drm/i915: Look for an active kernel context before switching

We were not very carefully checking to see if an older request on the
engine was an earlier switch-to-kernel-context before deciding to emit a
new switch. The end result would be that we could get into a permanent
loop of trying to emit a new request to perform the switch simply to
flush the existing switch.

What we need is a means of tracking the completion of each timeline
versus the kernel context, that is to detect if a more recent request
has been submitted that would result in a switch away from the kernel
context. To realise this, we need only to look in our syncmap on the
kernel context and check that we have synchronized against all active
rings.

v2: Since all ringbuffer clients currently share the same timeline, we do
have to use the gem_context to distinguish clients.

As a bonus, include all the tracing used to debug the death inside
suspend.

v3: Test, test, test. Construct a selftest to exercise and assert the
expected behaviour that multiple switch-to-contexts do not emit
redundant requests.

Reported-by: Mika Kuoppala <mika.kuoppala@intel.com>
Fixes: a89d1f921c15 ("drm/i915: Split i915_gem_timeline into individual timelines")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180524081135.15278-1-chris@chris-wilson.co.uk

6 files changed, 231 insertions, 15 deletions

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 03874b50ada9..05f44ca35a06 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3703,6 +3703,9 @@ static int wait_for_engines(struct drm_i915_private *i915)
 
 int i915_gem_wait_for_idle(struct drm_i915_private *i915, unsigned int flags)
 {
+	GEM_TRACE("flags=%x (%s)\n",
+		  flags, flags & I915_WAIT_LOCKED ? "locked" : "unlocked");
+
 	/* If the device is asleep, we have no requests outstanding */
 	if (!READ_ONCE(i915->gt.awake))
 		return 0;
@@ -3719,6 +3722,7 @@ int i915_gem_wait_for_idle(struct drm_i915_private *i915, unsigned int flags)
 				return err;
 		}
 		i915_retire_requests(i915);
+		GEM_BUG_ON(i915->gt.active_requests);
 
 		return wait_for_engines(i915);
 	} else {
@@ -4901,6 +4905,7 @@ static void assert_kernel_context_is_current(struct drm_i915_private *i915)
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 
+	GEM_BUG_ON(i915->gt.active_requests);
 	for_each_engine(engine, i915, id) {
 		GEM_BUG_ON(__i915_gem_active_peek(&engine->timeline.last_request));
 		GEM_BUG_ON(engine->last_retired_context->gem_context != kctx);
@@ -4932,6 +4937,8 @@ int i915_gem_suspend(struct drm_i915_private *dev_priv)
 	struct drm_device *dev = &dev_priv->drm;
 	int ret;
 
+	GEM_TRACE("\n");
+
 	intel_runtime_pm_get(dev_priv);
 	intel_suspend_gt_powersave(dev_priv);
 
diff --git a/drivers/gpu/drm/i915/i915_gem.h b/drivers/gpu/drm/i915/i915_gem.h
index 5bf24cfc218c..62ee4e385365 100644
--- a/drivers/gpu/drm/i915/i915_gem.h
+++ b/drivers/gpu/drm/i915/i915_gem.h
@@ -63,9 +63,12 @@ struct drm_i915_private;
 #if IS_ENABLED(CONFIG_DRM_I915_TRACE_GEM)
 #define GEM_TRACE(...) trace_printk(__VA_ARGS__)
 #define GEM_TRACE_DUMP() ftrace_dump(DUMP_ALL)
+#define GEM_TRACE_DUMP_ON(expr) \
+	do { if (expr) ftrace_dump(DUMP_ALL); } while (0)
 #else
 #define GEM_TRACE(...) do { } while (0)
 #define GEM_TRACE_DUMP() do { } while (0)
+#define GEM_TRACE_DUMP_ON(expr) BUILD_BUG_ON_INVALID(expr)
 #endif
 
 #define I915_NUM_ENGINES 8
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index b69b18ef8120..45393f6e0208 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -576,30 +576,72 @@ last_request_on_engine(struct i915_timeline *timeline,
 {
 	struct i915_request *rq;
 
-	if (timeline == &engine->timeline)
-		return NULL;
+	GEM_BUG_ON(timeline == &engine->timeline);
 
 	rq = i915_gem_active_raw(&timeline->last_request,
 				 &engine->i915->drm.struct_mutex);
-	if (rq && rq->engine == engine)
+	if (rq && rq->engine == engine) {
+		GEM_TRACE("last request for %s on engine %s: %llx:%d\n",
+			  timeline->name, engine->name,
+			  rq->fence.context, rq->fence.seqno);
+		GEM_BUG_ON(rq->timeline != timeline);
 		return rq;
+	}
 
 	return NULL;
 }
 
-static bool engine_has_idle_kernel_context(struct intel_engine_cs *engine)
+static bool engine_has_kernel_context_barrier(struct intel_engine_cs *engine)
 {
-	struct list_head * const active_rings = &engine->i915->gt.active_rings;
+	struct drm_i915_private *i915 = engine->i915;
+	const struct intel_context * const ce =
+		to_intel_context(i915->kernel_context, engine);
+	struct i915_timeline *barrier = ce->ring->timeline;
 	struct intel_ring *ring;
+	bool any_active = false;
 
-	lockdep_assert_held(&engine->i915->drm.struct_mutex);
+	lockdep_assert_held(&i915->drm.struct_mutex);
+	list_for_each_entry(ring, &i915->gt.active_rings, active_link) {
+		struct i915_request *rq;
+
+		rq = last_request_on_engine(ring->timeline, engine);
+		if (!rq)
+			continue;
 
-	list_for_each_entry(ring, active_rings, active_link) {
-		if (last_request_on_engine(ring->timeline, engine))
+		any_active = true;
+
+		if (rq->gem_context == i915->kernel_context)
+			continue;
+
+		/*
+		 * Was this request submitted after the previous
+		 * switch-to-kernel-context?
+		 */
+		if (!i915_timeline_sync_is_later(barrier, &rq->fence)) {
+			GEM_TRACE("%s needs barrier for %llx:%d\n",
+				  ring->timeline->name,
+				  rq->fence.context,
+				  rq->fence.seqno);
 			return false;
+		}
+
+		GEM_TRACE("%s has barrier after %llx:%d\n",
+			  ring->timeline->name,
+			  rq->fence.context,
+			  rq->fence.seqno);
 	}
 
-	return intel_engine_has_kernel_context(engine);
+	/*
+	 * If any other timeline was still active and behind the last barrier,
+	 * then our last switch-to-kernel-context must still be queued and
+	 * will run last (leaving the engine in the kernel context when it
+	 * eventually idles).
+	 */
+	if (any_active)
+		return true;
+
+	/* The engine is idle; check that it is idling in the kernel context. */
+	return engine->last_retired_context == ce;
 }
 
 int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915)
@@ -607,7 +649,10 @@ int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915)
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 
+	GEM_TRACE("\n");
+
 	lockdep_assert_held(&i915->drm.struct_mutex);
+	GEM_BUG_ON(!i915->kernel_context);
 
 	i915_retire_requests(i915);
 
@@ -615,9 +660,12 @@ int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915)
 		struct intel_ring *ring;
 		struct i915_request *rq;
 
-		if (engine_has_idle_kernel_context(engine))
+		GEM_BUG_ON(!to_intel_context(i915->kernel_context, engine));
+		if (engine_has_kernel_context_barrier(engine))
 			continue;
 
+		GEM_TRACE("emit barrier on %s\n", engine->name);
+
 		rq = i915_request_alloc(engine, i915->kernel_context);
 		if (IS_ERR(rq))
 			return PTR_ERR(rq);
@@ -627,10 +675,20 @@ int i915_gem_switch_to_kernel_context(struct drm_i915_private *i915)
 			struct i915_request *prev;
 
 			prev = last_request_on_engine(ring->timeline, engine);
-			if (prev)
-				i915_sw_fence_await_sw_fence_gfp(&rq->submit,
-								 &prev->submit,
-								 I915_FENCE_GFP);
+			if (!prev)
+				continue;
+
+			if (prev->gem_context == i915->kernel_context)
+				continue;
+
+			GEM_TRACE("add barrier on %s for %llx:%d\n",
+				  engine->name,
+				  prev->fence.context,
+				  prev->fence.seqno);
+			i915_sw_fence_await_sw_fence_gfp(&rq->submit,
+							 &prev->submit,
+							 I915_FENCE_GFP);
+			i915_timeline_sync_set(rq->timeline, &prev->fence);
 		}
 
 		/*
diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index fc499bcbd105..f187250e60c6 100644
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -320,6 +320,7 @@ static void advance_ring(struct i915_request *request)
 		 * is just about to be. Either works, if we miss the last two
 		 * noops - they are safe to be replayed on a reset.
 		 */
+		GEM_TRACE("marking %s as inactive\n", ring->timeline->name);
 		tail = READ_ONCE(request->tail);
 		list_del(&ring->active_link);
 	} else {
@@ -1095,8 +1096,10 @@ void __i915_request_add(struct i915_request *request, bool flush_caches)
 	i915_gem_active_set(&timeline->last_request, request);
 
 	list_add_tail(&request->ring_link, &ring->request_list);
-	if (list_is_first(&request->ring_link, &ring->request_list))
+	if (list_is_first(&request->ring_link, &ring->request_list)) {
+		GEM_TRACE("marking %s as active\n", ring->timeline->name);
 		list_add(&ring->active_link, &request->i915->gt.active_rings);
+	}
 	request->emitted_jiffies = jiffies;
 
 	/*
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/selftests/i915_gem_context.c
index ddb03f009232..b39392a00a6f 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_context.c
@@ -26,6 +26,7 @@
 #include "igt_flush_test.h"
 
 #include "mock_drm.h"
+#include "mock_gem_device.h"
 #include "huge_gem_object.h"
 
 #define DW_PER_PAGE (PAGE_SIZE / sizeof(u32))
@@ -420,6 +421,130 @@ out_unlock:
 	return err;
 }
 
+static __maybe_unused const char *
+__engine_name(struct drm_i915_private *i915, unsigned int engines)
+{
+	struct intel_engine_cs *engine;
+	unsigned int tmp;
+
+	if (engines == ALL_ENGINES)
+		return "all";
+
+	for_each_engine_masked(engine, i915, engines, tmp)
+		return engine->name;
+
+	return "none";
+}
+
+static int __igt_switch_to_kernel_context(struct drm_i915_private *i915,
+					  struct i915_gem_context *ctx,
+					  unsigned int engines)
+{
+	struct intel_engine_cs *engine;
+	unsigned int tmp;
+	int err;
+
+	GEM_TRACE("Testing %s\n", __engine_name(i915, engines));
+	for_each_engine_masked(engine, i915, engines, tmp) {
+		struct i915_request *rq;
+
+		rq = i915_request_alloc(engine, ctx);
+		if (IS_ERR(rq))
+			return PTR_ERR(rq);
+
+		i915_request_add(rq);
+	}
+
+	err = i915_gem_switch_to_kernel_context(i915);
+	if (err)
+		return err;
+
+	for_each_engine_masked(engine, i915, engines, tmp) {
+		if (!engine_has_kernel_context_barrier(engine)) {
+			pr_err("kernel context not last on engine %s!\n",
+			       engine->name);
+			return -EINVAL;
+		}
+	}
+
+	err = i915_gem_wait_for_idle(i915, I915_WAIT_LOCKED);
+	if (err)
+		return err;
+
+	GEM_BUG_ON(i915->gt.active_requests);
+	for_each_engine_masked(engine, i915, engines, tmp) {
+		if (engine->last_retired_context->gem_context != i915->kernel_context) {
+			pr_err("engine %s not idling in kernel context!\n",
+			       engine->name);
+			return -EINVAL;
+		}
+	}
+
+	err = i915_gem_switch_to_kernel_context(i915);
+	if (err)
+		return err;
+
+	if (i915->gt.active_requests) {
+		pr_err("switch-to-kernel-context emitted %d requests even though it should already be idling in the kernel context\n",
+		       i915->gt.active_requests);
+		return -EINVAL;
+	}
+
+	for_each_engine_masked(engine, i915, engines, tmp) {
+		if (!intel_engine_has_kernel_context(engine)) {
+			pr_err("kernel context not last on engine %s!\n",
+			       engine->name);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int igt_switch_to_kernel_context(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct intel_engine_cs *engine;
+	struct i915_gem_context *ctx;
+	enum intel_engine_id id;
+	int err;
+
+	/*
+	 * A core premise of switching to the kernel context is that
+	 * if an engine is already idling in the kernel context, we
+	 * do not emit another request and wake it up. The other being
+	 * that we do indeed end up idling in the kernel context.
+	 */
+
+	mutex_lock(&i915->drm.struct_mutex);
+	ctx = kernel_context(i915);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto out_unlock;
+	}
+
+	/* First check idling each individual engine */
+	for_each_engine(engine, i915, id) {
+		err = __igt_switch_to_kernel_context(i915, ctx, BIT(id));
+		if (err)
+			goto out_unlock;
+	}
+
+	/* Now en masse */
+	err = __igt_switch_to_kernel_context(i915, ctx, ALL_ENGINES);
+	if (err)
+		goto out_unlock;
+
+out_unlock:
+	GEM_TRACE_DUMP_ON(err);
+	if (igt_flush_test(i915, I915_WAIT_LOCKED))
+		err = -EIO;
+	mutex_unlock(&i915->drm.struct_mutex);
+
+	kernel_context_close(ctx);
+	return err;
+}
+
 static int fake_aliasing_ppgtt_enable(struct drm_i915_private *i915)
 {
 	struct drm_i915_gem_object *obj;
@@ -447,9 +572,28 @@ static void fake_aliasing_ppgtt_disable(struct drm_i915_private *i915)
 	i915_gem_fini_aliasing_ppgtt(i915);
 }
 
+int i915_gem_context_mock_selftests(void)
+{
+	static const struct i915_subtest tests[] = {
+		SUBTEST(igt_switch_to_kernel_context),
+	};
+	struct drm_i915_private *i915;
+	int err;
+
+	i915 = mock_gem_device();
+	if (!i915)
+		return -ENOMEM;
+
+	err = i915_subtests(tests, i915);
+
+	drm_dev_unref(&i915->drm);
+	return err;
+}
+
 int i915_gem_context_live_selftests(struct drm_i915_private *dev_priv)
 {
 	static const struct i915_subtest tests[] = {
+		SUBTEST(igt_switch_to_kernel_context),
 		SUBTEST(igt_ctx_exec),
 	};
 	bool fake_alias = false;
diff --git a/drivers/gpu/drm/i915/selftests/i915_mock_selftests.h b/drivers/gpu/drm/i915/selftests/i915_mock_selftests.h
index d16d74178e9d..1b70208eeea7 100644
--- a/drivers/gpu/drm/i915/selftests/i915_mock_selftests.h
+++ b/drivers/gpu/drm/i915/selftests/i915_mock_selftests.h
@@ -24,3 +24,4 @@ selftest(vma, i915_vma_mock_selftests)
 selftest(evict, i915_gem_evict_mock_selftests)
 selftest(gtt, i915_gem_gtt_mock_selftests)
 selftest(hugepages, i915_gem_huge_page_mock_selftests)
+selftest(contexts, i915_gem_context_mock_selftests)