[x264-devel] Re: x264: svn commit r618 (pengvado)

[Limin Wang]

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

Hi there,

+#define array_non_zero(a) array_non_zero_int(a, sizeof(a))
+static inline int array_non_zero_int( void *v, int i_count )

How about to use array_non_zero_int32 since int may be 2 byte
in some platform?

So like below code need use int32_t instead of int.
    struct
    {
        DECLARE_ALIGNED( int, luma16x16_dc[16], 16 );
        DECLARE_ALIGNED( int, chroma_dc[2][4], 16 );
        // FIXME merge with union
        DECLARE_ALIGNED( int, luma8x8[4][64], 16 );
        union
        {
            DECLARE_ALIGNED( int, residual_ac[15], 16 );
            DECLARE_ALIGNED( int, luma4x4[16], 16 );
        } block[16+8];
    } dct;

In addition, dct2x2 is defined as int16_t type, so I think it's better
to define and use array_non_zero_int16()?

Thanks,
Limin

* Subversion daemon <svn at videolan.org> [2007-01-12 00:55:52 +0100]:

> r618 | pengvado | 2007-01-12 00:55:51 +0100 (Fri, 12 Jan 2007) | 2 lines
> Changed paths:
>    M /trunk/encoder/macroblock.c
>    M /trunk/encoder/macroblock.h
> 
> make array_non_zero() compatible with -fstrict-aliasing
> 
>  > http://trac.videolan.org/x264/changeset/618
> 
> -- 
> This is the x264-devel mailing-list
> To unsubscribe, go to: http://developers.videolan.org/lists.html
> 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iQEVAwUBRabqREztbf7dKiuoAQJ+twf/bTQgcv/PbpSiCOKFdQbtnt/o/r3YXCXg
tZ9I9jDC/CtkvCjZmvhjZf0NXOZQnmijdfsM9KSEDwgKwcrMm/OuUggL+FaHketl
Re8LzMv4dj7vB6kv/Hxsl0X+9dMh8g6z9aoNyrfZc+0j4ARHAx1aZ1FinAhM7qzX
R5ellD21Ia4R+dqBs1aV0+u/IONw36YnjoPauLlisr6ER2UhK84gPrBWhXD8v8dn
8Ymr+AJNFM98oAVNaCdnDRBbLoBj4Av2J3srfkrRhRleM7X4gB/47oJGVAg8HvRn
L3GPDfHD5bsoB8No76JwOzdyuxCM9fsZvNAay4VXbgaXoQP9VzMH5Q==
=hW6l
-----END PGP SIGNATURE-----

-- 
This is the x264-devel mailing-list
To unsubscribe, go to: http://developers.videolan.org/lists.html