Application performance degradation on cgroup2
Issue
- Application performance degradation happens on cgroup2.
- Perf report shows the spin_lock contention is in observation from cgroup_rstat_flush_irqsafe.
Overhead sys usr Shared Object
- 68.61% 68.61% 0.00% [kernel.vmlinux] ▒
+ 33.54% acpi_processor_ffh_cstate_enter ▒
- 6.17% native_queued_spin_lock_slowpath ▒
- 5.72% _raw_spin_lock_irqsave ▒
- 5.56% cgroup_rstat_flush_irqsafe ▒
- mem_cgroup_wb_stats ▒
- 5.55% wb_over_bg_thresh ▒
+ 3.51% wb_workfn ▒
+ 2.04% wb_writeback ▒
+ 3.76% update_blocked_averages ▒
+ 1.04% __update_load_avg_cfs_rq ▒
+ 0.88% copy_user_enhanced_fast_string ▒
+ 0.85% __d_lookup_rcu ▒
+ 0.60% update_sd_lb_stats.constprop.121 ▒
Environment
- Red Hat Enterprise Linux 9.1
kernel-5.14.0-162.12.1.el9_1
- Red Hat Enterprise Linux 8.6
kernel-4.18.0-372.9.1.el8
- cgroup v2
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.