In frame parallel decode, libvpx decoder decodes several frames on all
cpus in parallel fashion. If not being flushed, it will only return frame
when all the cpus are busy. If getting flushed, it will return all the
frames in the decoder. Compare with current serial decode mode in which
libvpx decoder is idle between decode calls, libvpx decoder is busy
between decode calls.
Current frame parallel decode will only speed up the decoding for frame
parallel encoded videos. For non frame parallel encoded videos, frame
parallel decode is slower than serial decode due to lack of loopfilter
worker thread.
There are still some known issues that need to be addressed. For example:
decode frame parallel videos with segmentation enabled is not right sometimes.
* frame-parallel:
Add error handling for frame parallel decode and unit test for that.
Fix a bug in frame parallel decode and add a unit test for that.
Add two test vectors to test frame parallel decode.
Add key frame seeking to webmdec and webm_video_source.
Implement frame parallel decode for VP9.
Increase the thread test range to cover 5, 6, 7, 8 threads.
Fix a bug in adding frame parallel unit test.
Add VP9 frame-parallel unit test.
Manually pick "Make the api behavior conform to api spec." from master branch.
Move vp9_dec_build_inter_predictors_* to decoder folder.
Add segmentation map array for current and last frame segmentation.
Include the right header for VP9 worker thread.
Move vp9_thread.* to common.
ctrl_get_reference does not need user_priv.
Seperate the frame buffers from VP9 encoder/decoder structure.
Revert "Revert "Revert "Revert 3 patches from Hangyu to get Chrome to build:"""
Conflicts:
test/codec_factory.h
test/decode_test_driver.cc
test/decode_test_driver.h
test/invalid_file_test.cc
test/test-data.sha1
test/test.mk
test/test_vectors.cc
vp8/vp8_dx_iface.c
vp9/common/vp9_alloccommon.c
vp9/common/vp9_entropymode.c
vp9/common/vp9_loopfilter_thread.c
vp9/common/vp9_loopfilter_thread.h
vp9/common/vp9_mvref_common.c
vp9/common/vp9_onyxc_int.h
vp9/common/vp9_reconinter.c
vp9/decoder/vp9_decodeframe.c
vp9/decoder/vp9_decodeframe.h
vp9/decoder/vp9_decodemv.c
vp9/decoder/vp9_decoder.c
vp9/decoder/vp9_decoder.h
vp9/encoder/vp9_encoder.c
vp9/encoder/vp9_pickmode.c
vp9/encoder/vp9_rdopt.c
vp9/vp9_cx_iface.c
vp9/vp9_dx_iface.c
This reverts commit a18da9760a.
Change-Id: I361442ffec1586d036ea2e0ee97ce4f077585f02
In frame parallel decode, libvpx decoder decodes several frames on all
cpus in parallel fashion. If not being flushed, it will only return frame
when all the cpus are busy. If getting flushed, it will return all the
frames in the decoder. Compare with current serial decode mode in which
libvpx decoder is idle between decode calls, libvpx decoder is busy
between decode calls. VP9 frame parallel decode is >30% faster than serial
decode with tile parallel threading which will makes devices play 1080P
VP9 videos more easily.
* frame-parallel:
Add error handling for frame parallel decode and unit test for that.
Fix a bug in frame parallel decode and add a unit test for that.
Add two test vectors to test frame parallel decode.
Add key frame seeking to webmdec and webm_video_source.
Implement frame parallel decode for VP9.
Increase the thread test range to cover 5, 6, 7, 8 threads.
Fix a bug in adding frame parallel unit test.
Add VP9 frame-parallel unit test.
Manually pick "Make the api behavior conform to api spec." from master branch.
Move vp9_dec_build_inter_predictors_* to decoder folder.
Add segmentation map array for current and last frame segmentation.
Include the right header for VP9 worker thread.
Move vp9_thread.* to common.
ctrl_get_reference does not need user_priv.
Seperate the frame buffers from VP9 encoder/decoder structure.
Revert "Revert "Revert "Revert 3 patches from Hangyu to get Chrome to build:"""
Conflicts:
test/codec_factory.h
test/decode_test_driver.cc
test/decode_test_driver.h
test/invalid_file_test.cc
test/test-data.sha1
test/test.mk
test/test_vectors.cc
vp8/vp8_dx_iface.c
vp9/common/vp9_alloccommon.c
vp9/common/vp9_entropymode.c
vp9/common/vp9_loopfilter_thread.c
vp9/common/vp9_loopfilter_thread.h
vp9/common/vp9_mvref_common.c
vp9/common/vp9_onyxc_int.h
vp9/common/vp9_reconinter.c
vp9/decoder/vp9_decodeframe.c
vp9/decoder/vp9_decodeframe.h
vp9/decoder/vp9_decodemv.c
vp9/decoder/vp9_decoder.c
vp9/decoder/vp9_decoder.h
vp9/encoder/vp9_encoder.c
vp9/encoder/vp9_pickmode.c
vp9/encoder/vp9_rdopt.c
vp9/vp9_cx_iface.c
vp9/vp9_dx_iface.c
Change-Id: Ib92eb35851c172d0624970e312ed515054e5ca64
This commit enables sub8x8 inter block coding for RTC mode. The
use of sub8x8 blocks can be turned on by allowing
choose_partitioning function to select 4x4/4x8/8x4 block sizes.
Change-Id: Ifbf1fb3888fe4c094fc85158ac3aa89867d8494a
This commit explicitly set the second reference frame type to be
NONE in key frame coding mode. This fixes a subtle dependency of
reference motion vector used by next inter frame on mode_info
reset before key frame coding.
Change-Id: I5ff0359753fdc9992b0bfe889490f7a32d7d5f6a
Where there is very subtle motion, especially when combined
with low spatial complexity, the codec sometimes fails to quickly
pick up the ambient motion field.
Once it has been established though the field propagates well using
Nearest and Near MV.
This patch looks specifically at the case where the Nearest and Near
have not been established as non zero vectors and in this case
discounts the cost of searching for a new vector in the rd code.
This will almost certainly have some implications in terms of encode
speed but it should be possible to mitigate the impact in a subsequent
using first pass stats and the local spatial complexity.
Average results for test sets approximately neutral.
Change-Id: I44a29e20f11f7ab10f8c93ffbdc50183d9801524
These speed-up features for key frame coding are only turned on
in the settings of hybrid non-RD and RD mode decision. It provides
about 20% speed-up to the hybrid key frame coding at the expense
of certain compression performance loss. For vidyo1, the key frame
coding statistics are changed
9838F, 35.020 dB, 61677 us -> 9920F, 34.834 dB, 47556 us
Overall rtc set compression performance is down by -0.257%.
Change-Id: I0025447fda26bb7855e982955642b5f55d71b51f
This patch modified struct VP9_COMP. Created a struct ThreadData
to include data that need to be copied for each thread. In
multiple thread case, one thread processes one tile. all threads
share one copy of VP9_COMP,
(refer to VP9_COMP *cpi in the code)
but each thread has its own copy of ThreadData,
(refer to ThreadData *td in the code).
Therefore, within the scope of encode_tiles(), both cpi and td
need to be passed as function parameters.
In single thread case, the FRAME_COUNTS pointer in ThreadData
points to "counts" in VP9_COMMON.
Change-Id: Ib37908b2d8e2c0f4f9c18f38017df5ce60e8b13e
Similar to mask_filter, the filter_cache in RD_OPT struct can be
moved out, and declared as a local variable since it is only
used in pick_inter_mode functions.
Change-Id: I412b99cca82bade07ac912064ec03dd1de6b2c17
The mask_filter in RD_OPT struct is used to record rd result in
filter decision. It is only used in pick_inter_mode functions,
and is removed from the struct and declared as a local variable.
Change-Id: I3c95c8632ba7241591ce00ef2ef5677b5e297d7b
Modified VP9_COMP struct to include MACROBLOCK *mb. This change
makes it feasible in multi-thread case to allocate a mb for each
thread.
Change-Id: I624d6d1aa9c132362200753e5d90b581b1738d6e
This patch allocated frame contexts outside VP9_COMMON. This allows
multiple threads to share the same copy of frame contexts, and
reduces the overhead. It also guarantees the correct update of
these contexts during bitstream packing. This patch doesn't change
encoding result.
Change-Id: Ic181a2460b891d1d587278a6d02d8057b9dbd353
Using 4 threads, frame parallel decode is ~3x faster than single thread
decode and around 30% faster than tile parallel decode for frame parallel
encoded video on both Android and desktop with 4 threads. Decode speed is
scalable to threads too which means decode could be even faster with more threads.
Change-Id: Ia0a549aaa3e83b5a17b31d8299aa496ea4f21e3e
When early termination is triggered, properly reset the rate cost
to invalid value to avoid potential ioc issue.
Change-Id: I3444390be2e49a34bb02cf8a74c33d5dbd96d88d
This commit fixes an ioc issue that will happen when the cumulative
variables are not in effective use. The fix discards these
redundant additions.
Change-Id: Idbac5bfb989c0cedc5f8a323effce938519b2457
This commit makes a struct that contains rate value, distortion
value, and the rate-distortion cost. The goal is to provide a
better interface for rate-distortion related operation. It is
first used in rd_pick_partition and saves a few RDCOST calculations.
Change-Id: I1a6ab7b35282d3c80195af59b6810e577544691f
- Some fixes to surface fit.
- Returns variance function as cost rather than sad in the
pattern search and diamond search functions. Only
vp9_pattern_search_sad function used in bigdia search
uses sad as integer 1-away costs.
- Deploys SUBPEL_TREE_PRUNED_MORE for speed 4+.
Results:
derf [Speed 3]: About +0.036% in coding efficiency without any
discernible speed loss.
derf [Speed 4]: About 2-3% faster at -0.199% loss in coding efficiency.
derf [Speed 5]: About 3-4% faster at -0.149% loss in coding efficiency.
Change-Id: I8462f94f6adb46966ca964f2bd0400977357fd63
In model_rd_for_sb function, the spatial domain SSE and variance
are checked to see if transform coefficients are quantized to 0.
Besides that, this patch adds another set of thresholds that are
much more strict. These thresholds are used to conduct a partition
block level check to measure if all its TX blocks are skippable
for YUV planes. If it is true, x->skip is set for this partition
block, and thus its mode search is terminated.
This speeds up the encoding at very low prediction error case,
such as screen sharing application. This patch covers what
rd_encode_breakout_test() does, so that function is removed.
Borg test at speed 3 shows:
For stdhd set, psnr: +0.008%, ssim: +0.014%;
For derf set, psnr: +0.018%, ssim: +0.025%.
No noticeable speed change.
Change-Id: I4e5f15cf10016a282a68e35175ff854b28195944
Fixed an encoder crash. Set skip_txfm to 0 for cases that skip_txfm
isn't calculated. Put memcpy of skip_txfm at right place.
Change-Id: Ib3b6afc1b251a85b2a853c8138fb3393f48cfef6
The functions b_width_log2 and b_height_log2 only do direct
table fetch. This commit unifies such use cases by using the
table directly and removes these functions.
Change-Id: I3103fc6ba959c1182886a2799d21b8b77c8a7b6b
This commit fixes a buffer pointer mis-use in store_coding_context.
The compression performance for stdhd set of speed 3 is improved by
0.097%. It fixes issue 869.
Change-Id: Idc59e22035eaf39f7133ca04174894374d647ff7
It is possible that the GOLDEN reference frame is not avaiable, in
which setting the predicted mv will be associated with a residual
value of INT_MAX. This commit checks this condition before
left shift and comparison with that of ALTREF frame, to avoid
overflow issue.
Change-Id: Ib98c3149dbdd016f2fe5beaafb13f67d469dd07c
This commit enables the encoder to skip split partition search if
the bigger block size has all non-zero quantized coefficients in low
frequency area and the total rate cost is below a certain threshold.
It logarithmatically scales the rate threshold according to the
current block size. For speed 3, the compression performance loss:
derf -0.093%
stdhd -0.066%
Local experiments show 4% - 20% encoding speed-up for speed 3.
blue_sky_1080p, 1500 kbps
51051 b/f, 35.891 dB, 67236 ms ->
50554 b/f, 35.857 dB, 59270 ms (12% speed-up)
old_town_cross_720p, 1500 kbps
14431 b/f, 36.249 dB, 57687 ms ->
14108 b/f, 36.172 dB, 46586 ms (19% speed-up)
pedestrian_area_1080p, 1500 kbps
50812 b/f, 40.124 dB, 100439 ms ->
50755 b/f, 40.118 dB, 96549 ms (4% speed-up)
mobile_calendar_720p, 1000 kbps
10352 b/f, 35.055 dB, 51837 ms ->
10172 b/f, 35.003 dB, 44076 ms (15% speed-up)
Change-Id: I412e34db49060775b3b89ba1738522317c3239c8
Miscellaneous bug-fixes for high bitdepth functionality.
With this patch, high bit-depth profiles become mostly functional,
except for an intermittent assert failure issue that is being
tracked.
Change-Id: I6a7fcbdcf1e5b09842e88535f8442d2e1230748c
Block transform skipping was implemented based on DCT's energy
conservation property. Modified the thresholds using zero bin
parameters. AC and DC coefficients were checked separately to
allow better identifying of skippable blocks.
Borg test at speed 3 showed:
stdhd set: psnr gain: 0.153%, ssim gain: 0.051%;
derf set: psnr gain: 0.023%, ssim gain: 0.036%
For most test clips, the encoding speedup is 1% - 2%.
parkrun(720p): 7.5% speedup, park_joy(1080p): 3.5% speedup.
Change-Id: If28eb81113a077414f5ca7b021c14f9069b373bb
For regular inter frames, if the distance from GOLDEN_FRAME is larger
than 2 and if the predicted motion vector of LAST_FRAME gives lower
sse than that of GOLDEN_FRAME, skip the GOLDE_FRAME mode checking in
the rate-distortion optimization. It provides about 5% speed-up at
expense of -0.137% and -0.230% performance down for speed 3. Local
experiment results:
pedestrian 1080p 2000 kbps
66712 b/f, 40.908 dB, 113688 ms ->
66768 b/f, 40.911 dB, 108752 ms
blue_sky 1080p 2000 kbps
51054 b/f, 35.894 dB, 70406 ms ->
51051 b/f, 35.891 dB, 67236 ms
old_town_cross 720p 1500 kbps
14412 b/f, 36.252 dB, 60690 ms ->
14431 b/f, 36.249 dB, 57346 ms
Change-Id: Idfcafe7f63da7a4896602fc60bd7093f0f0d82ca
This commit enables the encoder to skip checking ALTREF inter modes
in ARF coding, if the predicted motion vectors suggest that the
GOLDEN_FRAME provides higher prediction accuracy than ALTREF_FRAME.
It improves the speed 3 encoding speed by about 5%, at the expense
of compression performance loss -0.041% and -0.225% for derf and
stdhd, respectively.
pedestrian_area 1080p 2000 kbps
66705 b/f, 40.909 dB, 118738 ms ->
66732 b/f, 40.908 dB, 113688 ms
old_town_cross 720p 1500 kbps
14427 b/f, 36.256 dB, 62746 ms ->
14412 b/f, 36.252 dB, 60690 ms
blue_sky 1080p 1500 kbps
51026 b/f, 35.897 dB, 73310 ms ->
50921 b/f, 35.893 dB, 70406 ms
bus CIF 1000 kbps
21301 b/f, 34.841 dB, 7326 ms ->
21248 b/f, 34.837 dB, 7196 ms
Change-Id: I76cf88b4d655e1ee3c0cb03c8a5745493040e8d2
The rd_thresholds are adaptively changed based on best mode tested.
It was only changed for the same block size, this commit makes the
adaptation for similar block sizes too. The commit also made minor
adjustment and code cleanups.
The impact on encoding time for _ped:
118089 ms -> 111927 ms
The impact on compression:
derf: -0.339%
stdhd: -0.303%
Change-Id: I8817fed1102350497f2ec631849e43f753878e5d
mi_grid_* are arrays of pointer to pointer. They save the pointers that point
to the MIs in cm->mi. But they are unnecessary and complicated. The original
goal was to remove MODE_INFO_t copy. But with an extra MODE_INFO_t pointer
inside MODE_INFO_t, same goal could be achieved.
This commit totally removes the mi_grid_* structures. But there are still
many dummy MODE_INFO_t inside cm->mi which are a waste of memory. Next commit
will do on-demand MODE_INFO_t allocation in order to save these memories.
Change-Id: I3a05cf1610679fed26e0b2eadd315a9ae91afdd6
If the partition block is skippable, which means no coefficients
for Y, U, and V planes, its skip flag is set to 1. No quality
change (verified by borg tests), and no noticeable speed change.
Change-Id: I9231f720f8dd6364384cf05aa148ca24d75450f1
This commit enforces ARF validation check for compound inter modes.
It avoids potential access to ARF in the encoding process if it
is not allowed.
Change-Id: I055fec946b5d19d97937dc9001e1e564923e2439
The valid reference frame check in sub8x8 rate-distortion
optimization search has been included in the ref_frame_skip_mask
scheme. This commit removes the later further validation checks
that are not in effect.
Change-Id: I853b477c44037d3dc0afec6cbfce08a96c597a75
This commit replaces the best_ref_index table fetch with the use
of best_mbmode in vp9_rd_pick_inter_mode_sub8x8.
Change-Id: I882ee9ee6a8c0e61befcca1f4dba6d2ea8de8f13
This commit unifies the reference frame control in the rate-
distortion optimization search loop of sub8x8 block size to remove
the control dependency on mode search order.
Change-Id: I3a174099f71a7cc176ede9fd60e2374243ae9232
Adds various high bitdepth transform functions and tests.
Much of the changes are related to using typedefs tran_low_t
and tran_high_t for the final transform cofficients and intermediate
stages of the transform computation respectively rather than fixed
types int16_t/int. When vp9_highbitdepth configure flag is off,
these map tp int16_t/int32_t, but when the flag is on, they map
to int32_t/int64_t to make space for needed extra precision.
Change-Id: I3c56de79e15b904d6f655b62ffae170729befdd8
The variable best_inter_rd is effectively not in use in the rate-
distortion mode search loops of both regular block sizes and sub8x8
block sizes.
Change-Id: I178f909f8c9629772e13adc6257908653b2adf31
hkuang
Jingning Han
Johann
Jim Bankoski
Paul Wilkins
James Zern
Yunqing Wang
Hui Su
Deb Mukherjee
Alex Converse
Yaowu Xu