Merge requests · VideoLAN / dav1d · GitLab

x86: add AVX2 SIMD for ipred.cfl_ac[444]

!954 · created Mar 25, 2020 by Ronald S. Bultje 0.7.0
performance x86
- Merged
- 4
- 1
updated Aug 09, 2021
x86: add AVX2/SSSE3 SIMD for generate_grain_uv_{422,444}

!952 · created Mar 24, 2020 by Ronald S. Bultje 0.7.0
performance x86
- Merged
- 2
updated Aug 09, 2021
Film grain SSSE3 optimizations

!866 · created Dec 21, 2019 by Ronald S. Bultje 0.6.0
performance x86
- Merged
- 14
- 1
updated Aug 09, 2021
Small improvements to ipred_h/ssse3, and make mc/avg/w_avg/mask work on 32bit

!473 · created Dec 06, 2018 by Ronald S. Bultje
performance x86
- Merged
- 10
- 1
updated Aug 09, 2021
x86: add AVX2/SSSE3 version of mc.resize()

!958 · created Mar 27, 2020 by Ronald S. Bultje 0.7.0
performance x86
- Merged
- 8
- 1
updated Aug 09, 2021
mc: add HBD/SSSE3 mc.emu_edge optimizations

!1220 · created Jun 09, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- Approved
updated Aug 09, 2021
Add SSSE3 HBD filmgrain assembly optimizations

!1215 · created Jun 04, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- 4
- 1
- Approved
updated Aug 09, 2021
Add 10/12-bit deblock SSSE3 implementation

!1210 · created Jun 01, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- 5
- 1
- Approved
updated Aug 09, 2021
x86/filmgrain: add fguv_32x32xn_i444/422 HBD AVX2 & SSSE3

!1250 · created Jul 19, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- Approved
updated Aug 09, 2021
x86/film_grain: make generate_grain_y/uv_420 32-bit compatible

!1242 · created Jul 08, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- Approved
updated Aug 09, 2021
x86: add AVX2/SSSE3 versions for filmgrain.fguv_32x32xn[422/444]

!962 · created Mar 29, 2020 by Ronald S. Bultje 0.7.0
performance x86
- Merged
- 4
- 1
updated Aug 09, 2021
x86: add 10/12-bpc AVX2 version of mc.emu_edge.

!1196 · created May 11, 2021 by Ronald S. Bultje 0.9.0
performance x86
- Merged
- 2
- 1
- Approved
updated Aug 09, 2021
Do avx2/hbd scaling*grain multiplication in 16bit instead of 32bit

!1213 · created Jun 03, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- Approved
updated Aug 09, 2021
x86/deblock: make hbd/ssse3 implementations 32bit-compatible 0 of 2 checklist items completed

!1237 · created Jul 05, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- Approved
updated Aug 09, 2021
x86/filmgrain: add generate_grain_uv_i422/i444 HBD AVX2 & SSSE3

!1245 · created Jul 14, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- Approved
updated Aug 09, 2021
x86: don't use vptest in SSSE3 version

!963 · created Mar 31, 2020 by Ronald S. Bultje 0.7.0
bug performance x86
- Merged
- 1
updated Aug 09, 2021
x86/deblock_{hbd_,}avx2: use vpblendvb instead of pand/pandn/por in flat16/8/6

!1238 · created Jul 05, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- 4
- Approved
updated Aug 09, 2021
x86/itx: 16x{4,8,16} inverse transforms hbd/sse4

!1255 · created Jul 30, 2021 by Ronald S. Bultje 0.9.2
performance x86
- Merged
- 4
- Approved
updated Aug 09, 2021
x86/filmgrain: make fg{y,uv_i420}_32x32xn HBD/SSSE3 32bit-compatible

!1244 · created Jul 13, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- Approved
updated Aug 09, 2021
x86/itx: change function signatures of itx_4x4 to 0 GPRs

!1249 · created Jul 17, 2021 by Ronald S. Bultje 0.9.1
performance x86
- Merged
- 1
- Approved
updated Aug 09, 2021

Prev
1
…
8
9
10
11
12
13
14
15
16
…
20
Next

VideoLAN code repository instance