Disallow the comparison between videos with different bit depth
Merge "vp9-svc: Define rc scene change flag per superframe"
Merge "doxygen: fix --disable-examples"
Merge "Remove -Wextra suppression."
Merge "Add build_motion_field()"
vp9-svc: Define rc scene change flag per superframe
Remove -Wextra suppression.
Merge "vp8: Fix potential use-after-free in mfqe."
Merge "Remove unused code in tiny_ssim"
vp8: Fix potential use-after-free in mfqe.
vp8-mfqe: Increase initial frame# threshold
Merge "Make the use of tpl model controlled by the encoder params"
vp9-svc: On scene change: only reset TL in flexible mode.
Make the use of tpl model controlled by the encoder params
Merge "update libwebm to libwebm-1.0.0.27-352-g6ab9fcf"
Merge "Replace mv_arr by pyramid_mv_arr"
Merge changes I44da4884,I36e3bcae
update libwebm to libwebm-1.0.0.27-352-g6ab9fcf
Replace mv_arr by pyramid_mv_arr
Merge "test/svc_end_to_end_test: fix SetConfig() signature"
vp8: Fix to enabling MFQE
Merge "Rescale arf bit budget calculation"
Merge "Use undamped adjustment for rate correction factors"
Merge "Refactor svc_*_test.cc"
Use undamped adjustment for rate correction factors
Change interface of motion_compensated_prediction
Move prepare_nb_full_mvs to vp9_mcomp.c
Merge "Fix intra_count_low calculation in first pass"
Merge "Clean up condition logics in rc_pick_q_and_bounds_two_pass()"
Merge "Remove redundant code about motion vector test"
test/svc_end_to_end_test: fix SetConfig() signature
Merge "test/svc_*_test: fix SetConfig() signature"
Merge "Add enum definition for subpel search precision"
Merge "apply -Wextra to third_party/"
Clean up condition logics in rc_pick_q_and_bounds_two_pass()
Remove redundant code about motion vector test
Merge "vp9 screen: Update motion search offset when set to NSTEP."
Merge "vp9 svc: add test for scaling partition on 1080p crash."
test/svc_*_test: fix SetConfig() signature
Merge "Add satd avx2 implementation"
Merge "test/*: use std::*tuple"
Merge "Optimize RDMult"
Merge changes Id10f72b3,Icde1b01e,I391aa322
vp9 svc: add test for scaling partition on 1080p crash.
vp9 screen: Update motion search offset when set to NSTEP.
Add enum definition for subpel search precision
Merge "Add high bit Hadamard 16x16 avx2 implementation"
Add high bit Hadamard 32x32 avx2 implementation
Merge "vp9 decoder: cleanup on exit if no available frame buffer."
Merge "configure: test -std=c++11 before enabling unit tests"
vp9 decoder: cleanup on exit if no available frame buffer.
Implement find_prev_nb_full_mvs
Pass mv_num into vp9_nb_mvs_inconsistency()
Merge "Fix DoS in Error Streams"
Merge "remove old visual studio support"
Merge "Refactor datarate svc test."
Merge "quantize neon: fix hbd builds"
Merge "Fix overflow in calculating highbd SSIM"
Merge "Add Parse and Recon Split functions"
Rescale arf bit budget calculation
Add high bit Hadamard 16x16 avx2 implementation
Fix overflow in calculating highbd SSIM
Merge "Reland "third_party/googletest: update to v1.8.1""
Merge "vp9: force refresh of long term ref when denoiser reset."
Fix intra_count_low calculation in first pass
Merge "Clean up rc_pick_q_and_bounds_two_pass()"
Refactor datarate svc test.
configure: test -std=c++11 before enabling unit tests
Fix the parameter of vp9_full_pixel_diamond_new
vp9: force refresh of long term ref when denoiser reset.
Merge "Add high bit Hadamard 8x8 avx2 implementation"
Merge "vp9: Rename post_encode drop function."
Fix DoS in Error Streams
Add Parse and Recon Split functions
vp9-svc: Fix to postencode drop for layers.
vp9: Overshoot detection for skipped base layer.
vp9: Rename post_encode drop function.
Add high bit Hadamard 8x8 avx2 implementation
Merge "Simplify constant q mode qp selection"
Merge "Use 16 bit ints in Hadamard highbd col8 first pass"
Merge changes Ic80def57,I61a2f8bf
Merge changes I18680413,Iebe38092
quantize 32x32: saturate dqcoeff on x86
Clean up rc_pick_q_and_bounds_two_pass()
Simplify constant q mode qp selection
Factor key frame qp selection from two-pass qp and bound decision
Revert "Optimize RDMULT values for key frames"
Consider mv inconsistency in single_motion_search
Change the interface of vp9_full_pixel_diamond_new