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

Les articles publiés sur le site

• x86 : Share the mbtree_propagate_list macro with aarch64

  3 septembre 2015, par Martin Storsjö

  x86: Share the mbtree_propagate_list macro with aarch64
  
  This avoids having to duplicate the same code for all architectures
  that implement only the internal part of this function in assembler.
  

  • [DH] common/aarch64/mc-c.c
  • [DH] common/mc.c
  • [DH] common/mc.h
  • [DH] common/x86/mc-c.c

• arm : Implement luma intra deblocking

  2 septembre 2015, par Martin Storsjö

  arm: Implement luma intra deblocking
  
  checkasm timing       Cortex-A7      A8     A9
  deblock_luma_intra[0]_c      5988    4653   4316
  deblock_luma_intra[0]_neon   3103    2170   2128
  deblock_luma_intra[1]_c      7119    5905   5347
  deblock_luma_intra[1]_neon   2068    1381   1412
  
  This includes extra optimizations by Janne Grunau.
  
  Timings from a separate build, on Exynos 5422:
  
                        Cortex-A7     A15
  deblock_luma_intra[0]_c      6627   3300
  deblock_luma_intra[0]_neon   3059   1128
  deblock_luma_intra[1]_c      7314   4128
  deblock_luma_intra[1]_neon   2038   720
  

  • [DH] common/arm/deblock-a.S
  • [DH] common/deblock.c

• arm : Implement some neon 8x16c intra predict functions

  31 août 2015, par Martin Storsjö

  arm: Implement some neon 8x16c intra predict functions
  
  checkasm timing       Cortex-A7      A8     A9
  intra_predict_8x16c_dct_c    862     540    590
  intra_predict_8x16c_dct_neon 608     511    657
  intra_predict_8x16c_h_c      972     707    719
  intra_predict_8x16c_h_neon   722     656    672
  intra_predict_8x16c_p_c      10183   9819   8655
  intra_predict_8x16c_p_neon   2622    1972   1983
  

  • [DH] common/arm/predict-a.S
  • [DH] common/arm/predict-c.c
  • [DH] common/arm/predict.h
  • [DH] common/predict.c

• checkasm : arm : Check register clobbering

  28 août 2015, par Martin Storsjö

  checkasm: arm: Check register clobbering
  
  Cast the function pointer to a different type signature, to
  be able to use uint64_t as return type (instead of intptr_t) for
  those calls that require it.
  
  Use two separate functions, depending on whether neon is available.
  

  • [DH] Makefile
  • [DH] tools/checkasm-arm.S
  • [DH] tools/checkasm.c

• arm : Implement x264_plane_copy_neon

  27 août 2015, par Martin Storsjö

  arm: Implement x264_plane_copy_neon
  
  checkasm timing       Cortex-A7      A8     A9
  plane_copy_c                 13124   10925  9106
  plane_copy_neon              7349    5103   8945
  

  • [DH] common/arm/mc-a.S
  • [DH] common/arm/mc-c.c