Hi Michal,

On 25.04.2024 19:46, Matthew Brost wrote:

On Thu, Apr 25, 2024 at 02:18:56PM +0200, Nirmoy Das wrote:

Add engine name to the engine reset and cat error log
which should be useful while debugging.

Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
---
 drivers/gpu/drm/xe/xe_guc_submit.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index c7d38469fb46..245e29d095c0 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -1655,7 +1655,7 @@ int xe_guc_exec_queue_reset_handler(struct xe_guc *guc, u32 *msg, u32 len)
 	if (unlikely(!q))
 		return -EPROTO;
 
-	drm_info(&xe->drm, "Engine reset: guc_id=%d", guc_id);
+	drm_info(&xe->drm, "Engine reset: name=%s, guc_id=%d", q->hwe->name, guc_id);

I don't think q->hwe->name name is useful as it might not actually be
exec queue is running. I'd drop that, and replace with string indicating
the hardware engine class. 

and maybe it's also a good time to switch into GT-oriented logs:

-	struct xe_device *xe = guc_to_xe(guc);
+	struct xe_gt *gt = guc_to_gt(guc);

	xe_gt_info(gt, "Engine reset....

Yes, gt log will throw more useful info. I will switch to xe_gt_info/xe_gt_dbg.

Thanks,

Nirmoy

 
 	/* FIXME: Do error capture, most likely async */
 
@@ -1690,7 +1690,8 @@ int xe_guc_exec_queue_memory_cat_error_handler(struct xe_guc *guc, u32 *msg,
 	if (unlikely(!q))
 		return -EPROTO;
 
-	drm_dbg(&xe->drm, "Engine memory cat error: guc_id=%d", guc_id);
+	drm_dbg(&xe->drm, "Engine memory cat error: name=%s, guc_id=%d",
+		q->hwe->name, guc_id);

Same here.

Matt

 	trace_xe_exec_queue_memory_cat_error(q);
 
 	/* Treat the same as engine reset */
-- 
2.42.0