If we happen to pull a running task on other cpu, a WARNING will happen. But the very task will be pulled to this cpu finally. This pulling is dangerous since the task's private data could be broken. Fix it by skipping this kind of task.
Signed-off-by: Yong Zhang <yong.zh...@windriver.com> Cc: Ingo Molnar <mi...@redhat.com> Cc: Peter Zijlstra <pet...@infradead.org> --- kernel/sched/rt.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 01970c8..e76d278 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -1665,7 +1665,15 @@ static int pull_rt_task(struct rq *this_rq) * the to-be-scheduled task? */ if (p && (p->prio < this_rq->rt.highest_prio.curr)) { - WARN_ON(p == src_rq->curr); + /* + * if we happen to choose a running task on other + * cpu, skip it and find another one. + */ + if (unlikely(p == src_rq->curr)) { + WARN_ON_ONCE(1); + goto skip; + } + WARN_ON(!p->on_rq); /* -- 1.7.9.5 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/