Re: [FFmpeg-devel] [PATCH 0/4] More H.264 assembly (the sequel) [version 2]

2016-12-06 Thread Ronald S. Bultje
Hi,

On Tue, Dec 6, 2016 at 7:04 AM, James Darnley  wrote:

> On 2016-12-05 19:32, James Darnley wrote:
> > Fixed the problem Michael highlighted.  Dropped the intra functions
> until it
> > becomes clear why their performance is unexpected. Updated the
> benchmarks with
> > results from a Nehalem and used (slightly) more accurate data.
> >
> > Regarding the age of MMX:  I have written it so unless someone tells me
> to
> > remove it I will keep the code.  However, I will probably not write any
> more
> > going forward.


The "age" of mmx has been brought up for a while now, but latest intel CPUs
still support it, so I agree it's fine for now.

If nobody raises objections or makes further comments I will push these,
> including the AVX, later today.


OK.

Ronald
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel


Re: [FFmpeg-devel] [PATCH 0/4] More H.264 assembly (the sequel) [version 2]

2016-12-06 Thread James Darnley
On 2016-12-05 19:32, James Darnley wrote:
> Fixed the problem Michael highlighted.  Dropped the intra functions until it
> becomes clear why their performance is unexpected. Updated the benchmarks with
> results from a Nehalem and used (slightly) more accurate data.
> 
> Regarding the age of MMX:  I have written it so unless someone tells me to
> remove it I will keep the code.  However, I will probably not write any more
> going forward.
> 
> James Darnley (4):
>   avcodec/h264: clean up and expand x86 function definitions
>   whitespace changes after last commit
>   avcodec/h264: mmx2, sse2, avx 10-bit h chroma deblock/loop filter
>   avcodec/h264: mmx2, sse2, avx 10-bit 4:2:2 h chroma deblock/loop
> filter
> 
>  libavcodec/x86/h264_deblock_10bit.asm | 157 
> ++
>  libavcodec/x86/h264dsp_init.c |  68 +--
>  2 files changed, 200 insertions(+), 25 deletions(-)
> 

If nobody raises objections or makes further comments I will push these,
including the AVX, later today.

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel


[FFmpeg-devel] [PATCH 0/4] More H.264 assembly (the sequel) [version 2]

2016-12-05 Thread James Darnley
Fixed the problem Michael highlighted.  Dropped the intra functions until it
becomes clear why their performance is unexpected. Updated the benchmarks with
results from a Nehalem and used (slightly) more accurate data.

Regarding the age of MMX:  I have written it so unless someone tells me to
remove it I will keep the code.  However, I will probably not write any more
going forward.

James Darnley (4):
  avcodec/h264: clean up and expand x86 function definitions
  whitespace changes after last commit
  avcodec/h264: mmx2, sse2, avx 10-bit h chroma deblock/loop filter
  avcodec/h264: mmx2, sse2, avx 10-bit 4:2:2 h chroma deblock/loop
filter

 libavcodec/x86/h264_deblock_10bit.asm | 157 ++
 libavcodec/x86/h264dsp_init.c |  68 +--
 2 files changed, 200 insertions(+), 25 deletions(-)

-- 
2.10.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel