From: Cheng Jian cj.chengjian@huawei.com

hulk inclusion category: bugfix Bugzilla: 47618 CVE: NA

----------------------------------------

e221d028bb ("sched,rt: fix isolated CPUs leaving root_task_group indefinitely throttled") only fixes isolated CPUs leaving root_task_group, and not fix all other ordinary task_groutask_group.

In some scenarios where we need attach task bind to isolated CPUs in task_group, the same problem will occur.

Isolated CPUs and non-isolate CPUs are not in the same root_domain. and the hrtimer only check the cpumask of this_rq's root_domain. so when the handler of RT_BANDWIDTH hrtimer is running on the isolated CPU, it will leaved the non-isolated CPUs indefinitely throttled. Because bandwidth period hrtimer can't resume them. and viceversa.

Let the bandwidth timer check all the rt_rq of cpu_online_mask.

Signed-off-by: Cheng Jian cj.chengjian@huawei.com Reviewed-by: Xie XiuQi xiexiuqi@huawei.com Signed-off-by: zhangyi (F) yi.zhang@huawei.com Signed-off-by: Lu Jialin lujialin4@huawei.com Reviewed-by: xiu jianfeng xiujianfeng@huawei.com Signed-off-by: Zheng Zengkai zhengzengkai@huawei.com Signed-off-by: Zhao Wenhui zhaowenhui8@huawei.com --- kernel/sched/rt.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index c21b2da3735a..3285125a512f 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -906,16 +906,14 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) span = sched_rt_period_mask(); #ifdef CONFIG_RT_GROUP_SCHED /* - * FIXME: isolated CPUs should really leave the root task group, - * whether they are isolcpus or were isolated via cpusets, lest - * the timer run on a CPU which does not service all runqueues, - * potentially leaving other CPUs indefinitely throttled. If - * isolation is really required, the user will turn the throttle - * off to kill the perturbations it causes anyway. Meanwhile, - * this maintains functionality for boot and/or troubleshooting. + * When the tasks in the task_group run on either isolated + * CPUs or non-isolated CPUs, whether they are isolcpus or + * were isolated via cpusets, check all the online rt_rq + * to lest the timer run on a CPU which does not service + * all runqueues, potentially leaving other CPUs indefinitely + * throttled. */ - if (rt_b == &root_task_group.rt_bandwidth) - span = cpu_online_mask; + span = cpu_online_mask; #endif for_each_cpu(i, span) { int enqueue = 0;