[linux-2.6-block.git] / drivers / cpuidle / cpuidle-riscv-sbi.c

// SPDX-License-Identifier: GPL-2.0-only
/*
 * RISC-V SBI CPU idle driver.
 *
 * Copyright (c) 2021 Western Digital Corporation or its affiliates.
 * Copyright (c) 2022 Ventana Micro Systems Inc.
 */

#define pr_fmt(fmt) "cpuidle-riscv-sbi: " fmt

#include <linux/cpuhotplug.h>
#include <linux/cpuidle.h>
#include <linux/cpumask.h>
#include <linux/cpu_pm.h>
#include <linux/cpu_cooling.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/of.h>
#include <linux/slab.h>
#include <linux/platform_device.h>
#include <linux/pm_domain.h>
#include <linux/pm_runtime.h>
#include <asm/cpuidle.h>
#include <asm/sbi.h>
#include <asm/smp.h>
#include <asm/suspend.h>

#include "dt_idle_states.h"
#include "dt_idle_genpd.h"

struct sbi_cpuidle_data {
	u32 *states;
	struct device *dev;
};

struct sbi_domain_state {
	bool available;
	u32 state;
};

static DEFINE_PER_CPU_READ_MOSTLY(struct sbi_cpuidle_data, sbi_cpuidle_data);
static DEFINE_PER_CPU(struct sbi_domain_state, domain_state);
static bool sbi_cpuidle_use_osi;
static bool sbi_cpuidle_use_cpuhp;
static bool sbi_cpuidle_pd_allow_domain_state;

static inline void sbi_set_domain_state(u32 state)
{
	struct sbi_domain_state *data = this_cpu_ptr(&domain_state);

	data->available = true;
	data->state = state;
}

static inline u32 sbi_get_domain_state(void)
{
	struct sbi_domain_state *data = this_cpu_ptr(&domain_state);

	return data->state;
}

static inline void sbi_clear_domain_state(void)
{
	struct sbi_domain_state *data = this_cpu_ptr(&domain_state);

	data->available = false;
}

static inline bool sbi_is_domain_state_available(void)
{
	struct sbi_domain_state *data = this_cpu_ptr(&domain_state);

	return data->available;
}

static int sbi_suspend_finisher(unsigned long suspend_type,
				unsigned long resume_addr,
				unsigned long opaque)
{
	struct sbiret ret;

	ret = sbi_ecall(SBI_EXT_HSM, SBI_EXT_HSM_HART_SUSPEND,
			suspend_type, resume_addr, opaque, 0, 0, 0);

	return (ret.error) ? sbi_err_map_linux_errno(ret.error) : 0;
}

static int sbi_suspend(u32 state)
{
	if (state & SBI_HSM_SUSP_NON_RET_BIT)
		return cpu_suspend(state, sbi_suspend_finisher);
	else
		return sbi_suspend_finisher(state, 0, 0);
}

static __cpuidle int sbi_cpuidle_enter_state(struct cpuidle_device *dev,
					     struct cpuidle_driver *drv, int idx)
{
	u32 *states = __this_cpu_read(sbi_cpuidle_data.states);
	u32 state = states[idx];

	if (state & SBI_HSM_SUSP_NON_RET_BIT)
		return CPU_PM_CPU_IDLE_ENTER_PARAM(sbi_suspend, idx, state);
	else
		return CPU_PM_CPU_IDLE_ENTER_RETENTION_PARAM(sbi_suspend,
							     idx, state);
}

static __cpuidle int __sbi_enter_domain_idle_state(struct cpuidle_device *dev,
						   struct cpuidle_driver *drv, int idx,
						   bool s2idle)
{
	struct sbi_cpuidle_data *data = this_cpu_ptr(&sbi_cpuidle_data);
	u32 *states = data->states;
	struct device *pd_dev = data->dev;
	u32 state;
	int ret;

	ret = cpu_pm_enter();
	if (ret)
		return -1;

	/* Do runtime PM to manage a hierarchical CPU toplogy. */
	if (s2idle)
		dev_pm_genpd_suspend(pd_dev);
	else
		pm_runtime_put_sync_suspend(pd_dev);

	ct_cpuidle_enter();

	if (sbi_is_domain_state_available())
		state = sbi_get_domain_state();
	else
		state = states[idx];

	ret = sbi_suspend(state) ? -1 : idx;

	ct_cpuidle_exit();

	if (s2idle)
		dev_pm_genpd_resume(pd_dev);
	else
		pm_runtime_get_sync(pd_dev);

	cpu_pm_exit();

	/* Clear the domain state to start fresh when back from idle. */
	sbi_clear_domain_state();
	return ret;
}

static int sbi_enter_domain_idle_state(struct cpuidle_device *dev,
				       struct cpuidle_driver *drv, int idx)
{
	return __sbi_enter_domain_idle_state(dev, drv, idx, false);
}

static int sbi_enter_s2idle_domain_idle_state(struct cpuidle_device *dev,
					      struct cpuidle_driver *drv,
					      int idx)
{
	return __sbi_enter_domain_idle_state(dev, drv, idx, true);
}

static int sbi_cpuidle_cpuhp_up(unsigned int cpu)
{
	struct device *pd_dev = __this_cpu_read(sbi_cpuidle_data.dev);

	if (pd_dev)
		pm_runtime_get_sync(pd_dev);

	return 0;
}

static int sbi_cpuidle_cpuhp_down(unsigned int cpu)
{
	struct device *pd_dev = __this_cpu_read(sbi_cpuidle_data.dev);

	if (pd_dev) {
		pm_runtime_put_sync(pd_dev);
		/* Clear domain state to start fresh at next online. */
		sbi_clear_domain_state();
	}

	return 0;
}

static void sbi_idle_init_cpuhp(void)
{
	int err;

	if (!sbi_cpuidle_use_cpuhp)
		return;

	err = cpuhp_setup_state_nocalls(CPUHP_AP_CPU_PM_STARTING,
					"cpuidle/sbi:online",
					sbi_cpuidle_cpuhp_up,
					sbi_cpuidle_cpuhp_down);
	if (err)
		pr_warn("Failed %d while setup cpuhp state\n", err);
}

static const struct of_device_id sbi_cpuidle_state_match[] = {
	{ .compatible = "riscv,idle-state",
	  .data = sbi_cpuidle_enter_state },
	{ },
};

static bool sbi_suspend_state_is_valid(u32 state)
{
	if (state > SBI_HSM_SUSPEND_RET_DEFAULT &&
	    state < SBI_HSM_SUSPEND_RET_PLATFORM)
		return false;
	if (state > SBI_HSM_SUSPEND_NON_RET_DEFAULT &&
	    state < SBI_HSM_SUSPEND_NON_RET_PLATFORM)
		return false;
	return true;
}

static int sbi_dt_parse_state_node(struct device_node *np, u32 *state)
{
	int err = of_property_read_u32(np, "riscv,sbi-suspend-param", state);

	if (err) {
		pr_warn("%pOF missing riscv,sbi-suspend-param property\n", np);
		return err;
	}

	if (!sbi_suspend_state_is_valid(*state)) {
		pr_warn("Invalid SBI suspend state %#x\n", *state);
		return -EINVAL;
	}

	return 0;
}

static int sbi_dt_cpu_init_topology(struct cpuidle_driver *drv,
				     struct sbi_cpuidle_data *data,
				     unsigned int state_count, int cpu)
{
	/* Currently limit the hierarchical topology to be used in OSI mode. */
	if (!sbi_cpuidle_use_osi)
		return 0;

	data->dev = dt_idle_attach_cpu(cpu, "sbi");
	if (IS_ERR_OR_NULL(data->dev))
		return PTR_ERR_OR_ZERO(data->dev);

	/*
	 * Using the deepest state for the CPU to trigger a potential selection
	 * of a shared state for the domain, assumes the domain states are all
	 * deeper states.
	 */
	drv->states[state_count - 1].flags |= CPUIDLE_FLAG_RCU_IDLE;
	drv->states[state_count - 1].enter = sbi_enter_domain_idle_state;
	drv->states[state_count - 1].enter_s2idle =
					sbi_enter_s2idle_domain_idle_state;
	sbi_cpuidle_use_cpuhp = true;

	return 0;
}

static int sbi_cpuidle_dt_init_states(struct device *dev,
					struct cpuidle_driver *drv,
					unsigned int cpu,
					unsigned int state_count)
{
	struct sbi_cpuidle_data *data = per_cpu_ptr(&sbi_cpuidle_data, cpu);
	struct device_node *state_node;
	struct device_node *cpu_node;
	u32 *states;
	int i, ret;

	cpu_node = of_cpu_device_node_get(cpu);
	if (!cpu_node)
		return -ENODEV;

	states = devm_kcalloc(dev, state_count, sizeof(*states), GFP_KERNEL);
	if (!states) {
		ret = -ENOMEM;
		goto fail;
	}

	/* Parse SBI specific details from state DT nodes */
	for (i = 1; i < state_count; i++) {
		state_node = of_get_cpu_state_node(cpu_node, i - 1);
		if (!state_node)
			break;

		ret = sbi_dt_parse_state_node(state_node, &states[i]);
		of_node_put(state_node);

		if (ret)
			return ret;

		pr_debug("sbi-state %#x index %d\n", states[i], i);
	}
	if (i != state_count) {
		ret = -ENODEV;
		goto fail;
	}

	/* Initialize optional data, used for the hierarchical topology. */
	ret = sbi_dt_cpu_init_topology(drv, data, state_count, cpu);
	if (ret < 0)
		return ret;

	/* Store states in the per-cpu struct. */
	data->states = states;

fail:
	of_node_put(cpu_node);

	return ret;
}

static void sbi_cpuidle_deinit_cpu(int cpu)
{
	struct sbi_cpuidle_data *data = per_cpu_ptr(&sbi_cpuidle_data, cpu);

	dt_idle_detach_cpu(data->dev);
	sbi_cpuidle_use_cpuhp = false;
}

static int sbi_cpuidle_init_cpu(struct device *dev, int cpu)
{
	struct cpuidle_driver *drv;
	unsigned int state_count = 0;
	int ret = 0;

	drv = devm_kzalloc(dev, sizeof(*drv), GFP_KERNEL);
	if (!drv)
		return -ENOMEM;

	drv->name = "sbi_cpuidle";
	drv->owner = THIS_MODULE;
	drv->cpumask = (struct cpumask *)cpumask_of(cpu);

	/* RISC-V architectural WFI to be represented as state index 0. */
	drv->states[0].enter = sbi_cpuidle_enter_state;
	drv->states[0].exit_latency = 1;
	drv->states[0].target_residency = 1;
	drv->states[0].power_usage = UINT_MAX;
	strcpy(drv->states[0].name, "WFI");
	strcpy(drv->states[0].desc, "RISC-V WFI");

	/*
	 * If no DT idle states are detected (ret == 0) let the driver
	 * initialization fail accordingly since there is no reason to
	 * initialize the idle driver if only wfi is supported, the
	 * default archictectural back-end already executes wfi
	 * on idle entry.
	 */
	ret = dt_init_idle_driver(drv, sbi_cpuidle_state_match, 1);
	if (ret <= 0) {
		pr_debug("HART%ld: failed to parse DT idle states\n",
			 cpuid_to_hartid_map(cpu));
		return ret ? : -ENODEV;
	}
	state_count = ret + 1; /* Include WFI state as well */

	/* Initialize idle states from DT. */
	ret = sbi_cpuidle_dt_init_states(dev, drv, cpu, state_count);
	if (ret) {
		pr_err("HART%ld: failed to init idle states\n",
		       cpuid_to_hartid_map(cpu));
		return ret;
	}

	ret = cpuidle_register(drv, NULL);
	if (ret)
		goto deinit;

	cpuidle_cooling_register(drv);

	return 0;
deinit:
	sbi_cpuidle_deinit_cpu(cpu);
	return ret;
}

static void sbi_cpuidle_domain_sync_state(struct device *dev)
{
	/*
	 * All devices have now been attached/probed to the PM domain
	 * topology, hence it's fine to allow domain states to be picked.
	 */
	sbi_cpuidle_pd_allow_domain_state = true;
}

#ifdef CONFIG_DT_IDLE_GENPD

static int sbi_cpuidle_pd_power_off(struct generic_pm_domain *pd)
{
	struct genpd_power_state *state = &pd->states[pd->state_idx];
	u32 *pd_state;

	if (!state->data)
		return 0;

	if (!sbi_cpuidle_pd_allow_domain_state)
		return -EBUSY;

	/* OSI mode is enabled, set the corresponding domain state. */
	pd_state = state->data;
	sbi_set_domain_state(*pd_state);

	return 0;
}

struct sbi_pd_provider {
	struct list_head link;
	struct device_node *node;
};

static LIST_HEAD(sbi_pd_providers);

static int sbi_pd_init(struct device_node *np)
{
	struct generic_pm_domain *pd;
	struct sbi_pd_provider *pd_provider;
	struct dev_power_governor *pd_gov;
	int ret = -ENOMEM;

	pd = dt_idle_pd_alloc(np, sbi_dt_parse_state_node);
	if (!pd)
		goto out;

	pd_provider = kzalloc(sizeof(*pd_provider), GFP_KERNEL);
	if (!pd_provider)
		goto free_pd;

	pd->flags |= GENPD_FLAG_IRQ_SAFE | GENPD_FLAG_CPU_DOMAIN;

	/* Allow power off when OSI is available. */
	if (sbi_cpuidle_use_osi)
		pd->power_off = sbi_cpuidle_pd_power_off;
	else
		pd->flags |= GENPD_FLAG_ALWAYS_ON;

	/* Use governor for CPU PM domains if it has some states to manage. */
	pd_gov = pd->states ? &pm_domain_cpu_gov : NULL;

	ret = pm_genpd_init(pd, pd_gov, false);
	if (ret)
		goto free_pd_prov;

	ret = of_genpd_add_provider_simple(np, pd);
	if (ret)
		goto remove_pd;

	pd_provider->node = of_node_get(np);
	list_add(&pd_provider->link, &sbi_pd_providers);

	pr_debug("init PM domain %s\n", pd->name);
	return 0;

remove_pd:
	pm_genpd_remove(pd);
free_pd_prov:
	kfree(pd_provider);
free_pd:
	dt_idle_pd_free(pd);
out:
	pr_err("failed to init PM domain ret=%d %pOF\n", ret, np);
	return ret;
}

static void sbi_pd_remove(void)
{
	struct sbi_pd_provider *pd_provider, *it;
	struct generic_pm_domain *genpd;

	list_for_each_entry_safe(pd_provider, it, &sbi_pd_providers, link) {
		of_genpd_del_provider(pd_provider->node);

		genpd = of_genpd_remove_last(pd_provider->node);
		if (!IS_ERR(genpd))
			kfree(genpd);

		of_node_put(pd_provider->node);
		list_del(&pd_provider->link);
		kfree(pd_provider);
	}
}

static int sbi_genpd_probe(struct device_node *np)
{
	struct device_node *node;
	int ret = 0, pd_count = 0;

	if (!np)
		return -ENODEV;

	/*
	 * Parse child nodes for the "#power-domain-cells" property and
	 * initialize a genpd/genpd-of-provider pair when it's found.
	 */
	for_each_child_of_node(np, node) {
		if (!of_property_present(node, "#power-domain-cells"))
			continue;

		ret = sbi_pd_init(node);
		if (ret)
			goto put_node;

		pd_count++;
	}

	/* Bail out if not using the hierarchical CPU topology. */
	if (!pd_count)
		goto no_pd;

	/* Link genpd masters/subdomains to model the CPU topology. */
	ret = dt_idle_pd_init_topology(np);
	if (ret)
		goto remove_pd;

	return 0;

put_node:
	of_node_put(node);
remove_pd:
	sbi_pd_remove();
	pr_err("failed to create CPU PM domains ret=%d\n", ret);
no_pd:
	return ret;
}

#else

static inline int sbi_genpd_probe(struct device_node *np)
{
	return 0;
}

#endif

static int sbi_cpuidle_probe(struct platform_device *pdev)
{
	int cpu, ret;
	struct cpuidle_driver *drv;
	struct cpuidle_device *dev;
	struct device_node *np, *pds_node;

	/* Detect OSI support based on CPU DT nodes */
	sbi_cpuidle_use_osi = true;
	for_each_possible_cpu(cpu) {
		np = of_cpu_device_node_get(cpu);
		if (np &&
		    of_property_present(np, "power-domains") &&
		    of_property_present(np, "power-domain-names")) {
			continue;
		} else {
			sbi_cpuidle_use_osi = false;
			break;
		}
	}

	/* Populate generic power domains from DT nodes */
	pds_node = of_find_node_by_path("/cpus/power-domains");
	if (pds_node) {
		ret = sbi_genpd_probe(pds_node);
		of_node_put(pds_node);
		if (ret)
			return ret;
	}

	/* Initialize CPU idle driver for each CPU */
	for_each_possible_cpu(cpu) {
		ret = sbi_cpuidle_init_cpu(&pdev->dev, cpu);
		if (ret) {
			pr_debug("HART%ld: idle driver init failed\n",
				 cpuid_to_hartid_map(cpu));
			goto out_fail;
		}
	}

	/* Setup CPU hotplut notifiers */
	sbi_idle_init_cpuhp();

	pr_info("idle driver registered for all CPUs\n");

	return 0;

out_fail:
	while (--cpu >= 0) {
		dev = per_cpu(cpuidle_devices, cpu);
		drv = cpuidle_get_cpu_driver(dev);
		cpuidle_unregister(drv);
		sbi_cpuidle_deinit_cpu(cpu);
	}

	return ret;
}

static struct platform_driver sbi_cpuidle_driver = {
	.probe = sbi_cpuidle_probe,
	.driver = {
		.name = "sbi-cpuidle",
		.sync_state = sbi_cpuidle_domain_sync_state,
	},
};

static int __init sbi_cpuidle_init(void)
{
	int ret;
	struct platform_device *pdev;

	/*
	 * The SBI HSM suspend function is only available when:
	 * 1) SBI version is 0.3 or higher
	 * 2) SBI HSM extension is available
	 */
	if ((sbi_spec_version < sbi_mk_version(0, 3)) ||
	    !sbi_probe_extension(SBI_EXT_HSM)) {
		pr_info("HSM suspend not available\n");
		return 0;
	}

	ret = platform_driver_register(&sbi_cpuidle_driver);
	if (ret)
		return ret;

	pdev = platform_device_register_simple("sbi-cpuidle",
						-1, NULL, 0);
	if (IS_ERR(pdev)) {
		platform_driver_unregister(&sbi_cpuidle_driver);
		return PTR_ERR(pdev);
	}

	return 0;
}
device_initcall(sbi_cpuidle_init);
Commit	Line	Data
6abf32f1 AP	1	// SPDX-License-Identifier: GPL-2.0-only
	2	/*
	3	* RISC-V SBI CPU idle driver.
	4	*
	5	* Copyright (c) 2021 Western Digital Corporation or its affiliates.
	6	* Copyright (c) 2022 Ventana Micro Systems Inc.
	7	*/
	8
	9	#define pr_fmt(fmt) "cpuidle-riscv-sbi: " fmt
	10
24760e43	11	#include <linux/cpuhotplug.h>
6abf32f1 AP	12	#include <linux/cpuidle.h>
	13	#include <linux/cpumask.h>
	14	#include <linux/cpu_pm.h>
	15	#include <linux/cpu_cooling.h>
	16	#include <linux/kernel.h>
	17	#include <linux/module.h>
	18	#include <linux/of.h>
6abf32f1 AP	19	#include <linux/slab.h>
	20	#include <linux/platform_device.h>
	21	#include <linux/pm_domain.h>
	22	#include <linux/pm_runtime.h>
	23	#include <asm/cpuidle.h>
	24	#include <asm/sbi.h>
f81f7861	25	#include <asm/smp.h>
6abf32f1 AP	26	#include <asm/suspend.h>
	27
	28	#include "dt_idle_states.h"
	29	#include "dt_idle_genpd.h"
	30
	31	struct sbi_cpuidle_data {
	32	u32 *states;
	33	struct device *dev;
	34	};
	35
	36	struct sbi_domain_state {
	37	bool available;
	38	u32 state;
	39	};
	40
	41	static DEFINE_PER_CPU_READ_MOSTLY(struct sbi_cpuidle_data, sbi_cpuidle_data);
	42	static DEFINE_PER_CPU(struct sbi_domain_state, domain_state);
	43	static bool sbi_cpuidle_use_osi;
	44	static bool sbi_cpuidle_use_cpuhp;
	45	static bool sbi_cpuidle_pd_allow_domain_state;
	46
	47	static inline void sbi_set_domain_state(u32 state)
	48	{
	49	struct sbi_domain_state *data = this_cpu_ptr(&domain_state);
	50
	51	data->available = true;
	52	data->state = state;
	53	}
	54
	55	static inline u32 sbi_get_domain_state(void)
	56	{
	57	struct sbi_domain_state *data = this_cpu_ptr(&domain_state);
	58
	59	return data->state;
	60	}
	61
	62	static inline void sbi_clear_domain_state(void)
	63	{
	64	struct sbi_domain_state *data = this_cpu_ptr(&domain_state);
	65
	66	data->available = false;
	67	}
	68
	69	static inline bool sbi_is_domain_state_available(void)
	70	{
	71	struct sbi_domain_state *data = this_cpu_ptr(&domain_state);
	72
	73	return data->available;
	74	}
	75
	76	static int sbi_suspend_finisher(unsigned long suspend_type,
	77	unsigned long resume_addr,
	78	unsigned long opaque)
	79	{
	80	struct sbiret ret;
	81
	82	ret = sbi_ecall(SBI_EXT_HSM, SBI_EXT_HSM_HART_SUSPEND,
	83	suspend_type, resume_addr, opaque, 0, 0, 0);
	84
	85	return (ret.error) ? sbi_err_map_linux_errno(ret.error) : 0;
	86	}
	87
	88	static int sbi_suspend(u32 state)
	89	{
90	if (state & SBI_HSM_SUSP_NON_RET_BIT)
91	return cpu_suspend(state, sbi_suspend_finisher);
92	else
93	return sbi_suspend_finisher(state, 0, 0);
94	}
95
69e26b4f PZ	96	static __cpuidle int sbi_cpuidle_enter_state(struct cpuidle_device *dev,
69e26b4f PZ	97	struct cpuidle_driver *drv, int idx)
6abf32f1 AP	98	{
6abf32f1 AP	99	u32 *states = __this_cpu_read(sbi_cpuidle_data.states);
cfadbb9d	100	u32 state = states[idx];
6abf32f1	101
cfadbb9d AP	102	if (state & SBI_HSM_SUSP_NON_RET_BIT)
	103	return CPU_PM_CPU_IDLE_ENTER_PARAM(sbi_suspend, idx, state);
	104	else
	105	return CPU_PM_CPU_IDLE_ENTER_RETENTION_PARAM(sbi_suspend,
	106	idx, state);
6abf32f1 AP	107	}
6abf32f1 AP	108
69e26b4f PZ	109	static __cpuidle int __sbi_enter_domain_idle_state(struct cpuidle_device *dev,
	110	struct cpuidle_driver *drv, int idx,
	111	bool s2idle)
6abf32f1 AP	112	{
	113	struct sbi_cpuidle_data *data = this_cpu_ptr(&sbi_cpuidle_data);
	114	u32 *states = data->states;
	115	struct device *pd_dev = data->dev;
	116	u32 state;
	117	int ret;
	118
	119	ret = cpu_pm_enter();
	120	if (ret)
	121	return -1;
	122
	123	/* Do runtime PM to manage a hierarchical CPU toplogy. */
6abf32f1 AP	124	if (s2idle)
	125	dev_pm_genpd_suspend(pd_dev);
	126	else
	127	pm_runtime_put_sync_suspend(pd_dev);
8e9ab9e8	128
a01353cf	129	ct_cpuidle_enter();
6abf32f1 AP	130
	131	if (sbi_is_domain_state_available())
	132	state = sbi_get_domain_state();
	133	else
	134	state = states[idx];
	135
	136	ret = sbi_suspend(state) ? -1 : idx;
	137
a01353cf	138	ct_cpuidle_exit();
8e9ab9e8	139
6abf32f1 AP	140	if (s2idle)
	141	dev_pm_genpd_resume(pd_dev);
	142	else
	143	pm_runtime_get_sync(pd_dev);
6abf32f1 AP	144
	145	cpu_pm_exit();
	146
	147	/* Clear the domain state to start fresh when back from idle. */
	148	sbi_clear_domain_state();
	149	return ret;
	150	}
	151
	152	static int sbi_enter_domain_idle_state(struct cpuidle_device *dev,
	153	struct cpuidle_driver *drv, int idx)
	154	{
	155	return __sbi_enter_domain_idle_state(dev, drv, idx, false);
	156	}
	157
	158	static int sbi_enter_s2idle_domain_idle_state(struct cpuidle_device *dev,
	159	struct cpuidle_driver *drv,
	160	int idx)
	161	{
	162	return __sbi_enter_domain_idle_state(dev, drv, idx, true);
	163	}
	164
	165	static int sbi_cpuidle_cpuhp_up(unsigned int cpu)
	166	{
	167	struct device *pd_dev = __this_cpu_read(sbi_cpuidle_data.dev);
	168
	169	if (pd_dev)
	170	pm_runtime_get_sync(pd_dev);
	171
	172	return 0;
	173	}
	174
	175	static int sbi_cpuidle_cpuhp_down(unsigned int cpu)
	176	{
	177	struct device *pd_dev = __this_cpu_read(sbi_cpuidle_data.dev);
	178
	179	if (pd_dev) {
	180	pm_runtime_put_sync(pd_dev);
	181	/* Clear domain state to start fresh at next online. */
	182	sbi_clear_domain_state();
	183	}
	184
	185	return 0;
	186	}
	187
	188	static void sbi_idle_init_cpuhp(void)
	189	{
	190	int err;
	191
	192	if (!sbi_cpuidle_use_cpuhp)
	193	return;
	194
	195	err = cpuhp_setup_state_nocalls(CPUHP_AP_CPU_PM_STARTING,
	196	"cpuidle/sbi:online",
	197	sbi_cpuidle_cpuhp_up,
	198	sbi_cpuidle_cpuhp_down);
	199	if (err)
	200	pr_warn("Failed %d while setup cpuhp state\n", err);
	201	}
	202
	203	static const struct of_device_id sbi_cpuidle_state_match[] = {
	204	{ .compatible = "riscv,idle-state",
	205	.data = sbi_cpuidle_enter_state },
	206	{ },
	207	};
208
209	static bool sbi_suspend_state_is_valid(u32 state)
210	{
211	if (state > SBI_HSM_SUSPEND_RET_DEFAULT &&
212	state < SBI_HSM_SUSPEND_RET_PLATFORM)
213	return false;
214	if (state > SBI_HSM_SUSPEND_NON_RET_DEFAULT &&
215	state < SBI_HSM_SUSPEND_NON_RET_PLATFORM)
216	return false;
217	return true;
218	}
219
220	static int sbi_dt_parse_state_node(struct device_node np, u32 state)
221	{
222	int err = of_property_read_u32(np, "riscv,sbi-suspend-param", state);
223
224	if (err) {
225	pr_warn("%pOF missing riscv,sbi-suspend-param property\n", np);
226	return err;
227	}
228
229	if (!sbi_suspend_state_is_valid(*state)) {
230	pr_warn("Invalid SBI suspend state %#x\n", *state);
231	return -EINVAL;
232	}
233
234	return 0;
235	}
236
237	static int sbi_dt_cpu_init_topology(struct cpuidle_driver *drv,
238	struct sbi_cpuidle_data *data,
239	unsigned int state_count, int cpu)
240	{
241	/* Currently limit the hierarchical topology to be used in OSI mode. */
242	if (!sbi_cpuidle_use_osi)
243	return 0;
244
245	data->dev = dt_idle_attach_cpu(cpu, "sbi");
246	if (IS_ERR_OR_NULL(data->dev))
247	return PTR_ERR_OR_ZERO(data->dev);
248
249	/*
250	* Using the deepest state for the CPU to trigger a potential selection
251	* of a shared state for the domain, assumes the domain states are all
252	* deeper states.
253	*/
8e9ab9e8	254	drv->states[state_count - 1].flags \|= CPUIDLE_FLAG_RCU_IDLE;
6abf32f1 AP	255	drv->states[state_count - 1].enter = sbi_enter_domain_idle_state;
	256	drv->states[state_count - 1].enter_s2idle =
	257	sbi_enter_s2idle_domain_idle_state;
	258	sbi_cpuidle_use_cpuhp = true;
	259
	260	return 0;
	261	}
	262
	263	static int sbi_cpuidle_dt_init_states(struct device *dev,
	264	struct cpuidle_driver *drv,
	265	unsigned int cpu,
	266	unsigned int state_count)
	267	{
	268	struct sbi_cpuidle_data *data = per_cpu_ptr(&sbi_cpuidle_data, cpu);
	269	struct device_node *state_node;
	270	struct device_node *cpu_node;
	271	u32 *states;
	272	int i, ret;
	273
	274	cpu_node = of_cpu_device_node_get(cpu);
	275	if (!cpu_node)
	276	return -ENODEV;
	277
	278	states = devm_kcalloc(dev, state_count, sizeof(*states), GFP_KERNEL);
	279	if (!states) {
	280	ret = -ENOMEM;
	281	goto fail;
	282	}
	283
	284	/* Parse SBI specific details from state DT nodes */
	285	for (i = 1; i < state_count; i++) {
	286	state_node = of_get_cpu_state_node(cpu_node, i - 1);
	287	if (!state_node)
	288	break;
	289
	290	ret = sbi_dt_parse_state_node(state_node, &states[i]);
	291	of_node_put(state_node);
	292
	293	if (ret)
	294	return ret;
	295
	296	pr_debug("sbi-state %#x index %d\n", states[i], i);
	297	}
	298	if (i != state_count) {
	299	ret = -ENODEV;
	300	goto fail;
	301	}
	302
	303	/* Initialize optional data, used for the hierarchical topology. */
	304	ret = sbi_dt_cpu_init_topology(drv, data, state_count, cpu);
	305	if (ret < 0)
	306	return ret;
	307
	308	/* Store states in the per-cpu struct. */
	309	data->states = states;
	310
	311	fail:
	312	of_node_put(cpu_node);
	313
	314	return ret;
	315	}
	316
	317	static void sbi_cpuidle_deinit_cpu(int cpu)
	318	{
319	struct sbi_cpuidle_data *data = per_cpu_ptr(&sbi_cpuidle_data, cpu);
320
321	dt_idle_detach_cpu(data->dev);
322	sbi_cpuidle_use_cpuhp = false;
323	}
324
325	static int sbi_cpuidle_init_cpu(struct device *dev, int cpu)
326	{
327	struct cpuidle_driver *drv;
328	unsigned int state_count = 0;
329	int ret = 0;
330
331	drv = devm_kzalloc(dev, sizeof(*drv), GFP_KERNEL);
332	if (!drv)
333	return -ENOMEM;
334
335	drv->name = "sbi_cpuidle";
336	drv->owner = THIS_MODULE;
337	drv->cpumask = (struct cpumask *)cpumask_of(cpu);
338
339	/* RISC-V architectural WFI to be represented as state index 0. */
340	drv->states[0].enter = sbi_cpuidle_enter_state;
341	drv->states[0].exit_latency = 1;
342	drv->states[0].target_residency = 1;
343	drv->states[0].power_usage = UINT_MAX;
344	strcpy(drv->states[0].name, "WFI");
345	strcpy(drv->states[0].desc, "RISC-V WFI");
346
347	/*
348	* If no DT idle states are detected (ret == 0) let the driver
349	* initialization fail accordingly since there is no reason to
350	* initialize the idle driver if only wfi is supported, the
351	* default archictectural back-end already executes wfi
352	* on idle entry.
353	*/
354	ret = dt_init_idle_driver(drv, sbi_cpuidle_state_match, 1);
355	if (ret <= 0) {
356	pr_debug("HART%ld: failed to parse DT idle states\n",
357	cpuid_to_hartid_map(cpu));
358	return ret ? : -ENODEV;
359	}
360	state_count = ret + 1; /* Include WFI state as well */
361
362	/* Initialize idle states from DT. */
363	ret = sbi_cpuidle_dt_init_states(dev, drv, cpu, state_count);
364	if (ret) {
365	pr_err("HART%ld: failed to init idle states\n",
366	cpuid_to_hartid_map(cpu));
367	return ret;
368	}
369
370	ret = cpuidle_register(drv, NULL);
371	if (ret)
372	goto deinit;
373
374	cpuidle_cooling_register(drv);
375
376	return 0;
377	deinit:
378	sbi_cpuidle_deinit_cpu(cpu);
379	return ret;
380	}
381
382	static void sbi_cpuidle_domain_sync_state(struct device *dev)
383	{
384	/*
385	* All devices have now been attached/probed to the PM domain
386	* topology, hence it's fine to allow domain states to be picked.
387	*/
388	sbi_cpuidle_pd_allow_domain_state = true;
389	}
390
391	#ifdef CONFIG_DT_IDLE_GENPD
392
393	static int sbi_cpuidle_pd_power_off(struct generic_pm_domain *pd)
394	{
395	struct genpd_power_state *state = &pd->states[pd->state_idx];
396	u32 *pd_state;
397
398	if (!state->data)
399	return 0;
400
401	if (!sbi_cpuidle_pd_allow_domain_state)
402	return -EBUSY;
403
404	/* OSI mode is enabled, set the corresponding domain state. */
405	pd_state = state->data;
406	sbi_set_domain_state(*pd_state);
407
408	return 0;
409	}
410
411	struct sbi_pd_provider {
412	struct list_head link;
413	struct device_node *node;
414	};
415
416	static LIST_HEAD(sbi_pd_providers);
417
418	static int sbi_pd_init(struct device_node *np)
419	{
420	struct generic_pm_domain *pd;
421	struct sbi_pd_provider *pd_provider;
422	struct dev_power_governor *pd_gov;
a6653fb5	423	int ret = -ENOMEM;
6abf32f1 AP	424
	425	pd = dt_idle_pd_alloc(np, sbi_dt_parse_state_node);
	426	if (!pd)
	427	goto out;
	428
	429	pd_provider = kzalloc(sizeof(*pd_provider), GFP_KERNEL);
	430	if (!pd_provider)
	431	goto free_pd;
	432
	433	pd->flags \|= GENPD_FLAG_IRQ_SAFE \| GENPD_FLAG_CPU_DOMAIN;
	434
	435	/* Allow power off when OSI is available. */
	436	if (sbi_cpuidle_use_osi)
	437	pd->power_off = sbi_cpuidle_pd_power_off;
	438	else
	439	pd->flags \|= GENPD_FLAG_ALWAYS_ON;
	440
	441	/* Use governor for CPU PM domains if it has some states to manage. */
a6653fb5	442	pd_gov = pd->states ? &pm_domain_cpu_gov : NULL;
6abf32f1 AP	443
	444	ret = pm_genpd_init(pd, pd_gov, false);
	445	if (ret)
	446	goto free_pd_prov;
	447
	448	ret = of_genpd_add_provider_simple(np, pd);
	449	if (ret)
	450	goto remove_pd;
	451
	452	pd_provider->node = of_node_get(np);
	453	list_add(&pd_provider->link, &sbi_pd_providers);
	454
	455	pr_debug("init PM domain %s\n", pd->name);
	456	return 0;
	457
	458	remove_pd:
	459	pm_genpd_remove(pd);
	460	free_pd_prov:
	461	kfree(pd_provider);
	462	free_pd:
	463	dt_idle_pd_free(pd);
	464	out:
	465	pr_err("failed to init PM domain ret=%d %pOF\n", ret, np);
	466	return ret;
	467	}
	468
	469	static void sbi_pd_remove(void)
	470	{
	471	struct sbi_pd_provider pd_provider, it;
	472	struct generic_pm_domain *genpd;
	473
	474	list_for_each_entry_safe(pd_provider, it, &sbi_pd_providers, link) {
	475	of_genpd_del_provider(pd_provider->node);
	476
	477	genpd = of_genpd_remove_last(pd_provider->node);
	478	if (!IS_ERR(genpd))
	479	kfree(genpd);
	480
	481	of_node_put(pd_provider->node);
	482	list_del(&pd_provider->link);
	483	kfree(pd_provider);
	484	}
	485	}
	486
	487	static int sbi_genpd_probe(struct device_node *np)
	488	{
	489	struct device_node *node;
	490	int ret = 0, pd_count = 0;
	491
	492	if (!np)
	493	return -ENODEV;
	494
	495	/*
	496	* Parse child nodes for the "#power-domain-cells" property and
	497	* initialize a genpd/genpd-of-provider pair when it's found.
	498	*/
	499	for_each_child_of_node(np, node) {
f914bfdd	500	if (!of_property_present(node, "#power-domain-cells"))
6abf32f1 AP	501	continue;
	502
	503	ret = sbi_pd_init(node);
	504	if (ret)
	505	goto put_node;
	506
	507	pd_count++;
	508	}
	509
	510	/* Bail out if not using the hierarchical CPU topology. */
	511	if (!pd_count)
	512	goto no_pd;
	513
	514	/* Link genpd masters/subdomains to model the CPU topology. */
	515	ret = dt_idle_pd_init_topology(np);
	516	if (ret)
	517	goto remove_pd;
	518
	519	return 0;
	520
	521	put_node:
	522	of_node_put(node);
	523	remove_pd:
	524	sbi_pd_remove();
	525	pr_err("failed to create CPU PM domains ret=%d\n", ret);
	526	no_pd:
	527	return ret;
	528	}
	529
	530	#else
	531
	532	static inline int sbi_genpd_probe(struct device_node *np)
	533	{
	534	return 0;
	535	}
	536
	537	#endif
	538
	539	static int sbi_cpuidle_probe(struct platform_device *pdev)
	540	{
	541	int cpu, ret;
	542	struct cpuidle_driver *drv;
	543	struct cpuidle_device *dev;
	544	struct device_node np, pds_node;
	545
	546	/* Detect OSI support based on CPU DT nodes */
	547	sbi_cpuidle_use_osi = true;
	548	for_each_possible_cpu(cpu) {
	549	np = of_cpu_device_node_get(cpu);
	550	if (np &&
f914bfdd RH	551	of_property_present(np, "power-domains") &&
f914bfdd RH	552	of_property_present(np, "power-domain-names")) {
6abf32f1 AP	553	continue;
	554	} else {
	555	sbi_cpuidle_use_osi = false;
	556	break;
	557	}
	558	}
	559
	560	/* Populate generic power domains from DT nodes */
	561	pds_node = of_find_node_by_path("/cpus/power-domains");
	562	if (pds_node) {
	563	ret = sbi_genpd_probe(pds_node);
	564	of_node_put(pds_node);
	565	if (ret)
	566	return ret;
	567	}
	568
	569	/* Initialize CPU idle driver for each CPU */
	570	for_each_possible_cpu(cpu) {
	571	ret = sbi_cpuidle_init_cpu(&pdev->dev, cpu);
	572	if (ret) {
	573	pr_debug("HART%ld: idle driver init failed\n",
	574	cpuid_to_hartid_map(cpu));
	575	goto out_fail;
	576	}
	577	}
	578
	579	/* Setup CPU hotplut notifiers */
	580	sbi_idle_init_cpuhp();
	581
	582	pr_info("idle driver registered for all CPUs\n");
	583
	584	return 0;
	585
	586	out_fail:
	587	while (--cpu >= 0) {
	588	dev = per_cpu(cpuidle_devices, cpu);
	589	drv = cpuidle_get_cpu_driver(dev);
	590	cpuidle_unregister(drv);
	591	sbi_cpuidle_deinit_cpu(cpu);
	592	}
	593
	594	return ret;
	595	}
	596
	597	static struct platform_driver sbi_cpuidle_driver = {
	598	.probe = sbi_cpuidle_probe,
	599	.driver = {
	600	.name = "sbi-cpuidle",
	601	.sync_state = sbi_cpuidle_domain_sync_state,
	602	},
	603	};
	604
	605	static int __init sbi_cpuidle_init(void)
	606	{
	607	int ret;
	608	struct platform_device *pdev;
	609
	610	/*
	611	* The SBI HSM suspend function is only available when:
	612	* 1) SBI version is 0.3 or higher
	613	* 2) SBI HSM extension is available
	614	*/
	615	if ((sbi_spec_version < sbi_mk_version(0, 3)) \|\|
41cad828	616	!sbi_probe_extension(SBI_EXT_HSM)) {
6abf32f1 AP	617	pr_info("HSM suspend not available\n");
	618	return 0;
	619	}
	620
	621	ret = platform_driver_register(&sbi_cpuidle_driver);
	622	if (ret)
	623	return ret;
	624
	625	pdev = platform_device_register_simple("sbi-cpuidle",
	626	-1, NULL, 0);
	627	if (IS_ERR(pdev)) {
	628	platform_driver_unregister(&sbi_cpuidle_driver);
	629	return PTR_ERR(pdev);
	630	}
	631
	632	return 0;
	633	}
	634	device_initcall(sbi_cpuidle_init);