generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
8,550 Commits 55 Branches 16 Tags 63 MiB
bb61327b98
Commit Graph

4391 Commits

Author SHA1 Message Date
Paul Wilkins
bb61327b98 Merge "vp9_rdopt.c: Use int64_t for dist_sum and rate_sum in model_rd_for_sb()." 2014-02-20 02:03:49 -08:00
Dmitry Kovalev
3ba5df4911 Merge "Adding is_mv_valid() function." 2014-02-19 13:39:21 -08:00
Alex Converse
43b7959dbc Merge "Do not use the value of 'rate_mv' if uninitialized." 2014-02-19 12:41:14 -08:00
James Zern
10f2db2b1f Merge "vp9: normalize DECLARE_ALIGNED use on global tables" 2014-02-19 11:38:47 -08:00
Paul Wilkins
98bf47df16 Merge "Remove comment from calls to vp9_clear_system_state." 2014-02-19 05:14:51 -08:00
Dmitry Kovalev
9b75f381cf Adding is_mv_valid() function. 
Change-Id: I9d036244b558765b252d8c6681b22721cb2e51bb
 2014-02-19 13:57:18 +01:00
Dmitry Kovalev
a7f8624458 Cleaning up vp9_onyx_if.c. 
Change-Id: I389790f1ad0dd40816a7ae1ddad4a5c46278cff3
 2014-02-19 10:59:53 +01:00
Dmitry Kovalev
35bd886864 Merge "Cleaning up pack_inter_mode_mvs() function." 2014-02-19 01:04:36 -08:00
Dmitry Kovalev
4e06ee39b0 Merge "Removing unused tree_update_hist[] array." 2014-02-19 01:02:51 -08:00
Dmitry Kovalev
54164c0e07 Merge "Cleaning up vp9_encodemb.c." 2014-02-19 01:02:30 -08:00
James Zern
b78c219c80 vp9: normalize DECLARE_ALIGNED use on global tables 
- place extern within the macro
- use in the header only

Change-Id: I4274b345d8af9ef329c0eb9553a3ddaad70d1d26
 2014-02-18 22:57:43 -08:00
Tom Finegan
a0e495579e vp9_rdopt.c: Use int64_t for dist_sum and rate_sum in model_rd_for_sb(). 
Removes unnecessary casts and silences warnings.

Change-Id: Ic516a909a2b3748831be6717f02d86ca60190af3
 2014-02-18 22:31:11 -08:00
Yaowu Xu
f4ce20c530 Removed unnecessary type cast 
Change-Id: I0f2d5155ec042e9b5cbe9a9d6b186b121ed7737f
 2014-02-18 21:03:47 -08:00
James Zern
26c8e720ca Merge "vp9_filter: move table alignment decl's to header" 2014-02-18 20:15:33 -08:00
Alex Converse
2ceaf10e07 Do not use the value of 'rate_mv' if uninitialized. 
This happens when exiting single_motion_search() early due to adaptive_motion_search.

Change-Id: Ic396e7e31e59ff219bb66a459c873c9e5b17400d
 2014-02-18 17:04:42 -08:00
Adrian Grange
9b9c33a2b8 Remove comment from calls to vp9_clear_system_state. 
Removed the comment "// __asm emms;" from all calls
to vp9_clear_system_state.

Change-Id: Ib4dae69c1cdf3f185b04184a870cd70c4d703f10
 2014-02-18 15:43:42 -08:00
Jingning Han
e3178d0c1c Merge "Enable sub-pixel motion search for rtc mode" 2014-02-18 14:52:13 -08:00
Jingning Han
f11a85127b Merge "Initialize internal buffer pointers in rtc mode" 2014-02-18 14:51:38 -08:00
Yunqing Wang
0cc71c9c9f Merge "SSSE3 convolution optimization" 2014-02-18 12:55:34 -08:00
Yunqing Wang
ad8d4454f0 Merge "AVX2 SubPixel Variance Optimization" 2014-02-18 12:18:13 -08:00
Jingning Han
6478673933 Enable sub-pixel motion search for rtc mode 
Run sub-pixel motion search when NEWMV gives lower rate-distortion
cost. This improves coding performance of derf set by 8%, std-hd by
2.2%.

Change-Id: Ife50f7fda8463927784fe59a41cc439c833e941a
 2014-02-18 12:07:55 -08:00
Johann
201398c3cd Merge "Replace vqshrun by vqmovun if shift #0 bit" 2014-02-17 20:53:12 -08:00
Yaowu Xu
671514ebee Merge "vp9_onyx_if.c resolve vs warnings" 2014-02-17 08:37:25 -08:00
Adrian Grange
bed0be5b22 Merge "Change post- to pre-increment." 2014-02-17 07:27:32 -08:00
Dmitry Kovalev
9141272235 Cleaning up vp9_encodemb.c. 
Change-Id: Ia5b35494e4e18758f9f6c172526db83a87fdaa26
 2014-02-17 12:34:02 +01:00
Jim Bankoski
6c5f6b306b vp9_onyx_if.c resolve vs warnings 
Change-Id: I02cc50648b62768e20d2f465b5ad09648b1abe93
 2014-02-15 21:02:18 -08:00
James Zern
cca8a54cdd Merge "A couple more V.S. warnings silenced." 2014-02-15 10:32:57 -08:00
James Zern
8092080216 vp9_filter: move table alignment decl's to header 
avoids mismatched alignment warnings in visual studio builds

Change-Id: I2cedb8042fd47e708bde3f7168a6fb4bd9aaa569
 2014-02-15 10:18:24 -08:00
Jim Bankoski
24025c0477 vp9_cx_iface vp9_dx_iface vpxdec vs warnings 
Change-Id: I747982e7d1157a8b45f4034ddf207306f9f957e0
 2014-02-15 09:14:02 -08:00
Tom Finegan
6a6c427710 vp9/encoder/vp9_onyx_if: Fix compute_qdelta_by_rate() warnings 
- Rename and make static
  s/vp9_compute_qdelta_by_rate/compute_qdelta_by_rate/
- Make base_q_index an integer.
- Add a cast.

Change-Id: Iea8d1397fd2717e7373b182ec51f5db960ef2cca
 2014-02-14 21:14:33 -08:00
James Yu
e486488ce8 Replace vqshrun by vqmovun if shift #0 bit 
Change-Id: Ifabb8c7ec0c327fea9d6739cab10addb060ff435
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-02-14 21:03:40 -08:00
Paul Wilkins
8618c70683 A couple more V.S. warnings silenced. 
Change-Id: Ica1b583d69810182f621de757d2543b2a3b35566
 2014-02-14 20:34:14 -08:00
Johann
4378503665 Merge "Remove redundant arm neon instructions." 2014-02-14 20:02:51 -08:00
Dmitry Kovalev
a6bd65080f Removing unused tree_update_hist[] array. 
Change-Id: If2e8e06c74d065323523904ff1771590fd48cabc
 2014-02-14 17:22:34 -08:00
Adrian Grange
120f2144d2 Change post- to pre-increment. 
Change-Id: I29e4d3a3c8cceb18825371bb78c83c0e7dd86e0f
 2014-02-14 16:15:13 -08:00
Adrian Grange
d2095e0c6d Merge "Cleanup some comments." 2014-02-14 16:14:58 -08:00
levytamar82
52dac5d1cb AVX2 SubPixel Variance Optimization 
Optimizing 2 functions to process 32 elements in parallel instead of 16:
1. vp9_sub_pixel_variance64x64
2. vp9_sub_pixel_variance32x32
both of those function were calling vp9_sub_pixel_variance16xh_ssse3
instead of calling that function, it calls vp9_sub_pixel_variance32xh_avx2
that is written in avx2 and process 32 elements in parallel.
This Optimization gave 70% function level gain and 2% user level gain

Change-Id: I4f5cb386b346ff6c878a094e1c3b37e418e50bde
 2014-02-14 16:59:11 -07:00
Deb Mukherjee
0876302e4e Merge "Make cpi->speed always positive" 2014-02-14 15:53:30 -08:00
Deb Mukherjee
5306eda9da Merge "Enforce max-intra-pct in one-pass cbr mode" 2014-02-14 15:53:19 -08:00
Adrian Grange
b7be30eb36 Cleanup some comments. 
Change-Id: I568861ba1d43620865ad9a98a97eef37a51fd856
 2014-02-14 15:05:30 -08:00
Yaowu Xu
ecf392a155 Merge "minor spelling cleanup in comments" 2014-02-14 14:29:35 -08:00
Deb Mukherjee
56835f1bdd Make cpi->speed always positive 
Minor fix to ensure that positive and negative cpu_used
values produce the same output.

Change-Id: I9157d05cb9a7c0bd57a4d8a9f268205a825b9b0b
 2014-02-14 14:28:58 -08:00
Deb Mukherjee
12ae6eaf21 Enforce max-intra-pct in one-pass cbr mode 
This was inadvertently left out in a previous refactoring step.

Change-Id: Idf3350e95152f876a1a1a4591dd30c8981856c96
 2014-02-14 14:27:02 -08:00
Jim Bankoski
323a7120b9 Merge "vp9_rdopt.c : Silence more V.S. compiler warnings." 2014-02-14 14:23:51 -08:00
Jim Bankoski
f25dec89b8 Merge "vp9_bitstream.c resolve warnings in vs" 2014-02-14 14:17:39 -08:00
Jim Bankoski
57704bff94 Merge "vp9_pickmode.c resolve vs issue" 2014-02-14 14:16:18 -08:00
Jim Bankoski
8dc39a7540 Merge "vp9_vaq.c resolve vs warnings" 2014-02-14 14:15:38 -08:00
levytamar82
3068d7d944 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 is done only for 64 bit

Change-Id: Ic07fce2f9360329b4f2d956efda1480ae958766b
 2014-02-14 15:08:42 -07:00
Paul Wilkins
40e374bbae vp9_rdopt.c : Silence more V.S. compiler warnings. 
Change-Id: I151c6c9396b8ea1ce2e05b80359f60f0d38af138
 2014-02-14 11:28:37 -08:00
Jim Bankoski
00717e804c vp9_bitstream.c resolve warnings in vs 
Change-Id: I65677b7822204dc7344c67b8080e90633afc13dc
 2014-02-14 11:19:39 -08:00