1. 13 Apr, 2016 1 commit
  2. 12 Apr, 2016 2 commits
  3. 22 Feb, 2016 1 commit
  4. 19 Feb, 2016 1 commit
  5. 18 Feb, 2016 1 commit
  6. 14 Feb, 2016 2 commits
  7. 31 Jan, 2016 3 commits
  8. 23 Jan, 2016 1 commit
  9. 21 Jan, 2016 2 commits
  10. 07 Jan, 2016 2 commits
  11. 29 Dec, 2015 1 commit
    • Janne Grunau's avatar
      checkasm: x86: post commit review fixes · f0f54117
      Janne Grunau authored
      Check the full FPU tag word instead of only the lower half and simplify
      the comparison.
      Use upper-case function base name as macro name to instantiate both
      checked_call variants.
      f0f54117
  12. 24 Dec, 2015 1 commit
  13. 21 Dec, 2015 5 commits
  14. 14 Dec, 2015 1 commit
    • Janne Grunau's avatar
      arm: add a cpu flag for the VFPv2 vector mode · e2710e79
      Janne Grunau authored
      The vector mode was deprecated in ARMv7-A/VFPv3 and various cpu
      implementations do not support it in hardware. Vector mode code will
      depending the OS either be emulated in software or result in an illegal
      instruction on cpus which does not support it. This was not really
      problem in practice since NEON implementations of the same functions are
      preferred. It will however become a problem for checkasm which tests
      every cpu flag separately.
      
      Since this is a cpu feature newer cpu do not support anymore the
      behaviour of this flag differs from the other flags. It can be only
      activated by runtime cpu feature selection.
      e2710e79
  15. 05 Dec, 2015 1 commit
  16. 08 Nov, 2015 1 commit
  17. 28 Oct, 2015 1 commit
  18. 13 Oct, 2015 1 commit
    • Ronald S. Bultje's avatar
      vp9: add itxfm_add eob shortcuts to 10/12bpp functions. · eb4b5ff7
      Ronald S. Bultje authored
      These aren't quite as helpful as the ones in 8bpp, since over there,
      we can use pmulhrsw, but here the coefficients have too many bits to
      be able to take advantage of pmulhrsw. However, we can still skip
      cols for which all coefs are 0, and instead just zero the input data
      for the row itx. This helps a few % on overall decoding speed.
      eb4b5ff7
  19. 06 Oct, 2015 1 commit
  20. 04 Oct, 2015 1 commit
  21. 03 Oct, 2015 2 commits
  22. 28 Sep, 2015 5 commits
  23. 27 Sep, 2015 1 commit
    • Henrik Gramner's avatar
      checkasm/x86: Correctly handle variadic functions · 7ca1de5b
      Henrik Gramner authored
      The System V ABI on x86-64 specifies that the al register contains an upper
      bound of the number of arguments passed in vector registers when calling
      variadic functions, so we aren't allowed to clobber it.
      
      checkasm_fail_func() is a variadic function so also zero al before calling it.
      7ca1de5b
  24. 26 Sep, 2015 2 commits