drm/i915: Move missed interrupt detection from hangcheck to breadcrumbs

In commit 2529d57050af ("drm/i915: Drop racy markup of missed-irqs from
idle-worker") the racy detection of missed interrupts was removed when
we went idle. This however opened up the issue that the stuck waiters
were not being reported, causing a test case failure. If we move the
stuck waiter detection out of hangcheck and into the breadcrumb
mechanims (i.e. the waiter) itself, we can avoid this issue entirely.
This leaves hangcheck looking for a stuck GPU (inspecting for request
advancement and HEAD motion), and breadcrumbs looking for a stuck
waiter - hopefully make both easier to understand by their segregation.

v2: Reduce the error message as we now run independently of hangcheck,
and the hanging batch used by igt also counts as a stuck waiter causing
extra warnings in dmesg.
v3: Move the breadcrumb's hangcheck kickstart to the first missed wait.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97104
Fixes: 2529d57050af (waiter"drm/i915: Drop racy markup of missed-irqs...")
Testcase: igt/drv_missed_irq
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470761272-1245-2-git-send-email-chris@chris-wilson.co.uk

1 files changed, 1 insertions, 25 deletions

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 591f452ece68..ebb83d5a448b 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -972,10 +972,8 @@ static void ironlake_rps_change_irq_handler(struct drm_i915_private *dev_priv)
 static void notify_ring(struct intel_engine_cs *engine)
 {
 	smp_store_mb(engine->breadcrumbs.irq_posted, true);
-	if (intel_engine_wakeup(engine)) {
+	if (intel_engine_wakeup(engine))
 		trace_i915_gem_request_notify(engine);
-		engine->breadcrumbs.irq_wakeups++;
-	}
 }
 
 static void vlv_c0_read(struct drm_i915_private *dev_priv,
@@ -3044,22 +3042,6 @@ engine_stuck(struct intel_engine_cs *engine, u64 acthd)
 	return HANGCHECK_HUNG;
 }
 
-static unsigned long kick_waiters(struct intel_engine_cs *engine)
-{
-	struct drm_i915_private *i915 = engine->i915;
-	unsigned long irq_count = READ_ONCE(engine->breadcrumbs.irq_wakeups);
-
-	if (engine->hangcheck.user_interrupts == irq_count &&
-	    !test_and_set_bit(engine->id, &i915->gpu_error.missed_irq_rings)) {
-		if (!test_bit(engine->id, &i915->gpu_error.test_irq_rings))
-			DRM_ERROR("Hangcheck timer elapsed... %s idle\n",
-				  engine->name);
-
-		intel_engine_enable_fake_irq(engine);
-	}
-
-	return irq_count;
-}
 /*
  * This is called when the chip hasn't reported back with completed
  * batchbuffers in a long time. We keep track per ring seqno progress and
@@ -3097,7 +3079,6 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 		bool busy = intel_engine_has_waiter(engine);
 		u64 acthd;
 		u32 seqno;
-		unsigned user_interrupts;
 
 		semaphore_clear_deadlocks(dev_priv);
 
@@ -3114,15 +3095,11 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 		acthd = intel_engine_get_active_head(engine);
 		seqno = intel_engine_get_seqno(engine);
 
-		/* Reset stuck interrupts between batch advances */
-		user_interrupts = 0;
-
 		if (engine->hangcheck.seqno == seqno) {
 			if (!intel_engine_is_active(engine)) {
 				engine->hangcheck.action = HANGCHECK_IDLE;
 				if (busy) {
 					/* Safeguard against driver failure */
-					user_interrupts = kick_waiters(engine);
 					engine->hangcheck.score += BUSY;
 				}
 			} else {
@@ -3185,7 +3162,6 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 
 		engine->hangcheck.seqno = seqno;
 		engine->hangcheck.acthd = acthd;
-		engine->hangcheck.user_interrupts = user_interrupts;
 		busy_count += busy;
 	}