Re: [PATCH v3 4/5] drm/xe/xe_drm_ras: Wire get-counter support into xe_drm_ras for CRI

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

On Tue, Apr 21, 2026 at 08:21:00PM +0530, Riana Tauro wrote:
> Hook CRI get-counter support to xe_drm_ras to allow userspace to query
> counters if supported. When userspace queries a drm_ras error counter,
> fetch the latest counter value from system controller.
> 
> Integrate this with xe_drm_ras.
> 
> Usage :
> 
> Query all error counter value using ynl
> 
> $ sudo ynl --family drm_ras --dump get-error-counter --json \
> '{"node-id":0}'
> [{'error-id': 1, 'error-name': 'core-compute', 'error-value': 0},
>  {'error-id': 2, 'error-name': 'soc-internal', 'error-value': 0},
>  {'error-id': 3, 'error-name': 'device-memory', 'error-value': 0},
>  {'error-id': 4, 'error-name': 'pcie', 'error-value': 0},
>  {'error-id': 5, 'error-name': 'fabric', 'error-value': 0}]
> 
> Query single error counter value using ynl
> 
> $ sudo ynl --family drm_ras  --do get-error-counter --json \
>   '{"node-id":1, "error-id":1}'

Nit: Align spacing with above command.

> {'error-id': 1, 'error-name': 'core-compute', 'error-value': 2}
> 
> Signed-off-by: Riana Tauro <riana.tauro@intel.com>
> ---
> v2: split patches (Raag)
> ---
>  drivers/gpu/drm/xe/xe_drm_ras.c | 19 ++++++++++++-------
>  1 file changed, 12 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_drm_ras.c b/drivers/gpu/drm/xe/xe_drm_ras.c
> index e07dc23a155e..7040ff0f9539 100644
> --- a/drivers/gpu/drm/xe/xe_drm_ras.c
> +++ b/drivers/gpu/drm/xe/xe_drm_ras.c
> @@ -11,19 +11,28 @@
>  
>  #include "xe_device_types.h"
>  #include "xe_drm_ras.h"
> +#include "xe_ras.h"
>  
>  static const char * const error_components[] = DRM_XE_RAS_ERROR_COMPONENT_NAMES;
>  static const char * const error_severity[] = DRM_XE_RAS_ERROR_SEVERITY_NAMES;
>  
> -static int hw_query_error_counter(struct xe_drm_ras_counter *info,
> +static int hw_query_error_counter(struct xe_device *xe,
> +				  const enum drm_xe_ras_error_severity severity,
>  				  u32 error_id, const char **name, u32 *val)
>  {
> +	struct xe_drm_ras *ras = &xe->ras;
> +	struct xe_drm_ras_counter *info = ras->info[severity];

Use int ret = 0 ...

>  	if (!info || !info[error_id].name)
>  		return -ENOENT;
>  
>  	*name = info[error_id].name;
>  	*val = atomic_read(&info[error_id].counter);

Move this as an else case below and ...

> +	/* Fetch counter from system controller if supported */
> +	if (xe->info.has_sysctrl)
> +		return xe_ras_get_error_counter(xe, severity, error_id, val);

ret = get_counter();

>  	return 0;

return ret;

Raag

>  }