Re: [libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-08-23 Thread Justin Ruggles
On 08/21/2012 03:53 PM, Ronald S. Bultje wrote: Hi, On Mon, Aug 6, 2012 at 9:22 AM, Justin Ruggles justin.rugg...@gmail.com wrote: +%if cpuflag(sse2) +mulps m0, m6, [srcq ] +mulps m1, m6, [srcq+src1q] +mulps m2, m6, [srcq+src2q] +mulps

Re: [libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-08-23 Thread Ronald S. Bultje
Hi, On Thu, Aug 23, 2012 at 12:37 PM, Justin Ruggles justin.rugg...@gmail.com wrote: On 08/21/2012 03:53 PM, Ronald S. Bultje wrote: On Mon, Aug 6, 2012 at 9:22 AM, Justin Ruggles justin.rugg...@gmail.com wrote: +%if cpuflag(sse2) +mulps m0, m6, [srcq ] +mulps

Re: [libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-08-23 Thread Justin Ruggles
On 08/23/2012 04:49 PM, Ronald S. Bultje wrote: Hi, On Thu, Aug 23, 2012 at 12:37 PM, Justin Ruggles justin.rugg...@gmail.com wrote: On 08/21/2012 03:53 PM, Ronald S. Bultje wrote: On Mon, Aug 6, 2012 at 9:22 AM, Justin Ruggles justin.rugg...@gmail.com wrote: +%if cpuflag(sse2) +

Re: [libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-08-23 Thread Justin Ruggles
On 08/23/2012 05:00 PM, Justin Ruggles wrote: On 08/23/2012 04:49 PM, Ronald S. Bultje wrote: Hi, On Thu, Aug 23, 2012 at 12:37 PM, Justin Ruggles justin.rugg...@gmail.com wrote: On 08/21/2012 03:53 PM, Ronald S. Bultje wrote: On Mon, Aug 6, 2012 at 9:22 AM, Justin Ruggles

Re: [libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-08-23 Thread Ronald S. Bultje
Hi, On Thu, Aug 23, 2012 at 3:02 PM, Justin Ruggles justin.rugg...@gmail.com wrote: For future use... the best I can come up with is 6x vpperm. m3 = 0, 1, x, 3, 4, x, 6, 7 m4 = x, 9, 10, x, 12, 13, x, 15 m5 = 16, 17, x, 19, 20, x, 22, 23 Then 3x vpperm to fill-in the gaps.

Re: [libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-08-21 Thread Ronald S. Bultje
Hi, On Mon, Aug 6, 2012 at 9:22 AM, Justin Ruggles justin.rugg...@gmail.com wrote: +%if cpuflag(sse2) +mulps m0, m6, [srcq ] +mulps m1, m6, [srcq+src1q] +mulps m2, m6, [srcq+src2q] +mulps m3, m6, [srcq+src3q] +mulps m4, m6,

Re: [libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-08-06 Thread Loren Merritt
On Mon, 6 Aug 2012, Justin Ruggles wrote: --- a/libavresample/x86/audio_convert.asm +++ b/libavresample/x86/audio_convert.asm @@ -581,6 +581,120 @@ CONV_FLTP_TO_S16_2CH INIT_XMM ssse3 CONV_FLTP_TO_S16_2CH +;--

[libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-08-06 Thread Justin Ruggles
--- libavresample/x86/audio_convert.asm| 114 libavresample/x86/audio_convert_init.c | 15 2 files changed, 129 insertions(+), 0 deletions(-) diff --git a/libavresample/x86/audio_convert.asm b/libavresample/x86/audio_convert.asm index

[libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-08-05 Thread Justin Ruggles
--- libavresample/x86/audio_convert.asm| 114 libavresample/x86/audio_convert_init.c | 15 2 files changed, 129 insertions(+), 0 deletions(-) diff --git a/libavresample/x86/audio_convert.asm b/libavresample/x86/audio_convert.asm index

Re: [libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-07-24 Thread Ronald S. Bultje
Hi, On Sat, Jul 14, 2012 at 9:29 PM, Justin Ruggles justin.rugg...@gmail.com wrote: +%else ; sse +mova xmm0, [srcq ] +mova xmm1, [srcq+src1q] +mova xmm2, [srcq+src2q] +mova xmm3, [srcq+src3q] +mova xmm4, [srcq+src4q] +mova xmm5,

Re: [libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-07-16 Thread Loren Merritt
On Sun, 15 Jul 2012, Justin Ruggles wrote: +%else ; sse +mova xmm0, [srcq ] +mova xmm1, [srcq+src1q] +mova xmm2, [srcq+src2q] +mova xmm3, [srcq+src3q] +mova xmm4, [srcq+src4q] +mova xmm5, [srcq+src5q] movq. And your code should

Re: [libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-07-16 Thread Justin Ruggles
On 07/16/2012 08:28 AM, Loren Merritt wrote: On Sun, 15 Jul 2012, Justin Ruggles wrote: +%else ; sse +mova xmm0, [srcq ] +mova xmm1, [srcq+src1q] +mova xmm2, [srcq+src2q] +mova xmm3, [srcq+src3q] +mova xmm4, [srcq+src4q] +mova

[libav-devel] [PATCH 06/15] lavr: x86: optimized 6-channel fltp to s16 conversion

2012-07-14 Thread Justin Ruggles
--- libavresample/x86/audio_convert.asm| 118 libavresample/x86/audio_convert_init.c | 15 2 files changed, 133 insertions(+), 0 deletions(-) diff --git a/libavresample/x86/audio_convert.asm b/libavresample/x86/audio_convert.asm index