GitLab

Performance drop by 0.084% on lowres

Change-Id: I2bcaae96b68033a0af7a1da988505623bc14ed94

When convolve_round is on, av1_convolve_2d_facade will be used for
interpolation rather than av1_convolve. Will remove the experiment
code of convolve_round experiment from av1_convolve in another CL.

So far we use 4-bit rounding in the intermediate stage on top of using
post rounding for compound mode after the last stage.

This will give us roughly 0.45% gain on lowres , 0.39% on midres and
roughly 0.6-0.7% on hdres
Altogether, is 1.15% on lowresm, 0.74% on midres and roughly 1.7-1.8% on
hdres

Note that there no restriction usage of 12-tap filter in the CL.
Adding that, we will lose roughly 0.1% again on lowres.

Change-Id: I6332e1d888e28a3b3ddc29711817d66e52cb5cdf

Change-Id: I02a019b1c11b3ee4ae32d424cb6ef2f8ae554824

Move declaration of filter_params_x/y outside of if/else block

Change-Id: I4f908872b7ff85b440a12a535d939a3c137aaab5

By turning on CONVOLVE_POST_ROUNDING, in the compound inter
prediction mode, FILTER_BITS rounding is moved after the summation
of two predictions.

Note that the post rounding is only applied on non-sub8x8 block

       PSNR     BDRate
lowres -0.808%  -0.673%

Change-Id: Ib91304e6122c24d832a582ab9f5757d33eac876c

We need uint16_t buf for storing no-rounding prediction.
Add uint16_t buf in conv_params for that.
This CL let us avoid changing interface of convolve functions.

Change-Id: I079fad911327f40ffb98e17c73e7638b1719c975

Those functions includes
av1_make_inter_predictor
av1_build_inter_predictor
inter_predictor

Change-Id: Ide3b744277cf30964e8b352fc8de91365d7217a8

Use a round flag in ConvolveParams to indicate if the destination buffer
has the result rounded by FILTER_BITS or not.
This CL is part of the goal of reducing interpolation rounding error in
compound prediction mode.

Change-Id: I49e522a89a67a771f5a6e7fbbc609e97923aecb6

Change-Id: Ia4779ffb47de333d670ae110cbdfb6cc567da910

When both directions pick sharp filter, horizontal direction use
12-tap sharp filter and vertical direction uses 8-tap sharp filter.

BDRate performance drop slightly.
       BDRate
lowres -0.083%
midres -0.073%
hdres  -0.016%

Change-Id: I6dc075af98f6b4fae558827424a7dd8f38d56503

BDRate varies within +-0.04%

Change-Id: I76f440c479d411c09ef39a19b46eb8dbc5330efb

Change-Id: Ic836798f6e582ee82964fa36ac778eef0c301eb3

Change-Id: Id3936ffd8146c6806fbe93a5bebc23bf5cd8b5f5

Change-Id: I2e3315209559320bbae0ab7e88c9ee5df9e53ef8

Change-Id: Iaf11f439eb4baf2e0c7bf2a19c7c1c012ae7f57b

Let aom_convolve8_### SIMD implementation support any block width.
Turn on SIMD optimization when interpolation filter types on two
directions are different.

This will reduce 30% of encoding time when dual_filter and ext_interp
both on.

Change-Id: I539dbb2737f01835034b7269656a15b2058fa3cc

Performance drop
BDRate
lowres -0.116%
midres -0.073%
hdres  -0.056%

Change-Id: Ic90caf9b8f6fb9d9fd6f9c0e80436a7c468a3c97

There is 0.003% BDRate change on lowres dataset.
Change-Id: Ie62a5ea07bdcfe0a62f37e8f981382df4cc59918

Generate simd filter structure in av1_convolve_init()
This will provide flexibility of changing filter coefficients.

Change-Id: If79f84c56483aa08c894d6b12e2b6ce10147f0ce

Change-Id: I2047adc4c147201ce0ce3c533fe2861cbff1002c

Change-Id: Ie84bdf90c31b12977d32baacfc8086c1fdd96e65

this matches style guidelines and stabilizes successive runs of
clang-format across the tree. remaining types should be address in
successive commits.

Change-Id: I6ad3f69cf0a22cb9a9b895b272195f891f71170f

Cherry-Picked the following commits:
0defd8f2 Changed "WebM" to "AOMedia" & "webm" to "aomedia"
54e66767 Replace "VPx" by "AVx"
5082a369 Change "Vpx" to "Avx"
7df44f17 Replace "Vp9" w/ "Av1"
967f722f Remove kVp9CodecId
828f30ce Change "Vp8" to "AOM"
030b5ffc AUTHORS regenerated
2524caee Add ref-mv experimental flag
016762be Change copyright notice to AOMedia form
81e55269 Replace vp9 w/ av1
9b94565b Add missing files
fa8ca9f2 Change "vp9" to "av1"
ec838b76  Convert "vp8" to "aom"
80edfa01 Change "VP9" to "AV1"
d1a11fb9 Change "vp8" to "aom"
7b582513 Point to WebM test data
dd1a5c8d Replace "VP8" with "AOM"
ff00fc0f Change "VPX" to "AOM"
01dee0bb Change "vp10" to "av1" in source code
cebe6f0c Convert "vpx" to "aom"
17b05679 rename vp10*.mk to av1_*.mk
fe5f8a8a rename files vp10_* to av1_*

Change-Id: I6fc3d18eb11fc171e46140c836ad5339cf6c9419

Manually cherry-picked commits:
ceef0583 libvpx->libaom part2
3d26d916 libvpx -> libaom
cfea7dd7 vp10/ -> av1/
3a8eff73 Fix a build issue for a test
bf4202ed Rename vpx to aom

Change-Id: I1b0eb5a40796e3aaf41c58984b4229a439a597dc

Change-Id: I01d8241eba3ccaf4d06c00a51df2d17c126f6f9d

fortunately, the call site was calling the function with
the correct parameter order.

Change-Id: Ia48099c18288a2416c8b9a7062d2b8d417fd07df

- For experiment EXT_INTERP under high bit depth.
- Add unit test to verify bit-exact.
- Speed performance improvement:
  On Xeon E5-2680, park_joy_1080p_12.y4m, 50 frames, encoding time
  drops from 6682503 ms to 5390270 ms.

Change-Id: Iea4debf5414f3accf1eb5672abeab56a0539ac77

quiets some -Wmissing-prototypes warnings

BUG=b/29584271

Change-Id: I9174728459fcabb6d9ac0028ae58029e52c0da92

- Fix the over-writing bug in horizontal filtering as width = 2.
- Fix 10-tap vertical filtering which no longer reads one row of
  pixel above the block.
- Fix 10-tap filter zero padding.
- Encoder speed slow down ~4.0%, compared to,
  81ad9536 Convolution vertical filter SSSE3 optimization

Change-Id: I9bb294a4529300081c29bf284e6bc6eb081cc536

- Apply 8-pixel vertical filtering direction parallelism.
- Add unit tests to verify bit exact.
- Encoder speed improves ~29% (enable EXT_INTERP) on Xeon E5-2680.
- Combinational cycle count of vp10_convolve() drops from 26.06%
  to 6.73%.

Change-Id: Ic1ae48f8fb1909991577947a8c00d07832737e57

- Apply signal direction/4-pixel vertical/8-pixel vertical
  parallelism.
- Add unit test to verify the bit exact result.
- Overall encoding time improves ~24% on Xeon E5-2680 CPU.

Change-Id: I104dcbfd43451476fee1f94cd16ca5f965878e59

This commit makes the filter extension in highbd aware of the
dual filter and ext-interp experiments to prevent enc/dec mismatch
when both experiments are turned on.

Change-Id: I11ac1f041bd5f73d61e839d6386d9c5d008da3f7

The amount of border extension needed in the first stage inter
filtering is decided by the length of the second stage filter
kernel.

Change-Id: Icddbc58c02234d5df09ff0eeebcf166ffe689203

This commit fixes the compiler error in high bit-depth inter
predictor when dual filter type experiment is turned on.

Change-Id: I404a76a246477f2fcffc38a3275007d5dfe229cd

Make the bit-stream level support per direction filter type coding
for motion compensated reference.

Change-Id: I61a2360b301075f6734cfd9711b7ae68f214174d

If --enable-ext-partition is used at build time, the superblock size
(sometimes also referred to as coding unit (CU) size) is extended to
128x128 pixels.

Change-Id: Ie09cec6b7e8d765b7555ff5d80974aab60803f3a

Includes various cosmetic changes and refactoring including
naming the sharp filters differently (since they are no longer
8-tap).

Change-Id: Ida5a19ca0daa9f6a64a6734394c685b2a4a2564a

Priviously, we do 12-tap interpolation even there is no sub pixel,
This could cause a bug becuase decoder doesn't extend border when there
is no sub pixel. In this situation, if we still do interpolation, we
will access the border extension which doesn't exist and cause a
memory error

Change-Id: I55b879722f0a10c5d13261bd9617a75c826a2418

Change-Id: I1a5b1e15a188ef11594d0c6ac0dbd42aac59cfca