diff options
| author | levytamar82 <levytamar82@gmail.com> | 2013-11-21 15:49:29 -0700 |
|---|---|---|
| committer | levytamar82 <levytamar82@gmail.com> | 2014-01-09 12:27:51 -0700 |
| commit | 511d218c60b9b6c1ab9383db746815e907af0359 (patch) | |
| tree | a7cbf64477adac2433384293d88d08f27c373fec /vp9/vp9_common.mk | |
| parent | a622ed554f7072268e4c8d0b8f26d2e8865c2b3b (diff) | |
| download | libvpx-511d218c60b9b6c1ab9383db746815e907af0359.tar libvpx-511d218c60b9b6c1ab9383db746815e907af0359.tar.gz libvpx-511d218c60b9b6c1ab9383db746815e907af0359.tar.bz2 libvpx-511d218c60b9b6c1ab9383db746815e907af0359.zip | |
SSSE3 convolution optimization
Optimizing all SSSE3 assembly for convolution:
1. vp9_filter_block1d4_h8_sse2
2. vp9_filter_block1d8_h8_sse2
3. vp9_filter_block1d16_h8_sse2
4. vp9_filter_block1d4_v8_sse2
5. vp9_filter_block1d8_v8_sse2
6. vp9_filter_block1d16_v8_sse2
my optimization include:
-processing 2x8 elements in one 128 bit register instead of processing
8 elements in one 128 bit register.
-removing unecessary loads.
This optimization gives between 2.4% user level gain for 480p input
and 1.6% user level gain for 720p.
This Optimization done only for 64bit.
Change-Id: Icb586dc0c938b56699864fcee6c52fd43b36b969
Diffstat (limited to 'vp9/vp9_common.mk')
| -rw-r--r-- | vp9/vp9_common.mk | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/vp9/vp9_common.mk b/vp9/vp9_common.mk index eefbd1ac9..8a8b63d97 100644 --- a/vp9/vp9_common.mk +++ b/vp9/vp9_common.mk @@ -77,6 +77,7 @@ VP9_COMMON_SRCS-$(CONFIG_VP9_POSTPROC) += common/vp9_postproc.c VP9_COMMON_SRCS-$(HAVE_MMX) += common/x86/vp9_loopfilter_mmx.asm VP9_COMMON_SRCS-$(HAVE_SSE2) += common/x86/vp9_subpixel_8t_sse2.asm VP9_COMMON_SRCS-$(HAVE_SSSE3) += common/x86/vp9_subpixel_8t_ssse3.asm +VP9_COMMON_SRCS-$(HAVE_SSSE3) += common/x86/vp9_subpixel_8t_intrin_ssse3.c ifeq ($(CONFIG_VP9_POSTPROC),yes) VP9_COMMON_SRCS-$(HAVE_MMX) += common/x86/vp9_postproc_mmx.asm VP9_COMMON_SRCS-$(HAVE_SSE2) += common/x86/vp9_postproc_sse2.asm |