Correct bug in loopfilter initialization
Merge "Cleanup inside vp9_reconinter.c."
Merge "Adds a new subpel motion function"
Merge "Removing unneeded intermediate entropy_nodes_adapt var."
Removing unneeded intermediate entropy_nodes_adapt var.
Adds a new subpel motion function
Moved fast motion search level decision to function
Simplify & fix potential bug in rd_pick_partition
Merge "Use low precision 32x32fdct for encodemb in speed1"
Cleanup inside vp9_reconinter.c.
Merge "Using only one scale function in scale_factors struct."
Merge "Adding ss_size_lookup table."
Use low precision 32x32fdct for encodemb in speed1
Adding ss_size_lookup table.
Merge "Removing NMS_STATS defines."
Neon version of vp9_short_idct4x4_add.
Merge "Clean ups of the subpel search functions"
Clean ups of the subpel search functions
Merge "Motion vector code cleanup."
Using only one scale function in scale_factors struct.
Merge "Place holder for high-precision 32x32 fdct"
sse3 intrapred x86inc protected
Merge "Flexible support for various pattern searches"
sad + miscellaneous updates
Merge "Inlining vp9_get_pred_probs_switchable_interp function."
Flexible support for various pattern searches
Place holder for high-precision 32x32 fdct
Inlining vp9_get_pred_probs_switchable_interp function.
Merge "Move fdct32x32 SSE2 implementation in separate file."
Merge "intrapred x86inc guards"
Motion vector code cleanup.
Merge "Finally removing all old block size constants."
Merge "Changing the order switchable filter enum constants."
Merge "Removing unused functions."
Merge "Add variance based mode/skipping"
Move fdct32x32 SSE2 implementation in separate file.
block error / x86inc mods
reworked config for use_x86_inc
Merge changes I082959ab,Ib6932640
Finally removing all old block size constants.
fixed script problem with config_force_x86_inc
Merge "Begin to restrict x86inc.asm usage"
Add variance based mode/skipping
Merge "cleanups after bw bh code"
vp9/decoder: threaded row-based loop filter
vp9/decoder: add thread worker
Changing the order switchable filter enum constants.
Update README
Begin to restrict x86inc.asm usage
Replacing long block size enum values with shorter ones (2).
Merge "Cleaning up vp9_build_inter_predictor function."
Merge "Replacing "txfm" with "tx" in identifiers."
reworked find_mv_ref
Replacing "txfm" with "tx" in identifiers.
Removing NMS_STATS defines.
Merge "Adding is_inter_block function."
Cleaning up vp9_build_inter_predictor function.
Removing unused functions.
Merge "Changing function arg type from int_mv* to MV*."
Merge "Cleanups around allow_high_precision_mv flag."
Adding is_inter_block function.
Merge "Cleaning up set_contexts_on_border function."
Merge "Add more checking to using_small_partition_info"
Cleaning up set_contexts_on_border function.
Changing function arg type from int_mv* to MV*.
Moving struct loop_filter_info from *.h to *.c file.
Merge "Fixed typos and added a few explanatory comments"
Merge "Changed name of rd_pick_intra4x4mby_modes"
Merge "vp9: neon: convolve: replace some insns with simpler equivalents"
Merge "vp9: neon: convolve: simplify branching to C fallbacks"
Merge "vp9: neon: optimise loads in horiz convolve functions"
Merge "vp9: neon: add vp9_mb_lpf_* functions"
Cleanups around allow_high_precision_mv flag.
Replacing long block size enum values with shorter ones.
Merge "Comment out 2 unused speed features"
vp9: neon: convolve: replace some insns with simpler equivalents
vp9: neon: convolve: simplify branching to C fallbacks
vp9: neon: optimise loads in horiz convolve functions
vp9: neon: add vp9_mb_lpf_* functions
Merge "Cleanup: replacing xd->seg with seg, and xd->lf with lf."
Merge "Cleanup: reusing clamp_mv function."
Merge "Remove unused vp9_short_idct10_32x32_add"
Cleanup: replacing xd->seg with seg, and xd->lf with lf.
Merge "Cleanup: removing unused function arguments."
Cleanup: reusing clamp_mv function.
Merge "Nice looking motion vector clamping functions."
Merge "Adds a source variance computation function"
Merge "vp9_get_pred_context_intra_inter cleanup."
Cleanup: removing unused function arguments.
Nice looking motion vector clamping functions.
Adds a source variance computation function
Remove unused vp9_short_idct10_32x32_add
Merge "Optimize 32x32 2D inverse DCT for speed-up"
Add more checking to using_small_partition_info
Comment out 2 unused speed features
Merge "Adding missing const to vp9_extra_bits array."