drm/i915: Use spinlocks for checking when to waitboost

In commit 1854d5ca0dd7a9fc11243ff220a3e93fce2b4d3e
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Apr 7 16:20:32 2015 +0100

    drm/i915: Deminish contribution of wait-boosting from clients

we removed an atomic timer based check for allowing waitboosting and
moved it below the mutex taken during RPS. However, that mutex can be
held for long periods of time on Vallyview/Cherryview as communication
with the PCU is slow. As clients may frequently wait for results (e.g.
such as tranform feedback) we introduced contention between the client
and the RPS worker. We can take advantage of the RPS worker, by
switching the wait boost decision to use spin locks and defer the
actual reclocking to the worker.

Fixes a regression of up to 45% on Baytrail and Baswell!

v2 (Daniel):
- Use max_freq_softlimit instead of the not-yet-merged boost
  frequency.
- Don't inject a fake irq into the boost work, instead treat
  client_boost as just another legit waker.

v3: Drop the now unused mask (Chris).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90112
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> (v1)
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

1 files changed, 13 insertions, 6 deletions

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 707e2ca8fbd8..e6bb72dca3ff 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1086,8 +1086,9 @@ static void gen6_pm_rps_work(struct work_struct *work)
 {
 	struct drm_i915_private *dev_priv =
 		container_of(work, struct drm_i915_private, rps.work);
+	bool client_boost;
+	int new_delay, adj, min, max;
 	u32 pm_iir;
-	int new_delay, adj;
 
 	spin_lock_irq(&dev_priv->irq_lock);
 	/* Speed up work cancelation during disabling rps interrupts. */
@@ -1099,12 +1100,14 @@ static void gen6_pm_rps_work(struct work_struct *work)
 	dev_priv->rps.pm_iir = 0;
 	/* Make sure not to corrupt PMIMR state used by ringbuffer on GEN6 */
 	gen6_enable_pm_irq(dev_priv, dev_priv->pm_rps_events);
+	client_boost = dev_priv->rps.client_boost;
+	dev_priv->rps.client_boost = false;
 	spin_unlock_irq(&dev_priv->irq_lock);
 
 	/* Make sure we didn't queue anything we're not going to process. */
 	WARN_ON(pm_iir & ~dev_priv->pm_rps_events);
 
-	if ((pm_iir & dev_priv->pm_rps_events) == 0)
+	if ((pm_iir & dev_priv->pm_rps_events) == 0 && !client_boost)
 		return;
 
 	mutex_lock(&dev_priv->rps.hw_lock);
@@ -1113,7 +1116,13 @@ static void gen6_pm_rps_work(struct work_struct *work)
 
 	adj = dev_priv->rps.last_adj;
 	new_delay = dev_priv->rps.cur_freq;
-	if (pm_iir & GEN6_PM_RP_UP_THRESHOLD) {
+	min = dev_priv->rps.min_freq_softlimit;
+	max = dev_priv->rps.max_freq_softlimit;
+
+	if (client_boost) {
+		new_delay = dev_priv->rps.max_freq_softlimit;
+		adj = 0;
+	} else if (pm_iir & GEN6_PM_RP_UP_THRESHOLD) {
 		if (adj > 0)
 			adj *= 2;
 		else /* CHV needs even encode values */
@@ -1149,9 +1158,7 @@ static void gen6_pm_rps_work(struct work_struct *work)
 	 * interrupt
 	 */
 	new_delay += adj;
-	new_delay = clamp_t(int, new_delay,
-			    dev_priv->rps.min_freq_softlimit,
-			    dev_priv->rps.max_freq_softlimit);
+	new_delay = clamp_t(int, new_delay, min, max);
 
 	intel_set_rps(dev_priv->dev, new_delay);