restore in the way we're doing it in d3d7. I think
the code has value as documentation though.

I don't have a strong opinion on this either, but my preference is to
have it in the tree rather than in the archives.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQIcBAEBAgAGBQJU8FhHAAoJEN0/YqbEcdMwtMMQAJWSekh9Yc/6mrj/3JsbhkDy
UwlwExx11ldu3h5GwCph7ozjdVuzVTiU1+MqDd5opsSA3rCOEidVByNBCziCh6NH
gsDt5VCYWD+Ri3ynGZsGrfI8xNGdffBbyke5irWnZnT49wE5BetprM6+k0q1ti/P
bBEiv8J6jkCayBCQ1Ehjgvb6fL1GAcF9HMtC7nVKOuNKWD+OYawaUPdkKhEtWD3+
Qau8tEqrIhwswloBy6UDqxLkazzTQDoTi8/LXmDhnzuDk/hDqarlqpYMgFz0C5vM
SF0Y1ADxxTEtDdODnmrwu21OkNq8kWqgv8eEdDWowDIGpsV07x4AI/9BpSaA3NcM
FA2563EsBDIb+uF3xZ+PCtxBYfXgvF2gl3hiaH9tSgLVrBdGLIi2ofbAJ7tzCuTO
7VF3Otjhldmi84UEbg/Xc3vHAb1uJnzweNEy1p03nrgUHBxfKFaASilp1PNOmkWz
kHgYdzKSm34I3+Jw1hbIeOx2pFPS3kZU4xuQYhg3RtKWUU5OWiOlqNiF987I9Whk
+srjOwDaNnkG8E/dWJhqgDXxIbafU7q7UX/j9plRnp1vVysVNbYI5N6z7J5S4WKE
roYGg0QYHaBE9ibLh9TOUpu30sV6EHtk25nAOBhVADgy7QamssfB9+tqMY9/jRDX
0nVuTYn9jiO6Z/cRLZfr
=2n3Z
-----END PGP SIGNATURE-----

[PATCH 1/2] d3d9/tests: Extend the FPU setup test.

[Henri Verbeet]

On 27 February 2015 at 12:43, Stefan Dösinger  wrote:
> Am 2015-02-27 um 12:34 schrieb Henri Verbeet:
> > Not that I would terribly mind if this got comitted, but I have to
> > wonder if this really has a lot of value as a regression test?
>
> As a regression test not really, except for cases where someone wants
> to add FPU CW set / restore in the way we're doing it in d3d7. I think
> the code has value as documentation though.
>
> I don't have a strong opinion on this either, but my preference is to
> have it in the tree rather than in the archives.

I guess it's mostly a matter of how Alexandre feels about this kind of
thing then.

[PATCH 1/2] d3d9/tests: Extend the FPU setup test.

[Alexandre Julliard]

Henri Verbeet  writes:
> On 27 February 2015 at 12:43, Stefan Dösinger  wrote:
> > Am 2015-02-27 um 12:34 schrieb Henri Verbeet:
> > > Not that I would terribly mind if this got comitted, but I have to
> > > wonder if this really has a lot of value as a regression test?
> >
> > As a regression test not really, except for cases where someone wants
> > to add FPU CW set / restore in the way we're doing it in d3d7. I think
> > the code has value as documentation though.
> >
> > I don't have a strong opinion on this either, but my preference is to
> > have it in the tree rather than in the archives.
>
> I guess it's mostly a matter of how Alexandre feels about this kind of
> thing then.

I don't mind having them in. At least they don't prevent me from using
the computer while running, unlike some of the other d3d tests ;-)