Re: [RFC PATCH 3/3] drm/xe: Trigger queue cleanup if not in wedged mode 2

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

On Fri, Dec 12, 2025 at 03:34:44PM -0800, Matthew Brost wrote:
> The intent of wedging a device is to allow queues to continue running
> only in wedged mode 2. In other modes, queues should initiate cleanup
> and signal all remaining fences. Fix xe_guc_submit_wedge to correctly
> clean up queues when wedge mode != 2.
> 
> Fixes: 7dbe8af13c18 ("drm/xe: Wedge the entire device")
> Cc: stable@vger.kernel.org
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_guc_submit.c | 31 ++++++++++++++++++------------
>  1 file changed, 19 insertions(+), 12 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> index 857375be9a84..1eef93d474f0 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> @@ -1277,6 +1277,7 @@ static void disable_scheduling_deregister(struct xe_guc *guc,
>   */
>  void xe_guc_submit_wedge(struct xe_guc *guc)
>  {
> +	struct xe_device *xe = guc_to_xe(guc);
>  	struct xe_gt *gt = guc_to_gt(guc);
>  	struct xe_exec_queue *q;
>  	unsigned long index;
> @@ -1291,19 +1292,25 @@ void xe_guc_submit_wedge(struct xe_guc *guc)
>  	if (!guc->submission_state.initialized)
>  		return;
>  
> -	err = devm_add_action_or_reset(guc_to_xe(guc)->drm.dev,
> -				       guc_submit_wedged_fini, guc);
> -	if (err) {
> -		xe_gt_err(gt, "Failed to register clean-up on wedged.mode=2; "
> -			  "Although device is wedged.\n");
> -		return;
> -	}
> +	if (xe->wedged.mode == 2) {
> +		err = devm_add_action_or_reset(guc_to_xe(guc)->drm.dev,
> +					       guc_submit_wedged_fini, guc);
> +		if (err) {
> +			xe_gt_err(gt, "Failed to register clean-up on wedged.mode=2; "
> +				  "Although device is wedged.\n");
> +			return;
> +		}
>  
> -	mutex_lock(&guc->submission_state.lock);
> -	xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
> -		if (xe_exec_queue_get_unless_zero(q))
> -			set_exec_queue_wedged(q);
> -	mutex_unlock(&guc->submission_state.lock);
> +		mutex_lock(&guc->submission_state.lock);
> +		xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
> +			if (xe_exec_queue_get_unless_zero(q))
> +				set_exec_queue_wedged(q);
> +		mutex_unlock(&guc->submission_state.lock);
> +	} else {
> +		/* Forcefully kill any remaining exec queues, signal fences */
> +		xe_guc_submit_stop(guc);
> +		xe_guc_submit_pause_abort(guc);

This is basically the prerequisite[1] as we decided at the time, but
perhaps I failed to notice it. I'm wondering if we should also redirect
page faults to dummy page as per prerequisites section?

[1] https://lore.kernel.org/dri-devel/20250204070528.1919158-2-raag.jadav@intel.com/

Raag