]> granicus.if.org Git - libvpx/log
libvpx
8 years agovpx_mem: Refactor code
Urvang Joshi [Tue, 26 Jul 2016 00:12:14 +0000 (17:12 -0700)]
vpx_mem: Refactor code

Change-Id: I2da9cd5da48ae97e770bccfd1233bcc70b484688
(cherry picked from aom: 83c95f5 aom_mem: Refactor code)

8 years agovp8_decoder_create_threads: check sem/pthread returns
James Zern [Wed, 24 Aug 2016 01:46:31 +0000 (18:46 -0700)]
vp8_decoder_create_threads: check sem/pthread returns

Change-Id: I353da4a2f988ca51d48d0ca91236e8cc0bb48ff5

8 years agovp8_create_decoder_instances: add missing setjmp
James Zern [Wed, 24 Aug 2016 01:29:42 +0000 (18:29 -0700)]
vp8_create_decoder_instances: add missing setjmp

vp8_decoder_create_threads() has allocations that expect one is set.

Change-Id: I423f2153a2969c88d48ba45cc9ead4a01443ce65

8 years agovp8: fix decoder crash with invalid leading keyframes
James Zern [Tue, 23 Aug 2016 22:51:32 +0000 (15:51 -0700)]
vp8: fix decoder crash with invalid leading keyframes

decoding the same invalid keyframe twice would result in a crash as the
second time through the decoder would be assumed to have been
initialized as there was no resolution change. in this case the
resolution was itself invalid (0x6), but vp8_peek_si() was only failing
in the case of 0x0.
invalid-vp80-00-comprehensive-018.ivf.2kf_0x6.ivf tests this case by
duplicating the first keyframe and additionally adds a valid one to
ensure decoding can resume without error.

BUG=b/30593765

Change-Id: If0859035908b7870d67a7f3f646b5a080252eb6d

8 years agoMerge "Disable split mode in 4k video encoding"
Yunqing Wang [Tue, 23 Aug 2016 15:35:32 +0000 (15:35 +0000)]
Merge "Disable split mode in 4k video encoding"

8 years agoDisable split mode in 4k video encoding
Yunqing Wang [Tue, 23 Aug 2016 02:46:44 +0000 (19:46 -0700)]
Disable split mode in 4k video encoding

Disabled the split mode while encoding 4k video to speed
up the encoder.

Borg test result on 4k set:
Overall PSNR: +0.029%; SSIM: +0.009%.
Average encoder speedup at speed 2 is 2.5%.

Change-Id: I1519c658f07c3ac838affbe5aff0ed9b94f3f8f4

8 years agoCorrect CHECK_MEM_ERROR macro
Yury Gitman [Mon, 22 Aug 2016 21:25:00 +0000 (14:25 -0700)]
Correct CHECK_MEM_ERROR macro

The previous macro doesn't work with &cpi->common as a first argument

Change-Id: Ic3f5c49a94cf8b17de6569811b957c963341bb58

8 years agoMerge "Revert "vp8: Move loopfilter synchronization to end of encode_frame call.""
Marco Paniconi [Mon, 22 Aug 2016 15:46:57 +0000 (15:46 +0000)]
Merge "Revert "vp8: Move loopfilter synchronization to end of encode_frame call.""

8 years agoRevert "vp8: Move loopfilter synchronization to end of encode_frame call."
Marco Paniconi [Mon, 22 Aug 2016 15:46:20 +0000 (15:46 +0000)]
Revert "vp8: Move loopfilter synchronization to end of encode_frame call."

This reverts commit c2fe9acceda922ca1d9f0d6185b340560b93597a.

This change break linux browser test in chromium:
https://build.chromium.org/p/chromium.webrtc/builders/Linux%20Tester

Change-Id: I226782fad480c17a99ec6c785ad93cf4ab88f0ae

8 years agoMerge "Adjust speed features for 4k video encoding"
Yunqing Wang [Fri, 19 Aug 2016 23:11:05 +0000 (23:11 +0000)]
Merge "Adjust speed features for 4k video encoding"

8 years agoAdjust speed features for 4k video encoding
Yunqing Wang [Fri, 19 Aug 2016 16:30:32 +0000 (09:30 -0700)]
Adjust speed features for 4k video encoding

Adjusted speed 2 features to speed up 4k video encoding.
BDBR results from borg test:
PSNR: +0.313%; SSIM: +0.268%.
Average speedup: 8.5%

Change-Id: I1e2695a01fb3f3817c1df4480e184c2aed8f2eba

8 years agovp9_pickmode: quiet float conversion warnings
James Zern [Fri, 19 Aug 2016 08:28:01 +0000 (01:28 -0700)]
vp9_pickmode: quiet float conversion warnings

Change-Id: I591e4f958955b3f2edb2f95a83c54cd83c8ef075

8 years agovp9_alloc_context_buffers: clear cm->mi* on failure
James Zern [Thu, 18 Aug 2016 22:29:56 +0000 (15:29 -0700)]
vp9_alloc_context_buffers: clear cm->mi* on failure

this fixes a crash in vp9_dec_setup_mi() via
vp9_init_context_buffers() should decoding continue and the decoder
resyncs on a smaller frame

BUG=b/30593752

Change-Id: I9ce8d94abe89bcd058697e8bd8599690e61bd380

8 years agoMerge "vp9 svc: SVC encoder speed up."
Jacky Chen [Thu, 18 Aug 2016 21:21:29 +0000 (21:21 +0000)]
Merge "vp9 svc: SVC encoder speed up."

8 years agoMerge "Remove '-chromium' flag from ads2gas_apple.pl"
Johann Koenig [Thu, 18 Aug 2016 19:54:54 +0000 (19:54 +0000)]
Merge "Remove '-chromium' flag from ads2gas_apple.pl"

8 years agovp9 svc: SVC encoder speed up.
JackyChen [Fri, 5 Aug 2016 18:10:29 +0000 (11:10 -0700)]
vp9 svc: SVC encoder speed up.

Bias towards base_mv and skip 1/4 pixel motion search when using base mv.
2~3% speed up for 2 spatial layers, 3~5% speed up for 3 spatial layers.
PSNR loss:
(2 layers) 0.07dB for gips_stationary, 0.04dB for gips_motion;
(3 layers) 0.07dB for gips_stationary, 0.06dB for gips_motion.

Change-Id: I773acbda080c301cabe8cd259f842bcc5b8bc999

8 years agoMerge "vp9 non-rd pickmode: Add limit on newmv-last and golden bias."
Marco Paniconi [Thu, 18 Aug 2016 18:03:48 +0000 (18:03 +0000)]
Merge "vp9 non-rd pickmode: Add limit on newmv-last and golden bias."

8 years agoMerge "vp8: Move loopfilter synchronization to end of encode_frame call."
Marco Paniconi [Thu, 18 Aug 2016 02:46:31 +0000 (02:46 +0000)]
Merge "vp8: Move loopfilter synchronization to end of encode_frame call."

8 years agovp9 non-rd pickmode: Add limit on newmv-last and golden bias.
Marco [Mon, 15 Aug 2016 17:22:40 +0000 (10:22 -0700)]
vp9 non-rd pickmode: Add limit on newmv-last and golden bias.

Add option, for newmv-last, to limit the rd-threshold update for early exit,
under a source varianace condition.
This can improve visual quality in low texture moving areas,
like forehead/faces.

Also add bias against golden to improve the speed/fps,
will little/negligible loss in quality.

Only affects CBR mode, non-svc, non-screen-content.

Change-Id: I3a5229eee860c71499a6fd464c450b167b07534d

8 years agoRemove '-chromium' flag from ads2gas_apple.pl
Johann [Wed, 17 Aug 2016 21:05:16 +0000 (14:05 -0700)]
Remove '-chromium' flag from ads2gas_apple.pl

The flag was added because Apple clang and Chromium clang disagreed
for certain versions of instructions.

qsubaddx, qaddsubx, ldrneb and ldrneh were used in armv6 assembly
which was removed in d55724fae9cb27e070add7952394fc0427ef2061

vqshrun was used in some neon assembly but superseded by
dcbfacbb984cdcd5b5a5030103305fa3669dfece

.include was used for obj_int_extract/asm_offsets and removed in
6eec73a747f3d14d327cdc33279de96d0deb48c8

Change-Id: I32f4c9b536d0318482101c0b8e91e42b8f545f18

8 years agoAdd casting to fix warning.
paulwilkins [Wed, 17 Aug 2016 09:50:18 +0000 (10:50 +0100)]
Add casting to fix warning.

Frame bits can safely be stored int but group bits
(kf or arf) use 64bit.

Change-Id: I0800f2a28070f8749110a95721c116fc56987885

8 years agoAdd {} to try and keep Jenkins happy.
paulwilkins [Wed, 17 Aug 2016 10:14:44 +0000 (11:14 +0100)]
Add {} to try and keep Jenkins happy.

Change-Id: If1ca3cf83e058317c9751d7da6caa7cd75eb6845

8 years agovp8: Move loopfilter synchronization to end of encode_frame call.
Marco [Tue, 16 Aug 2016 18:21:20 +0000 (11:21 -0700)]
vp8: Move loopfilter synchronization to end of encode_frame call.

Change-Id: I5bdfea7f51df1f1fa5d9c1597e96988acce6c2f2

8 years agoNEON asm of vpx_lpf_{horizontal,vertical}_8_dual_neon()
Linfeng Zhang [Sat, 13 Aug 2016 01:14:21 +0000 (18:14 -0700)]
NEON asm of vpx_lpf_{horizontal,vertical}_8_dual_neon()

Also expose the NEON intrinsics version.

BUG=webm:1261, webm:1266.

Change-Id: I8c4ae658467dcf66ebf7a75982b2ef712dbb4535

8 years agoChange default recode rule for good speed 0 and best.
paulwilkins [Wed, 10 Aug 2016 13:00:52 +0000 (14:00 +0100)]
Change default recode rule for good speed 0 and best.

Changes the default recode rule for Speed 0 and best quality
from ALLOW_RECODE to ALLOW_RECODE_KFARFGF.

Tested on the NF, hdres, midres and lowres test sets, this setting
when combined with patch I40cb559... now performs "as well" in
metrics terms (in fact it came out a tiny amount better overall)
but encode time is 9.6%  faster (measured as the average
from 27 mid rate local encodes on clips in the derf/lowres set.

Change-Id: I8c781c0cdfa3a9929cd9406d15582fce47d6ae3b

8 years agoChange to recode rules.
paulwilkins [Tue, 9 Aug 2016 09:50:33 +0000 (10:50 +0100)]
Change to recode rules.

Allow recodes for the first inter frame in each arf group
even when the recode rule is set to ALLOW_RECODE_KFARFGF.

Small gains of 0.05%.

Change-Id: I40cb559d36a2bf0ebf5cf758c3f92e452b480577

8 years agoMerge "Modified ARF group allocation."
Paul Wilkins [Mon, 15 Aug 2016 09:42:29 +0000 (09:42 +0000)]
Merge "Modified ARF group allocation."

8 years agoMerge "Fix another motion vector out of range bug"
Yunqing Wang [Fri, 12 Aug 2016 23:52:14 +0000 (23:52 +0000)]
Merge "Fix another motion vector out of range bug"

8 years agoMerge "variance_impl_avx2: restore table layout"
James Zern [Fri, 12 Aug 2016 23:02:27 +0000 (23:02 +0000)]
Merge "variance_impl_avx2: restore table layout"

8 years agovariance_impl_avx2: restore table layout
James Zern [Fri, 12 Aug 2016 06:23:16 +0000 (23:23 -0700)]
variance_impl_avx2: restore table layout

disable clang-format for bilinear_filters_avx2

restores the row layout prior to:
099bd7f vpx_dsp: apply clang-format
but keeps the justification used by clang-format

Change-Id: Icf1733a37edb807e74c26b23a93963c03bd08fd7

8 years agoNEON intrinsics for 4 loopfilter functions
Linfeng Zhang [Wed, 3 Aug 2016 18:42:33 +0000 (11:42 -0700)]
NEON intrinsics for 4 loopfilter functions

New NEON intrinsics functions:
vpx_lpf_horizontal_edge_8_neon()
vpx_lpf_horizontal_edge_16_neon()
vpx_lpf_vertical_16_neon()
vpx_lpf_vertical_16_dual_neon()

BUG=webm:1262, webm:1263, webm:1264, webm:1265.

Change-Id: I7a2aff2a358b22277429329adec606e08efbc8cb

8 years agoFix another motion vector out of range bug
Yunqing Wang [Fri, 12 Aug 2016 00:34:20 +0000 (17:34 -0700)]
Fix another motion vector out of range bug

This patch fixed a motion vector out of range bug:
vpxenc: ../libvpx/vp9/encoder/vp9_mcomp.c:69:
 mv_cost: Assertion `mv->col >= -((1 << (11 + 1 + 2)) - 1) &&
 mv->col < ((1 << (11 + 1 + 2)) - 1)' failed.

For blocks that returned without having full-pixel search, the original
MV limits were not restored, which caused the failure. Moved the set
MV limit function down to fix the bug.

Change-Id: Id7d798fc7214e95c6e4846c588f0233fcf1a4223

8 years agovp8: Fix denoiser setting in multi-res sample encoder.
Marco [Wed, 10 Aug 2016 23:20:46 +0000 (16:20 -0700)]
vp8: Fix denoiser setting in multi-res sample encoder.

Change-Id: I9222f3b252e5ed883659f1a14cd705944ee9da07

8 years agoModified ARF group allocation.
paulwilkins [Tue, 9 Aug 2016 09:10:16 +0000 (10:10 +0100)]
Modified ARF group allocation.

Small average gains in the range 0.05 - 0.1

Change-Id: I30e85c04be615cc84726427c5057388b20a6ff60

8 years agoAlign thread entry point stack
Aleksey Vasenev [Fri, 29 Jul 2016 08:28:50 +0000 (11:28 +0300)]
Align thread entry point stack

_beginthreadex does not align the stack on 16-byte boundary as expected
by gcc.

On x86 targets, the force_align_arg_pointer attribute may be applied to
individual function definitions, generating an alternate prologue and
epilogue that realigns the run-time stack if necessary. This supports
mixing legacy codes that run with a 4-byte aligned stack with modern
codes that keep a 16-byte stack for SSE compatibility.
https://gcc.gnu.org/onlinedocs/gcc/x86-Function-Attributes.html

Change-Id: Ie4e4ab32948c238fa87054d5664189972ca6708e
Signed-off-by: Aleksey Vasenev <margtu-fivt@ya.ru>
8 years agoMerge changes I1d3edbdb,I8b49fd05
James Zern [Wed, 10 Aug 2016 00:05:54 +0000 (00:05 +0000)]
Merge changes I1d3edbdb,I8b49fd05

* changes:
  tests: use scoped_ptr for local video source vars
  y4m_test: init members in the constructor

8 years agoMerge "Refactor mv limits."
Alex Converse [Tue, 9 Aug 2016 17:12:50 +0000 (17:12 +0000)]
Merge "Refactor mv limits."

8 years agotests: use scoped_ptr for local video source vars
James Zern [Mon, 8 Aug 2016 21:43:14 +0000 (14:43 -0700)]
tests: use scoped_ptr for local video source vars

prevents leak warnings on ASSERT*() failures

Change-Id: I1d3edbdbb18dbbe3b17691971348a8121cf09afa

8 years agoMerge "Add cpi parameter for forcing segmentation update"
Yury Gitman [Mon, 8 Aug 2016 21:29:40 +0000 (21:29 +0000)]
Merge "Add cpi parameter for forcing segmentation update"

8 years agoy4m_test: init members in the constructor
James Zern [Mon, 8 Aug 2016 21:27:34 +0000 (14:27 -0700)]
y4m_test: init members in the constructor

prevents use of an uninitialized value in the deconstructor should the
test fail before tmpfile_ is set.

Change-Id: I8b49fd05f0d05e055fdf653bd46983d30f466a68

8 years agoAdd cpi parameter for forcing segmentation update
Yury Gitman [Fri, 15 Jul 2016 22:38:52 +0000 (15:38 -0700)]
Add cpi parameter for forcing segmentation update

Change-Id: I1b0bcb1ffe7604117bfaa0b9989d0e25ff04d28c

8 years agoMerge changes from topic 'clang-tidy'
James Zern [Mon, 8 Aug 2016 20:12:42 +0000 (20:12 +0000)]
Merge changes from topic 'clang-tidy'

* changes:
  *_perf_test.cc: correct DoDecode signature
  test: apply clang-tidy google-readability-braces-around-statements

8 years agoRefactor mv limits.
Alex Converse [Mon, 8 Aug 2016 18:42:27 +0000 (11:42 -0700)]
Refactor mv limits.

Change-Id: Ifebdc9ef37850508eb4b8e572fd0f6026ab04987

8 years agoMerge "Fix a motion vector out of range bug"
Yunqing Wang [Mon, 8 Aug 2016 17:59:47 +0000 (17:59 +0000)]
Merge "Fix a motion vector out of range bug"

8 years ago*_perf_test.cc: correct DoDecode signature
James Zern [Sat, 6 Aug 2016 03:21:02 +0000 (20:21 -0700)]
*_perf_test.cc: correct DoDecode signature

+ delete unused kMaxPsnr from decode_perf_test.cc

Change-Id: Id93347631e7870491069a8b7c5bb1f6b2828425f

8 years agotest: apply clang-tidy google-readability-braces-around-statements
clang-format [Fri, 5 Aug 2016 02:28:34 +0000 (19:28 -0700)]
test: apply clang-tidy google-readability-braces-around-statements

applied against a x86_64 configure with and without
--enable-vp9-highbitdepth

clang-tidy-3.7.1 \
  -checks='-*,google-readability-braces-around-statements' \
  -header-filter='.*' -fix
+ clang-format afterward

Change-Id: Ia2993ec64cf1eb3505d3bfb39068d9e44cfbce8d

8 years agoRemove duplicates in Loop8Test6Param and Loop8Test9Param
Linfeng Zhang [Thu, 4 Aug 2016 20:22:39 +0000 (13:22 -0700)]
Remove duplicates in Loop8Test6Param and Loop8Test9Param

Extract the duplicated data generation code in OperationCheck() of
Loop8Test6Param and Loop8Test9Param, and put in function InitInput().

Change-Id: Ied39ba4ee86b50501cc5d10ebf54f5333c4708f0

8 years agoMerge "remove tools/vpx-style.sh"
James Zern [Sat, 6 Aug 2016 01:23:13 +0000 (01:23 +0000)]
Merge "remove tools/vpx-style.sh"

8 years agoMerge changes Ice037acb,I806af11b,I344a7dd0,Ib7cb87fa
James Zern [Sat, 6 Aug 2016 01:20:52 +0000 (01:20 +0000)]
Merge changes Ice037acb,I806af11b,I344a7dd0,Ib7cb87fa

* changes:
  vp9: normalize vpx_enc_frame_flags_t usage
  args.c: add some explicit casts
  webmdec: quiet -Wshorten-64-to-32 warning
  test/decode_test_driver: rm unused deadline member

8 years agoFix a bug in test/lpf_8_test.cc
Linfeng Zhang [Sat, 6 Aug 2016 00:28:58 +0000 (17:28 -0700)]
Fix a bug in test/lpf_8_test.cc

This bug is introduced in 36608af524bb65e39177f229301111f669bf0d20,
where buffer tmp_s is not fully initialized.

Change-Id: I125b966cf054a82bc63c72647cdd463f434eda17

8 years agoFix a motion vector out of range bug
Yunqing Wang [Fri, 5 Aug 2016 22:09:13 +0000 (15:09 -0700)]
Fix a motion vector out of range bug

This patch fixed a motion vector(MV) out of range bug, which was caused
by not restoring the original values of the MV min/max thresholds after
the sub8x8 full pixel motion search. It occurred rarely and only was seen
while encoding a 4k clip for 200 frames.

BUG=webm:1271

Change-Id: Ibc4e0de80846f297431923cef8a0c80fe8dcc6a5

8 years agovp9: normalize vpx_enc_frame_flags_t usage
James Zern [Fri, 5 Aug 2016 06:24:18 +0000 (23:24 -0700)]
vp9: normalize vpx_enc_frame_flags_t usage

quiets -Wshorten-64-to-32 warnings

Change-Id: Ice037acb675d1d81bfedf2dfcfa91a8a29a19dfd

8 years agoargs.c: add some explicit casts
James Zern [Fri, 5 Aug 2016 06:23:00 +0000 (23:23 -0700)]
args.c: add some explicit casts

values are range checked before returning; quiets -Wshorten-64-to-32
warnings

Change-Id: I806af11b2aaf6760c7ab234a2fe2fdf40e7bece7

8 years agowebmdec: quiet -Wshorten-64-to-32 warning
James Zern [Fri, 5 Aug 2016 06:13:17 +0000 (23:13 -0700)]
webmdec: quiet -Wshorten-64-to-32 warning

track->GetNumber() will fit in an int in well-behaved files

Change-Id: I344a7dd05d04daf3df2d67358ea69f8014a03a5b

8 years agotest/decode_test_driver: rm unused deadline member
James Zern [Fri, 5 Aug 2016 06:11:26 +0000 (23:11 -0700)]
test/decode_test_driver: rm unused deadline member

has the side-effect of removing some lint and -Wshorten-64-to-32
warnings

Change-Id: Ib7cb87fa65cd65534096921f243d15288e97256d

8 years agoremove tools/vpx-style.sh
James Zern [Fri, 5 Aug 2016 03:17:09 +0000 (20:17 -0700)]
remove tools/vpx-style.sh

update ftfy.sh to use clang-format

Change-Id: I8ac740c5b3842beed2b8878fbe506f381f4c57e4

8 years agoMerge changes I6ef79702,Id332c641,I354b5d22,I84438013
Johann Koenig [Thu, 4 Aug 2016 22:30:46 +0000 (22:30 +0000)]
Merge changes I6ef79702,Id332c641,I354b5d22,I84438013

* changes:
  Use common transpose for vpx_idct32x32_1024_add_neon
  Use common transpose for vpx_idct8x8_[12|64]_add_neon
  Use common transpose for vp9_iht8x8_add_neon
  Use common transpose for vpx_idct16x16_[10|256]_add_neon

8 years agoMerge "Remove armv6 target"
Johann Koenig [Thu, 4 Aug 2016 22:21:13 +0000 (22:21 +0000)]
Merge "Remove armv6 target"

8 years agoMerge "correct break placement"
James Zern [Thu, 4 Aug 2016 22:19:30 +0000 (22:19 +0000)]
Merge "correct break placement"

8 years agoUse common transpose for vpx_idct32x32_1024_add_neon
Johann [Fri, 29 Jul 2016 15:29:55 +0000 (08:29 -0700)]
Use common transpose for vpx_idct32x32_1024_add_neon

Change-Id: I6ef7970206d588761ebe80005aecd35365ec50ff

8 years agoUse common transpose for vpx_idct8x8_[12|64]_add_neon
Johann [Fri, 29 Jul 2016 15:29:07 +0000 (08:29 -0700)]
Use common transpose for vpx_idct8x8_[12|64]_add_neon

Change-Id: Id332c641f05336ef9a45e17493ff149fd0a168f0

8 years agoUse common transpose for vp9_iht8x8_add_neon
Johann [Fri, 29 Jul 2016 14:57:35 +0000 (07:57 -0700)]
Use common transpose for vp9_iht8x8_add_neon

Change-Id: I354b5d22130d76b0eceda0748db1f871f58fa372

8 years agoUse common transpose for vpx_idct16x16_[10|256]_add_neon
Johann [Thu, 28 Jul 2016 22:58:05 +0000 (15:58 -0700)]
Use common transpose for vpx_idct16x16_[10|256]_add_neon

Change-Id: I84438013f483e82084d33ba9a63c33273d35fcaa

8 years agoMerge "Extract neon transpose for re-use"
Johann Koenig [Thu, 4 Aug 2016 20:12:38 +0000 (20:12 +0000)]
Merge "Extract neon transpose for re-use"

8 years agoMerge "vp9_ratectrl.c: apply clang-format"
James Zern [Thu, 4 Aug 2016 20:01:45 +0000 (20:01 +0000)]
Merge "vp9_ratectrl.c: apply clang-format"

8 years agocorrect break placement
James Zern [Thu, 4 Aug 2016 19:55:23 +0000 (12:55 -0700)]
correct break placement

these should be placed within {}s when present

Change-Id: Ia775fac5373603e77360398f19b07958fb43f476

8 years agoMerge "Don't expand to Q register for 4x4 intrapred"
Johann Koenig [Thu, 4 Aug 2016 19:55:49 +0000 (19:55 +0000)]
Merge "Don't expand to Q register for 4x4 intrapred"

8 years agoRemove armv6 target
Johann [Fri, 29 Jul 2016 19:31:40 +0000 (12:31 -0700)]
Remove armv6 target

Change-Id: I1fa81cc9cabf362a185fc3a53f1e58de533a41e5

8 years agoMerge "Pad 'Left' when building under ASan"
Johann Koenig [Thu, 4 Aug 2016 19:27:44 +0000 (19:27 +0000)]
Merge "Pad 'Left' when building under ASan"

8 years agoMerge "Update Loop8Test{6,9}Param to test filter8() in mb_lpf_vertical_edge_w()"
Linfeng Zhang [Thu, 4 Aug 2016 19:21:22 +0000 (19:21 +0000)]
Merge "Update Loop8Test{6,9}Param to test filter8() in mb_lpf_vertical_edge_w()"

8 years agoExtract neon transpose for re-use
Johann [Thu, 28 Jul 2016 22:47:53 +0000 (15:47 -0700)]
Extract neon transpose for re-use

Change-Id: I5e1c7f4c80d1c6f7fd582ac468c6eaaa3603a06c

8 years agoMerge changes from topic 'clang-format'
James Zern [Thu, 4 Aug 2016 19:03:02 +0000 (19:03 +0000)]
Merge changes from topic 'clang-format'

* changes:
  README: add a note about clang-format
  README: update target list
  README: fix typo

8 years agovp9_ratectrl.c: apply clang-format
clang-format [Thu, 4 Aug 2016 18:57:00 +0000 (11:57 -0700)]
vp9_ratectrl.c: apply clang-format

after:
ff0a87c vp9 1pass vbr: Adjustment to gf interval.

Change-Id: I1296e53e601bf0c2b562e3a34082ac45c294a5f1

8 years agoDon't expand to Q register for 4x4 intrapred
Johann [Wed, 27 Jul 2016 21:24:14 +0000 (14:24 -0700)]
Don't expand to Q register for 4x4 intrapred

The code was expanding to Q registers so that vqrshn could be used, for
vector quad round shift and narrow. If 4 values are added together,
there is a shift by 2. If 8 values, a shift by 3. Since this accounts
for any possibility of overflow, we can skip the narrowing shift.

This allows keeping the values in D registers and casting the 16 bit
value to 8 bits.

Change-Id: I8d9cfa07176271f492c116ffa6a7b351af0b8751

8 years agoUpdate Loop8Test{6,9}Param to test filter8() in mb_lpf_vertical_edge_w()
Linfeng Zhang [Wed, 3 Aug 2016 18:02:15 +0000 (11:02 -0700)]
Update Loop8Test{6,9}Param to test filter8() in mb_lpf_vertical_edge_w()

One branch of filter8() in mb_lpf_vertical_edge_w() was not tested.

Change-Id: I194202d771d9acd6b4e5e600ee2bae89986b49f3

8 years agoMerge "vp9 1pass vbr: Adjustment to gf interval."
Marco Paniconi [Thu, 4 Aug 2016 17:50:55 +0000 (17:50 +0000)]
Merge "vp9 1pass vbr: Adjustment to gf interval."

8 years agoFix msvc compiler warnings
Yaowu Xu [Wed, 3 Aug 2016 21:58:46 +0000 (14:58 -0700)]
Fix msvc compiler warnings

MSVC 2013 complained about using 32 shift where 64 bit shift should be
used.

Change-Id: I7a2b165d1a92d3c0a91dd4511b27aba7709b5e55

8 years agoMerge "Resolve -Wshorten-64-to-32 warnings in prob.h."
James Zern [Thu, 4 Aug 2016 00:38:08 +0000 (00:38 +0000)]
Merge "Resolve -Wshorten-64-to-32 warnings in prob.h."

8 years agoREADME: add a note about clang-format
James Zern [Thu, 28 Jul 2016 05:16:51 +0000 (22:16 -0700)]
README: add a note about clang-format

Change-Id: I835401e3befffcbc68e7d2bdd2fd556a19948e91

8 years agoREADME: update target list
James Zern [Thu, 28 Jul 2016 05:10:09 +0000 (22:10 -0700)]
README: update target list

Change-Id: I80293720a5f12bc2449ceaadbb2ad0f924141552

8 years agoREADME: fix typo
James Zern [Thu, 28 Jul 2016 05:10:02 +0000 (22:10 -0700)]
README: fix typo

Change-Id: I2c3ecc62b1fd1e600b3d70b623c8b11e1e8e4d13

8 years agoMerge "test: apply clang-format"
James Zern [Thu, 4 Aug 2016 00:27:59 +0000 (00:27 +0000)]
Merge "test: apply clang-format"

8 years agoMerge "vp9/decoder,vp9/*.[hc]: apply clang-format"
James Zern [Thu, 4 Aug 2016 00:22:59 +0000 (00:22 +0000)]
Merge "vp9/decoder,vp9/*.[hc]: apply clang-format"

8 years agoPad 'Left' when building under ASan
Johann [Wed, 27 Jul 2016 21:19:20 +0000 (14:19 -0700)]
Pad 'Left' when building under ASan

The neon intrinsics are not able to load just the 4 values that are
used. In vpx_dsp/arm/intrapred_neon.c:dc_4x4 it loads 8 values for both
the 'above' and 'left' computations, but only uses the sum of the first
4 values.

BUG=webm:1268

Change-Id: I937113d7e3a21e25bebde3593de0446bf6b0115a

8 years agovp9 1pass vbr: Adjustment to gf interval.
Marco [Wed, 3 Aug 2016 18:39:52 +0000 (11:39 -0700)]
vp9 1pass vbr: Adjustment to gf interval.

Increase the minimum distance.
Reduces the overshoot somewhat on some clips,
small gain in avgPSNR (~0.1%) on ytlive set.

Change-Id: Id5ddde20c2907dbdb536e79542eff775019c142b

8 years agovp9/decoder,vp9/*.[hc]: apply clang-format
clang-format [Wed, 27 Jul 2016 03:20:13 +0000 (20:20 -0700)]
vp9/decoder,vp9/*.[hc]: apply clang-format

Change-Id: Ic38ea06c7b2fb3e8e94a4c0910e82672a1acaea7

8 years agoMerge "vp9 svc: Fix a valgrind error."
Yaowu Xu [Wed, 3 Aug 2016 20:53:04 +0000 (20:53 +0000)]
Merge "vp9 svc: Fix a valgrind error."

8 years agovp9/common: apply clang-format
clang-format [Tue, 26 Jul 2016 23:52:55 +0000 (16:52 -0700)]
vp9/common: apply clang-format

Change-Id: Ie0f150fdcfcbf7c4db52d3a08bc8238ed1c72e3b

8 years agovp9/encoder: apply clang-format
clang-format [Wed, 27 Jul 2016 03:43:23 +0000 (20:43 -0700)]
vp9/encoder: apply clang-format

Change-Id: I45d9fb4013f50766b24363a86365e8063e8954c2

8 years agovp9 svc: Fix a valgrind error.
JackyChen [Tue, 2 Aug 2016 22:45:44 +0000 (15:45 -0700)]
vp9 svc: Fix a valgrind error.

This error was introduced by the patch:
8ce67d7 vp9 svc: Enable different speed setting for each spatial layer.
To use svc, svc_param should be cleared to 0 at the beginning.

Change-Id: I222f03ddae8a50e84b4690b78263abb742fae91e

8 years agoResolve -Wshorten-64-to-32 warnings in prob.h.
Alex Converse [Thu, 28 Jul 2016 16:53:23 +0000 (09:53 -0700)]
Resolve -Wshorten-64-to-32 warnings in prob.h.

Change-Id: I1244ee908d81467f0fc8a8fce979fc8077a325b4

8 years agoMerge "Resolve -Wshorten-64-to-32 in variance."
Alex Converse [Tue, 2 Aug 2016 22:26:55 +0000 (22:26 +0000)]
Merge "Resolve -Wshorten-64-to-32 in variance."

8 years agochange HBD pixel value from uint8_t to uint16_t
Yaowu Xu [Tue, 2 Aug 2016 18:01:39 +0000 (11:01 -0700)]
change HBD pixel value from uint8_t to uint16_t

This fixes a regression in 10/12 bit encoding results.

Change-Id: I438877352a41aae0a864a8d9979afe4aa2061d81

8 years agoAdd pointer conversion for HBD buffers
Yaowu Xu [Mon, 1 Aug 2016 22:56:43 +0000 (15:56 -0700)]
Add pointer conversion for HBD buffers

This fixes a crash in HBD build.

Change-Id: I7f688f50227323e69bba65df0d56f4360f01771b

8 years agoMerge "Unfork 8-bit in HBD path in vp9_model_rd_from_var_lapndz callers."
Alex Converse [Mon, 1 Aug 2016 16:42:39 +0000 (16:42 +0000)]
Merge "Unfork 8-bit in HBD path in vp9_model_rd_from_var_lapndz callers."

8 years agoMerge "Cache optimizations in optimize_b()."
Alex Converse [Mon, 1 Aug 2016 16:30:05 +0000 (16:30 +0000)]
Merge "Cache optimizations in optimize_b()."

8 years agoCache optimizations in optimize_b().
Alex Converse [Fri, 29 Jul 2016 19:06:49 +0000 (12:06 -0700)]
Cache optimizations in optimize_b().

Move best index into the token state. Shrink it down to one byte. This
is more cache friendly (access are group together) and uses less total
memory.

Results in 4% fewer cycles in optimize_b().

Change-Id: I75db484fb3dc82f59928d54b659d79c80ee40452

8 years agoMerge "replace by VSTM/VLDM to reduce one of VST1/VLD1"
Johann Koenig [Fri, 29 Jul 2016 14:25:10 +0000 (14:25 +0000)]
Merge "replace by VSTM/VLDM to reduce one of VST1/VLD1"

8 years agoreplace by VSTM/VLDM to reduce one of VST1/VLD1
Min Chen [Wed, 27 Jul 2016 19:22:30 +0000 (14:22 -0500)]
replace by VSTM/VLDM to reduce one of VST1/VLD1

Change-Id: I596567570580babb1a52925541d1fd1045c352f5

8 years agovp8: Switch skin model to mode 0 to save some cycle.
JackyChen [Thu, 28 Jul 2016 20:27:08 +0000 (13:27 -0700)]
vp8: Switch skin model to mode 0 to save some cycle.

This change will speed up vp8 encoder by 1.5% ~ 2% on linux. No
much speed change on Mac.

Change-Id: Id957f19ddd89805baa2af84c5027d52d9a48553f