[igt-dev] [PATCH i-g-t] i915/gem_ctx_persistence: Exercise cleanup after disabling heartbeats

[Chris Wilson]

We expose the heartbeat interval on each engine, allowing the sysadim to
disable them if they prefer avoiding any interruption for their GPU
tasks. A caveat to allowing the contexts to run without checks is that
we require such contexts to be non-persistent and so cleaned up on
closure (including abnormal process termination). However, we also need
to flush any persistent contexts that are still inflight at that time,
lest they continue to run unchecked.

Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
---
 tests/i915/gem_ctx_persistence.c | 92 ++++++++++++++++++++++++++++++++
 1 file changed, 92 insertions(+)

diff --git a/tests/i915/gem_ctx_persistence.c b/tests/i915/gem_ctx_persistence.c
index e73a3e6a0..ca676d845 100644
--- a/tests/i915/gem_ctx_persistence.c
+++ b/tests/i915/gem_ctx_persistence.c
@@ -426,6 +426,87 @@ static void test_nohangcheck_hang(int i915)
 	close(dir);
 }
 
+static bool set_heartbeat(int i915, const char *name, unsigned int value)
+{
+	return gem_engine_property_printf(i915, name,
+					  "heartbeat_interval_ms",
+					  "%d", value) > 0;
+}
+
+static void test_noheartbeat_many(int i915, int count, unsigned int flags)
+{
+	cleanup(i915);
+
+	/*
+	 * If the user disables the heartbeat, after leaving behind
+	 * a number of long running *persistent* contexts, check they get
+	 * cleaned up.
+	 */
+
+	for_each_engine(e, i915) {
+		igt_spin_t *spin[count];
+
+		if (!set_heartbeat(i915, e->full_name, 100))
+			continue;
+
+		for (int n = 0; n < ARRAY_SIZE(spin); n++) {
+			uint32_t ctx;
+
+			ctx = gem_context_create(i915);
+			spin[n] = igt_spin_new(i915, ctx, .engine = eb_ring(e),
+					       .flags = (IGT_SPIN_FENCE_OUT |
+							 flags));
+			gem_context_destroy(i915, ctx);
+		}
+
+		if (set_heartbeat(i915, e->full_name, 0)) {
+			for (int n = 0; n < ARRAY_SIZE(spin); n++) {
+				igt_assert_eq(wait_for_status(i915, spin[n]->out_fence, reset_timeout_ms),
+					      -EIO);
+			}
+		}
+
+		for (int n = 0; n < ARRAY_SIZE(spin); n++)
+			igt_spin_free(i915, spin[n]);
+
+		set_heartbeat(i915, e->full_name, 2500);
+		cleanup(i915);
+	}
+}
+
+static void test_noheartbeat_close(int i915, unsigned int flags)
+{
+	cleanup(i915);
+
+	/*
+	 * Check that non-persistent contexts are also cleaned up if we
+	 * close the context while they are active, but the engine's
+	 * heartbeat has already been disabled.
+	 */
+
+	for_each_engine(e, i915) {
+		igt_spin_t *spin;
+		uint32_t ctx;
+
+		if (!set_heartbeat(i915, e->full_name, 0))
+			continue;
+
+		ctx = gem_context_create(i915);
+		gem_context_set_persistence(i915, ctx, false);
+		spin = igt_spin_new(i915, ctx, .engine = eb_ring(e),
+				    .flags = (IGT_SPIN_FENCE_OUT | flags));
+		gem_context_destroy(i915, ctx);
+
+		set_heartbeat(i915, e->full_name, 2500);
+
+		igt_assert_eq(wait_for_status(i915, spin->out_fence, reset_timeout_ms),
+			      -EIO);
+
+		igt_spin_free(i915, spin);
+		cleanup(i915);
+	}
+}
+
 static void test_nonpersistent_file(int i915)
 {
 	int debugfs = i915;
@@ -1157,6 +1238,17 @@ igt_main
 	igt_subtest("hang")
 		test_nohangcheck_hang(i915);
 
+	igt_subtest("heartbeat-stop")
+		test_noheartbeat_many(i915, 1, 0);
+	igt_subtest("heartbeat-hang")
+		test_noheartbeat_many(i915, 1, IGT_SPIN_NO_PREEMPTION);
+	igt_subtest("heartbeat-many")
+		test_noheartbeat_many(i915, 16, 0);
+	igt_subtest("heartbeat-close")
+		test_noheartbeat_close(i915, 0);
+	igt_subtest("heartbeat-hostile")
+		test_noheartbeat_close(i915, IGT_SPIN_NO_PREEMPTION);
+
 	igt_subtest_group {
 		igt_fixture
 			gem_require_contexts(i915);
-- 
2.28.0