[Intel-gfx] [PATCH] drm/i915: Don't accumate hangcheck score on forward progress

[Mika Kuoppala]

If the actual head has progressed forward inside a batch (request),
don't accumulate hangcheck score.

Signed-off-by: Mika Kuoppala <mika.kuoppala at intel.com>
---
 drivers/gpu/drm/i915/i915_gpu_error.c   |    2 ++
 drivers/gpu/drm/i915/i915_irq.c         |   15 ++++++++++++---
 drivers/gpu/drm/i915/intel_ringbuffer.h |    2 ++
 3 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 0b3f694..0eb4470 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -229,6 +229,8 @@ static const char *hangcheck_action_to_str(enum intel_ring_hangcheck_action a)
 		return "wait";
 	case HANGCHECK_ACTIVE:
 		return "active";
+	case HANGCHECK_ACTIVE_LOOP:
+		return "active_loop";
 	case HANGCHECK_KICK:
 		return "kick";
 	case HANGCHECK_HUNG:
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index cdab0b4..008632e 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -3187,8 +3187,14 @@ ring_stuck(struct intel_engine_cs *ring, u64 acthd)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 tmp;
 
-	if (ring->hangcheck.acthd != acthd)
-		return HANGCHECK_ACTIVE;
+	if (acthd != ring->hangcheck.acthd) {
+		if (acthd > ring->hangcheck.max_acthd) {
+			ring->hangcheck.max_acthd = acthd;
+			return HANGCHECK_ACTIVE;
+		}
+
+		return HANGCHECK_ACTIVE_LOOP;
+	}
 
 	if (IS_GEN2(dev))
 		return HANGCHECK_HUNG;
@@ -3299,8 +3305,9 @@ static void i915_hangcheck_elapsed(unsigned long data)
 				switch (ring->hangcheck.action) {
 				case HANGCHECK_IDLE:
 				case HANGCHECK_WAIT:
-					break;
 				case HANGCHECK_ACTIVE:
+					break;
+				case HANGCHECK_ACTIVE_LOOP:
 					ring->hangcheck.score += BUSY;
 					break;
 				case HANGCHECK_KICK:
@@ -3320,6 +3327,8 @@ static void i915_hangcheck_elapsed(unsigned long data)
 			 */
 			if (ring->hangcheck.score > 0)
 				ring->hangcheck.score--;
+
+			ring->hangcheck.acthd = ring->hangcheck.max_acthd = 0;
 		}
 
 		ring->hangcheck.seqno = seqno;
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index ed59410..70525d0 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -70,6 +70,7 @@ enum intel_ring_hangcheck_action {
 	HANGCHECK_IDLE = 0,
 	HANGCHECK_WAIT,
 	HANGCHECK_ACTIVE,
+	HANGCHECK_ACTIVE_LOOP,
 	HANGCHECK_KICK,
 	HANGCHECK_HUNG,
 };
@@ -78,6 +79,7 @@ enum intel_ring_hangcheck_action {
 
 struct intel_ring_hangcheck {
 	u64 acthd;
+	u64 max_acthd;
 	u32 seqno;
 	int score;
 	enum intel_ring_hangcheck_action action;
-- 
1.7.9.5