Re: [PATCH 0/6] MR6360: d3dx9: Beginning of pixel format handling changes.

[Matteo Bruni (＠Mystral)]

Matteo Bruni (@Mystral) commented about dlls/d3dx9_36/tests/surface.c:
> -
>        for (x = 0; x < 4; ++x)
> -
>            check_readback_pixel_4bpp(&surface_rb, x, y, src_pixels[(y * 4) + x], FALSE);
> -
>    }
> -
>    release_surface_readback(&surface_rb);
> -
> -
>    /*
> -
>     * Load the DXT block as a premultiplied alpha format. It's
> -
>     * demultiplied on upload.
> -
>     */
> -
>    hr = D3DXLoadSurfaceFromMemory(decomp_surf, NULL, NULL, dxt_block, pma_dxt_fmt, 16, NULL, &src_rect, D3DX_FILTER_NONE, 0);
> -
>    ok(hr == D3D_OK, "Unexpected hr %#lx.\n", hr);
> -
>    IDirect3DSurface9_LockRect(decomp_surf, &lock_rect, NULL, D3DLOCK_READONLY);
> -
>    for (y = 0; y < 4; ++y)
> -
>    {
> -
>        for (x = 0; x < 4; ++x)
> -
>            todo_wine check_pixel_4bpp(&lock_rect, x, y, 0x00000000);

Can we use a more interesting test image so that we get non-black output here? I realize this means using two different test images and / or expected results for the two tests.