diff options
| author | Ronald S. Bultje <rsbultje@gmail.com> | 2010-07-31 23:13:15 +0000 |
|---|---|---|
| committer | Ronald S. Bultje <rsbultje@gmail.com> | 2010-07-31 23:13:15 +0000 |
| commit | 6341838f3ca69c7850aa11b067165ef544cead95 (patch) | |
| tree | 7914c26ff9b26b9f544024e56b5032254d27f2b9 /libavcodec/x86/vp8dsp-init.c | |
| parent | ace7f813cd4b2bc092bd827f7e8257368781e9bb (diff) | |
| download | ffmpeg-6341838f3ca69c7850aa11b067165ef544cead95.tar.gz | |
Use word-writing instead of dword-writing (with two cached but otherwise
unchanged bytes) in the horizontal simple loopfilter. This makes the filter
quite a bit faster in itself (~30 cycles less on Core1), probably mostly
because we don't need a complex 4x4 transpose, but only a simple byte
interleave. Also allows using pextrw on SSE4, which speeds up even more
(e.g. 25% faster on Core i7).
Originally committed as revision 24638 to svn://svn.ffmpeg.org/ffmpeg/trunk
Diffstat (limited to 'libavcodec/x86/vp8dsp-init.c')
| -rw-r--r-- | libavcodec/x86/vp8dsp-init.c | 4 |
1 files changed, 3 insertions, 1 deletions
diff --git a/libavcodec/x86/vp8dsp-init.c b/libavcodec/x86/vp8dsp-init.c index 4f40bdd11f..4bf49364e7 100644 --- a/libavcodec/x86/vp8dsp-init.c +++ b/libavcodec/x86/vp8dsp-init.c @@ -346,7 +346,6 @@ av_cold void ff_vp8dsp_init_x86(VP8DSPContext* c) VP8_BILINEAR_MC_FUNC(1, 8, sse2); c->vp8_v_loop_filter_simple = ff_vp8_v_loop_filter_simple_sse2; - c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_sse2; c->vp8_v_loop_filter16y_inner = ff_vp8_v_loop_filter16y_inner_sse2; c->vp8_v_loop_filter8uv_inner = ff_vp8_v_loop_filter8uv_inner_sse2; @@ -358,6 +357,8 @@ av_cold void ff_vp8dsp_init_x86(VP8DSPContext* c) if (mm_flags & FF_MM_SSE2) { c->vp8_idct_dc_add4y = ff_vp8_idct_dc_add4y_sse2; + c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_sse2; + c->vp8_h_loop_filter16y_inner = ff_vp8_h_loop_filter16y_inner_sse2; c->vp8_h_loop_filter8uv_inner = ff_vp8_h_loop_filter8uv_inner_sse2; @@ -390,6 +391,7 @@ av_cold void ff_vp8dsp_init_x86(VP8DSPContext* c) if (mm_flags & FF_MM_SSE4) { c->vp8_idct_dc_add = ff_vp8_idct_dc_add_sse4; + c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_sse4; c->vp8_h_loop_filter16y = ff_vp8_h_loop_filter16y_mbedge_sse4; c->vp8_h_loop_filter8uv = ff_vp8_h_loop_filter8uv_mbedge_sse4; } |