On 2022/8/26 1:26, Paul E. McKenney wrote:
> On Thu, Aug 25, 2022 at 05:23:11PM +0800, Zhen Lei wrote:
>> When CONFIG_RCU_NOCB_CPU_DEFAULT_ALL=y or CONFIG_NO_HZ_FULL=y, additional
>> CPUs need to be added to 'rcu_nocb_mask'. But 'rcu_nocb_mask' may be not
>> available now, due to 'rcu_nocbs' is not specified. Check and initialize
>> 'rcu_nocb_mask' before using it. This code simplification strictly follows
>> this logic, compared with old implementations, unnecessary crossovers are
>> avoided and easy to understand.
>>
>> Signed-off-by: Zhen Lei <thunder.leizhen@xxxxxxxxxx>
>
> Much nicer, thank you!
>
> As usual, I could not resist the urge to wordsmith. Could you please
> check to make sure that I did not mess anything up?
Okay, thank you very much. The boot option name "rcu_nocb_mask=" is
incorrect, should be "rcu_nocbs=".
>
> Thanx, Paul
>
> ------------------------------------------------------------------------
>
> commit 4ac3b3d1a19943b1522c0b1d0895aefbb80ec294
> Author: Zhen Lei <thunder.leizhen@xxxxxxxxxx>
> Date: Thu Aug 25 17:23:11 2022 +0800
>
> rcu: Simplify rcu_init_nohz() cpumask handling
>
> In kernels built with either CONFIG_RCU_NOCB_CPU_DEFAULT_ALL=y or
> CONFIG_NO_HZ_FULL=y, additional CPUs must be added to rcu_nocb_mask.
> Except that kernels booted without the rcu_nocb_mask= will not have
rcu_nocb_mask= ----> rcu_nocbs=
> allocated rcu_nocb_mask. And the current rcu_init_nohz() function uses
> its need_rcu_nocb_mask and offload_all local variables to track the
> rcu_nocb and nohz_full state.
>
> But there is a much simpler approach, namely creating a cpumask pointer
> to track the default and then using cpumask_available() to check the
> rcu_nocb_mask state. This commit takes this approach, thereby simplifying
> and shortening the rcu_init_nohz() function.
>
> Signed-off-by: Zhen Lei <thunder.leizhen@xxxxxxxxxx>
> Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
>
> diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h
> index 0a5f0ef414845..c8167be2288fa 100644
> --- a/kernel/rcu/tree_nocb.h
> +++ b/kernel/rcu/tree_nocb.h
> @@ -1210,45 +1210,31 @@ EXPORT_SYMBOL_GPL(rcu_nocb_cpu_offload);
> void __init rcu_init_nohz(void)
> {
> int cpu;
> - bool need_rcu_nocb_mask = false;
> - bool offload_all = false;
> struct rcu_data *rdp;
> + const struct cpumask *cpumask = NULL;
>
> #if defined(CONFIG_RCU_NOCB_CPU_DEFAULT_ALL)
> - if (!rcu_state.nocb_is_setup) {
> - need_rcu_nocb_mask = true;
> - offload_all = true;
> - }
> -#endif /* #if defined(CONFIG_RCU_NOCB_CPU_DEFAULT_ALL) */
> -
> -#if defined(CONFIG_NO_HZ_FULL)
> - if (tick_nohz_full_running && !cpumask_empty(tick_nohz_full_mask)) {
> - need_rcu_nocb_mask = true;
> - offload_all = false; /* NO_HZ_FULL has its own mask. */
> - }
> -#endif /* #if defined(CONFIG_NO_HZ_FULL) */
> + cpumask = cpu_possible_mask;
> +#elif defined(CONFIG_NO_HZ_FULL)
> + if (tick_nohz_full_running && !cpumask_empty(tick_nohz_full_mask))
> + cpumask = tick_nohz_full_mask;
> +#endif
>
> - if (need_rcu_nocb_mask) {
> + if (cpumask) {
> if (!cpumask_available(rcu_nocb_mask)) {
> if (!zalloc_cpumask_var(&rcu_nocb_mask, GFP_KERNEL)) {
> pr_info("rcu_nocb_mask allocation failed, callback offloading disabled.\n");
> return;
> }
> }
> +
> + cpumask_or(rcu_nocb_mask, rcu_nocb_mask, cpumask);
> rcu_state.nocb_is_setup = true;
> }
>
> if (!rcu_state.nocb_is_setup)
> return;
>
> -#if defined(CONFIG_NO_HZ_FULL)
> - if (tick_nohz_full_running)
> - cpumask_or(rcu_nocb_mask, rcu_nocb_mask, tick_nohz_full_mask);
> -#endif /* #if defined(CONFIG_NO_HZ_FULL) */
> -
> - if (offload_all)
> - cpumask_setall(rcu_nocb_mask);
> -
> if (!cpumask_subset(rcu_nocb_mask, cpu_possible_mask)) {
> pr_info("\tNote: kernel parameter 'rcu_nocbs=', 'nohz_full', or 'isolcpus=' contains nonexistent CPUs.\n");
> cpumask_and(rcu_nocb_mask, cpu_possible_mask,
> .
>
--
Regards,
Zhen Lei
