v2.3.0 to v3.0.0 · Alliance for Open Media / SVT-AV1 · GitLab

Commits on Source 184

4e640fd2 Remove unused dec_svt_av1_highbd_warp_affine
Oct 29, 2024
75d71898 Remove 12-bit test cases for highbd_warp_affine
Oct 29, 2024
c09f5854 Add Neon implementation of highbd_warp_affine
Oct 29, 2024
9806f2a8 Add SVE implementation of highbd_warp_affine
Oct 29, 2024
277aab62 Hoist filter load if alpha == beta == 0 in warp_affine_neon
Oct 29, 2024
248d622c Add I8MM implementation of svt_av1_warp_affine
Oct 29, 2024
ea0e3fc8 Add SVE implementation of svt_av1_warp_affine
Oct 29, 2024
dd6f8ecd Add Neon implementation of HBD wiener_convolve_add_src
Nov 04, 2024
9364e94e Refactor intrapred_edge_filter_test.cc
Nov 04, 2024
65134b39 Add Neon implementation of svt_av1_filter_intra_edge
Nov 04, 2024
24279a5d Add Neon implementation of svt_av1_filter_intra_edge_high
Nov 04, 2024
ae1051ae Add Neon implementation of svt_av1_wedge_compute_delta_squares
Nov 05, 2024
3dba5fb4 Add SVE implementation of svt_av1_wedge_sse_from_residuals
Nov 05, 2024
97ddda5e Add SVE implementation of svt_av1_wedge_sign_from_residuals
Nov 05, 2024
dda33a04 Cleanup various code
Nov 05, 2024
65b63dcb Remove all 12-bit tests from CompoundUtilTest.cc
Nov 05, 2024
5b828722 Add Neon implementation of build_compound_diffwtd_mask
Nov 05, 2024
715eb000 Add Neon implementation of build_compound_diffwtd_mask_d16
Nov 05, 2024
108b271b Add Neon implementation of build_compound_diffwtd_mask_highbd
Nov 05, 2024
2f310724 Fix AArch64 SIMD implementations of highbd_warp_affine
Nov 06, 2024
6e5927c4 Move the DotProd impl. of svt_aom_sse to its own file
Nov 06, 2024
0ee45a24 Move svt_aom_highbd_sse_neon to a separate file
Nov 06, 2024
2129bc56 Port libaom Neon implementation of svt_aom_highbd_sse
Nov 06, 2024
04f0d200 Add SVE implementation of svt_aom_highbd_sse
Nov 06, 2024
3f124cda Fix input size for svt_full_distortion_kernel16_bits unit tests
Nov 06, 2024
2a3fd82a Use highbd sse helper for svt_full_distortion_kernel16_bits_neon
Nov 06, 2024
746e9e97 Add DotProd implementation of svt_spatial_full_kernel_distortion
Nov 06, 2024
33a2e714 Add SVE implementation of svt_full_distortion_kernel16_bits
Nov 06, 2024
8d9f4ff4 Optimize svt_full_distortion_kernel32_bits_neon
Nov 06, 2024
fde75679 Optimize svt_aom_blend_a64_hmask_neon
Nov 06, 2024
eb8375cb Optimize svt_aom_blend_a64_vmask_neon
Nov 06, 2024
c7b25688 Optimize svt_aom_lowbd_blend_a64_d16_mask_neon
Nov 06, 2024
8e7713a6 Add a Neon implementation of svt_aom_highbd_blend_a64_mask
Nov 06, 2024
482a9199 Add Neon implementation of svt_aom_highbd_blend_a64_d16_mask_neon
Nov 06, 2024
4d5f6f23 Delete svt_aom_highbd_blend_a64_h/vmask_8bit
Nov 06, 2024
11c6f55a Add Neon implementation of svt_aom_highbd_blend_a64_hmask_16bit
Nov 06, 2024
56c450fb Add Neon implementation of svt_aom_highbd_blend_a64_vmask_16bit
Nov 06, 2024
b8a7cf31 Inline and optimize helpers for Neon blend functions
Nov 06, 2024
fdcb8858 Change input type of blend_a64_d16_mask unit tests
Nov 06, 2024
fbdbf661 Add lossless coding support
Nov 08, 2024
a0c1bbb0 Add a safeguard against using qm when lossless coding is used. Avoid checking...
Nov 08, 2024
63c95718 Fix svt_av1_fwht4x4() for ARM architecture
Nov 08, 2024
2bac495e Add Neon implementation of highbd_smooth_v_predictor functions
Nov 08, 2024
fc327d84 Add Neon implementation of highbd_smooth_h_predictor functions
Nov 08, 2024
15721c37 Add Neon implementation of highbd_smooth_predictor functions
Nov 08, 2024
b60948c7 Add Neon implementation of highbd_v_predictor functions
Nov 08, 2024
1bc8869e Add Neon implementation of highbd_h_predictor functions
Nov 08, 2024
7e8afe68 Add Neon implementation of highbd_paeth_predictor functions
Nov 08, 2024
674fcee9 Add Neon implementation of highbd_dc_predictor functions
Nov 08, 2024
b5b64294 Add Neon implementation of highbd_dc_left_predictor functions
Nov 08, 2024
8886b51b Add Neon implementation of highbd_dc_top_predictor functions
Nov 08, 2024
4c2db316 Add Neon implementation of highbd_dc_128_predictor functions
Nov 08, 2024
3b0353f8 Add Neon implementation of av1_highbd_dr_prediction_z1_neon
Nov 08, 2024
d6b86525 Add Neon implementation of av1_highbd_dr_prediction_z2_neon
Nov 08, 2024
b2c25169 Add Neon implementation of av1_highbd_dr_prediction_z3_neon
Nov 08, 2024
94a1a85c CMakeLists.txt: Fix typo: OLD_CMAKE_{REQURED => REQUIRED}_FLAGS
Nov 14, 2024
b4dd9074 Add Neon implementation of svt_aom_highbd_quantize_b
Nov 21, 2024
c642bb04 Add Neon implementation of svt_av1_highbd_quantize_fp
Nov 21, 2024
c4b5c7d1 AVX2: Improve variance function performance
Nov 25, 2024
7faf0462 still-picture coding
Nov 26, 2024
66fb0f59 API: Add asm flag for AVX512 extensions in Icelake
Nov 27, 2024
032f75a9 AVX2: Use optimal AVX lane swizzle for each task
Nov 28, 2024
0fc1d842 AVX512: Add sad helper functions
Dec 03, 2024
425e4bdf Cleanup: Remove svt_nxm_sad_kernel_sub_sampled
Dec 04, 2024
c39f1bdc Optimize Neon implementation of svt_av1_fwd_txfm2d_4x4
Dec 05, 2024
4b19ff38 Optimize Neon implementation of svt_av1_fwd_txfm2d_8x8
Dec 05, 2024
6f29e6d6 Optimize Neon implementation of svt_av1_fwd_txfm2d_16x16
Dec 05, 2024
6bb3e61f Optimize Neon implementation of svt_av1_fwd_txfm2d_4xh
Dec 05, 2024
f72d298b Optimize Neon implementation of svt_av1_fwd_txfm2d_8xh
Dec 05, 2024
3089b747 Optimize Neon implementation of svt_av1_fwd_txfm2d_16xh
Dec 05, 2024
88eeeb52 Optimize Neon implementation of svt_av1_fwd_txfm2d_32xh
Dec 05, 2024
8f1f1b0d Optimize Neon implementation of svt_av1_fwd_txfm2d_64xh
Dec 05, 2024
4f910ed0 Tune presets and optimize mds0 early exits
Dec 05, 2024
3c9d117e Optimize GM search
Dec 05, 2024
a9fd1325 Tune Presets and Optimize ME, CDEF, lambda modulation
Dec 05, 2024
5485a23a Fix superres b64_total_count update
Dec 05, 2024
09d2f729 Optimize low-delay and fast-decode modes
Dec 05, 2024
e135fd2e Shift Presets
Dec 05, 2024
bcc280d2 Tune presets and fix unit tests
Dec 18, 2024
288aa181 Add the ability to specify a qp-offset for the start-up GOP
Dec 19, 2024
1ec05bb2 Add Neon implementation of svt_av1_fwd_txfm_4x4_N4
Dec 19, 2024
7cc85a1a Add Neon implementation of svt_sv1_fwd_txfm2d_8x8_N4
Dec 19, 2024
f71fbe4c Optimize Neon implementation of svt_av1_fwd_txfm2d_16x16_N4
Dec 19, 2024
3b196e69 Optimize Neon implementation of svt_av1_fwd_txfm2d_32x32_N4
Dec 19, 2024
6eb67651 Optimize Neon implementation of svt_av1_fwd_txfm2d_64x64_N4
Dec 19, 2024
38191530 Add Neon implementation of svt_av1_fwd_txfm2d_4x4_N2
Dec 19, 2024
52a3c6d2 Add Neon implementation of svt_av1_fwd_txfm2d_8x8_N2
Dec 19, 2024
499ed949 Add Neon implementation of svt_av1_fwd_txfm2d_16x16_N2
Dec 19, 2024
e1d6a0c3 Add Neon implementation of svt_av1_fwd_txfm2d_32x32_N2
Dec 19, 2024
c07e504d Add Neon implementation of svt_av1_fwd_txfm2d_64x64_N2
Dec 19, 2024
c87e36c5 Add Neon implementation of svt_aom_sum_squares_i16
Dec 20, 2024
17bea173 Add SVE implementation of svt_aom_sum_squares_i16
Dec 20, 2024
45ccefd8 Add Neon implementation of svt_aom_highbd_lpf_horizontal_4
Dec 20, 2024
1579b98a Add Neon implementation of svt_aom_highbd_lpf_horizontal_6
Dec 20, 2024
0654d55e Add Neon implementation of svt_aom_highbd_lpf_horizontal_8
Dec 20, 2024
68c2d0fd Add Neon implementation of svt_aom_highbd_lpf_horizontal_14
Dec 20, 2024
e63e725a Add Neon implementation of svt_aom_highbd_lpf_vertical_4
Dec 20, 2024
65508a6a Add Neon implementation of svt_aom_highbd_lpf_vertical_6
Dec 20, 2024
1050e3ad Add Neon implementation of svt_aom_highbd_lpf_vertical_8
Dec 20, 2024
783c3f1f Add Neon implementation of svt_aom_highbd_lpf_vertical_14
Dec 20, 2024

84 additional commits have been omitted to prevent performance issues.