aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/snowdata.h
diff options
context:
space:
mode:
authorChristophe Gisquet <christophe.gisquet@gmail.com>2013-04-09 22:16:36 +0200
committerMichael Niedermayer <michaelni@gmx.at>2013-04-19 13:19:45 +0200
commit76c7277385120741914168d02b57a7a1acf87a9b (patch)
tree1a0821b6a1b2aae976515ed3423a05ac9cd245f5 /libavcodec/snowdata.h
parent380cfce2b2138a0513f7c054134458b3b4c92fd4 (diff)
downloadffmpeg-76c7277385120741914168d02b57a7a1acf87a9b.tar.gz
x86: sbrdsp: implement SSE2 hf_apply_noise
233 to 105 cycles on Arrandale and Win64. Replacing the multiplication by s_m[m] by a pand and a pxor with appropriate vectors is slower. Unrolling is a 15 cycles win. A SSE version was 4 cycles slower. Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Diffstat (limited to 'libavcodec/snowdata.h')
0 files changed, 0 insertions, 0 deletions