Fix valgrind failure on uninitialized variables.
vp9: SVC: Force subpel search off under certain conditions.
Merge "vp9: Speed >8: Set subpel_search_method for low motion."
vp9: Speed >8: Set subpel_search_method for low motion.
Fix corruption in skin map debugging output yuv.
vp8 skin detection: Fix visual studio build failure.
vp8: Clean up skin detection.
Merge "comp_avg_pred neon: used by sub pixel avg variance"
Merge "Write skin map of vp8 skin detection for debug."
Merge "Update vpx_highbd_idct4x4_16_add_sse2()"
comp_avg_pred neon: used by sub pixel avg variance
Merge "comp_avg_pred: alignment"
Merge "remove DECLARE_ALIGNED from neon code"
Write skin map of vp8 skin detection for debug.
Update vpx_highbd_idct4x4_16_add_sse2()
Merge "Fix vp8 race when build --enable-vp9-highbitdepth."
Merge "subpel variance neon: reduce stack usage"
Merge "Use vdup instead of vmov"
remove DECLARE_ALIGNED from neon code
Fix vp8 race when build --enable-vp9-highbitdepth.
vp9: SVC: Fix to condiiton on using source_sad.
Merge "vp9: Use source_sad only on top temporal enhancement layer."
Merge "vp9: SVC: Enable copy partition for SVC speed >= 7."
vp9: Use source_sad only on top temporal enhancement layer.
Refactor: Move vp8 skin detection to new files.
vp9: SVC: Enable copy partition for SVC speed >= 7.
Merge changes Iaab2b9a1,Idfb458d3
Merge changes I31fa6ef8,I228c6f29
Merge "partial_idct_test,InitInput: fix rollover in mult"
partial_idct_test,InitInput: fix rollover in mult
subpel variance neon: reduce stack usage
Update inv_txfm_sse2.h and inv_txfm_sse2.c
Update InitInput() in test/partial_idct_test.cc
Add support for Visual Studio 2017
Merge "Add vpx_highbd_idct{4x4,8x8,16x16}_1_add_sse2"
sub pel avg variance neon: 4x block sizes
sub pel variance neon: 4x block sizes
sub pel avg variance neon: add neon optimizations
sub pel variance neon: normalize variable names
variance neon: assert overflow conditions
Add vpx_highbd_idct{4x4,8x8,16x16}_1_add_sse2
Merge changes Ib8dd96f7,Ie9854b77
Merge "vp9: Adjustments to cyclic refresh for high motion."
vp9: Adjustments to cyclic refresh for high motion.
vp9: Speed >= 8: Modify condition for low-resoln.
Merge "neon 4 byte helper functions"
Merge "neon fdct: 4x4 implementation"
Merge "Changes to modified error."
vp9: SVC: Modify condition to allow for copy partition.
Merge "vp9: Make copy partition work for SVC and dynamic resize."
vp9: Make copy partition work for SVC and dynamic resize.
Merge "vp9: Modify ChangingDropFrameThresh unittest."
Merge "Update partial idct testing code"
Merge "Experiment. Store first pass errors as per MB values."
use memcpy for unaligned neon stores
Experiment. Store first pass errors as per MB values.
vp9: Modify ChangingDropFrameThresh unittest.
Update partial idct testing code
Merge "Revert "Add visibility="protected" attribute for global variables referenced in asm files.""
Merge "Revert "Revert "vp8: Real-time mode: reduce mode_check_freq thresh for speed 10."""
Revert "Revert "vp8: Real-time mode: reduce mode_check_freq thresh for speed 10.""
Merge "Revert "vp8: Real-time mode: reduce mode_check_freq thresh for speed 10.""
Revert "vp8: Real-time mode: reduce mode_check_freq thresh for speed 10."
Add transpose_32bit_4x4() and rename transpose_4x4() for vpx_dsp/x86
Revert "Add visibility="protected" attribute for global variables referenced in asm files."
Merge "'protected' visibility unsupported on macho"
Merge "move neon load/stores to a new file"
'protected' visibility unsupported on macho
move neon load/stores to a new file
vp8: Real-time mode: reduce mode_check_freq thresh for speed 10.
Merge "vp9: speed 8: Fix seg fault in partition copy when drop frames."
Merge "Speed up encoding by skipping altref recode"
vp9: speed 8: Fix seg fault in partition copy when drop frames.
Add visibility="protected" attribute for global variables referenced in asm files.
Merge "vp9: Use INTERP_FILTER for filter_type in vp9_rtcd_defs.pl"
Merge changes I1b54a7a5,I3028bdad,I59788cd9
ppc: Add get_mb_ss_vsx
ppc: Add get4x4sse_cs_vsx
ppc: Add comp_avg_pred_vsx
ppc: Add vpx_sad64x32/64_vsx
ppc Add vpx_sad32x16/32/64_vsx
Merge "ppc: Add vpx_sad16x8/16/32_vsx"
Merge "vp9: Adjust speed features for speed 8 at low resoln."
Merge "vp9: SVC: Increase the partiiton and acskip thresholds"
Merge "vp9; Adjust noise estimation thresholds."
vp9: SVC: Increase the partiiton and acskip thresholds
Merge "Tuning of factor used to calculate Q range in two pass."
Merge "vp9: Fix ubsan failure in denoiser."
Tuning of factor used to calculate Q range in two pass.
vp9: SVC: allow for setting the interp_filter in non-rd pickmode.
Merge "vp9: SVC: Fix setting in sample encoder."
Speed up encoding by skipping altref recode
Merge "Update specializations of idct functions"