1. 24 Jan, 2018 1 commit
  2. 23 Jan, 2018 1 commit
  3. 14 Dec, 2017 1 commit
  4. 05 Sep, 2017 1 commit
  5. 08 Mar, 2017 1 commit
  6. 03 Mar, 2017 1 commit
  7. 23 Feb, 2017 1 commit
    • Manojkumar Bhosale's avatar
      Add MSA optimized Interpolate/MergeUV/Misc functions · 45b176d1
      Manojkumar Bhosale authored
      BUG=libyuv:634
      
      Change-Id: If8d60bd57f01fe95bc2fd26196466574195cc126
      
      Performance Gain (vs C auto-vectorized)
      InterpolateRow_MSA      - ~3.3x
      InterpolateRow_Any_MSA  - ~2.5x
      ARGBSetRow_MSA          - ~1.0x
      ARGBSetRow_Any_MSA      - ~1.0x
      ARGBToRGB24Row_MSA      - ~1.9x
      ARGBToRGB24Row_Any_MSA  - ~1.6x
      MergeUVRow_MSA          - ~1.6x
      MergeUVRow_Any_MSA      - ~1.2x
      
      Performance Gain (vs C non-vectorized)
      InterpolateRow_MSA      - ~11.3x
      InterpolateRow_Any_MSA  - ~ 7.9x
      ARGBSetRow_MSA          - ~ 6.2x
      ARGBSetRow_Any_MSA      - ~ 4.0x
      ARGBToRGB24Row_MSA      - ~ 9.9x
      ARGBToRGB24Row_Any_MSA  - ~ 8.4x
      MergeUVRow_MSA          - ~12.7x
      MergeUVRow_Any_MSA      - ~ 8.0x
      
      Change-Id: If8d60bd57f01fe95bc2fd26196466574195cc126
      Reviewed-on: https://chromium-review.googlesource.com/445817Reviewed-by: 's avatarFrank Barchard <fbarchard@google.com>
      Commit-Queue: Frank Barchard <fbarchard@google.com>
      45b176d1
  8. 15 Feb, 2017 1 commit
  9. 08 Nov, 2016 1 commit
  10. 15 Sep, 2016 1 commit
  11. 24 Jun, 2016 2 commits
  12. 05 Feb, 2016 1 commit
  13. 06 Jan, 2016 1 commit
  14. 17 Dec, 2015 1 commit
  15. 09 Jun, 2015 1 commit
    • fbarchard@google.com's avatar
      Box filter for YUV use rows with accumulation buffer for better memory behavior.… · 05416e2d
      fbarchard@google.com authored
      Box filter for YUV use rows with accumulation buffer for better memory behavior.  The old code would do columns accumulated into registers, and then store the result once.  This was slow from a memory point of view.  The new code does a row of source at a time, updating an accumulation buffer every row.  The accumulation buffer is small, and should fit cache.  Before each accumulation of N rows, the buffer needs to be reset to zero.  If the memset is a bottleneck, it would be faster to do the first row without an add, storing to the accumulation buffer, and then add for the remaining rows.
      BUG=425
      TESTED=out\release\libyuv_unittest --gtest_filter=*ScaleTo1x1*
      R=harryjin@google.com
      
      Review URL: https://webrtc-codereview.appspot.com/52659004
      
      git-svn-id: http://libyuv.googlecode.com/svn/trunk@1428 16f28f9a-4ce2-e073-06de-1de4eb20be90
      05416e2d
  16. 22 Apr, 2015 1 commit
  17. 14 Apr, 2015 1 commit
  18. 07 Feb, 2015 1 commit
  19. 24 Oct, 2014 1 commit
  20. 01 Oct, 2014 1 commit
  21. 20 May, 2014 1 commit
  22. 10 Jan, 2014 1 commit
  23. 09 Jan, 2014 1 commit
  24. 07 Jan, 2014 3 commits
  25. 03 Jan, 2014 6 commits
  26. 02 Jan, 2014 1 commit
  27. 09 Dec, 2013 1 commit
  28. 08 Dec, 2013 1 commit
  29. 07 Dec, 2013 1 commit
  30. 06 Dec, 2013 2 commits
  31. 05 Dec, 2013 1 commit