mirror of
https://github.com/xenia-project/FFmpeg.git
synced 2024-11-24 12:09:55 +00:00
a4d4bad75c
This work is sponsored by, and copyright, Google.
The plain pixel put/copy functions are used from the 8 bit version,
for the double size (e.g. put16 uses ff_vp9_copy32_neon), and a new
copy128 is added.
Compared with the 8 bit version, the filters can no longer use the
trick to accumulate in 16 bit with only saturation at the end, but now
the accumulators need to be 32 bit. This avoids the need to keep track
of which filter index is the largest though, reducing the size of the
executable code for these filters.
For the horizontal filters, we only do 4 or 8 pixels wide in parallel
(while doing two rows at a time), since we don't have enough register
space to filter 16 pixels wide.
For the vertical filters, we still do 4 and 8 pixels in parallel just
as in the 8 bit case, but we need to store the output after every 2
rows instead of after every 4 rows.
Examples of relative speedup compared to the C version, from checkasm:
Cortex A7 A8 A9 A53
vp9_avg4_10bpp_neon: 2.25 2.44 3.05 2.16
vp9_avg8_10bpp_neon: 3.66 8.48 3.86 3.50
vp9_avg16_10bpp_neon: 3.39 8.26 3.37 2.72
vp9_avg32_10bpp_neon: 4.03 10.20 4.07 3.42
vp9_avg64_10bpp_neon: 4.15 10.01 4.13 3.70
vp9_avg_8tap_smooth_4h_10bpp_neon: 3.38 6.22 3.41 4.75
vp9_avg_8tap_smooth_4hv_10bpp_neon: 3.89 6.39 4.30 5.32
vp9_avg_8tap_smooth_4v_10bpp_neon: 5.32 9.73 6.34 7.31
vp9_avg_8tap_smooth_8h_10bpp_neon: 4.45 9.40 4.68 6.87
vp9_avg_8tap_smooth_8hv_10bpp_neon: 4.64 8.91 5.44 6.47
vp9_avg_8tap_smooth_8v_10bpp_neon: 6.44 13.42 8.68 8.79
vp9_avg_8tap_smooth_64h_10bpp_neon: 4.66 9.02 4.84 7.71
vp9_avg_8tap_smooth_64hv_10bpp_neon: 4.61 9.14 4.92 7.10
vp9_avg_8tap_smooth_64v_10bpp_neon: 6.90 14.13 9.57 10.41
vp9_put4_10bpp_neon: 1.33 1.46 2.09 1.33
vp9_put8_10bpp_neon: 1.57 3.42 1.83 1.84
vp9_put16_10bpp_neon: 1.55 4.78 2.17 1.89
vp9_put32_10bpp_neon: 2.06 5.35 2.14 2.30
vp9_put64_10bpp_neon: 3.00 2.41 1.95 1.66
vp9_put_8tap_smooth_4h_10bpp_neon: 3.19 5.81 3.31 4.63
vp9_put_8tap_smooth_4hv_10bpp_neon: 3.86 6.22 4.32 5.21
vp9_put_8tap_smooth_4v_10bpp_neon: 5.40 9.77 6.08 7.21
vp9_put_8tap_smooth_8h_10bpp_neon: 4.22 8.41 4.46 6.63
vp9_put_8tap_smooth_8hv_10bpp_neon: 4.56 8.51 5.39 6.25
vp9_put_8tap_smooth_8v_10bpp_neon: 6.60 12.43 8.17 8.89
vp9_put_8tap_smooth_64h_10bpp_neon: 4.41 8.59 4.54 7.49
vp9_put_8tap_smooth_64hv_10bpp_neon: 4.43 8.58 5.34 6.63
vp9_put_8tap_smooth_64v_10bpp_neon: 7.26 13.92 9.27 10.92
For the larger 8tap filters, the speedup vs C code is around 4-14x.
Signed-off-by: Martin Storsjö <martin@martin.st>
149 lines
8.0 KiB
Makefile
149 lines
8.0 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_neon.o \
|
|
arm/vp9lpf_neon.o \
|
|
arm/vp9mc_16bpp_neon.o \
|
|
arm/vp9mc_neon.o
|