Function decomposition inside vp9_decodemv.c file.
Merge "tests: clear system state after non-API calls"
Merge "Make fdct32 computation flow within 16bit range"
tests: clear system state after non-API calls
Merge "Code cleanup inside the decoder code."
Merge "Removing vp9_invtrans.{c, h} files."
Make fdct32 computation flow within 16bit range
Merge "Move subpixel variance function from common/ to encoder/."
Merge "Use assembly-optimized variance functions in sub_pixel_{avg}_var()."
Merge "vpx_ports/x86.h: de-dup #elif block"
convolve_test: align filter arrays
vpx_ports/x86.h: de-dup #elif block
Code cleanup inside the decoder code.
Move subpixel variance function from common/ to encoder/.
Removing vp9_invtrans.{c, h} files.
Use assembly-optimized variance functions in sub_pixel_{avg}_var().
Merge "Fixing compilation error on Mac OS."
Merge "Fix typo ('weight' instead of 'width')."
Fix typo ('weight' instead of 'width').
Fixing compilation error on Mac OS.
Merge "Removed hardcoded global->limit"
Removed hardcoded global->limit
Change the encryption feature to use a callback for decryption.
Merge "Add vp9 test vectors unit test"
Merge "Remove unused need_to_clamp_mvs"
Merge "vpxdec: add --loops option"
vs9 build: avoid deprecated /Wp64 flag
Merge "Fix type mismatch in array definition"
Merge "Remove constant vp9_coef_update_prob table"
Fix type mismatch in array definition
Remove constant vp9_coef_update_prob table
Merge "Enable sse2 version of sad8x4/4x8"
Enable sse2 version of sad8x4/4x8
Merge "Avoid inverse transform/recon without residual"
Some cleanups in rd motion search
Avoid inverse transform/recon without residual
Merge "Fix win64 warning."
Enable sse2 version of sad8x4/4x8
Merge "libs.mk: allow tests to be sharded"
libs.mk: allow tests to be sharded
Add vp9 test vectors unit test
Quick modifications to mb loopfilter intrinsic functions
Remove unused need_to_clamp_mvs
vpxdec: add --loops option
Merge "Quick modifications to wide loopfilter intrinsic functions"
Implement SSE version for sad4x8x4d and SSE2 version for sad8x4x4d.
Quick modifications to wide loopfilter intrinsic functions
Merge changes I86fe51b0,I4c9a9e0f
Merge "Fix compile warnings on windows."
Merge "Enable mmx loop filter routines"
Merge "Fix row tiling."
Merge "Fix chroma output when scaling"
Remove unused vp9_idct_add_{y,uv}_block
Fix the comments in boundary block partition check
Remove some unused loopfilter code
Fix compile warnings on windows.
Merge "Quick fix to stop vpxdec infinite loop"
Quick fix to stop vpxdec infinite loop
Enable mmx loop filter routines
Fix chroma output when scaling
VP9 profile 0 release candidate
Merge branch 'master' into experimental
Trap reference frames of invalid size
decode_tiles(): validate buffer reads
Merge "Minor change in forward updates" into experimental
Merge "Fix mismatch check output" into experimental
Merge "Disallow wide loopfilter on some chroma borders" into experimental
Disallow wide loopfilter on some chroma borders
Minor change in forward updates
Fix rd partition search for corner blocks
Merge "halloc: Cast the offsetof macro to ptrdiff_t before negating"
Fix partition coding of corner block
Adds a zero check in model_rd function
Merge "Using network byte order (big-endian) to encode tile size." into experimental
Merge "New probs for filters/tx_size and a few others" into experimental
New probs for filters/tx_size and a few others
Merge "Remove remnants of VP8 profiles/versions" into experimental
Using network byte order (big-endian) to encode tile size.
Remove remnants of VP8 profiles/versions
print debugging info from mode info struct
Fix a crash in vpxdec
Merge the new loopfilter experiment
Merge "Loopfilter: Fix chroma edge selection" into experimental
Merge "Fix use of get_uv_tx_size in loopfilter" into experimental
Merge "Implement intra-coded frames" into experimental
Cosmetic cleanups of filters
Implement intra-coded frames
Loopfilter: Fix chroma edge selection
Fix use of get_uv_tx_size in loopfilter
Rd check on segment level reference mode.
Allow non-zeromv if ref_frame=intra with segmentation skip/ref enabled.
Fix crash on RD iterations with segmentation enabled.
Merge "Fixed point reference picture scaling" into experimental
Merge "border mvref issue" into experimental