GitLab

Various additional changes were made to make the experiment
compatible with misc_fixes.

derflr: +0.979%
hevcmr: +0.865%

Speed-wise with --enable-supertx the encoder is only about 10%
slower than without. Decoding impact is about 30% slowdown.

Note this does not work with ext-tx or var-tx yet. That is
a TODO.

Change-Id: If25af4241a7a9efbd28f58eda3c4f044c7a7ef4b

derflr +0.021%
hevclr +0.207%
hevcmr +0.035%
stdhd  +0.042%

Change-Id: Ic750df93bcc0a261a66a9b19d939a5cd61a6b516

Change-Id: I2aec63d8a600e319d037b764b0609092bce1e483

ext-intra vs nextgenv2 baseline:
derflr +1.12% (was +1.06%)
hevcmr +2.26% (was +2.15%)

Change-Id: I6cc7612d0d7e81e200aa962988db1ea7680626d7

* changes:
  Code clean of highbd_tm_predictor_16x16
  Code clean of highbd_dc_predictor_32x32

Remove the ARCH_X86_64 constraint. No performance hit on both
big core and small core.

Change-Id: I39860b62b7a0ae4acaafdca7d68f3e5820133a81

Remove the ARCH_X86_64 constraint.

Change-Id: I0139f8e998cc5525df55161c2054008d21ac24d4

Remove the ARCH_X86_64 constraint.

Change-Id: I7d2545fc4f24eb352cf3e03082fc4d48d46fbb09

For coding block sizes <=16X16, if the block is determined to be skin,
then always allow for that block to be candidate for refresh. So if that
block happens to be on the boost segment(s), segment won't get reset to 0
and delta-q will be applied.

PSNR/SSIM metrics neutral (little/no change) on RTC clips.
Speed increase small/negligible (< 1%).
Some visual improvement on faces in a few RTC clips.

Change-Id: I6bf0fce6f39d820b491ce05d7c017ad168fce7d6

reduce the register count by 1 to avoid xmm6 and unnecessarily
penalizing the other users of the base macro

Change-Id: I59605c9a41a31c1b74f67ec06a40d1a7f92c4699

Replace MMX with SSE2, reduce mem access to left neighbor,
loop unrolled.

Change-Id: I941be915af809025f121ecc6c6443f73c9903e70

MMX replaced with SSE2, same performance.

Change-Id: I2ab8f30a71e5fadbbc172fb385093dec1e11a696

MMX replaced with SSE2, same performance.

Change-Id: Ic57855254e26757191933c948fac6aa047fadafc

In 32-bit build with --enable-shared, there is a lot of
register pressure and register src_strideq is reused.
The code needs to use the stack based version of src_stride,
but this doesn't compile when used in an lea instruction.

This patch also fixes a related segmentation fault caused by the
implementation using src_strideq even though it has been
reused.

This patch also fixes the HBD subpel variance tests that fail
when compiled without disable-optimizations.
These failures were caused by local variables in the assembler
routines colliding with the caller's stack frame.

Change-Id: Ice9d4dafdcbdc6038ad5ee7c1c09a8f06deca362

H/V intra mode was only enabled for bsize < 16x16,
enable it also for bsize=16x16.

Metrics are neutral with this change:
Overall very small gain (0.1%), small visual gain on some RTC clips.

Change-Id: Ib2d7a44382433bfc11cf324aa3cc5c382ea9e088

Change-Id: I309c4e16fd305bcfa590d14f957a9598d23c7ee6

For the experiment of EXT_REFS, removed the previous special handling
on the new last 3 references, i.e. LAST2_FRAME, LAST3_FRAME, and
LAST4_FRAME, at the decoder, so that these new last references are
treated the same way as the other 3 references (LAST_FRAME,
GOLDEN_FRAME, and ALTREF_FRAME). Encoder changes have been made
accordingly to realize this flexibility.

Change-Id: Ic6546f9443b4377bb7e7b101bfa3e70a8b8d1c65

Replace MMX with SSE2, reduce psadbw ops which may help Silvermont.

Change-Id: Ic7aec15245c9e5b2f3903dc7631f38e60be7c93d

Change-Id: I3ec08e10d9ebf6d8b8a03004a320523f926e5cc4

This also fixes a compiling error under --enable-vp9_highbitdepth.

Change-Id: I9d1dcb95d3336d797eb3c23a4702c30b04355357