drm/i915: Convert hangcheck from a timer into a delayed work item

When run as a timer, i915_hangcheck_elapsed() must adhere to all the
rules of running in a softirq context. This is advantageous to us as we
want to minimise the risk that a driver bug will prevent us from
detecting a hung GPU. However, that is irrelevant if the driver bug
prevents us from resetting and recovering. Still it is prudent not to
rely on mutexes inside the checker, but given the coarseness of
dev->struct_mutex doing so is extremely hard.

Give in and run from a work queue, i.e. outside of softirq.

v2: Use own workqueue to avoid deadlocks (Daniel)
    Cleanup commit msg and add comment to i915_queue_hangcheck() (Chris)

Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Daniel Vetter <dnaiel.vetter@ffwll.chm>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> (v1)
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
[danvet: Remove accidental kerneldoc comment starter, to appease the 0
day builder.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

1 files changed, 16 insertions, 14 deletions

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 2399eaed2ca3..23bfe2232b6a 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2974,7 +2974,7 @@ ring_stuck(struct intel_engine_cs *ring, u64 acthd)
 	return HANGCHECK_HUNG;
 }
 
-/**
+/*
  * 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
  * if there are no progress, hangcheck score for that ring is increased.
@@ -2982,10 +2982,12 @@ ring_stuck(struct intel_engine_cs *ring, u64 acthd)
  * we kick the ring. If we see no progress on three subsequent calls
  * we assume chip is wedged and try to fix it by resetting the chip.
  */
-static void i915_hangcheck_elapsed(unsigned long data)
+static void i915_hangcheck_elapsed(struct work_struct *work)
 {
-	struct drm_device *dev = (struct drm_device *)data;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv =
+		container_of(work, typeof(*dev_priv),
+			     gpu_error.hangcheck_work.work);
+	struct drm_device *dev = dev_priv->dev;
 	struct intel_engine_cs *ring;
 	int i;
 	int busy_count = 0, rings_hung = 0;
@@ -3099,17 +3101,18 @@ static void i915_hangcheck_elapsed(unsigned long data)
 
 void i915_queue_hangcheck(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct timer_list *timer = &dev_priv->gpu_error.hangcheck_timer;
+	struct i915_gpu_error *e = &to_i915(dev)->gpu_error;
 
 	if (!i915.enable_hangcheck)
 		return;
 
-	/* Don't continually defer the hangcheck, but make sure it is active */
-	if (timer_pending(timer))
-		return;
-	mod_timer(timer,
-		  round_jiffies_up(jiffies + DRM_I915_HANGCHECK_JIFFIES));
+	/* Don't continually defer the hangcheck so that it is always run at
+	 * least once after work has been scheduled on any ring. Otherwise,
+	 * we will ignore a hung ring if a second ring is kept busy.
+	 */
+
+	queue_delayed_work(e->hangcheck_wq, &e->hangcheck_work,
+			   round_jiffies_up_relative(DRM_I915_HANGCHECK_JIFFIES));
 }
 
 static void ibx_irq_reset(struct drm_device *dev)
@@ -4353,9 +4356,8 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 	else
 		dev_priv->pm_rps_events = GEN6_PM_RPS_EVENTS;
 
-	setup_timer(&dev_priv->gpu_error.hangcheck_timer,
-		    i915_hangcheck_elapsed,
-		    (unsigned long) dev);
+	INIT_DELAYED_WORK(&dev_priv->gpu_error.hangcheck_work,
+			  i915_hangcheck_elapsed);
 	INIT_DELAYED_WORK(&dev_priv->hotplug_reenable_work,
 			  intel_hpd_irq_reenable_work);