[FFmpeg-devel] [PATCH] SSE3/4 implementation of flac_encode_residual_lpc

Bobby Bingham uhmmmm
Fri May 29 06:13:21 CEST 2009


On Sat, 23 May 2009 13:49:47 -0400
Jason Garrett-Glaser <darkshikari at gmail.com> wrote:

> 
> > "movlhps    %%xmm3,  %%xmm5         \n\t"
> > "movhlps    %%xmm4,  %%xmm5         \n\t"
> 
> Have you tried replacing this by movdqa/shufpd?  That should have one
> less latency.  This is actually the first place I've ever seen shufpd
> be potentially useful.
> 

As discussed on IRC, this is slower on Athlon 64.

Attached is a version I hope is about ready for inclusion.  Provides an
overall encoding speedup of ~30% at compression_level=12.

-- 
Bobby Bingham
??????????????????????
-------------- next part --------------
A non-text attachment was scrubbed...
Name: flac_sse2.patch
Type: text/x-patch
Size: 11924 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090528/422a2682/attachment.bin>



More information about the ffmpeg-devel mailing list