[0/7] Allow privileged user to map the OA buffer

Message

Umesh Nerlige Ramappa Nov. 13, 2020, 6:03 p.m. UTC

Allow user to map the OA buffer and also trigger reports into it.

CI fixes:
v1: Fixes a memory corruption due to addition of OA whitelist on demand.
v2: Spinlock when applying whitelist
v3: Use uncore->lock. Do not check for wal->count when applying whitelist.
v4: Refresh and rerun with newly added test (forked access).
v5:
- Split patches into smaller units
- Grow the wal->list only for the engine that needs it.
- Bring back the wal->count check when applying whitelist during resume
v6: Fix checkpatch and sparse warnings
v7:
- Fix mem leak
- Grow wal->list only if wal->count is not aligned to WAL_LIST_CHUNK

Signed-off-by: Umesh Nerlige Ramappa <[e-mail]>
Test-with: [e-mail]

Piotr Maciejewski (1):
  drm/i915/perf: Ensure observation logic is not clock gated

Umesh Nerlige Ramappa (6):
  drm/i915/gt: Lock intel_engine_apply_whitelist with uncore->lock
  drm/i915/gt: Add a reference to the engine in i915_wa_list
  drm/i915/perf: Whitelist OA report trigger registers
  drm/i915/gt: Refactor _wa_add to reuse wa_index and wa_list_grow
  drm/i915/perf: Whitelist OA counter and buffer registers
  drm/i915/perf: Map OA buffer to user space for gen12 performance query

 drivers/gpu/drm/i915/gem/i915_gem_mman.c      |   2 +-
 drivers/gpu/drm/i915/gem/i915_gem_mman.h      |   2 +
 drivers/gpu/drm/i915/gt/intel_workarounds.c   | 244 ++++++++++++++----
 drivers/gpu/drm/i915/gt/intel_workarounds.h   |   7 +
 .../gpu/drm/i915/gt/intel_workarounds_types.h |   6 +
 .../gpu/drm/i915/gt/selftest_workarounds.c    |   4 +-
 drivers/gpu/drm/i915/i915_perf.c              | 228 +++++++++++++++-
 drivers/gpu/drm/i915/i915_perf_types.h        |   8 +
 drivers/gpu/drm/i915/i915_reg.h               |  10 +
 include/uapi/drm/i915_drm.h                   |  33 +++
 10 files changed, 485 insertions(+), 59 deletions(-)