aom_dsp · 55f1ef00c1c76c3bbb5481f69f8981db6fdd6e39 · Xiph.Org / aom-rav1e

Highbd parallel_deblocking sse2 optimization

Yi Luo authored 7 years ago

- Decoder speed improves ~13.7% (baseline + parallel_deblocking).
- Highbd loopfilter AVX2 version works when this experiment is
  disabled.

Change-Id: I5d56b137a1d52236a4735656c370d57ef71ae043

6f5569f3

History

6f5569f3 7 years ago

History

Name	Last commit	Last update
..
arm
mips
simd
x86
add_noise.c
ans.h
ansreader.h
answriter.h
aom_convolve.c
aom_convolve.h
aom_dsp.cmake
aom_dsp.mk
aom_dsp_common.h
aom_dsp_rtcd.c
aom_dsp_rtcd_defs.pl
aom_filter.h
aom_simd.h
aom_simd_inline.h
avg.c
binary_codes_reader.c
binary_codes_reader.h
binary_codes_writer.c
binary_codes_writer.h
bitreader.h
bitreader_buffer.c
bitreader_buffer.h
bitwriter.h
bitwriter_buffer.c
bitwriter_buffer.h
blend.h
blend_a64_hmask.c
blend_a64_mask.c
blend_a64_vmask.c
buf_ans.c
buf_ans.h
daalaboolreader.c
daalaboolreader.h
daalaboolwriter.c
daalaboolwriter.h
entcode.c
entcode.h
entdec.c
entdec.h
entenc.c
entenc.h
fastssim.c
fwd_txfm.c
fwd_txfm.h
intrapred.c
inv_txfm.c
inv_txfm.h
loopfilter.c
postproc.h
prob.c
prob.h
psnr.c
psnr.h
psnrhvs.c
quantize.c
quantize.h
sad.c
ssim.c
ssim.h
subtract.c
sum_squares.c
txfm_common.h
variance.c
variance.h