git.videolan.org Git - x264.git/summary

Les articles publiés sur le site

• Remove compatibility workarounds

  6 juin 2017, par Anton Mitrofanov

  Remove compatibility workarounds
  
  This will break decoding with older versions of FFmpeg/Libav.
  

  • [DH] encoder/encoder.c

• Fix 8x8dct in lossless encoding

  5 juin 2017, par Anton Mitrofanov

  Fix 8x8dct in lossless encoding
  
  Change V and H intra prediction in lossless (TransformBypassModeFlag == 1)
  macroblocks to correctly adhere to the specification. Affects lossless
  encoding with 8x8dct or mix of lossless with normal macroblocks.
  
  8x8dct has already been disabled in lossless mode for some time due to
  being out-of-spec but this will allow us to re-enable it again.
  

  • [DH] encoder/macroblock.c

• x86 : AVX-512 add8x8_idct

  1er juin 2017, par Henrik Gramner

  x86: AVX-512 add8x8_idct
  

  • [DH] common/dct.c
  • [DH] common/x86/dct-a.asm
  • [DH] common/x86/dct.h

• Fix CABAC+8x8dct in 4:4:4

  30 mai 2017, par Anton Mitrofanov

  Fix CABAC+8x8dct in 4:4:4
  
  Use the correct ctxIdxInc calculation for coded_block_flag.
  

  • [DH] common/common.h
  • [DH] common/macroblock.c
  • [DH] encoder/cabac.c
  • [DH] encoder/encoder.c
  • [DH] x264.h

• aarch64 : Update the var2 functions to the new signature

  29 mai 2017, par Martin Storsjö

  aarch64: Update the var2 functions to the new signature
  
  The existing functions could easily be used by just calling them
  twice - this would give the following cycle numbers from checkasm:
  
  var2_8x8_c:      4110
  var2_8x8_neon:   1505
  var2_8x16_c:     8019
  var2_8x16_neon:  2545
  
  However, by merging both passes into the same function, we get the
  following speedup:
  var2_8x8_neon:   1205
  var2_8x16_neon:  2327
  

  • [DH] common/aarch64/pixel-a.S
  • [DH] common/aarch64/pixel.h
  • [DH] common/pixel.c