查询与进程调度(CFS)相关信息
目录
查询与进程相关的调度信息
查看CFS调度信息
CPU相关的信息
CFS就绪队列的总运行时间
实时队列与deadline调度的相关信息
所有进程相关的信息
查询与进程相关的调度信息
进程的nice值，优先级，调度策略,vruntime等信息。在proc目录下，以下是查询147号进程的调度信息。
cat /proc/3675/sched
update-notifier (3675, #threads: 4) //进程名称，ID：3675 以及线程数4
-------------------------------------------------------------------
se.exec_start : 26628458.370080 //进程总执行时间
se.vruntime : 89527.417357 //虚拟运行时间
se.sum_exec_runtime : 142.850903 //总运行时间
se.nr_migrations : 9 //发生9次迁移，
nr_switches : 727 //727次上下文切换
nr_voluntary_switches : 648 //主动调度次数
nr_involuntary_switches : 79 //被抢占次数
se.load.weight : 1048576//进程权重
se.runnable_weight : 1048576//
se.avg.load_sum : 58
se.avg.runnable_load_sum : 58
se.avg.util_sum : 59392
se.avg.load_avg : 0//量化负载
se.avg.runnable_load_avg : 0
se.avg.util_avg : 0 //进程的量化能力
se.avg.last_update_time : 26628458370048
se.avg.util_est.ewma : 9
se.avg.util_est.enqueued : 0
policy : 0//调度策略 SCHED_NORMAL
prio : 120//优先级
clock-delta : 14
mm->numa_scan_seq : 0
numa_pages_migrated : 0
numa_preferred_nid : -1
total_numa_faults : 0
current_node=0, numa_group_id=0
numa_faults node=0 task_private=0 task_shared=0 group_private=0 group_shared=0
root@ubuntu:/home/wy/release#
调度相关的统计 需要配置CONFIG_SCHEDSTATS
先执行 echo 1 > /proc/sys/kernel/sched_schedstats
再次查看，多出来好多statistics信息
cat /proc/3675/sched
update-notifier (3675, #threads: 4)
-------------------------------------------------------------------
se.exec_start : 26666000.266305
se.vruntime : 48799.679246
se.sum_exec_runtime : 144.183408
se.nr_migrations : 11
se.statistics.sum_sleep_runtime : 0.000000
se.statistics.wait_start : 0.000000
se.statistics.sleep_start : 0.000000
se.statistics.block_start : 0.000000
se.statistics.sleep_max : 0.000000
se.statistics.block_max : 0.000000
se.statistics.exec_max : 0.000000
se.statistics.slice_max : 0.000000
se.statistics.wait_max : 0.000000
se.statistics.wait_sum : 0.000000
se.statistics.wait_count : 0
se.statistics.iowait_sum : 0.000000
se.statistics.iowait_count : 0
se.statistics.nr_migrations_cold : 0
se.statistics.nr_failed_migrations_affine : 0
se.statistics.nr_failed_migrations_running : 0
se.statistics.nr_failed_migrations_hot : 0
se.statistics.nr_forced_migrations : 0
se.statistics.nr_wakeups : 0
se.statistics.nr_wakeups_sync : 0
se.statistics.nr_wakeups_migrate : 0
se.statistics.nr_wakeups_local : 0
se.statistics.nr_wakeups_remote : 0
se.statistics.nr_wakeups_affine : 0
se.statistics.nr_wakeups_affine_attempts : 0
se.statistics.nr_wakeups_passive : 0
se.statistics.nr_wakeups_idle : 0
avg_atom : 0.194579
avg_per_cpu : 13.107582
nr_switches : 741
nr_voluntary_switches : 662
nr_involuntary_switches : 79
se.load.weight : 1048576
se.runnable_weight : 1048576
se.avg.load_sum : 98
se.avg.runnable_load_sum : 98
se.avg.util_sum : 100391
se.avg.load_avg : 0
se.avg.runnable_load_avg : 0
se.avg.util_avg : 0
se.avg.last_update_time : 26666000266240
se.avg.util_est.ewma : 8
se.avg.util_est.enqueued : 1
policy : 0
prio : 120
clock-delta : 15
mm->numa_scan_seq : 0
numa_pages_migrated : 0
numa_preferred_nid : -1
total_numa_faults : 0
current_node=0, numa_group_id=0
numa_faults node=0 task_private=0 task_shared=0 group_private=0 group_shared=0
查看CFS调度信息
Sched Debug Version: v0.11, 5.4.0-26-generic #30-Ubuntu
ktime : 2270227.937775
sched_clk : 2270675.741916
cpu_clk : 2270509.286495
jiffies : 4295459310
sched_clock_stable() : 1
sysctl_sched
.sysctl_sched_latency : 18.000000
.sysctl_sched_min_granularity : 2.250000
.sysctl_sched_wakeup_granularity : 3.000000
.sysctl_sched_child_runs_first : 0
.sysctl_sched_features : 2059067
.sysctl_sched_tunable_scaling : 1 (logarithmic)
- 和调度相关的sysctl_sched的值
- 调度周期sysctl_sched_latency为 18ms.
- 调度最小粒度为2.25ms
- 唤醒最小粒度 3ms
- fork调用完成之后，禁止子进程先运行
CPU相关的信息
cpu#0, 2904.003 MHz
.nr_running : 1
.nr_switches : 360530
.nr_load_updates : 0
.nr_uninterruptible : 3
.next_balance : 4295.459320
.curr->pid : 11123
.clock : 2270510.005225
.clock_task : 2270510.005225
.avg_idle : 1000000
.max_idle_balance_cost : 500000
- nr_running 有1个进程在就绪队列里
- nr_switches 就绪队列发生进程切换的次数
- nr_load_updates 就绪队列里的cpu_load[]平滑负载更新的次数。
- next_balance 下一次做负载均衡的时间。
- curr->pid，正在运行的PID
- clock和clock_task：当前系统采样的时刻。
CFS就绪队列的总运行时间
cfs_rq[0]:/system.slice/containerd.service
.exec_clock : 0.000000
.MIN_vruntime : 0.000001
.min_vruntime : 820.991457
.max_vruntime : 0.000001
.spread : 0.000000
.spread0 : -127322.729964
.nr_spread_over : 0
.nr_running : 0
.load : 0
.runnable_weight : 0
.load_avg : 0
.runnable_load_avg : 0
.util_avg : 0
.util_est_enqueued : 0
.removed.load_avg : 0
.removed.util_avg : 0
.removed.runnable_sum : 0
.tg_load_avg_contrib : 0
.tg_load_avg : 0
.throttled : 0
.throttle_count : 0
.se->exec_start : 2270382.435692
.se->vruntime : 57637.719433
.se->sum_exec_runtime : 1692.348700
.se->load.weight : 1048576
.se->runnable_weight : 2
.se->avg.load_avg : 0
- exec_clock :CFS就绪队列的总运行时间；
- min_vruntime:表示CFS就绪队列的红黑树中最左边节点的 vruntime；
- max_vrunrime:CFS就绪队列的红黑树中最右边节点的vruntime；
- nr_running :CFS就绪队列中的进程数
- load：CFS就绪队列的总权重
- runnable_weight:CFS就绪队列中可运行状态的进程总权重
- load_avg:调度队列中的量化负载，这是CFS就绪队列中所有进程的量化负载之和
- runnable_load_avg:CFS就绪队列里所有可运行状态下的进程总量化负载
- util_avg：CFS就绪队列当前的量化计算能力
实时队列与deadline调度的相关信息
rt_rq[0]:/system.slice/cups-browsed.service
.rt_nr_running : 0
.rt_nr_migratory : 0
.rt_throttled : 0
.rt_time : 0.000000
.rt_runtime : 0.000000
dl_rq[0]:
.dl_nr_running : 0
.dl_nr_migratory : 0
.dl_bw->bw : 996147
.dl_bw->total_bw : 0
所有进程相关的信息
runnable tasks:
S task PID tree-key switches prio wait-time sum-exec sum-sleep
-----------------------------------------------------------------------------------------------------------
S kthreadd 2 127693.921719 338 120 0.000000 61.877558 0.000000 0 0 /
I rcu_gp 3 14.125023 2 100 0.000000 0.003748 0.000000 0 0 /
I rcu_par_gp 4 16.125400 2 100 0.000000 0.002300 0.000000 0 0 /
I kworker/0:0H 6 3901.743231 7 100 0.000000 0.123376 0.000000 0 0 /
I mm_percpu_wq 9 21.695937 2 100 0.000000 0.003137 0.000000 0 0 /
S ksoftirqd/0 10 127884.405645 1528 120 0.000000 55.672180 0.000000 0 0 /
S migration/0 12 0.000000 744 0 0.000000 28.640041 0.000000 0 0 /
S idle_inject/0 13 0.000000 3 49 0.000000 0.005038 0.000000 0 0 /
S cpuhp/0 14 9840.872379 10 120 0.000000 0.317368 0.000000 0 0 /
I kthrotld 154 2441.635038 2 100 0.000000 0.033174 0.000000 0 0 /
S irq/26-pciehp 157 0.000000 3 49 0.000000 0.215568 0.000000 0 0 /
S irq/30-pciehp 161 0.000000 3 49 0.000000 0.140429 0.000000 0 0 /
S irq/34-pciehp 165 0.000000 3 49 0.000000 0.190453 0.000000 0 0 /
S irq/38-pciehp 169 0.000000 3 49 0.000000 0.139489 0.000000 0 0 /
S irq/42-pciehp 173 0.000000 3 49 0.000000 0.139024 0.000000 0 0 /
S irq/46-pciehp 177 0.000000 3 49 0.000000 0.177426 0.000000 0 0 /
S irq/50-pciehp 181 0.000000 3 49 0.000000 0.086663 0.000000 0 0 /
S irq/54-pciehp 185 0.000000 3 49 0.000000 0.122572 0.000000 0 0 /
I scsi_tmf_1 192 2450.674661 2 100 0.000000 0.040843 0.000000 0 0 /
Ivfio-irqfd-clea 194 2450.675783 2 100 0.000000 0.078805 0.000000 0 0 /
Icharger_manager 227 2589.239305 2 100 0.000000 0.035306 0.000000 0 0 /
I mpt/0 297 3102.875515 2 100 0.000000 0.134915 0.000000 0 0 /
Ikworker/u256:28 365 128003.470404 5326 120 0.000000 333.547859 0.000000 0 0 /
I kworker/0:1H 377 127827.693809 1669 100 0.000000 242.858239 0.000000 0 0 /
Svmware-vmblock- 492 1.857542 5 120 0.000000 0.497505 0.000000 0 0 /system.slice/run-vmblock\x2dfuse.mount
S loop6 494 7943.387963 84 100 0.000000 21.071145 0.000000 0 0 /
S loop8 497 22417.738300 78 100 0.000000 16.332343 0.000000 0 0 /