[FFmpeg-devel] [PATCH 0/6] avx functions for h264 chroma deblocking

[James Almer]

On 2/20/2017 12:33 PM, James Darnley wrote:
> 6 more functions which eke out a little more speed.
> 
> James Darnley (6):
>   avcodec/h264: add avx 8-bit chroma v deblock/loop filter
>   avcodec/h264: add avx 8-bit 4:2:0 chroma h deblock/loop filter
>   avcodec/h264: add avx 8-bit 4:2:2 chroma h deblock/loop filter
>   avcodec/h264: add avx 8-bit chroma v intra deblock/loop filter
>   avcodec/h264: add avx 8-bit 4:2:0 chroma h intra deblock/loop filter
>   avcodec/h264: add avx 8-bit 4:2:2 chroma h intra deblock/loop filter
> 
>  libavcodec/x86/h264_deblock.asm | 195 ++++++++++++++++++++++++++++++++++++++++
>  libavcodec/x86/h264dsp_init.c   |  10 +++
>  2 files changed, 205 insertions(+)

All of these can and should be SSE2 first and foremost, and AVX only if
there's any gain on top of it from using the three operand version of the
instructions.