[v3,4/4] powercap/drivers/dtpm: Add CPU energy model based support
diff mbox series

Message ID 20201123214208.11007-5-daniel.lezcano@linaro.org
State New, archived
Headers show
Series
  • powercap/dtpm: Add the DTPM framework
Related show

Commit Message

Daniel Lezcano Nov. 23, 2020, 9:42 p.m. UTC
With the powercap dtpm controller, we are able to plug devices with
power limitation features in the tree.

The following patch introduces the CPU power limitation based on the
energy model and the performance states.

The power limitation is done at the performance domain level. If some
CPUs are unplugged, the corresponding power will be subtracted from
the performance domain total power.

It is up to the platform to initialize the dtpm tree and add the CPU.

Here is an example to create a simple tree with one root node called
"pkg" and the CPU's performance domains.

static int dtpm_register_pkg(struct dtpm_descr *descr)
{
	struct dtpm *pkg;
	int ret;

	pkg = dtpm_alloc(NULL);
	if (!pkg)
		return -ENOMEM;

	ret = dtpm_register(descr->name, pkg, descr->parent);
	if (ret)
		return ret;

	return dtpm_register_cpu(pkg);
}

static struct dtpm_descr descr = {
	.name = "pkg",
	.init = dtpm_register_pkg,
};
DTPM_DECLARE(descr);

Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
---
 drivers/powercap/Kconfig    |   7 +
 drivers/powercap/Makefile   |   1 +
 drivers/powercap/dtpm_cpu.c | 252 ++++++++++++++++++++++++++++++++++++
 include/linux/cpuhotplug.h  |   1 +
 include/linux/dtpm.h        |   2 +
 5 files changed, 263 insertions(+)
 create mode 100644 drivers/powercap/dtpm_cpu.c

Comments

Lukasz Luba Nov. 26, 2020, 10:06 a.m. UTC | #1
Hi Daniel,

On 11/23/20 9:42 PM, Daniel Lezcano wrote:
> With the powercap dtpm controller, we are able to plug devices with
> power limitation features in the tree.
> 

[snip]

> +
> +static void pd_release(struct dtpm *dtpm)
> +{
> +	struct dtpm_cpu *dtpm_cpu = dtpm->private;
> +

Maybe it's worth to add:
------------------->8----------------
if (freq_qos_request_active(&dtpm_cpu->qos_req))
	freq_qos_remove_request(&dtpm_cpu->qos_req);
-------------------8<---------------

If we are trying to unregister dtpm in error path due to freq_qos
registration failure, a warning would be emitted from freq_qos.

> +	freq_qos_remove_request(&dtpm_cpu->qos_req);
> +	kfree(dtpm_cpu);
> +}

[snip]

> +
> +static int cpuhp_dtpm_cpu_online(unsigned int cpu)
> +{
> +	struct dtpm *dtpm;
> +	struct dtpm_cpu *dtpm_cpu;
> +	struct cpufreq_policy *policy;
> +	struct em_perf_domain *pd;
> +	char name[CPUFREQ_NAME_LEN];
> +	int ret;
> +
> +	policy = cpufreq_cpu_get(cpu);
> +
> +	if (!policy)
> +		return 0;
> +
> +	pd = em_cpu_get(cpu);
> +	if (!pd)
> +		return -EINVAL;
> +
> +	dtpm = per_cpu(dtpm_per_cpu, cpu);
> +	if (dtpm)
> +		return power_add(dtpm, pd);
> +
> +	dtpm = dtpm_alloc(&dtpm_ops);
> +	if (!dtpm)
> +		return -EINVAL;
> +
> +	dtpm_cpu = kzalloc(sizeof(dtpm_cpu), GFP_KERNEL);
> +	if (!dtpm_cpu) {
> +		kfree(dtpm);
> +		return -ENOMEM;
> +	}
> +
> +	dtpm->private = dtpm_cpu;
> +	dtpm_cpu->cpu = cpu;
> +
> +	for_each_cpu(cpu, policy->related_cpus)
> +		per_cpu(dtpm_per_cpu, cpu) = dtpm;
> +
> +	sprintf(name, "cpu%d", dtpm_cpu->cpu);
> +
> +	ret = dtpm_register(name, dtpm, __parent);
> +	if (ret)
> +		goto out_kfree_dtpm_cpu;
> +
> +	ret = power_add(dtpm, pd);
> +	if (ret)
> +		goto out_power_sub;

Shouldn't we call dtpm_unregister() instead?
The dtpm_unregister() would remove the zone, which IIUC we
are currently missing.

> +
> +	ret = freq_qos_add_request(&policy->constraints,
> +				   &dtpm_cpu->qos_req, FREQ_QOS_MAX,
> +				   pd->table[pd->nr_perf_states - 1].frequency);
> +	if (ret)
> +		goto out_dtpm_unregister;

Could this trigger different steps, starting from out_power_sub_v2
below?

> +
> +	return 0;
> +
> +out_dtpm_unregister:
> +	dtpm_unregister(dtpm);
> +	dtpm_cpu = NULL; /* Already freed by the release ops */
> +out_power_sub:
> +	power_sub(dtpm, pd);

I would change the order of these two above into something like:

out_power_sub_v2:
	power_sub(dtpm, pd);
out_dtpm_unregister_v2:
	dtpm_unregister(dtpm);
	dtpm_cpu = NULL;

> +out_kfree_dtpm_cpu:
> +	for_each_cpu(cpu, policy->related_cpus)
> +		per_cpu(dtpm_per_cpu, cpu) = NULL;
> +	kfree(dtpm_cpu);
> +
> +	return ret;
> +}

IIUC power_sub() would decrement the power and set it to 0 for that
dtmp, then the dtpm_unregister() would also try to decrement the power,
but by the value of 0. So it should be safe.

Regards,
Lukasz
Daniel Lezcano Nov. 26, 2020, 3:06 p.m. UTC | #2
Hi Lukasz,


On 26/11/2020 11:06, Lukasz Luba wrote:
> Hi Daniel,
> 
> On 11/23/20 9:42 PM, Daniel Lezcano wrote:
>> With the powercap dtpm controller, we are able to plug devices with
>> power limitation features in the tree.
>>
> 
> [snip]
> 
>> +
>> +static void pd_release(struct dtpm *dtpm)
>> +{
>> +    struct dtpm_cpu *dtpm_cpu = dtpm->private;
>> +
> 
> Maybe it's worth to add:
> ------------------->8----------------
> if (freq_qos_request_active(&dtpm_cpu->qos_req))
>     freq_qos_remove_request(&dtpm_cpu->qos_req);
> -------------------8<---------------
> 
> If we are trying to unregister dtpm in error path due to freq_qos
> registration failure, a warning would be emitted from freq_qos.

Ah yes, good point.

>> +    freq_qos_remove_request(&dtpm_cpu->qos_req);
>> +    kfree(dtpm_cpu);
>> +}
> 
> [snip]
> 
>> +
>> +static int cpuhp_dtpm_cpu_online(unsigned int cpu)
>> +{
>> +    struct dtpm *dtpm;
>> +    struct dtpm_cpu *dtpm_cpu;
>> +    struct cpufreq_policy *policy;
>> +    struct em_perf_domain *pd;
>> +    char name[CPUFREQ_NAME_LEN];
>> +    int ret;
>> +
>> +    policy = cpufreq_cpu_get(cpu);
>> +
>> +    if (!policy)
>> +        return 0;
>> +
>> +    pd = em_cpu_get(cpu);
>> +    if (!pd)
>> +        return -EINVAL;
>> +
>> +    dtpm = per_cpu(dtpm_per_cpu, cpu);
>> +    if (dtpm)
>> +        return power_add(dtpm, pd);
>> +
>> +    dtpm = dtpm_alloc(&dtpm_ops);
>> +    if (!dtpm)
>> +        return -EINVAL;
>> +
>> +    dtpm_cpu = kzalloc(sizeof(dtpm_cpu), GFP_KERNEL);
>> +    if (!dtpm_cpu) {
>> +        kfree(dtpm);
>> +        return -ENOMEM;
>> +    }
>> +
>> +    dtpm->private = dtpm_cpu;
>> +    dtpm_cpu->cpu = cpu;
>> +
>> +    for_each_cpu(cpu, policy->related_cpus)
>> +        per_cpu(dtpm_per_cpu, cpu) = dtpm;
>> +
>> +    sprintf(name, "cpu%d", dtpm_cpu->cpu);
>> +
>> +    ret = dtpm_register(name, dtpm, __parent);
>> +    if (ret)
>> +        goto out_kfree_dtpm_cpu;
>> +
>> +    ret = power_add(dtpm, pd);
>> +    if (ret)
>> +        goto out_power_sub;
> 
> Shouldn't we call dtpm_unregister() instead?
> The dtpm_unregister() would remove the zone, which IIUC we
> are currently missing.
> 
>> +
>> +    ret = freq_qos_add_request(&policy->constraints,
>> +                   &dtpm_cpu->qos_req, FREQ_QOS_MAX,
>> +                   pd->table[pd->nr_perf_states - 1].frequency);
>> +    if (ret)
>> +        goto out_dtpm_unregister;
> 
> Could this trigger different steps, starting from out_power_sub_v2
> below?
> 
>> +
>> +    return 0;
>> +
>> +out_dtpm_unregister:
>> +    dtpm_unregister(dtpm);
>> +    dtpm_cpu = NULL; /* Already freed by the release ops */
>> +out_power_sub:
>> +    power_sub(dtpm, pd);
> 
> I would change the order of these two above into something like:

Ok, I'll revisit the rollback routine.

> out_power_sub_v2:
>     power_sub(dtpm, pd);
> out_dtpm_unregister_v2:
>     dtpm_unregister(dtpm);
>     dtpm_cpu = NULL;
> 
>> +out_kfree_dtpm_cpu:
>> +    for_each_cpu(cpu, policy->related_cpus)
>> +        per_cpu(dtpm_per_cpu, cpu) = NULL;
>> +    kfree(dtpm_cpu);
>> +
>> +    return ret;
>> +}
> 
> IIUC power_sub() would decrement the power and set it to 0 for that
> dtmp, then the dtpm_unregister() would also try to decrement the power,
> but by the value of 0. So it should be safe.

Right.


Thanks for the review

Patch
diff mbox series

diff --git a/drivers/powercap/Kconfig b/drivers/powercap/Kconfig
index cc1953bd8bed..20b4325c6161 100644
--- a/drivers/powercap/Kconfig
+++ b/drivers/powercap/Kconfig
@@ -49,4 +49,11 @@  config DTPM
 	help
 	  This enables support for the power capping for the dynamic
 	  thermal power management userspace engine.
+
+config DTPM_CPU
+	bool "Add CPU power capping based on the energy model"
+	depends on DTPM && ENERGY_MODEL
+	help
+	  This enables support for CPU power limitation based on
+	  energy model.
 endif
diff --git a/drivers/powercap/Makefile b/drivers/powercap/Makefile
index 6482ac52054d..fabcf388a8d3 100644
--- a/drivers/powercap/Makefile
+++ b/drivers/powercap/Makefile
@@ -1,5 +1,6 @@ 
 # SPDX-License-Identifier: GPL-2.0-only
 obj-$(CONFIG_DTPM) += dtpm.o
+obj-$(CONFIG_DTPM_CPU) += dtpm_cpu.o
 obj-$(CONFIG_POWERCAP)	+= powercap_sys.o
 obj-$(CONFIG_INTEL_RAPL_CORE) += intel_rapl_common.o
 obj-$(CONFIG_INTEL_RAPL) += intel_rapl_msr.o
diff --git a/drivers/powercap/dtpm_cpu.c b/drivers/powercap/dtpm_cpu.c
new file mode 100644
index 000000000000..81f0996dd8ad
--- /dev/null
+++ b/drivers/powercap/dtpm_cpu.c
@@ -0,0 +1,252 @@ 
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright 2020 Linaro Limited
+ *
+ * Author: Daniel Lezcano <daniel.lezcano@linaro.org>
+ *
+ * The DTPM CPU is based on the energy model. It hooks the CPU in the
+ * DTPM tree which in turns update the power number by propagating the
+ * power number from the CPU energy model information to the parents.
+ *
+ * The association between the power and the performance state, allows
+ * to set the power of the CPU at the OPP granularity.
+ *
+ * The CPU hotplug is supported and the power numbers will be updated
+ * if a CPU is hot plugged / unplugged.
+ */
+#include <linux/cpumask.h>
+#include <linux/cpufreq.h>
+#include <linux/cpuhotplug.h>
+#include <linux/dtpm.h>
+#include <linux/energy_model.h>
+#include <linux/pm_qos.h>
+#include <linux/slab.h>
+#include <linux/units.h>
+
+static struct dtpm *__parent;
+
+static DEFINE_PER_CPU(struct dtpm *, dtpm_per_cpu);
+
+struct dtpm_cpu {
+	struct freq_qos_request qos_req;
+	int cpu;
+};
+
+/*
+ * When a new CPU is inserted at hotplug or boot time, add the power
+ * contribution and update the dtpm tree.
+ */
+static int power_add(struct dtpm *dtpm, struct em_perf_domain *em)
+{
+	u64 power_min, power_max;
+
+	power_min = em->table[0].power;
+	power_min *= MICROWATT_PER_MILLIWATT;
+	power_min += dtpm->power_min;
+
+	power_max = em->table[em->nr_perf_states - 1].power;
+	power_max *= MICROWATT_PER_MILLIWATT;
+	power_max += dtpm->power_max;
+
+	return dtpm_update_power(dtpm, power_min, power_max);
+}
+
+/*
+ * When a CPU is unplugged, remove its power contribution from the
+ * dtpm tree.
+ */
+static int power_sub(struct dtpm *dtpm, struct em_perf_domain *em)
+{
+	u64 power_min, power_max;
+
+	power_min = em->table[0].power;
+	power_min *= MICROWATT_PER_MILLIWATT;
+	power_min = dtpm->power_min - power_min;
+
+	power_max = em->table[em->nr_perf_states - 1].power;
+	power_max *= MICROWATT_PER_MILLIWATT;
+	power_max = dtpm->power_max - power_max;
+
+	return dtpm_update_power(dtpm, power_min, power_max);
+}
+
+static u64 set_pd_power_limit(struct dtpm *dtpm, u64 power_limit)
+{
+	struct dtpm_cpu *dtpm_cpu = dtpm->private;
+	struct em_perf_domain *pd;
+	struct cpumask cpus;
+	unsigned long freq;
+	u64 power;
+	int i, nr_cpus;
+
+	pd = em_cpu_get(dtpm_cpu->cpu);
+
+	cpumask_and(&cpus, cpu_online_mask, to_cpumask(pd->cpus));
+
+	nr_cpus = cpumask_weight(&cpus);
+
+	for (i = 0; i < pd->nr_perf_states; i++) {
+
+		power = pd->table[i].power * MICROWATT_PER_MILLIWATT * nr_cpus;
+
+		if (power > power_limit)
+			break;
+	}
+
+	freq = pd->table[i - 1].frequency;
+
+	freq_qos_update_request(&dtpm_cpu->qos_req, freq);
+
+	power_limit = pd->table[i - 1].power *
+		MICROWATT_PER_MILLIWATT * nr_cpus;
+
+	return power_limit;
+}
+
+static u64 get_pd_power_uw(struct dtpm *dtpm)
+{
+	struct dtpm_cpu *dtpm_cpu = dtpm->private;
+	struct em_perf_domain *pd;
+	struct cpumask cpus;
+	unsigned long freq;
+	int i, nr_cpus;
+
+	pd = em_cpu_get(dtpm_cpu->cpu);
+	freq = cpufreq_quick_get(dtpm_cpu->cpu);
+	cpumask_and(&cpus, cpu_online_mask, to_cpumask(pd->cpus));
+	nr_cpus = cpumask_weight(&cpus);
+
+	for (i = 0; i < pd->nr_perf_states; i++) {
+
+		if (pd->table[i].frequency < freq)
+			continue;
+
+		return pd->table[i].power *
+			MICROWATT_PER_MILLIWATT * nr_cpus;
+	}
+
+	return 0;
+}
+
+static void pd_release(struct dtpm *dtpm)
+{
+	struct dtpm_cpu *dtpm_cpu = dtpm->private;
+
+	freq_qos_remove_request(&dtpm_cpu->qos_req);
+	kfree(dtpm_cpu);
+}
+
+static struct dtpm_ops dtpm_ops = {
+	.set_power_uw = set_pd_power_limit,
+	.get_power_uw = get_pd_power_uw,
+	.release = pd_release,
+};
+
+static int cpuhp_dtpm_cpu_offline(unsigned int cpu)
+{
+	struct cpufreq_policy *policy;
+	struct em_perf_domain *pd;
+	struct dtpm *dtpm;
+
+	policy = cpufreq_cpu_get(cpu);
+
+	if (!policy)
+		return 0;
+
+	pd = em_cpu_get(cpu);
+	if (!pd)
+		return -EINVAL;
+
+	dtpm = per_cpu(dtpm_per_cpu, cpu);
+
+	power_sub(dtpm, pd);
+
+	if (cpumask_weight(policy->cpus) != 1)
+		return 0;
+
+	for_each_cpu(cpu, policy->related_cpus)
+		per_cpu(dtpm_per_cpu, cpu) = NULL;
+
+	dtpm_unregister(dtpm);
+
+	return 0;
+}
+
+static int cpuhp_dtpm_cpu_online(unsigned int cpu)
+{
+	struct dtpm *dtpm;
+	struct dtpm_cpu *dtpm_cpu;
+	struct cpufreq_policy *policy;
+	struct em_perf_domain *pd;
+	char name[CPUFREQ_NAME_LEN];
+	int ret;
+
+	policy = cpufreq_cpu_get(cpu);
+
+	if (!policy)
+		return 0;
+
+	pd = em_cpu_get(cpu);
+	if (!pd)
+		return -EINVAL;
+
+	dtpm = per_cpu(dtpm_per_cpu, cpu);
+	if (dtpm)
+		return power_add(dtpm, pd);
+
+	dtpm = dtpm_alloc(&dtpm_ops);
+	if (!dtpm)
+		return -EINVAL;
+
+	dtpm_cpu = kzalloc(sizeof(dtpm_cpu), GFP_KERNEL);
+	if (!dtpm_cpu) {
+		kfree(dtpm);
+		return -ENOMEM;
+	}
+
+	dtpm->private = dtpm_cpu;
+	dtpm_cpu->cpu = cpu;
+
+	for_each_cpu(cpu, policy->related_cpus)
+		per_cpu(dtpm_per_cpu, cpu) = dtpm;
+
+	sprintf(name, "cpu%d", dtpm_cpu->cpu);
+
+	ret = dtpm_register(name, dtpm, __parent);
+	if (ret)
+		goto out_kfree_dtpm_cpu;
+
+	ret = power_add(dtpm, pd);
+	if (ret)
+		goto out_power_sub;
+
+	ret = freq_qos_add_request(&policy->constraints,
+				   &dtpm_cpu->qos_req, FREQ_QOS_MAX,
+				   pd->table[pd->nr_perf_states - 1].frequency);
+	if (ret)
+		goto out_dtpm_unregister;
+
+	return 0;
+
+out_dtpm_unregister:
+	dtpm_unregister(dtpm);
+	dtpm_cpu = NULL; /* Already freed by the release ops */
+out_power_sub:
+	power_sub(dtpm, pd);
+out_kfree_dtpm_cpu:
+	for_each_cpu(cpu, policy->related_cpus)
+		per_cpu(dtpm_per_cpu, cpu) = NULL;
+	kfree(dtpm_cpu);
+
+	return ret;
+}
+
+int dtpm_register_cpu(struct dtpm *parent)
+{
+	__parent = parent;
+
+	return cpuhp_setup_state(CPUHP_AP_DTPM_CPU_ONLINE,
+				 "dtpm_cpu:online",
+				 cpuhp_dtpm_cpu_online,
+				 cpuhp_dtpm_cpu_offline);
+}
diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
index bc56287a1ed1..72fd8db62342 100644
--- a/include/linux/cpuhotplug.h
+++ b/include/linux/cpuhotplug.h
@@ -192,6 +192,7 @@  enum cpuhp_state {
 	CPUHP_AP_ONLINE_DYN_END		= CPUHP_AP_ONLINE_DYN + 30,
 	CPUHP_AP_X86_HPET_ONLINE,
 	CPUHP_AP_X86_KVM_CLK_ONLINE,
+	CPUHP_AP_DTPM_CPU_ONLINE,
 	CPUHP_AP_ACTIVE,
 	CPUHP_ONLINE,
 };
diff --git a/include/linux/dtpm.h b/include/linux/dtpm.h
index 84b087b4ac01..723fe51f1628 100644
--- a/include/linux/dtpm.h
+++ b/include/linux/dtpm.h
@@ -73,4 +73,6 @@  void dtpm_unregister(struct dtpm *dtpm);
 
 int dtpm_register(const char *name, struct dtpm *dtpm, struct dtpm *parent);
 
+int dtpm_register_cpu(struct dtpm *parent);
+
 #endif