Re: [PATCH 2/2] drm/xe/guc: Prefer GT oriented logs in submit code

[Matthew Brost <matthew.brost@intel.com>]

On Thu, Nov 07, 2024 at 08:47:41PM +0100, Michal Wajdeczko wrote:
> For better diagnostics, use xe_gt_err() instead of drm_err().
> 
> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
> Cc: Matthew Brost <matthew.brost@intel.com>

Reviewed-by: Matthew Brost <matthew.brost@intel.com>

> ---
>  drivers/gpu/drm/xe/xe_guc_submit.c | 25 ++++++++++++-------------
>  1 file changed, 12 insertions(+), 13 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> index 0311d9e50d06..c70b75a3eb85 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> @@ -763,7 +763,6 @@ static void disable_scheduling_deregister(struct xe_guc *guc,
>  					  struct xe_exec_queue *q)
>  {
>  	MAKE_SCHED_CONTEXT_ACTION(q, DISABLE);
> -	struct xe_device *xe = guc_to_xe(guc);
>  	int ret;
>  
>  	set_min_preemption_timeout(guc, q);
> @@ -773,7 +772,7 @@ static void disable_scheduling_deregister(struct xe_guc *guc,
>  	if (!ret) {
>  		struct xe_gpu_scheduler *sched = &q->guc->sched;
>  
> -		drm_warn(&xe->drm, "Pending enable failed to respond");
> +		xe_gt_warn(q->gt, "Pending enable failed to respond\n");
>  		xe_sched_submission_start(sched);
>  		xe_gt_reset_async(q->gt);
>  		xe_sched_tdr_queue_imm(sched);
> @@ -817,7 +816,7 @@ static void xe_guc_exec_queue_trigger_cleanup(struct xe_exec_queue *q)
>   */
>  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;
>  	int err;
> @@ -827,7 +826,8 @@ void xe_guc_submit_wedge(struct xe_guc *guc)
>  	err = devm_add_action_or_reset(guc_to_xe(guc)->drm.dev,
>  				       guc_submit_wedged_fini, guc);
>  	if (err) {
> -		drm_err(&xe->drm, "Failed to register xe_guc_submit clean-up on wedged.mode=2. Although device is wedged.\n");
> +		xe_gt_err(gt, "Failed to register clean-up on wedged.mode=2; "
> +			  "Although device is wedged.\n");
>  		return;
>  	}
>  
> @@ -859,7 +859,6 @@ static void xe_guc_exec_queue_lr_cleanup(struct work_struct *w)
>  		container_of(w, struct xe_guc_exec_queue, lr_tdr);
>  	struct xe_exec_queue *q = ge->q;
>  	struct xe_guc *guc = exec_queue_to_guc(q);
> -	struct xe_device *xe = guc_to_xe(guc);
>  	struct xe_gpu_scheduler *sched = &ge->sched;
>  	bool wedged;
>  
> @@ -897,7 +896,7 @@ static void xe_guc_exec_queue_lr_cleanup(struct work_struct *w)
>  					 !exec_queue_pending_disable(q) ||
>  					 xe_guc_read_stopped(guc), HZ * 5);
>  		if (!ret) {
> -			drm_warn(&xe->drm, "Schedule disable failed to respond");
> +			xe_gt_warn(q->gt, "Schedule disable failed to respond\n");
>  			xe_sched_submission_start(sched);
>  			xe_gt_reset_async(q->gt);
>  			return;
> @@ -1801,17 +1800,17 @@ int xe_guc_submit_start(struct xe_guc *guc)
>  static struct xe_exec_queue *
>  g2h_exec_queue_lookup(struct xe_guc *guc, u32 guc_id)
>  {
> -	struct xe_device *xe = guc_to_xe(guc);
> +	struct xe_gt *gt = guc_to_gt(guc);
>  	struct xe_exec_queue *q;
>  
>  	if (unlikely(guc_id >= GUC_ID_MAX)) {
> -		drm_err(&xe->drm, "Invalid guc_id %u", guc_id);
> +		xe_gt_err(gt, "Invalid guc_id %u\n", guc_id);
>  		return NULL;
>  	}
>  
>  	q = xa_load(&guc->submission_state.exec_queue_lookup, guc_id);
>  	if (unlikely(!q)) {
> -		drm_err(&xe->drm, "Not engine present for guc_id %u", guc_id);
> +		xe_gt_err(gt, "Not engine present for guc_id %u\n", guc_id);
>  		return NULL;
>  	}
>  
> @@ -2039,7 +2038,7 @@ int xe_guc_exec_queue_memory_cat_error_handler(struct xe_guc *guc, u32 *msg,
>  
>  int xe_guc_exec_queue_reset_failure_handler(struct xe_guc *guc, u32 *msg, u32 len)
>  {
> -	struct xe_device *xe = guc_to_xe(guc);
> +	struct xe_gt *gt = guc_to_gt(guc);
>  	u8 guc_class, instance;
>  	u32 reason;
>  
> @@ -2051,10 +2050,10 @@ int xe_guc_exec_queue_reset_failure_handler(struct xe_guc *guc, u32 *msg, u32 le
>  	reason = msg[2];
>  
>  	/* Unexpected failure of a hardware feature, log an actual error */
> -	drm_err(&xe->drm, "GuC engine reset request failed on %d:%d because 0x%08X",
> -		guc_class, instance, reason);
> +	xe_gt_err(gt, "GuC engine reset request failed on %d:%d because 0x%08X",
> +		  guc_class, instance, reason);
>  
> -	xe_gt_reset_async(guc_to_gt(guc));
> +	xe_gt_reset_async(gt);
>  
>  	return 0;
>  }
> -- 
> 2.43.0
>