ffmpeg | branch: master | Zhao Zhili | Wed Sep 11
21:40:59 2024 +0800| [3f84d1d1fb75e4192c35664b4dd4741e8381b667] | committer:
Nuo Mi
aarch64/vvc: Add avg
avg_8_2x2_c: 0.2 ( 1.00x)
avg_8_2x2_neon: 0.2 ( 1.00x)
ffmpeg | branch: master | Zhao Zhili | Wed Sep 11
15:49:39 2024 +0800| [41a1885f7adc37f176ae2cf52f7a11457354e5fe] | committer:
Nuo Mi
aarch64/vvc: Add put_epel_h
put_chroma_h_8_4x4_c:0.2 ( 1.00x)
put_chroma_h_8_4x4_neon: 0.2 (
ffmpeg | branch: master | Zhao Zhili | Wed Sep 11
16:42:54 2024 +0800| [0dcf204e5d7df388e477fd40c6d4eb4609c7c671] | committer:
Nuo Mi
aarch64/vvc: Add put_epel_h i8mm
put_chroma_h_8_4x4_c:0.4 ( 1.00x)
put_chroma_h_8_4x4_neon:
ffmpeg | branch: master | Zhao Zhili | Wed Sep 11
15:03:33 2024 +0800| [260e1b4b62275a4f5913a55a20581f3567cd5b55] | committer:
Nuo Mi
aarch64/vvc: Add sad
sad_8x16_c: 0.8 ( 1.00x)
sad_8x16_neon: 0.2 ( 3.00x)
ffmpeg | branch: master | Zhao Zhili | Wed Sep 11
00:59:27 2024 +0800| [5ac692580319ba3ef3dadabcb26cb9403bc7f985] | committer:
Nuo Mi
aarch64/vvc: Add put_qpel_hv
With Apple M1 (no i8mm):
put_luma_hv_8_4x4_c: 2.2 ( 1.00x)
put_luma_hv_8_4x4_neon:
ffmpeg | branch: master | Zhao Zhili | Wed Sep 11
17:13:38 2024 +0800| [1be5a2374f224a1fc5e429aafe2ac8fc4efb548c] | committer:
Nuo Mi
aarch64/vvc: Add put_epel_hv
On Apple M1:
put_chroma_hv_8_4x4_c: 1.7 ( 1.00x)
put_chroma_hv_8_4x4_neon:
ffmpeg | branch: master | Zhao Zhili | Tue Sep 10
20:49:58 2024 +0800| [a0b52afd3227064d4127b52313986e3e858b8292] | committer:
Nuo Mi
aarch64/vvc: Add put_qpel_vx
put_luma_v_8_4x4_c: 1.0 ( 1.00x)
put_luma_v_8_4x4_neon: 0.0
ffmpeg | branch: master | Zhao Zhili | Sat Sep 7
16:28:34 2024 +0800| [9f6c8eb41226af7d9f6d9bb2ad2979849e7a3dc2] | committer:
Nuo Mi
aarch64/vvc: Add put_qpel_hx i8mm
Benchmark on Android pixel 8 with -fno-vectorize
put_luma_h_8_4x4_c: 0.2 ( 1.00x)
put_lu
ffmpeg | branch: master | Zhao Zhili | Sat Sep 7
14:07:17 2024 +0800| [25448d17160fe844c79d4ec8e659a704c9a57c9d] | committer:
Nuo Mi
aarch64/vvc: Add put_pel/put_pel_uni/put_pel_uni_w
put_luma_pixels_8_4x4_c: 0.2 ( 1.00x)
put_luma_pixels_8_4x4_neon:
ffmpeg | branch: master | Zhao Zhili | Fri Sep 6
17:58:18 2024 +0800| [20f2bf5530472e4e8ef5be3f7d83a0455a1cac42] | committer:
Nuo Mi
aarch64/vvc: Add put_qpel_h_* and put_qpel_uni_h_*
Just share hevc implementation.
checkasm --test=vvc_mc --benchmark:
put_luma_h_8_4x4_c:
ffmpeg | branch: master | Zhao Zhili | Tue Sep 10
11:11:10 2024 +0800| [b051bc7cb8e4cc81fb313d9e75b169b09f3184ed] | committer:
Nuo Mi
aarch64/h26x: Remove duplicate b.eq instruction
b.eq is added by calc_all after each calc.
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=b051bc7c
ffmpeg | branch: master | Zhao Zhili | Fri Sep 6
16:48:20 2024 +0800| [46f07ce7d1e646ac98ba31897c2a76afc979bdf9] | committer:
Nuo Mi
aarch64/hevc: Move epel/qpel to h26x directory
So vvc can reuse the implementation.
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=46f07ce7d1e646a
ffmpeg | branch: master | Zhao Zhili | Sun Sep 8
00:48:14 2024 +0800| [11443cc9b1ed0d1db901bad55bd1f0819e8187af] | committer:
Nuo Mi
avcodec/hevc: ff_hevc_(qpel/epel)_filters are signed type
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=11443cc9b1ed0d1db901bad55bd1f0819e8187af
-
ffmpeg | branch: master | Zhao Zhili | Sun Sep 1
23:52:18 2024 +0800| [8beafb56564c3876aa4491980fb7a823c71e370f] | committer:
Nuo Mi
aarch64/hevc: Simplify function prototypes by macro
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=8beafb56564c3876aa4491980fb7a823c71e370f
---
l
14 matches
Mail list logo