Commit Graph

7278 Commits

Author SHA1 Message Date
Frank Galligan
97d1258375 Revert "Add 16 wide neon horz loopfilter."
The change caused mismatches with some test vectors on neon.

Original CL: https://gerrit.chromium.org/gerrit/#/c/67863/

Change-Id: I913891636d53783e93cb1865ca78ded1821dc4b0
2013-11-21 14:01:33 -08:00
Yunqing Wang
e002bb99a8 Merge "Add filter_selectively_vert_row2 to enable parallel loopfiltering" 2013-11-21 11:25:55 -08:00
hkuang
370bf116a2 Merge "Remove unnecessary eob checking." 2013-11-21 11:24:02 -08:00
Frank Galligan
2dd77580c0 Merge "Add 16 wide neon horz loopfilter." 2013-11-21 10:29:30 -08:00
Yunqing Wang
b5e6d6cccf Add filter_selectively_vert_row2 to enable parallel loopfiltering
Added filter_selectively_vert_row2 to be ready for parallel
loopfiltering in vertical direction. This change did 2-row
filtering at a time. If 2 vertically adjacent 8x8 blocks do same
type of filtering, we can do 16-pixel filtering in parallel.

Next, we need to provide 16-pixel loopfiltering functions in c
and optimized versions for codec speedup.

Change-Id: Idf97bbdd70566e55bd30e1fd25cb8544e33291be
2013-11-21 09:53:15 -08:00
Yunqing Wang
6c4964602a Merge "Correct ssse3 8/16-pixel wide sub-pixel filter calculation" 2013-11-21 09:40:02 -08:00
Frank Galligan
98de15137e Add 16 wide neon horz loopfilter.
Add support to do 16 pixel horizontal filtering in Neon.
Nexus devices saw about 0.5% decode speed increase.

Change-Id: I2993f6c2d49f31fa74976879eeaa289fd3f4e15d
2013-11-21 09:39:36 -08:00
Tom Finegan
44dd3274da vpxenc: Warn users about incorrect quantizer settings.
Also, clean up stylistically questionable code near my changes.

Change-Id: I92c96a274cb339b7b74174a608f94ae86aba8354
2013-11-20 17:18:28 -08:00
Dmitry Kovalev
77a865d970 Merge "Removing old code." 2013-11-20 14:43:03 -08:00
Dmitry Kovalev
a218a96784 Merge "Adding MV_FP_SIZE constant." 2013-11-20 14:39:58 -08:00
Dmitry Kovalev
d54893da1d Merge "Using is_inter_block() and has_second_ref() functions." 2013-11-20 14:39:50 -08:00
Dmitry Kovalev
87ff7f2af3 Removing old code.
Change-Id: I67d1681c7b17661deb792c5e6a9e2014a73ff9b7
2013-11-20 14:05:21 -08:00
Yunqing Wang
256cf7ee7d Correct ssse3 8/16-pixel wide sub-pixel filter calculation
Although no mismatch was indicated for 8/16 wide sub-pixel filters
in issue 661, they had similar problems that could cause mismatch
potentially. This patch fixed calculations in HORIZx8/16
and VERTx8/16.

Change-Id: I169961c9d40a20340995b7d22aafc89ccf30bfca
2013-11-20 12:52:56 -08:00
Jim Bankoski
302c33e49f Merge "Clean up removal of vp9_pareto8 table." 2013-11-20 12:30:03 -08:00
Dmitry Kovalev
1a69eed2c4 Using is_inter_block() and has_second_ref() functions.
Change-Id: Iadd771a33c8874f3b774923bca4da3c8fe5429ee
2013-11-20 12:08:10 -08:00
Dmitry Kovalev
4956fcd31b Adding MV_FP_SIZE constant.
Change-Id: I98d750ee92ff51fb714980418ea28be3b1d0f3c6
2013-11-20 12:07:57 -08:00
Yunqing Wang
82f8ae4c5b Merge "Support for extended feature flags enumeration leaf in CPUID instruction" 2013-11-20 12:01:39 -08:00
hkuang
6debc446e0 Remove unnecessary eob checking.
Change-Id: Ia568f70bddc1a2b62141a0197459119ca74c22b5
2013-11-20 11:58:11 -08:00
Jim Bankoski
25aae73a30 Merge "remove the model and copy in pack_mb_tokens" 2013-11-20 11:34:30 -08:00
Jim Bankoski
5bbb0c6295 Clean up removal of vp9_pareto8 table.
Change-Id: I5556e8d1fc150be8a3e93af21900829b59a500dc
2013-11-20 11:17:26 -08:00
Erik Niemeyer
9f26861147 Support for extended feature flags enumeration leaf in CPUID instruction
This CL fixes an overcite with the AVX2 support CL previously
merged (Change-Id: Idc03f3fca4bf2d0afd33631ea1d3caf8fc34ec29) that
prevented runtime execution of AVX2 code in WebM. 

Background:
Starting with the Sandybridge processor, the CPUID instruction was
enhanced to add various extended feature flag enumeration leaves.
Reading these leaves requires an additional input value for the CPUID
instruction which is stored in ECX. This change adds this second input
value for all ARCH_X86 and ARCH_x86_64 targets to the CPUID macros,
allowing checks of EBX bit 5 for AVX2 support. This capability will be
required moving forward to check for future processor features.

Change-Id: Ie9d872bc9ff68dad4b6578e4544e4dfd0ae26c36
2013-11-20 11:10:54 -08:00
Jingning Han
81b9fd4310 Merge "Take out assertion from inverse transforms" 2013-11-20 10:55:27 -08:00
Jim Bankoski
03276bf6e6 remove the model and copy in pack_mb_tokens
Change-Id: I00a5203c8ed76c184d936fccf93d76e7c06773d3
2013-11-20 10:06:04 -08:00
Yunqing Wang
0ef63f596d Fix stack pointer in sub-pixel filters
In commit "3d50da5397d20abc932d81453b26cde758293a40", the stack
pointer was modified while aligning the stack, and it needed to
be pop out at the end.

Change-Id: I062971e195f1f2ab9d0ab5fb84dcf215a0fcaa67
2013-11-20 09:42:44 -08:00
Guillaume Martres
b00057c88a Merge "vpxenc: add --aq-mode flag to control adaptive quantization" 2013-11-20 08:13:28 -08:00
Marco Paniconi
d486427cf1 Undo the vp8 change in "Reduce loop filter in..."
Patch in https://gerrit.chromium.org/gerrit/#/c/41176/
was merged into repository by mistake.

Change-Id: I235c71af26bb2d72698c8aac2301e5a7e9c5f960
2013-11-19 17:16:00 -08:00
Jim Bankoski
7a8a68e2bd Merge "scan order table lookup same for encoder and decoder" 2013-11-19 16:22:48 -08:00
Yunqing Wang
e8f8e77642 Merge "Fix decoder mismatch with ssse3 enabled" 2013-11-19 16:19:32 -08:00
Jingning Han
75673cfc3d Merge "Use restore_dst_buf in handle_inter_mode" 2013-11-19 16:19:04 -08:00
Dmitry Kovalev
e8346f8cf7 Merge "Cleaning up probability/cost functions." 2013-11-19 16:08:16 -08:00
Yaowu Xu
dd04ff506b Merge "Move vp9_setup_interp_filter() to encoder" 2013-11-19 16:01:19 -08:00
Jingning Han
82c32fe1b5 Use restore_dst_buf in handle_inter_mode
There are many places in handle_inter_mode that need to restore the
dst buffer pointers, due to buffer pointer swap and early rd search
breakout. This commit wraps these operations into an inline function
for clean-up.

Change-Id: I0462e8c41c8bc3cd8db07395489cac03d8e5be54
2013-11-19 15:33:16 -08:00
Jim Bankoski
d6667dd54f scan order table lookup same for encoder and decoder
Change-Id: I473947b5ca70b7a81151926284bff86f8555492a
2013-11-19 15:31:43 -08:00
Tom Finegan
a944deae97 Merge "vpxdec: Relocate WebM input support." 2013-11-19 15:30:41 -08:00
Yunqing Wang
3d50da5397 Fix decoder mismatch with ssse3 enabled
This patch fixed issue 661: "Decoder produces mismatched outputs
with ssse3 enabled and disabled." In sub-pixel filters, a pixel
value was multiplied by a filter coefficient, and the results
were added up. The order of adding up these multiplications had to
be arranged carefully to prevent incorrect overflowing.

Change-Id: Id08af4200fea9e1b896fc40157b8651c2c7e80f2
2013-11-19 15:10:04 -08:00
Dmitry Kovalev
65cee2f01a Merge "Simplifying partition context calculation." 2013-11-19 15:09:01 -08:00
Dmitry Kovalev
86b629efb6 Merge "Calculating dst pointer only once per transform block." 2013-11-19 15:05:46 -08:00
Dmitry Kovalev
e55d3e6d1b Cleaning up probability/cost functions.
Change-Id: Ifad4b0e6355ce49fcc6f470becc080e8069452ee
2013-11-19 14:59:12 -08:00
Jim Bankoski
60aba6558f Merge "entropy code speedup" 2013-11-19 14:58:44 -08:00
Yaowu Xu
df78fea166 Move vp9_setup_interp_filter() to encoder
As it is used in encoder only.

Change-Id: I5f2a8abbe72bb18cbf6ce36a3dc7e132aeae8ec2
2013-11-19 14:57:58 -08:00
Jim Bankoski
69541e1dec Merge "Reduce loop filter in cyclic refresh." 2013-11-19 14:42:32 -08:00
Yaowu Xu
f92cfa1ca6 Merge "Move vp9_sadmxn.h from common to encoder" 2013-11-19 14:41:33 -08:00
Jim Bankoski
8cf352abac entropy code speedup
Change-Id: Ic316d3374ff9a2b43897272260947d56765a0fdd
2013-11-19 14:31:38 -08:00
Jim Bankoski
ff4f1c4b76 scan order / neighbors converted to lookup
Change-Id: I64b189dfeee1cf3e90134a1a93497072f3361e5e
2013-11-19 12:55:44 -08:00
Yaowu Xu
30b03050a2 Move vp9_sadmxn.h from common to encoder
Change-Id: I6f6ba91b1b8b280902b171472314d665aa0baf0b
2013-11-19 12:46:08 -08:00
Yaowu Xu
5b3a8563de Merge "Fix a bug in vpxenc reading raw input frame" 2013-11-19 11:26:02 -08:00
Dmitry Kovalev
f6ec323906 Simplifying partition context calculation.
Reversing bit order of partition_context_lookup, and modifying accordingly
update_partition_context() and partition_plane_context().

Change-Id: I64a11f1a94962a3bf217de2f50698cb781db71a5
2013-11-19 11:17:30 -08:00
Johann
296b4be859 Merge "Disable avx/avx2 for Visual Studio 2010" 2013-11-19 11:17:20 -08:00
Yunqing Wang
f16fb829e6 Merge "Improve vp9_iht4x4_16_add_sse2 (x1.341)" 2013-11-19 11:11:47 -08:00
Yaowu Xu
a7c7e78aa5 Fix a bug in vpxenc reading raw input frame
The bug was introduced in 00a35aab. The reading of raw yuv input frame
was off by 4 bytes.

Change-Id: I6923ea5528aa529a47a06b64adca8f94847f19a6
2013-11-19 10:28:30 -08:00