mirror of
https://git.ffmpeg.org/ffmpeg.git
synced 2024-12-24 08:12:44 +00:00
1e5d87eec3
This work is sponsored by, and copyright, Google.
This is pretty much similar to the 8 bpp version, but in some senses
simpler. All input pixels are 16 bits, and all intermediates also fit
in 16 bits, so there's no lengthening/narrowing in the filter at all.
For the full 16 pixel wide filter, we can only process 4 pixels at a time
(using an implementation very much similar to the one for 8 bpp),
but we can do 8 pixels at a time for the 4 and 8 pixel wide filters with
a different implementation of the core filter.
Examples of relative speedup compared to the C version, from checkasm:
Cortex A7 A8 A9 A53
vp9_loop_filter_h_4_8_10bpp_neon: 1.83 2.16 1.40 2.09
vp9_loop_filter_h_8_8_10bpp_neon: 1.39 1.67 1.24 1.70
vp9_loop_filter_h_16_8_10bpp_neon: 1.56 1.47 1.10 1.81
vp9_loop_filter_h_16_16_10bpp_neon: 1.94 1.69 1.33 2.24
vp9_loop_filter_mix2_h_44_16_10bpp_neon: 2.01 2.27 1.67 2.39
vp9_loop_filter_mix2_h_48_16_10bpp_neon: 1.84 2.06 1.45 2.19
vp9_loop_filter_mix2_h_84_16_10bpp_neon: 1.89 2.20 1.47 2.29
vp9_loop_filter_mix2_h_88_16_10bpp_neon: 1.69 2.12 1.47 2.08
vp9_loop_filter_mix2_v_44_16_10bpp_neon: 3.16 3.98 2.50 4.05
vp9_loop_filter_mix2_v_48_16_10bpp_neon: 2.84 3.64 2.25 3.77
vp9_loop_filter_mix2_v_84_16_10bpp_neon: 2.65 3.45 2.16 3.54
vp9_loop_filter_mix2_v_88_16_10bpp_neon: 2.55 3.30 2.16 3.55
vp9_loop_filter_v_4_8_10bpp_neon: 2.85 3.97 2.24 3.68
vp9_loop_filter_v_8_8_10bpp_neon: 2.27 3.19 1.96 3.08
vp9_loop_filter_v_16_8_10bpp_neon: 3.42 2.74 2.26 4.40
vp9_loop_filter_v_16_16_10bpp_neon: 2.86 2.44 1.93 3.88
The speedup vs C code measured in checkasm is around 1.1-4x.
These numbers are quite inconclusive though, since the checkasm test
runs multiple filterings on top of each other, so later rounds might
end up with different codepaths (different decisions on which filter
to apply, based on input pixel differences).
Based on START_TIMER/STOP_TIMER wrapping around a few individual
functions, the speedup vs C code is around 2-4x.
Signed-off-by: Martin Storsjö <martin@martin.st>
151 lines
8.1 KiB
Makefile
151 lines
8.1 KiB
Makefile
ARCH_HEADERS = mathops.h
|
|
|
|
# subsystems
|
|
OBJS-$(CONFIG_AC3DSP) += arm/ac3dsp_init_arm.o \
|
|
arm/ac3dsp_arm.o
|
|
OBJS-$(CONFIG_AUDIODSP) += arm/audiodsp_init_arm.o
|
|
OBJS-$(CONFIG_BLOCKDSP) += arm/blockdsp_init_arm.o
|
|
OBJS-$(CONFIG_FFT) += arm/fft_init_arm.o \
|
|
arm/fft_fixed_init_arm.o
|
|
OBJS-$(CONFIG_FLACDSP) += arm/flacdsp_init_arm.o \
|
|
arm/flacdsp_arm.o
|
|
OBJS-$(CONFIG_FMTCONVERT) += arm/fmtconvert_init_arm.o
|
|
OBJS-$(CONFIG_G722DSP) += arm/g722dsp_init_arm.o
|
|
OBJS-$(CONFIG_H264CHROMA) += arm/h264chroma_init_arm.o
|
|
OBJS-$(CONFIG_H264DSP) += arm/h264dsp_init_arm.o
|
|
OBJS-$(CONFIG_H264PRED) += arm/h264pred_init_arm.o
|
|
OBJS-$(CONFIG_H264QPEL) += arm/h264qpel_init_arm.o
|
|
OBJS-$(CONFIG_HPELDSP) += arm/hpeldsp_init_arm.o \
|
|
arm/hpeldsp_arm.o
|
|
OBJS-$(CONFIG_IDCTDSP) += arm/idctdsp_init_arm.o \
|
|
arm/idctdsp_arm.o \
|
|
arm/jrevdct_arm.o \
|
|
arm/simple_idct_arm.o
|
|
OBJS-$(CONFIG_LLAUDDSP) += arm/lossless_audiodsp_init_arm.o
|
|
OBJS-$(CONFIG_ME_CMP) += arm/me_cmp_init_arm.o
|
|
OBJS-$(CONFIG_MPEGAUDIODSP) += arm/mpegaudiodsp_init_arm.o
|
|
OBJS-$(CONFIG_MPEGVIDEO) += arm/mpegvideo_arm.o
|
|
OBJS-$(CONFIG_MPEGVIDEOENC) += arm/mpegvideoencdsp_init_arm.o
|
|
OBJS-$(CONFIG_NEON_CLOBBER_TEST) += arm/neontest.o
|
|
OBJS-$(CONFIG_PIXBLOCKDSP) += arm/pixblockdsp_init_arm.o
|
|
OBJS-$(CONFIG_RDFT) += arm/rdft_init_arm.o
|
|
OBJS-$(CONFIG_RV34DSP) += arm/rv34dsp_init_arm.o
|
|
OBJS-$(CONFIG_VC1DSP) += arm/vc1dsp_init_arm.o
|
|
OBJS-$(CONFIG_VIDEODSP) += arm/videodsp_init_arm.o
|
|
OBJS-$(CONFIG_VP3DSP) += arm/vp3dsp_init_arm.o
|
|
OBJS-$(CONFIG_VP8DSP) += arm/vp8dsp_init_arm.o
|
|
|
|
# decoders/encoders
|
|
OBJS-$(CONFIG_AAC_DECODER) += arm/aacpsdsp_init_arm.o \
|
|
arm/sbrdsp_init_arm.o
|
|
OBJS-$(CONFIG_DCA_DECODER) += arm/synth_filter_init_arm.o
|
|
OBJS-$(CONFIG_HEVC_DECODER) += arm/hevcdsp_init_arm.o
|
|
OBJS-$(CONFIG_MLP_DECODER) += arm/mlpdsp_init_arm.o
|
|
OBJS-$(CONFIG_RV40_DECODER) += arm/rv40dsp_init_arm.o
|
|
OBJS-$(CONFIG_VORBIS_DECODER) += arm/vorbisdsp_init_arm.o
|
|
OBJS-$(CONFIG_VP6_DECODER) += arm/vp6dsp_init_arm.o
|
|
OBJS-$(CONFIG_VP9_DECODER) += arm/vp9dsp_init_10bpp_arm.o \
|
|
arm/vp9dsp_init_12bpp_arm.o \
|
|
arm/vp9dsp_init_arm.o
|
|
|
|
|
|
# ARMv5 optimizations
|
|
# subsystems
|
|
ARMV5TE-OBJS-$(CONFIG_IDCTDSP) += arm/idctdsp_init_armv5te.o \
|
|
arm/simple_idct_armv5te.o
|
|
ARMV5TE-OBJS-$(CONFIG_MPEGVIDEO) += arm/mpegvideo_armv5te.o \
|
|
arm/mpegvideo_armv5te_s.o
|
|
ARMV5TE-OBJS-$(CONFIG_VIDEODSP) += arm/videodsp_init_armv5te.o \
|
|
arm/videodsp_armv5te.o
|
|
|
|
# decoders/encoders
|
|
ARMV5TE-OBJS-$(CONFIG_MLP_DECODER) += arm/mlpdsp_armv5te.o
|
|
|
|
|
|
# ARMv6 optimizations
|
|
# subsystems
|
|
ARMV6-OBJS-$(CONFIG_AC3DSP) += arm/ac3dsp_armv6.o
|
|
ARMV6-OBJS-$(CONFIG_HPELDSP) += arm/hpeldsp_init_armv6.o \
|
|
arm/hpeldsp_armv6.o
|
|
ARMV6-OBJS-$(CONFIG_IDCTDSP) += arm/idctdsp_init_armv6.o \
|
|
arm/idctdsp_armv6.o \
|
|
arm/simple_idct_armv6.o
|
|
ARMV6-OBJS-$(CONFIG_ME_CMP) += arm/me_cmp_armv6.o
|
|
ARMV6-OBJS-$(CONFIG_MPEGAUDIODSP) += arm/mpegaudiodsp_fixed_armv6.o
|
|
ARMV6-OBJS-$(CONFIG_MPEGVIDEOENC) += arm/mpegvideoencdsp_armv6.o
|
|
ARMV6-OBJS-$(CONFIG_PIXBLOCKDSP) += arm/pixblockdsp_armv6.o
|
|
ARMV6-OBJS-$(CONFIG_STARTCODE) += arm/startcode_armv6.o
|
|
ARMV6-OBJS-$(CONFIG_VP8DSP) += arm/vp8_armv6.o \
|
|
arm/vp8dsp_init_armv6.o \
|
|
arm/vp8dsp_armv6.o
|
|
|
|
# decoders/encoders
|
|
ARMV6-OBJS-$(CONFIG_MLP_DECODER) += arm/mlpdsp_armv6.o
|
|
|
|
|
|
# VFP optimizations
|
|
|
|
# subsystems
|
|
VFP-OBJS-$(CONFIG_FFT) += arm/fft_vfp.o
|
|
VFP-OBJS-$(CONFIG_FMTCONVERT) += arm/fmtconvert_vfp.o
|
|
VFP-OBJS-$(CONFIG_MDCT) += arm/mdct_vfp.o
|
|
|
|
# decoders/encoders
|
|
VFP-OBJS-$(CONFIG_DCA_DECODER) += arm/synth_filter_vfp.o
|
|
|
|
|
|
# NEON optimizations
|
|
|
|
# subsystems
|
|
NEON-OBJS-$(CONFIG_AC3DSP) += arm/ac3dsp_neon.o
|
|
NEON-OBJS-$(CONFIG_AUDIODSP) += arm/audiodsp_init_neon.o \
|
|
arm/audiodsp_neon.o \
|
|
arm/int_neon.o
|
|
NEON-OBJS-$(CONFIG_BLOCKDSP) += arm/blockdsp_init_neon.o \
|
|
arm/blockdsp_neon.o
|
|
NEON-OBJS-$(CONFIG_FFT) += arm/fft_neon.o \
|
|
arm/fft_fixed_neon.o
|
|
NEON-OBJS-$(CONFIG_FMTCONVERT) += arm/fmtconvert_neon.o
|
|
NEON-OBJS-$(CONFIG_G722DSP) += arm/g722dsp_neon.o
|
|
NEON-OBJS-$(CONFIG_H264CHROMA) += arm/h264cmc_neon.o
|
|
NEON-OBJS-$(CONFIG_H264DSP) += arm/h264dsp_neon.o \
|
|
arm/h264idct_neon.o
|
|
NEON-OBJS-$(CONFIG_H264PRED) += arm/h264pred_neon.o
|
|
NEON-OBJS-$(CONFIG_H264QPEL) += arm/h264qpel_neon.o \
|
|
arm/hpeldsp_neon.o
|
|
NEON-OBJS-$(CONFIG_HPELDSP) += arm/hpeldsp_init_neon.o \
|
|
arm/hpeldsp_neon.o
|
|
NEON-OBJS-$(CONFIG_IDCTDSP) += arm/idctdsp_init_neon.o \
|
|
arm/idctdsp_neon.o \
|
|
arm/simple_idct_neon.o
|
|
NEON-OBJS-$(CONFIG_MDCT) += arm/mdct_neon.o \
|
|
arm/mdct_fixed_neon.o
|
|
NEON-OBJS-$(CONFIG_MPEGVIDEO) += arm/mpegvideo_neon.o
|
|
NEON-OBJS-$(CONFIG_RDFT) += arm/rdft_neon.o
|
|
NEON-OBJS-$(CONFIG_VC1DSP) += arm/vc1dsp_init_neon.o \
|
|
arm/vc1dsp_neon.o
|
|
NEON-OBJS-$(CONFIG_VP3DSP) += arm/vp3dsp_neon.o
|
|
NEON-OBJS-$(CONFIG_VP8DSP) += arm/vp8dsp_init_neon.o \
|
|
arm/vp8dsp_neon.o
|
|
|
|
# decoders/encoders
|
|
NEON-OBJS-$(CONFIG_AAC_DECODER) += arm/aacpsdsp_neon.o \
|
|
arm/sbrdsp_neon.o
|
|
NEON-OBJS-$(CONFIG_LLAUDDSP) += arm/lossless_audiodsp_neon.o
|
|
NEON-OBJS-$(CONFIG_DCA_DECODER) += arm/synth_filter_neon.o
|
|
NEON-OBJS-$(CONFIG_HEVC_DECODER) += arm/hevcdsp_init_neon.o \
|
|
arm/hevcdsp_deblock_neon.o \
|
|
arm/hevcdsp_idct_neon.o \
|
|
arm/hevcdsp_qpel_neon.o
|
|
NEON-OBJS-$(CONFIG_RV30_DECODER) += arm/rv34dsp_neon.o
|
|
NEON-OBJS-$(CONFIG_RV40_DECODER) += arm/rv34dsp_neon.o \
|
|
arm/rv40dsp_neon.o
|
|
NEON-OBJS-$(CONFIG_VORBIS_DECODER) += arm/vorbisdsp_neon.o
|
|
NEON-OBJS-$(CONFIG_VP6_DECODER) += arm/vp6dsp_neon.o
|
|
NEON-OBJS-$(CONFIG_VP9_DECODER) += arm/vp9itxfm_16bpp_neon.o \
|
|
arm/vp9itxfm_neon.o \
|
|
arm/vp9lpf_16bpp_neon.o \
|
|
arm/vp9lpf_neon.o \
|
|
arm/vp9mc_16bpp_neon.o \
|
|
arm/vp9mc_neon.o
|