]>
granicus.if.org Git - libvpx/log
Paul Wilkins [Wed, 7 May 2014 11:53:27 +0000 (12:53 +0100)]
Revert "Add an MMX fwht4x4"
Includes changes that are not compatible with VS windows builds.
Amongst other things stdint.h is not supported in VS.
This reverts commit
89fbf3de501b5d7fd90047192521eae3198705cd .
Change-Id: Ifa86d7df250578d1ada9b539c9ff12ed0c523cdd
Johann [Tue, 6 May 2014 19:25:33 +0000 (12:25 -0700)]
Revert "VP8 for ARMv8 by using NEON intrinsics 08"
This reverts commit
a5d79f43b963ced59b462206faf3b7857bdeff7b
There is an issue with gcc 4.6 in the Android NDK:
loopfilter_neon.c: In function 'vp8_loop_filter_vertical_edge_y_neon':
loopfilter_neon.c:394:1: error: insn does not satisfy its constraints:
Change-Id: I2b8c6ee3fa595c152ac3a5c08dd79bd9770c7b52
Vignesh Venkatasubramanian [Tue, 6 May 2014 18:15:57 +0000 (11:15 -0700)]
Merge "third_party/libwebm: pull from upstream"
Alex Converse [Tue, 6 May 2014 18:12:27 +0000 (11:12 -0700)]
Merge "Add an MMX fwht4x4"
Alex Converse [Tue, 6 May 2014 18:12:23 +0000 (11:12 -0700)]
Merge "Add test for 4x4 WHT."
Jingning Han [Tue, 6 May 2014 16:17:22 +0000 (09:17 -0700)]
Merge "SSSE3 implementation of full inverse 8x8 2D-DCT"
Minghai Shang [Tue, 6 May 2014 06:22:09 +0000 (23:22 -0700)]
Merge "[spatial svc] Fix one extra frame count during flush"
Minghai Shang [Tue, 6 May 2014 06:22:03 +0000 (23:22 -0700)]
Merge "[spatial svc] Use VPX_DL_GOOD_QUALITY for vp9_spatial_scalable_encoder"
Minghai Shang [Tue, 6 May 2014 06:21:53 +0000 (23:21 -0700)]
Merge "[spatial svc] No need to code full width and height for non key frame"
Dmitry Kovalev [Tue, 6 May 2014 05:01:02 +0000 (22:01 -0700)]
Merge "Cleaning up vp9_full_range_search_c() function."
Dmitry Kovalev [Fri, 2 May 2014 01:40:24 +0000 (18:40 -0700)]
Making vp9_get_sse_sum_{8x8, 16x16} static.
Change-Id: Ifb7937c977308c682986f0ce9645a0807d2aa46a
Dmitry Kovalev [Tue, 6 May 2014 02:10:25 +0000 (19:10 -0700)]
Merge "Moving costs from MACROBLOCK to VP9_COMP."
Dmitry Kovalev [Tue, 6 May 2014 02:09:57 +0000 (19:09 -0700)]
Merge "Cleaning up vp9_variance.c"
Adrian Grange [Mon, 5 May 2014 22:41:10 +0000 (15:41 -0700)]
Merge changes Ibfa447fe,I771809ba
* changes:
Fix generic-gnu target build
Fix rounding in ARNR calculation
Alex Converse [Mon, 5 May 2014 21:10:41 +0000 (14:10 -0700)]
Add an MMX fwht4x4
7% faster encoding a desktop lossless at RT speed 4.
Change-Id: I41627f5b737752616b6512bb91a36ec45995bf64
Alex Converse [Mon, 5 May 2014 20:50:12 +0000 (13:50 -0700)]
Add test for 4x4 WHT.
Change-Id: I97ba3152f9b44d23a74c6144809142810d07e368
Vignesh Venkatasubramanian [Fri, 2 May 2014 18:44:35 +0000 (11:44 -0700)]
third_party/libwebm: pull from upstream
Pulling libwebm from upstream
Changes from upstream:
249629d make Mkv(Reader|Writer)(FILE*) explicit
7f3cda4 mkvparser: fix a bunch of windows warnings
5c06178 Merge "clang-format on mkvparser.[ch]pp"
4df111e clang-format on mkvparser.[ch]pp
7b24501 clang-format re-run.
c6767b9 Change AlignTrailingComments to false in .clang-format
9097a06 Merge "muxer: Reject file if TrackType is never specified"
eddf974 Merge "clang-format on mkvmuxertypes.hpp and webmids.hpp"
def325c muxer: Reject file if TrackType is never specified
41f869c Merge "clang-format on webvttparser.(cc|h)"
fd0be37 clang-format on webvttparser.(cc|h)
207d8a1 Merge "clang-format on mkvmuxerutil.[ch]pp"
02429eb Merge "clang-format on mkvwriter.[ch]pp"
0cf7b1b Merge "clang-format on mkvreader.[ch]pp"
2e80fed Merge "clang-format on sample.cpp"
3402e12 Merge "clang-format on sample_muxer.cpp"
1a685db Merge "clang-format on sample_muxer_metadata.(cc|h)"
6634c7f Merge "clang-format on vttreader.cc"
7566004 Merge "clang-format on vttdemux.cc"
9915b84 clang-format on mkvreader.[ch]pp
7437254 clang-format on mkvmuxertypes.hpp and webmids.hpp
0d5a98c clang-format on sample_muxer.cpp
e3485c9 clang-format on vttdemux.cc
46cc823 clang-format on dumpvtt.cc
5218bd2 clang-format on vttreader.cc
1a0130d clang-format on sample_muxer_metadata.(cc|h)
867f189 clang-format on sample.cpp
4c7bec5 clang-format on mkvwriter.[ch]pp
9ead078 clang-format on mkvmuxerutil.[ch]pp
fb6b6e6 clang-format on mkvmuxer.[ch]pp
ce77592 Update .clang-format to allow short functions in one line
0a24fe4 Merge "Add support for DateUTC and DefaultDuration in MKV Muxer."
11d5b66 Merge "Add .clang-format"
a1a3b14 Add .clang-format
0fcec38 Add support for DateUTC and DefaultDuration in MKV Muxer.
Change-Id: Ia0ed161ffc3d63c2eba8ed145707ffe543617976
Johann [Mon, 5 May 2014 20:09:43 +0000 (13:09 -0700)]
Merge "VP8 for ARMv8 by using NEON intrinsics 16"
Johann [Mon, 5 May 2014 20:09:12 +0000 (13:09 -0700)]
Merge "VP8 for ARMv8 by using NEON intrinsics 15"
Tom Finegan [Wed, 30 Apr 2014 21:33:24 +0000 (14:33 -0700)]
Add test for postproc example.
Change-Id: I3446d2ebc23e6f1ed650dd5c82648041febe3c4d
Adrian Grange [Mon, 5 May 2014 17:44:48 +0000 (10:44 -0700)]
Fix generic-gnu target build
Added macro to conditionally compile some of the
post-processing functions only when CONFIG_POSTPROC
is defined.
This was causing the build for the generic-gnu target
to fail.
Change-Id: Ibfa447feceb7a0528135025f105be48f97e9965c
Adrian Grange [Mon, 5 May 2014 16:39:24 +0000 (09:39 -0700)]
Fix rounding in ARNR calculation
The rounding of the ARNR filter output prior to
normalization by the filter strength was incorrect
when strength = 0.
In this case 1 << (strength - 1) would not create the
required rounding of 0, rather it would outrange. This
patch fixes this issue.
Change-Id: I771809ba34d6052b17d34c870ea11ff67b418dab
Jingning Han [Fri, 2 May 2014 01:34:46 +0000 (18:34 -0700)]
SSSE3 implementation of full inverse 8x8 2D-DCT
This commit enables SSSE3 version full inverse 8x8 2D-DCT and
reconstruction. It makes the runtime of vp9_idct8x8_64_add down
from 256 cycles (SSE2) to 246 cycles.
Change-Id: I0600feac894d6a443a3c9d18daf34156d4e225c3
Johann [Mon, 5 May 2014 14:08:08 +0000 (07:08 -0700)]
Merge "VP8 for ARMv8 by using NEON intrinsics 14"
Johann [Mon, 5 May 2014 12:36:54 +0000 (05:36 -0700)]
Merge changes Iaf7d6b0a,Iece0bf56
* changes:
Use INLINE and include vpx_config.h instead of plain 'inline'
Use vreinterpret instead of casting neon vector types
Scott LaVarnway [Mon, 5 May 2014 12:31:24 +0000 (05:31 -0700)]
Merge "Remove struct params from vp8_denoiser_filter"
Martin Storsjo [Sun, 4 May 2014 16:09:30 +0000 (19:09 +0300)]
Use INLINE and include vpx_config.h instead of plain 'inline'
This fixes compilation with MSVC.
Change-Id: Iaf7d6b0a0134968a6addf315fde6d852f298db8c
Martin Storsjo [Sun, 4 May 2014 16:09:12 +0000 (19:09 +0300)]
Use vreinterpret instead of casting neon vector types
MSVC doesn't support casting neon vector types but requires using
vreinterpret.
Change-Id: Iece0bf5632567efd7f37f527abea38afeab4926d
James Yu [Wed, 18 Dec 2013 10:47:48 +0000 (18:47 +0800)]
VP8 for ARMv8 by using NEON intrinsics 16
Add variance_neon.c
- vp8_variance16x16_neon
- vp8_variance16x8_neon
- vp8_variance8x16_neon
- vp8_variance8x8_neon
Change-Id: Idfb9c96134a1c6a696a98ce68b4f7ed593a00660
Signed-off-by: James Yu <james.yu@linaro.org>
James Yu [Tue, 17 Dec 2013 14:11:35 +0000 (22:11 +0800)]
VP8 for ARMv8 by using NEON intrinsics 15
Add idct_dequant_0_2x_neon.c
- idct_dequant_0_2x_neon
Change-Id: I8e129172ef1b2517cf72ff267788921f1a792586
Signed-off-by: James Yu <james.yu@linaro.org>
James Yu [Tue, 17 Dec 2013 14:09:51 +0000 (22:09 +0800)]
VP8 for ARMv8 by using NEON intrinsics 14
Add sixtappredict_neon.c
- vp8_sixtap_predict16x16_neon
- vp8_sixtap_predict8x8_neon
- vp8_sixtap_predict8x4_neon
- vp8_sixtap_predict4x4_neon
Change-Id: I3b02fce48ae2e6c6099041ba5ddd7b090f1463b9
Signed-off-by: James Yu <james.yu@linaro.org>
James Yu [Tue, 17 Dec 2013 15:21:50 +0000 (23:21 +0800)]
VP8 for ARMv8 by using NEON intrinsics 13
Add shortidct4x4llm_neon.c
- vp8_short_idct4x4llm_neon
Change-Id: I5a734bbffca8dacf8633c2b0ff07b98aa2f438ba
Signed-off-by: James Yu <james.yu@linaro.org>
Johann [Sun, 4 May 2014 02:06:55 +0000 (19:06 -0700)]
Merge "VP8 for ARMv8 by using NEON intrinsics 12"
Johann [Sun, 4 May 2014 02:05:26 +0000 (19:05 -0700)]
Merge "VP8 for ARMv8 by using NEON intrinsics 11"
Johann [Sun, 4 May 2014 02:04:57 +0000 (19:04 -0700)]
Merge "VP8 for ARMv8 by using NEON intrinsics 10"
Johann [Sun, 4 May 2014 02:04:18 +0000 (19:04 -0700)]
Merge "VP8 for ARMv8 by using NEON intrinsics 09"
Johann [Sun, 4 May 2014 02:03:39 +0000 (19:03 -0700)]
Merge "vp9 register checks only apply to vp9"
James Yu [Tue, 17 Dec 2013 14:03:09 +0000 (22:03 +0800)]
VP8 for ARMv8 by using NEON intrinsics 12
Add sad_neon.c
- vp8_sad16x16_neon
- vp8_sad16x8_neon
- vp8_sad8x8_neon
- vp8_sad8x16_neon
- vp8_sad4x4_neon
Change-Id: I08eaae49ec03fb91b394354660a5df0367cea311
Signed-off-by: James Yu <james.yu@linaro.org>
James Yu [Tue, 17 Dec 2013 15:16:03 +0000 (23:16 +0800)]
VP8 for ARMv8 by using NEON intrinsics 11
Add mbloopfilter_neon.c
- vp8_mbloop_filter_horizontal_edge_y_neon
- vp8_mbloop_filter_horizontal_edge_uv_neon
- vp8_mbloop_filter_vertical_edge_y_neon
- vp8_mbloop_filter_vertical_edge_uv_neon
Change-Id: Ia9084e0892d4d49412d9cf2b165a0f719f2382d7
Signed-off-by: James Yu <james.yu@linaro.org>
Johann [Sat, 3 May 2014 11:21:16 +0000 (04:21 -0700)]
Merge "Remove asm_offsets dependency in quantize_b_ssse3"
James Yu [Tue, 17 Dec 2013 13:49:53 +0000 (21:49 +0800)]
VP8 for ARMv8 by using NEON intrinsics 10
Add loopfiltersimpleverticaledge_neon.c
- vp8_loop_filter_bvs_neon
- vp8_loop_filter_mbvs_neon
Change-Id: I7cf0a161ad4ae37c881b94cc0122f895d3baae79
Signed-off-by: James Yu <james.yu@linaro.org>
James Yu [Tue, 17 Dec 2013 13:48:07 +0000 (21:48 +0800)]
VP8 for ARMv8 by using NEON intrinsics 09
Add loopfiltersimplehorizontaledge_neon.c
- vp8_loop_filter_bhs_neon
- vp8_loop_filter_mbhs_neon
Change-Id: I77f9721b20585da8bf3869a3850ff0ae4b4bfeea
Signed-off-by: James Yu <james.yu@linaro.org>
Johann [Sat, 3 May 2014 11:10:18 +0000 (04:10 -0700)]
Merge "VP8 for ARMv8 by using NEON intrinsics 08"
Johann [Sat, 3 May 2014 11:08:54 +0000 (04:08 -0700)]
Merge "VP8 for ARMv8 by using NEON intrinsics 07"
Minghai Shang [Fri, 2 May 2014 19:09:13 +0000 (12:09 -0700)]
[spatial svc] Fix one extra frame count during flush
Change-Id: Ia096aafa8aca4f9e58450c4b861393d8f0a15afe
Minghai Shang [Fri, 2 May 2014 18:54:45 +0000 (11:54 -0700)]
[spatial svc] Use VPX_DL_GOOD_QUALITY for vp9_spatial_scalable_encoder
Change-Id: I759f2d7c0e75b44ff4522a5ce57328b1c92730aa
Vignesh Venkatasubramanian [Fri, 2 May 2014 18:39:29 +0000 (11:39 -0700)]
Merge "third_party/libwebm: Add Android.mk"
Minghai Shang [Thu, 1 May 2014 19:00:39 +0000 (12:00 -0700)]
[spatial svc] No need to code full width and height for non key frame
Change-Id: I62ab0f4346b4157a90dc5b5f73ab5e597d69c1bd
Dmitry Kovalev [Fri, 2 May 2014 18:08:11 +0000 (11:08 -0700)]
Cleaning up vp9_variance.c
Change-Id: Ie2b9c9881085053c191f3f7b8253e283edef573b
Scott LaVarnway [Fri, 2 May 2014 17:31:52 +0000 (10:31 -0700)]
Remove struct params from vp8_denoiser_filter
This eliminates the asm_offsets dependency for future
all-assembly versions of this function.
Change-Id: I3227073ecfcb8ee6e593934fab941e9081abdda0
Scott LaVarnway [Fri, 2 May 2014 16:59:59 +0000 (09:59 -0700)]
Merge "Improved intrinsic version of vp8_denoiser_filter_neon"
James Yu [Tue, 17 Dec 2013 13:46:08 +0000 (21:46 +0800)]
VP8 for ARMv8 by using NEON intrinsics 08
Add loopfilter_neon.c
- vp8_loop_filter_horizontal_edge_y_neon
- vp8_loop_filter_horizontal_edge_uv_neon
- vp8_loop_filter_vertical_edge_y_neon
- vp8_loop_filter_vertical_edge_uv_neon
Change-Id: I50b57dedabd42d2a3c183c1738cc5346f0e71ed8
Signed-off-by: James Yu <james.yu@linaro.org>
James Yu [Tue, 17 Dec 2013 14:47:58 +0000 (22:47 +0800)]
VP8 for ARMv8 by using NEON intrinsics 07
Add iwalsh_neon.c
- vp8_short_inv_walsh4x4_neon
Change-Id: I8beda6ce11ad8ce9e80cc0a38d40161938359162
Signed-off-by: James Yu <james.yu@linaro.org>
Adrian Grange [Fri, 2 May 2014 15:13:06 +0000 (08:13 -0700)]
Merge "Revert "Force ARNR filtering to be centered on the ARF frame""
Adrian Grange [Fri, 2 May 2014 15:12:00 +0000 (08:12 -0700)]
Merge "Fix mode selection bug when ARNR filtering disabled"
Johann [Fri, 2 May 2014 15:11:54 +0000 (08:11 -0700)]
vp9 register checks only apply to vp9
Disable register checks when vp9 is not configured. Soon vp8 assembly
will move to intrinsics, obviating this check.
This will still run the check when vp9 is enabled.
Change-Id: I90f50d22cb8c15e9c07f2c8e830e08de7fce0689
Johann [Wed, 3 Apr 2013 21:49:30 +0000 (14:49 -0700)]
Remove asm_offsets dependency in quantize_b_ssse3
Replace it with some intrinsic code and inline assembly.
Change-Id: I81b4df146db3d01039059be7dae31083e2943b97
Johann [Fri, 2 May 2014 14:15:38 +0000 (07:15 -0700)]
Merge "Only hide assembly functions for Chrome"
Johann [Fri, 2 May 2014 13:42:34 +0000 (06:42 -0700)]
Merge "VP8 for ARMv8 by using NEON intrinsics 06"
Debargha Mukherjee [Fri, 2 May 2014 11:03:29 +0000 (04:03 -0700)]
Merge "Switch the default 2 pass vbr variant."
James Yu [Tue, 17 Dec 2013 13:23:08 +0000 (21:23 +0800)]
VP8 for ARMv8 by using NEON intrinsics 06
Add idct_dequant_full_2x_neon.c
- idct_dequant_full_2x_neon
==== Summary of apply VP8 decode patch series ====
Benchmark on Samsung Chromebook, Cortex-A15, 1.7GHz, Dual core
Toolchain: linaro-1.13.1-4.8-2014.01
Compile argument: CROSS=arm-linux-gnueabihf- ../libvpx/configure
--target=armv7-linux-gcc --prefix=$HOME/out
--enable-shared --cpu=cortex-a7
Test argument: vpxdec --summary --noblit ./tears_of_steel_1080p.webm
NEON assembly 46.68 (fps)
Apply patch 06 46.65, -0.03
Apply patch 07 46.86, +0.21
Apply patch 08 46.58, -0.28
Apply patch 09 46.57, -0.01
Apply patch 10 46.51, -0.06
Apply patch 11 46.13, -0.38
Apply patch 12 45.42, -0.71
Apply patch 13 46.06, +0.64
Apply patch 14 45.19, -0.87
Apply patch 15 45.93, +0.74
Apply patch 16 45.48, -0.45
Apply patch 17 45.84, +0.36
Apply patch 18 45.91, +0.07 <= With all NEON intrinsics patches
Total -0.77 fps, 1.65% performance regression
Change-Id: I77bfc9eaccfb97b8d401e949ceff8795e26ca6b7
Signed-off-by: James Yu <james.yu@linaro.org>
Dmitry Kovalev [Fri, 2 May 2014 01:28:34 +0000 (18:28 -0700)]
Merge "Moving #defines to *.c from *.h."
Dmitry Kovalev [Fri, 2 May 2014 01:28:21 +0000 (18:28 -0700)]
Merge "Adding vp9_temporal_filter_init() function."
Dmitry Kovalev [Fri, 2 May 2014 01:28:12 +0000 (18:28 -0700)]
Merge "Replacing int_mv with MV."
Dmitry Kovalev [Fri, 2 May 2014 01:27:56 +0000 (18:27 -0700)]
Merge "Simplifying vp9_is_upper_layer_key_frame()."
Johann [Fri, 2 May 2014 00:43:34 +0000 (17:43 -0700)]
Only hide assembly functions for Chrome
Match x86_abi_support.asm configuration
Change-Id: Ic0d03a23961e6858cf5153389ec8afa0fae3307a
Dmitry Kovalev [Thu, 1 May 2014 23:12:23 +0000 (16:12 -0700)]
Moving costs from MACROBLOCK to VP9_COMP.
Change-Id: I61471dd0f77d1547abec13cbf9670e1c4eb9131a
Dmitry Kovalev [Thu, 1 May 2014 21:50:45 +0000 (14:50 -0700)]
Merge "Removing half-variance asm functions which are not used."
Adrian Grange [Thu, 1 May 2014 21:31:50 +0000 (14:31 -0700)]
Fix mode selection bug when ARNR filtering disabled
When ARNR filtering is disabled, by setting
arnr_max_frames=0, mode_skip_mask was being set to
-1 for the ARF frame resulting in no mode being
selected for the block.
The intent is to restrict the reference frame to the
previous ARF frame and the mode to one of ZEROMV,
NEARMV or NEARESTMV.
Change-Id: Ifc3920b153142cd01d422910c94d2f20ffb6f129
Paul Wilkins [Thu, 1 May 2014 11:10:25 +0000 (12:10 +0100)]
Switch the default 2 pass vbr variant.
On balance Deb's modified rate control for VBR seems
to be outperforming especially on some low motion YT
clips so I have switched this to be the default mode for
now.
Change-Id: I0713d430cad6425ac5c48fccdf332e12814ee44a
Dmitry Kovalev [Thu, 1 May 2014 18:41:01 +0000 (11:41 -0700)]
Simplifying vp9_is_upper_layer_key_frame().
Change-Id: Ib56df7cd282dadbfd202de23f0c746a93b5ce63e
Dmitry Kovalev [Thu, 1 May 2014 18:35:50 +0000 (11:35 -0700)]
Adding vp9_temporal_filter_init() function.
Change-Id: I0d50354111df79b74aafcd3bb7dc14df3c14733a
Minghai Shang [Thu, 1 May 2014 17:53:47 +0000 (10:53 -0700)]
Merge "[svc rc] RC improvement for key frames in upper layers for spatial svc."
Tom Finegan [Wed, 30 Apr 2014 18:03:15 +0000 (11:03 -0700)]
Add --show-program-output option to shell tests.
When used --show-program-output shows the output from the programs run
during testing.
Change-Id: I15a47c43d1fcf0243c8df1a75d0d2a584ae1f08f
Dmitry Kovalev [Wed, 30 Apr 2014 22:50:34 +0000 (15:50 -0700)]
Merge "Removing unused alt_activity_measure() function."
Dmitry Kovalev [Wed, 30 Apr 2014 22:50:26 +0000 (15:50 -0700)]
Merge "Removing unused typedefs."
Dmitry Kovalev [Wed, 30 Apr 2014 22:50:13 +0000 (15:50 -0700)]
Replacing int_mv with MV.
Change-Id: Idccb530c814cb8a2fb9f7d0c11eaef25044efe5e
Dmitry Kovalev [Wed, 30 Apr 2014 22:40:24 +0000 (15:40 -0700)]
Cleaning up vp9_full_range_search_c() function.
Change-Id: Ifc9114aeacd493cfa04d4cb3d071bf1de80e0568
Minghai Shang [Thu, 24 Apr 2014 20:31:46 +0000 (13:31 -0700)]
[svc rc] RC improvement for key frames in upper layers for spatial svc.
Change-Id: Id6ab59e505be28cd4eb9f1fe114feb47debe0539
Dmitry Kovalev [Wed, 30 Apr 2014 22:01:26 +0000 (15:01 -0700)]
Merge "Using crop_{width, height} instead of {width, height}."
Jim Bankoski [Wed, 30 Apr 2014 21:43:56 +0000 (14:43 -0700)]
Merge "corrupted frames shouldn't affect stats"
Debargha Mukherjee [Wed, 30 Apr 2014 20:59:12 +0000 (13:59 -0700)]
Merge "Added 16bit vpx_img_fmt_t enums"
Jingning Han [Wed, 30 Apr 2014 20:41:36 +0000 (13:41 -0700)]
Merge "Enable SSSE3 implementation of 8x8 forward 2D-DCT"
Dmitry Kovalev [Wed, 30 Apr 2014 19:40:03 +0000 (12:40 -0700)]
Moving #defines to *.c from *.h.
Change-Id: I7cc6f441f414ca1b4d95dad3f789fff6faf8c3c4
Dmitry Kovalev [Wed, 30 Apr 2014 19:34:09 +0000 (12:34 -0700)]
Removing unused typedefs.
Change-Id: I34ebc59980cf661ed658555e245bf0a93e5c3373
Dmitry Kovalev [Wed, 30 Apr 2014 19:21:54 +0000 (12:21 -0700)]
Removing half-variance asm functions which are not used.
Corresponding C functions were removed in
I99695564a3aa9bc8c79ac0a551d257e2ff3ad3c3
Change-Id: I50a5575065a7a9e41904eb2161afd739def927db
Tom Finegan [Wed, 30 Apr 2014 18:19:37 +0000 (11:19 -0700)]
Merge "tools_common.sh: Add vlog()"
Tom Finegan [Tue, 29 Apr 2014 03:56:37 +0000 (20:56 -0700)]
tools_common.sh: Add vlog()
Add a verbose logging function instead of checking
$VPX_TEST_VERBOSE_OUTPUT in multiple places.
Change-Id: I82618809f0964f696ed17ca4d99d8d7d252232f4
Minghai Shang [Wed, 30 Apr 2014 18:02:54 +0000 (11:02 -0700)]
Merge "[svc rc] Increase worst quality for enhancement layers"
Jim Bankoski [Wed, 30 Apr 2014 17:13:29 +0000 (10:13 -0700)]
corrupted frames shouldn't affect stats
Don't update the stats if we have a corrupted frame.
Change-Id: I65a13adc50e0389b4201d3b671f0225195dfaff4
TODO: Test case that shows this problem.
Peter de Rivaz [Wed, 30 Apr 2014 15:14:11 +0000 (16:14 +0100)]
Added 16bit vpx_img_fmt_t enums
Added in preparation for modifications to support high bitdepth
operations.
Change-Id: I1ad403ea8886cb84020ff06807ae25e2e4bff608
Yunqing Wang [Wed, 30 Apr 2014 16:17:18 +0000 (09:17 -0700)]
Merge "Remove VP8 save_reg_neon function"
Scott LaVarnway [Wed, 30 Apr 2014 13:58:16 +0000 (06:58 -0700)]
Improved intrinsic version of vp8_denoiser_filter_neon
Used horizonal add instructions instead of adding
byte lanes. The encoder performance improved by
~4% for the test clip used.
Change-Id: Iaddd10403fcffb5b3f53b1f591ab2fe0ff002c08
Dmitry Kovalev [Tue, 29 Apr 2014 23:59:47 +0000 (16:59 -0700)]
Merge "Adding search_site_config struct."
Dmitry Kovalev [Tue, 29 Apr 2014 23:58:22 +0000 (16:58 -0700)]
Removing unused alt_activity_measure() function.
Change-Id: I6b520553cb5334b44356dc4651a2dbc1cb93cca5
Dmitry Kovalev [Tue, 29 Apr 2014 23:37:17 +0000 (16:37 -0700)]
Merge "Cleaning up vp9_full_search_sadx8() function."
Yunqing Wang [Tue, 29 Apr 2014 21:36:51 +0000 (14:36 -0700)]
Remove VP8 save_reg_neon function
This patch did a cleanup following the commit "Save NEON registers
in VP8 NEON functions". The pushing/poping of callee-saved NEON
registers was moved into individual NEON functions. Therefore,
we don't need to save those registers at the beginning of codec.
The related code was removed.
Change-Id: I5648166514fc9beffb780aa138495597731f49ea
Dmitry Kovalev [Tue, 29 Apr 2014 23:12:44 +0000 (16:12 -0700)]
Using crop_{width, height} instead of {width, height}.
Change-Id: I6dc9741cdcd700f5c4a387f58da7feb58dd4bbda
Jingning Han [Wed, 23 Apr 2014 16:13:45 +0000 (09:13 -0700)]
Enable SSSE3 implementation of 8x8 forward 2D-DCT
Assembly implementation of ssse3 8x8 forward 2D-DCT. The current
version is turned on only for x86_64. The average unit runtime
goes from 157 cycles down to 136 cycles, i.e., about 12.8% faster.
This translates into about 1.5% speed-up for pedestrian_area 1080p
at speed 2.
Change-Id: I0f12435857e9425ed7ce12541344dfa16837f4f4
Dmitry Kovalev [Tue, 29 Apr 2014 22:09:38 +0000 (15:09 -0700)]
Merge "scale_and_extend_frame{,_nonnormative}() cleanup."