Re: [PATCH v1] drm/xe/ras: Fix sysctrl and drm-ras discrepancy

["Tauro, Riana" <riana.tauro@intel.com>]

On 08-09-2026 13:34, Raag Jadav wrote:
> There are cases where a reported error is valid for sysctrl but it's not
> exposed as a valid uapi through drm-ras. We allow such error to be logged
> but it's redundant to try to send an error-event for it. Early return such
> cases and prevent misleading drm-ras error-event failures.
>
> Fixes: 6a9493e47075 ("drm/xe/xe_ras: Report correctable/uncorrectable error events to userspace")
> Signed-off-by: Raag Jadav <raag.jadav@intel.com>
> ---
>   drivers/gpu/drm/xe/xe_ras.c | 10 +++++++---
>   1 file changed, 7 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_ras.c b/drivers/gpu/drm/xe/xe_ras.c
> index 7a85735c57d5..522d891da4a0 100644
> --- a/drivers/gpu/drm/xe/xe_ras.c
> +++ b/drivers/gpu/drm/xe/xe_ras.c
> @@ -288,6 +288,13 @@ static void ras_send_error_event(struct xe_device *xe, u8 severity, u8 component
>   	u32 value;
>   	int ret;
>   
> +	drm_severity = xe_to_drm_ras_severity(severity);
> +	drm_component = xe_to_drm_ras_component(component);
> +
> +	/* Not all sysctrl reported errors are valid uapi */
> +	if (drm_severity == DRM_XE_RAS_ERR_SEV_MAX || drm_component == DRM_XE_RAS_ERR_COMP_MAX)
> +		return;

We should add this check in xe_drm_ras_event instead.

Thanks
Riana

> +
>   	counter.common.severity = severity;
>   	counter.common.component = component;
>   
> @@ -295,9 +302,6 @@ static void ras_send_error_event(struct xe_device *xe, u8 severity, u8 component
>   	if (ret)
>   		return;
>   
> -	drm_severity = xe_to_drm_ras_severity(severity);
> -	drm_component = xe_to_drm_ras_component(component);
> -
>   	xe_drm_ras_event(xe, drm_component, drm_severity, value);
>   }
>