Re: [PATCH] drm/amdgpu: replace VM fault error by info logs

[Felix Kuehling <felix.kuehling@xxxxxxx>]

On 2022-04-22 17:15, Alex Sierra wrote:
> This is not a kernel error. These logs are caused by VM faults that
> could not be handled. Typically, generated by user mode applications.
>
> Signed-off-by: Alex Sierra <alex.sierra@xxxxxxx>
> ---
>   drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 20 ++++++++++----------

Please update the other GMC versions as well to keep things consistent.

Regards,
  Felix


>   1 file changed, 10 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
> index 22761a3bb818..98c8de7307be 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
> @@ -582,14 +582,14 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
>   	memset(&task_info, 0, sizeof(struct amdgpu_task_info));
>   	amdgpu_vm_get_task_info(adev, entry->pasid, &task_info);
>   
> -	dev_err(adev->dev,
> +	dev_info(adev->dev,
>   		"[%s] %s page fault (src_id:%u ring:%u vmid:%u "
>   		"pasid:%u, for process %s pid %d thread %s pid %d)\n",
>   		hub_name, retry_fault ? "retry" : "no-retry",
>   		entry->src_id, entry->ring_id, entry->vmid,
>   		entry->pasid, task_info.process_name, task_info.tgid,
>   		task_info.task_name, task_info.pid);
> -	dev_err(adev->dev, "  in page starting at address 0x%016llx from IH client 0x%x (%s)\n",
> +	dev_info(adev->dev, "  in page starting at address 0x%016llx from IH client 0x%x (%s)\n",
>   		addr, entry->client_id,
>   		soc15_ih_clientid_name[entry->client_id]);
>   
> @@ -611,11 +611,11 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
>   	WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
>   
>   
> -	dev_err(adev->dev,
> +	dev_info(adev->dev,
>   		"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
>   		status);
>   	if (hub == &adev->vmhub[AMDGPU_GFXHUB_0]) {
> -		dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
> +		dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
>   			cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" :
>   			gfxhub_client_ids[cid],
>   			cid);
> @@ -648,22 +648,22 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
>   			mmhub_cid = NULL;
>   			break;
>   		}
> -		dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
> +		dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
>   			mmhub_cid ? mmhub_cid : "unknown", cid);
>   	}
> -	dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n",
> +	dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n",
>   		REG_GET_FIELD(status,
>   		VM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS));
> -	dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n",
> +	dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n",
>   		REG_GET_FIELD(status,
>   		VM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR));
> -	dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n",
> +	dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n",
>   		REG_GET_FIELD(status,
>   		VM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS));
> -	dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n",
> +	dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n",
>   		REG_GET_FIELD(status,
>   		VM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR));
> -	dev_err(adev->dev, "\t RW: 0x%x\n", rw);
> +	dev_info(adev->dev, "\t RW: 0x%x\n", rw);
>   	return 0;
>   }
>