Commit Graph

7830 Commits

Author SHA1 Message Date
Dmitry Kovalev
9c9fdf30e6 Merge "Adding raw_read_frame() function to vpxdec." 2014-01-13 11:46:01 -08:00
James Zern
ff0d073d88 Merge "test_libvpx: fix negative CPU filter" 2014-01-13 11:25:59 -08:00
Dmitry Kovalev
4ae8ac7be8 Merge "Cleaning up and fixing psnr calculation code." 2014-01-13 10:59:26 -08:00
Dmitry Kovalev
f0fde24379 Adding raw_read_frame() function to vpxdec.
Change-Id: Ie2a4606daf35b327d6f2ac8d7fd8f6cacf4c5b6a
2014-01-13 10:54:57 -08:00
Yaowu Xu
6fcabdd20c Merge "Enable reference frame masking for rt mode" 2014-01-13 10:52:00 -08:00
Yaowu Xu
5e5d4c0ea7 Enable reference frame masking for rt mode
Reference frame masking helped good quality mode to gain about 5% in
encoding speed, this commit enable it for rt mode to gain the speed
improvement.

In addition, this commit move the speed feature setup to a separate
function.

Change-Id: I015e8f78bbb21dd43ae183b9b9355bea2ccda9c5
2014-01-13 10:06:17 -08:00
Johann
7c52a3a25b Merge "mips dsp-ase r2 vp9 decoder fixed bug in extend frame module (seg fault)" 2014-01-13 09:22:15 -08:00
Paul Wilkins
a00dad39bd No arf right before real scene cut.
To reduce pulsing we now allow an arf just before forced key frames
and at the end of a clip or section (which may be stitched to
another clip or section). However, this does not make sense for
key frames arising from real scene cuts.

Change from original patch reflects other recent changes in regard
to alignment of gf/arf and kf groups.

Change-Id: I074a91d1207e9b3e28085af982f6718aa599775f
2014-01-13 11:35:02 +00:00
Paul Wilkins
603075fa04 Further rate control tweaks and fixes.
Further fixes regarding min and max rate.
Bug fixes re kf group bits and last kf group.

Change-Id: Iaafd719d30a489e135a3c55851ce8c632091a436
2014-01-13 11:34:34 +00:00
Parag Salasakar
59fa8cfd5f mips dsp-ase r2 vp9 decoder fixed bug in extend frame module (seg fault)
Change-Id: Ic46da8897d2fb5351e22d6d74805fd9d78c81b32
2014-01-13 14:44:18 +05:30
James Zern
f83c12b540 Merge "cosmetics: vp9_reconinter.h: make some variables const" 2014-01-11 12:39:32 -08:00
James Zern
01236769ab Merge "datarate_test: fix some type conversions" 2014-01-11 12:38:36 -08:00
James Zern
ada9dd7d84 test_libvpx: fix negative CPU filter
negative patterns begin with _one_ '-', not one before each pattern.

https://code.google.com/p/googletest/wiki/V1_7_AdvancedGuide#Running_a_Subset_of_the_Tests

Change-Id: I6fd527b31bc2571dfdaa453cbbb57c77b3409563
2014-01-10 19:08:44 -08:00
Johann
2c60f87506 Merge "Apply neon flags to intrinsic files" 2014-01-10 17:04:56 -08:00
Dmitry Kovalev
ca677a3859 Merge "Cleaning up vp9_rc_postencode_update() function." 2014-01-10 16:53:35 -08:00
Dmitry Kovalev
4def0a8125 Cleaning up and fixing psnr calculation code.
Introducing calc_psnr() which calculates psnr between two yv12 buffers.
Previously we incorrectly used width/height instead of
crop_width/crop_height to calculate number of samples -- fixed.

Change-Id: Iecda01980555de55ad347e0276e6641c793fa56c
2014-01-10 16:09:56 -08:00
Dmitry Kovalev
6ed59ecc8a Merge "Cleaning up vp9_dx_iface.c." 2014-01-10 16:01:54 -08:00
Jingning Han
29736b889b Merge "Declare setup_buffer_inter in vp9_rdopt.h" 2014-01-10 16:01:07 -08:00
Jingning Han
4f969ccc1b Merge "Enable skipping reference frame check in rd loop" 2014-01-10 16:00:56 -08:00
Dmitry Kovalev
aea2864471 Merge "Removing mi_height_log2_lookup table." 2014-01-10 16:00:56 -08:00
Jim Bankoski
6439aa5a0e explain speed features
Added comments to explain what the various speed features do, and removed
1 that was clearly unused.

Change-Id: Icd37a536072ddafedbfaefcecbe48979f6d10faf
2014-01-10 14:10:34 -08:00
Jingning Han
db2b350d33 Declare setup_buffer_inter in vp9_rdopt.h
This funtion initializes buffer pointers and first stage motion vector
prediction. It will be needed by both regular rate-distortion
optimization loop and the non-RD mode decision. Hence move its
declaration in vp9_rdopt.h

Change-Id: I64e8b6316c9d05f20756a62721533a2e4d158235
2014-01-10 13:52:40 -08:00
Dmitry Kovalev
96be0a50ab Removing mi_height_log2_lookup table.
Change-Id: I1f0ae2edc3a96b33c0494d165ae756a8feba6184
2014-01-10 13:29:47 -08:00
Johann
dadf350551 Apply neon flags to intrinsic files
Filter out files ending in _neon.c and append .neon so the Android build
system knows to apply -mfpu=neon

Change-Id: Ib67277e5920bfcaeda7c4aa16cd1001b11d59305
2014-01-10 12:16:59 -08:00
Marco Paniconi
21a0c1f38f Merge "Don't use gf_update by default for 1-pass CBR." 2014-01-10 11:43:20 -08:00
Dmitry Kovalev
21ededd4f4 Cleaning up vp9_dx_iface.c.
Change-Id: I6a0dfb95c55ee6cadc7b1675782c7830e5c7caaf
2014-01-10 11:01:52 -08:00
hkuang
b5af9d2905 Merge "Fix Issue #679: vp9 C loop filter produces valgrind warning." 2014-01-10 10:59:00 -08:00
Dmitry Kovalev
447eece329 Cleaning up vp9_rc_postencode_update() function.
Change-Id: I02e44c10660fdb9201a802ad19ceb64756feeebe
2014-01-10 10:44:20 -08:00
Marco Paniconi
c46538d45e Don't use gf_update by default for 1-pass CBR.
Change-Id: I5df6abceb0a2a69706feadeb820b593cae88f573
2014-01-10 10:40:12 -08:00
Dmitry Kovalev
ed364b2114 Merge "Adding {get, set}_rate_correction_factor() functions." 2014-01-10 10:30:04 -08:00
hkuang
66c6f7bf61 Fix Issue #679: vp9 C loop filter produces valgrind warning.
Fix the valgrind error due to access uninitialized
memory in loopfilter.

Change-Id: I52fccf5ede845ee1f4c13d3bd909b8f220c0bdff
2014-01-10 10:24:21 -08:00
Marco Paniconi
a260369aa8 Merge "Keep buffer clipped to maximum in change_config." 2014-01-10 09:33:33 -08:00
Paul Wilkins
b645257121 Revert "SSSE3 convolution optimization"
This reverts commit 511d218c60.

In current form intrinsics break borg build.

Change-Id: Ied37936af841250ecff449802e69a3d3761c91b9
2014-01-10 13:38:26 +00:00
Jingning Han
d66c748635 Enable skipping reference frame check in rd loop
This commit allows encoder to compare the SAD cost associated with
the best motion vector predictor, per frame. If one reference frame
has this cost more than 4 times of the best SAD cost given by other
reference frames, skip NEARESTMV, NEARMV, ZEROMV mode check of this
reference frame.

This setting is turned on in speed 2 and above. Compression quality
change in speed 2:
derf  -0.014%
yt    -0.097%
hd    -0.023%
stdhd  0.046%

It reduces the speed 2 runtime of test sequences:
pedestrian_area_1080p 4000 kbps 310763 ms -> 303595 ms
bluesky_1080p 6000 kbps         259852 ms -> 251920 ms

Change-Id: I7f59cf79503d51836d61d56d50dc5bdf0e502e22
2014-01-09 18:25:53 -08:00
Jingning Han
a4c94a94cc Merge "Optimze inv 16x16 DCT with 10 non-zero coeffs - P2" 2014-01-09 18:17:25 -08:00
Jingning Han
faa2ba86cc Merge "Optimze inv 16x16 DCT with 10 non-zero coeffs - P1" 2014-01-09 18:17:12 -08:00
Deb Mukherjee
36c8daed58 Merge "Cleanups on refresh flags" 2014-01-09 17:38:45 -08:00
Deb Mukherjee
412e4954c1 Cleanups on refresh flags
Cleanups on frame refresh flags and external overrides.

Change-Id: Ia6a56fe1bde906b1dc3fcbf4ef1c7b207cd2df2d
2014-01-09 17:00:23 -08:00
Johann
e8192cf633 Merge "Use the correct member for initialization" 2014-01-09 15:21:19 -08:00
Yaowu Xu
b1d81e19d8 Merge "Simplify set_rt_speed_feature()" 2014-01-09 15:02:24 -08:00
Marco Paniconi
193fa5c8ba Keep buffer clipped to maximum in change_config.
Under a configuration change, where the bitrate suddenly decreases,
the buffer level may be larger than maximum allowed (for that first frame to be encoded after change_config).
This change keeps it clipped to its maximum level.

Change-Id: I4d0b5b3d1fd8148600dd39e02bd630c9464baba5
2014-01-09 14:33:40 -08:00
Dmitry Kovalev
c8e8d3a461 Merge "Renaming 'Sharpness' to 'sharpness'." 2014-01-09 13:42:55 -08:00
Yaowu Xu
2d381d76d8 Simplify set_rt_speed_feature()
1. Made speed choices to be progressive
2. Adjusted rt speed settings to achieve better speed/quality

Overall, rt-5 gained 2.5% in compression/quality, encoding time of 720p
niklas clip goes from 137,052ms to 121,874ms

Change-Id: Ia6e7e1e15225395a868a2f1059c3db8e266e1600
2014-01-09 13:02:15 -08:00
Jingning Han
af31b27aae Optimze inv 16x16 DCT with 10 non-zero coeffs - P2
This commit further optimizes SSE2 operations in the second 1-D
inverse 16x16 DCT, with (<10) non-zero coefficients. The average
runtime of this module goes down from 779 cycles -> 725 cycles.

Change-Id: Iac31b123640d9b1e8f906e770702936b71f0ba7f
2014-01-09 12:46:09 -08:00
Yunqing Wang
f3b9b97c0e Merge "SSSE3 convolution optimization" 2014-01-09 12:39:47 -08:00
levytamar82
511d218c60 SSSE3 convolution optimization
Optimizing all SSSE3 assembly for convolution:
1. vp9_filter_block1d4_h8_sse2
2. vp9_filter_block1d8_h8_sse2
3. vp9_filter_block1d16_h8_sse2
4. vp9_filter_block1d4_v8_sse2
5. vp9_filter_block1d8_v8_sse2
6. vp9_filter_block1d16_v8_sse2
my optimization include:
-processing 2x8 elements in one 128 bit register instead of processing
8 elements in one 128 bit register.
-removing unecessary loads.
This optimization gives between 2.4% user level gain for 480p input
and 1.6% user level gain for 720p.
This Optimization done only for 64bit.

Change-Id: Icb586dc0c938b56699864fcee6c52fd43b36b969
2014-01-09 12:27:51 -07:00
Dmitry Kovalev
6d812d6f24 Merge "Removing examples code generation and making them static." 2014-01-09 11:15:46 -08:00
Dmitry Kovalev
42647fc9fe Merge "Using VP9_COMMON instead of VP9_COMP." 2014-01-09 11:15:29 -08:00
Johann
c8a2aaa7e7 Merge "VP8 for ARMv8 by using NEON intrinsics 01" 2014-01-09 10:39:05 -08:00
James Yu
79395e16cf VP8 for ARMv8 by using NEON intrinsics 01
Add bilinearpredict_neon_intrinsics.c
- vp8_bilinear_predict4x4_neon
- vp8_bilinear_predict8x4_neon
- vp8_bilinear_predict8x8_neon
- vp8_bilinear_predict16x16_neon

Change-Id: I33dfa502881219841b442dda32b73220e51b716b
Signed-off-by: James Yu <james.yu@linaro.org>
2014-01-09 09:56:22 -08:00