This commit replaces SAD cost with modeled rate-distortion cost
for non-RD mode decision. It translates the prediction residual
SSE into estimate rate and reconstruction distorion costs, hence
capturing the quantization setting effect. The compression
performance of speed -7 for rtc set is improved by 14.79%.

Change-Id: Ifda014eb0501d13109fe7f92680bf1410b463632

Change-Id: I76e39f7c5545ba4e0cd537f530c574531356e94a

Removes some unused variables and assignments

Change-Id: I228f43d658ce50237d65ce8dc8ab2ccf0a0b21ca

Change-Id: I03451c88536bc498edddbe0cd9773ff79da085c2

Change-Id: Ibabd8440ff5c9dd7f300f317e2c9d50e8e84aab9

Change-Id: I9717ef611ef9c39b109b2358c9771bf7fae2dd50

Change-Id: Ib4d850227cef35694c663feb157765a2cbf84699

Change-Id: If175d97990454b171b6abeddb76d142497484487

Change-Id: If90c1bc822873156d4e38fca1938e4907f6c95f0

Change-Id: Ie5f69d39b49a4169cc731d4487e4668fb5af4b4f

Change-Id: I3ed0a95645a66be069ce92a1fad8083a87d01001

Change-Id: I7f3bcca103f0b1f6b3c064b61472543de9a8288a

Properly set intra mode information entries.

Change-Id: Ie3a8992e415dcdd9087a55993bbca06087a0107a

The frame level transform size selection is done inside
encode_frame_internal().

Change-Id: Id7e5e417a3a3f88f875bf12b21bb8029bc6d0d82

Set speed features before running frame encoding. This avoids
redundant RD threshold calculation in key frame coding.

Change-Id: If8e3cf2c02976baa59b310c1c23af9eea0c46e36

* Remove all non-DC intra modes for BLOCK_32x32 and up
* Remove all intra modes for blocks bigger than BLOCK_32x32
* Remove ZEROMV for BLOCK_32x32 and up
* Only consider NEARESTMV for blocks bigger than BLOCK_32x32

Change-Id: Ia18351a238213e2f072f9e481d622949346a245f

Change-Id: Ib719a9f74924718715592e0238bc391ae239d026

Change-Id: Ib49d8dbc67c590f22a1a70251ff607c9f38febd7

The core motion estimation fucntions all return sad now consistently.
The only exception is vp9_full_pixel_diamond(), however the core diamond
and refining search routines called from vp9_full_pixel_diamond() also
return SAD. If variance of pred error + mv cost is desired it must be
calculated explicitly outside these functions. For very fast encoding,
hopefully this will eliminate some redundant computations.

Also suggests reimplementing FAST_HEX with the vp9_pattern_search
framework. It is not exactly the same as the existing FAST_HEX, but
performance is slightly better and speed is very similar. Enables
removing a lot of duplicate code.

Change-Id: I152736393438c25bdf7e96b37cbb8ce330f4f94a

Change-Id: I16dd4d4aaae8ce6a482da3c9d142f41fe9155e82

* speed improvment of 30 percent achieved
* multiplies and adds remain the same
* non-arithmetic instructions minimized by hand, by:
   -expanding 2 pass loop
   -removing irrelivant "shuffles"
   -combining last two rounding steps
* further improvments may be possible

Change-Id: Idec2c3f52910c48e6a0e0f9aefed5cae31b0b8c0

Change-Id: I4a54b12e5229705222c5a101258b9d1f81e2948d

Change-Id: I9423b543e1be414e5c9e10480b813f06e6b88f8a

This patch adds a new speed feature which doesn't do the rather
expensive entropy context lookup or save to the table,  while
doing costing.

The speed up on desktop36p.y4m is around 10% other clips much less.

On the RTC test set this was + 1% in overall datarate.

Change-Id: Ia5144bbf45270671e7be9c8e4055369909e2f738

This gets more accurate mode hit stats. It's also the first step to
handling ZEROMV not being allowed more intelligently.

Change-Id: I5de6734507b5177bf73e9ddbad923f218c39f3e4

Reset FAST_HEX search_method, which was lost because of code merging.

Change-Id: Ic2b2cb9ec2f1751e96f7db0f9226310f605fcc83

intra_y_mode_mask is already enforced for the sub8x8 case.
intra_uv_mode_mask is already enforced for all sizes.

Change-Id: Ia9dd14701cb49873c2e8f24eb5f8b255eaf76a1f

Clear warning caused by implied conversion from int32 to int64.

Change-Id: I473b37a54984cbfb22702eb3f712465881ff86e8

Change-Id: I5a01d0dccc9afc87e2adbb5ddc7d44379fe1f125

There is one parameter that is never used, therefore is removed.

Change-Id: I595722e7c5731534c72831315359e4dce9c21763

Two paramemters are not in use, therefore are removed.

Change-Id: I41eb601b7b4bf2feedb13b3f240315656d8694f9

Change-Id: Ia319bd924401677b5c30b0f46d1ab2be6a1f446e

Change-Id: I111c5bd0dd47db7b1f6952c18056bee548bb42e9

Change-Id: I6d47618e8aee6e6c2fe1cf0b8bb4da0b2a0bab66

In addition, a local variable was renamed to avoid confusion.

Change-Id: Id4c497f9cfa219e8a414aa9fee9a85af5f147249

The function has evolved over time, now only calls vp9_rtcd(), so this
commit removes the function and changes to call vp9_rtcd() directly.

Change-Id: I8cfa6190daa4b28f6f3d1e11bb3a07f9c95322bf

The parameter was never used.

Change-Id: I99fd9aad697917ddbe3d10c71cae344228fafc31

Optimizing 2 functions to process 32 elements in parallel instead of 16:
1. vp9_sub_pixel_avg_variance64x64
2. vp9_sub_pixel_avg_variance32x32
both of those function were calling vp9_sub_pixel_avg_variance16xh_ssse3
instead of calling that function, it calls vp9_sub_pixel_avg_variance32xh_avx2
that is written in avx2 and process 32 elements in parallel.
This Optimization gave 80% function level gain and 2% user level gain

Change-Id: Iea694654e1b7612dc6ed11e2626208c2179502c8

Change-Id: I7255b3bc47d760333f58ac4878becbcc8ad30967

Change-Id: I909a81991aa0cc18afe6526542c7513837445213