Fix testPutSingleInterrupt under Android Marshmallow by working around what seems to be a long-since-fixed ART bug.

The test broke after d417c5b, going undetected at presubmit time because a tool decided to skip some of our tests to save resources.

RELNOTES=n/a
PiperOrigin-RevId: 825661598

Author: cpovirk

android/guava-tests/test/com/google/common/util/concurrent/UninterruptiblesTest.java

@@ -78,6 +78,19 @@ protected void setUp() {
               + "Some test probably didn't clear the interrupt state");
     }
 
+    /*
+     * b/456222735: Initialize Truth up front. Android Marshmallow appears to sometimes clear the
+     * interrupt bit when requesting class initialization, breaking our tests that check that the
+     * interrupt bit is set when appropriate.
+     *
+     * Merely calling assert_(), while apparently enough to clear the interrupt in my experiments,
+     * is not enough to make the test reliably pass. (Presumably it leaves more classes to be
+     * initialized later, at which point they cause more clearing of the interrupt?) It's not
+     * obvious that the following assertion is necessarily enough, either, but in practice, it seems
+     * to work, at least with the current set of Truth classes that this test uses.
+     */
+    assertThat(1L).isGreaterThan(0);
+
     tearDownStack.addTearDown(
         new TearDown() {
           @Override

guava-tests/test/com/google/common/util/concurrent/UninterruptiblesTest.java

@@ -78,6 +78,19 @@ protected void setUp() {
               + "Some test probably didn't clear the interrupt state");
     }
 
+    /*
+     * b/456222735: Initialize Truth up front. Android Marshmallow appears to sometimes clear the
+     * interrupt bit when requesting class initialization, breaking our tests that check that the
+     * interrupt bit is set when appropriate.
+     *
+     * Merely calling assert_(), while apparently enough to clear the interrupt in my experiments,
+     * is not enough to make the test reliably pass. (Presumably it leaves more classes to be
+     * initialized later, at which point they cause more clearing of the interrupt?) It's not
+     * obvious that the following assertion is necessarily enough, either, but in practice, it seems
+     * to work, at least with the current set of Truth classes that this test uses.
+     */
+    assertThat(1L).isGreaterThan(0);
+
     tearDownStack.addTearDown(
         new TearDown() {
           @Override