From: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
If only a subset of events is enabled we can afford to suspend
the sampling timer when the GPU is idle and so save some cycles
and power.
v2: Rebase and limit timer even more.
v3: Rebase.
v4: Rebase.
v5: Skip action if perf PMU failed to register.
v6: Checkpatch cleanup.
v7:
* Add a common helper to start the timer if needed. (Chris Wilson)
* Add comment explaining bitwise logic in pmu_needs_timer.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
---
drivers/gpu/drm/i915/i915_drv.h | 8 +++
drivers/gpu/drm/i915/i915_gem.c | 1 +
drivers/gpu/drm/i915/i915_gem_request.c | 1 +
drivers/gpu/drm/i915/i915_pmu.c | 88 +++++++++++++++++++++++++++++----
4 files changed, 88 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 824372037672..0e0df3b72944 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2252,6 +2252,10 @@ struct i915_pmu {
*/
unsigned int enable_count[I915_PMU_MASK_BITS];
/**
+ * @timer_enabled: Should the internal sampling timer be running.
+ */
+ bool timer_enabled;
+ /**
* @sample: Current counter value for i915 events which need sampling.
*
* These counters are updated from the i915 PMU sampling timer.
@@ -4002,9 +4006,13 @@ extern void i915_perf_unregister(struct drm_i915_private *dev_priv);
#ifdef CONFIG_PERF_EVENTS
void i915_pmu_register(struct drm_i915_private *i915);
void i915_pmu_unregister(struct drm_i915_private *i915);
+void i915_pmu_gt_idle(struct drm_i915_private *i915);
+void i915_pmu_gt_active(struct drm_i915_private *i915);
#else
static inline void i915_pmu_register(struct drm_i915_private *i915) {}
static inline void i915_pmu_unregister(struct drm_i915_private *i915) {}
+static inline void i915_pmu_gt_idle(struct drm_i915_private *i915) {}
+static inline void i915_pmu_gt_active(struct drm_i915_private *i915) {}
#endif
/* i915_suspend.c */
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 3250dfaa192b..27ae16abed86 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3227,6 +3227,7 @@ i915_gem_idle_work_handler(struct work_struct *work)
intel_engines_mark_idle(dev_priv);
i915_gem_timelines_mark_idle(dev_priv);
+ i915_pmu_gt_idle(dev_priv);
GEM_BUG_ON(!dev_priv->gt.awake);
dev_priv->gt.awake = false;
diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c
index 813a3b546d6e..18a1e379253e 100644
--- a/drivers/gpu/drm/i915/i915_gem_request.c
+++ b/drivers/gpu/drm/i915/i915_gem_request.c
@@ -258,6 +258,7 @@ static void mark_busy(struct drm_i915_private *i915)
i915_update_gfx_val(i915);
if (INTEL_GEN(i915) >= 6)
gen6_rps_busy(i915);
+ i915_pmu_gt_active(i915);
queue_delayed_work(i915->wq,
&i915->gt.retire_work,
diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index b0a942f8da74..b7de6fe3cac7 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -90,6 +90,75 @@ static unsigned int event_enabled_bit(struct perf_event *event)
return config_enabled_bit(event->attr.config);
}
+static bool pmu_needs_timer(struct drm_i915_private *i915, bool gpu_active)
+{
+ u64 enable;
+
+ /**
+ * Only some counters need the sampling timer.
+ *
+ * We start with a bitmask of all currently enabled events.
+ */
+ enable = i915->pmu.enable;
+
+ /**
+ * Mask out all the ones which do not need the timer, or in
+ * other words keep all the ones that could need the timer.
+ */
+ enable &= config_enabled_mask(I915_PMU_ACTUAL_FREQUENCY) |
+ config_enabled_mask(I915_PMU_REQUESTED_FREQUENCY) |
+ ENGINE_SAMPLE_MASK;
+
+ /**
+ * When the GPU is idle per-engine counters do not need to be
+ * running so clear those bits out.
+ */
+ if (!gpu_active)
+ enable &= ~ENGINE_SAMPLE_MASK;
+
+ /**
+ * If some bits remain it means we need the sampling timer running.
+ */
+ return enable;
+}
+
+void i915_pmu_gt_idle(struct drm_i915_private *i915)
+{
+ if (!i915->pmu.base.event_init)
+ return;
+
+ spin_lock_irq(&i915->pmu.lock);
+ /*
+ * Signal sampling timer to stop if only engine events are enabled and
+ * GPU went idle.
+ */
+ i915->pmu.timer_enabled = pmu_needs_timer(i915, false);
+ spin_unlock_irq(&i915->pmu.lock);
+}
+
+static void __i915_pmu_maybe_start_timer(struct drm_i915_private *i915)
+{
+ if (!i915->pmu.timer_enabled && pmu_needs_timer(i915, true)) {
+ i915->pmu.timer_enabled = true;
+ hrtimer_start_range_ns(&i915->pmu.timer,
+ ns_to_ktime(PERIOD), 0,
+ HRTIMER_MODE_REL_PINNED);
+ }
+}
+
+void i915_pmu_gt_active(struct drm_i915_private *i915)
+{
+ if (!i915->pmu.base.event_init)
+ return;
+
+ spin_lock_irq(&i915->pmu.lock);
+ /*
+ * Re-enable sampling timer when GPU goes active.
+ */
+ __i915_pmu_maybe_start_timer(i915);
+ spin_unlock_irq(&i915->pmu.lock);
+}
+
static bool grab_forcewake(struct drm_i915_private *i915, bool fw)
{
if (!fw)
@@ -181,7 +250,7 @@ static enum hrtimer_restart i915_sample(struct hrtimer *hrtimer)
struct drm_i915_private *i915 =
container_of(hrtimer, struct drm_i915_private, pmu.timer);
- if (i915->pmu.enable == 0)
+ if (!READ_ONCE(i915->pmu.timer_enabled))
return HRTIMER_NORESTART;
engines_sample(i915);
@@ -379,14 +448,6 @@ static void i915_pmu_enable(struct perf_event *event)
spin_lock_irqsave(&i915->pmu.lock, flags);
/*
- * Start the sampling timer when enabling the first event.
- */
- if (i915->pmu.enable == 0)
- hrtimer_start_range_ns(&i915->pmu.timer,
- ns_to_ktime(PERIOD), 0,
- HRTIMER_MODE_REL_PINNED);
-
- /*
* Update the bitmask of enabled events and increment
* the event reference counter.
*/
@@ -396,6 +457,11 @@ static void i915_pmu_enable(struct perf_event *event)
i915->pmu.enable_count[bit]++;
/*
+ * Start the sampling timer if needed and not already enabled.
+ */
+ __i915_pmu_maybe_start_timer(i915);
+
+ /*
* For per-engine events the bitmask and reference counting
* is stored per engine.
*/
@@ -457,8 +523,10 @@ static void i915_pmu_disable(struct perf_event *event)
* Decrement the reference count and clear the enabled
* bitmask when the last listener on an event goes away.
*/
- if (--i915->pmu.enable_count[bit] == 0)
+ if (--i915->pmu.enable_count[bit] == 0) {
i915->pmu.enable &= ~BIT_ULL(bit);
+ i915->pmu.timer_enabled &= pmu_needs_timer(i915, true);
+ }
spin_unlock_irqrestore(&i915->pmu.lock, flags);
}
--
2.9.5
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
