Merge tag 'pm-5.20-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

Pull power management updates from Rafael Wysocki:
 "These are mostly minor improvements all over including new CPU IDs for
  the Intel RAPL driver, an Energy Model rework to use micro-Watt as the
  power unit, cpufreq fixes and cleanus, cpuidle updates, devfreq
  updates, documentation cleanups and a new version of the pm-graph
  suite of utilities.

  Specifics:

   - Make cpufreq_show_cpus() more straightforward (Viresh Kumar).

   - Drop unnecessary CPU hotplug locking from store() used by cpufreq
     sysfs attributes (Viresh Kumar).

   - Make the ACPI cpufreq driver support the boost control interface on
     Zhaoxin/Centaur processors (Tony W Wang-oc).

   - Print a warning message on attempts to free an active cpufreq
     policy which should never happen (Viresh Kumar).

   - Fix grammar in the Kconfig help text for the loongson2 cpufreq
     driver (Randy Dunlap).

   - Use cpumask_var_t for an on-stack CPU mask in the ondemand cpufreq
     governor (Zhao Liu).

   - Add trace points for guest_halt_poll_ns grow/shrink to the haltpoll
     cpuidle driver (Eiichi Tsukata).

   - Modify intel_idle to treat C1 and C1E as independent idle states on
     Sapphire Rapids (Artem Bityutskiy).

   - Extend support for wakeirq to callback wrappers used during system
     suspend and resume (Ulf Hansson).

   - Defer waiting for device probe before loading a hibernation image
     till the first actual device access to avoid possible deadlocks
     reported by syzbot (Tetsuo Handa).

   - Unify device_init_wakeup() for PM_SLEEP and !PM_SLEEP (Bjorn
     Helgaas).

   - Add Raptor Lake-P to the list of processors supported by the Intel
     RAPL driver (George D Sworo).

   - Add Alder Lake-N and Raptor Lake-P to the list of processors for
     which Power Limit4 is supported in the Intel RAPL driver (Sumeet
     Pawnikar).

   - Make pm_genpd_remove() check genpd_debugfs_dir against NULL before
     attempting to remove it (Hsin-Yi Wang).

   - Change the Energy Model code to represent power in micro-Watts and
     adjust its users accordingly (Lukasz Luba).

   - Add new devfreq driver for Mediatek CCI (Cache Coherent
     Interconnect) (Johnson Wang).

   - Convert the Samsung Exynos SoC Bus bindings to DT schema of
     exynos-bus.c (Krzysztof Kozlowski).

   - Address kernel-doc warnings by adding the description for unused
     function parameters in devfreq core (Mauro Carvalho Chehab).

   - Use NULL to pass a null pointer rather than zero according to the
     function propotype in imx-bus.c (Colin Ian King).

   - Print error message instead of error interger value in
     tegra30-devfreq.c (Dmitry Osipenko).

   - Add checks to prevent setting negative frequency QoS limits for
     CPUs (Shivnandan Kumar).

   - Update the pm-graph suite of utilities to the latest revision 5.9
     including multiple improvements (Todd Brandt).

   - Drop pme_interrupt reference from the PCI power management
     documentation (Mario Limonciello)"

* tag 'pm-5.20-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (27 commits)
  powercap: RAPL: Add Power Limit4 support for Alder Lake-N and Raptor Lake-P
  PM: QoS: Add check to make sure CPU freq is non-negative
  PM: hibernate: defer device probing when resuming from hibernation
  intel_idle: make SPR C1 and C1E be independent
  cpufreq: ondemand: Use cpumask_var_t for on-stack cpu mask
  cpufreq: loongson2: fix Kconfig "its" grammar
  pm-graph v5.9
  cpufreq: Warn users while freeing active policy
  cpufreq: scmi: Support the power scale in micro-Watts in SCMI v3.1
  firmware: arm_scmi: Get detailed power scale from perf
  Documentation: EM: Switch to micro-Watts scale
  PM: EM: convert power field to micro-Watts precision and align drivers
  PM / devfreq: tegra30: Add error message for devm_devfreq_add_device()
  PM / devfreq: imx-bus: use NULL to pass a null pointer rather than zero
  PM / devfreq: shut up kernel-doc warnings
  dt-bindings: interconnect: samsung,exynos-bus: convert to dtschema
  PM / devfreq: mediatek: Introduce MediaTek CCI devfreq driver
  dt-bindings: interconnect: Add MediaTek CCI dt-bindings
  PM: domains: Ensure genpd_debugfs_dir exists before remove
  PM: runtime: Extend support for wakeirq for force_suspend|resume
  ...

24 files changed, 570 insertions, 108 deletions

diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c
index 739e52cd4aba..55a10e6d4e2a 100644
--- a/drivers/base/power/domain.c
+++ b/drivers/base/power/domain.c
@@ -222,6 +222,9 @@ static void genpd_debug_remove(struct generic_pm_domain *genpd)
 {
 	struct dentry *d;
 
+	if (!genpd_debugfs_dir)
+		return;
+
 	d = debugfs_lookup(genpd->name, genpd_debugfs_dir);
 	debugfs_remove(d);
 }
diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c
index 949907e2e242..997be3ac20a7 100644
--- a/drivers/base/power/runtime.c
+++ b/drivers/base/power/runtime.c
@@ -1862,10 +1862,13 @@ int pm_runtime_force_suspend(struct device *dev)
 
 	callback = RPM_GET_CALLBACK(dev, runtime_suspend);
 
+	dev_pm_enable_wake_irq_check(dev, true);
 	ret = callback ? callback(dev) : 0;
 	if (ret)
 		goto err;
 
+	dev_pm_enable_wake_irq_complete(dev);
+
 	/*
 	 * If the device can stay in suspend after the system-wide transition
 	 * to the working state that will follow, drop the children counter of
@@ -1882,6 +1885,7 @@ int pm_runtime_force_suspend(struct device *dev)
 	return 0;
 
 err:
+	dev_pm_disable_wake_irq_check(dev, true);
 	pm_runtime_enable(dev);
 	return ret;
 }
@@ -1915,9 +1919,11 @@ int pm_runtime_force_resume(struct device *dev)
 
 	callback = RPM_GET_CALLBACK(dev, runtime_resume);
 
+	dev_pm_disable_wake_irq_check(dev, false);
 	ret = callback ? callback(dev) : 0;
 	if (ret) {
 		pm_runtime_set_suspended(dev);
+		dev_pm_enable_wake_irq_check(dev, false);
 		goto out;
 	}
 
diff --git a/drivers/base/power/wakeup.c b/drivers/base/power/wakeup.c
index 11a4ffe91367..e3befa2c1b66 100644
--- a/drivers/base/power/wakeup.c
+++ b/drivers/base/power/wakeup.c
@@ -501,36 +501,6 @@ void device_set_wakeup_capable(struct device *dev, bool capable)
 EXPORT_SYMBOL_GPL(device_set_wakeup_capable);
 
 /**
- * device_init_wakeup - Device wakeup initialization.
- * @dev: Device to handle.
- * @enable: Whether or not to enable @dev as a wakeup device.
- *
- * By default, most devices should leave wakeup disabled.  The exceptions are
- * devices that everyone expects to be wakeup sources: keyboards, power buttons,
- * possibly network interfaces, etc.  Also, devices that don't generate their
- * own wakeup requests but merely forward requests from one bus to another
- * (like PCI bridges) should have wakeup enabled by default.
- */
-int device_init_wakeup(struct device *dev, bool enable)
-{
-	int ret = 0;
-
-	if (!dev)
-		return -EINVAL;
-
-	if (enable) {
-		device_set_wakeup_capable(dev, true);
-		ret = device_wakeup_enable(dev);
-	} else {
-		device_wakeup_disable(dev);
-		device_set_wakeup_capable(dev, false);
-	}
-
-	return ret;
-}
-EXPORT_SYMBOL_GPL(device_init_wakeup);
-
-/**
  * device_set_wakeup_enable - Enable or disable a device to wake up the system.
  * @dev: Device to handle.
  * @enable: enable/disable flag
diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
index c3038cdc6865..2a84fc63371e 100644
--- a/drivers/cpufreq/Kconfig
+++ b/drivers/cpufreq/Kconfig
@@ -268,7 +268,7 @@ config LOONGSON2_CPUFREQ
 	  This option adds a CPUFreq driver for loongson processors which
 	  support software configurable cpu frequency.
 
-	  Loongson2F and it's successors support this feature.
+	  Loongson2F and its successors support this feature.
 
 	  If in doubt, say N.
 
diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c
index 3d514b82d055..1bb2b90ebb21 100644
--- a/drivers/cpufreq/acpi-cpufreq.c
+++ b/drivers/cpufreq/acpi-cpufreq.c
@@ -78,6 +78,8 @@ static bool boost_state(unsigned int cpu)
 
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_INTEL:
+	case X86_VENDOR_CENTAUR:
+	case X86_VENDOR_ZHAOXIN:
 		rdmsr_on_cpu(cpu, MSR_IA32_MISC_ENABLE, &lo, &hi);
 		msr = lo | ((u64)hi << 32);
 		return !(msr & MSR_IA32_MISC_ENABLE_TURBO_DISABLE);
@@ -97,6 +99,8 @@ static int boost_set_msr(bool enable)
 
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_INTEL:
+	case X86_VENDOR_CENTAUR:
+	case X86_VENDOR_ZHAOXIN:
 		msr_addr = MSR_IA32_MISC_ENABLE;
 		msr_mask = MSR_IA32_MISC_ENABLE_TURBO_DISABLE;
 		break;
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 2cad42774164..954eef26685f 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -843,12 +843,14 @@ ssize_t cpufreq_show_cpus(const struct cpumask *mask, char *buf)
 	unsigned int cpu;
 
 	for_each_cpu(cpu, mask) {
-		if (i)
-			i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), " ");
-		i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), "%u", cpu);
+		i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), "%u ", cpu);
 		if (i >= (PAGE_SIZE - 5))
 			break;
 	}
+
+	/* Remove the extra space at the end */
+	i--;
+
 	i += sprintf(&buf[i], "\n");
 	return i;
 }
@@ -971,21 +973,10 @@ static ssize_t store(struct kobject *kobj, struct attribute *attr,
 	if (!fattr->store)
 		return -EIO;
 
-	/*
-	 * cpus_read_trylock() is used here to work around a circular lock
-	 * dependency problem with respect to the cpufreq_register_driver().
-	 */
-	if (!cpus_read_trylock())
-		return -EBUSY;
-
-	if (cpu_online(policy->cpu)) {
-		down_write(&policy->rwsem);
-		if (likely(!policy_is_inactive(policy)))
-			ret = fattr->store(policy, buf, count);
-		up_write(&policy->rwsem);
-	}
-
-	cpus_read_unlock();
+	down_write(&policy->rwsem);
+	if (likely(!policy_is_inactive(policy)))
+		ret = fattr->store(policy, buf, count);
+	up_write(&policy->rwsem);
 
 	return ret;
 }
@@ -1282,6 +1273,13 @@ static void cpufreq_policy_free(struct cpufreq_policy *policy)
 	unsigned long flags;
 	int cpu;
 
+	/*
+	 * The callers must ensure the policy is inactive by now, to avoid any
+	 * races with show()/store() callbacks.
+	 */
+	if (unlikely(!policy_is_inactive(policy)))
+		pr_warn("%s: Freeing active policy\n", __func__);
+
 	/* Remove policy from list */
 	write_lock_irqsave(&cpufreq_driver_lock, flags);
 	list_del(&policy->policy_list);
@@ -1536,8 +1534,6 @@ out_destroy_policy:
 	for_each_cpu(j, policy->real_cpus)
 		remove_cpu_dev_symlink(policy, j, get_cpu_device(j));
 
-	cpumask_clear(policy->cpus);
-
 out_offline_policy:
 	if (cpufreq_driver->offline)
 		cpufreq_driver->offline(policy);
@@ -1547,6 +1543,7 @@ out_exit_policy:
 		cpufreq_driver->exit(policy);
 
 out_free_policy:
+	cpumask_clear(policy->cpus);
 	up_write(&policy->rwsem);
 
 	cpufreq_policy_free(policy);
diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index e8fbf970ff07..c52d19d67557 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -416,10 +416,13 @@ static struct dbs_governor od_dbs_gov = {
 static void od_set_powersave_bias(unsigned int powersave_bias)
 {
 	unsigned int cpu;
-	cpumask_t done;
+	cpumask_var_t done;
+
+	if (!alloc_cpumask_var(&done, GFP_KERNEL))
+		return;
 
 	default_powersave_bias = powersave_bias;
-	cpumask_clear(&done);
+	cpumask_clear(done);
 
 	cpus_read_lock();
 	for_each_online_cpu(cpu) {
@@ -428,7 +431,7 @@ static void od_set_powersave_bias(unsigned int powersave_bias)
 		struct dbs_data *dbs_data;
 		struct od_dbs_tuners *od_tuners;
 
-		if (cpumask_test_cpu(cpu, &done))
+		if (cpumask_test_cpu(cpu, done))
 			continue;
 
 		policy = cpufreq_cpu_get_raw(cpu);
@@ -439,13 +442,15 @@ static void od_set_powersave_bias(unsigned int powersave_bias)
 		if (!policy_dbs)
 			continue;
 
-		cpumask_or(&done, &done, policy->cpus);
+		cpumask_or(done, done, policy->cpus);
 
 		dbs_data = policy_dbs->dbs_data;
 		od_tuners = dbs_data->tuners;
 		od_tuners->powersave_bias = default_powersave_bias;
 	}
 	cpus_read_unlock();
+
+	free_cpumask_var(done);
 }
 
 void od_register_powersave_bias_handler(unsigned int (*f)
diff --git a/drivers/cpufreq/mediatek-cpufreq-hw.c b/drivers/cpufreq/mediatek-cpufreq-hw.c
index 813cccbfe934..f0e0a35c7f21 100644
--- a/drivers/cpufreq/mediatek-cpufreq-hw.c
+++ b/drivers/cpufreq/mediatek-cpufreq-hw.c
@@ -51,7 +51,7 @@ static const u16 cpufreq_mtk_offsets[REG_ARRAY_SIZE] = {
 };
 
 static int __maybe_unused
-mtk_cpufreq_get_cpu_power(struct device *cpu_dev, unsigned long *mW,
+mtk_cpufreq_get_cpu_power(struct device *cpu_dev, unsigned long *uW,
 			  unsigned long *KHz)
 {
 	struct mtk_cpufreq_data *data;
@@ -71,8 +71,9 @@ mtk_cpufreq_get_cpu_power(struct device *cpu_dev, unsigned long *mW,
 	i--;
 
 	*KHz = data->table[i].frequency;
-	*mW = readl_relaxed(data->reg_bases[REG_EM_POWER_TBL] +
-			    i * LUT_ROW_SIZE) / 1000;
+	/* Provide micro-Watts value to the Energy Model */
+	*uW = readl_relaxed(data->reg_bases[REG_EM_POWER_TBL] +
+			    i * LUT_ROW_SIZE);
 
 	return 0;
 }
diff --git a/drivers/cpufreq/scmi-cpufreq.c b/drivers/cpufreq/scmi-cpufreq.c
index 6d2a4cf46db7..513a071845c2 100644
--- a/drivers/cpufreq/scmi-cpufreq.c
+++ b/drivers/cpufreq/scmi-cpufreq.c
@@ -19,6 +19,7 @@
 #include <linux/slab.h>
 #include <linux/scmi_protocol.h>
 #include <linux/types.h>
+#include <linux/units.h>
 
 struct scmi_data {
 	int domain_id;
@@ -99,6 +100,7 @@ static int __maybe_unused
 scmi_get_cpu_power(struct device *cpu_dev, unsigned long *power,
 		   unsigned long *KHz)
 {
+	enum scmi_power_scale power_scale = perf_ops->power_scale_get(ph);
 	unsigned long Hz;
 	int ret, domain;
 
@@ -112,6 +114,10 @@ scmi_get_cpu_power(struct device *cpu_dev, unsigned long *power,
 	if (ret)
 		return ret;
 
+	/* Convert the power to uW if it is mW (ignore bogoW) */
+	if (power_scale == SCMI_POWER_MILLIWATTS)
+		*power *= MICROWATT_PER_MILLIWATT;
+
 	/* The EM framework specifies the frequency in KHz. */
 	*KHz = Hz / 1000;
 
@@ -249,8 +255,9 @@ static int scmi_cpufreq_exit(struct cpufreq_policy *policy)
 static void scmi_cpufreq_register_em(struct cpufreq_policy *policy)
 {
 	struct em_data_callback em_cb = EM_DATA_CB(scmi_get_cpu_power);
-	bool power_scale_mw = perf_ops->power_scale_mw_get(ph);
+	enum scmi_power_scale power_scale = perf_ops->power_scale_get(ph);
 	struct scmi_data *priv = policy->driver_data;
+	bool em_power_scale = false;
 
 	/*
 	 * This callback will be called for each policy, but we don't need to
@@ -262,9 +269,13 @@ static void scmi_cpufreq_register_em(struct cpufreq_policy *policy)
 	if (!priv->nr_opp)
 		return;
 
+	if (power_scale == SCMI_POWER_MILLIWATTS
+	    || power_scale == SCMI_POWER_MICROWATTS)
+		em_power_scale = true;
+
 	em_dev_register_perf_domain(get_cpu_device(policy->cpu), priv->nr_opp,
 				    &em_cb, priv->opp_shared_cpus,
-				    power_scale_mw);
+				    em_power_scale);
 }
 
 static struct cpufreq_driver scmi_cpufreq_driver = {
diff --git a/drivers/cpuidle/governors/haltpoll.c b/drivers/cpuidle/governors/haltpoll.c
index cb2a96eafc02..1dff3a52917d 100644
--- a/drivers/cpuidle/governors/haltpoll.c
+++ b/drivers/cpuidle/governors/haltpoll.c
@@ -19,6 +19,7 @@
 #include <linux/sched.h>
 #include <linux/module.h>
 #include <linux/kvm_para.h>
+#include <trace/events/power.h>
 
 static unsigned int guest_halt_poll_ns __read_mostly = 200000;
 module_param(guest_halt_poll_ns, uint, 0644);
@@ -90,6 +91,7 @@ static void adjust_poll_limit(struct cpuidle_device *dev, u64 block_ns)
 		if (val > guest_halt_poll_ns)
 			val = guest_halt_poll_ns;
 
+		trace_guest_halt_poll_ns_grow(val, dev->poll_limit_ns);
 		dev->poll_limit_ns = val;
 	} else if (block_ns > guest_halt_poll_ns &&
 		   guest_halt_poll_allow_shrink) {
@@ -100,6 +102,7 @@ static void adjust_poll_limit(struct cpuidle_device *dev, u64 block_ns)
 			val = 0;
 		else
 			val /= shrink;
+		trace_guest_halt_poll_ns_shrink(val, dev->poll_limit_ns);
 		dev->poll_limit_ns = val;
 	}
 }
diff --git a/drivers/devfreq/Kconfig b/drivers/devfreq/Kconfig
index 87eb2b837e68..9754d8b31621 100644
--- a/drivers/devfreq/Kconfig
+++ b/drivers/devfreq/Kconfig
@@ -120,6 +120,16 @@ config ARM_TEGRA_DEVFREQ
 	  It reads ACTMON counters of memory controllers and adjusts the
 	  operating frequencies and voltages with OPP support.
 
+config ARM_MEDIATEK_CCI_DEVFREQ
+	tristate "MEDIATEK CCI DEVFREQ Driver"
+	depends on ARM_MEDIATEK_CPUFREQ || COMPILE_TEST
+	select DEVFREQ_GOV_PASSIVE
+	help
+	  This adds a devfreq driver for MediaTek Cache Coherent Interconnect
+	  which is shared the same regulators with the cpu cluster. It can track
+	  buck voltages and update a proper CCI frequency. Use the notification
+	  to get the regulator status.
+
 config ARM_RK3399_DMC_DEVFREQ
 	tristate "ARM RK3399 DMC DEVFREQ Driver"
 	depends on (ARCH_ROCKCHIP && HAVE_ARM_SMCCC) || \
diff --git a/drivers/devfreq/Makefile b/drivers/devfreq/Makefile
index 0b6be92a25d9..bf40d04928d0 100644
--- a/drivers/devfreq/Makefile
+++ b/drivers/devfreq/Makefile
@@ -11,6 +11,7 @@ obj-$(CONFIG_DEVFREQ_GOV_PASSIVE)	+= governor_passive.o
 obj-$(CONFIG_ARM_EXYNOS_BUS_DEVFREQ)	+= exynos-bus.o
 obj-$(CONFIG_ARM_IMX_BUS_DEVFREQ)	+= imx-bus.o
 obj-$(CONFIG_ARM_IMX8M_DDRC_DEVFREQ)	+= imx8m-ddrc.o
+obj-$(CONFIG_ARM_MEDIATEK_CCI_DEVFREQ)	+= mtk-cci-devfreq.o
 obj-$(CONFIG_ARM_RK3399_DMC_DEVFREQ)	+= rk3399_dmc.o
 obj-$(CONFIG_ARM_SUN8I_A33_MBUS_DEVFREQ)	+= sun8i-a33-mbus.o
 obj-$(CONFIG_ARM_TEGRA_DEVFREQ)		+= tegra30-devfreq.o
diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c
index 9602141bb8ec..63347a5ae599 100644
--- a/drivers/devfreq/devfreq.c
+++ b/drivers/devfreq/devfreq.c
@@ -696,6 +696,8 @@ static int qos_notifier_call(struct devfreq *devfreq)
 /**
  * qos_min_notifier_call() - Callback for QoS min_freq changes.
  * @nb:		Should be devfreq->nb_min
+ * @val:	not used
+ * @ptr:	not used
  */
 static int qos_min_notifier_call(struct notifier_block *nb,
 					 unsigned long val, void *ptr)
@@ -706,6 +708,8 @@ static int qos_min_notifier_call(struct notifier_block *nb,
 /**
  * qos_max_notifier_call() - Callback for QoS max_freq changes.
  * @nb:		Should be devfreq->nb_max
+ * @val:	not used
+ * @ptr:	not used
  */
 static int qos_max_notifier_call(struct notifier_block *nb,
 					 unsigned long val, void *ptr)
diff --git a/drivers/devfreq/imx-bus.c b/drivers/devfreq/imx-bus.c
index f3f6e25053ed..f87067fc574d 100644
--- a/drivers/devfreq/imx-bus.c
+++ b/drivers/devfreq/imx-bus.c
@@ -59,7 +59,7 @@ static int imx_bus_init_icc(struct device *dev)
 	struct imx_bus *priv = dev_get_drvdata(dev);
 	const char *icc_driver_name;
 
-	if (!of_get_property(dev->of_node, "#interconnect-cells", 0))
+	if (!of_get_property(dev->of_node, "#interconnect-cells", NULL))
 		return 0;
 	if (!IS_ENABLED(CONFIG_INTERCONNECT_IMX)) {
 		dev_warn(dev, "imx interconnect drivers disabled\n");
diff --git a/drivers/devfreq/mtk-cci-devfreq.c b/drivers/devfreq/mtk-cci-devfreq.c
new file mode 100644
index 000000000000..71abb3fbd042
--- /dev/null
+++ b/drivers/devfreq/mtk-cci-devfreq.c
@@ -0,0 +1,440 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2022 MediaTek Inc.
+ */
+
+#include <linux/clk.h>
+#include <linux/devfreq.h>
+#include <linux/minmax.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include <linux/pm_opp.h>
+#include <linux/regulator/consumer.h>
+
+struct mtk_ccifreq_platform_data {
+	int min_volt_shift;
+	int max_volt_shift;
+	int proc_max_volt;
+	int sram_min_volt;
+	int sram_max_volt;
+};
+
+struct mtk_ccifreq_drv {
+	struct device *dev;
+	struct devfreq *devfreq;
+	struct regulator *proc_reg;
+	struct regulator *sram_reg;
+	struct clk *cci_clk;
+	struct clk *inter_clk;
+	int inter_voltage;
+	unsigned long pre_freq;
+	/* Avoid race condition for regulators between notify and policy */
+	struct mutex reg_lock;
+	struct notifier_block opp_nb;
+	const struct mtk_ccifreq_platform_data *soc_data;
+	int vtrack_max;
+};
+
+static int mtk_ccifreq_set_voltage(struct mtk_ccifreq_drv *drv, int new_voltage)
+{
+	const struct mtk_ccifreq_platform_data *soc_data = drv->soc_data;
+	struct device *dev = drv->dev;
+	int pre_voltage, pre_vsram, new_vsram, vsram, voltage, ret;
+	int retry_max = drv->vtrack_max;
+
+	if (!drv->sram_reg) {
+		ret = regulator_set_voltage(drv->proc_reg, new_voltage,
+					    drv->soc_data->proc_max_volt);
+		return ret;
+	}
+
+	pre_voltage = regulator_get_voltage(drv->proc_reg);
+	if (pre_voltage < 0) {
+		dev_err(dev, "invalid vproc value: %d\n", pre_voltage);
+		return pre_voltage;
+	}
+
+	pre_vsram = regulator_get_voltage(drv->sram_reg);
+	if (pre_vsram < 0) {
+		dev_err(dev, "invalid vsram value: %d\n", pre_vsram);
+		return pre_vsram;
+	}
+
+	new_vsram = clamp(new_voltage + soc_data->min_volt_shift,
+			  soc_data->sram_min_volt, soc_data->sram_max_volt);
+
+	do {
+		if (pre_voltage <= new_voltage) {
+			vsram = clamp(pre_voltage + soc_data->max_volt_shift,
+				      soc_data->sram_min_volt, new_vsram);
+			ret = regulator_set_voltage(drv->sram_reg, vsram,
+						    soc_data->sram_max_volt);
+			if (ret)
+				return ret;
+
+			if (vsram == soc_data->sram_max_volt ||
+			    new_vsram == soc_data->sram_min_volt)
+				voltage = new_voltage;
+			else
+				voltage = vsram - soc_data->min_volt_shift;
+
+			ret = regulator_set_voltage(drv->proc_reg, voltage,
+						    soc_data->proc_max_volt);
+			if (ret) {
+				regulator_set_voltage(drv->sram_reg, pre_vsram,
+						      soc_data->sram_max_volt);
+				return ret;
+			}
+		} else if (pre_voltage > new_voltage) {
+			voltage = max(new_voltage,
+				      pre_vsram - soc_data->max_volt_shift);
+			ret = regulator_set_voltage(drv->proc_reg, voltage,
+						    soc_data->proc_max_volt);
+			if (ret)
+				return ret;
+
+			if (voltage == new_voltage)
+				vsram = new_vsram;
+			else
+				vsram = max(new_vsram,
+					    voltage + soc_data->min_volt_shift);
+
+			ret = regulator_set_voltage(drv->sram_reg, vsram,
+						    soc_data->sram_max_volt);
+			if (ret) {
+				regulator_set_voltage(drv->proc_reg, pre_voltage,
+						      soc_data->proc_max_volt);
+				return ret;
+			}
+		}
+
+		pre_voltage = voltage;
+		pre_vsram = vsram;
+
+		if (--retry_max < 0) {
+			dev_err(dev,
+				"over loop count, failed to set voltage\n");
+			return -EINVAL;
+		}
+	} while (voltage != new_voltage || vsram != new_vsram);
+
+	return 0;
+}
+
+static int mtk_ccifreq_target(struct device *dev, unsigned long *freq,
+			      u32 flags)
+{
+	struct mtk_ccifreq_drv *drv = dev_get_drvdata(dev);
+	struct clk *cci_pll = clk_get_parent(drv->cci_clk);
+	struct dev_pm_opp *opp;
+	unsigned long opp_rate;
+	int voltage, pre_voltage, inter_voltage, target_voltage, ret;
+
+	if (!drv)
+		return -EINVAL;
+
+	if (drv->pre_freq == *freq)
+		return 0;
+
+	inter_voltage = drv->inter_voltage;
+
+	opp_rate = *freq;
+	opp = devfreq_recommended_opp(dev, &opp_rate, 1);
+	if (IS_ERR(opp)) {
+		dev_err(dev, "failed to find opp for freq: %ld\n", opp_rate);
+		return PTR_ERR(opp);
+	}
+
+	mutex_lock(&drv->reg_lock);
+
+	voltage = dev_pm_opp_get_voltage(opp);
+	dev_pm_opp_put(opp);
+
+	pre_voltage = regulator_get_voltage(drv->proc_reg);
+	if (pre_voltage < 0) {
+		dev_err(dev, "invalid vproc value: %d\n", pre_voltage);
+		ret = pre_voltage;
+		goto out_unlock;
+	}
+
+	/* scale up: set voltage first then freq. */
+	target_voltage = max(inter_voltage, voltage);
+	if (pre_voltage <= target_voltage) {
+		ret = mtk_ccifreq_set_voltage(drv, target_voltage);
+		if (ret) {
+			dev_err(dev, "failed to scale up voltage\n");
+			goto out_restore_voltage;
+		}
+	}
+
+	/* switch the cci clock to intermediate clock source. */
+	ret = clk_set_parent(drv->cci_clk, drv->inter_clk);
+	if (ret) {
+		dev_err(dev, "failed to re-parent cci clock\n");
+		goto out_restore_voltage;
+	}
+
+	/* set the original clock to target rate. */
+	ret = clk_set_rate(cci_pll, *freq);
+	if (ret) {
+		dev_err(dev, "failed to set cci pll rate: %d\n", ret);
+		clk_set_parent(drv->cci_clk, cci_pll);
+		goto out_restore_voltage;
+	}
+
+	/* switch the cci clock back to the original clock source. */
+	ret = clk_set_parent(drv->cci_clk, cci_pll);
+	if (ret) {
+		dev_err(dev, "failed to re-parent cci clock\n");
+		mtk_ccifreq_set_voltage(drv, inter_voltage);
+		goto out_unlock;
+	}
+
+	/*
+	 * If the new voltage is lower than the intermediate voltage or the
+	 * original voltage, scale down to the new voltage.
+	 */
+	if (voltage < inter_voltage || voltage < pre_voltage) {
+		ret = mtk_ccifreq_set_voltage(drv, voltage);
+		if (ret) {
+			dev_err(dev, "failed to scale down voltage\n");
+			goto out_unlock;
+		}
+	}
+
+	drv->pre_freq = *freq;
+	mutex_unlock(&drv->reg_lock);
+
+	return 0;
+
+out_restore_voltage:
+	mtk_ccifreq_set_voltage(drv, pre_voltage);
+
+out_unlock:
+	mutex_unlock(&drv->reg_lock);
+	return ret;
+}
+
+static int mtk_ccifreq_opp_notifier(struct notifier_block *nb,
+				    unsigned long event, void *data)
+{
+	struct dev_pm_opp *opp = data;
+	struct mtk_ccifreq_drv *drv;
+	unsigned long freq, volt;
+
+	drv = container_of(nb, struct mtk_ccifreq_drv, opp_nb);
+
+	if (event == OPP_EVENT_ADJUST_VOLTAGE) {
+		freq = dev_pm_opp_get_freq(opp);
+
+		mutex_lock(&drv->reg_lock);
+		/* current opp item is changed */
+		if (freq == drv->pre_freq) {
+			volt = dev_pm_opp_get_voltage(opp);
+			mtk_ccifreq_set_voltage(drv, volt);
+		}
+		mutex_unlock(&drv->reg_lock);
+	}
+
+	return 0;
+}
+
+static struct devfreq_dev_profile mtk_ccifreq_profile = {
+	.target = mtk_ccifreq_target,
+};
+
+static int mtk_ccifreq_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct mtk_ccifreq_drv *drv;
+	struct devfreq_passive_data *passive_data;
+	struct dev_pm_opp *opp;
+	unsigned long rate, opp_volt;
+	int ret;
+
+	drv = devm_kzalloc(dev, sizeof(*drv), GFP_KERNEL);
+	if (!drv)
+		return -ENOMEM;
+
+	drv->dev = dev;
+	drv->soc_data = (const struct mtk_ccifreq_platform_data *)
+				of_device_get_match_data(&pdev->dev);
+	mutex_init(&drv->reg_lock);
+	platform_set_drvdata(pdev, drv);
+
+	drv->cci_clk = devm_clk_get(dev, "cci");
+	if (IS_ERR(drv->cci_clk)) {
+		ret = PTR_ERR(drv->cci_clk);
+		return dev_err_probe(dev, ret, "failed to get cci clk\n");
+	}
+
+	drv->inter_clk = devm_clk_get(dev, "intermediate");
+	if (IS_ERR(drv->inter_clk)) {
+		ret = PTR_ERR(drv->inter_clk);
+		return dev_err_probe(dev, ret,
+				     "failed to get intermediate clk\n");
+	}
+
+	drv->proc_reg = devm_regulator_get_optional(dev, "proc");
+	if (IS_ERR(drv->proc_reg)) {
+		ret = PTR_ERR(drv->proc_reg);
+		return dev_err_probe(dev, ret,
+				     "failed to get proc regulator\n");
+	}
+
+	ret = regulator_enable(drv->proc_reg);
+	if (ret) {
+		dev_err(dev, "failed to enable proc regulator\n");
+		return ret;
+	}
+
+	drv->sram_reg = devm_regulator_get_optional(dev, "sram");
+	if (IS_ERR(drv->sram_reg))
+		drv->sram_reg = NULL;
+	else {
+		ret = regulator_enable