]> granicus.if.org Git - libvpx/log
libvpx
10 years agoMerge "Fix the cpuid macro for x86_64 non-gcc build"
Yaowu Xu [Fri, 22 Nov 2013 18:03:51 +0000 (10:03 -0800)]
Merge "Fix the cpuid macro for x86_64 non-gcc build"

10 years agoMerge changes Id1698a35,Idcabd0b9
Jim Bankoski [Fri, 22 Nov 2013 16:16:17 +0000 (08:16 -0800)]
Merge changes Id1698a35,Idcabd0b9

* changes:
  detokenization speedups
  Don't write 0's to token_cache

10 years agoMerge "Refactoring of rate control - part 1"
Deb Mukherjee [Fri, 22 Nov 2013 16:06:48 +0000 (08:06 -0800)]
Merge "Refactoring of rate control - part 1"

10 years agoRefactoring of rate control - part 1
Deb Mukherjee [Wed, 6 Nov 2013 21:13:59 +0000 (13:13 -0800)]
Refactoring of rate control - part 1

Moves all rate control variables to a separate structure,
removes some currently unused variables,
moves some rate control functions to vp9_ratectrl.c,
and splits the encode_frame_to_data_rate function.

Change-Id: I4ed54c24764b3b6de2dd676484f01473724ab52b

10 years agoFix the cpuid macro for x86_64 non-gcc build
Yaowu Xu [Fri, 22 Nov 2013 01:39:33 +0000 (17:39 -0800)]
Fix the cpuid macro for x86_64 non-gcc build

Change-Id: I0c44800db10db8d74c1ddfe89abecfd1c53d0f8d

10 years agoMerge "vpxenc: Add vpxenc.h and move/rename the global_config struct"
Tom Finegan [Fri, 22 Nov 2013 01:56:26 +0000 (17:56 -0800)]
Merge "vpxenc: Add vpxenc.h and move/rename the global_config struct"

10 years agodetokenization speedups
Jim Bankoski [Fri, 22 Nov 2013 00:55:22 +0000 (16:55 -0800)]
detokenization speedups

removed unnecessary ifs and branches ..

Change-Id: Id1698a35292659388f48926791024d1400f2cea9

10 years agoMerge "Using num_4x4_blocks_* instead of b_{width, height}_log2."
Dmitry Kovalev [Fri, 22 Nov 2013 00:48:34 +0000 (16:48 -0800)]
Merge "Using num_4x4_blocks_* instead of b_{width, height}_log2."

10 years agovpxenc: Add vpxenc.h and move/rename the global_config struct
Tom Finegan [Fri, 22 Nov 2013 00:46:40 +0000 (16:46 -0800)]
vpxenc: Add vpxenc.h and move/rename the global_config struct

- Rename the struct to VpxEncoderConfig.
- The idea behind this is to enable checking the global settings against
  stream specific settings in source files other than vpxenc.c.

Change-Id: Ic736cbb714845b9466acb34671780d65b83ad1a8

10 years agoMerge "Removing plane_block_{width, height} functions."
Dmitry Kovalev [Fri, 22 Nov 2013 00:37:27 +0000 (16:37 -0800)]
Merge "Removing plane_block_{width, height} functions."

10 years agoMerge "Using txfrm_block_to_raster_xy() in encoder."
Dmitry Kovalev [Fri, 22 Nov 2013 00:24:22 +0000 (16:24 -0800)]
Merge "Using txfrm_block_to_raster_xy() in encoder."

10 years agoUsing num_4x4_blocks_* instead of b_{width, height}_log2.
Dmitry Kovalev [Thu, 21 Nov 2013 23:53:06 +0000 (15:53 -0800)]
Using num_4x4_blocks_* instead of b_{width, height}_log2.

Change-Id: I9ea3946c17b19f511565cd771037abe7db8b3ddb

10 years agoMerge "Removing PARAMS macro for consistency"
Joshua Litt [Thu, 21 Nov 2013 23:06:51 +0000 (15:06 -0800)]
Merge "Removing PARAMS macro for consistency"

10 years agoMerge "Revert "Add 16 wide neon horz loopfilter.""
Frank Galligan [Thu, 21 Nov 2013 23:06:17 +0000 (15:06 -0800)]
Merge "Revert "Add 16 wide neon horz loopfilter.""

10 years agoRevert "Add 16 wide neon horz loopfilter."
Frank Galligan [Thu, 21 Nov 2013 22:01:33 +0000 (14:01 -0800)]
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

10 years agoDon't write 0's to token_cache
Jim Bankoski [Thu, 21 Nov 2013 20:52:15 +0000 (12:52 -0800)]
Don't write 0's to token_cache

This code only updates the token_cache if the result is non0.

Change-Id: Idcabd0b993a926fea9c29dbec134b9c5c4859b40

10 years agoSyncing update_coef_probs() implementation with decoder.
Dmitry Kovalev [Thu, 21 Nov 2013 20:36:02 +0000 (12:36 -0800)]
Syncing update_coef_probs() implementation with decoder.

Using for loop based on max_tx_size instead of separate checks. Combining
build_coeff_contexts() with update_coef_probs().

Change-Id: Ie335a7db29830677fbc14478a9c190d3c1068665

10 years agoMerge "Add filter_selectively_vert_row2 to enable parallel loopfiltering"
Yunqing Wang [Thu, 21 Nov 2013 19:25:55 +0000 (11:25 -0800)]
Merge "Add filter_selectively_vert_row2 to enable parallel loopfiltering"

10 years agoMerge "Remove unnecessary eob checking."
hkuang [Thu, 21 Nov 2013 19:24:02 +0000 (11:24 -0800)]
Merge "Remove unnecessary eob checking."

10 years agoMerge "Add 16 wide neon horz loopfilter."
Frank Galligan [Thu, 21 Nov 2013 18:29:30 +0000 (10:29 -0800)]
Merge "Add 16 wide neon horz loopfilter."

10 years agoAdd filter_selectively_vert_row2 to enable parallel loopfiltering
Yunqing Wang [Fri, 15 Nov 2013 19:04:09 +0000 (11:04 -0800)]
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

10 years agoMerge "Correct ssse3 8/16-pixel wide sub-pixel filter calculation"
Yunqing Wang [Thu, 21 Nov 2013 17:40:02 +0000 (09:40 -0800)]
Merge "Correct ssse3 8/16-pixel wide sub-pixel filter calculation"

10 years agoAdd 16 wide neon horz loopfilter.
Frank Galligan [Tue, 19 Nov 2013 18:33:32 +0000 (10:33 -0800)]
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

10 years agovpxenc: Warn users about incorrect quantizer settings.
Tom Finegan [Thu, 21 Nov 2013 01:18:28 +0000 (17:18 -0800)]
vpxenc: Warn users about incorrect quantizer settings.

Also, clean up stylistically questionable code near my changes.

Change-Id: I92c96a274cb339b7b74174a608f94ae86aba8354

10 years agoMerge "Removing old code."
Dmitry Kovalev [Wed, 20 Nov 2013 22:43:03 +0000 (14:43 -0800)]
Merge "Removing old code."

10 years agoMerge "Adding MV_FP_SIZE constant."
Dmitry Kovalev [Wed, 20 Nov 2013 22:39:58 +0000 (14:39 -0800)]
Merge "Adding MV_FP_SIZE constant."

10 years agoMerge "Using is_inter_block() and has_second_ref() functions."
Dmitry Kovalev [Wed, 20 Nov 2013 22:39:50 +0000 (14:39 -0800)]
Merge "Using is_inter_block() and has_second_ref() functions."

10 years agoRemoving old code.
Dmitry Kovalev [Wed, 20 Nov 2013 22:05:21 +0000 (14:05 -0800)]
Removing old code.

Change-Id: I67d1681c7b17661deb792c5e6a9e2014a73ff9b7

10 years agoUsing txfrm_block_to_raster_xy() in encoder.
Dmitry Kovalev [Wed, 20 Nov 2013 21:58:21 +0000 (13:58 -0800)]
Using txfrm_block_to_raster_xy() in encoder.

Change-Id: Ibe847000467fe46bf8ce87d8f1ef8f2d5ad1eaf4

10 years agoCorrect ssse3 8/16-pixel wide sub-pixel filter calculation
Yunqing Wang [Wed, 20 Nov 2013 20:52:56 +0000 (12:52 -0800)]
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

10 years agoRemoving plane_block_{width, height} functions.
Dmitry Kovalev [Wed, 20 Nov 2013 20:39:29 +0000 (12:39 -0800)]
Removing plane_block_{width, height} functions.

Change-Id: I29c0dfcf41a1253d5e2a0d2ff740c0c38ebaa5a2

10 years agoMerge "Clean up removal of vp9_pareto8 table."
Jim Bankoski [Wed, 20 Nov 2013 20:30:03 +0000 (12:30 -0800)]
Merge "Clean up removal of vp9_pareto8 table."

10 years agoUsing is_inter_block() and has_second_ref() functions.
Dmitry Kovalev [Wed, 20 Nov 2013 04:25:55 +0000 (20:25 -0800)]
Using is_inter_block() and has_second_ref() functions.

Change-Id: Iadd771a33c8874f3b774923bca4da3c8fe5429ee

10 years agoAdding MV_FP_SIZE constant.
Dmitry Kovalev [Wed, 20 Nov 2013 04:18:01 +0000 (20:18 -0800)]
Adding MV_FP_SIZE constant.

Change-Id: I98d750ee92ff51fb714980418ea28be3b1d0f3c6

10 years agoMerge "Support for extended feature flags enumeration leaf in CPUID instruction"
Yunqing Wang [Wed, 20 Nov 2013 20:01:39 +0000 (12:01 -0800)]
Merge "Support for extended feature flags enumeration leaf in CPUID instruction"

10 years agoRemove unnecessary eob checking.
hkuang [Wed, 20 Nov 2013 19:22:00 +0000 (11:22 -0800)]
Remove unnecessary eob checking.

Change-Id: Ia568f70bddc1a2b62141a0197459119ca74c22b5

10 years agoMerge "remove the model and copy in pack_mb_tokens"
Jim Bankoski [Wed, 20 Nov 2013 19:34:30 +0000 (11:34 -0800)]
Merge "remove the model and copy in pack_mb_tokens"

10 years agoClean up removal of vp9_pareto8 table.
Jim Bankoski [Wed, 20 Nov 2013 19:17:26 +0000 (11:17 -0800)]
Clean up removal of vp9_pareto8 table.

Change-Id: I5556e8d1fc150be8a3e93af21900829b59a500dc

10 years agoSupport for extended feature flags enumeration leaf in CPUID instruction
Erik Niemeyer [Wed, 20 Nov 2013 04:11:57 +0000 (21:11 -0700)]
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

10 years agoMerge "Take out assertion from inverse transforms"
Jingning Han [Wed, 20 Nov 2013 18:55:27 +0000 (10:55 -0800)]
Merge "Take out assertion from inverse transforms"

10 years agoremove the model and copy in pack_mb_tokens
Jim Bankoski [Wed, 20 Nov 2013 18:06:04 +0000 (10:06 -0800)]
remove the model and copy in pack_mb_tokens

Change-Id: I00a5203c8ed76c184d936fccf93d76e7c06773d3

10 years agoFix stack pointer in sub-pixel filters
Yunqing Wang [Wed, 20 Nov 2013 17:42:44 +0000 (09:42 -0800)]
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

10 years agoMerge "vpxenc: add --aq-mode flag to control adaptive quantization"
Guillaume Martres [Wed, 20 Nov 2013 16:13:28 +0000 (08:13 -0800)]
Merge "vpxenc: add --aq-mode flag to control adaptive quantization"

10 years agoUndo the vp8 change in "Reduce loop filter in..."
Marco Paniconi [Wed, 20 Nov 2013 01:10:57 +0000 (17:10 -0800)]
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

10 years agoMerge "scan order table lookup same for encoder and decoder"
Jim Bankoski [Wed, 20 Nov 2013 00:22:48 +0000 (16:22 -0800)]
Merge "scan order table lookup  same for encoder and decoder"

10 years agoMerge "Fix decoder mismatch with ssse3 enabled"
Yunqing Wang [Wed, 20 Nov 2013 00:19:32 +0000 (16:19 -0800)]
Merge "Fix decoder mismatch with ssse3 enabled"

10 years agoMerge "Use restore_dst_buf in handle_inter_mode"
Jingning Han [Wed, 20 Nov 2013 00:19:04 +0000 (16:19 -0800)]
Merge "Use restore_dst_buf in handle_inter_mode"

10 years agoMerge "Cleaning up probability/cost functions."
Dmitry Kovalev [Wed, 20 Nov 2013 00:08:16 +0000 (16:08 -0800)]
Merge "Cleaning up probability/cost functions."

10 years agoMerge "Move vp9_setup_interp_filter() to encoder"
Yaowu Xu [Wed, 20 Nov 2013 00:01:19 +0000 (16:01 -0800)]
Merge "Move vp9_setup_interp_filter() to encoder"

10 years agoUse restore_dst_buf in handle_inter_mode
Jingning Han [Tue, 19 Nov 2013 23:29:22 +0000 (15:29 -0800)]
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

10 years agoscan order table lookup same for encoder and decoder
Jim Bankoski [Tue, 19 Nov 2013 23:31:43 +0000 (15:31 -0800)]
scan order table lookup  same for encoder and decoder

Change-Id: I473947b5ca70b7a81151926284bff86f8555492a

10 years agoMerge "vpxdec: Relocate WebM input support."
Tom Finegan [Tue, 19 Nov 2013 23:30:41 +0000 (15:30 -0800)]
Merge "vpxdec: Relocate WebM input support."

10 years agoFix decoder mismatch with ssse3 enabled
Yunqing Wang [Tue, 19 Nov 2013 22:29:25 +0000 (14:29 -0800)]
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

10 years agoMerge "Simplifying partition context calculation."
Dmitry Kovalev [Tue, 19 Nov 2013 23:09:01 +0000 (15:09 -0800)]
Merge "Simplifying partition context calculation."

10 years agoMerge "Calculating dst pointer only once per transform block."
Dmitry Kovalev [Tue, 19 Nov 2013 23:05:46 +0000 (15:05 -0800)]
Merge "Calculating dst pointer only once per transform block."

10 years agoCleaning up probability/cost functions.
Dmitry Kovalev [Tue, 19 Nov 2013 22:59:12 +0000 (14:59 -0800)]
Cleaning up probability/cost functions.

Change-Id: Ifad4b0e6355ce49fcc6f470becc080e8069452ee

10 years agoMerge "entropy code speedup"
Jim Bankoski [Tue, 19 Nov 2013 22:58:44 +0000 (14:58 -0800)]
Merge "entropy code speedup"

10 years agoMove vp9_setup_interp_filter() to encoder
Yaowu Xu [Tue, 19 Nov 2013 22:57:58 +0000 (14:57 -0800)]
Move vp9_setup_interp_filter() to encoder

As it is used in encoder only.

Change-Id: I5f2a8abbe72bb18cbf6ce36a3dc7e132aeae8ec2

10 years agoMerge "Reduce loop filter in cyclic refresh."
Jim Bankoski [Tue, 19 Nov 2013 22:42:32 +0000 (14:42 -0800)]
Merge "Reduce loop filter in cyclic refresh."

10 years agoMerge "Move vp9_sadmxn.h from common to encoder"
Yaowu Xu [Tue, 19 Nov 2013 22:41:33 +0000 (14:41 -0800)]
Merge "Move vp9_sadmxn.h from common to encoder"

10 years agoentropy code speedup
Jim Bankoski [Tue, 19 Nov 2013 22:31:38 +0000 (14:31 -0800)]
entropy code speedup

Change-Id: Ic316d3374ff9a2b43897272260947d56765a0fdd

10 years agoscan order / neighbors converted to lookup
Jim Bankoski [Tue, 19 Nov 2013 20:50:48 +0000 (12:50 -0800)]
scan order / neighbors converted to lookup

Change-Id: I64b189dfeee1cf3e90134a1a93497072f3361e5e

10 years agoMove vp9_sadmxn.h from common to encoder
Yaowu Xu [Tue, 19 Nov 2013 20:46:08 +0000 (12:46 -0800)]
Move vp9_sadmxn.h from common to encoder

Change-Id: I6f6ba91b1b8b280902b171472314d665aa0baf0b

10 years agoMerge "Fix a bug in vpxenc reading raw input frame"
Yaowu Xu [Tue, 19 Nov 2013 19:26:02 +0000 (11:26 -0800)]
Merge "Fix a bug in vpxenc reading raw input frame"

10 years agoSimplifying partition context calculation.
Dmitry Kovalev [Tue, 19 Nov 2013 19:17:30 +0000 (11:17 -0800)]
Simplifying partition context calculation.

Reversing bit order of partition_context_lookup, and modifying accordingly
update_partition_context() and partition_plane_context().

Change-Id: I64a11f1a94962a3bf217de2f50698cb781db71a5

10 years agoMerge "Disable avx/avx2 for Visual Studio 2010"
Johann [Tue, 19 Nov 2013 19:17:19 +0000 (11:17 -0800)]
Merge "Disable avx/avx2 for Visual Studio 2010"

10 years agoMerge "Improve vp9_iht4x4_16_add_sse2 (x1.341)"
Yunqing Wang [Tue, 19 Nov 2013 19:11:47 +0000 (11:11 -0800)]
Merge "Improve vp9_iht4x4_16_add_sse2 (x1.341)"

10 years agoFix a bug in vpxenc reading raw input frame
Yaowu Xu [Tue, 19 Nov 2013 18:17:04 +0000 (10:17 -0800)]
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

10 years agovpxdec: Relocate WebM input support.
Tom Finegan [Mon, 18 Nov 2013 22:39:51 +0000 (14:39 -0800)]
vpxdec: Relocate WebM input support.

- Move it to webmdec.c and webmdec.h.
- Also, tidy up obvious style nits in the vicinity of code I was
  already touching.

Change-Id: Ie2898d06e73c1e9030d9c8d465b73ee7edc3c02a

10 years agoRemoving PARAMS macro for consistency
Joshua Litt [Tue, 19 Nov 2013 01:07:55 +0000 (17:07 -0800)]
Removing PARAMS macro for consistency

Change-Id: I23ed873a6c47b15491a2ffbcdd4f0fdeef1207a0

10 years agoRemoving raster_block_offset_uint8() function.
Dmitry Kovalev [Tue, 19 Nov 2013 03:00:49 +0000 (19:00 -0800)]
Removing raster_block_offset_uint8() function.

There is no need to use that function, it is much clear to pass offset
directly to the buffer.

Change-Id: I9026cb0c5094c46f97df5d7f7daeb952f2843b24

10 years agoMerge "Finally removing txfrm_block_to_raster_block() function."
Dmitry Kovalev [Tue, 19 Nov 2013 02:43:16 +0000 (18:43 -0800)]
Merge "Finally removing txfrm_block_to_raster_block() function."

10 years agoCalculating dst pointer only once per transform block.
Dmitry Kovalev [Tue, 19 Nov 2013 02:37:53 +0000 (18:37 -0800)]
Calculating dst pointer only once per transform block.

Change-Id: I23fea0a2e85be8373600e3e2dae98d36acde389c

10 years agoMerge "Cleaning up vp9_entropy.c file."
Dmitry Kovalev [Tue, 19 Nov 2013 02:04:56 +0000 (18:04 -0800)]
Merge "Cleaning up vp9_entropy.c file."

10 years agoImprove vp9_iht4x4_16_add_sse2 (x1.341)
Abo Talib Mahfoodh [Tue, 19 Nov 2013 01:51:20 +0000 (20:51 -0500)]
Improve vp9_iht4x4_16_add_sse2 (x1.341)

This rebase is a better implementation of the previous ones.

Modifications are done to reduce the total clock cycle.
Speedup: 1.341
Compiled with -O3
Tested with: park_joy_420_720p50.y4m

Change-Id: I940eaf283f60597ca0d9d2e13d518878d55ff02d

10 years agoCleaning up vp9_entropy.c file.
Dmitry Kovalev [Tue, 19 Nov 2013 01:18:14 +0000 (17:18 -0800)]
Cleaning up vp9_entropy.c file.

Change-Id: I568f5e2d4ef2f2affe013ba1691ffb546f1fe8c6

10 years agoDecoder performance test added to unit tests
Joshua Litt [Fri, 15 Nov 2013 20:29:26 +0000 (12:29 -0800)]
Decoder performance test added to unit tests

Change-Id: Id578a5fe2039631cefd82dc2ef98cc62683194c3

10 years agoMerge "vpxdec: Include frame number when decode fails."
Tom Finegan [Tue, 19 Nov 2013 00:23:20 +0000 (16:23 -0800)]
Merge "vpxdec: Include frame number when decode fails."

10 years agovpxdec: Include frame number when decode fails.
Tom Finegan [Mon, 18 Nov 2013 23:50:58 +0000 (15:50 -0800)]
vpxdec: Include frame number when decode fails.

Change-Id: I6ea460af884d522319735e4416a2dd66c2f35d27

10 years agoMerge "Fixed a bug in commit a4a5a210"
Yaowu Xu [Mon, 18 Nov 2013 23:43:41 +0000 (15:43 -0800)]
Merge "Fixed a bug in commit a4a5a210"

10 years agoMerge "Move vp9_extend.{h,c} from common to encoder"
Yaowu Xu [Mon, 18 Nov 2013 23:43:32 +0000 (15:43 -0800)]
Merge "Move vp9_extend.{h,c} from common to encoder"

10 years agoFixed a bug in commit a4a5a210
Yaowu Xu [Mon, 18 Nov 2013 22:44:38 +0000 (14:44 -0800)]
Fixed a bug in commit a4a5a210

Commit a4a5a210 enabled lossless coding, but the commit incorrectly
disabled the usage of skip in encoder even when skip should be used.
This commit make sure that skip is enabled even in lossless mode.

Change-Id: I276954f952c6ac68f17a316ebc72f09001228a08

10 years agoDisable avx/avx2 for Visual Studio 2010
Johann [Mon, 18 Nov 2013 21:30:19 +0000 (13:30 -0800)]
Disable avx/avx2 for Visual Studio 2010

VS2010 only supports avx. There is currently no avx code
in libvpx so don't create a special case for it.

Change-Id: Iacb10ea4762155412e04f23904b4324d01451fbd

10 years agoMove vp9_extend.{h,c} from common to encoder
Yaowu Xu [Mon, 18 Nov 2013 20:36:55 +0000 (12:36 -0800)]
Move vp9_extend.{h,c} from common to encoder

Since they used in encoder only. This commit also re-order includes
for the files that include vp9_extend.h

Change-Id: I929fc113f2135d3198cd1fc6a17434e5a2f8a459

10 years agoMerge "Constrain encoder motion search range"
Jingning Han [Mon, 18 Nov 2013 20:35:34 +0000 (12:35 -0800)]
Merge "Constrain encoder motion search range"

10 years agoConstrain encoder motion search range
Jingning Han [Sat, 16 Nov 2013 04:32:03 +0000 (20:32 -0800)]
Constrain encoder motion search range

Explicitly constrain the upper limit of motion search range (in the
unit of full pixel) to be [-1023, +1023]. It is intended to control
the effective motion search range for 4K sequences.

Change-Id: I645539c70885eec0f155781f439d97d333336e88

10 years agoMerge "Do horizontal loopfiltering in parallel"
Yunqing Wang [Mon, 18 Nov 2013 18:03:41 +0000 (10:03 -0800)]
Merge "Do horizontal loopfiltering in parallel"

10 years agoMerge "Add support for VC++2013"
Yaowu Xu [Mon, 18 Nov 2013 17:32:19 +0000 (09:32 -0800)]
Merge "Add support for VC++2013"

10 years agopartition context update speedup
Jim Bankoski [Sun, 17 Nov 2013 14:58:08 +0000 (06:58 -0800)]
partition context update speedup

This removes a lot of operations in setting partition context...

Change-Id: I365e6f5607ece85190cb21443988816dfa510ce3

10 years agovpxdec: Restore IVF support.
Tom Finegan [Sat, 16 Nov 2013 16:31:20 +0000 (08:31 -0800)]
vpxdec: Restore IVF support.

Refactored IVF frame reading code out into ivf_read_frame(). Forgot
to actually make the function call in read_frame().

Change-Id: Ie9f6917e70bd26d0352a761932465c60a29a1f81

10 years agoDo horizontal loopfiltering in parallel
Yunqing Wang [Wed, 13 Nov 2013 00:51:15 +0000 (16:51 -0800)]
Do horizontal loopfiltering in parallel

This patch followed "Rewrite filter_selectively_horiz for parallel
loopfiltering" commit, and added x86 SSE2 optimization to do
16-pixel filtering in parallel. Also, corrected the declaration
of aligned arrays. For 8-pixel-in-parallel case, improved the
calculation of the masks and filters. Updated the threshold loading
since the thresholds were already duplicated. Updated neon C functions
to call neon loopfilters twice.

Using tulip clip, tests showed it gave a ~1.5% decoder speed gain.

Change-Id: Id02638626ac27a4b0e0b09d71792a24c0499bd35

10 years agoMerge "Let the idct vp9_idct32x32_34_add = vp9_idct32x32_1024_add on arm until we...
hkuang [Fri, 15 Nov 2013 23:45:49 +0000 (15:45 -0800)]
Merge "Let the idct vp9_idct32x32_34_add = vp9_idct32x32_1024_add on arm until we implenment real vp9_idct32x32_34_add_neon."

10 years agoTake out assertion from inverse transforms
Jingning Han [Fri, 15 Nov 2013 23:21:38 +0000 (15:21 -0800)]
Take out assertion from inverse transforms

Separate the rounding and right shift operations of forward transform
from those of inverse transform. Take out the assertion check from
inverse transforms. If the transform coefficients were constructed to
cause intermediate steps of inverse transform overflow, the codec will
just let it overflow without breaking the decoding flow.

Change-Id: I73cfc3706c4e840fc543a77cbc4cdb0b05d07730

10 years agoMerge "Renamed two files"
Yaowu Xu [Fri, 15 Nov 2013 23:20:23 +0000 (15:20 -0800)]
Merge "Renamed two files"

10 years agoLet the idct vp9_idct32x32_34_add = vp9_idct32x32_1024_add
hkuang [Fri, 15 Nov 2013 22:43:22 +0000 (14:43 -0800)]
Let the idct vp9_idct32x32_34_add = vp9_idct32x32_1024_add
on arm until we implenment real vp9_idct32x32_34_add_neon.

This issue is due to commit 47665452f0da3c11427ecb4852535e1787bb0c5b
Merge "Add 32x32 idct function for eob<=34 case".

Change-Id: I56b5f0abc20e7dd1bba521f78a995e85d65ea296

10 years agoMerge "Removing vp9_encodeintra.{h, c} files."
Dmitry Kovalev [Fri, 15 Nov 2013 21:38:20 +0000 (13:38 -0800)]
Merge "Removing vp9_encodeintra.{h, c} files."

10 years agoRenamed two files
Yaowu Xu [Fri, 15 Nov 2013 20:48:43 +0000 (12:48 -0800)]
Renamed two files

from vp9_decodframe.{c,h} to vp9_decodeframe.{c,h}

Change-Id: I21ac4b14fc90246e3f16bd90c52c12d126d791f8

10 years agoRemoving vp9_encodeintra.{h, c} files.
Dmitry Kovalev [Fri, 15 Nov 2013 20:17:16 +0000 (12:17 -0800)]
Removing vp9_encodeintra.{h, c} files.

There was only one function in *.c file, so moving it to vp9_encodemb.c.

Change-Id: I728859d08b3d6c05c33c1c5b21f0ea1d0e0f83af

10 years agoMerge "vpx[dec|enc]: Extract IVF support from the apps."
Tom Finegan [Fri, 15 Nov 2013 19:21:57 +0000 (11:21 -0800)]
Merge "vpx[dec|enc]: Extract IVF support from the apps."

10 years agoPrepare 1.3.0 release:
Paweł Hajdan [Fri, 15 Nov 2013 18:48:09 +0000 (10:48 -0800)]
Prepare 1.3.0 release:

- update .mailmap and AUTHORS
- update CHANGELOG

Change-Id: I3d75e8a6f87cabcbebc20e3ee5f3cd7db4d2e84d