[PATCH 1/2] powerpc/perf/hv-24x7: Add cpu hotplug support

[Gautham R Shenoy]

Hello Kajol,

On Thu, Jun 18, 2020 at 05:57:12PM +0530, Kajol Jain wrote:
> Patch here adds cpu hotplug functions to hv_24x7 pmu.
> A new cpuhp_state "CPUHP_AP_PERF_POWERPC_HV_24x7_ONLINE" enum
> is added.
> 
> The online function update the cpumask only if its NULL.
> As the primary intention for adding hotplug support
> is to desiginate a CPU to make HCALL to collect the
> count data.
> 
> The offline function test and clear corresponding cpu in a cpumask
> and update cpumask to any other active cpu.
> 
> With this patchset, perf tool side does not need "-C <cpu>"
> to be added.
> 
> Signed-off-by: Kajol Jain <kjain at linux.ibm.com>
> ---
>  arch/powerpc/perf/hv-24x7.c | 45 +++++++++++++++++++++++++++++++++++++
>  include/linux/cpuhotplug.h  |  1 +
>  2 files changed, 46 insertions(+)
> 
> diff --git a/arch/powerpc/perf/hv-24x7.c b/arch/powerpc/perf/hv-24x7.c
> index db213eb7cb02..fdc4ae155d60 100644
> --- a/arch/powerpc/perf/hv-24x7.c
> +++ b/arch/powerpc/perf/hv-24x7.c
> @@ -31,6 +31,8 @@ static int interface_version;
>  /* Whether we have to aggregate result data for some domains. */
>  static bool aggregate_result_elements;
> 
> +static cpumask_t hv_24x7_cpumask;
> +
>  static bool domain_is_valid(unsigned domain)
>  {
>  	switch (domain) {
> @@ -1641,6 +1643,44 @@ static struct pmu h_24x7_pmu = {
>  	.capabilities = PERF_PMU_CAP_NO_EXCLUDE,
>  };
> 
> +static int ppc_hv_24x7_cpu_online(unsigned int cpu)
> +{
> +	/* Make this CPU the designated target for counter collection */
> +	if (cpumask_empty(&hv_24x7_cpumask))
> +		cpumask_set_cpu(cpu, &hv_24x7_cpumask);
> +
> +	return 0;
> +}
> +
> +static int ppc_hv_24x7_cpu_offline(unsigned int cpu)
> +{
> +	int target = -1;
> +
> +	/* Check if exiting cpu is used for collecting 24x7 events */
> +	if (!cpumask_test_and_clear_cpu(cpu, &hv_24x7_cpumask))
> +		return 0;
> +
> +	/* Find a new cpu to collect 24x7 events */
> +	target = cpumask_any_but(cpu_active_mask, cpu);

cpumask_any_but() typically picks the first CPU in cpu_active_mask
that is not @cpu.


> +
> +	if (target < 0 || target >= nr_cpu_ids)
> +		return -1;
> +
> +	/* Migrate 24x7 events to the new target */
> +	cpumask_set_cpu(target, &hv_24x7_cpumask);
> +	perf_pmu_migrate_context(&h_24x7_pmu, cpu, target);


On a system with N CPUs numbered [O..N-1], can you please verify if
the time required to sequentially offline CPUs [0..N-2] ,in that
order, increase with this patch ?

I am asking this because we have encountered this problem once before
at a customer site and the commit 9c9f8fb71fee ("powerpc/perf: Use
cpumask_last() to determine the designated cpu for nest/core units.")
was introduced to fix that problem.

> +
> +	return 0;
> +}
> +
> +static int hv_24x7_cpu_hotplug_init(void)
> +{
> +	return cpuhp_setup_state(CPUHP_AP_PERF_POWERPC_HV_24x7_ONLINE,
> +			  "perf/powerpc/hv_24x7:online",
> +			  ppc_hv_24x7_cpu_online,
> +			  ppc_hv_24x7_cpu_offline);
> +}
> +
>  static int hv_24x7_init(void)
>  {
>  	int r;
> @@ -1685,6 +1725,11 @@ static int hv_24x7_init(void)
>  	if (r)
>  		return r;
> 
> +	/* init cpuhotplug */
> +	r = hv_24x7_cpu_hotplug_init();
> +	if (r)
> +		pr_err("hv_24x7: CPU hotplug init failed\n");
> +
>  	r = perf_pmu_register(&h_24x7_pmu, h_24x7_pmu.name, -1);
>  	if (r)
>  		return r;
> diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
> index 8377afef8806..16ed8f6f8774 100644
> --- a/include/linux/cpuhotplug.h
> +++ b/include/linux/cpuhotplug.h
> @@ -180,6 +180,7 @@ enum cpuhp_state {
>  	CPUHP_AP_PERF_POWERPC_CORE_IMC_ONLINE,
>  	CPUHP_AP_PERF_POWERPC_THREAD_IMC_ONLINE,
>  	CPUHP_AP_PERF_POWERPC_TRACE_IMC_ONLINE,
> +	CPUHP_AP_PERF_POWERPC_HV_24x7_ONLINE,
>  	CPUHP_AP_WATCHDOG_ONLINE,
>  	CPUHP_AP_WORKQUEUE_ONLINE,
>  	CPUHP_AP_RCUTREE_ONLINE,
> -- 
> 2.18.2
>