drm/i915/gt: Always report the sample time for busy-stats

Return the monotonic timestamp (ktime_get()) at the time of sampling the
busy-time. This is used in preference to taking ktime_get() separately
before or after the read seqlock as there can be some large variance in
reported timestamps. For selftests trying to ascertain that we are
reporting accurate to within a few microseconds, even a small delay
leads to the test failing.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200617130916.15261-2-chris@chris-wilson.co.uk

4 files changed, 21 insertions, 21 deletions

diff --git a/drivers/gpu/drm/i915/gt/intel_engine.h b/drivers/gpu/drm/i915/gt/intel_engine.h
index 791897f8d847..a9249a23903a 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine.h
@@ -334,7 +334,8 @@ void intel_engine_dump(struct intel_engine_cs *engine,
 		       struct drm_printer *m,
 		       const char *header, ...);
 
-ktime_t intel_engine_get_busy_time(struct intel_engine_cs *engine);
+ktime_t intel_engine_get_busy_time(struct intel_engine_cs *engine,
+				   ktime_t *now);
 
 struct i915_request *
 intel_engine_find_active_request(struct intel_engine_cs *engine);
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
index 045179c65c44..c62b3cbdbbf9 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
@@ -1595,7 +1595,8 @@ void intel_engine_dump(struct intel_engine_cs *engine,
 	intel_engine_print_breadcrumbs(engine, m);
 }
 
-static ktime_t __intel_engine_get_busy_time(struct intel_engine_cs *engine)
+static ktime_t __intel_engine_get_busy_time(struct intel_engine_cs *engine,
+					    ktime_t *now)
 {
 	ktime_t total = engine->stats.total;
 
@@ -1603,9 +1604,9 @@ static ktime_t __intel_engine_get_busy_time(struct intel_engine_cs *engine)
 	 * If the engine is executing something at the moment
 	 * add it to the total.
 	 */
+	*now = ktime_get();
 	if (atomic_read(&engine->stats.active))
-		total = ktime_add(total,
-				  ktime_sub(ktime_get(), engine->stats.start));
+		total = ktime_add(total, ktime_sub(*now, engine->stats.start));
 
 	return total;
 }
@@ -1613,17 +1614,18 @@ static ktime_t __intel_engine_get_busy_time(struct intel_engine_cs *engine)
 /**
  * intel_engine_get_busy_time() - Return current accumulated engine busyness
  * @engine: engine to report on
+ * @now: monotonic timestamp of sampling
  *
  * Returns accumulated time @engine was busy since engine stats were enabled.
  */
-ktime_t intel_engine_get_busy_time(struct intel_engine_cs *engine)
+ktime_t intel_engine_get_busy_time(struct intel_engine_cs *engine, ktime_t *now)
 {
 	unsigned int seq;
 	ktime_t total;
 
 	do {
 		seq = read_seqbegin(&engine->stats.lock);
-		total = __intel_engine_get_busy_time(engine);
+		total = __intel_engine_get_busy_time(engine, now);
 	} while (read_seqretry(&engine->stats.lock, seq));
 
 	return total;
diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c
index 2f59fc6df3c2..bdece932592b 100644
--- a/drivers/gpu/drm/i915/gt/intel_rps.c
+++ b/drivers/gpu/drm/i915/gt/intel_rps.c
@@ -53,13 +53,13 @@ static void rps_timer(struct timer_list *t)
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 	s64 max_busy[3] = {};
-	ktime_t dt, last;
+	ktime_t dt, timestamp, last;
 
 	for_each_engine(engine, rps_to_gt(rps), id) {
 		s64 busy;
 		int i;
 
-		dt = intel_engine_get_busy_time(engine);
+		dt = intel_engine_get_busy_time(engine, &timestamp);
 		last = engine->stats.rps;
 		engine->stats.rps = dt;
 
@@ -70,15 +70,14 @@ static void rps_timer(struct timer_list *t)
 		}
 	}
 
-	dt = ktime_get();
 	last = rps->pm_timestamp;
-	rps->pm_timestamp = dt;
+	rps->pm_timestamp = timestamp;
 
 	if (intel_rps_is_active(rps)) {
 		s64 busy;
 		int i;
 
-		dt = ktime_sub(dt, last);
+		dt = ktime_sub(timestamp, last);
 
 		/*
 		 * Our goal is to evaluate each engine independently, so we run
diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_pm.c b/drivers/gpu/drm/i915/gt/selftest_engine_pm.c
index dd54dcb5cca2..b08fc5390e8a 100644
--- a/drivers/gpu/drm/i915/gt/selftest_engine_pm.c
+++ b/drivers/gpu/drm/i915/gt/selftest_engine_pm.c
@@ -29,8 +29,8 @@ static int live_engine_busy_stats(void *arg)
 	GEM_BUG_ON(intel_gt_pm_is_awake(gt));
 	for_each_engine(engine, gt, id) {
 		struct i915_request *rq;
-		ktime_t de;
-		u64 dt;
+		ktime_t de, dt;
+		ktime_t t[2];
 
 		if (!intel_engine_supports_stats(engine))
 			continue;
@@ -47,12 +47,11 @@ static int live_engine_busy_stats(void *arg)
 
 		ENGINE_TRACE(engine, "measuring idle time\n");
 		preempt_disable();
-		dt = ktime_to_ns(ktime_get());
-		de = intel_engine_get_busy_time(engine);
+		de = intel_engine_get_busy_time(engine, &t[0]);
 		udelay(100);
-		de = ktime_sub(intel_engine_get_busy_time(engine), de);
-		dt = ktime_to_ns(ktime_get()) - dt;
+		de = ktime_sub(intel_engine_get_busy_time(engine, &t[1]), de);
 		preempt_enable();
+		dt = ktime_sub(t[1], t[0]);
 		if (de < 0 || de > 10) {
 			pr_err("%s: reported %lldns [%d%%] busyness while sleeping [for %lldns]\n",
 			       engine->name,
@@ -80,12 +79,11 @@ static int live_engine_busy_stats(void *arg)
 
 		ENGINE_TRACE(engine, "measuring busy time\n");
 		preempt_disable();
-		dt = ktime_to_ns(ktime_get());
-		de = intel_engine_get_busy_time(engine);
+		de = intel_engine_get_busy_time(engine, &t[0]);
 		udelay(100);
-		de = ktime_sub(intel_engine_get_busy_time(engine), de);
-		dt = ktime_to_ns(ktime_get()) - dt;
+		de = ktime_sub(intel_engine_get_busy_time(engine, &t[1]), de);
 		preempt_enable();
+		dt = ktime_sub(t[1], t[0]);
 		if (100 * de < 95 * dt || 95 * de > 100 * dt) {
 			pr_err("%s: reported %lldns [%d%%] busyness while spinning [for %lldns]\n",
 			       engine->name,