1. 02 Jan, 2017 1 commit
  2. 29 Feb, 2016 1 commit
  3. 14 Feb, 2016 1 commit
  4. 02 Feb, 2016 1 commit
  5. 22 Oct, 2015 2 commits
  6. 05 Nov, 2012 2 commits
  7. 31 Oct, 2012 1 commit
  8. 30 Oct, 2012 2 commits
  9. 12 Aug, 2012 1 commit
  10. 29 Jun, 2012 1 commit
    • Mans Rullgard's avatar
      x86: vc1: fix and enable optimised loop filter · f2fd1678
      Mans Rullgard authored
      The problem is that the ssse3 psign instruction does the wrong
      thing here.  Commit ea60dfe2 incorrectly removed a macro emulating
      this instruction for pre-ssse3 code.  However, the emulation is
      incorrect, and the code relies on the behaviour of the macro.
      Specifically, the psign sets destination elements to zero where
      the corresponding source element is zero, whereas the emulation
      only negates destination elements where the source is negative.
      
      Furthermore, the PSIGNW_MMX macro in x86util.asm is totally bogus,
      which is why the original VC-1 code had an additional right shift
      when using it.  Since the psign instruction cannot be used here,
      skip all the macro hell and use the working instruction sequence
      directly.
      
      None of this was noticed due a stray return statement in
      ff_vc1dsp_init_mmx() which meant that only the mmx version of the
      loop filter was ever used (before being removed in ea60dfe2).
      Signed-off-by: 's avatarMans Rullgard <mans@mansr.com>
      f2fd1678
  11. 12 May, 2012 1 commit
  12. 10 Dec, 2011 1 commit
  13. 19 Oct, 2011 1 commit
  14. 15 Aug, 2011 1 commit
  15. 12 Aug, 2011 2 commits
  16. 14 May, 2011 1 commit
  17. 19 Mar, 2011 1 commit
  18. 05 Sep, 2010 1 commit
  19. 11 Jul, 2010 1 commit