在 2022/1/13 上午4:42, Tejun Heo 写道:

Hello,

On Tue, Jan 11, 2022 at 05:56:01PM +0800, Cruz Zhao wrote:

quoted

+#ifdef CONFIG_SCHED_CORE
+void cpuacct_account_forceidle(int cpu, struct task_struct *tsk, u64 cputime)
+{
+	struct cpuacct *ca;
+	u64 *fi;
+
+	rcu_read_lock();
+	/*
+	 * We have hold rq->core->__lock here, which protects ca->forceidle
+	 * percpu.
+	 */
+	for (ca = task_ca(tsk); ca; ca = parent_ca(ca)) {
+		fi = per_cpu_ptr(ca->forceidle, cpu);
+		*fi += cputime;
+	}

Please don't do this. Use rstat and integrate it with other stats.

Thanks.

Thanks for suggestions, I'll try to do this using rstat. BTW, is it ok
to integrate it with cgroup_base_stat?

Best,
Cruz Zhao