[PATCH 1/5] drm/amdgpu: Add log info for umc_v12_0 and smu_v13_0_6

YiPeng Chai YiPeng.Chai at amd.com
Tue Jan 16 08:20:30 UTC 2024


Add log info for umc_v12_0 and smu_v13_0_6.

Signed-off-by: YiPeng Chai <YiPeng.Chai at amd.com>
---
 drivers/gpu/drm/amd/amdgpu/umc_v12_0.c              | 11 +++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_events.c             |  6 +++++-
 .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c    | 13 +++++++++++++
 3 files changed, 29 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
index 6423dca5b777..fa2168f1d3bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
@@ -91,6 +91,17 @@ static void umc_v12_0_reset_error_count(struct amdgpu_device *adev)
 
 bool umc_v12_0_is_deferred_error(struct amdgpu_device *adev, uint64_t mc_umc_status)
 {
+	dev_info(adev->dev,
+		"MCA_UMC_STATUS(0x%llx): Val:%llu, Poison:%llu, Deferred:%llu, PCC:%llu, UC:%llu, TCC:%llu\n",
+		mc_umc_status,
+		REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val),
+		REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Poison),
+		REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred),
+		REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, PCC),
+		REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UC),
+		REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, TCC)
+	);
+
 	return (amdgpu_ras_is_poison_mode_supported(adev) &&
 		(REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) &&
 		(REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1));
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index 11923964ce9a..51bb98db5d7a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -1297,8 +1297,10 @@ void kfd_signal_poison_consumed_event(struct kfd_node *dev, u32 pasid)
 	uint32_t id = KFD_FIRST_NONSIGNAL_EVENT_ID;
 	int user_gpu_id;
 
-	if (!p)
+	if (!p) {
+		dev_warn(dev->adev->dev, "Not find process with pasid:%d\n", pasid);
 		return; /* Presumably process exited. */
+	}
 
 	user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id);
 	if (unlikely(user_gpu_id == -EINVAL)) {
@@ -1334,6 +1336,8 @@ void kfd_signal_poison_consumed_event(struct kfd_node *dev, u32 pasid)
 		}
 	}
 
+	dev_warn(dev->adev->dev, "Send SIGBUS to process %s(pasid:%d)\n",
+		p->lead_thread->comm, pasid);
 	rcu_read_unlock();
 
 	/* user application will handle SIGBUS signal */
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 952a983da49a..cee8ee5afcb6 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -2406,10 +2406,23 @@ static int smu_v13_0_6_get_valid_mca_count(struct smu_context *smu, enum amdgpu_
 
 	ret = smu_cmn_send_smc_msg(smu, msg, count);
 	if (ret) {
+		dev_err(smu->adev->dev, "%s(%d) failed to query %s MCA count, ret:%d\n",
+			(msg == SMU_MSG_QueryValidMcaCeCount) ?
+				"SMU_MSG_QueryValidMcaCeCount" : "SMU_MSG_QueryValidMcaCount",
+			msg,
+			(msg == SMU_MSG_QueryValidMcaCeCount) ? "CE" : "UE",
+			ret);
 		*count = 0;
 		return ret;
 	}
 
+	dev_info(smu->adev->dev, "MSG %s(%d) query %s MCA count result:%u\n",
+		(msg == SMU_MSG_QueryValidMcaCeCount) ?
+			"SMU_MSG_QueryValidMcaCeCount" : "SMU_MSG_QueryValidMcaCount",
+		msg,
+		(msg == SMU_MSG_QueryValidMcaCeCount) ? "CE" : "UE",
+		*count);
+
 	return 0;
 }
 
-- 
2.34.1



More information about the amd-gfx mailing list