FFmpeg/libavcodec/aarch64/vvc
Krzysztof Pyrkosz 71a91485fa avcodec/aarch64/vvc: Optimize NEON version of vvc_dmvr
This patch replaces blocks of instructions performing rounding and
widening shifts with one-liners achieving the same result.

Before and after on A78
dmvr_8_12x20_neon:                                      86.2 ( 6.90x)
dmvr_8_20x12_neon:                                      94.8 ( 5.93x)
dmvr_8_20x20_neon:                                     141.5 ( 6.50x)
dmvr_12_12x20_neon:                                    158.0 ( 3.76x)
dmvr_12_20x12_neon:                                    151.2 ( 3.73x)
dmvr_12_20x20_neon:                                    247.2 ( 3.71x)
dmvr_hv_8_12x20_neon:                                  423.2 ( 3.75x)
dmvr_hv_8_20x12_neon:                                  434.0 ( 3.69x)
dmvr_hv_8_20x20_neon:                                  706.0 ( 3.69x)

dmvr_8_12x20_neon:                                      77.2 ( 7.70x)
dmvr_8_20x12_neon:                                      66.5 ( 8.49x)
dmvr_8_20x20_neon:                                      92.2 ( 9.90x)
dmvr_12_12x20_neon:                                     80.2 ( 7.38x)
dmvr_12_20x12_neon:                                     58.2 ( 9.59x)
dmvr_12_20x20_neon:                                     90.0 (10.15x)
dmvr_hv_8_12x20_neon:                                  369.0 ( 4.34x)
dmvr_hv_8_20x12_neon:                                  355.8 ( 4.49x)
dmvr_hv_8_20x20_neon:                                  574.2 ( 4.51x)

Signed-off-by: Martin Storsjö <martin@martin.st>
2025-03-04 10:35:31 +02:00
..
alf.S aarch64/vvc: Fix clip in alf 2024-12-10 21:00:47 +08:00
alf_template.c
dsp_init.c aarch64/vvc: Add apply_bdof 2024-12-21 11:54:44 +08:00
inter.S avcodec/aarch64/vvc: Optimize NEON version of vvc_dmvr 2025-03-04 10:35:31 +02:00
Makefile aarch64/vvc: Add avg 2024-09-14 16:36:34 +08:00
of_template.c aarch64/vvc: Add apply_bdof 2024-12-21 11:54:44 +08:00
sad.S aarch64/vvc: Add sad 2024-09-14 16:36:34 +08:00