GitLab

Clear the compiler errors when both high bit-depth and recursive
transform block partition experiments are enabled.

Change-Id: If0b6396851f10c28b4f26350322ccd1ba2fc9aff

It makes the encoder 5% faster for CIF clips and 12% faster for
1080p clips.

Change-Id: I073408dbb4d50675a79db8794fe73975ac957b91

This commit re-designs the recursive transform block partition
rate-distortion optimization framework. It allows the encoder to
improve speed by 10%.

Change-Id: I6dd3a7dd428a530d8012e5c6ddc40e650c8b392b

This commit enables the rate-distortion optimization for recursive
transform block coding scheme.

Change-Id: Id6a8336ca847bb3af1e94cbfb51db1f4da12d38f

This commit makes the rate-distortion optimization for chroma
component support the recursive transform block coding scheme.

Change-Id: I1bfed6d05b0ebb3905cb625222401e2ccbae10f3

This commit allows the decoder to recursively parse and rebuild
the pixel blocks.

Change-Id: I510f3a30ae7cdad5b70725c66882b00a0594e96f

Add the row and column index to the argument list of unit functions
called by foreach_transformed_block wrapper. This avoids the
repeated internal parsing according to the block index.

Change-Id: I42b3578eac258ebaba7a7c74f684de9abab521a6

Change-Id: Id7f7a9b2e53fc0074b55d58143f296afad6b844e

Remove unnecessary transform type lookups and unused codes.

Change-Id: Ib52d26690468996b1501b419d919643c8ea5ecaa

ext-tx on derflr +2.30% (was +1.84%)

Change-Id: Ic91565cacc38e7a8e1200d054ed7bf99295fe19e

ext-tx on derflr: +1.756% (was +1.648)

Change-Id: I8a87970fa589e8f5f96db7aa68ec9b6c98e20188

Adds an early termination to the ext_tx search, and also
implements the DST transforms more efficiently.

About 4 times faster with the ext-tx experiment.

There is a 0.09% drop in performance on derflr from 1.735% to
1.648%, but worth it with the speedup achieved.

Change-Id: I2ede9d69c557f25e0a76cd5d701cc0e36e825c7c

Resolved Conflicts in the following files:
        configure
        vp10/common/idct.c
        vp10/encoder/dct.c
        vp10/encoder/encodemb.c
        vp10/encoder/rdopt.c

Change-Id: I4cb3986b0b80de65c722ca29d53a0a57f5a94316

Change-Id: Ic161b8b257a02c1c43e515d830c1051d0de074de

At this point, ext-tx compapred to baseline +1.735%.

Change-Id: Ia16ac293e2cc87e06a0d898c1d52a8f3495ff814

Consider tha case in which skipping transform coefficients is more
efficient.

derflr +0.13%
hevclr +0.11%
hevcmr +0.14%
hevchr +0.22%

with ext-tx, the impact is -0.02%.

Change-Id: I0aa2965cf9e152396623c2fee62545bd3a3a7f07

No impact on performance.

Change-Id: Ib0420b190c9d83ef47b14ea78d5918a6a5078e3a

derflr: +1.732% (8-bit)

Change-Id: I9c04c8249646ff96eacacfa1dcb0bd118c04e84a

Change-Id: Ie3daed0b282b43ef81d2f8797ac1f6e8bde7d65e

This condition is not effectively in use. The actual reference
frame masking is done in other route.

Change-Id: Ia59c843bcac7243dada92f0f67658d7ce43df5e8

Change-Id: Ic39d4523e78863c816b0fc85f56ea5ae5e0b3310

Take out speed features that affect the compression performance
to simplify the coding route. This commit removes the motion field
mode search used in speed 3.

Change-Id: Ifdf6862cb1ece8261125a56d9d89bcef60758c00

Change-Id: I61359787fdacdeb245e2798031a6e06e4afb83e0

Does not include DST1 yet.

derflr: +1.437 (8-bit internal), +7.243 (12-bit internal)
with --enable-ext-tx

Change-Id: I91f1759fd2de794755eb6384cda52e80e979cb7d

Change-Id: I2e387a06484a06301f3cd6600c4ba2f4335b61ee

prevents redeclaration warnings;
vp8 has its own define which will be resolved in a future commit

Change-Id: Ic941fef3dd4262fcdce48b73075fe6b375f11c9c

Code cleanup.

Change-Id: I668bd5a4e5fbe96969d51d9ee777fc5f1d8abfe6

derflr +0.202%
hevclf +0.207%
hevcmr +0.095%
hevchr +0.077%

Tested locally on several derf sequences, speed (encoder + decoder)
is slower by less than 1%.

It is part of the EXT_TX experiment, which is to be continued to
explore different transform variants.

Change-Id: I05d44994a62106538a9a241ed8d89bd7c5d14761

stdhd +0.226 hevchr +0.091 hevcmr +0.052 derflr +0.033

Change-Id: I84034209c5760609a99bd6e0ce55e02534b72cac

Prepare for adding more transform varieties (EXT_TX and TX_SKIP in nextgen).

Change-Id: I2dfe024f6be7a92078775917092ed62abc2e7d1e

This makes it easier to add new transform types and scan orders
to VP10 in the future.

Change-Id: I94874ddc9b19928d7820d57e94e2af04adf51efe

Change-Id: I83b5c69621f9f28b742e5b13517d4e5b99c6cd26

Change-Id: I651b7bee90f33581368853da81f9622805ccc0ea

Change-Id: Ia069da11eebb271063e9eb837bdb3e7175ecce13

Change-Id: Ic5004f8bdc1c2b025b598e80374ee1f286ea95ee

Remove the vp9_ prefix from vp10 file names.

Change-Id: I513a211b286a57d6126fc1b0fbfd6405120014f1

This commit folks the VP9 and VP10 codebase and makes libvpx
support VP8, VP9, and VP10.

Change-Id: I81782e0b809acb3c9844bee8c8ec8f4d5e8fa356

Use system_state.h in vpx_dsp and remove unneeded includes of
vp9_systemdependent.h.

Change-Id: I92557ec6dd5aa790160b4f31fe7967db0d7ec3c4

Change-Id: I77e397ac9f594c9c4c1db442e334a6ea5f53f588

Change-Id: I83ed3422f1f4009675ad2f5c4b7236bc7b83b30e