The kernel cmdline parameter rcutree.kthread_prio elevates the priority of various RCU service threads as stated by kernel-parameters.txt: Set the SCHED_FIFO priority of the RCU per-CPU kthreads (rcuc/N). This value is also used for the priority of the RCU boost threads (rcub/N) and for the RCU grace-period kthreads (rcu_bh, rcu_preempt, and rcu_sched). What the parameter does not do, at least in kernel 5.4, is to elevate the priority of the rcuop threads that perform the work of rcuc threads for cores offloaded via rcu_nocbs. Is there a reason why doing so would be a bad idea, or is the differing treatment of rcuc and rcuop threads merely an oversight? If the latter, would it make sense to patch the kernel to make the behavior more uniform? Thanks, Alison Chaiken Aurora Innovation
