1. 01 Mar, 2018 1 commit
    • Chris Wilson's avatar
      lib: Provide an accelerated routine for readback from WC · 6a06d014
      Chris Wilson authored
      Reading from WC is awfully slow as each access is uncached and so
      performed synchronously, stalling for the memory load. x86 did introduce
      some new instructions in SSE 4.1 to provide a small internal buffer to
      accelerate reading back a cacheline at a time from uncached memory, for
      this purpose.
      
      v2: Don't be lazy and handle misalignment.
      v3: Switch out of sse41 before emitting the generic memcpy routine
      v4: Replace opencoded memcpy_from_wc
      v5: Always flush the internal buffer before use (Eric)
      v6: Assume bulk moves, so check for dst alignment.
      v7: Use _mm_fence for _buitlin_ia32_mfence for consistency, remove
      superfluous defines (Ville)
      Signed-off-by: Chris Wilson's avatarChris Wilson <chris@chris-wilson.co.uk>
      Cc: Eric Anholt <eric@anholt.net>
      Reviewed-by: Ville Syrjälä's avatarVille Syrjälä <ville.syrjala@linux.intel.com>
      6a06d014
  2. 06 Nov, 2017 1 commit
  3. 03 Oct, 2017 1 commit
  4. 11 Jul, 2016 1 commit
  5. 25 Feb, 2015 1 commit
  6. 29 Oct, 2014 1 commit
  7. 01 Nov, 2013 1 commit
  8. 13 Sep, 2013 1 commit
  9. 20 Aug, 2013 1 commit
  10. 17 Aug, 2013 1 commit