Stowage/ffmpeg
2
0
Fork 0
mirror of https://git.ffmpeg.org/ffmpeg.git synced 2026-02-15 15:20:22 +00:00
Code Issues Projects Releases Packages Wiki Activity
ffmpeg/libavcodec/aarch64
History
Krzysztof Pyrkosz f9b8f30680 avcodec/aarch64/vvc: Optimize vvc_avg{8, 10, 12} 
This patch replaces integer widening with halving addition, and
multi-step "emulated" rounding shift with a single asm instruction doing
exactly that.

Benchmarks before and after:
A78
avg_8_64x64_neon:                                     2686.2 ( 6.12x)
avg_8_128x128_neon:                                  10734.2 ( 5.88x)
avg_10_64x64_neon:                                    2536.8 ( 5.40x)
avg_10_128x128_neon:                                 10079.0 ( 5.22x)
avg_12_64x64_neon:                                    2548.2 ( 5.38x)
avg_12_128x128_neon:                                 10133.8 ( 5.19x)

avg_8_64x64_neon:                                      897.8 (18.26x)
avg_8_128x128_neon:                                   3608.5 (17.37x)
avg_10_32x32_neon:                                     444.2 ( 8.51x)
avg_10_64x64_neon:                                    1711.8 ( 8.00x)
avg_12_64x64_neon:                                    1706.2 ( 8.02x)
avg_12_128x128_neon:                                  7010.0 ( 7.46x)

A72
avg_8_64x64_neon:                                     5823.4 ( 3.88x)
avg_8_128x128_neon:                                  17430.5 ( 4.73x)
avg_10_64x64_neon:                                    5228.1 ( 3.71x)
avg_10_128x128_neon:                                 16722.2 ( 4.17x)
avg_12_64x64_neon:                                    5379.1 ( 3.51x)
avg_12_128x128_neon:                                 16715.7 ( 4.17x)

avg_8_64x64_neon:                                     2006.5 (10.61x)
avg_8_128x128_neon:                                   9158.7 ( 8.96x)
avg_10_64x64_neon:                                    3357.7 ( 5.60x)
avg_10_128x128_neon:                                 12411.7 ( 5.56x)
avg_12_64x64_neon:                                    3317.5 ( 5.67x)
avg_12_128x128_neon:                                 12358.5 ( 5.58x)

A53
avg_8_64x64_neon:                                     8327.8 ( 5.18x)
avg_8_128x128_neon:                                  31631.3 ( 5.34x)
avg_10_64x64_neon:                                    8783.5 ( 4.98x)
avg_10_128x128_neon:                                 32617.0 ( 5.25x)
avg_12_64x64_neon:                                    8686.0 ( 5.06x)
avg_12_128x128_neon:                                 32487.5 ( 5.25x)

avg_8_64x64_neon:                                     6032.3 ( 7.17x)
avg_8_128x128_neon:                                  22008.5 ( 7.69x)
avg_10_64x64_neon:                                    7738.0 ( 5.68x)
avg_10_128x128_neon:                                 27813.8 ( 6.14x)
avg_12_64x64_neon:                                    7844.5 ( 5.60x)
avg_12_128x128_neon:                                 26999.5 ( 6.34x)

Signed-off-by: Martin Storsjö <martin@martin.st>
2025-03-07 15:51:20 +02:00
..
h26x aarch64: h26x: Fix the indentation of one function 2024-09-26 13:42:11 +03:00
vvc avcodec/aarch64/vvc: Optimize vvc_avg{8, 10, 12} 2025-03-07 15:51:20 +02:00
aacencdsp_init.c avcodec/aarch64/aacencdsp: NEON implementation 2025-01-28 10:44:40 +02:00
aacencdsp_neon.S avcodec/aarch64/aacencdsp: NEON implementation 2025-01-28 10:44:40 +02:00
aacpsdsp_init_aarch64.c Include attributes.h directly 2021-04-19 14:34:10 +02:00
aacpsdsp_neon.S aarch64: Reindent all assembly to 8/24 column indentation 2023-10-21 23:25:54 +03:00
ac3dsp_init_aarch64.c avcodec/ac3: Implement sum_square_butterfly_float for aarch64 NEON 2024-04-08 13:36:40 +03:00
ac3dsp_neon.S avcodec/aarch64/ac3dsp_neon.S: Optimize ac3_sum_square_butterfly_int32_neon 2025-03-02 01:17:53 +02:00
cabac.h
fdct.h lavc/aarch64/fdct: add neon-optimized fdct for aarch64 2024-05-13 14:54:10 +02:00
fdctdsp_init_aarch64.c lavc/aarch64/fdct: add neon-optimized fdct for aarch64 2024-05-13 14:54:10 +02:00
fdctdsp_neon.S lavc/aarch64/fdct: add neon-optimized fdct for aarch64 2024-05-13 14:54:10 +02:00
fmtconvert_init.c avcodec/fmtconvert: Remove unused AVCodecContext parameter 2022-09-21 20:26:40 +02:00
fmtconvert_neon.S
h264chroma_init_aarch64.c avcodec/h264chroma: Constify src in h264_chroma_mc_func 2022-08-05 03:02:13 +02:00
h264cmc_neon.S aarch64: Lowercase UXTW/SXTW and similar flags 2023-10-21 23:25:23 +03:00
h264dsp_init_aarch64.c lavc/aarch64: h264, add chroma loop filters for 10bit 2021-08-21 00:06:26 +03:00
h264dsp_neon.S aarch64: Make the indentation more consistent 2023-10-21 23:25:29 +03:00
h264idct_neon.S aarch64: Lowercase UXTW/SXTW and similar flags 2023-10-21 23:25:23 +03:00
h264pred_init.c lavc/aarch64: add pred functions for 10-bit 2021-08-21 00:06:26 +03:00
h264pred_neon.S lavc/aarch64: Fix ff_pred16x16_plane_neon_10 2024-12-17 14:50:29 +02:00
h264qpel_init_aarch64.c lavc/aarch64: h264qpel, add 10-bit lowpass_8_10 based functions 2023-12-07 23:20:14 +02:00
h264qpel_neon.S lavc/aarch64: h264qpel, add 10-bit lowpass_8_10 based functions 2023-12-07 23:20:14 +02:00
hevcdsp_deblock_neon.S avcodec/aarch64/hevc: add luma deblock NEON 2024-02-28 10:14:58 +01:00
hevcdsp_idct_neon.S aarch64/hevcdsp_idct_neon: Add implementation for idct dc 12 2025-03-04 17:01:58 +08:00
hevcdsp_init_aarch64.c aarch64/hevcdsp_idct_neon: Add implementation for idct dc 12 2025-03-04 17:01:58 +08:00
hpeldsp_init_aarch64.c
hpeldsp_neon.S aarch64: Consistently use lowercase for vector element specifiers 2023-10-21 23:25:18 +03:00
idct.h avcodec/aarch64/idct: Add missing stddef 2022-02-21 13:10:04 +01:00
idctdsp_init_aarch64.c lavc/aarch64: fix include for cpu.h 2024-05-13 14:50:38 +02:00
idctdsp_neon.S avcodec/idctdsp: Arm 64-bit NEON block add and clamp fast paths 2022-04-01 10:03:34 +03:00
Makefile avcodec/aarch64/aacencdsp: NEON implementation 2025-01-28 10:44:40 +02:00
me_cmp_init_aarch64.c avcodec/aarch64/me_cmp: add dotprod implementations of sse16 and vsse_intra16 2024-08-17 15:31:48 +02:00
me_cmp_neon.S avcodec/aarch64/me_cmp: add dotprod implementations of sse16 and vsse_intra16 2024-08-17 15:31:48 +02:00
mpegaudiodsp_init.c
mpegaudiodsp_neon.S lavc/hevcdsp_qpel_neon: using movi.16b instead of movi.2d 2023-11-28 15:54:49 +02:00
mpegvideoencdsp_init.c avcodec/mpegvideoencdsp: convert stride parameters from int to ptrdiff_t 2024-09-01 13:42:30 +02:00
mpegvideoencdsp_neon.S avcodec/mpegvideoencdsp: convert stride parameters from int to ptrdiff_t 2024-09-01 13:42:30 +02:00
neon.S aarch64: Consistently use lowercase for vector element specifiers 2023-10-21 23:25:18 +03:00
neontest.c avcodec: Remove deprecated old encode/decode APIs 2021-04-27 10:43:12 -03:00
opusdsp_init.c lavc/opus*: move to opus/ subdir 2024-09-02 11:56:53 +02:00
opusdsp_neon.S avcodec/aarch64/opusdsp_neon: Simplify opus_postfilter_neon 2025-02-10 14:55:16 +02:00
pixblockdsp_init_aarch64.c
pixblockdsp_neon.S
rv40dsp_init_aarch64.c avcodec/h264chroma: Constify src in h264_chroma_mc_func 2022-08-05 03:02:13 +02:00
sbrdsp_init_aarch64.c
sbrdsp_neon.S aarch64: Consistently use lowercase for vector element specifiers 2023-10-21 23:25:18 +03:00
simple_idct_neon.S aarch64: Consistently use lowercase for vector element specifiers 2023-10-21 23:25:18 +03:00
synth_filter_init.c avcodec: Remove DCT, FFT, MDCT and RDFT 2023-10-01 02:25:09 +02:00
synth_filter_neon.S avcodec: Remove DCT, FFT, MDCT and RDFT 2023-10-01 02:25:09 +02:00
vc1dsp_init_aarch64.c avcodec/h264chroma: Constify src in h264_chroma_mc_func 2022-08-05 03:02:13 +02:00
vc1dsp_neon.S avcodec/vc1: Arm 64-bit NEON unescape fast path 2022-04-01 10:03:34 +03:00
videodsp.S lavc/aarch64: fix relocation out of range error 2021-09-25 21:55:29 +03:00
videodsp_init.c avcodec/videodsp: Constify buf in VideoDSPContext.prefetch 2022-07-31 03:14:34 +02:00
vorbisdsp_init.c lavc/vorbisdsp: use ptrdiff_t rather than intptr_t 2022-09-19 13:51:00 -03:00
vorbisdsp_neon.S
vp8dsp.h avcodec/vp8dsp: Constify src in vp8_mc_func 2022-09-11 20:57:51 +02:00
vp8dsp_init_aarch64.c
vp8dsp_neon.S aarch64: Make the indentation more consistent 2023-10-21 23:25:29 +03:00
vp9dsp_init.h
vp9dsp_init_10bpp_aarch64.c
vp9dsp_init_12bpp_aarch64.c
vp9dsp_init_16bpp_aarch64_template.c
vp9dsp_init_aarch64.c
vp9itxfm_16bpp_neon.S aarch64: Use ret x<n> instead of br x<n> where possible 2021-11-16 13:43:56 +02:00
vp9itxfm_neon.S aarch64: Implement stack spilling in a consistent way. 2022-10-11 09:12:02 +02:00
vp9lpf_16bpp_neon.S aarch64: Implement stack spilling in a consistent way. 2022-10-11 09:12:02 +02:00
vp9lpf_neon.S aarch64: Implement stack spilling in a consistent way. 2022-10-11 09:12:02 +02:00
vp9mc_16bpp_neon.S
vp9mc_aarch64.S
vp9mc_neon.S aarch64: vp9mc: Load only 12 pixels in the 4 pixel wide horizontal filter 2025-01-03 17:53:46 -05:00