]> granicus.if.org Git - libvpx/log
libvpx
10 years agoMerge "Revert "Revert "VP8 for ARMv8 by using NEON intrinsics 08"""
Johann [Tue, 2 Sep 2014 16:28:14 +0000 (09:28 -0700)]
Merge "Revert "Revert "VP8 for ARMv8 by using NEON intrinsics 08"""

10 years agoMerge "Removing lookup_next_frame_stats()."
Dmitry Kovalev [Tue, 2 Sep 2014 16:25:16 +0000 (09:25 -0700)]
Merge "Removing lookup_next_frame_stats()."

10 years agoMerge "Skip intra mode tests depending on inter residuals"
Jingning Han [Tue, 2 Sep 2014 15:32:52 +0000 (08:32 -0700)]
Merge "Skip intra mode tests depending on inter residuals"

10 years agoMerge "Replacing asm 8x8 variance calculation with intrinsics."
Dmitry Kovalev [Mon, 1 Sep 2014 01:39:46 +0000 (18:39 -0700)]
Merge "Replacing asm 8x8 variance calculation with intrinsics."

10 years agoRemoving dummy_packing member from VP9_COMP.
Dmitry Kovalev [Fri, 22 Aug 2014 00:28:44 +0000 (17:28 -0700)]
Removing dummy_packing member from VP9_COMP.

Change-Id: I571ce84c97087f8a1a36a10058393bfdcefbf72a

10 years agoReplacing asm 8x8 variance calculation with intrinsics.
Dmitry Kovalev [Fri, 29 Aug 2014 21:25:31 +0000 (14:25 -0700)]
Replacing asm 8x8 variance calculation with intrinsics.

New code is 10% faster for 64-bit and 25% faster for 32-bit. Compiled
using clang.

Change-Id: I8ba1544c30dd6f3ca479db806384317549650dfc

10 years agoMerge "Fix int64_t to unsigned int conversion warnings"
Jingning Han [Sat, 30 Aug 2014 00:15:46 +0000 (17:15 -0700)]
Merge "Fix int64_t to unsigned int conversion warnings"

10 years agoMerge "Extend block level sse to support multiple txfm blocks"
Jingning Han [Sat, 30 Aug 2014 00:15:30 +0000 (17:15 -0700)]
Merge "Extend block level sse to support multiple txfm blocks"

10 years agoFix int64_t to unsigned int conversion warnings
Jingning Han [Thu, 28 Aug 2014 16:21:58 +0000 (09:21 -0700)]
Fix int64_t to unsigned int conversion warnings

Use unsigned int type to store the sse in the pixel domain. The
precision is sufficient to handle sse of block size up to 64x64.
The transform domain version however needs int64_t, since there is
a transfer gain applied in the forward transformation that might
cause unsigned int overflow.

Change-Id: Ifef97c38597e426262290f35341fbb093cf0a079

10 years agoMerge "Removing variance MMX code."
Dmitry Kovalev [Fri, 29 Aug 2014 21:08:02 +0000 (14:08 -0700)]
Merge "Removing variance MMX code."

10 years agoMerge "vp9: sync workers at the start of decode_tiles_mt()"
James Zern [Fri, 29 Aug 2014 21:07:37 +0000 (14:07 -0700)]
Merge "vp9: sync workers at the start of decode_tiles_mt()"

10 years agoMerge "vp9: fix m/t loop filter invalid free"
James Zern [Fri, 29 Aug 2014 21:07:02 +0000 (14:07 -0700)]
Merge "vp9: fix m/t loop filter invalid free"

10 years agoMerge "Minor fix in vp9_encoder.h"
Yunqing Wang [Fri, 29 Aug 2014 20:44:10 +0000 (13:44 -0700)]
Merge "Minor fix in vp9_encoder.h"

10 years agoRevert "Revert "VP8 for ARMv8 by using NEON intrinsics 08""
Scott LaVarnway [Fri, 29 Aug 2014 20:29:36 +0000 (13:29 -0700)]
Revert "Revert "VP8 for ARMv8 by using NEON intrinsics 08""

This reverts commit 928ff03889dadc3f63883553b443c08e625b4885

Compiles with 4.6 now.

Change-Id: Ib455da1098bb0e0623248be07579882a425fcbd1

10 years agoMinor fix in vp9_encoder.h
Yunqing Wang [Fri, 29 Aug 2014 18:27:24 +0000 (11:27 -0700)]
Minor fix in vp9_encoder.h

Added the missing "int".

Change-Id: I7c8af3dee700837b40f010d53e1431a59370ae3a

10 years agovp9: fix m/t loop filter invalid free
James Zern [Fri, 29 Aug 2014 00:11:31 +0000 (17:11 -0700)]
vp9: fix m/t loop filter invalid free

store the number of allocated rows in VP9LfSync, the calculated values
can not be relied on when dealing with corrupt material.

Change-Id: I13b8bcec9738c299a71df726772ab7ac05511e5b

10 years agoRemoving variance MMX code.
Dmitry Kovalev [Fri, 29 Aug 2014 00:31:00 +0000 (17:31 -0700)]
Removing variance MMX code.

Removed functions:
* vp9_mse16x16_mmx
* vp9_get_mb_ss_mmx
* vp9_get4x4var_mmx
* vp9_get8x8var_mmx
* vp9_variance4x4_mmx
* vp9_variance8x8_mmx
* vp9_variance16x16_mmx
* vp9_variance16x8_mmx
* vp9_variance8x16_mmx

They all have SSE2 equivalent.

Change-Id: I3796f2477c4f59b35b4828f46a300c16e62a2615

10 years agoSkip intra mode tests depending on inter residuals
Jingning Han [Fri, 29 Aug 2014 05:58:30 +0000 (22:58 -0700)]
Skip intra mode tests depending on inter residuals

This commit allows encoder to skip intra coding mode test, when
the known inter residual is less than the source variance. It
reduces the runtime of speed 3 for test clips:
bus cif 1000 kbps: 8587 ms -> 8260 ms, 3.8% speed-up
pedestrian 1080p 2000 kbps: 161381 ms -> 155241 ms, 3.7% speed-up.

The compression performance is down by
derf   -0.36%
stdhd  -0.25%

Change-Id: I75ce1e035b4da2153cb1ac14111d1a07c05a735d

10 years agoExtend block level sse to support multiple txfm blocks
Jingning Han [Thu, 28 Aug 2014 16:09:37 +0000 (09:09 -0700)]
Extend block level sse to support multiple txfm blocks

This commit extends the sse and forward transform computation flag
to support the case 64x64 blocks where there are 4 32x32 2D-DCT
blocks.

Change-Id: I86a3e805dfaa0f3abd812f590520c71aa0e40473

10 years agoMerge "vp9: fix crash in inline loopfilter w/corrupt file"
James Zern [Fri, 29 Aug 2014 01:37:30 +0000 (18:37 -0700)]
Merge "vp9: fix crash in inline loopfilter w/corrupt file"

10 years agoMerge "vp9: fix crash in mt loopfilter w/corrupt file"
James Zern [Fri, 29 Aug 2014 01:36:31 +0000 (18:36 -0700)]
Merge "vp9: fix crash in mt loopfilter w/corrupt file"

10 years agovp9: sync workers at the start of decode_tiles_mt()
James Zern [Fri, 29 Aug 2014 00:50:38 +0000 (17:50 -0700)]
vp9: sync workers at the start of decode_tiles_mt()

prevents any problems resuming decode after decoding a corrupt frame

Change-Id: Ib7eb1b5c062aebe71074fef1ece32a32822c16be

10 years agoMerge "Implementing 4x4 variance calculation with SSE2."
Dmitry Kovalev [Fri, 29 Aug 2014 00:25:46 +0000 (17:25 -0700)]
Merge "Implementing 4x4 variance calculation with SSE2."

10 years agoImplementing 4x4 variance calculation with SSE2.
Dmitry Kovalev [Thu, 28 Aug 2014 00:39:44 +0000 (17:39 -0700)]
Implementing 4x4 variance calculation with SSE2.

New SSE2 function is three times faster than MMX one.

Change-Id: I4f387ce9f75b88379176ec7bdc62d86eb5f70fbe

10 years agoRemoving alg_priv from vpx_codec_priv struct.
Dmitry Kovalev [Thu, 21 Aug 2014 00:02:10 +0000 (17:02 -0700)]
Removing alg_priv from vpx_codec_priv struct.

In order to understand memory layout consider the declaration of the
following structs. The first one is a part of our API:

struct vpx_codec_ctx {
  // ...
  struct vpx_codec_priv *priv;
};

The second one is defined in vpx_codec_internal.h:

struct vpx_codec_priv {
  // ...
};

The following struct is defined 4 times for encoder/decoder VP8/VP9:

struct vpx_codec_alg_priv {
  struct vpx_codec_priv base;
  // ...
};

Private data allocation for the given ctx:

struct vpx_codec_ctx *ctx = <get>
struct vpx_codec_alg_priv *alg_priv = <allocate>
ctx->priv = (struct vpx_codec_priv *)alg_priv;

The cast works because vpx_codec_alg_priv has a
vpx_codec_priv instance as a first member 'base'.

Change-Id: I10d1afc8c9a7dfda50baade8c7b0296678bdb0d0

10 years agoMerge "Removing unused arnr_type from VP9EncoderConfig and vp9_extracfg."
Dmitry Kovalev [Thu, 28 Aug 2014 20:50:05 +0000 (13:50 -0700)]
Merge "Removing unused arnr_type from VP9EncoderConfig and vp9_extracfg."

10 years agoMerge "Early termination in encoding partition search"
Yunqing Wang [Thu, 28 Aug 2014 20:49:39 +0000 (13:49 -0700)]
Merge "Early termination in encoding partition search"

10 years agoMerge "vp8_multi_resolution_encoder: Huge cleanup."
Dmitry Kovalev [Thu, 28 Aug 2014 19:13:29 +0000 (12:13 -0700)]
Merge "vp8_multi_resolution_encoder: Huge cleanup."

10 years agoMerge "test/vp8_multi_resolution_encoder.sh: Add missing devnull redirect."
James Zern [Thu, 28 Aug 2014 18:59:30 +0000 (11:59 -0700)]
Merge "test/vp8_multi_resolution_encoder.sh: Add missing devnull redirect."

10 years agoMerge "Removing unused debug code under WRITE_RECON_BUFFER."
Dmitry Kovalev [Thu, 28 Aug 2014 18:46:45 +0000 (11:46 -0700)]
Merge "Removing unused debug code under WRITE_RECON_BUFFER."

10 years agoMerge "Converting configure_skippable_frame() to is_skippable_frame()."
Dmitry Kovalev [Thu, 28 Aug 2014 18:45:32 +0000 (11:45 -0700)]
Merge "Converting configure_skippable_frame() to is_skippable_frame()."

10 years agoMerge "Renaming vpxenc option from frame_boost to frame-boost."
Dmitry Kovalev [Thu, 28 Aug 2014 18:42:23 +0000 (11:42 -0700)]
Merge "Renaming vpxenc option from frame_boost to frame-boost."

10 years agoMerge "Adding several vpxenc parameter ranges to help output."
Dmitry Kovalev [Thu, 28 Aug 2014 18:42:04 +0000 (11:42 -0700)]
Merge "Adding several vpxenc parameter ranges to help output."

10 years agoEarly termination in encoding partition search
Yunqing Wang [Fri, 15 Aug 2014 00:25:21 +0000 (17:25 -0700)]
Early termination in encoding partition search

In the partition search, the encoder checks all possible
partitionings in the superblock's partition search tree.
This patch proposed a set of criteria for partition search
early termination, which effectively decided whether or
not to terminate the search in current branch based on the
"skippable" result of the quantized transform coefficients.
The "skippable" information was gathered during the
partition mode search, and no overhead calculations were
introduced.

This patch gives significant encoding speed gains without
sacrificing the quality.

Borg test results:
1. At speed 1,
   stdhd set: psnr: +0.074%, ssim: +0.093%;
   derf set:  psnr: -0.024%, ssim: +0.011%;
2. At speed 2,
   stdhd set: psnr: +0.033%, ssim: +0.100%;
   derf set:  psnr: -0.062%, ssim: +0.003%;
3. At speed 3,
   stdhd set: psnr: +0.060%, ssim: +0.190%;
   derf set:  psnr: -0.064%, ssim: -0.002%;
4. At speed 4,
   stdhd set: psnr: +0.070%, ssim: +0.143%;
   derf set:  psnr: -0.104%, ssim: +0.039%;

The speedup ranges from several percent to 60+%.
                 speed1    speed2    speed3    speed4
(1080p, 100f):
old_town_cross:  48.2%     23.9%     20.8%     16.5%
park_joy:        11.4%     17.8%     29.4%     18.2%
pedestrian_area: 10.7%      4.0%      4.2%      2.4%
(720p, 200f):
mobcal:          68.1%     36.3%     34.4%     17.7%
parkrun:         15.8%     24.2%     37.1%     16.8%
shields:         45.1%     32.8%     30.1%      9.6%
(cif, 300f)
bus:              3.7%     10.4%     14.0%      7.9%
deadline:        13.6%     14.8%     12.6%     10.9%
mobile:           5.3%     11.5%     14.7%     10.7%

Change-Id: I246c38fb952ad762ce5e365711235b605f470a66

10 years agovp8_multi_resolution_encoder: Huge cleanup.
Dmitry Kovalev [Tue, 26 Aug 2014 22:54:00 +0000 (15:54 -0700)]
vp8_multi_resolution_encoder: Huge cleanup.

Change-Id: I65b2c1fbed5a306949843315999d10368a100431

10 years agotest/vp8_multi_resolution_encoder.sh: Add missing devnull redirect.
Tom Finegan [Thu, 28 Aug 2014 17:49:33 +0000 (10:49 -0700)]
test/vp8_multi_resolution_encoder.sh: Add missing devnull redirect.

Change-Id: Id60ae8178191e30603a5c8665793ace7eb9c8533

10 years agoAdding several vpxenc parameter ranges to help output.
Dmitry Kovalev [Tue, 26 Aug 2014 06:46:45 +0000 (23:46 -0700)]
Adding several vpxenc parameter ranges to help output.

Change-Id: Iec1a00203930a539ed584961aed4b2bfd7488255

10 years agoMerge "Updates vp9_pattern search to return integer sads"
Deb Mukherjee [Thu, 28 Aug 2014 16:38:56 +0000 (09:38 -0700)]
Merge "Updates vp9_pattern search to return integer sads"

10 years agoMerge "Replacing int_mv with MV."
Dmitry Kovalev [Thu, 28 Aug 2014 16:18:11 +0000 (09:18 -0700)]
Merge "Replacing int_mv with MV."

10 years agoUpdates vp9_pattern search to return integer sads
Deb Mukherjee [Wed, 27 Aug 2014 21:06:30 +0000 (14:06 -0700)]
Updates vp9_pattern search to return integer sads

Updates the vp9_pattern_search function to return integer one-away
neighbors' sad values, for subsequent use in speeding up the
sub-pel search. Also, removes code for the do_refine option
which is not being used currently.
Updates the integer and subpel functions to pass in a 5-element
sad list for output or input.

A new pruned sub-pel search algorithm is implemented that uses
the sad returned from the integer pel search. But it is not
deployed yet.

Change-Id: Ifa9f5ad024b5b660570366d2bd900343e1891520

10 years agoMerge "Re-work RD modeling based on inter frame prediction residual"
Jingning Han [Thu, 28 Aug 2014 01:48:49 +0000 (18:48 -0700)]
Merge "Re-work RD modeling based on inter frame prediction residual"

10 years agoMerge "Re-use switchable rate value in handle_inter_mode"
Jingning Han [Thu, 28 Aug 2014 01:48:41 +0000 (18:48 -0700)]
Merge "Re-use switchable rate value in handle_inter_mode"

10 years agoMerge "Add an early termination check in handle_inter_mode"
Jingning Han [Thu, 28 Aug 2014 01:48:32 +0000 (18:48 -0700)]
Merge "Add an early termination check in handle_inter_mode"

10 years agoMerge "Use max txfm size unit in rate-distortion cost modeling"
Jingning Han [Thu, 28 Aug 2014 01:48:24 +0000 (18:48 -0700)]
Merge "Use max txfm size unit in rate-distortion cost modeling"

10 years agovp9: fix crash in inline loopfilter w/corrupt file
James Zern [Wed, 27 Aug 2014 23:29:34 +0000 (16:29 -0700)]
vp9: fix crash in inline loopfilter w/corrupt file

attempting to decode a frame after the previous frame failed has the
potential of interrupting an earlier loop filter task

Change-Id: I6f2b1ddcdf5b89c3e2ee8caf5289dada2a087d66

10 years agoAdd test for vp8_multi_resolution_encoder.
Tom Finegan [Wed, 27 Aug 2014 22:48:44 +0000 (15:48 -0700)]
Add test for vp8_multi_resolution_encoder.

Change-Id: I94667a23046e3cf17c421ac97ca66ea57f846b5c

10 years agoRe-work RD modeling based on inter frame prediction residual
Jingning Han [Tue, 26 Aug 2014 23:52:56 +0000 (16:52 -0700)]
Re-work RD modeling based on inter frame prediction residual

This commit re-work the operation flow related to prediction
residual generation and the rate-distortion modeling. It saves one
call for model_rd_for_sb.

Change-Id: Icaf96c0ff09c903637ed5283448afe01d798195f

10 years agoRe-use switchable rate value in handle_inter_mode
Jingning Han [Tue, 26 Aug 2014 23:06:28 +0000 (16:06 -0700)]
Re-use switchable rate value in handle_inter_mode

The value of switchable rate has been stored in a local variable.
This change skips the second call to vp9_get_switchable_rate() by
reusing the local variable.

Change-Id: Ib7d3fef7621cc4bde94c6d6e6b3a71f1fd4559f2

10 years agoAdd an early termination check in handle_inter_mode
Jingning Han [Tue, 26 Aug 2014 22:53:56 +0000 (15:53 -0700)]
Add an early termination check in handle_inter_mode

Check the mode and motion vector cost. If it is already above
the existing best rate-distortion cost, skip the rest check process
on this mode.

Change-Id: Ie065cebdfda2a3be3be18b8e8b43dc29aaa8c179

10 years agoUse max txfm size unit in rate-distortion cost modeling
Jingning Han [Tue, 26 Aug 2014 19:34:54 +0000 (12:34 -0700)]
Use max txfm size unit in rate-distortion cost modeling

This commit makes the rate distortion modeling run in the unit of
maximum transform block size. No compression/speed change observed.
It is for the use of later fast forward transform purpose.

Change-Id: Ibaaedb69c765e8d0c5d5012f0ec07f36fd9f68fd

10 years agovp9: fix crash in mt loopfilter w/corrupt file
James Zern [Wed, 27 Aug 2014 19:36:23 +0000 (12:36 -0700)]
vp9: fix crash in mt loopfilter w/corrupt file

if the first frame was corrupt and loop filter not called, the next call
would assume the necessary allocations had been done and segfault when
accessing a NULL pointer

Change-Id: Ib6ef505e5c594e6f0fe65ab0700172bcf06b92a6

10 years agoAdd a separate datarate test for denoiser enabled.
Marco [Wed, 27 Aug 2014 16:40:30 +0000 (09:40 -0700)]
Add a separate datarate test for denoiser enabled.

Change-Id: Ibc91a46d11ad9c58e340b4346ed813a1be9b404a

10 years agoMerge "add a new interp filter search strategy."
Yaowu Xu [Wed, 27 Aug 2014 00:30:42 +0000 (17:30 -0700)]
Merge "add a new interp filter search strategy."

10 years agoMerge "Enable denoiser for vp8 datarate tests."
Marco [Wed, 27 Aug 2014 00:04:37 +0000 (17:04 -0700)]
Merge "Enable denoiser for vp8 datarate tests."

10 years agoEnable denoiser for vp8 datarate tests.
Marco [Tue, 26 Aug 2014 18:21:04 +0000 (11:21 -0700)]
Enable denoiser for vp8 datarate tests.

Keep one test, BasicBufferModel, with denoiser set off.

Change-Id: I762c2004b3adca24f677535cfa5b991108f31f01

10 years agoMerge "gen_msvs_(vcx)proj.sh: set yasm format explicitly"
James Zern [Tue, 26 Aug 2014 19:53:53 +0000 (12:53 -0700)]
Merge "gen_msvs_(vcx)proj.sh: set yasm format explicitly"

10 years agoMerge "vp8 fix: deallocate denoiser->yv12_last_source"
Marco [Tue, 26 Aug 2014 18:16:27 +0000 (11:16 -0700)]
Merge "vp8 fix: deallocate denoiser->yv12_last_source"

10 years agoReplacing int_mv with MV.
Dmitry Kovalev [Tue, 26 Aug 2014 17:53:05 +0000 (10:53 -0700)]
Replacing int_mv with MV.

Change-Id: I483a2fefc5f9ea4533dfd64448f3b6b426dd9eed

10 years agoMerge "Set scaled img format correctly in vpxdec"
Deb Mukherjee [Tue, 26 Aug 2014 17:36:47 +0000 (10:36 -0700)]
Merge "Set scaled img format correctly in vpxdec"

10 years agoMerge "twopass_encoder: Flush encoder."
Dmitry Kovalev [Tue, 26 Aug 2014 17:35:54 +0000 (10:35 -0700)]
Merge "twopass_encoder: Flush encoder."

10 years agoMerge "vpxenc.sh: Move some frequently used params into functions."
Tom Finegan [Tue, 26 Aug 2014 17:18:58 +0000 (10:18 -0700)]
Merge "vpxenc.sh: Move some frequently used params into functions."

10 years agovp8 fix: deallocate denoiser->yv12_last_source
Marco [Tue, 26 Aug 2014 17:12:43 +0000 (10:12 -0700)]
vp8 fix: deallocate denoiser->yv12_last_source

Change-Id: I7f3bc292b1b411e10586f0975143ae996ea78d1d

10 years agoadd a new interp filter search strategy.
Yaowu Xu [Wed, 20 Aug 2014 18:17:05 +0000 (11:17 -0700)]
add a new interp filter search strategy.

This commit addes a new strategy to reduce the search for optimal
interpolation filter type. The encoder counts and store how many each
filter type is selected and used for each of the reference frames.
A filter type that is rarely used for all three reference frames is
masked out to avoid computation.

The impact on compression is neglectible:
-0.02% on derf
+0.02% on stdhd

Encoding time is seen to reduce by 2~3%.

Change-Id: Ibafa92291b51185de40da513716222db4b230383

10 years agoRemoving unused arnr_type from VP9EncoderConfig and vp9_extracfg.
Dmitry Kovalev [Tue, 26 Aug 2014 06:48:52 +0000 (23:48 -0700)]
Removing unused arnr_type from VP9EncoderConfig and vp9_extracfg.

Change-Id: Icab9a4399c5687453f4bec14b8cb5000464335e5

10 years agotwopass_encoder: Flush encoder.
Dmitry Kovalev [Mon, 25 Aug 2014 22:29:20 +0000 (15:29 -0700)]
twopass_encoder: Flush encoder.

Change-Id: Ib04a545b43f3a36c6df7eaf3ed2d802cf41119bb

10 years agoMerge "Removing non-executable code from encoder_encode()."
Dmitry Kovalev [Tue, 26 Aug 2014 01:59:17 +0000 (18:59 -0700)]
Merge "Removing non-executable code from encoder_encode()."

10 years agoMerge "Removing tx_stepdown_count from VP9_COMP."
Dmitry Kovalev [Tue, 26 Aug 2014 01:37:40 +0000 (18:37 -0700)]
Merge "Removing tx_stepdown_count from VP9_COMP."

10 years agoMerge "Adding oxcf temp variable."
Dmitry Kovalev [Tue, 26 Aug 2014 01:37:33 +0000 (18:37 -0700)]
Merge "Adding oxcf temp variable."

10 years agoMerge "Adding is_keyframe temp var."
Dmitry Kovalev [Tue, 26 Aug 2014 01:36:59 +0000 (18:36 -0700)]
Merge "Adding is_keyframe temp var."

10 years agoRenaming vpxenc option from frame_boost to frame-boost.
Dmitry Kovalev [Tue, 26 Aug 2014 01:34:20 +0000 (18:34 -0700)]
Renaming vpxenc option from frame_boost to frame-boost.

We use dashes, no underscores.

Change-Id: I423eee946d10731856e15591eb6e9e4628e3ee6c

10 years agoRemoving non-executable code from encoder_encode().
Dmitry Kovalev [Mon, 25 Aug 2014 23:48:31 +0000 (16:48 -0700)]
Removing non-executable code from encoder_encode().

We don't output invisible frames with VP9.

Change-Id: I7b874d3ac454c1b2966d5d7d72e12a864b49afae

10 years agoMerge "Passing const cfg pointer to vpx_codec_{dec,enc}_init_ver()."
Dmitry Kovalev [Mon, 25 Aug 2014 22:17:46 +0000 (15:17 -0700)]
Merge "Passing const cfg pointer to vpx_codec_{dec,enc}_init_ver()."

10 years agovpxenc.sh: Move some frequently used params into functions.
Tom Finegan [Mon, 25 Aug 2014 21:57:42 +0000 (14:57 -0700)]
vpxenc.sh: Move some frequently used params into functions.

Echo args back from new util functions to decrease the repetition
in this file.

Change-Id: Id4d7c02bf81cf2c5261f58b560e033869a346677

10 years agoRemoving tx_stepdown_count from VP9_COMP.
Dmitry Kovalev [Mon, 25 Aug 2014 21:42:05 +0000 (14:42 -0700)]
Removing tx_stepdown_count from VP9_COMP.

The variable is never read.

Change-Id: I94141c1667fa5d10604cd6f83c5f64df107dee94

10 years agoMerge "[spatial svc]Multiple frame context feature"
Minghai Shang [Mon, 25 Aug 2014 21:29:49 +0000 (14:29 -0700)]
Merge "[spatial svc]Multiple frame context feature"

10 years agoMerge "vpxenc.sh: Make vpxenc() usage agree with implementation."
Tom Finegan [Mon, 25 Aug 2014 21:29:44 +0000 (14:29 -0700)]
Merge "vpxenc.sh: Make vpxenc() usage agree with implementation."

10 years agoMerge "twopass_encoder: Separate functions for first and second pass."
Dmitry Kovalev [Mon, 25 Aug 2014 21:29:14 +0000 (14:29 -0700)]
Merge "twopass_encoder: Separate functions for first and second pass."

10 years agovpxenc.sh: Make vpxenc() usage agree with implementation.
Tom Finegan [Thu, 21 Aug 2014 23:30:37 +0000 (16:30 -0700)]
vpxenc.sh: Make vpxenc() usage agree with implementation.

- Make $1 input file at all call sites.
- Actually use vpxenc_pipe().

Change-Id: I22f25719aeb3dce60863207b2e2a49e42e27ed06

10 years agoMerge "vpxenc.sh: Add real time tests."
Tom Finegan [Mon, 25 Aug 2014 21:27:11 +0000 (14:27 -0700)]
Merge "vpxenc.sh: Add real time tests."

10 years agoPassing const cfg pointer to vpx_codec_{dec,enc}_init_ver().
Dmitry Kovalev [Mon, 25 Aug 2014 21:02:53 +0000 (14:02 -0700)]
Passing const cfg pointer to vpx_codec_{dec,enc}_init_ver().

Change-Id: I3e319f581215ccb814afca247c1d911be231a4ac

10 years agoMerge "Replacing int_mv with MV inside the first pass code."
Dmitry Kovalev [Mon, 25 Aug 2014 20:56:14 +0000 (13:56 -0700)]
Merge "Replacing int_mv with MV inside the first pass code."

10 years agoMerge "Cleaning up is_background()."
Dmitry Kovalev [Mon, 25 Aug 2014 20:27:03 +0000 (13:27 -0700)]
Merge "Cleaning up is_background()."

10 years agotwopass_encoder: Separate functions for first and second pass.
Dmitry Kovalev [Mon, 25 Aug 2014 19:56:32 +0000 (12:56 -0700)]
twopass_encoder: Separate functions for first and second pass.

Change-Id: Ic5e3ff5f572ed55d7dd3df8c606e57daacf885dc

10 years agoMerge "Removing source field from VP9_COMP."
Dmitry Kovalev [Mon, 25 Aug 2014 19:53:19 +0000 (12:53 -0700)]
Merge "Removing source field from VP9_COMP."

10 years agoMerge "Removing is_best_mode() function."
Dmitry Kovalev [Mon, 25 Aug 2014 19:52:35 +0000 (12:52 -0700)]
Merge "Removing is_best_mode() function."

10 years agoCleaning up is_background().
Dmitry Kovalev [Mon, 25 Aug 2014 18:55:30 +0000 (11:55 -0700)]
Cleaning up is_background().

Change-Id: I2b9609dd22bacbf26e669f70bf155613b0316eb3

10 years ago[spatial svc]Multiple frame context feature
Minghai Shang [Mon, 18 Aug 2014 21:51:04 +0000 (14:51 -0700)]
[spatial svc]Multiple frame context feature

We can use one frame context for each layer so that we don't have
to reset the probs every frame. But we can't use prev_mi since we
may drop enhancement layers. So we have to generate a non vp9
compatible bitstream and modify it in the player.
1. We need to code all frames as invisible frame to let prev_mi
   not to be used. But in the bitstream we need to code the
   show_frame flag to 1 so that the publisher will know it's
   supposed to be a visible frame.
2. In the player we need to change the show_frame flag to 0 for
   all frames. Then add an one byte frame into the super frame
   to tell the decoder which layer we want to show.
Change-Id: I75b7304cf31f0ab952f043e33c034495e88f01f3

10 years agoMerge "enable wextra in vp9"
Jim Bankoski [Mon, 25 Aug 2014 13:40:22 +0000 (06:40 -0700)]
Merge "enable wextra in vp9"

10 years agoMerge "vp9_calc_ssim: remove unused parm to resolve warning"
Jim Bankoski [Mon, 25 Aug 2014 13:39:49 +0000 (06:39 -0700)]
Merge "vp9_calc_ssim: remove unused parm to resolve warning"

10 years agoMerge "vp9_denoiser_update_frame_stats: unused parm fixed"
Jim Bankoski [Mon, 25 Aug 2014 13:39:23 +0000 (06:39 -0700)]
Merge "vp9_denoiser_update_frame_stats: unused parm fixed"

10 years agoMerge "vp9_postproc.c: unused parameter warning resolved"
Jim Bankoski [Mon, 25 Aug 2014 13:39:12 +0000 (06:39 -0700)]
Merge "vp9_postproc.c: unused parameter warning resolved"

10 years agoinvalid_file_test: correct HandlePeekResult signature
James Zern [Sat, 23 Aug 2014 06:06:13 +0000 (23:06 -0700)]
invalid_file_test: correct HandlePeekResult signature

restore const lost in:
f0f9ab6 invalid_file_test.cc: remove unused param warnings

Change-Id: I0c5318f84c81e3dd7bb44468463faf9edd6bc18a

10 years agoMerge "vpxdec: fix --keep-going with --disable-vp8"
James Zern [Sat, 23 Aug 2014 02:52:52 +0000 (19:52 -0700)]
Merge "vpxdec: fix --keep-going with --disable-vp8"

10 years agoMerge "add libyuv/scale_neon64.cc"
James Zern [Sat, 23 Aug 2014 02:46:05 +0000 (19:46 -0700)]
Merge "add libyuv/scale_neon64.cc"

10 years agoMerge "active_map_test: use vpx_active_map_t() to initialize vars"
James Zern [Sat, 23 Aug 2014 02:44:58 +0000 (19:44 -0700)]
Merge "active_map_test: use vpx_active_map_t() to initialize vars"

10 years agoMerge "Choosing GOOD mode by default."
Dmitry Kovalev [Sat, 23 Aug 2014 01:43:34 +0000 (18:43 -0700)]
Merge "Choosing GOOD mode by default."

10 years agoadd libyuv/scale_neon64.cc
James Zern [Sat, 23 Aug 2014 01:16:44 +0000 (18:16 -0700)]
add libyuv/scale_neon64.cc

from r1051, missed in r1060 snapshot update

Change-Id: I3756d8d29fc0dd4c4491efa84b8f8837edde23a7

10 years agovpxdec: fix --keep-going with --disable-vp8
James Zern [Sat, 23 Aug 2014 01:01:28 +0000 (18:01 -0700)]
vpxdec: fix --keep-going with --disable-vp8

the parsing of this flag was mistakenly put in a CONFIG_VP8_DECODER
conditional block in:
95853db vpxdec: add --keep-going option

Change-Id: Ie83ca0399fd3f3d4b0a9d03b7ca5536b310e1f02

10 years agoMerge "Removing extraconfig_map from vp8_cx_iface.c."
Dmitry Kovalev [Sat, 23 Aug 2014 00:20:42 +0000 (17:20 -0700)]
Merge "Removing extraconfig_map from vp8_cx_iface.c."

10 years agoactive_map_test: use vpx_active_map_t() to initialize vars
James Zern [Sat, 23 Aug 2014 00:10:18 +0000 (17:10 -0700)]
active_map_test: use vpx_active_map_t() to initialize vars

0-initializes [1] and removes a warning for missing initializer fields

[1] http://en.cppreference.com/w/cpp/language/value_initialization

Change-Id: I886fa72c74679b52cdce2af3364172395a5b6a51