[FFmpeg-devel] [PATCH 0/5] Integer scalar product DSP functions

[Paul B Mahol]

On 5/8/13, Michael Niedermayer <michaelni at gmx.at> wrote:
> On Wed, May 08, 2013 at 02:51:46PM +0000, Christophe Gisquet wrote:
>> Those functions generally work in batch of 16 elements, but some cases
>> need batches of 8. The first 2 patches only matter for ra144, and
>> alternate
>> patch ra144-pad-buffer-and-use-scalarproduct_int16.patch is probably a
>> better
>> idea.
>>
>
>> wma pro lossless also would need such a modification, but I have never
>> found
>> any sample actually requiring it.
>
> If theres no file that uses it then its probably better not to
> compicate asm code but rather use the C code as your
> "wmaprolossless: switching back to C dsp function" patch does

tak decoder could make use of it, even adding support for batches
of 4 would be nice, instead of if/else code in decoder.

>
>
>>
>> Christophe Gisquet (5):
>>   dsputil: modify scalarproduct_int16 to handle mod8 numbers of loops.
>>   ra144: use scalarproduct_int16
>
>>   dsputil: allow scalarproduct_and_madd_int16 to handle mod8 number of
>>     loops
>
>>   wma lossless: reuse scalarproduct_and_madd_int16
>>   wmaprolossless: switching back to C dsp function
>
> these 2 patches both change wmalosslessdec.c but one calls it
> wmaprolossless
>
> [...]
> --
> Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
>
> Freedom in capitalist society always remains about the same as it was in
> ancient Greek republics: Freedom for slave owners. -- Vladimir Lenin
>