[Intel-gfx] [PATCH 0/4] Async flip optimization for DG2

Stanislav Lisovskiy stanislav.lisovskiy at intel.com
Fri Jan 21 08:06:11 UTC 2022


Limitting the WM levels to 0 for DG2 during async flips,
allows to slightly increase the performance, as recommended
by HW team.

Stanislav Lisovskiy (4):
  drm/i915: Pass plane to watermark calculation functions
  drm/i915: Introduce do_async_flip flag to intel_plane_state
  drm/i915: Use wm0 only during async flips for DG2
  drm/i915: Don't allocate extra ddb during async flip for DG2

 .../gpu/drm/i915/display/intel_atomic_plane.c |  5 +-
 .../gpu/drm/i915/display/intel_atomic_plane.h |  3 ++
 drivers/gpu/drm/i915/display/intel_display.c  | 31 +++++++++++-
 .../drm/i915/display/intel_display_types.h    |  3 ++
 drivers/gpu/drm/i915/intel_pm.c               | 49 ++++++++++++++++---
 5 files changed, 80 insertions(+), 11 deletions(-)

-- 
2.24.1.485.gad05a3d8e5



More information about the Intel-gfx mailing list