Skip to contentSkip to conversation

drm/i915/pmu: Fix PMU enable vs execlists tasklet race

PATCH Superseded Tvrtko Ursulin 3 participants 8 messages Feb 2, 2018 – Feb 6, 2018

Subsystems (2)
lore
Share & tools
lore

Tvrtko Ursulin

author maintainer
#0 lore raw

PATCH 1/1 drm/i915/pmu: Fix PMU enable vs execlists tasklet race

From: Tvrtko Ursulin <[email protected]>

Commit 99e48bf98dd0 ("drm/i915: Lock out execlist tasklet while peeking inside for busy-stats") added a tasklet_disable call in busy stats enabling, but we failed to understand that the PMU enable callback runs as an hard IRQ (IPI).

Consequence of this is that the PMU enable callback can interrupt the execlists tasklet, and will then deadlock when it calls intel_engine_stats_enable->tasklet_disable.

To fix this, I realized it is possible to move the engine stats enablement and disablement to PMU event init and destroy hooks. This allows for much simpler implementation since those hooks run in normal context (can sleep).

Fixes 99e48bf98dd0 ("drm/i915: Lock out execlist tasklet while peeking inside for busy-stats")

Testcase: igt/perf_pmu/enable-race-*
Cc: Chris Wilson <[email protected]>
Cc: Tvrtko Ursulin <[email protected]>
Cc: Jani Nikula <[email protected]>
Cc: Joonas Lahtinen <[email protected]>
Cc: Rodrigo Vivi <[email protected]>
Cc: [email protected]

---
 drivers/gpu/drm/i915/i915_pmu.c         | 95 +++++++++++----------------------
 drivers/gpu/drm/i915/intel_ringbuffer.h | 14 -----
 2 files changed, 31 insertions(+), 78 deletions(-)
drivers/gpu/drm/i915/i915_pmu.c +31 -64
index ecb0198bfb7a..e97860b44004 100644
@@ -287,7 +287,24 @@ static u64 count_interrupts(struct drm_i915_private *i915)
 
 static void i915_pmu_event_destroy(struct perf_event *event)
 {
+	struct drm_i915_private *i915 =
+		container_of(event->pmu, typeof(*i915), pmu.base);
+	struct intel_engine_cs *engine;
+
 	WARN_ON(event->parent);
+
+	if (!is_engine_event(event))
+		return;
+
+	engine = intel_engine_lookup_user(i915,
+					  engine_event_class(event),
+					  engine_event_instance(event));
+	if (WARN_ON_ONCE(!engine))
+		return;
+
+	if ((engine_event_sample(event) == I915_SAMPLE_BUSY) &&
+	    intel_engine_supports_stats(engine))
+		intel_disable_engine_stats(engine);
 }
 
 static int
@@ -340,13 +357,23 @@ static int engine_event_init(struct perf_event *event)
 	struct drm_i915_private *i915 =
 		container_of(event->pmu, typeof(*i915), pmu.base);
 	struct intel_engine_cs *engine;
+	u8 sample;
+	int ret;
 
 	engine = intel_engine_lookup_user(i915, engine_event_class(event),
 					  engine_event_instance(event));
 	if (!engine)
 		return -ENODEV;
 
-	return engine_event_status(engine, engine_event_sample(event));
+	sample = engine_event_sample(event);
+	ret = engine_event_status(engine, sample);
+	if (ret)
+		return ret;
+
+	if ((sample == I915_SAMPLE_BUSY) && intel_engine_supports_stats(engine))
+		ret = intel_enable_engine_stats(engine);
+
+	return ret;
 }
 
 static int i915_pmu_event_init(struct perf_event *event)
@@ -402,7 +429,7 @@ static u64 __i915_pmu_event_read(struct perf_event *event)
 		if (WARN_ON_ONCE(!engine)) {
 			/* Do nothing */
 		} else if (sample == I915_SAMPLE_BUSY &&
-			   engine->pmu.busy_stats) {
+			   intel_engine_supports_stats(engine)) {
 			val = ktime_to_ns(intel_engine_get_busy_time(engine));
 		} else {
 			val = engine->pmu.sample[sample].cur;
@@ -457,12 +484,6 @@ static void i915_pmu_event_read(struct perf_event *event)
 	local64_add(new - prev, &event->count);
 }
 
-static bool engine_needs_busy_stats(struct intel_engine_cs *engine)
-{
-	return intel_engine_supports_stats(engine) &&
-	       (engine->pmu.enable & BIT(I915_SAMPLE_BUSY));
-}
-
 static void i915_pmu_enable(struct perf_event *event)
 {
 	struct drm_i915_private *i915 =
@@ -502,21 +523,7 @@ static void i915_pmu_enable(struct perf_event *event)
 
 		GEM_BUG_ON(sample >= I915_PMU_SAMPLE_BITS);
 		GEM_BUG_ON(engine->pmu.enable_count[sample] == ~0);
-		if (engine->pmu.enable_count[sample]++ == 0) {
-			/*
-			 * Enable engine busy stats tracking if needed or
-			 * alternatively cancel the scheduled disable.
-			 *
-			 * If the delayed disable was pending, cancel it and
-			 * in this case do not enable since it already is.
-			 */
-			if (engine_needs_busy_stats(engine) &&
-			    !engine->pmu.busy_stats) {
-				engine->pmu.busy_stats = true;
-				if (!cancel_delayed_work(&engine->pmu.disable_busy_stats))
-					intel_enable_engine_stats(engine);
-			}
-		}
+		engine->pmu.enable_count[sample]++;
 	}
 
 	/*
@@ -529,14 +536,6 @@ static void i915_pmu_enable(struct perf_event *event)
 	spin_unlock_irqrestore(&i915->pmu.lock, flags);
 }
 
-static void __disable_busy_stats(struct work_struct *work)
-{
-	struct intel_engine_cs *engine =
-	       container_of(work, typeof(*engine), pmu.disable_busy_stats.work);
-
-	intel_disable_engine_stats(engine);
-}
-
 static void i915_pmu_disable(struct perf_event *event)
 {
 	struct drm_i915_private *i915 =
@@ -560,26 +559,8 @@ 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 (--engine->pmu.enable_count[sample] == 0) {
+		if (--engine->pmu.enable_count[sample] == 0)
 			engine->pmu.enable &= ~BIT(sample);
-			if (!engine_needs_busy_stats(engine) &&
-			    engine->pmu.busy_stats) {
-				engine->pmu.busy_stats = false;
-				/*
-				 * We request a delayed disable to handle the
-				 * rapid on/off cycles on events, which can
-				 * happen when tools like perf stat start, in a
-				 * nicer way.
-				 *
-				 * In addition, this also helps with busy stats
-				 * accuracy with background CPU offline/online
-				 * migration events.
-				 */
-				queue_delayed_work(system_wq,
-						   &engine->pmu.disable_busy_stats,
-						   round_jiffies_up_relative(HZ));
-			}
-		}
 	}
 
 	GEM_BUG_ON(bit >= I915_PMU_MASK_BITS);
@@ -956,8 +937,6 @@ static void i915_pmu_unregister_cpuhp_state(struct drm_i915_private *i915)
 
 void i915_pmu_register(struct drm_i915_private *i915)
 {
-	struct intel_engine_cs *engine;
-	enum intel_engine_id id;
 	int ret;
 
 	if (INTEL_GEN(i915) <= 2) {
@@ -985,10 +964,6 @@ void i915_pmu_register(struct drm_i915_private *i915)
 	hrtimer_init(&i915->pmu.timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	i915->pmu.timer.function = i915_sample;
 
-	for_each_engine(engine, i915, id)
-		INIT_DELAYED_WORK(&engine->pmu.disable_busy_stats,
-				  __disable_busy_stats);
-
 	ret = perf_pmu_register(&i915->pmu.base, "i915", -1);
 	if (ret)
 		goto err;
@@ -1009,9 +984,6 @@ void i915_pmu_register(struct drm_i915_private *i915)
 
 void i915_pmu_unregister(struct drm_i915_private *i915)
 {
-	struct intel_engine_cs *engine;
-	enum intel_engine_id id;
-
 	if (!i915->pmu.base.event_init)
 		return;
 
@@ -1019,11 +991,6 @@ void i915_pmu_unregister(struct drm_i915_private *i915)
 
 	hrtimer_cancel(&i915->pmu.timer);
 
-	for_each_engine(engine, i915, id) {
-		GEM_BUG_ON(engine->pmu.busy_stats);
-		flush_delayed_work(&engine->pmu.disable_busy_stats);
-	}
-
 	i915_pmu_unregister_cpuhp_state(i915);
 
 	perf_pmu_unregister(&i915->pmu.base);
drivers/gpu/drm/i915/intel_ringbuffer.h +0 -14
index c5ff203e42d6..a0e7a6c2a57c 100644
@@ -366,20 +366,6 @@ struct intel_engine_cs {
 		 */
 #define I915_ENGINE_SAMPLE_MAX (I915_SAMPLE_SEMA + 1)
 		struct i915_pmu_sample sample[I915_ENGINE_SAMPLE_MAX];
-		/**
-		 * @busy_stats: Has enablement of engine stats tracking been
-		 * 		requested.
-		 */
-		bool busy_stats;
-		/**
-		 * @disable_busy_stats: Work item for busy stats disabling.
-		 *
-		 * Same as with @enable_busy_stats action, with the difference
-		 * that we delay it in case there are rapid enable-disable
-		 * actions, which can happen during tool startup (like perf
-		 * stat).
-		 */
-		struct delayed_work disable_busy_stats;
 	} pmu;
 
 	/*
Signature · 6 lines
2.14.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#1 lore raw

✓ Fi.CI.BAT: success for drm/i915/pmu: Fix PMU enable vs execlists tasklet race

== Series Details ==

== Series Details ==

Series: drm/i915/pmu: Fix PMU enable vs execlists tasklet race
URL : https://patchwork.freedesktop.org/series/37575/
State : success

== Summary ==

Series 37575v1 drm/i915/pmu: Fix PMU enable vs execlists tasklet race
https://patchwork.freedesktop.org/api/1.0/series/37575/revisions/1/mbox/

Test gem_exec_suspend:

        Subgroup basic-s4-devices:
                incomplete -> PASS       (fi-bdw-5557u) fdo#104162

fdo#104162 https://bugs.freedesktop.org/show_bug.cgi?id=104162

fi-bdw-5557u     total:288  pass:267  dwarn:0   dfail:0   fail:0   skip:21  time:419s
fi-bdw-gvtdvm    total:288  pass:264  dwarn:0   dfail:0   fail:0   skip:24  time:426s
fi-blb-e6850     total:288  pass:223  dwarn:1   dfail:0   fail:0   skip:64  time:373s
fi-bsw-n3050     total:288  pass:242  dwarn:0   dfail:0   fail:0   skip:46  time:489s
fi-bwr-2160      total:288  pass:183  dwarn:0   dfail:0   fail:0   skip:105 time:284s
fi-bxt-dsi       total:288  pass:258  dwarn:0   dfail:0   fail:0   skip:30  time:485s
fi-bxt-j4205     total:288  pass:259  dwarn:0   dfail:0   fail:0   skip:29  time:481s
fi-byt-j1900     total:288  pass:253  dwarn:0   dfail:0   fail:0   skip:35  time:468s
fi-byt-n2820     total:288  pass:249  dwarn:0   dfail:0   fail:0   skip:39  time:458s
fi-cfl-s2        total:288  pass:262  dwarn:0   dfail:0   fail:0   skip:26  time:571s
fi-elk-e7500     total:288  pass:229  dwarn:0   dfail:0   fail:0   skip:59  time:415s
fi-gdg-551       total:288  pass:179  dwarn:0   dfail:0   fail:1   skip:108 time:287s
fi-glk-1         total:288  pass:260  dwarn:0   dfail:0   fail:0   skip:28  time:511s
fi-hsw-4770      total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:391s
fi-hsw-4770r     total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:403s
fi-ilk-650       total:288  pass:228  dwarn:0   dfail:0   fail:0   skip:60  time:413s
fi-ivb-3520m     total:288  pass:259  dwarn:0   dfail:0   fail:0   skip:29  time:458s
fi-ivb-3770      total:288  pass:255  dwarn:0   dfail:0   fail:0   skip:33  time:415s
fi-kbl-7500u     total:288  pass:263  dwarn:1   dfail:0   fail:0   skip:24  time:455s
fi-kbl-7560u     total:288  pass:269  dwarn:0   dfail:0   fail:0   skip:19  time:500s
fi-kbl-7567u     total:288  pass:268  dwarn:0   dfail:0   fail:0   skip:20  time:455s
fi-kbl-r         total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:500s
fi-pnv-d510      total:288  pass:222  dwarn:1   dfail:0   fail:0   skip:65  time:575s
fi-skl-6260u     total:288  pass:268  dwarn:0   dfail:0   fail:0   skip:20  time:427s
fi-skl-6600u     total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:516s
fi-skl-6700hq    total:288  pass:262  dwarn:0   dfail:0   fail:0   skip:26  time:529s
fi-skl-6700k2    total:288  pass:264  dwarn:0   dfail:0   fail:0   skip:24  time:492s
fi-skl-6770hq    total:288  pass:268  dwarn:0   dfail:0   fail:0   skip:20  time:477s
fi-skl-guc       total:288  pass:260  dwarn:0   dfail:0   fail:0   skip:28  time:415s
fi-skl-gvtdvm    total:288  pass:265  dwarn:0   dfail:0   fail:0   skip:23  time:431s
fi-snb-2520m     total:3    pass:2    dwarn:0   dfail:0   fail:0   skip:0  
fi-snb-2600      total:288  pass:248  dwarn:0   dfail:0   fail:0   skip:40  time:396s

Blacklisted hosts:
fi-glk-dsi total:288 pass:258 dwarn:0 dfail:0 fail:0 skip:30 time:467s

bde4e003ce549b8225142843e0d5aee19dd37d13 drm-tip: 2018y-02m-02d-15h-03m-15s UTC integration manifest 0056845d822a drm/i915/pmu: Fix PMU enable vs execlists tasklet race

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_7867/issues.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#2 lore raw

✓ Fi.CI.IGT: success for drm/i915/pmu: Fix PMU enable vs execlists tasklet race

In reply to Tvrtko Ursulin
== Series Details ==

== Series Details ==

Series: drm/i915/pmu: Fix PMU enable vs execlists tasklet race
URL : https://patchwork.freedesktop.org/series/37575/
State : success

== Summary ==

Test perf:

        Subgroup enable-disable:
                fail       -> PASS       (shard-apl) fdo#103715
        Subgroup blocking:
                fail       -> PASS       (shard-hsw) fdo#102252
Test kms_cursor_crc:
        Subgroup cursor-128x128-suspend:
                pass       -> INCOMPLETE (shard-snb) fdo#103880
                incomplete -> PASS       (shard-hsw) fdo#103540
Test perf_pmu:
        Subgroup busy-double-start-vcs0:
                incomplete -> PASS       (shard-apl)
Test kms_frontbuffer_tracking:
        Subgroup fbc-1p-shrfb-fliptrack:
                pass       -> FAIL       (shard-apl) fdo#103167
Test kms_flip:
        Subgroup plain-flip-fb-recreate-interruptible:
                pass       -> FAIL       (shard-hsw) fdo#100368 +1
        Subgroup 2x-plain-flip-ts-check:
                fail       -> PASS       (shard-hsw)
Test kms_sysfs_edid_timing:
                pass       -> WARN       (shard-apl) fdo#100047

fdo#103715 https://bugs.freedesktop.org/show_bug.cgi?id=103715
fdo#102252 https://bugs.freedesktop.org/show_bug.cgi?id=102252
fdo#103880 https://bugs.freedesktop.org/show_bug.cgi?id=103880
fdo#103540 https://bugs.freedesktop.org/show_bug.cgi?id=103540
fdo#103167 https://bugs.freedesktop.org/show_bug.cgi?id=103167
fdo#100368 https://bugs.freedesktop.org/show_bug.cgi?id=100368
fdo#100047 https://bugs.freedesktop.org/show_bug.cgi?id=100047

shard-apl        total:2836 pass:1749 dwarn:1   dfail:0   fail:21  skip:1064 time:12388s
shard-hsw        total:2836 pass:1732 dwarn:1   dfail:0   fail:12  skip:1090 time:11553s
shard-snb        total:2800 pass:1309 dwarn:1   dfail:0   fail:10  skip:1479 time:6086s

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_7867/shards.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#3 lore raw
In reply to Tvrtko Ursulin

Quoting Tvrtko Ursulin (2018-02-02 18:38:16)

Quoted 36 lines

From: Tvrtko Ursulin <[email protected]>

Commit 99e48bf98dd0 ("drm/i915: Lock out execlist tasklet while peeking inside for busy-stats") added a tasklet_disable call in busy stats enabling, but we failed to understand that the PMU enable callback runs as an hard IRQ (IPI).

Consequence of this is that the PMU enable callback can interrupt the execlists tasklet, and will then deadlock when it calls intel_engine_stats_enable->tasklet_disable.

To fix this, I realized it is possible to move the engine stats enablement and disablement to PMU event init and destroy hooks. This allows for much simpler implementation since those hooks run in normal context (can sleep).

Fixes 99e48bf98dd0 ("drm/i915: Lock out execlist tasklet while peeking inside for busy-stats")

Testcase: igt/perf_pmu/enable-race-*
Cc: Chris Wilson <[email protected]>
Cc: Tvrtko Ursulin <[email protected]>
Cc: Jani Nikula <[email protected]>
Cc: Joonas Lahtinen <[email protected]>
Cc: Rodrigo Vivi <[email protected]>
Cc: [email protected]

---
 drivers/gpu/drm/i915/i915_pmu.c         | 95 +++++++++++----------------------
 drivers/gpu/drm/i915/intel_ringbuffer.h | 14 -----
 2 files changed, 31 insertions(+), 78 deletions(-)
drivers/gpu/drm/i915/i915_pmu.c +0 -0
index ecb0198bfb7a..e97860b44004 100644
@@ -287,7 +287,24 @@ static u64 count_interrupts(struct drm_i915_private *i915)
 

drivers/gpu/drm/i915/i915_pmu.c static u64 count_interrupts(struct drm_i915_private *i915) · line 297

Quoted 10 lines
drivers/gpu/drm/i915/i915_pmu.c +7 -0
 static void i915_pmu_event_destroy(struct perf_event *event)
 {
+       struct drm_i915_private *i915 =
+               container_of(event->pmu, typeof(*i915), pmu.base);
+       struct intel_engine_cs *engine;
+
        WARN_ON(event->parent);
+
+       if (!is_engine_event(event))
+               return;

I think engine_event_destroy() for symmetry with engine_event_init() ?

drivers/gpu/drm/i915/i915_pmu.c static u64 count_interrupts(struct drm_i915_private *i915) · line 305

Quoted 8 lines
drivers/gpu/drm/i915/i915_pmu.c +8 -0
+
+       engine = intel_engine_lookup_user(i915,
+                                         engine_event_class(event),
+                                         engine_event_instance(event));
+       if (WARN_ON_ONCE(!engine))
+               return;
+
+       if ((engine_event_sample(event) == I915_SAMPLE_BUSY) &&

Rogue ()

Quoted 18 lines
drivers/gpu/drm/i915/i915_pmu.c +4 -1
+           intel_engine_supports_stats(engine))
+               intel_disable_engine_stats(engine);
 }
 
 static int
@@ -340,13 +357,23 @@ static int engine_event_init(struct perf_event *event)
        struct drm_i915_private *i915 =
                container_of(event->pmu, typeof(*i915), pmu.base);
        struct intel_engine_cs *engine;
+       u8 sample;
+       int ret;
 
        engine = intel_engine_lookup_user(i915, engine_event_class(event),
                                          engine_event_instance(event));
        if (!engine)
                return -ENODEV;
 
-       return engine_event_status(engine, engine_event_sample(event));

drivers/gpu/drm/i915/i915_pmu.c static int engine_event_init(struct perf_event *event) · line 377

Quoted 10 lines
drivers/gpu/drm/i915/i915_pmu.c +9 -0
+       sample = engine_event_sample(event);
+       ret = engine_event_status(engine, sample);
+       if (ret)
+               return ret;
+
+       if ((sample == I915_SAMPLE_BUSY) && intel_engine_supports_stats(engine))
+               ret = intel_enable_engine_stats(engine);
+
+       return ret;
 }

Yeah, that is disgustingly simpler. On a second look nothing springs to mind why this shouldn't work. Will sleep on it.
-Chris
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

author maintainer
#4 lore raw

PATCH v2 1/1 drm/i915/pmu: Fix PMU enable vs execlists tasklet race

From: Tvrtko Ursulin <[email protected]>

Commit 99e48bf98dd0 ("drm/i915: Lock out execlist tasklet while peeking inside for busy-stats") added a tasklet_disable call in busy stats enabling, but we failed to understand that the PMU enable callback runs as an hard IRQ (IPI).

Consequence of this is that the PMU enable callback can interrupt the execlists tasklet, and will then deadlock when it calls intel_engine_stats_enable->tasklet_disable.

To fix this, I realized it is possible to move the engine stats enablement and disablement to PMU event init and destroy hooks. This allows for much simpler implementation since those hooks run in normal context (can sleep).

v2: Extract engine_event_destroy. (Chris Wilson)

Fixes 99e48bf98dd0 ("drm/i915: Lock out execlist tasklet while peeking inside for busy-stats")

Testcase: igt/perf_pmu/enable-race-*
Cc: Chris Wilson <[email protected]>
Cc: Tvrtko Ursulin <[email protected]>
Cc: Jani Nikula <[email protected]>
Cc: Joonas Lahtinen <[email protected]>
Cc: Rodrigo Vivi <[email protected]>
Cc: [email protected]

---
 drivers/gpu/drm/i915/i915_pmu.c         | 98 ++++++++++++---------------------
 drivers/gpu/drm/i915/intel_ringbuffer.h | 14 -----
 2 files changed, 34 insertions(+), 78 deletions(-)
drivers/gpu/drm/i915/i915_pmu.c +34 -64
index ecb0198bfb7a..1c440460255d 100644
@@ -285,9 +285,29 @@ static u64 count_interrupts(struct drm_i915_private *i915)
 	return sum;
 }
 
+static void engine_event_destroy(struct perf_event *event)
+{
+	struct drm_i915_private *i915 =
+		container_of(event->pmu, typeof(*i915), pmu.base);
+	struct intel_engine_cs *engine;
+
+	engine = intel_engine_lookup_user(i915,
+					  engine_event_class(event),
+					  engine_event_instance(event));
+	if (WARN_ON_ONCE(!engine))
+		return;
+
+	if (engine_event_sample(event) == I915_SAMPLE_BUSY &&
+	    intel_engine_supports_stats(engine))
+		intel_disable_engine_stats(engine);
+}
+
 static void i915_pmu_event_destroy(struct perf_event *event)
 {
 	WARN_ON(event->parent);
+
+	if (is_engine_event(event))
+		engine_event_destroy(event);
 }
 
 static int
@@ -340,13 +360,23 @@ static int engine_event_init(struct perf_event *event)
 	struct drm_i915_private *i915 =
 		container_of(event->pmu, typeof(*i915), pmu.base);
 	struct intel_engine_cs *engine;
+	u8 sample;
+	int ret;
 
 	engine = intel_engine_lookup_user(i915, engine_event_class(event),
 					  engine_event_instance(event));
 	if (!engine)
 		return -ENODEV;
 
-	return engine_event_status(engine, engine_event_sample(event));
+	sample = engine_event_sample(event);
+	ret = engine_event_status(engine, sample);
+	if (ret)
+		return ret;
+
+	if (sample == I915_SAMPLE_BUSY && intel_engine_supports_stats(engine))
+		ret = intel_enable_engine_stats(engine);
+
+	return ret;
 }
 
 static int i915_pmu_event_init(struct perf_event *event)
@@ -402,7 +432,7 @@ static u64 __i915_pmu_event_read(struct perf_event *event)
 		if (WARN_ON_ONCE(!engine)) {
 			/* Do nothing */
 		} else if (sample == I915_SAMPLE_BUSY &&
-			   engine->pmu.busy_stats) {
+			   intel_engine_supports_stats(engine)) {
 			val = ktime_to_ns(intel_engine_get_busy_time(engine));
 		} else {
 			val = engine->pmu.sample[sample].cur;
@@ -457,12 +487,6 @@ static void i915_pmu_event_read(struct perf_event *event)
 	local64_add(new - prev, &event->count);
 }
 
-static bool engine_needs_busy_stats(struct intel_engine_cs *engine)
-{
-	return intel_engine_supports_stats(engine) &&
-	       (engine->pmu.enable & BIT(I915_SAMPLE_BUSY));
-}
-
 static void i915_pmu_enable(struct perf_event *event)
 {
 	struct drm_i915_private *i915 =
@@ -502,21 +526,7 @@ static void i915_pmu_enable(struct perf_event *event)
 
 		GEM_BUG_ON(sample >= I915_PMU_SAMPLE_BITS);
 		GEM_BUG_ON(engine->pmu.enable_count[sample] == ~0);
-		if (engine->pmu.enable_count[sample]++ == 0) {
-			/*
-			 * Enable engine busy stats tracking if needed or
-			 * alternatively cancel the scheduled disable.
-			 *
-			 * If the delayed disable was pending, cancel it and
-			 * in this case do not enable since it already is.
-			 */
-			if (engine_needs_busy_stats(engine) &&
-			    !engine->pmu.busy_stats) {
-				engine->pmu.busy_stats = true;
-				if (!cancel_delayed_work(&engine->pmu.disable_busy_stats))
-					intel_enable_engine_stats(engine);
-			}
-		}
+		engine->pmu.enable_count[sample]++;
 	}
 
 	/*
@@ -529,14 +539,6 @@ static void i915_pmu_enable(struct perf_event *event)
 	spin_unlock_irqrestore(&i915->pmu.lock, flags);
 }
 
-static void __disable_busy_stats(struct work_struct *work)
-{
-	struct intel_engine_cs *engine =
-	       container_of(work, typeof(*engine), pmu.disable_busy_stats.work);
-
-	intel_disable_engine_stats(engine);
-}
-
 static void i915_pmu_disable(struct perf_event *event)
 {
 	struct drm_i915_private *i915 =
@@ -560,26 +562,8 @@ 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 (--engine->pmu.enable_count[sample] == 0) {
+		if (--engine->pmu.enable_count[sample] == 0)
 			engine->pmu.enable &= ~BIT(sample);
-			if (!engine_needs_busy_stats(engine) &&
-			    engine->pmu.busy_stats) {
-				engine->pmu.busy_stats = false;
-				/*
-				 * We request a delayed disable to handle the
-				 * rapid on/off cycles on events, which can
-				 * happen when tools like perf stat start, in a
-				 * nicer way.
-				 *
-				 * In addition, this also helps with busy stats
-				 * accuracy with background CPU offline/online
-				 * migration events.
-				 */
-				queue_delayed_work(system_wq,
-						   &engine->pmu.disable_busy_stats,
-						   round_jiffies_up_relative(HZ));
-			}
-		}
 	}
 
 	GEM_BUG_ON(bit >= I915_PMU_MASK_BITS);
@@ -956,8 +940,6 @@ static void i915_pmu_unregister_cpuhp_state(struct drm_i915_private *i915)
 
 void i915_pmu_register(struct drm_i915_private *i915)
 {
-	struct intel_engine_cs *engine;
-	enum intel_engine_id id;
 	int ret;
 
 	if (INTEL_GEN(i915) <= 2) {
@@ -985,10 +967,6 @@ void i915_pmu_register(struct drm_i915_private *i915)
 	hrtimer_init(&i915->pmu.timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	i915->pmu.timer.function = i915_sample;
 
-	for_each_engine(engine, i915, id)
-		INIT_DELAYED_WORK(&engine->pmu.disable_busy_stats,
-				  __disable_busy_stats);
-
 	ret = perf_pmu_register(&i915->pmu.base, "i915", -1);
 	if (ret)
 		goto err;
@@ -1009,9 +987,6 @@ void i915_pmu_register(struct drm_i915_private *i915)
 
 void i915_pmu_unregister(struct drm_i915_private *i915)
 {
-	struct intel_engine_cs *engine;
-	enum intel_engine_id id;
-
 	if (!i915->pmu.base.event_init)
 		return;
 
@@ -1019,11 +994,6 @@ void i915_pmu_unregister(struct drm_i915_private *i915)
 
 	hrtimer_cancel(&i915->pmu.timer);
 
-	for_each_engine(engine, i915, id) {
-		GEM_BUG_ON(engine->pmu.busy_stats);
-		flush_delayed_work(&engine->pmu.disable_busy_stats);
-	}
-
 	i915_pmu_unregister_cpuhp_state(i915);
 
 	perf_pmu_unregister(&i915->pmu.base);
drivers/gpu/drm/i915/intel_ringbuffer.h +0 -14
index c5ff203e42d6..a0e7a6c2a57c 100644
@@ -366,20 +366,6 @@ struct intel_engine_cs {
 		 */
 #define I915_ENGINE_SAMPLE_MAX (I915_SAMPLE_SEMA + 1)
 		struct i915_pmu_sample sample[I915_ENGINE_SAMPLE_MAX];
-		/**
-		 * @busy_stats: Has enablement of engine stats tracking been
-		 * 		requested.
-		 */
-		bool busy_stats;
-		/**
-		 * @disable_busy_stats: Work item for busy stats disabling.
-		 *
-		 * Same as with @enable_busy_stats action, with the difference
-		 * that we delay it in case there are rapid enable-disable
-		 * actions, which can happen during tool startup (like perf
-		 * stat).
-		 */
-		struct delayed_work disable_busy_stats;
 	} pmu;
 
 	/*
Signature · 6 lines
2.14.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Reviewed-by on v2 1/1
#5 lore raw

Quoting Tvrtko Ursulin (2018-02-05 09:34:48)

Quoted 20 lines

From: Tvrtko Ursulin <[email protected]>

Commit 99e48bf98dd0 ("drm/i915: Lock out execlist tasklet while peeking inside for busy-stats") added a tasklet_disable call in busy stats enabling, but we failed to understand that the PMU enable callback runs as an hard IRQ (IPI).

Consequence of this is that the PMU enable callback can interrupt the execlists tasklet, and will then deadlock when it calls intel_engine_stats_enable->tasklet_disable.

To fix this, I realized it is possible to move the engine stats enablement and disablement to PMU event init and destroy hooks. This allows for much simpler implementation since those hooks run in normal context (can sleep).

v2: Extract engine_event_destroy. (Chris Wilson)

Fixes 99e48bf98dd0 ("drm/i915: Lock out execlist tasklet while peeking inside for busy-stats")
Quoted 7 lines

Testcase: igt/perf_pmu/enable-race-*
Cc: Chris Wilson <[email protected]>
Cc: Tvrtko Ursulin <[email protected]>
Cc: Jani Nikula <[email protected]>
Cc: Joonas Lahtinen <[email protected]>
Cc: Rodrigo Vivi <[email protected]>
Cc: [email protected]

I didn't have any nightmares about this,

-Chris
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#6 lore raw

✓ Fi.CI.BAT: success for drm/i915/pmu: Fix PMU enable vs execlists tasklet race (rev2)

In reply to Tvrtko Ursulin
== Series Details ==

== Series Details ==

Series: drm/i915/pmu: Fix PMU enable vs execlists tasklet race (rev2)
URL : https://patchwork.freedesktop.org/series/37575/
State : success

== Summary ==

Series 37575v2 drm/i915/pmu: Fix PMU enable vs execlists tasklet race
https://patchwork.freedesktop.org/api/1.0/series/37575/revisions/2/mbox/

Test gem_mmap_gtt:

        Subgroup basic-small-bo-tiledx:
                pass       -> FAIL       (fi-gdg-551) fdo#102575
Test kms_chamelium:
        Subgroup dp-crc-fast:
                pass       -> FAIL       (fi-kbl-7500u) fdo#103841

fdo#102575 https://bugs.freedesktop.org/show_bug.cgi?id=102575
fdo#103841 https://bugs.freedesktop.org/show_bug.cgi?id=103841

fi-bdw-5557u     total:288  pass:267  dwarn:0   dfail:0   fail:0   skip:21  time:416s
fi-bdw-gvtdvm    total:288  pass:264  dwarn:0   dfail:0   fail:0   skip:24  time:420s
fi-blb-e6850     total:288  pass:223  dwarn:1   dfail:0   fail:0   skip:64  time:369s
fi-bsw-n3050     total:288  pass:242  dwarn:0   dfail:0   fail:0   skip:46  time:483s
fi-bwr-2160      total:288  pass:183  dwarn:0   dfail:0   fail:0   skip:105 time:280s
fi-bxt-dsi       total:288  pass:258  dwarn:0   dfail:0   fail:0   skip:30  time:480s
fi-bxt-j4205     total:288  pass:259  dwarn:0   dfail:0   fail:0   skip:29  time:483s
fi-byt-j1900     total:288  pass:253  dwarn:0   dfail:0   fail:0   skip:35  time:464s
fi-byt-n2820     total:288  pass:249  dwarn:0   dfail:0   fail:0   skip:39  time:451s
fi-cfl-s2        total:288  pass:262  dwarn:0   dfail:0   fail:0   skip:26  time:563s
fi-elk-e7500     total:288  pass:229  dwarn:0   dfail:0   fail:0   skip:59  time:402s
fi-gdg-551       total:288  pass:179  dwarn:0   dfail:0   fail:1   skip:108 time:279s
fi-glk-1         total:288  pass:260  dwarn:0   dfail:0   fail:0   skip:28  time:509s
fi-hsw-4770      total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:390s
fi-hsw-4770r     total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:397s
fi-ilk-650       total:288  pass:228  dwarn:0   dfail:0   fail:0   skip:60  time:407s
fi-ivb-3520m     total:288  pass:259  dwarn:0   dfail:0   fail:0   skip:29  time:458s
fi-ivb-3770      total:288  pass:255  dwarn:0   dfail:0   fail:0   skip:33  time:413s
fi-kbl-7500u     total:288  pass:262  dwarn:1   dfail:0   fail:1   skip:24  time:456s
fi-kbl-7560u     total:288  pass:269  dwarn:0   dfail:0   fail:0   skip:19  time:495s
fi-kbl-7567u     total:288  pass:268  dwarn:0   dfail:0   fail:0   skip:20  time:450s
fi-kbl-r         total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:497s
fi-pnv-d510      total:288  pass:222  dwarn:1   dfail:0   fail:0   skip:65  time:575s
fi-skl-6260u     total:288  pass:268  dwarn:0   dfail:0   fail:0   skip:20  time:429s
fi-skl-6600u     total:288  pass:261  dwarn:0   dfail:0   fail:0   skip:27  time:503s
fi-skl-6700hq    total:288  pass:262  dwarn:0   dfail:0   fail:0   skip:26  time:525s
fi-skl-6700k2    total:288  pass:264  dwarn:0   dfail:0   fail:0   skip:24  time:482s
fi-skl-6770hq    total:288  pass:268  dwarn:0   dfail:0   fail:0   skip:20  time:490s
fi-skl-guc       total:288  pass:260  dwarn:0   dfail:0   fail:0   skip:28  time:414s
fi-skl-gvtdvm    total:288  pass:265  dwarn:0   dfail:0   fail:0   skip:23  time:430s
fi-snb-2520m     total:288  pass:248  dwarn:0   dfail:0   fail:0   skip:40  time:517s
fi-snb-2600      total:288  pass:248  dwarn:0   dfail:0   fail:0   skip:40  time:395s

Blacklisted hosts:
fi-glk-dsi total:288 pass:258 dwarn:0 dfail:0 fail:0 skip:30 time:464s

2e76a2952923eba64c4f9baf461613bc42ee997a drm-tip: 2018y-02m-02d-20h-33m-12s UTC integration manifest c0ee935c8efd drm/i915/pmu: Fix PMU enable vs execlists tasklet race

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_7883/issues.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#7 lore raw

✓ Fi.CI.IGT: success for drm/i915/pmu: Fix PMU enable vs execlists tasklet race (rev2)

In reply to Tvrtko Ursulin
== Series Details ==

== Series Details ==

Series: drm/i915/pmu: Fix PMU enable vs execlists tasklet race (rev2)
URL : https://patchwork.freedesktop.org/series/37575/
State : success

== Summary ==

Test kms_cursor_legacy:

        Subgroup flip-vs-cursor-atomic:
                fail       -> PASS       (shard-hsw) fdo#102670 +1
Test gem_eio:
        Subgroup in-flight-contexts:
                dmesg-warn -> PASS       (shard-snb) fdo#104058 +1
Test perf:
        Subgroup oa-exponents:
                pass       -> FAIL       (shard-apl) fdo#102254
Test kms_flip:
        Subgroup 2x-plain-flip-ts-check:
                fail       -> PASS       (shard-hsw) fdo#100368
        Subgroup 2x-flip-vs-expired-vblank-interruptible:
                fail       -> PASS       (shard-hsw) fdo#102887

fdo#102670 https://bugs.freedesktop.org/show_bug.cgi?id=102670
fdo#104058 https://bugs.freedesktop.org/show_bug.cgi?id=104058
fdo#102254 https://bugs.freedesktop.org/show_bug.cgi?id=102254
fdo#100368 https://bugs.freedesktop.org/show_bug.cgi?id=100368
fdo#102887 https://bugs.freedesktop.org/show_bug.cgi?id=102887

shard-apl        total:2836 pass:1750 dwarn:1   dfail:0   fail:21  skip:1064 time:12561s
shard-hsw        total:2836 pass:1733 dwarn:1   dfail:0   fail:11  skip:1090 time:11703s
shard-snb        total:2836 pass:1327 dwarn:2   dfail:0   fail:10  skip:1497 time:6407s

Blacklisted hosts:
shard-kbl total:2836 pass:1867 dwarn:5 dfail:0 fail:22 skip:942 time:9368s

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_7883/shards.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx