Linfeng Zhang authored Dec 11, 2017 and James Zern committed Dec 14, 2017

Move get_nz_map_ctx() out of get_coeff_cost(), so that the coeff_ctx
won't be recalculated in get_dist_cost_stats().

Change-Id: I511d66e3e1a768c77314b290dc6bb90a123ca541

Change-Id: Ieed2451060f68fee6e67bca7d836d27bc6483c0f

Correctly set the daala_tx_dst8 and daala_tx_dst32 experiment flags in
 both the configure and cmake build systems.
By default daala_tx_dst8 is disabled and daala_tx_dst32 is enabled.

Change-Id: I6c344f40d6603fd766403d915d6689d6c9650257

This reverts commit 0dc0ddc0.

Reason for revert: enc/dec mismatch was found and tracked down to this commit. 

BUG=aomedia:1123

Change-Id: I47ac289426e18d2118117d56c2cbc8930b78bda2

Check that the uv_mode is not UV_MODE_INVALID

Change-Id: Ibbe342397b47d94b0b6823d558a7ed713adef314

This prepares the way for a heuristic to truncate the search.

For a given joint sign value, each plane's alpha is coded
independently and so the alpha choice is also independent.
Rather than iterating over all possible pairs, first choose
for each plane separately and then merge the choices.
Merge with RD collection and continuously update the bound.

On subset1, then encoder output is identical for most sequences:
  PSNR | PSNR Cb | PSNR Cr | PSNR HVS |   SSIM | MS SSIM | CIEDE 2000
0.0002 | -0.0058 | -0.0016 |   0.0020 | 0.0025 |  0.0040 |    -0.0022

Change-Id: Ie7d1e1622d46eaa8cf42f50c9a6bec6e59bdbc39

Update make files, include paths etc.

Change-Id: I78153b28890c7610d65c846eb72cb9dacd30bc2e

Also added some missing 1:2 transforms to the test.

BUG=aomedia:1114

Change-Id: If5ee878c56ba32450931e00b70c00cd95b56359c

- Added all 1:2 and 1:4 transforms through a dynamically generated list.
- Reworked the code to support testing these rectangular transforms.

BUG=aomedia:1114

Change-Id: I7e83b48f02a530716d5e30103780c5c4f450cbbd

Usage of CPU by round_shift_array goes from 2.01% to 1.04%.
Overall encoding is slightly faster (~0.05%).

This means some of the intermediate array have to be aligned.
Also, these functions were moved to common header/source files.

BUG=aomedia:1106

Change-Id: I492c9b1f2e7339c6cb83cfe68a61218642654d1b

Now that the maximum partition size for CfL is 32X32, only one average
needs to be computed by partition. This greatly simplifies the the
cfl_subtract_average() function.

This does not change the bitstream, nor the encoder behavior.

Change-Id: Ib4866442154e4029f2d863e78a8d68d793b83d93

Based on the HW Subgroup call of December 4th 2017, we limit luma partition to
32X32.

Regression on Subset 1
  PSNR | PSNR Cb | PSNR Cr | PSNR HVS |    SSIM | MS SSIM | CIEDE 2000
0.0881 |  1.3504 |  1.2936 |   0.0572 |  0.0182 |  0.0227 |     0.5204

https://two.arewecompressedyet.com/?job=CfL-PartU%402017-12-12T15%3A39%3A36.794Z&job=CfL-Max32x32%402017-12-12T16%3A10%3A09.989Z

Change-Id: I7e3cfd68097c0bc24b1426348b5fd574c4f638a0

To avoid a cascade of encodes when performing CfL RDO,
we compute DC_PRED on the partition unit. To do so, we
change the tx_size of CfL to match the size of the
partition unit (i.e. CfL partitions only contain 1
transform block).

This change requires disabling CfL when a chroma
partition-unit-sized DC_PRED is unavailable
(i.e. 4:1, 1:4 partitions and chroma partitions > 32X32).

Results on Subset1 (compared to disabling 4:1 and 1:4 partU):
   PSNR | PSNR Cb | PSNR Cr | PSNR HVS |    SSIM | MS SSIM | CIEDE 2000
-0.1243 | -1.9286 | -2.0140 |  -0.1514 | -0.1512 | -0.1947 |    -0.8066

https://two.arewecompressedyet.com/?job=master%402017-12-12T14%3A53%3A01.451Z&job=CfL-PartU%402017-12-12T15%3A39%3A36.794Z

Change-Id: I2a4adde79c10089130775b8e0df5f9c198855cad

Change-Id: I4eadddf7c3e99a9a71fd858e46970c1b7d3cd713

Setting --test_decode=fatal no longer checks
U and V planes for monochrome video.

Change-Id: I4ff9f083855223d30bc1cd5d605d866cd5bd1cd7

Change-Id: I899cbbfe28a6d6e8718e18ecff310fe960c3f9bc

Account for the txb_optimize effect on level map context modeling
in the transform block rate distortion optimization search.
Properly update such information in the txb hash rd cost.

BUG=aomedia:1118

Change-Id: I16cf80c57fa37929f94291a7500d84f9e0778400

Previously the tx block order, i.e. the child index, was wrong for
rectangular partition blocks.

Re-eanbles tx rd hash for rectangular blocks, encoder speed up by 7%; no
compression quality loss.

BUG=aomedia:1106

Change-Id: I615b8c56744075088943d372e607c18795cedac4

Use av1_foreach_transformed_block_in_plane.

Change-Id: I22f92b601f6a90e92a93dbff8492df86c717ab6f

Change-Id: Iaf0dd93c4612968db258e78dda0e69cbb152047a

Change-Id: Ie39b11d0cfda205df95650cdced616db148be80a

Minor changes for compile warnings
Merge delta-q changes in bitstream

Change-Id: I732f55e58de3f7eed29b0fe4646714ee6f652324

with the rename in:
1dbe80bc Unify loopfilter function names

the first instance of aom_lpf_horizontal_16_dual_msa was shadowed.
rename this instance and make it private as was done in libvpx.

Change-Id: Ifadae6850c2b7c49ee8f4c988835464ecb3bde94

Also fix one issue with max-tile+LR when the top left tile is not the largest tile.

Change-Id: I721254f63f1a2c6e2c199e27ebaaaebe57234d0f

Blocks with uv_mode == UV_CFL_PRED were pointing past
the end of intra_mode_to_tx_type_context[].

Change-Id: Ia638470ce46f914883bfc3e03f35706ef3d76352

Merged convolve_round experiment and removed its config flag in the code.
Removed compound_round code.

Change-Id: Ic01856732d75cca65d3866383d3cc1dd572f8863

This unifies the coding of signed values in the uncompressed part,
and prevents having one coding for "minus zero".

Change-Id: I3bdc1ab23a33bec0d9a8fe98d5a6bf91df90ff15

HW simplifications for intra predicition.
Changed from 256 points interpolation to 32 points.
Changed values in dr_intra_derivative to be upscaled by 64
instead of 256 and adjusted the angles to be more evenly
spread out. Values are also kept within 10-bits.

AWCY results:
objective-1-fast-1, key-frame only, cpu-used=0
--disable-ext-partition --disable-ext-partition-types
   PSNR | PSNR Cb | PSNR Cr | PSNR HVS |    SSIM | MS SSIM | CIEDE 2000
-0.0476 |     N/A |  0.2705 |  -0.0741 | -0.1357 | -0.1960 |        N/A

Change-Id: I1e1712ab71dc47a3e898b1799ba9e648153735eb

Support the motion field projection for both interpolation and
extrapolation situations.

Change-Id: I74e2d3c8956e62a2731f342f0d7a60dbc6b91ed1

Also, remove warning prevention guard.

Change-Id: I7154f103669e0140068fa6bd7184a8ec34fc7ddf

Change-Id: I8f0f7b7ad996e331c1f1e499e0c6177de9024845

Also, make some locals 'const'.

Change-Id: I030c493b8bb7a1827df16a9feda38dd9dd1e424c

Change-Id: If7fe6c609ee33be4ba243dbb33596b227e313ecd

Change-Id: I7a5b1db07b24f1f97a3324d755ef0438755b1dd2

Change-Id: I834aaf5d043cf591ef3ba01abf7420f08080297d

Change-Id: I9c63b90fd79a925159fd41b05956176127271cd9

Based on av1_convolve_2d_sse2()

Change-Id: Ib648344687eabf5bb6738219206b84224a089aba

Also, fix this test by using 'txsize_sqr_up_map' instead of
'tx_size_sqr_map' in both code and test.

BUG=aomedia:1114

Change-Id: Id72b5e72a40f79ad32819224c99822b5a98538d9

get_nz_map_ctx_from_stats() is take TX_CLASS instead of TX_TYPE, this
commit makes sure right type of parameter is passed in.

Change-Id: I09b97a31890050189fc12f0133c0bde83fc79615

Make sure the ref dv is a valid dv.

Improves compression by about 0.3% on screen_content keyframes.

Change-Id: I3a20c1a04b0ebcad610276f21b410dcfc8ba4c4d