vp9-svc: Allow for nonzero motion on spatial reference.
Merge "vpx_dsp: [loongson] optimize variance v2."
Merge "pass 'win64' instead of 'x64' to the assembler"
vpx_dsp: [loongson] optimize variance v2.
pass 'win64' instead of 'x64' to the assembler
vp9-svc: Set num_inter_modes in non-rd pickmode.
Merge "mips msa optimize vpx_scaled_2d function"
Merge "Nonrd-pickmode: avoid duplicate computation of UV predictor."
Merge "decouple spatial-svc from encoder abi"
Nonrd-pickmode: avoid duplicate computation of UV predictor.
decouple spatial-svc from encoder abi
Fix to copy partition.
Merge changes Icd9c866b,I81717e47
Merge "vpx: [loongson] fix bug in var_filter_block2d_bil_16x"
Merge "Add PSNR Cb and Cr metric to opsnr.stt"
Merge "vp9-svc: Don't allow encode_breakout on golden ref."
vp8: [loongson] optimize regular quantize v2.
vp9-svc: Don't allow encode_breakout on golden ref.
Merge "Remove unnecessary includes of emmintrin_compat.h"
mips msa optimize vpx_scaled_2d function
vp9-svc: Clean conditon for allowing copy_partition.
vp8: [loongson] optimize vp8_short_fdct4x4_mmi v2.
Remove unnecessary includes of emmintrin_compat.h
vpx: [loongson] fix bug in var_filter_block2d_bil_16x
Add PSNR Cb and Cr metric to opsnr.stt
vp9-svc: Fix condition for setting downsampling filter.
vp9-svc: Fix to the layer buffer settings.
Merge "[CFI] Remove function pointer casts"
Merge "vp8 simulcast: fix compile warnings."
Merge "Optimize AVX2 get16x16var and get32x16var functions"
Merge "vp9 svc: fix a few compile warnings."
[CFI] Remove function pointer casts
vp8 simulcast: fix compile warnings.
vp9-svc: Enbale scale partition reference frames.
vp9 svc: fix a few compile warnings.
Optimize AVX2 get16x16var and get32x16var functions
Merge "Disable allow_partition_search_skip for speed 2."
Merge "Code cleanup."
Merge "Remove decay_accumulator clause from alt ref breakout."
Merge "Add clause to alt ref group breakout."
Merge "vp9: Fix mem rel for non-ref for external buffer."
Disable allow_partition_search_skip for speed 2.
vp9 svc: Rework/fix scale partitioning on boundary.
Remove decay_accumulator clause from alt ref breakout.
Add clause to alt ref group breakout.
fwd txfm ssse3: use GLOBAL() for loading constants
vp9-svc: Fix flag for usage of reuse-lowres partition
tiny_ssim.c : clang compile error fix
Merge "add 10 and 12 bit to tiny_ssim"
Merge "vp9 svc: Change conditions on VPX_ENCODER_ABI_VERSION."
New content type to improve grain retention.
Merge "Fix to frames considered in arf boost calculation."
Merge "CVBR command line option."
vp9 svc: Change conditions on VPX_ENCODER_ABI_VERSION.
add 10 and 12 bit to tiny_ssim
vpx: [x86] add vpx_satd_avx2()
Merge "vpx: [x86] add vp9_block_error_fp_avx2()"
Merge "vp9-svc: Avoid minmax variance for non-reference frames."
Merge "runtime error fix: bitdepth_conversion_avx2.h"
Merge "vp9: SVC feature to use partition from lower resolution."
Merge "fail early on oversize frames"
vp9-svc: Avoid minmax variance for non-reference frames.
vp9: Fix mem rel for non-ref for external buffer.
vp9: SVC feature to use partition from lower resolution.
runtime error fix: bitdepth_conversion_avx2.h
Fix to frames considered in arf boost calculation.
vpx: [x86] add vp9_block_error_fp_avx2()
Merge "vp9: Add nonref frame buffer test."
vp9: Add nonref frame buffer test.
Merge "Support building AVX-512 and implement sadx4 for AVX-512"
Nonrd_pickmode: avoid computing UV cost when early_term is set.
Support building AVX-512 and implement sadx4 for AVX-512
Compound prediction mode for nonrd pickmode.
Merge "vp9: Move allocation of vt2 after early exits."
Merge "vp9: Reduce stack usage of choose_partitioning."
vp9: Move allocation of vt2 after early exits.
vp9: Reduce stack usage of choose_partitioning.
Merge "vp9: Reduce stack usage of choose_partioning."
vp9: Reduce stack usage of choose_partioning.
vp8: correct if/else '{' placement
vpx: hadamard: use ptrdiff_t instead of int for stride
Merge "Optimize convolve8 SSSE3 and AVX2 intrinsics"
Optimize convolve8 SSSE3 and AVX2 intrinsics
Merge "vpx: [x86] vpx_hadamard_16x16_avx2() highbitdepth fix"
vp9-svc: Allow for adapt_rd_thresh with row-mt.
vpx: [x86] vpx_hadamard_16x16_avx2() highbitdepth fix
vpx: [x86] vpx_hadamard_16x16_avx2() improvements
Merge "vpx: [x86] add vpx_hadamard_16x16_avx2()"
Merge "Corpus VBR tweak for undershoot."
Merge "Increase precision of some debug stats output for corpus VBR."
Merge "Prevent double application of min rate in two pass."