Commit Graph

16724 Commits

Author SHA1 Message Date
Yunqing Wang
770c6663d6 Merge "Changes to facilitate row based multi-threading of ARNR filtering" 2017-02-01 22:04:15 +00:00
Johann Koenig
dc90501ba3 Merge changes I374dfc08,I7e15192e,Ica414007
* changes:
  hadamard highbd ssse3: use tran_low_t for coeff
  hadamard highbd neon: use tran_low_t for coeff
  hadamard highbd sse2: use tran_low_t for coeff
2017-02-01 21:56:36 +00:00
Ranjit Kumar Tulabandu
359a6796da Changes to facilitate row based multi-threading of ARNR filtering
Change-Id: I2fd72af00afbbeb903e4fe364611abcc148f2fbb
2017-02-01 13:03:52 -08:00
Johann Koenig
5cc0a364ae Merge "vp9_rdopt: declare 'c' closer to use" 2017-02-01 20:55:12 +00:00
Johann
bfd62cdaff vp9_rdopt: declare 'c' closer to use
Clears up static clang analysis warning regarding a dead store. Only
declare 'c' when it will be used.

Change-Id: I1ac0fc7f94bc44da63938c63cd1efcd6b95e0eb3
2017-02-01 19:58:24 +00:00
Johann Koenig
f60171bb4f Merge "deblock: annotate postproc parameters" 2017-02-01 19:57:29 +00:00
Johann
0f751ecee3 hadamard highbd ssse3: use tran_low_t for coeff
BUG=webm:1365

Change-Id: I374dfc08732932382043905f128e928b08cb4f57
2017-02-01 11:51:15 -08:00
Johann
1eb8a718bf hadamard highbd neon: use tran_low_t for coeff
BUG=webm:1365

Change-Id: I7e15192ead3a3631755b386f102c979f06e26279
2017-02-01 11:50:46 -08:00
Johann
2dac808dd1 hadamard highbd sse2: use tran_low_t for coeff
BUG=webm:1365

Change-Id: Ica414007d8412ceebfffa9e58e8416226a3fe934
2017-02-01 11:46:57 -08:00
Johann Koenig
3bda634576 Merge "quantize ssse3: remove unused pxor" 2017-02-01 19:41:41 +00:00
Jingning Han
a7949f2dd2 Make satd unit test support all bit-depth settings
Turn on satd unit test for c function in both regular and high
bit-depth settings.

Change-Id: I4b0c56addfb84964ede0da3ab760fe0ee640cfd0
2017-01-31 23:21:32 -08:00
Jingning Han
59917dd18e Unify the hadamard transform unit test for bit-depth settings
Unify the 8x8 and 16x16 Hadamard unit test system for both 8-bit
and high bit-depth settings.

Change-Id: I53373c1d43f3ced514ad1e53e03f0fb9b25d9ead
2017-01-31 23:21:32 -08:00
Jingning Han
969957f9f2 Fix real-time compression regression in hbd mode
This commit resolves the compression performance regression in
real-time encoding setting when high bit-depth mode is enabled.

The current solution temporarily disables the SIMD implementations
of vpx_satd, hadamard8x8, and hadamard16x16 in high bit-depth mode.

The commit makes the coding results bit-wise identical between
regular coding pipeline and high bit-depth at profile 0.

BUG=webm:1365

Change-Id: Icfb900821733749685370460a1a5a7e07f76f4bf
2017-01-31 23:17:09 -08:00
Johann
32f68cc58c deblock: annotate postproc parameters
Clears a clang static analyzer warning where 'cols' is assumed to be
less than 0, preventing the for loop from executing.

The assembly already requires that the size be 8 or 16 (U/V or Y plane)
and cols is a multiple of 8.

Change-Id: Ica4612690ead1638c94cfe56b306e87f8ce644f9
2017-01-31 15:58:57 -08:00
Johann Koenig
9efc42f4f8 Merge "Use Buffer class for post proc tests" 2017-01-31 15:28:28 +00:00
Kaustubh Raste
df7e1fecc1 Add mips msa vpx_minmax_8x8 function
average improvement ~4x-5x

Change-Id: I83aee9977534fddb8a9b80d31af646c0b6b1a8c3
2017-01-31 10:00:43 +05:30
Kaustubh Raste
280ad35553 Merge "Add mips msa vpx_vector_var function" 2017-01-31 02:34:51 +00:00
Johann
dcfff3ccc8 quantize ssse3: remove unused pxor
Change-Id: Ifa22d77fd530827de0b32ae71810dc2213ab2937
2017-01-30 17:02:57 -08:00
Marco
d47f257484 vp9: Modify bsize condition for using model_rd_large for speed 7.
In non-rd pickmode: Allow speed 7 to also use larger block size in
model_rd. Small change in behavior for speed 7.

Change-Id: I8c5523e424308e8f0bc71b3f6324dec42a464cc8
2017-01-30 11:16:51 -08:00
Yunqing Wang
106c620a23 Merge "Disable multi-threading in first pass for SVC encoding" 2017-01-28 19:29:01 +00:00
Kaustubh Raste
4ce20fb3f4 Add mips msa vpx_vector_var function
average improvement ~4x-5x

Change-Id: I2f63ef83d816052ca8dc42421e7e9d42f7a7af6b
2017-01-28 08:53:20 +00:00
Marco Paniconi
164db8278f Merge "vp9: Fix to pick_filter_level for highbitdepth build." 2017-01-27 22:47:45 +00:00
Jerome Jiang
9b1a2aa82b Merge "Add macOS Sierra support in configure" 2017-01-27 21:15:52 +00:00
Marco
d94d0ed12f vp9: Fix to pick_filter_level for highbitdepth build.
Change-Id: I53b3fa8bfc0a0717eb1b730c29f2b70060b1b1b7
2017-01-27 10:44:07 -08:00
Jerome Jiang
6d38ad4146 Add macOS Sierra support in configure
BUG=webm:1367

Change-Id: I3000b6d9f93ec49ca86d08151348d33d86bf0034
2017-01-27 10:41:38 -08:00
Ranjit Kumar Tulabandu
6985a0f516 Disable multi-threading in first pass for SVC encoding
BUG=webm:1366

Change-Id: I204ef8496884ba7c4debe64f23f50d298b4090c3
2017-01-27 09:41:53 -08:00
Marco Paniconi
ad1aad69fb Merge "vp9: Modify bsize condition for using model_rd_large." 2017-01-27 15:15:36 +00:00
Marco Paniconi
d0495132aa Merge "vp9: Fixes for usage of skin_map for high bit depth." 2017-01-27 15:15:15 +00:00
Marco
b16c77cdc4 vp9: Modify bsize condition for using model_rd_large.
In non-rd pickmode: small change in behavior for speed 6 and 7.
Remove condition on HIGHBITDEPTH flag.

Change-Id: I360a13fcc313d72612fe9b918162ef4bb278cdea
2017-01-26 22:45:27 -08:00
Kaustubh Raste
407fad2356 Add mips msa vpx Integer projection row/col functions
average improvement ~4x-5x

Change-Id: I17c41383250282b39f5ecae0197ef1df7de20801
2017-01-27 11:11:42 +05:30
Kaustubh Raste
c1553f859f Merge "Add mips msa vpx satd function" 2017-01-27 04:08:51 +00:00
Marco
db99840bf6 vp9: Fixes for usage of skin_map for high bit depth.
Also avoid noise_estimation and source_sad if use_highbitdepth is set.

Change-Id: I5fea396b8f8380ea377045d99ba22a52b92daa46
2017-01-26 19:57:59 -08:00
Johann
f380a1658d Use Buffer class for post proc tests
Add Buffer features for:
Setting the buffer to the output of an ACMRandom function.
Copying a buffer.
Comparing two buffers.
Printing two buffers.

Change-Id: Ib53fb602451a3abdcee279ea2b65b51fbc02d3df
2017-01-26 09:50:49 -08:00
Jerome Jiang
eacc3b4ccf Merge "vp9: Refactor copy partitioning to reduce duplication." 2017-01-26 17:46:11 +00:00
Jerome Jiang
fe4791b0d5 vp9: Refactor copy partitioning to reduce duplication.
Change-Id: Ia1b3c118adec5eccbd2900c8e4b9ea6b1e3e9b7c
2017-01-25 17:33:04 -08:00
Yunqing Wang
4d50dc5ab5 Merge "Remove marco MVC in mcomp.c" 2017-01-26 00:32:55 +00:00
Hui Su
37cd112b0f Merge "Fix an overflow warning in optimize_b()" 2017-01-25 22:49:30 +00:00
Marco
3b2d08a93b vp9-denoiser: Modify skip denoising condition for small blocks.
Skip denoising for blocks < 16x16, and for block = 16x16
skip denoising for low noise levels and width > 480 for now.
Allow for some speed-up in denoiser.

Change-Id: Ib46cefe4741962d145fa08775defea3a9c928567
2017-01-25 11:48:09 -08:00
hui su
519b2e48a8 Fix an overflow warning in optimize_b()
BUG=webm:1361

Change-Id: Ib840bf3b39f7b3c8c017d3488a83434e9a0f45f5
2017-01-25 10:54:39 -08:00
Jerome Jiang
70a3652693 Merge "vp9: Adjust threshold for y sad used in copying partition." 2017-01-25 17:54:15 +00:00
Yunqing Wang
a762cef917 Merge "Initialize errorperbit and sabperbit in ARNR filtering" 2017-01-25 16:43:02 +00:00
Yunqing Wang
633dbcb458 Merge "Multi-threading of first pass stats collection" 2017-01-25 16:40:32 +00:00
Jerome Jiang
3a7ad43fb8 vp9: Adjust threshold for y sad used in copying partition.
Visual quality improvement is observed for noisy clips. Little effects
on speed tests on Nexus 6.

Change-Id: Ib38e04002220708c34102de7b5c36e9940775d89
2017-01-24 17:20:05 -08:00
Ranjit Kumar Tulabandu
8b0c11c358 Multi-threading of first pass stats collection
(yunqingwang)
1. Rebased the patch. Incorporated recent first pass changes.
2. Turned on the first pass unit test.

Change-Id: Ia2f7ba8152d0b6dd6bf8efb9dfaf505ba7d8edee
2017-01-24 15:48:02 -08:00
Marco
8d0c8c5e6b vp9: Adjust some parameters in aq-mode=3 mode.
Increase the qp-delta, mainly for low resolutions,
excluding case of very low bitrates.

avgPSNR/SSSIM gain of ~3-5% on rtc_derf set.
Small change on rtc set.

Change-Id: Ice03d04bd0340404d1957666ef154fd64fed0606
2017-01-24 14:18:02 -08:00
Jerome Jiang
8e8e2d11bf Merge "vp9: Copy partition using avg_source_sad." 2017-01-24 20:58:09 +00:00
Jerome Jiang
ac1358cd56 vp9: Copy partition using avg_source_sad.
Affecting only speed 8.
Speed tests on Nexus 6 show 4% faster for QVGA and 2.4% faster for VGA.
Little/negligible quality regression observed on both rtc and rtc_derf sets.

Change-Id: I337f301a2db49a568d18ba7623160f7678399ae1
2017-01-24 10:31:22 -08:00
Yunqing Wang
91aa1fae2a Merge "Add the multi-threaded first pass encoder unit test" 2017-01-24 17:14:07 +00:00
Ranjit Kumar Tulabandu
75d2443bf0 Initialize errorperbit and sabperbit in ARNR filtering
(Yunqing)
This patch added the missing initialization in temporal filter.
Borg test BDRate results:
PSNR: -0.019%(lowres); -0.013%(hdres);
SSIM: -0.001%(lowres); -0.010%(hdres).
Other q values gave comparable but no better results.

Change-Id: I7ad0c18b39e6f558342688e2fe1e12fdb133ce9b
2017-01-24 08:58:17 -08:00
Kaustubh Raste
182ea677a0 Add mips msa vpx satd function
average improvement ~4x-5x

Change-Id: If8683d636fe2606d4ca1038e28185bca53bbe244
2017-01-24 10:44:22 +05:30