[FFmpeg-devel] x86: float dsp: SSE2 butterflies

[Michael Niedermayer]

On Mon, Apr 15, 2013 at 09:06:14PM +0200, Christophe Gisquet wrote:
> Passes fate-aac (where it was mostly tested), -mp3, -imc, -twinvq and
> -wmaencode.
> 
> --
> Christophe

>  float_dsp.asm    |   24 ++++++++++++++++++++++++
>  float_dsp_init.c |    3 +++
>  2 files changed, 27 insertions(+)
> e92a4ed8833e823683cc2f5cece69851de9c82f9  0006-x86-float-dsp-butterflies_float-SSE.patch
> From ef5a1cb2aefac77592d71135991f9a8ee1e4379d Mon Sep 17 00:00:00 2001
> From: Christophe Gisquet <christophe.gisquet at gmail.com>
> Date: Fri, 12 Apr 2013 21:07:01 +0200
> Subject: [PATCH 6/8] x86: float dsp: butterflies_float SSE
> 
> 97c -> 49c
> Some codecs could benefit from more unrolling, but AAC doesn't.

patch applied

ive also replaced the 2 lea by 2 adds, which should be simpler and
a tiny bit faster. But iam happy to revert that if its not faster
for you

Thanks


[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Asymptotically faster algorithms should always be preferred if you have
asymptotical amounts of data
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20130417/971d499d/attachment.asc>