On Tue, Nov 23, 2021 at 01:37:07AM +0100, Frederic Weisbecker wrote:
> If a user wants to boot without any CPU in offloaded mode initially but
> with the possibility to offload them later using cpusets, provide a way
> to simply pass an empty "rcu_nocbs" kernel parameter which will enforce
> the creation of dormant nocb kthreads.
Huh. This would have been a use for Yury Norov's "none" bitmask
specifier. ;-)
I pulled this one in with the usual wordsmithing.
Thanx, Paul
> Signed-off-by: Frederic Weisbecker <frederic@xxxxxxxxxx>
> Cc: Neeraj Upadhyay <quic_neeraju@xxxxxxxxxxx>
> Cc: Boqun Feng <boqun.feng@xxxxxxxxx>
> Cc: Uladzislau Rezki <urezki@xxxxxxxxx>
> Cc: Josh Triplett <josh@xxxxxxxxxxxxxxxx>
> Cc: Joel Fernandes <joel@xxxxxxxxxxxxxxxxx>
> ---
> .../admin-guide/kernel-parameters.txt | 26 ++++++++++++-------
> kernel/rcu/tree_nocb.h | 10 ++++---
> 2 files changed, 23 insertions(+), 13 deletions(-)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index cd860dc7c60b..6ff1a5f06383 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -4351,20 +4351,28 @@
> Disable the Correctable Errors Collector,
> see CONFIG_RAS_CEC help text.
>
> - rcu_nocbs= [KNL]
> - The argument is a cpu list, as described above.
> + rcu_nocbs[=cpu-list]
> + [KNL] The optional argument is a cpu list,
> + as described above.
>
> - In kernels built with CONFIG_RCU_NOCB_CPU=y, set
> - the specified list of CPUs to be no-callback CPUs.
> - Invocation of these CPUs' RCU callbacks will be
> - offloaded to "rcuox/N" kthreads created for that
> - purpose, where "x" is "p" for RCU-preempt, and
> - "s" for RCU-sched, and "N" is the CPU number.
> - This reduces OS jitter on the offloaded CPUs,
> + In kernels built with CONFIG_RCU_NOCB_CPU=y, enable the
> + no-callback CPU mode. Invocation of such CPUs' RCU
> + callbacks will be offloaded to "rcuox/N" kthreads
> + created for that purpose, where "x" is "p" for
> + RCU-preempt, and "s" for RCU-sched, and "N" is the CPU
> + number. This reduces OS jitter on the offloaded CPUs,
> which can be useful for HPC and real-time
> workloads. It can also improve energy efficiency
> for asymmetric multiprocessors.
>
> + If a cpulist is passed as an argument, the specified
> + list of CPUs is set to no-callback mode from boot.
> +
> + If otherwise the '=' sign and the cpulist arguments are
> + omitted, no CPU will be set to no-callback mode from
> + boot but cpuset will allow for toggling that mode at
> + runtime.
> +
> rcu_nocb_poll [KNL]
> Rather than requiring that offloaded CPUs
> (specified by rcu_nocbs= above) explicitly
> diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h
> index 9d37916278d4..d915780d40c8 100644
> --- a/kernel/rcu/tree_nocb.h
> +++ b/kernel/rcu/tree_nocb.h
> @@ -66,14 +66,16 @@ static bool rcu_nocb_is_setup;
> static int __init rcu_nocb_setup(char *str)
> {
> alloc_bootmem_cpumask_var(&rcu_nocb_mask);
> - if (cpulist_parse(str, rcu_nocb_mask)) {
> - pr_warn("rcu_nocbs= bad CPU range, all CPUs set\n");
> - cpumask_setall(rcu_nocb_mask);
> + if (*str == '=') {
> + if (cpulist_parse(++str, rcu_nocb_mask)) {
> + pr_warn("rcu_nocbs= bad CPU range, all CPUs set\n");
> + cpumask_setall(rcu_nocb_mask);
> + }
> }
> rcu_nocb_is_setup = true;
> return 1;
> }
> -__setup("rcu_nocbs=", rcu_nocb_setup);
> +__setup("rcu_nocbs", rcu_nocb_setup);
>
> static int __init parse_rcu_nocb_poll(char *arg)
> {
> --
> 2.25.1
>
