Jingning Han authored 9 years ago and James Zern committed 9 years ago

This commit fixes issue 1141. The issue was triggered in multi-tile
encoding. The change properly saves and restores the block context
information in the real-time mode selection process. It removes
several redundant memcpy operations in sub8x8 intra block mode search.

Change-Id: I35c9ad197f4bd500ec39b5fc833f052f19eee010

For dynamic resizing (whether the new codec size is determined internally
or externally set by user), we should for now keep rc.resize_allowed enabled.
This prevent the use of referene_masking for real-time mode
(in set_rt_speed_feature()).

Change-Id: Ibb7c3ff35be88afdf1a3c6db6693521766f177a3

to vp9_setup_pre_planes(), preventing the function
unscaled_value() from being called.  unscaled_value()
returns the same value that was passed in.  See
scaled_buffer_offset() in vp9_reconinter.h.

Change-Id: I2a6fbaf07972c2f212834929d29a2cbe72e399c3

Change-Id: I96d4764523b8d634eafb4cb3ed6eb3aee4052573

The bit to error transformation got doubled as a result of going from
8-bit to 9-bit costs (change d13385ce).

Use defines to derive the scale numbers and comment some of the fields.

derf: -0.023 BDRATE
hevcmr: +0.067 BDRATE
stdhd: +0.098 BDRATE
(These are substantially smaller than than the original gains from 8 to
9 bit costing.)

Change-Id: I6a2b3b029b2f1415e4f90a05709b2333ec0eea9b

When the codec frame size is the same as the reference frame size,
release the scaled reference before assigning it a new buf_idx.
Only affects 1 pass non-svc mode, where the scaled references are
release only under certain conditions (to prevent un-needed scaling
of the references every frame).

Modified a unittest that can trigger this bug without this change.

https://code.google.com/p/chromium/issues/detail?id=582598

Change-Id: I9a884e36ebd7608b1641ec2a469e20a4f829cf43

If the application changes frame size (external size changes),
and aq-mode=3 is on, reset the cyclic refresh.

Modify the TestExternalResize unittest (longer run with more resize
actions). Without this change an assert would be triggered on this
longer test.

Change-Id: I0eefd2cd7ffa0c557cca96ae30d607034a2599ce

This reduces the buffer allocation size for high bit depth buffers by
more than 70%.

Change-Id: Iab9007d9684e8ac30ce7fa94e61f038f01792fc5

Fixes an issue where the tx_type was not set correctly for
sub8x8 inter and intra blocks. In the current syntax, for
sub8x8 blocks, there is still a single tx_type that is
transmitted. Ideally, this should be searched for the best
rd performance, albeit at the expense of encode speed.
For now, we just set it to DCT_DCT. Previously it was left
incorrectly as what was used for the previous non sub8x8
block.

derflr: BDRATE -0.277%

Change-Id: If76ba903bfbfd4d374cf1ac7d1daee50e92f0edd

See https://bugs.chromium.org/p/webm/issues/detail?id=1145

Change-Id: I778ee07dc39a524e3f729bef47a7abeed51e0cee

Change-Id: I6c99ac6fd3735e24cf668479c55bff313c0b4970

Make this consistent with regular block size rate-distortion
optimization. It improves the compression performance:
derf    0.055%
hevcmr  0.129%

Change-Id: I112fe734f592c21bc7aa6efb7e3f269c4214ee7b

Use local variable.

Change-Id: I0d3df36cf4536958a0cda422f6c30da50f0e0bbf

Change-Id: Icc34a00759c95b7b8ac356cdcc4adae848b61431

For 1 pass real-time mode. No change in behavior as only last
and golden are used as references in 1 pass real-time mode.

Change-Id: Ie4655014eee1a8b271542f29d74b2c6f7fed54c9

the results along the top and left border are then stored with a moving
window into the vector.
~40-67% faster on ARM, ~40-77+% on x86 depending on the block size.

Change-Id: Iab369aa2946a3ae4eb7290d512868fe5db92dbc8

Change-Id: I3ac1070df2948f6d8637ade7192400be53a7675c

delete apply_cyclic_refresh_bitrate(). unused since:
3472cbbd vp9 aq-mode=3: Keep it on even at low bitrates.

Change-Id: I0fac9a31b59504e31000ac3a8f0b68e8d4320113

Change-Id: I6a27db6b9dbe3d8cb3d47ef30fbd52557171cadd

Change-Id: I9cc9780f506e025aea57485a9e21f0835faf173c

The definition is for the number of frames to check to determine the
recent decay rate, further to determine the next key frame in the
first pass of the encoder.

Change-Id: Ic696d6eb518a86fa296842273cf8767ef0b0e27a

when INLINE is defined and mips is not being targeted. otherwise keep
the old --enable-extra-warnings behavior

Change-Id: Iba576edbe5fca03efa56ce99eee11f9cafc573ad

in non-x86 builds

Change-Id: I9f6fa40d27f95a4afe92444f71c23dc9a15db466

* changes:
  yv12config.c: remove dead CONFIG_ALPHA code
  vp9_encoder.c: remove unused macro
  vpx_mem.c: remove unused macro
  vp9_svc_layercontext.c: relocate a macro definition
  vp9_encoder.c: protect SNPRINT* macros w/CONFIG check
  vp9_aq_360.c: remove some unused macros
  vp9_diamond_search_sad_avx.c: rename (un)likely macros
  vp9_resize.c: add missing include
  vp9_aq_complexity.c: remove unused macros
  vp9_detokenize.c: remove unused macros

* changes:
  dct16x16_test: quiet unused-function warning
  intrapred: protect functions w/CONFIG check x2
  dct32x32_test: remove unused function
  tests: quiet some warnings w/--disable-webm-io

-use larger threshold on y (as in vp8).
-add distance threshold for each cluster
-use larger skin distance threshold for first cluster
-add some early exist checks.

Keep default setting to model=0.

Change-Id: I1044b99ade4bb1f215a860a019a4d84cee2f7715

It improves the compression performance of VP9 by 0.1% across all
test sets. No speed change is observed.

Change-Id: I59338c5c9e67bae22188f35fc3afbfe2a6bba6b0