[PATCH 0/3] PF: migration-friendly auto-provisioning

[Michal Wajdeczko <michal.wajdeczko@intel.com>]

Instead of trying very hard to find the largest fair number of GuC
IDs that could be allocated for VFs on the current GT, pick some
smaller rounded down to power-of-two value that is more likely
to be provisioned in the same manner by the other PF instance.

 num VFs | num contexts | num doorbells
 --------+--------------+--------------
  63..32 | 1024         | 4
  31..16 | 2048         | 8
  15..8  | 4096         | 16
   7..4  | 8192         | 32
   3..2  | 16384        | 64
      1  | 32768        | 128  (regular PF)
      1  | 64512        | 240  (admin only PF)

Only if the above values can't be satisfied, fallback to estimated
largest available value.

Michal Wajdeczko (3):
  drm/xe/pf: Use migration-friendly context IDs auto-provisioning
  drm/xe/pf: Use migration-friendly doorbells auto-provisioning
  drm/xe/tests: Add KUnit tests for PF fair provisioning

 .../xe/tests/xe_gt_sriov_pf_config_kunit.c    | 162 ++++++++++++++++++
 drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c    |  37 ++++
 drivers/gpu/drm/xe/xe_sriov_pf_helpers.h      |  11 ++
 3 files changed, 210 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/tests/xe_gt_sriov_pf_config_kunit.c

-- 
2.47.1