[PATCH 7/7] drm/xe: Wire devcoredump to LR TDR
Matthew Brost
matthew.brost at intel.com
Fri Nov 8 17:43:12 UTC 2024
LR queue's can hang, cause engine reset, or cause IOMMU CAT errors.
Collect an error capture when this occurs.
Signed-off-by: Matthew Brost <matthew.brost at intel.com>
---
drivers/gpu/drm/xe/xe_guc_submit.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 3df8543deec1..a52e56ea42df 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -901,6 +901,7 @@ 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) {
+ xe_devcoredump(q, NULL);
drm_warn(&xe->drm, "Schedule disable failed to respond");
xe_sched_submission_start(sched);
xe_gt_reset_async(q->gt);
@@ -908,6 +909,9 @@ static void xe_guc_exec_queue_lr_cleanup(struct work_struct *w)
}
}
+ if (!exec_queue_killed(q) && !xe_lrc_ring_is_idle(q->lrc[0]))
+ xe_devcoredump(q, NULL);
+
xe_sched_submission_start(sched);
}
--
2.34.1
More information about the Intel-xe
mailing list