Merging upstream version 6.10.3.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

9 files changed, 729 insertions, 82 deletions

diff --git a/drivers/power/supply/ab8500_charger.c b/drivers/power/supply/ab8500_charger.c
index 9b34d1a60f..4b0ad1b4b4 100644
--- a/drivers/power/supply/ab8500_charger.c
+++ b/drivers/power/supply/ab8500_charger.c
@@ -488,8 +488,10 @@ static int ab8500_charger_get_ac_voltage(struct ab8500_charger *di)
 	/* Only measure voltage if the charger is connected */
 	if (di->ac.charger_connected) {
 		ret = iio_read_channel_processed(di->adc_main_charger_v, &vch);
-		if (ret < 0)
+		if (ret < 0) {
 			dev_err(di->dev, "%s ADC conv failed,\n", __func__);
+			return ret;
+		}
 	} else {
 		vch = 0;
 	}
@@ -540,8 +542,10 @@ static int ab8500_charger_get_vbus_voltage(struct ab8500_charger *di)
 	/* Only measure voltage if the charger is connected */
 	if (di->usb.charger_connected) {
 		ret = iio_read_channel_processed(di->adc_vbus_v, &vch);
-		if (ret < 0)
+		if (ret < 0) {
 			dev_err(di->dev, "%s ADC conv failed,\n", __func__);
+			return ret;
+		}
 	} else {
 		vch = 0;
 	}
@@ -563,8 +567,10 @@ static int ab8500_charger_get_usb_current(struct ab8500_charger *di)
 	/* Only measure current if the charger is online */
 	if (di->usb.charger_online) {
 		ret = iio_read_channel_processed(di->adc_usb_charger_c, &ich);
-		if (ret < 0)
+		if (ret < 0) {
 			dev_err(di->dev, "%s ADC conv failed,\n", __func__);
+			return ret;
+		}
 	} else {
 		ich = 0;
 	}
@@ -586,8 +592,10 @@ static int ab8500_charger_get_ac_current(struct ab8500_charger *di)
 	/* Only measure current if the charger is online */
 	if (di->ac.charger_online) {
 		ret = iio_read_channel_processed(di->adc_main_charger_c, &ich);
-		if (ret < 0)
+		if (ret < 0) {
 			dev_err(di->dev, "%s ADC conv failed,\n", __func__);
+			return ret;
+		}
 	} else {
 		ich = 0;
 	}
diff --git a/drivers/power/supply/bq27xxx_battery.c b/drivers/power/supply/bq27xxx_battery.c
index abca568344..750fda5433 100644
--- a/drivers/power/supply/bq27xxx_battery.c
+++ b/drivers/power/supply/bq27xxx_battery.c
@@ -1545,7 +1545,8 @@ static int bq27xxx_battery_read_soc(struct bq27xxx_device_info *di)
  * Return a battery charge value in µAh
  * Or < 0 if something fails.
  */
-static int bq27xxx_battery_read_charge(struct bq27xxx_device_info *di, u8 reg)
+static int bq27xxx_battery_read_charge(struct bq27xxx_device_info *di, u8 reg,
+				       union power_supply_propval *val)
 {
 	int charge;
 
@@ -1561,34 +1562,39 @@ static int bq27xxx_battery_read_charge(struct bq27xxx_device_info *di, u8 reg)
 	else
 		charge *= 1000;
 
-	return charge;
+	val->intval = charge;
+
+	return 0;
 }
 
 /*
  * Return the battery Nominal available capacity in µAh
  * Or < 0 if something fails.
  */
-static inline int bq27xxx_battery_read_nac(struct bq27xxx_device_info *di)
+static inline int bq27xxx_battery_read_nac(struct bq27xxx_device_info *di,
+					   union power_supply_propval *val)
 {
-	return bq27xxx_battery_read_charge(di, BQ27XXX_REG_NAC);
+	return bq27xxx_battery_read_charge(di, BQ27XXX_REG_NAC, val);
 }
 
 /*
  * Return the battery Remaining Capacity in µAh
  * Or < 0 if something fails.
  */
-static inline int bq27xxx_battery_read_rc(struct bq27xxx_device_info *di)
+static inline int bq27xxx_battery_read_rc(struct bq27xxx_device_info *di,
+					  union power_supply_propval *val)
 {
-	return bq27xxx_battery_read_charge(di, BQ27XXX_REG_RC);
+	return bq27xxx_battery_read_charge(di, BQ27XXX_REG_RC, val);
 }
 
 /*
  * Return the battery Full Charge Capacity in µAh
  * Or < 0 if something fails.
  */
-static inline int bq27xxx_battery_read_fcc(struct bq27xxx_device_info *di)
+static inline int bq27xxx_battery_read_fcc(struct bq27xxx_device_info *di,
+					   union power_supply_propval *val)
 {
-	return bq27xxx_battery_read_charge(di, BQ27XXX_REG_FCC);
+	return bq27xxx_battery_read_charge(di, BQ27XXX_REG_FCC, val);
 }
 
 /*
@@ -1633,7 +1639,8 @@ static int bq27xxx_battery_read_dcap(struct bq27xxx_device_info *di,
  * Return the battery Available energy in µWh
  * Or < 0 if something fails.
  */
-static int bq27xxx_battery_read_energy(struct bq27xxx_device_info *di)
+static int bq27xxx_battery_read_energy(struct bq27xxx_device_info *di,
+				       union power_supply_propval *val)
 {
 	int ae;
 
@@ -1648,14 +1655,17 @@ static int bq27xxx_battery_read_energy(struct bq27xxx_device_info *di)
 	else
 		ae *= 1000;
 
-	return ae;
+	val->intval = ae;
+
+	return 0;
 }
 
 /*
- * Return the battery temperature in tenths of degree Kelvin
+ * Return the battery temperature in tenths of degree Celsius
  * Or < 0 if something fails.
  */
-static int bq27xxx_battery_read_temperature(struct bq27xxx_device_info *di)
+static int bq27xxx_battery_read_temperature(struct bq27xxx_device_info *di,
+					    union power_supply_propval *val)
 {
 	int temp;
 
@@ -1668,14 +1678,20 @@ static int bq27xxx_battery_read_temperature(struct bq27xxx_device_info *di)
 	if (di->opts & BQ27XXX_O_ZERO)
 		temp = 5 * temp / 2;
 
-	return temp;
+	/* Convert decidegree Kelvin to Celsius */
+	temp -= 2731;
+
+	val->intval = temp;
+
+	return 0;
 }
 
 /*
  * Return the battery Cycle count total
  * Or < 0 if something fails.
  */
-static int bq27xxx_battery_read_cyct(struct bq27xxx_device_info *di)
+static int bq27xxx_battery_read_cyct(struct bq27xxx_device_info *di,
+				     union power_supply_propval *val)
 {
 	int cyct;
 
@@ -1683,14 +1699,17 @@ static int bq27xxx_battery_read_cyct(struct bq27xxx_device_info *di)
 	if (cyct < 0)
 		dev_err(di->dev, "error reading cycle count total\n");
 
-	return cyct;
+	val->intval = cyct;
+
+	return 0;
 }
 
 /*
  * Read a time register.
  * Return < 0 if something fails.
  */
-static int bq27xxx_battery_read_time(struct bq27xxx_device_info *di, u8 reg)
+static int bq27xxx_battery_read_time(struct bq27xxx_device_info *di, u8 reg,
+				     union power_supply_propval *val)
 {
 	int tval;
 
@@ -1704,7 +1723,9 @@ static int bq27xxx_battery_read_time(struct bq27xxx_device_info *di, u8 reg)
 	if (tval == 65535)
 		return -ENODATA;
 
-	return tval * 60;
+	val->intval = tval * 60;
+
+	return 0;
 }
 
 /*
@@ -1756,19 +1777,26 @@ static bool bq27xxx_battery_capacity_inaccurate(struct bq27xxx_device_info *di,
 		return false;
 }
 
-static int bq27xxx_battery_read_health(struct bq27xxx_device_info *di)
+static int bq27xxx_battery_read_health(struct bq27xxx_device_info *di,
+				       union power_supply_propval *val)
 {
+	int health;
+
 	/* Unlikely but important to return first */
 	if (unlikely(bq27xxx_battery_overtemp(di, di->cache.flags)))
-		return POWER_SUPPLY_HEALTH_OVERHEAT;
-	if (unlikely(bq27xxx_battery_undertemp(di, di->cache.flags)))
-		return POWER_SUPPLY_HEALTH_COLD;
-	if (unlikely(bq27xxx_battery_dead(di, di->cache.flags)))
-		return POWER_SUPPLY_HEALTH_DEAD;
-	if (unlikely(bq27xxx_battery_capacity_inaccurate(di, di->cache.flags)))
-		return POWER_SUPPLY_HEALTH_CALIBRATION_REQUIRED;
-
-	return POWER_SUPPLY_HEALTH_GOOD;
+		health = POWER_SUPPLY_HEALTH_OVERHEAT;
+	else if (unlikely(bq27xxx_battery_undertemp(di, di->cache.flags)))
+		health = POWER_SUPPLY_HEALTH_COLD;
+	else if (unlikely(bq27xxx_battery_dead(di, di->cache.flags)))
+		health = POWER_SUPPLY_HEALTH_DEAD;
+	else if (unlikely(bq27xxx_battery_capacity_inaccurate(di, di->cache.flags)))
+		health = POWER_SUPPLY_HEALTH_CALIBRATION_REQUIRED;
+	else
+		health = POWER_SUPPLY_HEALTH_GOOD;
+
+	val->intval = health;
+
+	return 0;
 }
 
 static bool bq27xxx_battery_is_full(struct bq27xxx_device_info *di, int flags)
@@ -1851,22 +1879,8 @@ static void bq27xxx_battery_update_unlocked(struct bq27xxx_device_info *di)
 	if ((cache.flags & 0xff) == 0xff)
 		cache.flags = -1; /* read error */
 	if (cache.flags >= 0) {
-		cache.temperature = bq27xxx_battery_read_temperature(di);
-		if (di->regs[BQ27XXX_REG_TTE] != INVALID_REG_ADDR)
-			cache.time_to_empty = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTE);
-		if (di->regs[BQ27XXX_REG_TTECP] != INVALID_REG_ADDR)
-			cache.time_to_empty_avg = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTECP);
-		if (di->regs[BQ27XXX_REG_TTF] != INVALID_REG_ADDR)
-			cache.time_to_full = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTF);
-
-		cache.charge_full = bq27xxx_battery_read_fcc(di);
 		cache.capacity = bq27xxx_battery_read_soc(di);
-		if (di->regs[BQ27XXX_REG_AE] != INVALID_REG_ADDR)
-			cache.energy = bq27xxx_battery_read_energy(di);
 		di->cache.flags = cache.flags;
-		cache.health = bq27xxx_battery_read_health(di);
-		if (di->regs[BQ27XXX_REG_CYCT] != INVALID_REG_ADDR)
-			cache.cycle_count = bq27xxx_battery_read_cyct(di);
 
 		/*
 		 * On gauges with signed current reporting the current must be
@@ -2038,18 +2052,16 @@ static int bq27xxx_battery_get_property(struct power_supply *psy,
 		ret = bq27xxx_battery_capacity_level(di, val);
 		break;
 	case POWER_SUPPLY_PROP_TEMP:
-		ret = bq27xxx_simple_value(di->cache.temperature, val);
-		if (ret == 0)
-			val->intval -= 2731; /* convert decidegree k to c */
+		ret = bq27xxx_battery_read_temperature(di, val);
 		break;
 	case POWER_SUPPLY_PROP_TIME_TO_EMPTY_NOW:
-		ret = bq27xxx_simple_value(di->cache.time_to_empty, val);
+		ret = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTE, val);
 		break;
 	case POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG:
-		ret = bq27xxx_simple_value(di->cache.time_to_empty_avg, val);
+		ret = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTECP, val);
 		break;
 	case POWER_SUPPLY_PROP_TIME_TO_FULL_NOW:
-		ret = bq27xxx_simple_value(di->cache.time_to_full, val);
+		ret = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTF, val);
 		break;
 	case POWER_SUPPLY_PROP_TECHNOLOGY:
 		if (di->opts & BQ27XXX_O_MUL_CHEM)
@@ -2059,12 +2071,12 @@ static int bq27xxx_battery_get_property(struct power_supply *psy,
 		break;
 	case POWER_SUPPLY_PROP_CHARGE_NOW:
 		if (di->regs[BQ27XXX_REG_NAC] != INVALID_REG_ADDR)
-			ret = bq27xxx_simple_value(bq27xxx_battery_read_nac(di), val);
+			ret = bq27xxx_battery_read_nac(di, val);
 		else
-			ret = bq27xxx_simple_value(bq27xxx_battery_read_rc(di), val);
+			ret = bq27xxx_battery_read_rc(di, val);
 		break;
 	case POWER_SUPPLY_PROP_CHARGE_FULL:
-		ret = bq27xxx_simple_value(di->cache.charge_full, val);
+		ret = bq27xxx_battery_read_fcc(di, val);
 		break;
 	case POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN:
 		ret = bq27xxx_battery_read_dcap(di, val);
@@ -2077,16 +2089,16 @@ static int bq27xxx_battery_get_property(struct power_supply *psy,
 	case POWER_SUPPLY_PROP_VOLTAGE_MIN_DESIGN:
 		return -EINVAL;
 	case POWER_SUPPLY_PROP_CYCLE_COUNT:
-		ret = bq27xxx_simple_value(di->cache.cycle_count, val);
+		ret = bq27xxx_battery_read_cyct(di, val);
 		break;
 	case POWER_SUPPLY_PROP_ENERGY_NOW:
-		ret = bq27xxx_simple_value(di->cache.energy, val);
+		ret = bq27xxx_battery_read_energy(di, val);
 		break;
 	case POWER_SUPPLY_PROP_POWER_AVG:
 		ret = bq27xxx_battery_pwr_avg(di, val);
 		break;
 	case POWER_SUPPLY_PROP_HEALTH:
-		ret = bq27xxx_simple_value(di->cache.health, val);
+		ret = bq27xxx_battery_read_health(di, val);
 		break;
 	case POWER_SUPPLY_PROP_MANUFACTURER:
 		val->strval = BQ27XXX_MANUFACTURER;
diff --git a/drivers/power/supply/cros_peripheral_charger.c b/drivers/power/supply/cros_peripheral_charger.c
index a204f2355b..d406f2a784 100644
--- a/drivers/power/supply/cros_peripheral_charger.c
+++ b/drivers/power/supply/cros_peripheral_charger.c
@@ -5,6 +5,7 @@
  * Copyright 2020 Google LLC.
  */
 
+#include <linux/mod_devicetable.h>
 #include <linux/module.h>
 #include <linux/notifier.h>
 #include <linux/platform_data/cros_ec_commands.h>
@@ -367,16 +368,22 @@ static int __maybe_unused cros_pchg_resume(struct device *dev)
 
 static SIMPLE_DEV_PM_OPS(cros_pchg_pm_ops, NULL, cros_pchg_resume);
 
+static const struct platform_device_id cros_pchg_id[] = {
+	{ DRV_NAME, 0 },
+	{}
+};
+MODULE_DEVICE_TABLE(platform, cros_pchg_id);
+
 static struct platform_driver cros_pchg_driver = {
 	.driver = {
 		.name = DRV_NAME,
 		.pm = &cros_pchg_pm_ops,
 	},
-	.probe = cros_pchg_probe
+	.probe = cros_pchg_probe,
+	.id_table = cros_pchg_id,
 };
 
 module_platform_driver(cros_pchg_driver);
 
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("ChromeOS EC peripheral device charger");
-MODULE_ALIAS("platform:" DRV_NAME);
diff --git a/drivers/power/supply/ingenic-battery.c b/drivers/power/supply/ingenic-battery.c
index 2e7fdfde47..0a40f425c2 100644
--- a/drivers/power/supply/ingenic-battery.c
+++ b/drivers/power/supply/ingenic-battery.c
@@ -31,8 +31,9 @@ static int ingenic_battery_get_property(struct power_supply *psy,
 
 	switch (psp) {
 	case POWER_SUPPLY_PROP_HEALTH:
-		ret = iio_read_channel_processed(bat->channel, &val->intval);
-		val->intval *= 1000;
+		ret = iio_read_channel_processed_scale(bat->channel,
+						       &val->intval,
+						       1000);
 		if (val->intval < info->voltage_min_design_uv)
 			val->intval = POWER_SUPPLY_HEALTH_DEAD;
 		else if (val->intval > info->voltage_max_design_uv)
@@ -41,8 +42,9 @@ static int ingenic_battery_get_property(struct power_supply *psy,
 			val->intval = POWER_SUPPLY_HEALTH_GOOD;
 		return ret;
 	case POWER_SUPPLY_PROP_VOLTAGE_NOW:
-		ret = iio_read_channel_processed(bat->channel, &val->intval);
-		val->intval *= 1000;
+		ret = iio_read_channel_processed_scale(bat->channel,
+						       &val->intval,
+						       1000);
 		return ret;
 	case POWER_SUPPLY_PROP_VOLTAGE_MIN_DESIGN:
 		val->intval = info->voltage_min_design_uv;
diff --git a/drivers/power/supply/sbs-manager.c b/drivers/power/supply/sbs-manager.c
index 9e4141cffb..933b04806d 100644
--- a/drivers/power/supply/sbs-manager.c
+++ b/drivers/power/supply/sbs-manager.c
@@ -358,7 +358,7 @@ static int sbsm_probe(struct i2c_client *client)
 	/* register muxed i2c channels. One for each supported battery */
 	for (i = 0; i < SBSM_MAX_BATS; ++i) {
 		if (data->supported_bats & BIT(i)) {
-			ret = i2c_mux_add_adapter(data->muxc, 0, i + 1, 0);
+			ret = i2c_mux_add_adapter(data->muxc, 0, i + 1);
 			if (ret)
 				break;
 		}
diff --git a/drivers/power/supply/test_power.c b/drivers/power/supply/test_power.c
index 0d0a77584c..442ceb7795 100644
--- a/drivers/power/supply/test_power.c
+++ b/drivers/power/supply/test_power.c
@@ -35,6 +35,8 @@ static int battery_capacity		= 50;
 static int battery_voltage		= 3300;
 static int battery_charge_counter	= -1000;
 static int battery_current		= -1600;
+static enum power_supply_charge_behaviour battery_charge_behaviour =
+	POWER_SUPPLY_CHARGE_BEHAVIOUR_AUTO;
 
 static bool module_initialized;
 
@@ -123,6 +125,9 @@ static int test_power_get_battery_property(struct power_supply *psy,
 	case POWER_SUPPLY_PROP_CURRENT_NOW:
 		val->intval = battery_current;
 		break;
+	case POWER_SUPPLY_PROP_CHARGE_BEHAVIOUR:
+		val->intval = battery_charge_behaviour;
+		break;
 	default:
 		pr_info("%s: some properties deliberately report errors.\n",
 			__func__);
@@ -131,6 +136,31 @@ static int test_power_get_battery_property(struct power_supply *psy,
 	return 0;
 }
 
+static int test_power_battery_property_is_writeable(struct power_supply *psy,
+						    enum power_supply_property psp)
+{
+	return psp == POWER_SUPPLY_PROP_CHARGE_BEHAVIOUR;
+}
+
+static int test_power_set_battery_property(struct power_supply *psy,
+					   enum power_supply_property psp,
+					   const union power_supply_propval *val)
+{
+	switch (psp) {
+	case POWER_SUPPLY_PROP_CHARGE_BEHAVIOUR:
+		if (val->intval < 0 ||
+		    val->intval >= BITS_PER_TYPE(typeof(psy->desc->charge_behaviours)) ||
+		    !(BIT(val->intval) & psy->desc->charge_behaviours)) {
+			return -EINVAL;
+		}
+		battery_charge_behaviour = val->intval;
+		break;
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
 static enum power_supply_property test_power_ac_props[] = {
 	POWER_SUPPLY_PROP_ONLINE,
 };
@@ -156,6 +186,7 @@ static enum power_supply_property test_power_battery_props[] = {
 	POWER_SUPPLY_PROP_VOLTAGE_NOW,
 	POWER_SUPPLY_PROP_CURRENT_AVG,
 	POWER_SUPPLY_PROP_CURRENT_NOW,
+	POWER_SUPPLY_PROP_CHARGE_BEHAVIOUR,
 };
 
 static char *test_power_ac_supplied_to[] = {
@@ -178,6 +209,11 @@ static const struct power_supply_desc test_power_desc[] = {
 		.properties = test_power_battery_props,
 		.num_properties = ARRAY_SIZE(test_power_battery_props),
 		.get_property = test_power_get_battery_property,
+		.set_property = test_power_set_battery_property,
+		.property_is_writeable = test_power_battery_property_is_writeable,
+		.charge_behaviours = BIT(POWER_SUPPLY_CHARGE_BEHAVIOUR_AUTO)
+				   | BIT(POWER_SUPPLY_CHARGE_BEHAVIOUR_INHIBIT_CHARGE)
+				   | BIT(POWER_SUPPLY_CHARGE_BEHAVIOUR_FORCE_DISCHARGE),
 	},
 	[TEST_USB] = {
 		.name = "test_usb",
diff --git a/drivers/powercap/dtpm_cpu.c b/drivers/powercap/dtpm_cpu.c
index bc90126f1b..6b6f51b215 100644
--- a/drivers/powercap/dtpm_cpu.c
+++ b/drivers/powercap/dtpm_cpu.c
@@ -43,13 +43,11 @@ static u64 set_pd_power_limit(struct dtpm *dtpm, u64 power_limit)
 	struct dtpm_cpu *dtpm_cpu = to_dtpm_cpu(dtpm);
 	struct em_perf_domain *pd = em_cpu_get(dtpm_cpu->cpu);
 	struct em_perf_state *table;
-	struct cpumask cpus;
 	unsigned long freq;
 	u64 power;
 	int i, nr_cpus;
 
-	cpumask_and(&cpus, cpu_online_mask, to_cpumask(pd->cpus));
-	nr_cpus = cpumask_weight(&cpus);
+	nr_cpus = cpumask_weight_and(cpu_online_mask, to_cpumask(pd->cpus));
 
 	rcu_read_lock();
 	table = em_perf_state_from_pd(pd);
@@ -123,11 +121,9 @@ static int update_pd_power_uw(struct dtpm *dtpm)
 	struct dtpm_cpu *dtpm_cpu = to_dtpm_cpu(dtpm);
 	struct em_perf_domain *em = em_cpu_get(dtpm_cpu->cpu);
 	struct em_perf_state *table;
-	struct cpumask cpus;
 	int nr_cpus;
 
-	cpumask_and(&cpus, cpu_online_mask, to_cpumask(em->cpus));
-	nr_cpus = cpumask_weight(&cpus);
+	nr_cpus = cpumask_weight_and(cpu_online_mask, to_cpumask(em->cpus));
 
 	rcu_read_lock();
 	table = em_perf_state_from_pd(em);
diff --git a/drivers/powercap/intel_rapl_common.c b/drivers/powercap/intel_rapl_common.c
index a28d54fd52..aac0744011 100644
--- a/drivers/powercap/intel_rapl_common.c
+++ b/drivers/powercap/intel_rapl_common.c
@@ -5,27 +5,29 @@
  */
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include <linux/bitmap.h>
 #include <linux/cleanup.h>
+#include <linux/cpu.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/intel_rapl.h>
 #include <linux/kernel.h>
-#include <linux/module.h>
 #include <linux/list.h>
-#include <linux/types.h>
-#include <linux/device.h>
-#include <linux/slab.h>
 #include <linux/log2.h>
-#include <linux/bitmap.h>
-#include <linux/delay.h>
-#include <linux/sysfs.h>
-#include <linux/cpu.h>
+#include <linux/module.h>
+#include <linux/nospec.h>
+#include <linux/perf_event.h>
+#include <linux/platform_device.h>
 #include <linux/powercap.h>
-#include <linux/suspend.h>
-#include <linux/intel_rapl.h>
 #include <linux/processor.h>
-#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/suspend.h>
+#include <linux/sysfs.h>
+#include <linux/types.h>
 
-#include <asm/iosf_mbi.h>
 #include <asm/cpu_device_id.h>
 #include <asm/intel-family.h>
+#include <asm/iosf_mbi.h>
 
 /* bitmasks for RAPL MSRs, used by primitive access functions */
 #define ENERGY_STATUS_MASK      0xffffffff
@@ -1263,6 +1265,7 @@ static const struct x86_cpu_id rapl_ids[] __initconst = {
 	X86_MATCH_INTEL_FAM6_MODEL(SAPPHIRERAPIDS_X,	&rapl_defaults_spr_server),
 	X86_MATCH_INTEL_FAM6_MODEL(EMERALDRAPIDS_X,	&rapl_defaults_spr_server),
 	X86_MATCH_INTEL_FAM6_MODEL(LUNARLAKE_M,		&rapl_defaults_core),
+	X86_MATCH_INTEL_FAM6_MODEL(ARROWLAKE_H,		&rapl_defaults_core),
 	X86_MATCH_INTEL_FAM6_MODEL(ARROWLAKE,		&rapl_defaults_core),
 	X86_MATCH_INTEL_FAM6_MODEL(LAKEFIELD,		&rapl_defaults_core),
 
@@ -1506,6 +1509,586 @@ static int rapl_detect_domains(struct rapl_package *rp)
 	return 0;
 }
 
+#ifdef CONFIG_PERF_EVENTS
+
+/*
+ * Support for RAPL PMU
+ *
+ * Register a PMU if any of the registered RAPL Packages have the requirement
+ * of exposing its energy counters via Perf PMU.
+ *
+ * PMU Name:
+ *	power
+ *
+ * Events:
+ *	Name		Event id	RAPL Domain
+ *	energy_cores	0x01		RAPL_DOMAIN_PP0
+ *	energy_pkg	0x02		RAPL_DOMAIN_PACKAGE
+ *	energy_ram	0x03		RAPL_DOMAIN_DRAM
+ *	energy_gpu	0x04		RAPL_DOMAIN_PP1
+ *	energy_psys	0x05		RAPL_DOMAIN_PLATFORM
+ *
+ * Unit:
+ *	Joules
+ *
+ * Scale:
+ *	2.3283064365386962890625e-10
+ *	The same RAPL domain in different RAPL Packages may have different
+ *	energy units. Use 2.3283064365386962890625e-10 (2^-32) Joules as
+ *	the fixed unit for all energy counters, and covert each hardware
+ *	counter increase to N times of PMU event counter increases.
+ *
+ * This is fully compatible with the current MSR RAPL PMU. This means that
+ * userspace programs like turbostat can use the same code to handle RAPL Perf
+ * PMU, no matter what RAPL Interface driver (MSR/TPMI, etc) is running
+ * underlying on the platform.
+ *
+ * Note that RAPL Packages can be probed/removed dynamically, and the events
+ * supported by each TPMI RAPL device can be different. Thus the RAPL PMU
+ * support is done on demand, which means
+ * 1. PMU is registered only if it is needed by a RAPL Package. PMU events for
+ *    unsupported counters are not exposed.
+ * 2. PMU is unregistered and registered when a new RAPL Package is probed and
+ *    supports new counters that are not supported by current PMU.
+ * 3. PMU is unregistered when all registered RAPL Packages don't need PMU.
+ */
+
+struct rapl_pmu {
+	struct pmu pmu;			/* Perf PMU structure */
+	u64 timer_ms;			/* Maximum expiration time to avoid counter overflow */
+	unsigned long domain_map;	/* Events supported by current registered PMU */
+	bool registered;		/* Whether the PMU has been registered or not */
+};
+
+static struct rapl_pmu rapl_pmu;
+
+/* PMU helpers */
+
+static int get_pmu_cpu(struct rapl_package *rp)
+{
+	int cpu;
+
+	if (!rp->has_pmu)
+		return nr_cpu_ids;
+
+	/* Only TPMI RAPL is supported for now */
+	if (rp->priv->type != RAPL_IF_TPMI)
+		return nr_cpu_ids;
+
+	/* TPMI RAPL uses any CPU in the package for PMU */
+	for_each_online_cpu(cpu)
+		if (topology_physical_package_id(cpu) == rp->id)
+			return cpu;
+
+	return nr_cpu_ids;
+}
+
+static bool is_rp_pmu_cpu(struct rapl_package *rp, int cpu)
+{
+	if (!rp->has_pmu)
+		return false;
+
+	/* Only TPMI RAPL is supported for now */
+	if (rp->priv->type != RAPL_IF_TPMI)
+		return false;
+
+	/* TPMI RAPL uses any CPU in the package for PMU */
+	return topology_physical_package_id(cpu) == rp->id;
+}
+
+static struct rapl_package_pmu_data *event_to_pmu_data(struct perf_event *event)
+{
+	struct rapl_package *rp = event->pmu_private;
+
+	return &rp->pmu_data;
+}
+
+/* PMU event callbacks */
+
+static u64 event_read_counter(struct perf_event *event)
+{
+	struct rapl_package *rp = event->pmu_private;
+	u64 val;
+	int ret;
+
+	/* Return 0 for unsupported events */
+	if (event->hw.idx < 0)
+		return 0;
+
+	ret = rapl_read_data_raw(&rp->domains[event->hw.idx], ENERGY_COUNTER, false, &val);
+
+	/* Return 0 for failed read */
+	if (ret)
+		return 0;
+
+	return val;
+}
+
+static void __rapl_pmu_event_start(struct perf_event *event)
+{
+	struct rapl_package_pmu_data *data = event_to_pmu_data(event);
+
+	if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
+		return;
+
+	event->hw.state = 0;
+
+	list_add_tail(&event->active_entry, &data->active_list);
+
+	local64_set(&event->hw.prev_count, event_read_counter(event));
+	if (++data->n_active == 1)
+		hrtimer_start(&data->hrtimer, data->timer_interval,
+			      HRTIMER_MODE_REL_PINNED);
+}
+
+static void rapl_pmu_event_start(struct perf_event *event, int mode)
+{
+	struct rapl_package_pmu_data *data = event_to_pmu_data(event);
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&data->lock, flags);
+	__rapl_pmu_event_start(event);
+	raw_spin_unlock_irqrestore(&data->lock, flags);
+}
+
+static u64 rapl_event_update(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct rapl_package_pmu_data *data = event_to_pmu_data(event);
+	u64 prev_raw_count, new_raw_count;
+	s64 delta, sdelta;
+
+	/*
+	 * Follow the generic code to drain hwc->prev_count.
+	 * The loop is not expected to run for multiple times.
+	 */
+	prev_raw_count = local64_read(&hwc->prev_count);
+	do {
+		new_raw_count = event_read_counter(event);
+	} while (!local64_try_cmpxchg(&hwc->prev_count,
+		&prev_raw_count, new_raw_count));
+
+
+	/*
+	 * Now we have the new raw value and have updated the prev
+	 * timestamp already. We can now calculate the elapsed delta
+	 * (event-)time and add that to the generic event.
+	 */
+	delta = new_raw_count - prev_raw_count;
+
+	/*
+	 * Scale delta to smallest unit (2^-32)
+	 * users must then scale back: count * 1/(1e9*2^32) to get Joules
+	 * or use ldexp(count, -32).
+	 * Watts = Joules/Time delta
+	 */
+	sdelta = delta * data->scale[event->hw.flags];
+
+	local64_add(sdelta, &event->count);
+
+	return new_raw_count;
+}
+
+static void rapl_pmu_event_stop(struct perf_event *event, int mode)
+{
+	struct rapl_package_pmu_data *data = event_to_pmu_data(event);
+	struct hw_perf_event *hwc = &event->hw;
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&data->lock, flags);
+
+	/* Mark event as deactivated and stopped */
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		WARN_ON_ONCE(data->n_active <= 0);
+		if (--data->n_active == 0)
+			hrtimer_cancel(&data->hrtimer);
+
+		list_del(&event->active_entry);
+
+		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+		hwc->state |= PERF_HES_STOPPED;
+	}
+
+	/* Check if update of sw counter is necessary */
+	if ((mode & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		/*
+		 * Drain the remaining delta count out of a event
+		 * that we are disabling:
+		 */
+		rapl_event_update(event);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
+
+	raw_spin_unlock_irqrestore(&data->lock, flags);
+}
+
+static int rapl_pmu_event_add(struct perf_event *event, int mode)
+{
+	struct rapl_package_pmu_data *data = event_to_pmu_data(event);
+	struct hw_perf_event *hwc = &event->hw;
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&data->lock, flags);
+
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+
+	if (mode & PERF_EF_START)
+		__rapl_pmu_event_start(event);
+
+	raw_spin_unlock_irqrestore(&data->lock, flags);
+
+	return 0;
+}
+
+static void rapl_pmu_event_del(struct perf_event *event, int flags)
+{
+	rapl_pmu_event_stop(event, PERF_EF_UPDATE);
+}
+
+/* RAPL PMU event ids, same as shown in sysfs */
+enum perf_rapl_events {
+	PERF_RAPL_PP0 = 1,	/* all cores */
+	PERF_RAPL_PKG,		/* entire package */
+	PERF_RAPL_RAM,		/* DRAM */
+	PERF_RAPL_PP1,		/* gpu */
+	PERF_RAPL_PSYS,		/* psys */
+	PERF_RAPL_MAX
+};
+#define RAPL_EVENT_MASK GENMASK(7, 0)
+
+static const int event_to_domain[PERF_RAPL_MAX] = {
+	[PERF_RAPL_PP0]		= RAPL_DOMAIN_PP0,
+	[PERF_RAPL_PKG]		= RAPL_DOMAIN_PACKAGE,
+	[PERF_RAPL_RAM]		= RAPL_DOMAIN_DRAM,
+	[PERF_RAPL_PP1]		= RAPL_DOMAIN_PP1,
+	[PERF_RAPL_PSYS]	= RAPL_DOMAIN_PLATFORM,
+};
+
+static int rapl_pmu_event_init(struct perf_event *event)
+{
+	struct rapl_package *pos, *rp = NULL;
+	u64 cfg = event->attr.config & RAPL_EVENT_MASK;
+	int domain, idx;
+
+	/* Only look at RAPL events */
+	if (event->attr.type != event->pmu->type)
+		return -ENOENT;
+
+	/* Check for supported events only */
+	if (!cfg || cfg >= PERF_RAPL_MAX)
+		return -EINVAL;
+
+	if (event->cpu < 0)
+		return -EINVAL;
+
+	/* Find out which Package the event belongs to */
+	list_for_each_entry(pos, &rapl_packages, plist) {
+		if (is_rp_pmu_cpu(pos, event->cpu)) {
+			rp = pos;
+			break;
+		}
+	}
+	if (!rp)
+		return -ENODEV;
+
+	/* Find out which RAPL Domain the event belongs to */
+	domain = event_to_domain[cfg];
+
+	event->event_caps |= PERF_EV_CAP_READ_ACTIVE_PKG;
+	event->pmu_private = rp;	/* Which package */
+	event->hw.flags = domain;	/* Which domain */
+
+	event->hw.idx = -1;
+	/* Find out the index in rp->domains[] to get domain pointer */
+	for (idx = 0; idx < rp->nr_domains; idx++) {
+		if (rp->domains[idx].id == domain) {
+			event->hw.idx = idx;
+			break;
+		}
+	}
+
+	return 0;
+}
+
+static void rapl_pmu_event_read(struct perf_event *event)
+{
+	rapl_event_update(event);
+}
+
+static enum hrtimer_restart rapl_hrtimer_handle(struct hrtimer *hrtimer)
+{
+	struct rapl_package_pmu_data *data =
+		container_of(hrtimer, struct rapl_package_pmu_data, hrtimer);
+	struct perf_event *event;
+	unsigned long flags;
+
+	if (!data->n_active)
+		return HRTIMER_NORESTART;
+
+	raw_spin_lock_irqsave(&data->lock, flags);
+
+	list_for_each_entry(event, &data->active_list, active_entry)
+		rapl_event_update(event);
+
+	raw_spin_unlock_irqrestore(&data->lock, flags);
+
+	hrtimer_forward_now(hrtimer, data->timer_interval);
+
+	return HRTIMER_RESTART;
+}
+
+/* PMU sysfs attributes */
+
+/*
+ * There are no default events, but we need to create "events" group (with
+ * empty attrs) before updating it with detected events.
+ */
+static struct attribute *attrs_empty[] = {
+	NULL,
+};
+
+static struct attribute_group pmu_events_group = {
+	.name = "events",
+	.attrs = attrs_empty,
+};
+
+static ssize_t cpumask_show(struct device *dev,
+			    struct device_attribute *attr, char *buf)
+{
+	struct rapl_package *rp;
+	cpumask_var_t cpu_mask;
+	int cpu;
+	int ret;
+
+	if (!alloc_cpumask_var(&cpu_mask, GFP_KERNEL))
+		return -ENOMEM;
+
+	cpus_read_lock();
+
+	cpumask_clear(cpu_mask);
+
+	/* Choose a cpu for each RAPL Package */
+	list_for_each_entry(rp, &rapl_packages, plist) {
+		cpu = get_pmu_cpu(rp);
+		if (cpu < nr_cpu_ids)
+			cpumask_set_cpu(cpu, cpu_mask);
+	}
+	cpus_read_unlock();
+
+	ret = cpumap_print_to_pagebuf(true, buf, cpu_mask);
+
+	free_cpumask_var(cpu_mask);
+
+	return ret;
+}
+
+static DEVICE_ATTR_RO(cpumask);
+
+static struct attribute *pmu_cpumask_attrs[] = {
+	&dev_attr_cpumask.attr,
+	NULL
+};
+
+static struct attribute_group pmu_cpumask_group = {
+	.attrs = pmu_cpumask_attrs,
+};
+
+PMU_FORMAT_ATTR(event, "config:0-7");
+static struct attribute *pmu_format_attr[] = {
+	&format_attr_event.attr,
+	NULL
+};
+
+static struct attribute_group pmu_format_group = {
+	.name = "format",
+	.attrs = pmu_format_attr,
+};
+
+static const struct attribute_group *pmu_attr_groups[] = {
+	&pmu_events_group,
+	&pmu_cpumask_group,
+	&pmu_format_group,
+	NULL
+};
+
+#define RAPL_EVENT_ATTR_STR(_name, v, str)					\
+static struct perf_pmu_events_attr event_attr_##v = {				\
+	.attr		= __ATTR(_name, 0444, perf_event_sysfs_show, NULL),	\
+	.event_str	= str,							\
+}
+
+RAPL_EVENT_ATTR_STR(energy-cores,	rapl_cores,	"event=0x01");
+RAPL_EVENT_ATTR_STR(energy-pkg,		rapl_pkg,	"event=0x02");
+RAPL_EVENT_ATTR_STR(energy-ram,		rapl_ram,	"event=0x03");
+RAPL_EVENT_ATTR_STR(energy-gpu,		rapl_gpu,	"event=0x04");
+RAPL_EVENT_ATTR_STR(energy-psys,	rapl_psys,	"event=0x05");
+
+RAPL_EVENT_ATTR_STR(energy-cores.unit,	rapl_unit_cores,	"Joules");
+RAPL_EVENT_ATTR_STR(energy-pkg.unit,	rapl_unit_pkg,		"Joules");
+RAPL_EVENT_ATTR_STR(energy-ram.unit,	rapl_unit_ram,		"Joules");
+RAPL_EVENT_ATTR_STR(energy-gpu.unit,	rapl_unit_gpu,		"Joules");
+RAPL_EVENT_ATTR_STR(energy-psys.unit,	rapl_unit_psys,		"Joules");
+
+RAPL_EVENT_ATTR_STR(energy-cores.scale,	rapl_scale_cores,	"2.3283064365386962890625e-10");
+RAPL_EVENT_ATTR_STR(energy-pkg.scale,	rapl_scale_pkg,		"2.3283064365386962890625e-10");
+RAPL_EVENT_ATTR_STR(energy-ram.scale,	rapl_scale_ram,		"2.3283064365386962890625e-10");
+RAPL_EVENT_ATTR_STR(energy-gpu.scale,	rapl_scale_gpu,		"2.3283064365386962890625e-10");
+RAPL_EVENT_ATTR_STR(energy-psys.scale,	rapl_scale_psys,	"2.3283064365386962890625e-10");
+
+#define RAPL_EVENT_GROUP(_name, domain)			\
+static struct attribute *pmu_attr_##_name[] = {		\
+	&event_attr_rapl_##_name.attr.attr,		\
+	&event_attr_rapl_unit_##_name.attr.attr,	\
+	&event_attr_rapl_scale_##_name.attr.attr,	\
+	NULL						\
+};							\
+static umode_t is_visible_##_name(struct kobject *kobj, struct attribute *attr, int event)	\
+{											\
+	return rapl_pmu.domain_map & BIT(domain) ? attr->mode : 0;	\
+}							\
+static struct attribute_group pmu_group_##_name = {	\
+	.name  = "events",				\
+	.attrs = pmu_attr_##_name,			\
+	.is_visible = is_visible_##_name,		\
+}
+
+RAPL_EVENT_GROUP(cores,	RAPL_DOMAIN_PP0);
+RAPL_EVENT_GROUP(pkg,	RAPL_DOMAIN_PACKAGE);
+RAPL_EVENT_GROUP(ram,	RAPL_DOMAIN_DRAM);
+RAPL_EVENT_GROUP(gpu,	RAPL_DOMAIN_PP1);
+RAPL_EVENT_GROUP(psys,	RAPL_DOMAIN_PLATFORM);
+
+static const struct attribute_group *pmu_attr_update[] = {
+	&pmu_group_cores,
+	&pmu_group_pkg,
+	&pmu_group_ram,
+	&pmu_group_gpu,
+	&pmu_group_psys,
+	NULL
+};
+
+static int rapl_pmu_update(struct rapl_package *rp)
+{
+	int ret = 0;
+
+	/* Return if PMU already covers all events supported by current RAPL Package */
+	if (rapl_pmu.registered && !(rp->domain_map & (~rapl_pmu.domain_map)))
+		goto end;
+
+	/* Unregister previous registered PMU */
+	if (rapl_pmu.registered)
+		perf_pmu_unregister(&rapl_pmu.pmu);
+
+	rapl_pmu.registered = false;
+	rapl_pmu.domain_map |= rp->domain_map;
+
+	memset(&rapl_pmu.pmu, 0, sizeof(struct pmu));
+	rapl_pmu.pmu.attr_groups = pmu_attr_groups;
+	rapl_pmu.pmu.attr_update = pmu_attr_update;
+	rapl_pmu.pmu.task_ctx_nr = perf_invalid_context;
+	rapl_pmu.pmu.event_init = rapl_pmu_event_init;
+	rapl_pmu.pmu.add = rapl_pmu_event_add;
+	rapl_pmu.pmu.del = rapl_pmu_event_del;
+	rapl_pmu.pmu.start = rapl_pmu_event_start;
+	rapl_pmu.pmu.stop = rapl_pmu_event_stop;
+	rapl_pmu.pmu.read = rapl_pmu_event_read;
+	rapl_pmu.pmu.module = THIS_MODULE;
+	rapl_pmu.pmu.capabilities = PERF_PMU_CAP_NO_EXCLUDE | PERF_PMU_CAP_NO_INTERRUPT;
+	ret = perf_pmu_register(&rapl_pmu.pmu, "power", -1);
+	if (ret) {
+		pr_info("Failed to register PMU\n");
+		return ret;
+	}
+
+	rapl_pmu.registered = true;
+end:
+	rp->has_pmu = true;
+	return ret;
+}
+
+int rapl_package_add_pmu(struct rapl_package *rp)
+{
+	struct rapl_package_pmu_data *data = &rp->pmu_data;
+	int idx;
+
+	if (rp->has_pmu)
+		return -EEXIST;
+
+	guard(cpus_read_lock)();
+
+	for (idx = 0; idx < rp->nr_domains; idx++) {
+		struct rapl_domain *rd = &rp->domains[idx];
+		int domain = rd->id;
+		u64 val;
+
+		if (!test_bit(domain, &rp->domain_map))
+			continue;
+
+		/*
+		 * The RAPL PMU granularity is 2^-32 Joules
+		 * data->scale[]: times of 2^-32 Joules for each ENERGY COUNTER increase
+		 */
+		val = rd->energy_unit * (1ULL << 32);
+		do_div(val, ENERGY_UNIT_SCALE * 1000000);
+		data->scale[domain] = val;
+
+		if (!rapl_pmu.timer_ms) {
+			struct rapl_primitive_info *rpi = get_rpi(rp, ENERGY_COUNTER);
+
+			/*
+			 * Calculate the timer rate:
+			 * Use reference of 200W for scaling the timeout to avoid counter
+			 * overflows.
+			 *
+			 * max_count = rpi->mask >> rpi->shift + 1
+			 * max_energy_pj = max_count * rd->energy_unit
+			 * max_time_sec = (max_energy_pj / 1000000000) / 200w
+			 *
+			 * rapl_pmu.timer_ms = max_time_sec * 1000 / 2
+			 */
+			val = (rpi->mask >> rpi->shift) + 1;
+			val *= rd->energy_unit;
+			do_div(val, 1000000 * 200 * 2);
+			rapl_pmu.timer_ms = val;
+
+			pr_debug("%llu ms overflow timer\n", rapl_pmu.timer_ms);
+		}
+
+		pr_debug("Domain %s: hw unit %lld * 2^-32 Joules\n", rd->name, data->scale[domain]);
+	}
+
+	/* Initialize per package PMU data */
+	raw_spin_lock_init(&data->lock);
+	INIT_LIST_HEAD(&data->active_list);
+	data->timer_interval = ms_to_ktime(rapl_pmu.timer_ms);
+	hrtimer_init(&data->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	data->hrtimer.function = rapl_hrtimer_handle;
+
+	return rapl_pmu_update(rp);
+}
+EXPORT_SYMBOL_GPL(rapl_package_add_pmu);
+
+void rapl_package_remove_pmu(struct rapl_package *rp)
+{
+	struct rapl_package *pos;
+
+	if (!rp->has_pmu)
+		return;
+
+	guard(cpus_read_lock)();
+
+	list_for_each_entry(pos, &rapl_packages, plist) {
+		/* PMU is still needed */
+		if (pos->has_pmu && pos != rp)
+			return;
+	}
+
+	perf_pmu_unregister(&rapl_pmu.pmu);
+	memset(&rapl_pmu, 0, sizeof(struct rapl_pmu));
+}
+EXPORT_SYMBOL_GPL(rapl_package_remove_pmu);
+#endif
+
 /* called from CPU hotplug notifier, hotplug lock held */
 void rapl_remove_package_cpuslocked(struct rapl_package *rp)
 {
diff --git a/drivers/powercap/intel_rapl_tpmi.c b/drivers/powercap/intel_rapl_tpmi.c
index f6b7f08597..947544e4d2 100644
--- a/drivers/powercap/intel_rapl_tpmi.c
+++ b/drivers/powercap/intel_rapl_tpmi.c
@@ -302,6 +302,8 @@ static int intel_rapl_tpmi_probe(struct auxiliary_device *auxdev,
 		goto err;
 	}
 
+	rapl_package_add_pmu(trp->rp);
+
 	auxiliary_set_drvdata(auxdev, trp);
 
 	return 0;
@@ -314,6 +316,7 @@ static void intel_rapl_tpmi_remove(struct auxiliary_device *auxdev)
 {
 	struct tpmi_rapl_package *trp = auxiliary_get_drvdata(auxdev);
 
+	rapl_package_remove_pmu(trp->rp);
 	rapl_remove_package(trp->rp);
 	trp_release(trp);
 }