Re: [PATCH 2/5] d3d9/tests: Allow the creation of SWVP devices

[Henri Verbeet]

On 15 January 2013 00:13, Stefan Dösinger <stefan(a)codeweavers.com> wrote:
> -static IDirect3DDevice9 *create_device(IDirect3D9 *d3d9, HWND device_window, HWND focus_window, BOOL windowed)
> +static IDirect3DDevice9 *create_device(IDirect3D9 *d3d9, HWND device_window, HWND focus_window, BOOL windowed,
> +        DWORD flags)
>  {
>      D3DPRESENT_PARAMETERS present_parameters = {0};
>      IDirect3DDevice9 *device;
> @@ -90,7 +91,7 @@ static IDirect3DDevice9 *create_device(IDirect3D9 *d3d9, HWND device_window, HWN
>      present_parameters.AutoDepthStencilFormat = D3DFMT_D24S8;
>
> if (SUCCEEDED(IDirect3D9_CreateDevice(d3d9, D3DADAPTER_DEFAULT, D3DDEVTYPE_HAL, focus_window,
> -            D3DCREATE_HARDWARE_VERTEXPROCESSING, &present_parameters, &device))) return device;
> +            flags, &present_parameters, &device))) return device;

If this fails, we're going to try hardware vertex processing again,
but I don't think this is something we want to specify for every
single test anyway. If this is just for a single test that needs
software vertex processing, just write it out. If you already know
there are going to be multiple tests like this, you could probably
introduce a function like create_swvp_device().

[PATCH 2/5] d3d9/tests: Allow the creation of SWVP devices

[Stefan Dösinger]

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

Am 2013-01-15 08:33, schrieb Henri Verbeet:
> If this fails, we're going to try hardware vertex processing
> again, but I don't think this is something we want to specify for
> every single test anyway. If this is just for a single test that
> needs software vertex processing, just write it out. If you already
> know there are going to be multiple tests like this, you could
> probably introduce a function like create_swvp_device().
> I don't have any other written tests that need swvp yet. There might
> be some related ones in the future(there are still apps that don't
> like our dynamic buffer handling), or some tests looking at
> PUREDEVICE, but I can refactor the code when I'm writing such tests.
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v2.0.19 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
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=ORos
-----END PGP SIGNATURE-----