vp9: Move allocation of vt2 after early exits.
vp9: Reduce stack usage of choose_partitioning.
vp9: Reduce stack usage of choose_partioning.
Merge "Add datarate test for vp8 ROI."
Merge "vp8: Enable use of ROI map."
Merge "Refactor x86/vpx_subpixel_8t_intrin_avx2.c"
Add datarate test for vp8 ROI.
Merge "vp8: [loongson] optimize idct with mmi"
Merge changes I17fff122,Ic149e3cb
Merge "Generalize CheckScalingFiltering in ConvolveTest"
vp8: Enable use of ROI map.
Refactor x86/vpx_subpixel_8t_intrin_avx2.c
vp8: [loongson] optimize idct with mmi
Add 4 to 3 scaling SSSE3 optimization
Adjust threshold in gf_boost for 1 pass vbr
Merge "Further Corpus VBR change."
Merge "Corpus Wide VBR test implementation."
Merge "vp8: [loongson] optimize dct with mmi"
Merge "Adjust to scene detection for 1 pass vbr."
Merge "vp9: use nonrd pick_intra for small blocks on keyframes."
Merge changes I38783d97,If5160c0c
Adjust to scene detection for 1 pass vbr.
vp8: [loongson] optimize dct with mmi
Merge "vp8: [loongson] optimize quantize with mmi"
vp9: use nonrd pick_intra for small blocks on keyframes.
vp8: [loongson] optimize quantize with mmi
Adjust threshold in datarate tests for 1 pass VBR
Test extreme inputs in frame scale functions
Merge "Fix alignment in vpx_image without external allocation."
Add 4 to 1 scaling x86 optimization
Fix alignment in vpx_image without external allocation.
Generalize CheckScalingFiltering in ConvolveTest
Adjustment to scene detection and key frame.
Merge changes I9d4c1af5,I882da3a0
Corpus Wide VBR test implementation.
Extend 16 wide AVX2 convolve8 code to support averaging.
Merge "Update vp9_scale_and_extend_frame_ssse3()"
Rename some inline functions in NEON scaling
Add AVX2 version of vpx_convolve8_avg.
Merge "ppc: Add vpx_idct32x32_1024_add_vsx"
Merge "Revert "Speed >=5 real-time: add TM intra mode for high_source_sad.""
Revert "Speed >=5 real-time: add TM intra mode for high_source_sad."
Adjust threshold in scene detection
Merge "Speed >=5 real-time: add TM intra mode for high_source_sad."
Merge "vpx_codec.h: namespace local defines"
Speed >=5 real-time: add TM intra mode for high_source_sad.
vpx_codec.h: namespace local defines
vp9_ethread_test: abort early/add more detailed output
Merge "Adjust threshold for adapt_partition for speed 6."
Merge "Avoid nonrd_pick_partition for speed >= 6."
Merge "Modify early exit for alt_ref in nonrd_pickmode."
Adjust threshold for adapt_partition for speed 6.
Merge changes Id6a8c549,Ib1e0650b,Ic369dd86
Avoid nonrd_pick_partition for speed >= 6.
Merge "Fix image width alignment. Enable ImageSizeSetting test."
Generalize 2:1 vp9_scale_and_extend_frame_ssse3()
Update vp9_scale_and_extend_frame_ssse3()
Modify early exit for alt_ref in nonrd_pickmode.
Enable arf usage for speed >= 6, 1 pass vbr.
vp9: 1 pass vbr: Limit qpdelta on high_source_sad.
Merge "vpx: fix nasm build errors"
vpx: fix nasm build errors
Refactor x86/vpx_subpixel_8t_intrin_ssse3.c
Add vpx_dsp/x86/mem_sse2.h
Use adapt_partition for ARF in 1 pass.
Add transpose_8bit_{4x4,8x8}() x86 optimization
Merge "ARF in 1 pass vbr: modify skip ref_frame in nonrd_pickmode."
ARF in 1 pass vbr: modify skip ref_frame in nonrd_pickmode.
Add 4 to 3 scaling NEON optimization
Specialize 4 to 3 frame scaling in C
Merge "vpxdsp: [x86] add highbd_d135_predictor functions"
ppc: Add vpx_idct32x32_1024_add_vsx
Fix partition selection in speed features for arf overlay frame.
Enable use_altref_onepass for speed 4 real-time mode.
vpxdsp: [x86] add highbd_d135_predictor functions
vpxdsp: [x86] add highbd_d117_predictor functions
Fix image width alignment. Enable ImageSizeSetting test.
Set rc->high_source_sad = 0 before scene detection.
Merge "vp9: Modification to adapt the ARF usage for 1 pass vbr"
vp9: Modification to adapt the ARF usage for 1 pass vbr
Merge "Add vpx_scaled_2d_neon()"
Add use_svc condition to the scene detection in 1 pass.
Merge "Revert "Remove the speed condition on scene detection in 1 pass code.""
Merge "vpxdsp: [x86] add highbd_d153_predictor functions"
Revert "Remove the speed condition on scene detection in 1 pass code."
Merge "fix signed integer overflow of idct"
Merge "vp9_dx_iface: Stop using iter parameter incorrectly"
fix signed integer overflow of idct
vpxdsp: [x86] add highbd_d153_predictor functions
Remove the speed condition on scene detection in 1 pass code.
vp9_dx_iface: Stop using iter parameter incorrectly
Merge "Add unit test to expose vp8 bug when width is set odd."
Merge "vp8: [loongson] optimize copymen with mmi"
Add unit test to expose vp8 bug when width is set odd.
Merge "SVC: Add setting for max_intra_rate_pct in sample encoder."
Merge changes Ib9105462,Idfac00ed,If8d8a0e2
Remove the speed condition in setting compute_source_sad.
Replace flag USE_ALTREF_FOR_ONE_PASS with speed feature.