[aarch64] INFO: rcu_sched detected expedited stalls on CPUs/tasks

From: Bruno Goncalves
Date: Fri Jun 03 2022 - 05:45:31 EST


Hello,

We recently started to hit this problem on some of our aarch64
machines. The stalls can happen even during boot.

[ 1086.949484] rcu: INFO: rcu_sched detected expedited stalls on
CPUs/tasks: { 23-... } 3 jiffies s: 3441 root: 0x2/.
[ 1086.949510] rcu: blocking rcu_node structures (internal RCU debug):
l=1:16-31:0x80/.
[ 1086.949524] Task dump for CPU 23:
[ 1086.949528] task:sugov:23 state:R running task stack:
0 pid: 2914 ppid: 2 flags:0x0000000a
[ 1086.949543] Call trace:
[ 1086.949546] __switch_to+0x104/0x19c
[ 1086.949568] __schedule+0x410/0x67c
[ 1086.949576] schedule+0x70/0xa8
[ 1086.949583] schedule_hrtimeout_range_clock+0x144/0x1d8
[ 1086.949592] schedule_hrtimeout_range+0x20/0x2c
[ 1086.949598] usleep_range_state+0x5c/0x80
[ 1086.949603] check_pcc_chan+0x7c/0xf4
[ 1086.949615] send_pcc_cmd+0x130/0x2a8
[ 1086.949619] cppc_set_perf+0x12c/0x22c
[ 1086.949624] cppc_cpufreq_set_target+0xf8/0x15c [cppc_cpufreq]
[ 1086.949645] __cpufreq_driver_target+0x94/0xfc
[ 1086.949658] sugov_work+0x98/0xe0
[ 1086.949675] kthread_worker_fn+0x124/0x2b8
[ 1086.949683] kthread+0xd4/0x558
[ 1086.949689] ret_from_fork+0x10/0x20

More logs:
https://s3.us-east-1.amazonaws.com/arr-cki-prod-datawarehouse-public/datawarehouse-public/2022/06/02/553734635/redhat:553734635_aarch64/tests/Storage_block_filesystem_fio_test/12073991_aarch64_1_dmesg.log

https://s3.us-east-1.amazonaws.com/arr-cki-prod-datawarehouse-public/datawarehouse-public/2022/06/02/553734635/redhat:553734635_aarch64/tests/Boot_test/12073991_aarch64_1_test_console.log

CKI issue tracker: https://datawarehouse.cki-project.org/issue/1259

Thanks,
Bruno Goncalves