Re: [PATCH] d3drm/tests: Add tests for IDirect3DRM*::LoadTexture (try 5).

[Stefan Dösinger]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Am 2015-12-27 um 10:23 schrieb Aaryaman Vasishta:
> -
>    if (info_header.biWidth % 4 != 0)
> -
>        expected = (info_header.biWidth & ~0x03) + 4;
> -
>    else
> -
>        expected = info_header.biWidth;

This is more complicated than it needs to be:

expected = (info_header.biWidth + 3) & ~3

should work as well.
> -
>                k++;
> -
>                /* Take into account padded bytes after every row for odd widths */
> -
>                if ((w % 2 == 1) && (j == w - 1))
> -
>                    k = (k & ~0x03) + 4;

Wouldn't this also apply for e.g. a width of 10 that is a multiple of 2, but not a multiple of 4?

The canonical way is to advance bytes_per_line bytes every line, e.g. buffer1[i * bytes_per_line + j]

Or do k += bytes_per_line in the outer loop and then just do buffer1[k + j] to avoid re-doing the multiplication every time. I'd expect the compiler to optimize it anyway though.

Other than that the patch looks good :-)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
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=BnPO
-----END PGP SIGNATURE-----

[PATCH] d3drm/tests: Add tests for IDirect3DRM*::LoadTexture (try 5).

[Aaryaman Vasishta]

On Fri, Jan 1, 2016 at 6:39 PM, Stefan Dösinger 
wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
>
> Am 2015-12-27 um 10:23 schrieb Aaryaman Vasishta:
> > -
> >    if (info_header.biWidth % 4 != 0)
> > -
> >        expected = (info_header.biWidth & ~0x03) + 4;
> > -
> >    else
> > -
> >        expected = info_header.biWidth;
>
> This is more complicated than it needs to be:
>
> expected = (info_header.biWidth + 3) & ~3
>
> should work as well.

Much better solution :)

I've sent try 6 with the changes.

Cheers,
Aaryaman