Update 32x32 idct sse2 and ssse3 optimizations.
Merge changes I4b4beab1,I02f74dec
Merge "temporal filter test: adjust inputs and runtime"
temporal filter test: adjust inputs and runtime
vp9 svc: Fix the stats output when sl = 1.
Merge "vp9: strip mb graph code"
disable SSSE3/VP9QuantizeTest* in hbd builds
vp9: strip mb graph code
Merge "vp9 SVC: Fix the denoiser frame buffer management."
Merge "Clean highbd idct x86 code with inline functions"
vp9 SVC: Fix the denoiser frame buffer management.
Merge "neon: vpx_quantize_b_32x32"
Merge "quantize: copy ssse3 optimizations to intrinsics"
Merge "vp9: Partition logic adjustment for speed 6 feature."
Clean highbd idct x86 code with inline functions
quantize test: use negative input
quantize: copy ssse3 optimizations to intrinsics
vp9: Partition logic adjustment for speed 6 feature.
Update 32x32 idct sse2 funcs, add partial case 135
Merge "quantize test: consolidate sizes"
Rename highbd_multiplication_and_add_xx() to highbd_butterfly_xx()
Replace multiplication_and_add() with butterfly() in idct x86 code
Update butterfly() in idct x86 optimizations.
Add vpx_highbd_idct16x16_{10, 38, 256}_add_sse4_1
Update for loop increment of idct x86 functions
Update high bitdepth 16x16 idct x86 code
vpx_dsp: merge avx2 variance files
Fix mips dspr2 6 tap filter clobber list
Merge "Rewrite vpx_idct16x16_{10,256}_add_sse2() and add case 38 function"
Merge "vpx_dsp: Use correct check for halfpel in"
Merge changes I76727df0,I66297d78,I1d000c6b
Rewrite vpx_idct16x16_{10,256}_add_sse2() and add case 38 function
vpx_dsp: Use correct check for halfpel in
Merge "Force the bit exactness in the first pass"
Merge "quantize test: add speed comparison"
Extract inlined 16x16 idct sse2 code into header file
Add transpose_32bit_8x4() sse2 optimization
Force the bit exactness in the first pass
vp8: Drop due to overshoot for non-screen content.
Merge "vpxdsp: variance_impl_avx2.c cleanup"
quantize test: add speed comparison
vpxdsp: variance_impl_avx2.c cleanup
Merge "loongson mmi configuration patch."
Merge "Rewrite vpx_highbd_idct8x8_{12,64}_add_sse2"
Merge "neon: vpx_quantize_b"
Update x86 idct optimization
Merge "Respond more rapidly to excessive local overshoot."
Merge "vp9: Adjust noise estimation for 360p."
Merge "highbd_inv_txfm_sse4: make << of neg. val a multiply"
loongson mmi configuration patch.
vp9: Adjust noise estimation for 360p.
Rewrite vpx_highbd_idct8x8_{12,64}_add_sse2
Merge "Revert "Revert "vp9: Speed feature to adapt partition based on source_sad."""
vp9: Fix denoising condition when pickmode partition is used.
Revert "Revert "vp9: Speed feature to adapt partition based on source_sad.""
highbd_inv_txfm_sse4: make << of neg. val a multiply
Merge "Revert "vp9: Speed feature to adapt partition based on source_sad.""
Revert "vp9: Speed feature to adapt partition based on source_sad."
Revert "Rewrite vpx_highbd_idct8x8_{12,64}_add_sse2"
Merge "vp9: Adjust logic in source sad for screen content."
Merge "vp9: Speed feature to adapt partition based on source_sad."
vp9: Speed feature to adapt partition based on source_sad.
Remove the DP version of vp9_optimize_b().
Merge changes Ia0e20f5f,I28150789,I35df041b,I221dff34
vp9: Adjust logic in source sad for screen content.
Revert "quantize ssse3: declare all variables"
Merge "quantize ssse3: declare all variables"
Merge "vp8: Remove isolated skin & non skin blocks."
Update vpx_idct16x16_10_add_sse2()
Add vpx_idct16x16_38_add_sse2()
Rewrite vpx_highbd_idct8x8_{12,64}_add_sse2
Refactor highbd idct 4x4 and 8x8 x86 functions
vp8: Remove isolated skin & non skin blocks.
inv_txfm_{sse2,ssse3}: clear conversion warnings
highbd_idct*_sse*.c: clear conversion warnings
vpx_variance16x16_sse2: correct cast order
vpx_get16x16var_avx2: correct cast order
set_var_thresh_from_histogram: prevent negative variance
vp8: Fix compile warning in vp8_multi_resolution_encoder.c
Merge "quantize test: promote RandRange() result to signed"
Merge "quantize test: lowbd functions do not pass in highbd"
Merge "vp9: Removed unused skin detection function."
Merge "quantize test: eob is output"
quantize test: promote RandRange() result to signed
quantize test: lowbd functions do not pass in highbd
Merge "Earmark extra space for VSX."
Merge "Roll libwebm: Fix android build failure with NDK r15b."
Merge "variance: call C comp_avg_pred"
Merge "vp9: Allocate alt-ref in denoiser for SVC."
Merge "vp9: Remove isolated skin & non-skin blocks."
Merge changes I62c2e313,Ibd7a0337,I94e1d886
variance: call C comp_avg_pred
Roll libwebm: Fix android build failure with NDK r15b.