[PATCH 0/4] GuC: Improve handling of the malformed G2H

Michal Wajdeczko michal.wajdeczko at intel.com
Tue Nov 5 17:30:28 UTC 2024


Inspired by this log found during dirty experiments with VF:

 [ ] xe 0000:00:02.0: [drm] *ERROR* Invalid guc_id 4294967295
 [ ] DMAR: [DMA Write NO_PASID] Request device [00:02.0] fault addr 0x0 [fault reason 0x02] Present bit in context entry is clear
 [ ] DMAR: DRHD: handling fault status reg 2
 [ ] xe 0000:00:02.0: [drm] *ERROR* GT0: G2H action 0x6000 failed (-EPROTO)
 [ ] xe 0000:00:02.0: [drm] GT0: Capture 1.1: CTB is dead - reason=0x202

Cc: Matthew Brost <matthew.brost at intel.com>

Michal Wajdeczko (4):
  drm/xe/guc: Log content of the failed G2H message
  drm/xe/guc: Drop redundant logs about invalid G2H length
  drm/xe/guc: Don't read data from G2H prior to length check
  drm/xe/guc: Don't treat GuC generic CAT error as protocol error

 drivers/gpu/drm/xe/xe_guc_ct.c     |  4 +--
 drivers/gpu/drm/xe/xe_guc_submit.c | 53 +++++++++++++++---------------
 2 files changed, 29 insertions(+), 28 deletions(-)

-- 
2.43.0



More information about the Intel-xe mailing list