]>
granicus.if.org Git - libvpx/log
Angie Chiang [Tue, 25 Oct 2016 16:24:23 +0000 (16:24 +0000)]
Merge "Refactor: Add macro LOG_SWITCHABLE_FILTERS" into nextgenv2
Angie Chiang [Mon, 24 Oct 2016 19:24:44 +0000 (12:24 -0700)]
Refactor: Add macro LOG_SWITCHABLE_FILTERS
Change-Id: I7593ff2f8949d8bc26ca1c8577faaefb09640b59
Debargha Mukherjee [Tue, 25 Oct 2016 16:21:20 +0000 (16:21 +0000)]
Merge "Fix to make intra_only frames decodable out of order" into nextgenv2
Yaowu Xu [Tue, 25 Oct 2016 16:04:32 +0000 (16:04 +0000)]
Merge "update_state_supertx: rename a shadowed variable" into nextgenv2
Yaowu Xu [Tue, 25 Oct 2016 16:03:47 +0000 (16:03 +0000)]
Merge "Move small fixes and refactoring for obmc pred from AV1" into nextgenv2
Yaowu Xu [Tue, 25 Oct 2016 16:03:40 +0000 (16:03 +0000)]
Merge "adapt_scan experiment" into nextgenv2
Thomas Daede [Mon, 19 Sep 2016 20:24:51 +0000 (13:24 -0700)]
Automatically upgrade profile to match input chroma subsampling.
This is a follow-up to
1195a396f6c53a5bd35559eed957c2aac855f7e .
Change-Id: I4db554e5d88290d55258062e29a1652707d91037
Yaowu Xu [Mon, 24 Oct 2016 19:55:54 +0000 (12:55 -0700)]
Remove select_tx_size from struct macroblock
It is no longer used anywhere.
Change-Id: I5d40664373d66821c5382f6155302b8719ce58c0
Guillaume Martres [Mon, 22 Aug 2016 02:02:33 +0000 (19:02 -0700)]
Avoid unnecessary reencode in choose_largest_tx_size
This change is similar to the one done for choose_tx_size_from_rd in
daf841b4a10ece1b6831300d79f271d00f9d027b
It gives a 4% speed-up on bus_cif.y4m with the following settings:
--cpu-used=4 -p 1 --end-usage=q --cq-level=40 --tile-columns=0 --tile-rows=0
Change-Id: Ic54fe4a066a2c0b5f6349d80cd13de8bb8ddcabc
Brennan Shacklett [Sat, 20 Aug 2016 00:34:11 +0000 (17:34 -0700)]
Remove encode_breakout and related speed features
Seems to be dead code
Change-Id: I17b3edc9e82d6a1da172a686522358a6b1a630e9
Peter de Rivaz [Tue, 25 Oct 2016 09:19:57 +0000 (10:19 +0100)]
Fix to make intra_only frames decodable out of order
last_frame_type is not well defined for intra_only frames
if we are decoding them out of order.
This change removes a dependency on last_frame_type for these frames.
Change-Id: I440cac68792714de222e192a0b3e75f6e1aa5e4b
Sarah Parker [Tue, 25 Oct 2016 02:00:48 +0000 (02:00 +0000)]
Merge "Extend warp_frame functions to average compound predictions" into nextgenv2
Angie Chiang [Fri, 21 Oct 2016 23:44:47 +0000 (16:44 -0700)]
adapt_scan experiment
Performance improvement
BDRate
lowres 0.921%
midres 0.730%
hdres 1.019%
Change-Id: I26208d6c0531937bff44de505b4ea355c7852802
Urvang Joshi [Tue, 25 Oct 2016 01:06:28 +0000 (01:06 +0000)]
Merge changes I2153c57e,I0e291edd into nextgenv2
* changes:
Palette: Generate encodings automatically from tree.
Palette + Ext-Intra: shadowed declaration fix.
Yue Chen [Mon, 24 Oct 2016 23:16:08 +0000 (23:16 +0000)]
Merge "Fix bugs in SUB8X8_MC" into nextgenv2
Jingning Han [Mon, 24 Oct 2016 22:24:04 +0000 (22:24 +0000)]
Merge "Refactor tx_size to pixel number mapping in reconintra.c" into nextgenv2
Jingning Han [Mon, 24 Oct 2016 22:23:59 +0000 (22:23 +0000)]
Merge "Simplify variable defs in av1_encode_block_intra" into nextgenv2
Jingning Han [Mon, 24 Oct 2016 22:23:40 +0000 (22:23 +0000)]
Merge "Refactor av1_predict_intra_block tx_size interface" into nextgenv2
Jingning Han [Mon, 24 Oct 2016 22:23:36 +0000 (22:23 +0000)]
Merge "Add block size in pixels lookup table" into nextgenv2
Angie Chiang [Mon, 24 Oct 2016 21:41:18 +0000 (21:41 +0000)]
Merge "Fix unsigned type error in aom_scale.c" into nextgenv2
Urvang Joshi [Mon, 24 Oct 2016 21:06:43 +0000 (14:06 -0700)]
Palette: Generate encodings automatically from tree.
Ran some manual sanity checks:
- Verified that the automatically generated encodings match the
hand-written encodings before the patch.
- Verified that the encoded bitstream before/after this patch is
identical.
Change-Id: I2153c57e463cff09c1d03d619b432fb1015199c3
Yue Chen [Fri, 21 Oct 2016 23:50:52 +0000 (16:50 -0700)]
Move small fixes and refactoring for obmc pred from AV1
Covering commits
1c263e0 and
79d8a07 from AOM codebase
Change-Id: I6400e5f99bbb2ef6584ef232d465e520230c06e0
Urvang Joshi [Mon, 24 Oct 2016 21:13:55 +0000 (14:13 -0700)]
Palette + Ext-Intra: shadowed declaration fix.
This shadowed declaration warning was generated when both experiments
are on.
Change-Id: I0e291eddeefabd68c5c3a0e5f8ac87706a82d55a
Jingning Han [Fri, 29 Jul 2016 17:48:34 +0000 (10:48 -0700)]
Prevent potential token buffer overflow in format 444
For a 16x16 pixel block, one needs to allocate 16x16 coefficient
tokens, plus up to 16 eob tokens, per plane. This commit increases
the token allocation size to cover the case where all the transform
blocks are of size 4x4 in format 444.
Change-Id: I5755e6a53771053d51163d01ec1d62e670c5009e
Thomas Daede [Fri, 26 Feb 2016 01:23:17 +0000 (17:23 -0800)]
Automatically set internal bit depth to at least the input bit depth.
Upgrade profile if required.
Change-Id: Ieb2b77d2446290a8fc749739247a01e8f0600c55
Jingning Han [Mon, 24 Oct 2016 21:01:29 +0000 (21:01 +0000)]
Merge "Add MAX_VARTX_DEPTH macro" into nextgenv2
Jingning Han [Mon, 24 Oct 2016 21:01:17 +0000 (21:01 +0000)]
Merge "Simplify the recursive transform block decoding" into nextgenv2
Yaowu Xu [Mon, 24 Oct 2016 20:57:55 +0000 (20:57 +0000)]
Merge "Correct data size estimation for odd size video" into nextgenv2
Debargha Mukherjee [Mon, 24 Oct 2016 19:53:41 +0000 (19:53 +0000)]
Merge "Fix a bug when combining new-quant + supertx" into nextgenv2
Yue Chen [Mon, 24 Oct 2016 19:21:48 +0000 (12:21 -0700)]
Fix bugs in SUB8X8_MC
Change-Id: Ia544974f83c6b7f9cdb148eeb13a6d0c6eb4ed24
Yaowu Xu [Mon, 24 Oct 2016 19:21:44 +0000 (19:21 +0000)]
Merge "Increase min size of compressed data" into nextgenv2
Yaowu Xu [Mon, 24 Oct 2016 19:21:29 +0000 (19:21 +0000)]
Merge "Avoid the use of uninitialized value in ActiveMap encoding route" into nextgenv2
Angie Chiang [Mon, 24 Oct 2016 16:34:53 +0000 (09:34 -0700)]
Fix unsigned type error in aom_scale.c
Avoid unary minus operator applied to unsigned type
Change-Id: I6986cd2b0ea236e0129ee94c02275593c287a87d
Yaowu Xu [Mon, 24 Oct 2016 18:34:29 +0000 (18:34 +0000)]
Merge "Use the actual inter prediction filter buffer in DRL" into nextgenv2
Yaowu Xu [Mon, 24 Oct 2016 18:25:55 +0000 (18:25 +0000)]
Merge "fdct4x4_test: fix unsigned overflow" into nextgenv2
Debargha Mukherjee [Mon, 24 Oct 2016 18:14:31 +0000 (18:14 +0000)]
Merge "Add bit accounting information for deringing" into nextgenv2
Yaowu Xu [Mon, 24 Oct 2016 18:10:32 +0000 (18:10 +0000)]
Merge "Use remove some magic numbers in aom_rans_merge_prob8_pdf." into nextgenv2
Jingning Han [Mon, 24 Oct 2016 18:04:47 +0000 (11:04 -0700)]
Refactor tx_size to pixel number mapping in reconintra.c
Change-Id: Id66a14a869df8317c5bbb693d14262326fe84206
Jingning Han [Mon, 24 Oct 2016 17:32:37 +0000 (10:32 -0700)]
Simplify variable defs in av1_encode_block_intra
Use direct table access to fetch the block size and transform size
in pixels.
Change-Id: Ia0093d5aed912be24996a06b0567bb2d873ec068
Jingning Han [Mon, 24 Oct 2016 17:27:28 +0000 (10:27 -0700)]
Refactor av1_predict_intra_block tx_size interface
Simplify the input arguments. Make direct use of the block size
in the unit of pixels.
Change-Id: Ifec9d90b4b4fa9605f93b4f93b8242f76f898b5f
Yaowu Xu [Tue, 12 Jul 2016 18:57:51 +0000 (11:57 -0700)]
Correct data size estimation for odd size video
Given the largest transform size is 32x32, this commmit changes size
estiiation based on the size rounding up to 32 multiples to avoid
insufficient buffer allocations.
BUG=https://bugs.chromium.org/p/aomedia/issues/detail?id=36
Change-Id: I6eab09dc6acdc0f5a6bcadb918d62c4852aae21f
Jingning Han [Mon, 24 Oct 2016 17:38:43 +0000 (10:38 -0700)]
Add MAX_VARTX_DEPTH macro
Change-Id: I85532cf88f91f0f0cb4d9cb4b2dbda8a181297bf
Yaowu Xu [Mon, 11 Jul 2016 18:38:53 +0000 (11:38 -0700)]
Increase min size of compressed data
This commit increases the minimum size for allocated buffer for
compressed data. The old size underestimated the size needed for
small images with width or height less than 64 pixels.
BUG=https://bugs.chromium.org/p/aomedia/issues/detail?id=31
Change-Id: Ia12507edc2be1e737ec49c32f64fd2ebf1eab41f
David Barker [Mon, 24 Oct 2016 09:55:35 +0000 (10:55 +0100)]
Fix a bug when combining new-quant + supertx
Previously, we assumed that av1_init_plane_quantizers is always called with
segment_id == xd->mi[0]->mbmi.segment_id (and use the latter to derive the value
of 'qindex' to use in the quantizer). But this is no longer true when supertx
is enabled. This patch instead remembers the value of 'qindex' derived from
the latest call to av1_init_plane_quantizers and uses that directly.
Change-Id: Ifa1c5bf74cad29942ff79b88ca92c231bc07f336
Jingning Han [Mon, 11 Jul 2016 18:24:24 +0000 (11:24 -0700)]
Avoid the use of uninitialized value in ActiveMap encoding route
This commit resets the transform size to be the maximum possible
value. It avoids out-of-boundary writing when the ActiveMap is
turned on.
Change-Id: I8302dd9a5c9fffaea3edf9ad33f72aa111999737
Angie Chiang [Mon, 24 Oct 2016 16:36:36 +0000 (16:36 +0000)]
Merge "Align frame contexts." into nextgenv2
Jingning Han [Mon, 24 Oct 2016 16:32:41 +0000 (09:32 -0700)]
Use the actual inter prediction filter buffer in DRL
This avoids an encoding segmentation fault in speed 5, due to the
use of uninitialized dummy inter prediction filter buffer in the
dynamic motion vector referencing scheme.
Change-Id: Icd888d46623e8abf34267838135eed8656d552e4
Yaowu Xu [Mon, 24 Oct 2016 16:21:09 +0000 (09:21 -0700)]
fdct4x4_test: fix unsigned overflow
The difference between src and dst will be signed, the error will be
unsigned. The change quiets -fsanitize=integer:
unsigned integer overflow:
4294967295 *
4294967295
Change-Id: I131cefcc9583ee8a5b98eb5182fd30e9c7237ea0
David Barker [Mon, 24 Oct 2016 12:25:23 +0000 (13:25 +0100)]
Add bit accounting information for deringing
It seems that when bit accounting was introduced in
https://chromium-review.googlesource.com/#/c/400658/
there was one place which was accidentally skipped, leading to build failures
with --enable-dering. This patch adds the missing information.
Change-Id: I59e1bd6f7e1d4fa58506ee7af307b845c78a7cbe
Alex Converse [Mon, 27 Jun 2016 08:37:10 +0000 (01:37 -0700)]
Use remove some magic numbers in aom_rans_merge_prob8_pdf.
Change-Id: I0cefae17642d7adf1b9bd637ecb81b437629aa0c
Jingning Han [Mon, 24 Oct 2016 15:57:28 +0000 (15:57 +0000)]
Merge "Limit the transform block partition depth" into nextgenv2
Jingning Han [Mon, 24 Oct 2016 15:57:17 +0000 (15:57 +0000)]
Merge "Allow frame level tx_mode switch" into nextgenv2
Jingning Han [Mon, 24 Oct 2016 15:56:33 +0000 (15:56 +0000)]
Merge "Separate intra and inter tx_size counting" into nextgenv2
Jingning Han [Mon, 24 Oct 2016 00:46:42 +0000 (17:46 -0700)]
Make set context function aware of rectangular transform block size
Account for the rectangular transform block size in setting the
context data.
Change-Id: Ic30a6a3eaaca4c945e0aab3acbaeb99aa48b0064
Yaowu Xu [Sun, 23 Oct 2016 03:15:37 +0000 (03:15 +0000)]
Merge "Add a runtime flag to enable bit accounting." into nextgenv2
Yaowu Xu [Sun, 23 Oct 2016 03:15:21 +0000 (03:15 +0000)]
Merge "Add a decoder control to retrieve accounting data." into nextgenv2
James Zern [Sat, 22 Oct 2016 19:42:43 +0000 (12:42 -0700)]
update_state_supertx: rename a shadowed variable
Change-Id: I0e5fa71a4b7cd03c9e28b434b1ea72b090ca6772
Jingning Han [Sat, 22 Oct 2016 03:41:05 +0000 (03:41 +0000)]
Merge "Fix comment typo in common_data.h" into nextgenv2
Jingning Han [Sat, 22 Oct 2016 03:40:53 +0000 (03:40 +0000)]
Merge "Refactor tx_size to pixel number mapping in reconintra.c" into nextgenv2
Jingning Han [Sat, 22 Oct 2016 03:40:43 +0000 (03:40 +0000)]
Merge "Refactor tx_size step use cases in decoder" into nextgenv2
Jingning Han [Sat, 22 Oct 2016 03:40:32 +0000 (03:40 +0000)]
Merge "Replace tx_size_1d with tx_size_wide/high" into nextgenv2
Yi Luo [Sat, 22 Oct 2016 01:46:09 +0000 (01:46 +0000)]
Merge "Fix avx2 16x16/32x32 fwd txfm coeff output on HBD" into nextgenv2
Yaowu Xu [Sat, 22 Oct 2016 01:18:35 +0000 (01:18 +0000)]
Merge "Fix compiler warning when CONFIG_ACCOUNTING enabled." into nextgenv2
Jingning Han [Sat, 22 Oct 2016 01:07:32 +0000 (01:07 +0000)]
Merge "Refactor decoder side qcoeff reset" into nextgenv2
Sarah Parker [Sat, 22 Oct 2016 00:06:37 +0000 (17:06 -0700)]
Extend warp_frame functions to average compound predictions
Change-Id: I400e95161d576510423880b5b9923a2307b5eb02
Angie Chiang [Fri, 21 Oct 2016 23:16:56 +0000 (16:16 -0700)]
Align frame contexts.
This will allow for aligned cdfs and scan orders inside.
Change-Id: I8ebcd64d55e41da20f518a39ae6ef192def70109
Angie Chiang [Sat, 22 Oct 2016 00:13:59 +0000 (17:13 -0700)]
Run clang-format on entropymv.c
Change-Id: Ic9f34e32e51f8a8a4426543bae0b92f5fab0792e
Jingning Han [Fri, 21 Oct 2016 23:40:47 +0000 (16:40 -0700)]
Add block size in pixels lookup table
This prepares for the next refactoring to support 2x2 transform
block sizes.
Change-Id: Ia06bc487da34e853ef9323cd13e3d482e819db43
Jingning Han [Fri, 21 Oct 2016 23:30:10 +0000 (16:30 -0700)]
Fix comment typo in common_data.h
"varios" -> "various"
Change-Id: If91a462dc009f701c48c2cfd7965cd71f61f2970
Nathan E. Egge [Wed, 5 Oct 2016 23:33:48 +0000 (19:33 -0400)]
Add a runtime flag to enable bit accounting.
By default, when building with --enable-accounting the bit accounting
code will collect statistics for every frame while decoding.
Collecting statistics can slow down decode time and we would eventually
like to enable the CONFIG_ACCOUNTING flag by default.
This patch adds a runtime flag so that bit accounting statistics are
only collected when actually needed.
Change-Id: I25d9eaf26ea132d61ace95b952872158c9ac29e7
Nathan E. Egge [Thu, 6 Oct 2016 01:28:36 +0000 (21:28 -0400)]
Add a decoder control to retrieve accounting data.
This decoder control requires AV1 to be compiled with --enable-accounting.
Note that bit accounting data is only available after a frame has been
decoded.
Change-Id: I8a15213d9f2587638e0edb62932738e985160e03
Nathan E. Egge [Wed, 5 Oct 2016 23:30:15 +0000 (19:30 -0400)]
Fix compiler warning when CONFIG_ACCOUNTING enabled.
ISO C90 forbids mixed declarations and code and the function
aom_accounting_set_context() was being called before the MB_MODE_INFO
declaration.
Change-Id: I8619525b1b2fd37753891bd310d9d59c881b8807
Nathan E. Egge [Thu, 8 Sep 2016 19:59:53 +0000 (15:59 -0400)]
Update class0_fp_cdf and fp_cdf tables once per frame.
Move computing the class0_fp_cdf and fp_cdf tables per coded mv
symbol to computing it only when the probabilities are updated.
Change-Id: Ib4957c8ab21e6189bcc3817a07b7681dfb343223
Nathan E. Egge [Thu, 8 Sep 2016 19:08:48 +0000 (15:08 -0400)]
Update class_cdf table once per frame.
Move computing the class_cdf table per coded mv class symbol to
computing it only when the probabilities are updated.
Change-Id: I6c4a9075817e8ba2e251f0e82436995f08f2ec5c
Nathan E. Egge [Thu, 8 Sep 2016 15:22:03 +0000 (11:22 -0400)]
Update joint_cdf table once per frame.
Move computing the joint_cdf table per coded mv joint symbol to
computing it only when the probabilities are updated.
Change-Id: If5d195f70e6fad7b60f69606c8386ad5e69657d2
Nathan E. Egge [Thu, 8 Sep 2016 14:41:20 +0000 (10:41 -0400)]
Update inter_mode_cdf tables once per frame.
Move computing the inter_mode_cdf tables per coded inter mode symbol to
computing them only when the probabilities are updated.
Change-Id: I7a7b059ee75723cb6f278ed82a20cf34c27915d8
Yaowu Xu [Fri, 21 Oct 2016 22:53:42 +0000 (22:53 +0000)]
Merge "Update uv_mode_cdf tables once per frame." into nextgenv2
Jingning Han [Fri, 21 Oct 2016 17:14:36 +0000 (10:14 -0700)]
Limit the transform block partition depth
Limit the recursive transform block partition depth to 2. For a
32x32 transform block unit, one can maximally go down to 8x8 transform
block size.
Change-Id: I2caa92bb2eee64762b7ecca8920259f7c50fb0aa
Yaowu Xu [Fri, 21 Oct 2016 22:44:09 +0000 (22:44 +0000)]
Merge "Update y_mode_cdf tables once per frame." into nextgenv2
Yaowu Xu [Fri, 21 Oct 2016 22:43:56 +0000 (22:43 +0000)]
Merge "Update kf_y_mode_cdf tables once per frame." into nextgenv2
Jingning Han [Thu, 20 Oct 2016 22:17:43 +0000 (15:17 -0700)]
Allow frame level tx_mode switch
Check the encoding statistics. If all the coding blocks use the
max transform size, skip transform size coding in the frame header.
Change-Id: I31cb16314e87f945d7e95a34a90a5536b3ed82d5
Jingning Han [Thu, 20 Oct 2016 19:00:15 +0000 (12:00 -0700)]
Separate intra and inter tx_size counting
Skip counting the inter transform block size distribution for
the intra transform block size coding.
Change-Id: Ifad9d843f57d069d0619a54d66ca18101e1b69f1
Jingning Han [Fri, 21 Oct 2016 19:55:54 +0000 (12:55 -0700)]
Simplify the recursive transform block decoding
Remove unneeded block index.
Change-Id: Ifceab4985d3ccd65d4c0a110de83a0b457ce5868
Jingning Han [Fri, 21 Oct 2016 20:04:31 +0000 (13:04 -0700)]
Refactor tx_size to pixel number mapping in reconintra.c
Change-Id: I1e4a43f5f08b76867240a207c60d7e85a8ffbb74
Jingning Han [Fri, 21 Oct 2016 19:44:29 +0000 (12:44 -0700)]
Refactor tx_size step use cases in decoder
Use lookup table to replace the arithmetic computation for transform
block step.
Change-Id: I1318d81bda9d7ffaf9d550acd19354b0615ede36
Jingning Han [Fri, 21 Oct 2016 19:06:02 +0000 (12:06 -0700)]
Replace tx_size_1d with tx_size_wide/high
This prepares the support to both rectangular and 2x2 transform
block sizes.
Change-Id: I3c2d4e317f6b627bb45d2273c278331bd976ee92
Jingning Han [Fri, 21 Oct 2016 18:48:15 +0000 (11:48 -0700)]
Refactor decoder side qcoeff reset
Allow the decoder to memset partial dequantized coefficient line
to zero.
Change-Id: I1f07dc7bf802958754502c1b5c819cc81e7a08cb
Yaowu Xu [Fri, 21 Oct 2016 21:51:50 +0000 (21:51 +0000)]
Merge "Pass AV1_COMMON into get_scan" into nextgenv2
Yaowu Xu [Fri, 21 Oct 2016 21:50:13 +0000 (21:50 +0000)]
Merge "Decoder performance improvement with daala_ec." into nextgenv2
Sarah Parker [Fri, 21 Oct 2016 21:16:54 +0000 (21:16 +0000)]
Merge "Fix logical vs bitwise & bug" into nextgenv2
Yi Luo [Fri, 21 Oct 2016 18:13:48 +0000 (11:13 -0700)]
Fix avx2 16x16/32x32 fwd txfm coeff output on HBD
Change-Id: Ida036defe5688894a63007a31aa2dd0b3f0b5d59
Jingning Han [Fri, 21 Oct 2016 21:11:30 +0000 (21:11 +0000)]
Merge "Fix unused variable error in intrapred.c" into nextgenv2
Jingning Han [Fri, 21 Oct 2016 19:39:22 +0000 (19:39 +0000)]
Merge "Refactor tx_size to pixel number in decodeframe.c" into nextgenv2
Nathan E. Egge [Thu, 8 Sep 2016 14:13:42 +0000 (10:13 -0400)]
Update uv_mode_cdf tables once per frame.
Move computing the uv_mode_cdf tables per coded intra mode symbol to
computing them only when the probabilities are updated.
Change-Id: I627b59d30726c913f5d7ba7753cb0446a12655bb
Nathan E. Egge [Thu, 8 Sep 2016 14:01:16 +0000 (10:01 -0400)]
Update y_mode_cdf tables once per frame.
Move computing the y_mode_cdf tables per coded intra mode symbol to
computing them only when the probabilities are updated.
Change-Id: I8c43d09b8ef5febe2a3ec64bd51d28bd78ea73ed
Nathan E. Egge [Wed, 7 Sep 2016 22:20:41 +0000 (18:20 -0400)]
Update kf_y_mode_cdf tables once per frame.
Move computing the kf_y_mode_cdf tables per coded intra mode symbol to
computing them only when the probabilities are updated.
Change-Id: I5999447050c2f7d5dbccde80bee05ecd1c5440ab
Nathan E. Egge [Fri, 9 Sep 2016 18:21:56 +0000 (14:21 -0400)]
Decoder performance improvement with daala_ec.
Cherry-pick Daala
b5020bee :
Remove redundant test in od_ec_decode_bool_q15().
Using a test that decodes 100M random binary symbols, making this change
produced a speed up of 8.81% with gcc-4.9.3 and 3.71% with clang-3.7.1,
both compiled with -O2.
Change-Id: If6d0077a56121a575ae53bcd4d1d9b7d800a317d
Yaowu Xu [Fri, 21 Oct 2016 18:46:29 +0000 (18:46 +0000)]
Merge "Use divide by multiply in the ans writer." into nextgenv2
Angie Chiang [Fri, 21 Oct 2016 18:02:09 +0000 (11:02 -0700)]
Pass AV1_COMMON into get_scan
This CL will facilitate adapt_scan experiment.
In adapt_scan experiment, dynamic scan order will be stored in
AV1_COMMON
Change-Id: I4763ea931b5e1af54d4f173971befeb01a4db335
Yaowu Xu [Fri, 21 Oct 2016 18:46:18 +0000 (18:46 +0000)]
Merge "Sub8x8 block chroma component inter prediction" into nextgenv2