Merge changes I93308a09,If85c36b4,I918eb36a
Merge "Fix minor bug in calculation of max arf group length."
Merge "Adjustment of GOP intra factor for multi-layer."
Merge "Add MID_OVERLAY_UPDATE frame type"
Merge "Refactor gf_overlay frame type update"
Merge "Revert "Merge "Adapt GOP size threshold to the allowed layer depth"""
Fix minor bug in calculation of max arf group length.
Adjustment of GOP intra factor for multi-layer.
Merge "Fix a loophole in nonrd_pick_partition()"
Revert "Merge "Adapt GOP size threshold to the allowed layer depth""
Merge "Remove deprecated variables from GF_GROUP structure"
Merge "Remove unused for-loop in multi-layer arf bit allocation"
Merge "Adapt GOP size threshold to the allowed layer depth"
Add MID_OVERLAY_UPDATE frame type
Refactor gf_overlay frame type update
Merge "add cfi sanitizer"
Fix a loophole in nonrd_pick_partition()
Remove deprecated variables from GF_GROUP structure
Remove unused for-loop in multi-layer arf bit allocation
Adapt GOP size threshold to the allowed layer depth
Add vpx_clear_system_state to new mv search func
Merge changes If96a8a1c,Iaf535fde,Icbde9880
CONFIG_WEBM_IO: include webmids.h
Call vp9_full_pixel_diamond_new in tpl mv search
Merge "Use layer dependent gfu_boost factor"
Add vp9_full_pixel_diamond_new
Add vp9_refining_search_sad_new
Add vp9_diamond_search_sad_new
Merge "Remove redundant code"
Use layer dependent gfu_boost factor
Merge "clang-format v6.0.1"
Merge "vp9,encoder: check pointers before member access"
Merge "sanitizer: sse2 - fix unaligned double stores"
sanitizer: sse2 - fix unaligned double stores
Merge "segfault: fix missing alignment declaration"
Merge "fix integer overflow caused by uninitialized memory"
Merge "Improve subpel MV search for speed 1"
fix integer overflow caused by uninitialized memory
segfault: fix missing alignment declaration
Revert "Revert "Revert "Loopfilter MultiThread Optimization"""
Merge "vp8: exit with bad fragment size in decoder."
Merge "Rework is_compound_allowed logic at encoder"
Merge "internal stats: fix mem leak and initialize memory"
Merge "better-hw-compatibility: fix out of bounds access"
better-hw-compatibility: fix out of bounds access
Merge "Revert "third_party/googletest: update to v1.8.1""
Revert "third_party/googletest: update to v1.8.1"
Merge "sanitizer: fix unaligned loads"
Merge "Use different corner detection score"
internal stats: fix mem leak and initialize memory
vp8: exit with bad fragment size in decoder.
Improve subpel MV search for speed 1
Rework is_compound_allowed logic at encoder
Merge "Generalize encoder comp_var_ref setting"
Merge "Skip checking compound modes with same sign bias for sub8x8"
Merge "Update the comp_refs counts"
Merge "Skip RD check for compound modes that have same sign bias"
Merge "Sync ref frame writing with decoder"
Merge "Add frame_start/end to gf_group"
sanitizer: fix unaligned loads
Merge "third_party/googletest: update to v1.8.1"
Merge "sanitizer: fix unaligned load/stores"
sanitizer: fix unaligned load/stores
Merge "fix UB when initializing parameterized tests"
Merge changes Ibbe7a1c1,I4333a207
Use different corner detection score
third_party/googletest: update to v1.8.1
fix UB when initializing parameterized tests
Generalize encoder comp_var_ref setting
Skip checking compound modes with same sign bias for sub8x8
Update the comp_refs counts
Add feature score for each block
Correct mv rows/cols bug in read_frame_dpl_stats
Merge "Fix buffer overrun of postproc_state.limits"
Merge "Fix stack corruption with x86 and --enable-pic"
Merge "Rename set_arf_sign_bias() to set_ref_sign_bias"
Merge "Re-work set_arf_sign_bias()"
Merge "Update frame index per buffer at encoder"
Merge "Remove unnecessary code"
Merge "Remove the SECOND_LEVEL_CHECKS_BEST macro"
Fix stack corruption with x86 and --enable-pic
Add frame_start/end to gf_group
Sync ref frame writing with decoder
Skip RD check for compound modes that have same sign bias
Remove the SECOND_LEVEL_CHECKS_BEST macro
Rename set_arf_sign_bias() to set_ref_sign_bias
Merge "Add a frame_index entry to RefCntBuffer"
Merge "Assign GOP frame offset to all the coding frames"
Merge "cosmetics: normalize include guards"
Re-work set_arf_sign_bias()
Update frame index per buffer at encoder
vpx_mem: allow VPX_MAX_ALLOCABLE_MEMORY to be overridden
Remove multi_arf_last_grp_enabled flag.
Remove multi_arf_allowed variable.