Re: [PATCH v5 09/13] drm/xe/xe_ras: Report soc internal errors using SIGID

[Raag Jadav <raag.jadav@intel.com>]

On Fri, Oct 23, 2026 at 10:04:31AM +0530, Badal Nilawar wrote:
> Use xe_log_* helpers to report soc internal errors.
> 
> Example:
> 	xe 0000:03:00.0: [drm] *ERROR* SIGID=205 FATAL
> (020600700000000000000000) [Hardware Error]: SOC_INTERNAL:
> PUNIT error detected: 0x00000002
> 
> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
> Cc: Riana Tauro <riana.tauro@intel.com>
> ---
> v5:
>   - Log unknown soc internal errors
> ---
>  drivers/gpu/drm/xe/xe_ras.c | 16 +++++++++-------
>  1 file changed, 9 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_ras.c b/drivers/gpu/drm/xe/xe_ras.c
> index 1f94dd1fb91b..4408fc530900 100644
> --- a/drivers/gpu/drm/xe/xe_ras.c
> +++ b/drivers/gpu/drm/xe/xe_ras.c
> @@ -608,7 +608,6 @@ static u8 handle_soc_internal_errors(struct xe_device *xe, struct xe_ras_error_a
>  {
>  	struct xe_ras_soc_error *info = (void *)arr->details;
>  	struct xe_ras_soc_error_source *source = &info->source;
> -	struct xe_ras_error_class *counter = &arr->counter;
>  
>  	if (source->csc) {
>  		struct xe_ras_csc_error *csc_error = (void *)info->details;
> @@ -622,9 +621,8 @@ static u8 handle_soc_internal_errors(struct xe_device *xe, struct xe_ras_error_a
>  		 * is required.
>  		 */
>  		if (csc_error->hec_fw_error) {
> -			xe_err(xe, "[RAS]: CSC %s detected: 0x%x\n",
> -			       sev_to_str(counter->common.severity),
> -			       csc_error->hec_fw_error);
> +			xe_log_comp_fatal(xe, SOC_INTERNAL, &arr->counter, sizeof(arr->counter),
> +					  "CSC error detected: 0x%08x\n", csc_error->hec_fw_error);
>  			xe_survivability_mode_runtime_enable(xe);
>  			return XE_RAS_RECOVERY_ACTION_DISCONNECT;
>  		}
> @@ -632,15 +630,19 @@ static u8 handle_soc_internal_errors(struct xe_device *xe, struct xe_ras_error_a
>  		struct xe_ras_ieh_error *ieh_error = (void *)info->details;
>  
>  		if (ieh_error->global_error_status & XE_RAS_SOC_IEH_PUNIT) {
> -			xe_err(xe, "[RAS]: PUNIT %s detected: 0x%x\n",
> -			       sev_to_str(counter->common.severity),
> -			       ieh_error->global_error_status);
> +			xe_log_comp_fatal(xe, SOC_INTERNAL, &arr->counter, sizeof(arr->counter),
> +					  "PUNIT error detected: 0x%08x\n",
> +					  ieh_error->global_error_status);
>  			punit_error_handler(xe);
>  			return XE_RAS_RECOVERY_ACTION_DISCONNECT;
>  		}
>  	}
>  
>  	/* For other SoC internal errors, request a reset as recovery mechanism */
> +	xe_log_comp(xe, ras_sev_to_cper_sev(arr->counter.common.severity), SOC_INTERNAL,
> +		    &arr->counter, sizeof(arr->counter),
> +		    "Unknown error detected (csc=%u ieh=%u)\n",

If the error is unknown, we may want to print the entire source using 0x%08x.

Raag

> +		    source->csc, source->ieh);
>  	return XE_RAS_RECOVERY_ACTION_RESET;
>  }
>  
> -- 
> 2.54.0
>