Total encoding time for first 50 frames of bus (speed 0) @ 1500kbps
goes 2min34.8 to 2min14.4, i.e. a 10.4% overall speedup. The code is
x86-64 only, it needs some minor modifications to be 32bit compatible,
because it uses 15 xmm registers, whereas 32bit only has 8.

Change-Id: I2df53770c2e850813ffa713e1a91b45b0082b904

Change-Id: I692d800af1f976c84a76f8bd66864c4b39540abc

Makes cost_coeffs() a lot faster:
4x4: 236 -> 181 cycles
8x8: 888 -> 588 cycles
16x16: 3550 -> 2483 cycles
32x32: 17392 -> 12010 cycles

Total encode time of first 50 frames of bus (speed 0) @ 1500kbps goes
from 2min51.6 to 2min43.9, i.e. 4.7% overall speedup.

Change-Id: I16b8d595946393c8dc661599550b3f37f5718896

4x4: 234 -> 236 cycles
8x8: 878 -> 888 cycles
16x16: 3664 -> 3550 cycles
32x32: 18134 -> 17392 cycles

Change-Id: I37a51bfbb0060a3a54f09c6045c14a989811ed78

Cycle timings for first 3 frames of bus (speed 0) at 1500kbps:
4x4: 298 -> 234 cycles
8x8: 1227 -> 878 cycles
16x16: 23426 -> 18134 cycles
32x32: 4906 -> 3664 cycles

Total encode time of first 50 frames of bus @ 1500kbps (speed 0) goes
from 3min0.7 to 2min51.6 seconds, i.e. 5.3% faster.

Change-Id: I68a0e1b530b0563b84a67342cca4b45146077e95

This commit replaces zrun_zbin_boost, a method of biasing non-zero
coefficients following runs of zero-coefficients to be rounded towards
zero, with an explicit skip-block choice in the RD loop.

The logic is basically that if individual coefficients should be rounded
towards zero (from a RD point of view), the trellis/optimize loop should
take care of it. If whole blocks should be zero (from a RD point of
view), a single RD check is much more efficient than a complete
serialization of the quantization loop.

Quality change: derf +0.5% psnr, +1.6% ssim; yt +0.6% psnr, +1.1% ssim.
SIMD for quantize will follow in a separate patch. Results for other
test sets pending.

Change-Id: Ife5fa641163ac5150ac428011e87188f1937c1f4

Change-Id: I379617c1c731a686b3f7e032b8805860c1055b12

This commit enables configurable reference buffer pointer for intra
predictor. This allows later removal of spatial dependency between
blocks inside a 64x64 superblock in the rate-distortion optimization
loop.

Change-Id: I02418c2077efe19adc86e046a6b49364a980f5b1

Also tweaks to other features and experiments with
what is on and off at different speed settings.

Change-Id: I3e1d0be0d195216bf17c2ac5df67f34ce0b306b2

Each frame we reset all adaptive thresholds to MAX
rather than base. As modes are picked their thresholds
drop down.

Change-Id: Ia37f03a73003c2d9bfcda57edea07205e9a0e5e8

Renamed cpi->sf.first_step to cpi->sf.reduce_first_step_size
and changed its meaning such that it is a delta applied to
reduce the default first step size (>> x) in the motion search
rather than an absolute value.

The default first step size is already changed according to the image
dimensions (smaller for smaller images). cpi->sf.reduce_first_step_size
now applies a further correction from the default.

Change-Id: Ia94e08bc24c67b604831f980909af7e982fcd16d

Remove vp9_intra4x4_predict(). Use the common intra prediction
function for all block sizes.

Change-Id: Ibd19d51dfa3da8bbdfb79ddeb81530b2e2089560

Change-Id: I453ed11b965e857a14c18ea5c0f4a0a48e7dc0d9

Removing block index (ib) parameter from get_tx_type_{8x8, 16x16}
functions.

Change-Id: Ia213335aae7a7cb027f97b9cc9b04519840250f1

Using MV instead of int_mv for function arguments.

Change-Id: Ic25e13dccbc98fac1fa1b3255127e00cca2a57f6

Change vp9_block_error() to return a 64bit error variable, change all
callers to expect a 64bit return value (this will prevent overflows,
which we basically don't check for at all right now). Remove duplicate
block_error() function, which fixed that through truncation. Remove
old (incompatible) mmx/sse2 block_error SIMD versions and replace with
a new one that returns a 64bit value.

Encoding time of first 50 frames of bus @ 1500kbps goes from 3min29 to
3min23, i.e. a 3% overall speedup.

Change-Id: Ib71ac5508b5ee8a80f1753cd85d72df1629abe68

Change-Id: I7960178c95c54d5c4497e44cfc8c493566294b34

Improves the rd modeling function and implements them using interpolation
from a table which is a little faster. Also uses sse as input to the
modeling function rather than var - since there is no dc prediction
used and as a result the sse works a little better.

derfraw300: +0.05%
Speedup: ~1%

Change-Id: I151353c6451e0e8fe3ae18ab9842f8f67e5151ff

Change-Id: Ie24489a4d39f3e53e816eeebf75a1c9c7d94515a

Change-Id: Ic924f07c6ab0c929c6cdf11880d3c625806e272c

No bitstream or output change - only cosmetics.

Change-Id: Ic8c1d7ad010a87dcf27d12a38cd7dd5adba683a7

Avoids divide-by-zero when variance is 0.

Change-Id: I3c7f526979046ff7d17714ce960fe81d6e1442a0

Change the argument of get_uv_tx_size() to be an MBMI pointer, so that the
correct column's MBMI can be passed to the function.

Change-Id: Ied6b8ec33b77cdd353119e8fd2d157811815fc98

Do not allow the rd code to check compound modes if
a segment level reference frame is selected.

Change-Id: I95f0c57789e0eaceed7caf227e94b4ba3130a06c

Change-Id: Ib5a95bb6ab643b276df3faa9bf99595e4a69ff18

Fixed point scaling factors are calculated once for each
reference frame by using integer division. Otherwise fixed point
scaling routines are used in all scaling calculations. This makes it
possible to calculate fixed point scaling factors on device driver
software and pass them to hardware and thus avoid division on hardware.

TODO:
 - Missing check for maximum frame dimensions
   (currently scaling uses 14 bits)
 - Missing check for maximum scaling ratio
   (upscaling 16:1, downscaling 2:1)

Problems:
 - Straightforward fixed point implementation can cause error +-1
   compared to integer division (i.e. in x_step_q4). Should only
   be an issue for frames larger than 16k.

Change-Id: I3cf4dabd610a4dc18da3bdb31ae244ebaf5d579c

Adds coding of transform size within a frame by use of context
of transform sizes selected in left and above blocks.

Also incorporates code for generating stats.

TODO: generate and incorporate new default stats

Change-Id: I6a7af099f6ad61d448521d9a51167aedaf638ed6

Simplify feature to only support a single reference frame
instead of a mask.

Change-Id: I5dd3a98c7a224aafb35708850ab82e2f220e68fb

Changes to the coding of transform sizes, along with forward
and backward probability updates.

Results:
derf300: +0.241%

Context based coding of transform sizes will be in a separate
patch.

Change-Id: I97241d60a926f014fee2de21fa4446ca56495756

Code intra/inter, then comp/single, then the ref frame selection.
Use contextualization for all steps. Don't code two past frames
in comp pred mode.

Change-Id: I4639a78cd5cccb283023265dbcc07898c3e7cf95

Split partition probabilities between keyframes and non-keyframes,
since they are fairly different. Also have per-blocksize interframe
y intramode probabilities, since these vary heavily between different
blocksizes.

Lastly, replace default probabilities for partitioning and intra modes
with new ones generated from current codec. Replace counts with actual
probabilities also.

Change-Id: I77ca996e25e4a28e03bdbc542f27a3e64ca1234f

Fix the calculation of step size in height.

Change-Id: I0e0c0175f141f5a41214ae51cef233d13942d3c5

Change-Id: Ieface458c83eb6e7ee95595d9fc662f372117c9a

Added structures to support independent rd thresholds
for different block sizes (and set experimental block
size correction factors).

Added structure to to allow dynamic adaptation of thresholds
per mode and per block size basis depending on how often
the mode/block size combination is seen (currently fixed factor).

Removed some unused variables.

TODO
- Adaptation of thresholds based on how often each mode chosen.
- The baseline mode values could also be adjusted based on
  the block size (e.g. for a particular intra mode use a low threshold
  for 4x4 prediction blocks but a relatively high value for 64x64.

Change-Id: Iddee65ff3324ee309815ae7c1c5a8584720e7568

Turn this feature off for some modes in  "good" quality.

Change-Id: I3f262d62cca8f01736b977af1465291e8be29f0a

This avoids encoding tokens for blocks that are entirely
in the UMV border. This changes the bitstream.

Change-Id: I32b4df46ac8a990d0c37cee92fd34f8ddd4fb6c9

This commit makes the coding/reconstruction operations of intra
coding rate-distortion loop for UV components consistent with those
of the encoding process.

key frame coding gains:
derf:   0.11%
stdhd:  0.42%

Change-Id: I8d49f83924a320e3689ef2d60096c49d7f0c7a40

NO bitstream change

Change-Id: I79f6146dac5fdd157051b6f8dc611c0b7b5e5f7f

This commit makes operations of the superblock intra coding rate
distortion optimization consistent with those used in the encoding
process. Given the test prediction mode and transform size, the rd
optimizer encodes and reconstructs each transformed block of the
superblock consecutively, then computes the total rate-distortion
costs accosicated with the current superblock to select the coding
decisions.

It achieves coding performance gains:
derf 0.353%
yt   1.111%

Change-Id: I0da2eb7a71361dfb8c1384927fc536b0c2790d07

Enable iterative motion search for compound inter-inter prediction
of block sizes 4x4/4x8/8x4 only when best coding quality is selected.
The iterative motion search provides about 0.1% gains for derf and
stdhd at this point, at the expense of longer runtime.

Change-Id: Idc03e7f827e51f1bb8d269bc3752ee297a6bbfe5