Re: [libav-devel] [PATCH 2/6] lavr: Add x86-optimized functions for s16 to flt conversion

2012-06-01 Thread Ronald S. Bultje
Hi, On Fri, Jun 1, 2012 at 12:08 PM, Justin Ruggles wrote: > --- >  libavresample/x86/audio_convert.asm    |   36 > >  libavresample/x86/audio_convert_init.c |    9 >  2 files changed, 45 insertions(+), 0 deletions(-) OK. Ronald ___

[libav-devel] [PATCH 2/6] lavr: Add x86-optimized functions for s16 to flt conversion

2012-06-01 Thread Justin Ruggles
--- libavresample/x86/audio_convert.asm| 36 libavresample/x86/audio_convert_init.c |9 2 files changed, 45 insertions(+), 0 deletions(-) diff --git a/libavresample/x86/audio_convert.asm b/libavresample/x86/audio_convert.asm index e2cfbf9..6e14

Re: [libav-devel] [PATCH 2/6] lavr: Add x86-optimized functions for s16 to flt conversion

2012-05-31 Thread Ronald S. Bultje
Hi, On Thu, May 31, 2012 at 6:58 PM, Justin Ruggles wrote: > --- >  libavresample/x86/audio_convert.asm    |   36 > >  libavresample/x86/audio_convert_init.c |    9 >  2 files changed, 45 insertions(+), 0 deletions(-) > > diff --git a/libavresample/x86/a

[libav-devel] [PATCH 2/6] lavr: Add x86-optimized functions for s16 to flt conversion

2012-05-31 Thread Justin Ruggles
--- libavresample/x86/audio_convert.asm| 36 libavresample/x86/audio_convert_init.c |9 2 files changed, 45 insertions(+), 0 deletions(-) diff --git a/libavresample/x86/audio_convert.asm b/libavresample/x86/audio_convert.asm index e2cfbf9..746a

Re: [libav-devel] [PATCH 2/6] lavr: Add x86-optimized functions for s16 to flt conversion

2012-05-02 Thread Justin Ruggles
On 05/01/2012 05:21 PM, Jason Garrett-Glaser wrote: > +%if cpuflag(sse4) > +pmovsxwd m0, [srcq+lenq ] > +pmovsxwd m1, [srcq+lenq+mmsize/2] > +%else > +mova m1, [srcq+lenq] > > Did you forget to load m0 in the non-sse4 version? No, it doesn't matter what's in m0. It

Re: [libav-devel] [PATCH 2/6] lavr: Add x86-optimized functions for s16 to flt conversion

2012-05-01 Thread Ronald S. Bultje
Hi, On Tue, May 1, 2012 at 2:21 PM, Jason Garrett-Glaser wrote: > +%if cpuflag(sse4) > +    pmovsxwd   m0, [srcq+lenq         ] > +    pmovsxwd   m1, [srcq+lenq+mmsize/2] > +%else > +    mova       m1, [srcq+lenq] > > Did you forget to load m0 in the non-sse4 version? punpcklwd m0, m1 Justin m

Re: [libav-devel] [PATCH 2/6] lavr: Add x86-optimized functions for s16 to flt conversion

2012-05-01 Thread Jason Garrett-Glaser
+%if cpuflag(sse4) +pmovsxwd m0, [srcq+lenq ] +pmovsxwd m1, [srcq+lenq+mmsize/2] +%else +mova m1, [srcq+lenq] Did you forget to load m0 in the non-sse4 version? Jason ___ libav-devel mailing list libav-devel@libav.org https

[libav-devel] [PATCH 2/6] lavr: Add x86-optimized functions for s16 to flt conversion

2012-05-01 Thread Justin Ruggles
--- Athlon64 C - 12412 SSE2 - 3706 SandyBridge C - 16440 SSE2 - 3116 SSE4 - 1002 libavresample/x86/audio_convert.asm| 43 libavresample/x86/audio_convert_init.c |9 ++ 2 files changed, 52 insertions(+), 0 deletions(-) diff --git a/libavres