Re: [PATCH 09/10] x86/intel_rdt: Hot cpu support for Cache Allocation

[Thomas Gleixner <tglx@linutronix.de>]

On Tue, 23 Jun 2015, Vikas Shivappa wrote:
> +/*
> + * cbm_update_msrs() - Updates all the existing IA32_L3_MASK_n MSRs
> + * which are one per CLOSid except IA32_L3_MASK_0 on the current package.
> + */
> +static inline void cbm_update_msrs(void)
> +{
> +	int maxid = boot_cpu_data.x86_cache_max_closid;
> +	unsigned int i;
> +
> +	/*
> +	 * At cpureset, all bits of IA32_L3_MASK_n are set.
> +	 * The index starts from one as there is no need
> +	 * to update IA32_L3_MASK_0 as it belongs to root cgroup
> +	 * whose cache mask is all 1s always.
> +	 */
> +	for (i = 1; i < maxid; i++) {
> +		if (ccmap[i].clos_refcnt)
> +			cbm_cpu_update((void *)i);

So this is called via

   intel_rdt_cpu_notifier(CPU_ONLINE)
     intel_rdt_cpu_start()
       cbm_update_msrs()
	 cbm_cpu_update()
	   wrmsr()

And runs on the CPU which initiated the hotplug and not on the newly
online cpu. Pretty useless, right?

> +	}
> +}
> +
> +static inline void intel_rdt_cpu_start(int cpu)
> +{
> +	struct intel_pqr_state *state = &per_cpu(pqr_state, cpu);
> +
> +	state->closid = 0;
> +	mutex_lock(&rdt_group_mutex);
> +	if (rdt_cpumask_update(cpu))
> +		cbm_update_msrs();
> +	mutex_unlock(&rdt_group_mutex);
> +}
> +
> +static void intel_rdt_cpu_exit(unsigned int cpu)
> +{
> +	int i;
> +
> +	mutex_lock(&rdt_group_mutex);
> +	if (!cpumask_test_and_clear_cpu(cpu, &rdt_cpumask)) {
> +		mutex_unlock(&rdt_group_mutex);
> +		return;
> +	}
> +
> +	i = cpumask_any_online_but(topology_core_cpumask(cpu), cpu);
> +	if (i < nr_cpu_ids)
> +		cpumask_set_cpu(i, &rdt_cpumask);
> +	mutex_unlock(&rdt_group_mutex);

Same issue as in the cqm patch.

Thanks,

	tglx