Merge "msvs_common.sh: make fix_file_list generic"
vp9: Move the initialization of some rc variables to rc_init.
Merge "Avoid an unsigned overflow in invert_quant"
Merge "vp9: Don't force block size to 8x8 in denoiser."
Merge "Port convolve test refactor to master."
Merge "bitreader: remove an unsigned overflow."
vp9: Don't force block size to 8x8 in denoiser.
Merge "Declare the operator== overload for M128A in the global namespace."
Declare the operator== overload for M128A in the global namespace.
msvs_common.sh: make fix_file_list generic
Merge changes If45b8931,Ieab273a7
Port convolve test refactor to master.
Tweak casts on vpx_sub_pixel_variance to avoid implicit overflow.
bitreader: remove an unsigned overflow.
Be explicit about overflow in vpx_variance16x16_sse2.
Remove casts on < 16x16 variance.
Avoid an unsigned overflow in invert_quant
Merge "TODO cleanup: Remove unused buffer_size"
Merge "webmdec.h: remove stale TODO"
Merge "TODO cleanup: s/EbmlGlobal/WebmOutputContext/"
TODO cleanup: Remove unused buffer_size
vp9_ethread_test: avoid crash when decoder is unavailable
resize_test: skip VP9 decode checks if it's disabled
vp8: Adjust consec_zeromv thresholds for skin detection.
Merge "VP9: adjust trellis quant optimization RD parameters"
Merge "VP9: enable trellis quantization optimization for intra blocks"
Merge "vp9: Fix condition to update consec_zero_mv."
Merge "vp9: Simplify the logic in denoiser SSE2 code."
Merge "Add the 64-bit CPU cycle count utility function"
Merge "Resolve a couple of TODOs in firstpass.c"
Merge "Experiment to adapt for net AQ offset."
VP9: adjust trellis quant optimization RD parameters
VP9: enable trellis quantization optimization for intra blocks
vp9: Simplify the logic in denoiser SSE2 code.
Merge "vp9: Fix a bug where best_sse_mv is not set properly."
Merge "vpx_minmax_8x8_neon and test"
Merge "vp9-denoiser: Bugfix and some adjustments for high noise case."
vp9: Fix consec_zero_mv in skin detection.
Merge "vp9: 1 pass vbr: Adjust gf boost/interval on motion level."
vp9: Fix condition to update consec_zero_mv.
webmdec.h: remove stale TODO
TODO cleanup: s/EbmlGlobal/WebmOutputContext/
Resolve a couple of TODOs in firstpass.c
Experiment to adapt for net AQ offset.
vp9: 1 pass vbr: Adjust gf boost/interval on motion level.
vp9: Fix a bug where best_sse_mv is not set properly.
vp9-denoiser: Bugfix and some adjustments for high noise case.
Merge "VP9: Do not call vp9_adjust_mask() in vp9_setup_mask()"
vp9: Add rc quantity to track amount of low motion in scene.
Merge "hadamard_test: align src/dst buffers"
Merge "vp9: Some adjustments to denoiser under high-noise condition."
VP9: Do not call vp9_adjust_mask() in vp9_setup_mask()
vp9-noise estimation: Add check on whether skin detection is enabled.
Merge "vp9: Move consec_zero_mv from cyclic refresh to cpi struct."
vp9: Some adjustments to denoiser under high-noise condition.
Add the 64-bit CPU cycle count utility function
vp9: Move consec_zero_mv from cyclic refresh to cpi struct.
Merge "hadamard 8x8 test"
hadamard_test: align src/dst buffers
hadamard 8x8 test
Merge "vp9_loopfilter.c : fix / clarify todo"
Merge "vp9_loopfilter.c: Todo clean up encoder should work like decoder."
Merge "vp9_loopfilter.c : todo cleanup"
vp9_loopfilter.c : fix / clarify todo
vp9_loopfilter.c: Todo clean up encoder should work like decoder.
vp9_encodemb.c: TODO clean up
vp9_loopfilter.c : todo cleanup
Merge "vp9_loopfilter.c: Todo try inloop calculation."
vp9_loopfilter.c: Todo try inloop calculation.
vp9: 1 pass vbr: Small fix on reset avg_intersize_gfint on scene cut.
vp9: 1 pass vbr: small adjustment to active_best factor for inter.
Merge "vp8: Update on sum_diff_threshold."
vp9: Adjust gf period for high average Q and overshoot.
vp8: Update on sum_diff_threshold.
vp9: 1 pass vbr: More even spacing for gf near key frame.
vp9: Adjustment to active_best_quality for inter_frame, 1 pass vbr.
Merge "Adjustment to prediction decay."
Merge "VP9: inline vp9_get_intra_inter_context()"
Merge "Enable vpx_idct32x32_1024_add_neon for neon as well, not only for neon_asm"
Enable vpx_idct32x32_1024_add_neon for neon as well, not only for neon_asm
VP9: inline vp9_get_intra_inter_context()
Apply 'const' to data not pointer
Merge "dct32x32_test: s/HAVE_NEON_ASM/HAVE_NEON/"
Merge "vp9: Adjustment to scene-cut detection."
vp9: Adjustment to scene-cut detection.
Fix Visual Studio build warning
Merge changes from topic 'arm64'
dct32x32_test: s/HAVE_NEON_ASM/HAVE_NEON/
vp9: Adjust threshold for scene-change detection.
Merge "Increase active map test coverage from RT speeds 0-5 to 0-8."
Merge "vp9: Fix to active_best for GF/ARF in 1 pass vbr."
Merge "Step towards making the 2-pass cq mode perceptual"
configure: Detect aarch64 toolchains automatically
configure: Add an arm64-linux-gcc target configuration
Increase active map test coverage from RT speeds 0-5 to 0-8.
vp9: Fix to active_best for GF/ARF in 1 pass vbr.
Merge "Remove obsolete segment skip checks from tokenization."
pickmode: only cost the skip flag once per prediction block
Adjustment to prediction decay.