Sometime in the last week or two, we've been seeing intermittent failures in the mullapudi2016_reorder test. Not yet clear if this is a major regression or just a performance-threshold test that we've gotten close enough to that it is now flaky. Should be investigated.