Disable DCHECK for no BLOCK_SHUTDOWN posted after TaskScheduler shutdown.

base/task_scheduler/task_tracker.cc

Index: base/task_scheduler/task_tracker.cc
diff --git a/base/task_scheduler/task_tracker.cc b/base/task_scheduler/task_tracker.cc
index ebf43152064e1a407d5a87629b509936b7aec768..4a7ca94ff7e877b4170f01bd22a60d88d3ccd880 100644
--- a/base/task_scheduler/task_tracker.cc
+++ b/base/task_scheduler/task_tracker.cc
@@ -405,12 +405,18 @@ bool TaskTracker::BeforePostTask(TaskShutdownBehavior shutdown_behavior) {
       // ordering bug. This aims to catch those early.
       DCHECK(shutdown_event_);
       if (shutdown_event_->IsSignaled()) {
+#if DCHECK_IS_ON()
+// clang-format off
         // TODO(robliao): http://crbug.com/698140. Since the service thread
         // doesn't stop processing its own tasks at shutdown, we may still
         // attempt to post a BLOCK_SHUTDOWN task in response to a
-        // FileDescriptorWatcher.
-#if DCHECK_IS_ON()
-        DCHECK(IsPostingBlockShutdownTaskAfterShutdownAllowed());
+        // FileDescriptorWatcher. Same is true for FilePathWatcher

[robliao, 2017/05/31 21:33:54]

Tab this one more to the right

[gab, 2017/06/01 13:52:30]

Why? This is aligned with the code inside the if as it should be.

[robliao, 2017/06/01 17:00:53]

Ah interesting! This was a rendering bug in mobile Safari where I was viewing
the code.

+        // (http://crbug.com/728235). Until it's possible for such services to
+        // post to non-BLOCK_SHUTDOWN sequences which are themselves funneled to
+        // the main execution sequence (a future plan for the post_task.h API),
+        // this DCHECK will be flaky and must be disabled.
+        // DCHECK(IsPostingBlockShutdownTaskAfterShutdownAllowed());
+// clang-format on
 #endif
         state_->DecrementNumTasksBlockingShutdown();
         return false;