|
|
|
|
# subsystems
|
|
|
|
|
OBJS-$(CONFIG_FFT) += aarch64/fft_init_aarch64.o
|
|
|
|
|
OBJS-$(CONFIG_FMTCONVERT) += aarch64/fmtconvert_init.o
|
|
|
|
|
OBJS-$(CONFIG_H264CHROMA) += aarch64/h264chroma_init_aarch64.o
|
|
|
|
|
OBJS-$(CONFIG_H264DSP) += aarch64/h264dsp_init_aarch64.o
|
|
|
|
|
OBJS-$(CONFIG_H264PRED) += aarch64/h264pred_init.o
|
|
|
|
|
OBJS-$(CONFIG_H264QPEL) += aarch64/h264qpel_init_aarch64.o
|
|
|
|
|
OBJS-$(CONFIG_HPELDSP) += aarch64/hpeldsp_init_aarch64.o
|
|
|
|
|
OBJS-$(CONFIG_IDCTDSP) += aarch64/idctdsp_init_aarch64.o
|
|
|
|
|
OBJS-$(CONFIG_MPEGAUDIODSP) += aarch64/mpegaudiodsp_init.o
|
|
|
|
|
OBJS-$(CONFIG_NEON_CLOBBER_TEST) += aarch64/neontest.o
|
|
|
|
|
OBJS-$(CONFIG_VIDEODSP) += aarch64/videodsp_init.o
|
|
|
|
|
OBJS-$(CONFIG_VP8DSP) += aarch64/vp8dsp_init_aarch64.o
|
|
|
|
|
|
|
|
|
|
# decoders/encoders
|
|
|
|
|
OBJS-$(CONFIG_AAC_DECODER) += aarch64/aacpsdsp_init_aarch64.o \
|
|
|
|
|
aarch64/sbrdsp_init_aarch64.o
|
|
|
|
|
OBJS-$(CONFIG_DCA_DECODER) += aarch64/synth_filter_init.o
|
aarch64/opusdsp: implement NEON accelerated postfilter and deemphasis
153372 UNITS in postfilter_c, 65536 runs, 0 skips
73164 UNITS in postfilter_neon, 65536 runs, 0 skips -> 2.1x speedup
80591 UNITS in deemphasis_c, 131072 runs, 0 skips
43969 UNITS in deemphasis_neon, 131072 runs, 0 skips -> 1.83x speedup
Total decoder speedup: ~15% on a Raspberry Pi 3 (from 28.1x to 33.5x realtime)
Deemphasis SIMD based on the following unrolling:
const float c1 = CELT_EMPH_COEFF, c2 = c1*c1, c3 = c2*c1, c4 = c3*c1;
float state = coeff;
for (int i = 0; i < len; i += 4) {
y[0] = x[0] + c1*state;
y[1] = x[1] + c2*state + c1*x[0];
y[2] = x[2] + c3*state + c1*x[1] + c2*x[0];
y[3] = x[3] + c4*state + c1*x[2] + c2*x[1] + c3*x[0];
state = y[3];
y += 4;
x += 4;
}
Unlike the x86 version, duplication is used instead of pslldq so
the structure and tables are different.
6 years ago
|
|
|
OBJS-$(CONFIG_OPUS_DECODER) += aarch64/opusdsp_init.o
|
|
|
|
|
OBJS-$(CONFIG_RV40_DECODER) += aarch64/rv40dsp_init_aarch64.o
|
|
|
|
|
OBJS-$(CONFIG_VC1DSP) += aarch64/vc1dsp_init_aarch64.o
|
|
|
|
|
OBJS-$(CONFIG_VORBIS_DECODER) += aarch64/vorbisdsp_init.o
|
|
|
|
|
OBJS-$(CONFIG_VP9_DECODER) += aarch64/vp9dsp_init_10bpp_aarch64.o \
|
|
|
|
|
aarch64/vp9dsp_init_12bpp_aarch64.o \
|
|
|
|
|
aarch64/vp9mc_aarch64.o \
|
|
|
|
|
aarch64/vp9dsp_init_aarch64.o
|
|
|
|
|
|
|
|
|
|
# ARMv8 optimizations
|
|
|
|
|
|
|
|
|
|
# subsystems
|
|
|
|
|
ARMV8-OBJS-$(CONFIG_VIDEODSP) += aarch64/videodsp.o
|
|
|
|
|
|
|
|
|
|
# NEON optimizations
|
|
|
|
|
|
|
|
|
|
# subsystems
|
|
|
|
|
NEON-OBJS-$(CONFIG_AAC_DECODER) += aarch64/sbrdsp_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_FFT) += aarch64/fft_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_FMTCONVERT) += aarch64/fmtconvert_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_H264CHROMA) += aarch64/h264cmc_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_H264DSP) += aarch64/h264dsp_neon.o \
|
|
|
|
|
aarch64/h264idct_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_H264PRED) += aarch64/h264pred_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_H264QPEL) += aarch64/h264qpel_neon.o \
|
|
|
|
|
aarch64/hpeldsp_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_HPELDSP) += aarch64/hpeldsp_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_IDCTDSP) += aarch64/simple_idct_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_MDCT) += aarch64/mdct_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_MPEGAUDIODSP) += aarch64/mpegaudiodsp_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_VP8DSP) += aarch64/vp8dsp_neon.o
|
|
|
|
|
|
|
|
|
|
# decoders/encoders
|
|
|
|
|
NEON-OBJS-$(CONFIG_AAC_DECODER) += aarch64/aacpsdsp_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_DCA_DECODER) += aarch64/synth_filter_neon.o
|
aarch64/opusdsp: implement NEON accelerated postfilter and deemphasis
153372 UNITS in postfilter_c, 65536 runs, 0 skips
73164 UNITS in postfilter_neon, 65536 runs, 0 skips -> 2.1x speedup
80591 UNITS in deemphasis_c, 131072 runs, 0 skips
43969 UNITS in deemphasis_neon, 131072 runs, 0 skips -> 1.83x speedup
Total decoder speedup: ~15% on a Raspberry Pi 3 (from 28.1x to 33.5x realtime)
Deemphasis SIMD based on the following unrolling:
const float c1 = CELT_EMPH_COEFF, c2 = c1*c1, c3 = c2*c1, c4 = c3*c1;
float state = coeff;
for (int i = 0; i < len; i += 4) {
y[0] = x[0] + c1*state;
y[1] = x[1] + c2*state + c1*x[0];
y[2] = x[2] + c3*state + c1*x[1] + c2*x[0];
y[3] = x[3] + c4*state + c1*x[2] + c2*x[1] + c3*x[0];
state = y[3];
y += 4;
x += 4;
}
Unlike the x86 version, duplication is used instead of pslldq so
the structure and tables are different.
6 years ago
|
|
|
NEON-OBJS-$(CONFIG_OPUS_DECODER) += aarch64/opusdsp_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_VORBIS_DECODER) += aarch64/vorbisdsp_neon.o
|
|
|
|
|
NEON-OBJS-$(CONFIG_VP9_DECODER) += aarch64/vp9itxfm_16bpp_neon.o \
|
|
|
|
|
aarch64/vp9itxfm_neon.o \
|
|
|
|
|
aarch64/vp9lpf_16bpp_neon.o \
|
aarch64: vp9: Implement NEON loop filters
This work is sponsored by, and copyright, Google.
These are ported from the ARM version; thanks to the larger
amount of registers available, we can do the loop filters with
16 pixels at a time. The implementation is fully templated, with
a single macro which can generate versions for both 8 and
16 pixels wide, for both 4, 8 and 16 pixels loop filters
(and the 4/8 mixed versions as well).
For the 8 pixel wide versions, it is pretty close in speed (the
v_4_8 and v_8_8 filters are the best examples of this; the h_4_8
and h_8_8 filters seem to get some gain in the load/transpose/store
part). For the 16 pixels wide ones, we get a speedup of around
1.2-1.4x compared to the 32 bit version.
Examples of runtimes vs the 32 bit version, on a Cortex A53:
ARM AArch64
vp9_loop_filter_h_4_8_neon: 144.0 127.2
vp9_loop_filter_h_8_8_neon: 207.0 182.5
vp9_loop_filter_h_16_8_neon: 415.0 328.7
vp9_loop_filter_h_16_16_neon: 672.0 558.6
vp9_loop_filter_mix2_h_44_16_neon: 302.0 203.5
vp9_loop_filter_mix2_h_48_16_neon: 365.0 305.2
vp9_loop_filter_mix2_h_84_16_neon: 365.0 305.2
vp9_loop_filter_mix2_h_88_16_neon: 376.0 305.2
vp9_loop_filter_mix2_v_44_16_neon: 193.2 128.2
vp9_loop_filter_mix2_v_48_16_neon: 246.7 218.4
vp9_loop_filter_mix2_v_84_16_neon: 248.0 218.5
vp9_loop_filter_mix2_v_88_16_neon: 302.0 218.2
vp9_loop_filter_v_4_8_neon: 89.0 88.7
vp9_loop_filter_v_8_8_neon: 141.0 137.7
vp9_loop_filter_v_16_8_neon: 295.0 272.7
vp9_loop_filter_v_16_16_neon: 546.0 453.7
The speedup vs C code in checkasm tests is around 2-7x, which is
pretty much the same as for the 32 bit version. Even if these functions
are faster than their 32 bit equivalent, the C version that we compare
to also became around 1.3-1.7x faster than the C version in 32 bit.
Based on START_TIMER/STOP_TIMER wrapping around a few individual
functions, the speedup vs C code is around 4-5x.
Examples of runtimes vs C on a Cortex A57 (for a slightly older version
of the patch):
A57 gcc-5.3 neon
loop_filter_h_4_8_neon: 256.6 93.4
loop_filter_h_8_8_neon: 307.3 139.1
loop_filter_h_16_8_neon: 340.1 254.1
loop_filter_h_16_16_neon: 827.0 407.9
loop_filter_mix2_h_44_16_neon: 524.5 155.4
loop_filter_mix2_h_48_16_neon: 644.5 173.3
loop_filter_mix2_h_84_16_neon: 630.5 222.0
loop_filter_mix2_h_88_16_neon: 697.3 222.0
loop_filter_mix2_v_44_16_neon: 598.5 100.6
loop_filter_mix2_v_48_16_neon: 651.5 127.0
loop_filter_mix2_v_84_16_neon: 591.5 167.1
loop_filter_mix2_v_88_16_neon: 855.1 166.7
loop_filter_v_4_8_neon: 271.7 65.3
loop_filter_v_8_8_neon: 312.5 106.9
loop_filter_v_16_8_neon: 473.3 206.5
loop_filter_v_16_16_neon: 976.1 327.8
The speed-up compared to the C functions is 2.5 to 6 and the cortex-a57
is again 30-50% faster than the cortex-a53.
This is an adapted cherry-pick from libav commits
9d2afd1eb8c5cc0633062430e66326dbf98c99e0 and
31756abe29eb039a11c59a42cb12e0cc2aef3b97.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
8 years ago
|
|
|
aarch64/vp9lpf_neon.o \
|
|
|
|
|
aarch64/vp9mc_16bpp_neon.o \
|
|
|
|
|
aarch64/vp9mc_neon.o
|
