Re: [FFmpeg-devel] [PATCH 3/4] x86: lossless audio: SSE4 madd 32bits

2016-05-07 Thread Michael Niedermayer
On Sat, May 07, 2016 at 05:58:21PM +0200, Paul B Mahol wrote: > On 5/1/16, Christophe Gisquet wrote: > > The unique user so far is wmalossless 24bits. The few samples tested show an > > order of 8, so more unrolling or an avx2 version do not make sense. > > > > Timings: 68 -> 49 cycles > > --- > >

Re: [FFmpeg-devel] [PATCH 3/4] x86: lossless audio: SSE4 madd 32bits

2016-05-07 Thread Paul B Mahol
On 5/1/16, Christophe Gisquet wrote: > The unique user so far is wmalossless 24bits. The few samples tested show an > order of 8, so more unrolling or an avx2 version do not make sense. > > Timings: 68 -> 49 cycles > --- > libavcodec/x86/lossless_audiodsp.asm| 33 > +++

[FFmpeg-devel] [PATCH 3/4] x86: lossless audio: SSE4 madd 32bits

2016-05-01 Thread Christophe Gisquet
The unique user so far is wmalossless 24bits. The few samples tested show an order of 8, so more unrolling or an avx2 version do not make sense. Timings: 68 -> 49 cycles --- libavcodec/x86/lossless_audiodsp.asm| 33 + libavcodec/x86/lossless_audiodsp_init.c |