In case of catastrophic errors GuC sends notification, which results in
cryptic message. Let's add handler which, for starters, dumps state
of affected engine.
Signed-off-by: Andrzej Hajda <andrzej.hajda@xxxxxxxxx>
---
.../gpu/drm/i915/gt/uc/abi/guc_actions_abi.h | 1 +
drivers/gpu/drm/i915/gt/uc/intel_guc.h | 2 ++
drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 3 ++
.../gpu/drm/i915/gt/uc/intel_guc_submission.c | 31 +++++++++++++++++++
4 files changed, 37 insertions(+)
diff --git a/drivers/gpu/drm/i915/gt/uc/abi/guc_actions_abi.h b/drivers/gpu/drm/i915/gt/uc/abi/guc_actions_abi.h
index f359bef046e0b2..f9a1c5642855e3 100644
--- a/drivers/gpu/drm/i915/gt/uc/abi/guc_actions_abi.h
+++ b/drivers/gpu/drm/i915/gt/uc/abi/guc_actions_abi.h
@@ -138,6 +138,7 @@ enum intel_guc_action {
INTEL_GUC_ACTION_REGISTER_CONTEXT_MULTI_LRC = 0x4601,
INTEL_GUC_ACTION_CLIENT_SOFT_RESET = 0x5507,
INTEL_GUC_ACTION_SET_ENG_UTIL_BUFF = 0x550A,
+ INTEL_GUC_ACTION_NOTIFY_MEMORY_CAT_ERROR = 0x6000,
INTEL_GUC_ACTION_STATE_CAPTURE_NOTIFICATION = 0x8002,
INTEL_GUC_ACTION_NOTIFY_FLUSH_LOG_BUFFER_TO_FILE = 0x8003,
INTEL_GUC_ACTION_NOTIFY_CRASH_DUMP_POSTED = 0x8004,
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.h b/drivers/gpu/drm/i915/gt/uc/intel_guc.h
index 804133df1ac9b4..61b412732d095a 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.h
@@ -445,6 +445,8 @@ int intel_guc_engine_failure_process_msg(struct intel_guc *guc,
const u32 *msg, u32 len);
int intel_guc_error_capture_process_msg(struct intel_guc *guc,
const u32 *msg, u32 len);
+int intel_guc_cat_error_process_msg(struct intel_guc *guc,
+ const u32 *msg, u32 len);
struct intel_engine_cs *
intel_guc_lookup_engine(struct intel_guc *guc, u8 guc_class, u8 instance);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
index 2b22065e87bf9a..f55f724e264407 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
@@ -1035,6 +1035,9 @@ static int ct_process_request(struct intel_guc_ct *ct, struct ct_incoming_msg *r
CT_ERROR(ct, "Received GuC exception notification!\n");
ret = 0;
break;
+ case INTEL_GUC_ACTION_NOTIFY_MEMORY_CAT_ERROR:
+ ret = intel_guc_cat_error_process_msg(guc, payload, len);
+ break;
default:
ret = -EOPNOTSUPP;
break;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
index 693b07a977893d..94f91dfa3ec456 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
@@ -4659,6 +4659,37 @@ int intel_guc_engine_failure_process_msg(struct intel_guc *guc,
return 0;
}
+int intel_guc_cat_error_process_msg(struct intel_guc *guc,
+ const u32 *msg, u32 len)
+{
+ struct drm_i915_private *i915 = guc_to_gt(guc)->i915;
+ struct intel_engine_cs *engine;
+ struct intel_context *ce;
+ struct drm_printer p;
+ unsigned long flags;
+ int ctx_id;
+
+ if (unlikely(len != 1)) {
+ drm_dbg(&i915->drm, "Invalid length %u", len);
+ return -EPROTO;
+ }
+ ctx_id = msg[0];
+
+ xa_lock_irqsave(&guc->context_lookup, flags);
+ ce = g2h_context_lookup(guc, ctx_id);
+ if (ce)
+ engine = ce->engine;
+ xa_unlock_irqrestore(&guc->context_lookup, flags);
+ if (unlikely(!ce || !engine))
+ return -EPROTO;
+
+ drm_err(&i915->drm, "%s: CAT error reported by GuC\n", engine->name);
+ p = drm_info_printer(i915->drm.dev);
+ intel_engine_dump(engine, &p, "%s\n", engine->name);
+
+ return 0;
+}
+
void intel_guc_find_hung_context(struct intel_engine_cs *engine)
{
struct intel_guc *guc = &engine->gt->uc.guc;
--
2.34.1
