Merge Intel thermal control drivers changes for v6.2

 - Add Raptor Lake-S support to the intel_tcc_cooling driver (Zhang
   Rui).

 - Make the intel_tcc_cooling driver detect TCC locking (Zhang Rui).

 - Address Coverity warning in intel_hfi_process_event() (Ricardo Neri).

 - Prevent accidental clearing of HFI in the package thermal interrupt
   status (Srinivas Pandruvada).

 - Protect the clearing of status bits in MSR_IA32_PACKAGE_THERM_STATUS
   and MSR_IA32_THERM_STATUS (Srinivas Pandruvada).

 - Allow the HFI interrupt handler to ACK an event for the same
   timestamp (Srinivas Pandruvada).

* thermal-intel:
  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: 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

Author: rafaeljw

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);

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 =

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);
 }
 

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 */

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);