[PATCH v2 0/3] Improve anti-pre-emption w/a for compute workloads

John.C.Harrison at Intel.com John.C.Harrison at Intel.com
Fri Feb 25 20:41:48 UTC 2022


From: John Harrison <John.C.Harrison at Intel.com>

Compute workloads are inherently not pre-emptible on current hardware.
Thus the pre-emption timeout was disabled as a workaround to prevent
unwanted resets. Instead, the hang detection was left to the heartbeat
and its (longer) timeout. This is undesirable with GuC submission as
the heartbeat is a full GT reset rather than a per engine reset and so
is much more destructive. Instead, just bump the pre-emption timeout
to a big value. Also, update the heartbeat to allow such a long
pre-emption delay in the final heartbeat period.

v2: Add clamping helpers.

Signed-off-by: John Harrison <John.C.Harrison at Intel.com>


John Harrison (3):
  drm/i915/guc: Limit scheduling properties to avoid overflow
  drm/i915/gt: Make the heartbeat play nice with long pre-emption
    timeouts
  drm/i915: Improve long running OCL w/a for GuC submission

 drivers/gpu/drm/i915/gt/intel_engine.h        |  6 ++
 drivers/gpu/drm/i915/gt/intel_engine_cs.c     | 91 ++++++++++++++++++-
 .../gpu/drm/i915/gt/intel_engine_heartbeat.c  | 16 ++++
 drivers/gpu/drm/i915/gt/sysfs_engines.c       | 25 +++--
 drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h   |  9 ++
 5 files changed, 134 insertions(+), 13 deletions(-)

-- 
2.25.1



More information about the dri-devel mailing list