Merge tag 'thermal-6.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

Pull thermal control updates from Rafael Wysocki:
 "These include thermal core fixes to protect thermal device operations
  against thermal device removal, other thermal core fixes and updates
  of Intel thermal control drivers.

  Specifics:

   - Fix race conditions related to thermal device operations that are
     not protected against thermal device removal (Guenter Roeck)

   - Fix error code in __thermal_cooling_device_register() (Dan
     Carpenter)

   - Validate new cooling device state (coming from user space) in
     cur_state_store() and reuse the max_state value from cooling device
     structure in the sysfs interface (Viresh Kumar)

   - Fix some possible name leaks in error paths in the thermal control
     core code (Yang Yingliang)

   - Detect TCC lock bit set in the intel_tcc_cooling driver and make it
     refuse to update the TCC offset in that case (Zhang Rui)

   - Add TCC cooling support for RaptorLake-S (Zhang Rui)

   - Prevent accidental clearing of HFI status by one of the other
     drivers using the same status register (Srinivas Pandruvada)

   - Protect clearing of thermal status bits in Intel thermal control
     drivers (Srinivas Pandruvada)

   - Allow the HFI thermal control driver to ACK an HFI event for the
     previously observed timestamp (Srinivas Pandruvada)

   - Remove a pointless die_id check from the HFI thermal driver and
     adjust the definition a data structure used by it (Ricardo Neri)"

* tag 'thermal-6.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  thermal: intel: hfi: Remove a pointless die_id check
  thermal: core: fix some possible name leaks in error paths
  thermal: intel: hfi: ACK HFI for the same timestamp
  thermal: intel: Protect clearing of thermal status bits
  thermal: intel: Prevent accidental clearing of HFI status
  thermal/core: Protect thermal device operations against thermal device removal
  thermal/core: Remove thermal_zone_set_trips()
  thermal/core: Protect sysfs accesses to thermal operations with thermal zone mutex
  thermal/core: Protect hwmon accesses to thermal operations with thermal zone mutex
  thermal/core: Introduce locked version of thermal_zone_device_update
  thermal/core: Move parameter validation from __thermal_zone_get_temp to thermal_zone_get_temp
  thermal/core: Ensure that thermal device is registered in thermal_zone_get_temp
  thermal/core: Delete device under thermal device zone lock
  thermal/core: Destroy thermal zone device mutex in release function
  thermal: intel: intel_tcc_cooling: Add TCC cooling support for RaptorLake-S
  thermal: intel: intel_tcc_cooling: Detect TCC lock bit
  thermal: intel: hfi: Improve the type of hfi_features::nr_table_pages
  thermal/core: fix error code in __thermal_cooling_device_register()
  thermal: sysfs: Reuse cdev->max_state
  thermal: Validate new state in cur_state_store()

11 files changed, 255 insertions, 142 deletions

diff --git a/drivers/thermal/gov_fair_share.c b/drivers/thermal/gov_fair_share.c
index a4ee4661e9cc..1cfeac16e7ac 100644
--- a/drivers/thermal/gov_fair_share.c
+++ b/drivers/thermal/gov_fair_share.c
@@ -49,11 +49,7 @@ static int get_trip_level(struct thermal_zone_device *tz)
 static long get_target_state(struct thermal_zone_device *tz,
 		struct thermal_cooling_device *cdev, int percentage, int level)
 {
-	unsigned long max_state;
-
-	cdev->ops->get_max_state(cdev, &max_state);
-
-	return (long)(percentage * level * max_state) / (100 * tz->num_trips);
+	return (long)(percentage * level * cdev->max_state) / (100 * tz->num_trips);
 }
 
 /**
diff --git a/drivers/thermal/intel/intel_hfi.c b/drivers/thermal/intel/intel_hfi.c
index a0640f762dc5..6e604bda2b93 100644
--- a/drivers/thermal/intel/intel_hfi.c
+++ b/drivers/thermal/intel/intel_hfi.c
@@ -42,9 +42,7 @@
 
 #include "../thermal_core.h"
 #include "intel_hfi.h"
-
-#define THERM_STATUS_CLEAR_PKG_MASK (BIT(1) | BIT(3) | BIT(5) | BIT(7) | \
-				     BIT(9) | BIT(11) | BIT(26))
+#include "thermal_interrupt.h"
 
 /* Hardware Feedback Interface MSR configuration bits */
 #define HW_FEEDBACK_PTR_VALID_BIT		BIT(0)
@@ -137,7 +135,7 @@ struct hfi_instance {
  * Parameters and supported features that are common to all HFI instances
  */
 struct hfi_features {
-	unsigned int	nr_table_pages;
+	size_t		nr_table_pages;
 	unsigned int	cpu_stride;
 	unsigned int	hdr_size;
 };
@@ -252,7 +250,7 @@ void intel_hfi_process_event(__u64 pkg_therm_status_msr_val)
 	struct hfi_instance *hfi_instance;
 	int cpu = smp_processor_id();
 	struct hfi_cpu_info *info;
-	u64 new_timestamp;
+	u64 new_timestamp, msr, hfi;
 
 	if (!pkg_therm_status_msr_val)
 		return;
@@ -281,9 +279,21 @@ void intel_hfi_process_event(__u64 pkg_therm_status_msr_val)
 	if (!raw_spin_trylock(&hfi_instance->event_lock))
 		return;
 
-	/* Skip duplicated updates. */
+	rdmsrl(MSR_IA32_PACKAGE_THERM_STATUS, msr);
+	hfi = msr & PACKAGE_THERM_STATUS_HFI_UPDATED;
+	if (!hfi) {
+		raw_spin_unlock(&hfi_instance->event_lock);
+		return;
+	}
+
+	/*
+	 * Ack duplicate update. Since there is an active HFI
+	 * status from HW, it must be a new event, not a case
+	 * where a lagging CPU entered the locked region.
+	 */
 	new_timestamp = *(u64 *)hfi_instance->hw_table;
 	if (*hfi_instance->timestamp == new_timestamp) {
+		thermal_clear_package_intr_status(PACKAGE_LEVEL, PACKAGE_THERM_STATUS_HFI_UPDATED);
 		raw_spin_unlock(&hfi_instance->event_lock);
 		return;
 	}
@@ -297,16 +307,14 @@ void intel_hfi_process_event(__u64 pkg_therm_status_msr_val)
 	memcpy(hfi_instance->local_table, hfi_instance->hw_table,
 	       hfi_features.nr_table_pages << PAGE_SHIFT);
 
-	raw_spin_unlock(&hfi_instance->table_lock);
-	raw_spin_unlock(&hfi_instance->event_lock);
-
 	/*
 	 * Let hardware know that we are done reading the HFI table and it is
 	 * free to update it again.
 	 */
-	pkg_therm_status_msr_val &= THERM_STATUS_CLEAR_PKG_MASK &
-				    ~PACKAGE_THERM_STATUS_HFI_UPDATED;
-	wrmsrl(MSR_IA32_PACKAGE_THERM_STATUS, pkg_therm_status_msr_val);
+	thermal_clear_package_intr_status(PACKAGE_LEVEL, PACKAGE_THERM_STATUS_HFI_UPDATED);
+
+	raw_spin_unlock(&hfi_instance->table_lock);
+	raw_spin_unlock(&hfi_instance->event_lock);
 
 	queue_delayed_work(hfi_updates_wq, &hfi_instance->update_work,
 			   HFI_UPDATE_INTERVAL);
@@ -371,7 +379,7 @@ void intel_hfi_online(unsigned int cpu)
 	die_id = topology_logical_die_id(cpu);
 	hfi_instance = info->hfi_instance;
 	if (!hfi_instance) {
-		if (die_id < 0 || die_id >= max_hfi_instances)
+		if (die_id >= max_hfi_instances)
 			return;
 
 		hfi_instance = &hfi_instances[die_id];
diff --git a/drivers/thermal/intel/intel_tcc_cooling.c b/drivers/thermal/intel/intel_tcc_cooling.c
index 95adac427b6f..a89e7e1890e4 100644
--- a/drivers/thermal/intel/intel_tcc_cooling.c
+++ b/drivers/thermal/intel/intel_tcc_cooling.c
@@ -14,6 +14,7 @@
 #define TCC_SHIFT 24
 #define TCC_MASK	(0x3fULL<<24)
 #define TCC_PROGRAMMABLE	BIT(30)
+#define TCC_LOCKED		BIT(31)
 
 static struct thermal_cooling_device *tcc_cdev;
 
@@ -84,6 +85,7 @@ static const struct x86_cpu_id tcc_ids[] __initconst = {
 	X86_MATCH_INTEL_FAM6_MODEL(ALDERLAKE_N, NULL),
 	X86_MATCH_INTEL_FAM6_MODEL(RAPTORLAKE, NULL),
 	X86_MATCH_INTEL_FAM6_MODEL(RAPTORLAKE_P, NULL),
+	X86_MATCH_INTEL_FAM6_MODEL(RAPTORLAKE_S, NULL),
 	{}
 };
 
@@ -108,6 +110,15 @@ static int __init tcc_cooling_init(void)
 	if (!(val & TCC_PROGRAMMABLE))
 		return -ENODEV;
 
+	err = rdmsrl_safe(MSR_IA32_TEMPERATURE_TARGET, &val);
+	if (err)
+		return err;
+
+	if (val & TCC_LOCKED) {
+		pr_info("TCC Offset locked\n");
+		return -ENODEV;
+	}
+
 	pr_info("Programmable TCC Offset detected\n");
 
 	tcc_cdev =
diff --git a/drivers/thermal/intel/therm_throt.c b/drivers/thermal/intel/therm_throt.c
index 8352083b87c7..4bb7fddaa143 100644
--- a/drivers/thermal/intel/therm_throt.c
+++ b/drivers/thermal/intel/therm_throt.c
@@ -190,32 +190,33 @@ static const struct attribute_group thermal_attr_group = {
 };
 #endif /* CONFIG_SYSFS */
 
-#define CORE_LEVEL	0
-#define PACKAGE_LEVEL	1
-
 #define THERM_THROT_POLL_INTERVAL	HZ
 #define THERM_STATUS_PROCHOT_LOG	BIT(1)
 
 #define THERM_STATUS_CLEAR_CORE_MASK (BIT(1) | BIT(3) | BIT(5) | BIT(7) | BIT(9) | BIT(11) | BIT(13) | BIT(15))
-#define THERM_STATUS_CLEAR_PKG_MASK  (BIT(1) | BIT(3) | BIT(5) | BIT(7) | BIT(9) | BIT(11))
+#define THERM_STATUS_CLEAR_PKG_MASK  (BIT(1) | BIT(3) | BIT(5) | BIT(7) | BIT(9) | BIT(11) | BIT(26))
 
-static void clear_therm_status_log(int level)
+/*
+ * Clear the bits in package thermal status register for bit = 1
+ * in bitmask
+ */
+void thermal_clear_package_intr_status(int level, u64 bit_mask)
 {
+	u64 msr_val;
 	int msr;
-	u64 mask, msr_val;
 
 	if (level == CORE_LEVEL) {
 		msr  = MSR_IA32_THERM_STATUS;
-		mask = THERM_STATUS_CLEAR_CORE_MASK;
+		msr_val = THERM_STATUS_CLEAR_CORE_MASK;
 	} else {
 		msr  = MSR_IA32_PACKAGE_THERM_STATUS;
-		mask = THERM_STATUS_CLEAR_PKG_MASK;
+		msr_val = THERM_STATUS_CLEAR_PKG_MASK;
 	}
 
-	rdmsrl(msr, msr_val);
-	msr_val &= mask;
-	wrmsrl(msr, msr_val & ~THERM_STATUS_PROCHOT_LOG);
+	msr_val &= ~bit_mask;
+	wrmsrl(msr, msr_val);
 }
+EXPORT_SYMBOL_GPL(thermal_clear_package_intr_status);
 
 static void get_therm_status(int level, bool *proc_hot, u8 *temp)
 {
@@ -295,7 +296,7 @@ static void __maybe_unused throttle_active_work(struct work_struct *work)
 	state->average = avg;
 
 re_arm:
-	clear_therm_status_log(state->level);
+	thermal_clear_package_intr_status(state->level, THERM_STATUS_PROCHOT_LOG);
 	schedule_delayed_work_on(this_cpu, &state->therm_work, THERM_THROT_POLL_INTERVAL);
 }
 
diff --git a/drivers/thermal/intel/thermal_interrupt.h b/drivers/thermal/intel/thermal_interrupt.h
index 01e7bed2ffc7..01dfd4cdb5df 100644
--- a/drivers/thermal/intel/thermal_interrupt.h
+++ b/drivers/thermal/intel/thermal_interrupt.h
@@ -2,6 +2,9 @@
 #ifndef _INTEL_THERMAL_INTERRUPT_H
 #define _INTEL_THERMAL_INTERRUPT_H
 
+#define CORE_LEVEL	0
+#define PACKAGE_LEVEL	1
+
 /* Interrupt Handler for package thermal thresholds */
 extern int (*platform_thermal_package_notify)(__u64 msr_val);
 
@@ -15,4 +18,7 @@ extern bool (*platform_thermal_package_rate_control)(void);
 /* Handle HWP interrupt */
 extern void notify_hwp_interrupt(void);
 
+/* Common function to clear Package thermal status register */
+extern void thermal_clear_package_intr_status(int level, u64 bit_mask);
+
 #endif /* _INTEL_THERMAL_INTERRUPT_H */
diff --git a/drivers/thermal/intel/x86_pkg_temp_thermal.c b/drivers/thermal/intel/x86_pkg_temp_thermal.c
index a0e234fce71a..84c3a116ed04 100644
--- a/drivers/thermal/intel/x86_pkg_temp_thermal.c
+++ b/drivers/thermal/intel/x86_pkg_temp_thermal.c
@@ -265,7 +265,6 @@ static void pkg_temp_thermal_threshold_work_fn(struct work_struct *work)
 	struct thermal_zone_device *tzone = NULL;
 	int cpu = smp_processor_id();
 	struct zone_device *zonedev;
-	u64 msr_val, wr_val;
 
 	mutex_lock(&thermal_zone_mutex);
 	raw_spin_lock_irq(&pkg_temp_lock);
@@ -279,12 +278,8 @@ static void pkg_temp_thermal_threshold_work_fn(struct work_struct *work)
 	}
 	zonedev->work_scheduled = false;
 
-	rdmsrl(MSR_IA32_PACKAGE_THERM_STATUS, msr_val);
-	wr_val = msr_val & ~(THERM_LOG_THRESHOLD0 | THERM_LOG_THRESHOLD1);
-	if (wr_val != msr_val) {
-		wrmsrl(MSR_IA32_PACKAGE_THERM_STATUS, wr_val);
-		tzone = zonedev->tzone;
-	}
+	thermal_clear_package_intr_status(PACKAGE_LEVEL, THERM_LOG_THRESHOLD0 | THERM_LOG_THRESHOLD1);
+	tzone = zonedev->tzone;
 
 	enable_pkg_thres_interrupt();
 	raw_spin_unlock_irq(&pkg_temp_lock);
diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c
index 117eeaf7dd24..f17ab2316dbd 100644
--- a/drivers/thermal/thermal_core.c
+++ b/drivers/thermal/thermal_core.c
@@ -203,6 +203,9 @@ int thermal_zone_device_set_policy(struct thermal_zone_device *tz,
 	mutex_lock(&thermal_governor_lock);
 	mutex_lock(&tz->lock);
 
+	if (!device_is_registered(&tz->device))
+		goto exit;
+
 	gov = __find_governor(strim(policy));
 	if (!gov)
 		goto exit;
@@ -403,6 +406,34 @@ static void thermal_zone_device_init(struct thermal_zone_device *tz)
 		pos->initialized = false;
 }
 
+void __thermal_zone_device_update(struct thermal_zone_device *tz,
+				  enum thermal_notify_event event)
+{
+	int count;
+
+	if (atomic_read(&in_suspend))
+		return;
+
+	if (WARN_ONCE(!tz->ops->get_temp,
+		      "'%s' must not be called without 'get_temp' ops set\n",
+		      __func__))
+		return;
+
+	if (!thermal_zone_device_is_enabled(tz))
+		return;
+
+	update_temperature(tz);
+
+	__thermal_zone_set_trips(tz);
+
+	tz->notify_event = event;
+
+	for (count = 0; count < tz->num_trips; count++)
+		handle_thermal_trip(tz, count);
+
+	monitor_thermal_zone(tz);
+}
+
 static int thermal_zone_device_set_mode(struct thermal_zone_device *tz,
 					enum thermal_device_mode mode)
 {
@@ -417,15 +448,21 @@ static int thermal_zone_device_set_mode(struct thermal_zone_device *tz,
 		return ret;
 	}
 
+	if (!device_is_registered(&tz->device)) {
+		mutex_unlock(&tz->lock);
+
+		return -ENODEV;
+	}
+
 	if (tz->ops->change_mode)
 		ret = tz->ops->change_mode(tz, mode);
 
 	if (!ret)
 		tz->mode = mode;
 
-	mutex_unlock(&tz->lock);
+	__thermal_zone_device_update(tz, THERMAL_EVENT_UNSPECIFIED);
 
-	thermal_zone_device_update(tz, THERMAL_EVENT_UNSPECIFIED);
+	mutex_unlock(&tz->lock);
 
 	if (mode == THERMAL_DEVICE_ENABLED)
 		thermal_notify_tz_enable(tz->id);
@@ -457,31 +494,9 @@ int thermal_zone_device_is_enabled(struct thermal_zone_device *tz)
 void thermal_zone_device_update(struct thermal_zone_device *tz,
 				enum thermal_notify_event event)
 {
-	int count;
-
-	if (atomic_read(&in_suspend))
-		return;
-
-	if (WARN_ONCE(!tz->ops->get_temp, "'%s' must not be called without "
-		      "'get_temp' ops set\n", __func__))
-		return;
-
 	mutex_lock(&tz->lock);
-
-	if (!thermal_zone_device_is_enabled(tz))
-		goto out;
-
-	update_temperature(tz);
-
-	__thermal_zone_set_trips(tz);
-
-	tz->notify_event = event;
-
-	for (count = 0; count < tz->num_trips; count++)
-		handle_thermal_trip(tz, count);
-
-	monitor_thermal_zone(tz);
-out:
+	if (device_is_registered(&tz->device))
+		__thermal_zone_device_update(tz, event);
 	mutex_unlock(&tz->lock);
 }
 EXPORT_SYMBOL_GPL(thermal_zone_device_update);
@@ -603,8 +618,7 @@ int thermal_zone_bind_cooling_device(struct thermal_zone_device *tz,
 	struct thermal_instance *pos;
 	struct thermal_zone_device *pos1;
 	struct thermal_cooling_device *pos2;
-	unsigned long max_state;
-	int result, ret;
+	int result;
 
 	if (trip >= tz->num_trips || trip < 0)
 		return -EINVAL;
@@ -621,15 +635,11 @@ int thermal_zone_bind_cooling_device(struct thermal_zone_device *tz,
 	if (tz != pos1 || cdev != pos2)
 		return -EINVAL;
 
-	ret = cdev->ops->get_max_state(cdev, &max_state);
-	if (ret)
-		return ret;
-
 	/* lower default 0, upper default max_state */
 	lower = lower == THERMAL_NO_LIMIT ? 0 : lower;
-	upper = upper == THERMAL_NO_LIMIT ? max_state : upper;
+	upper = upper == THERMAL_NO_LIMIT ? cdev->max_state : upper;
 
-	if (lower > upper || upper > max_state)
+	if (lower > upper || upper > cdev->max_state)
 		return -EINVAL;
 
 	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
@@ -759,6 +769,7 @@ static void thermal_release(struct device *dev)
 		     sizeof("thermal_zone") - 1)) {
 		tz = to_thermal_zone(dev);
 		thermal_zone_destroy_device_groups(tz);
+		mutex_destroy(&tz->lock);
 		kfree(tz);
 	} else if (!strncmp(dev_name(dev), "cooling_device",
 			    sizeof("cooling_device") - 1)) {
@@ -883,10 +894,6 @@ __thermal_cooling_device_register(struct device_node *np,
 	cdev->id = ret;
 	id = ret;
 
-	ret = dev_set_name(&cdev->device, "cooling_device%d", cdev->id);
-	if (ret)
-		goto out_ida_remove;
-
 	cdev->type = kstrdup(type ? type : "", GFP_KERNEL);
 	if (!cdev->type) {
 		ret = -ENOMEM;
@@ -900,7 +907,17 @@ __thermal_cooling_device_register(struct device_node *np,
 	cdev->updated = false;
 	cdev->device.class = &thermal_class;
 	cdev->devdata = devdata;
+
+	ret = cdev->ops->get_max_state(cdev, &cdev->max_state);
+	if (ret)
+		goto out_kfree_type;
+
 	thermal_cooling_device_setup_sysfs(cdev);
+	ret = dev_set_name(&cdev->device, "cooling_device%d", cdev->id);
+	if (ret) {
+		thermal_cooling_device_destroy_sysfs(cdev);
+		goto out_kfree_type;
+	}
 	ret = device_register(&cdev->device);
 	if (ret)
 		goto out_kfree_type;
@@ -1234,10 +1251,6 @@ thermal_zone_device_register_with_trips(const char *type, struct thermal_trip *t
 	tz->id = id;
 	strscpy(tz->type, type, sizeof(tz->type));
 
-	result = dev_set_name(&tz->device, "thermal_zone%d", tz->id);
-	if (result)
-		goto remove_id;
-
 	if (!ops->critical)
 		ops->critical = thermal_zone_device_critical;
 
@@ -1260,6 +1273,11 @@ thermal_zone_device_register_with_trips(const char *type, struct thermal_trip *t
 	/* A new thermal zone needs to be updated anyway. */
 	atomic_set(&tz->need_update, 1);
 
+	result = dev_set_name(&tz->device, "thermal_zone%d", tz->id);
+	if (result) {
+		thermal_zone_destroy_device_groups(tz);
+		goto remove_id;
+	}
 	result = device_register(&tz->device);
 	if (result)
 		goto release_device;
@@ -1390,8 +1408,12 @@ void thermal_zone_device_unregister(struct thermal_zone_device *tz)
 	thermal_remove_hwmon_sysfs(tz);
 	ida_free(&thermal_tz_ida, tz->id);
 	ida_destroy(&tz->ida);
-	mutex_destroy(&tz->lock);
-	device_unregister(&tz->device);
+
+	mutex_lock(&tz->lock);
+	device_del(&tz->device);
+	mutex_unlock(&tz->lock);
+
+	put_device(&tz->device);
 
 	thermal_notify_tz_delete(tz_id);
 }
diff --git a/drivers/thermal/thermal_core.h b/drivers/thermal/thermal_core.h
index 1571917bd3c8..b834cb273429 100644
--- a/drivers/thermal/thermal_core.h
+++ b/drivers/thermal/thermal_core.h
@@ -109,9 +109,10 @@ int thermal_register_governor(struct thermal_governor *);
 void thermal_unregister_governor(struct thermal_governor *);
 int thermal_zone_device_set_policy(struct thermal_zone_device *, char *);
 int thermal_build_list_of_policies(char *buf);
+void __thermal_zone_device_update(struct thermal_zone_device *tz,
+				  enum thermal_notify_event event);
 
 /* Helpers */
-void thermal_zone_set_trips(struct thermal_zone_device *tz);
 void __thermal_zone_set_trips(struct thermal_zone_device *tz);
 int __thermal_zone_get_temp(struct thermal_zone_device *tz, int *temp);
 
diff --git a/drivers/thermal/thermal_helpers.c b/drivers/thermal/thermal_helpers.c
index c65cdce8f856..56aa2e88f34f 100644
--- a/drivers/thermal/thermal_helpers.c
+++ b/drivers/thermal/thermal_helpers.c
@@ -64,6 +64,20 @@ get_thermal_instance(struct thermal_zone_device *tz,
 }
 EXPORT_SYMBOL(get_thermal_instance);
 
+/**
+ * __thermal_zone_get_temp() - returns the temperature of a thermal zone
+ * @tz: a valid pointer to a struct thermal_zone_device
+ * @temp: a valid pointer to where to store the resulting temperature.
+ *
+ * When a valid thermal zone reference is passed, it will fetch its
+ * temperature and fill @temp.
+ *
+ * Both tz and tz->ops must be valid pointers when calling this function,
+ * and the tz->ops->get_temp callback must be provided.
+ * The function must be called under tz->lock.
+ *
+ * Return: On success returns 0, an error code otherwise
+ */
 int __thermal_zone_get_temp(struct thermal_zone_device *tz, int *temp)
 {
 	int ret = -EINVAL;
@@ -73,9 +87,6 @@ int __thermal_zone_get_temp(struct thermal_zone_device *tz, int *temp)
 
 	lockdep_assert_held(&tz->lock);
 
-	if (!tz || IS_ERR(tz) || !tz->ops->get_temp)
-		return -EINVAL;
-
 	ret = tz->ops->get_temp(tz, temp);
 
 	if (IS_ENABLED(CONFIG_THERMAL_EMULATION) && tz->emul_temperature) {
@@ -114,14 +125,43 @@ int thermal_zone_get_temp(struct thermal_zone_device *tz, int *temp)
 {
 	int ret;
 
+	if (IS_ERR_OR_NULL(tz))
+		return -EINVAL;
+
 	mutex_lock(&tz->lock);
-	ret = __thermal_zone_get_temp(tz, temp);
+
+	if (!tz->ops->get_temp) {
+		ret = -EINVAL;
+		goto unlock;
+	}
+
+	if (device_is_registered(&tz->device))
+		ret = __thermal_zone_get_temp(tz, temp);
+	else
+		ret = -ENODEV;
+
+unlock:
 	mutex_unlock(&tz->lock);
 
 	return ret;
 }
 EXPORT_SYMBOL_GPL(thermal_zone_get_temp);
 
+/**
+ * __thermal_zone_set_trips - Computes the next trip points for the driver
+ * @tz: a pointer to a thermal zone device structure
+ *
+ * The function computes the next temperature boundaries by browsing
+ * the trip points. The result is the closer low and high trip points
+ * to the current temperature. These values are passed to the backend
+ * driver to let it set its own notification mechanism (usually an
+ * interrupt).
+ *
+ * This function must be called with tz->lock held. Both tz and tz->ops
+ * must be valid pointers.
+ *
+ * It does not return a value
+ */
 void __thermal_zone_set_trips(struct thermal_zone_device *tz)
 {
 	int low = -INT_MAX;
@@ -168,25 +208,6 @@ void __thermal_zone_set_trips(struct thermal_zone_device *tz)
 		dev_err(&tz->device, "Failed to set trips: %d\n", ret);
 }
 
-/**
- * thermal_zone_set_trips - Computes the next trip points for the driver
- * @tz: a pointer to a thermal zone device structure
- *
- * The function computes the next temperature boundaries by browsing
- * the trip points. The result is the closer low and high trip points
- * to the current temperature. These values are passed to the backend
- * driver to let it set its own notification mechanism (usually an
- * interrupt).
- *
- * It does not return a value
- */
-void thermal_zone_set_trips(struct thermal_zone_device *tz)
-{
-	mutex_lock(&tz->lock);
-	__thermal_zone_set_trips(tz);
-	mutex_unlock(&tz->lock);
-}
-
 static void thermal_cdev_set_cur_state(struct thermal_cooling_device *cdev,
 				       int target)
 {
diff --git a/drivers/thermal/thermal_hwmon.c b/drivers/thermal/thermal_hwmon.c
index f53f4ceb6a5d..c594c42bea6d 100644
--- a/drivers/thermal/thermal_hwmon.c
+++ b/drivers/thermal/thermal_hwmon.c
@@ -77,7 +77,15 @@ temp_crit_show(struct device *dev, struct device_attribute *attr, char *buf)
 	int temperature;
 	int ret;
 
-	ret = tz->ops->get_crit_temp(tz, &temperature);
+	mutex_lock(&tz->lock);
+
+	if (device_is_registered(&tz->device))
+		ret = tz->ops->get_crit_temp(tz, &temperature);
+	else
+		ret = -ENODEV;
+
+	mutex_unlock(&tz->lock);
+
 	if (ret)
 		return ret;
 
diff --git a/drivers/thermal/thermal_sysfs.c b/drivers/thermal/thermal_sysfs.c
index ec495c7dff03..d97f0bc0a26b 100644
--- a/drivers/thermal/thermal_sysfs.c
+++ b/drivers/thermal/thermal_sysfs.c
@@ -92,7 +92,14 @@ trip_point_type_show(struct device *dev, struct device_attribute *attr,
 	if (sscanf(attr->attr.name, "trip_point_%d_type", &trip) != 1)
 		return -EINVAL;
 
-	result = tz->ops->get_trip_type(tz, trip, &type);
+	mutex_lock(&tz->lock);
+
+	if (device_is_registered(dev))
+		result = tz->ops->get_trip_type(tz, trip, &type);
+	else
+		result = -ENODEV;
+
+	mutex_unlock(&tz->lock);
 	if (result)
 		return result;
 
@@ -128,10 +135,17 @@ trip_point_temp_store(struct device *dev, struct device_attribute *attr,
 	if (kstrtoint(buf, 10, &temperature))
 		return -EINVAL;
 
+	mutex_lock(&tz->lock);
+
+	if (!device_is_registered(dev)) {
+		ret = -ENODEV;
+		goto unlock;
+	}
+
 	if (tz->ops->set_trip_temp) {
 		ret = tz->ops->set_trip_temp(tz, trip, temperature);
 		if (ret)
-			return ret;
+			goto unlock;
 	}
 
 	if (tz->trips)
@@ -140,16 +154,22 @@ trip_point_temp_store(struct device *dev, struct device_attribute *attr,
 	if (tz->ops->get_trip_hyst) {
 		ret = tz->ops->get_trip_hyst(tz, trip, &hyst);
 		if (ret)
-			return ret;
+			goto unlock;
 	}
 
 	ret = tz->ops->get_trip_type(tz, trip, &type);
 	if (ret)
-		return ret;
+		goto unlock;
 
 	thermal_notify_tz_trip_change(tz->id, trip, type, temperature, hyst);
 
-	thermal_zone_device_update(tz, THERMAL_EVENT_UNSPECIFIED);
+	__thermal_zone_device_update(tz, THERMAL_EVENT_UNSPECIFIED);
+
+unlock:
+	mutex_unlock(&tz->lock);
+
+	if (ret)
+		return ret;
 
 	return count;
 }
@@ -168,7 +188,14 @@ trip_point_temp_show(struct device *dev, struct device_attribute *attr,
 	if (sscanf(attr->attr.name, "trip_point_%d_temp", &trip) != 1)
 		return -EINVAL;
 
-	ret = tz->ops->get_trip_temp(tz, trip, &temperature);
+	mutex_lock(&tz->lock);
+
+	if (device_is_registered(dev))
+		ret = tz->ops->get_trip_temp(tz, trip, &temperature);
+	else
+		ret = -ENODEV;
+
+	mutex_unlock(&tz->lock);
 
 	if (ret)
 		return ret;
@@ -193,6 +220,13 @@ trip_point_hyst_store(struct device *dev, struct device_attribute *attr,
 	if (kstrtoint(buf, 10, &temperature))
 		return -EINVAL;
 
+	mutex_lock(&tz->lock);
+
+	if (!device_is_registered(dev)) {
+		ret = -ENODEV;
+		goto unlock;
+	}
+
 	/*
 	 * We are not doing any check on the 'temperature' value
 	 * here. The driver implementing 'set_trip_hyst' has to
@@ -201,7 +235,10 @@ trip_point_hyst_store(struct device *dev, struct device_attribute *attr,
 	ret = tz->ops->set_trip_hyst(tz, trip, temperature);
 
 	if (!ret)
-		thermal_zone_set_trips(tz);
+		__thermal_zone_set_trips(tz);
+
+unlock:
+	mutex_unlock(&tz->lock);
 
 	return ret ? ret : count;
 }
@@ -220,7 +257,14 @@ trip_point_hyst_show(struct device *dev, struct device_attribute *attr,
 	if (sscanf(attr->attr.name, "trip_point_%d_hyst", &trip) != 1)
 		return -EINVAL;
 
-	ret = tz->ops->get_trip_hyst(tz, trip, &temperature);
+	mutex_lock(&tz->lock);
+
+	if (device_is_registered(dev))
+		ret = tz->ops->get_trip_hyst(tz, trip, &temperature);
+	else
+		ret = -ENODEV;
+
+	mutex_unlock(&tz->lock);
 
 	return ret ? ret : sprintf(buf, "%d\n", temperature);
 }
@@ -269,16 +313,23 @@ emul_temp_store(struct device *dev, struct device_attribute *attr,
 	if (kstrtoint(buf, 10, &temperature))
 		return -EINVAL;
 
-	if (!tz->ops->set_emul_temp) {
-		mutex_lock(&tz->lock);
+	mutex_lock(&tz->lock);
+
+	if (!device_is_registered(dev)) {
+		ret = -ENODEV;
+		goto unlock;
+	}
+
+	if (!tz->ops->set_emul_temp)
 		tz->emul_temperature = temperature;
-		mutex_unlock(&tz->lock);
-	} else {
+	else
 		ret = tz->ops->set_emul_temp(tz, temperature);
-	}
 
 	if (!ret)
-		thermal_zone_device_update(tz, THERMAL_EVENT_UNSPECIFIED);
+		__thermal_zone_device_update(tz, THERMAL_EVENT_UNSPECIFIED);
+
+unlock:
+	mutex_unlock(&tz->lock);
 
 	return ret ? ret : count;
 }
@@ -589,13 +640,8 @@ static ssize_t max_state_show(struct device *dev, struct device_attribute *attr,
 			      char *buf)
 {
 	struct thermal_cooling_device *cdev = to_cooling_device(dev);
-	unsigned long state;
-	int ret;
 
-	ret = cdev->ops->get_max_state(cdev, &state);
-	if (ret)
-		return ret;
-	return sprintf(buf, "%ld\n", state);
+	return sprintf(buf, "%ld\n", cdev->max_state);
 }
 
 static ssize_t cur_state_show(struct device *dev, struct device_attribute *attr,
@@ -625,6 +671,10 @@ cur_state_store(struct device *dev, struct device_attribute *attr,
 	if ((long)state < 0)
 		return -EINVAL;
 
+	/* Requested state should be less than max_state + 1 */
+	if (state > cdev->max_state)
+		return -EINVAL;
+
 	mutex_lock(&cdev->lock);
 
 	result = cdev->ops->set_cur_state(cdev, state);
@@ -662,7 +712,6 @@ struct cooling_dev_stats {
 	spinlock_t lock;
 	unsigned int total_trans;
 	unsigned long state;
-	unsigned long max_states;
 	ktime_t last_time;
 	ktime_t *time_in_state;
 	unsigned int *trans_table;
@@ -692,7 +741,7 @@ void thermal_cooling_device_stats_update(struct thermal_cooling_device *cdev,
 		goto unlock;
 
 	update_time_in_state(stats);
-	stats->trans_table[stats->state * stats->max_states + new_state]++;
+	stats->trans_table[stats->state * (cdev->max_state + 1) + new_state]++;
 	stats->state = new_state;
 	stats->total_trans++;
 
@@ -726,7 +775,7 @@ time_in_state_ms_show(struct device *dev, struct device_attribute *attr,
 	spin_lock(&stats->lock);
 	update_time_in_state(stats);
 
-	for (i = 0; i < stats->max_states; i++) {
+	for (i = 0; i <= cdev->max_state; i++) {
 		len += sprintf(buf + len, "state%u\t%llu\n", i,
 			       ktime_to_ms(stats->time_in_state[i]));
 	}
@@ -741,7 +790,7 @@ reset_store(struct device *dev, struct device_attribute *attr, const char *buf,
 {
 	struct thermal_cooling_device *cdev = to_cooling_device(dev);
 	struct cooling_dev_stats *stats = cdev->stats;
-	int i, states = stats->max_states;
+	int i, states = cdev->max_state + 1;
 
 	spin_lock(&stats->lock);
 
@@ -750,7 +799,7 @@ reset_store(struct device *dev, struct device_attribute *attr, const char *buf,
 	memset(stats->trans_table, 0,
 	       states * states * sizeof(*stats->trans_table));
 
-	for (i = 0; i < stats->max_states; i++)
+	for (i = 0; i < states; i++)
 		stats->time_in_state[i] = ktime_set(0, 0);
 
 	spin_unlock(&stats->lock);
@@ -768,7 +817,7 @@ static ssize_t trans_table_show(struct device *dev,
 
 	len += snprintf(buf + len, PAGE_SIZE - len, " From  :    To\n");
 	len += snprintf(buf + len, PAGE_SIZE - len, "       : ");
-	for (i = 0; i < stats->max_states; i++) {
+	for (i = 0; i <= cdev->max_state; i++) {
 		if (len >= PAGE_SIZE)
 			break;
 		len += snprintf(buf + len, PAGE_SIZE - len, "state%2u  ", i);
@@ -778,17 +827,17 @@ static ssize_t trans_table_show(struct device *dev,
 
 	len += snprintf(buf + len, PAGE_SIZE - len, "\n");
 
-	for (i = 0; i < stats->max_states; i++) {
+	for (i = 0; i <= cdev->max_state; i++) {
 		if (len >= PAGE_SIZE)
 			break;
 
 		len += snprintf(buf + len, PAGE_SIZE - len, "state%2u:", i);
 
-		for (j = 0; j < stats->max_states; j++) {
+		for (j = 0; j <= cdev->max_state; j++) {
 			if (len >= PAGE_SIZE)
 				break;
 			len += snprintf(buf + len, PAGE_SIZE - len, "%8u ",
-				stats->trans_table[i * stats->max_states + j]);
+				stats->trans_table[i * (cdev->max_state + 1) + j]);
 		}
 		if (len >= PAGE_SIZE)
 			break;
@@ -824,14 +873,10 @@ static void cooling_device_stats_setup(struct thermal_cooling_device *cdev)
 {
 	const struct attribute_group *stats_attr_group = NULL;
 	struct cooling_dev_stats *stats;
-	unsigned long states;
+	/* Total number of states is highest state + 1 */
+	unsigned long states = cdev->max_state + 1;
 	int var;
 
-	if (cdev->ops->get_max_state(cdev, &states))
-		goto out;
-
-	states++; /* Total number of states is highest state + 1 */
-
 	var = sizeof(*stats);
 	var += sizeof(*stats->time_in_state) * states;
 	var += sizeof(*stats->trans_table) * states * states;
@@ -844,7 +889,6 @@ static void cooling_device_stats_setup(struct thermal_cooling_device *cdev)
 	stats->trans_table = (unsigned int *)(stats->time_in_state + states);
 	cdev->stats = stats;
 	stats->last_time = ktime_get();
-	stats->max_states = states;
 
 	spin_lock_init(&stats->lock);