1. 08 Aug, 2017 6 commits
    • Tom Finegan's avatar
      Fix CONFIG_DAALA_DCT4/8/16 builds with CMake. · 67a7d9f7
      Tom Finegan authored
      Build aom_ports/x86/emms.asm for all x86 targets. It was
      previously guarded by HAVE_MMX, which violates assumptions
      established by the configure build.
      Change-Id: Icac2b0ba6272c61dbb71dfc3824a8cf391c251b6
    • Tom Finegan's avatar
      Fix CONFIG_BGSPRITE builds with CMake. · 33698326
      Tom Finegan authored
      Add av1/encoder/bgsprite.[ch] to the CMake build.
      Change-Id: Ic90c81c724be6814a3534f2f612173ac325d4423
    • Tom Finegan's avatar
      Fix CONFIG_AOM_QM builds with CMake. · 92d50233
      Tom Finegan authored
      Exclude test/quantize_func_test.cc from the build when
      CONFIG_AOM_QM or CONFIG_NEW_QUANT are enabled to mimic the
      configure build.
      Change-Id: Ib1358d466aedad4a784b4693f74babacf86cda7a
    • Tom Finegan's avatar
      Fix CONFIG_ANS builds with CMake. · 1d179187
      Tom Finegan authored
      Exclude test/binary_codes_test.cc from the build when
      CONFIG_ANS is enabled to mimic the configure build.
      Change-Id: Ia40dcc667173d50efb55a09b6ae863bf44e15608
    • Yushin Cho's avatar
      Make braces matchable · 4b92ab86
      Yushin Cho authored
      Change-Id: I7ef1fe3018d720663d7c32555f7d11e9ec1ceb7b
    • Tom Finegan's avatar
      Add sanitizer support to CMake build. · 8cc239b7
      Tom Finegan authored
      Enabled via addition of -DSANITIZE=<sanitizer> to CMake
      command line at generation time.
      Change-Id: I32b5fc547bcd0329703d9012aa864d6a9cd6cb64
  2. 07 Aug, 2017 8 commits
    • Yue Chen's avatar
      Compute mode costs from cdf · ba47c842
      Yue Chen authored
      Previously, a lot of costs are computed from per-frame-updated
      probs, which cannot reflect the actual entropy models even if we
      just consider initial entropy models in cm->fc for each frame.
      For most symbols, cost update is still on per frame basis in this
      patch, unless cdf is updated in RDO.
      Performance change: -0.011 lowres
      Change-Id: Id2faf88c7638ffe2fa414bac5d0b05cb1f65459a
    • Jingning Han's avatar
      Use the bottom right 4x4 block for 8x8 unit mv reference · 74978fcb
      Jingning Han authored
      When accessing an 8x8 block for motion vector referencing, always
      use the bottom right 4x4 block's motion information.
      Change-Id: I95e7235f9645ea3dea83482041b9fbd4e764b13f
    • Jingning Han's avatar
      Use 4x4 granularity for nearest neighbor mv search · d37b83d8
      Jingning Han authored
      Search the immediate neighbor for reference motion vectors in the
      unit of 4x4 block size. This improves the compression performance
      by 0.05%.
      Change-Id: Iaf04aa2c8986de6fbc22fd56e137c1f0bcb38185
    • Jingning Han's avatar
      Use 8x8 block mv search for sub8x8 blocks · 7f99477a
      Jingning Han authored
      Use 8x8 block search resolution for reference motion vectors for
      non-nearest neighbors of a sub8x8 coding block. The compression
      performance is slightly up by 0.05%.
      Change-Id: I83d99a2232b0bb92f87dd22a77a4005ce0c0c883
    • Cheng Chen's avatar
      Make braces consistent in encodeframe.c · 50b0f6ca
      Cheng Chen authored
      Enable vim to jump to correct corresponding braces, which
      is helpful especially when the function is long.
      Change-Id: Ibcf4f11df3e923e12bc33ec1656bb6dbd0451be6
    • David Barker's avatar
      Fix bug in highbd wedge-interintra prediction · b746c3b7
      David Barker authored
      When calling aom_(highbd_)blend_a64_mask for a subsampled plane,
      the mask is subsampled in-place. This means that the mask stride
      passed in needs to be the value before subsampling. Since the
      wedge masks are set up so that mask stride = mask width, this means
      that the stride should equal the width of the current mi block.
      However, for the wedge-interintra path in combine_interintra_highbd,
      we subsampled the mask stride, resulting in the wrong stride being
      used for chroma planes if subsampling_x == 1.
      The lowbd path already uses the correct stride, so this patch just
      copies the stride calculation over to the highbd path.
      Change-Id: I08a4120b11a38912c45d9e637a54e009eddd88ec
    • Wei-Ting Lin's avatar
      ncobmc-adapt-weight: remove bitstream for training · 7daf0422
      Wei-Ting Lin authored
      Change-Id: I08340afd1bc667b97d96e5540cf5c126bcfaaab4
    • Rupert Swarbrick's avatar
      Tidy up and generalise av1_predict_intra_block · 914d53a1
      Rupert Swarbrick authored
      This tidies up the code somewhat, factoring the snapshot/restore of
      rows and columns in the reference frame into a separate function. The
      variable names might also be a little easier to follow.
      The generalisation is that av1_predict_intra_block no longer requires
      that a rectangular block is exactly 2:1 or 1:2. The new code will also
      allow tiling a rectangular transform size (say two 1:2 transforms to
      make a 1:4 transform).
      Change-Id: Ie03cb01cce4645b49d943533c57a495dddc6e55d
  3. 05 Aug, 2017 4 commits
  4. 04 Aug, 2017 13 commits
  5. 03 Aug, 2017 9 commits
    • Angie Chiang's avatar
      Fix integer overflow in AV1Convolve2DTest · 9f65bd51
      Angie Chiang authored
      Change-Id: Ie8a0eda6de60d09bbda49eb101e278572af75613
    • Yaowu Xu's avatar
      Replace shift with multiply · bc83b642
      Yaowu Xu authored
      To avoid left shift of negative values.
      Change-Id: I8dacf99f162771a58bef1f839cafd0be9a8f5e86
    • Sarah Parker's avatar
      Add macros to turn off inter and intra mrc_dct separately · 2e08d96d
      Sarah Parker authored
      This will aid in testing different masking methods for inter
      and intra blocks.
      Change-Id: Ic038da77e55405e3303177e6cd260bd5e19311c1
    • hui su's avatar
      Calculate coeff token cost before encoding each SB · bd57abe0
      hui su authored
      Compression gain:
      -0.42	-0.15   	-0.38	-0.62
      Google testsets
      lowres -0.18%
      midres -0.32%
      No obvious encoding speed changes observed.
      Change-Id: I888cf4b22b591d79d80b2d77c9eb10d84e5de722
    • Yaowu Xu's avatar
      Avoid the use of INT_MAX in rate computation · 897dd193
      Yaowu Xu authored
      Change-Id: I278e84de26d995419b85039f7b78b8324cadf3a3
    • Zoe Liu's avatar
      Add a cast to avoid a Visual Studio build warning · 708e08b7
      Zoe Liu authored
      Change-Id: I82fe5145b83c690bff50e4104adcf22aeb973903
    • Sarah Parker's avatar
      Ensure ext-tx is enabled when mrc-tx is enabled · 33ef8832
      Sarah Parker authored
      When ext-tx is disabled, the mrc-tx implementation is not complete,
      so they must be enabled together for now.
      Change-Id: Ib049f0e15023272c44a905581842db0626cdf14d
    • hui su's avatar
      Calculate coeff token cost from CDF · c0cf71df
      hui su authored
      AWCY results:
      -0.09	-0.04	  -0.02	  -0.03
      On Google testsets:
      lowres  -0.18%
      midres  -0.20%
      Above results are obtained with
      --disable-ext-refs --disable-dual-filter --disable-loop-restoration
      --disable-global-motion --disable-warped-motion
      Change-Id: Iba58d5e5ec9a65d0afba29609aa2e379a80d7236
    • Yaowu Xu's avatar
      Prevent redundant down converting operation · 9b0f7038
      Yaowu Xu authored
      Adds a flag "buf_8bit_valid" to indicate if buf_8bit is synchronized
      with HBD buffer, the flag is then used to avoid possible redundant
      operation in down conversion the hbd buffer to buf_8bit.
      Also this commit also moves the operation to prepare source frame for
      coding outside of recode loop to avoid duplicate operations there.
      Change-Id: I67a1eb408b75bb734bd8a16fe6ffbd58c93a230f