thermal/intel_powerclamp: Convert to CPU hotplug state

This is a conversation to the new hotplug state machine with
the difference that CPU_DEAD becomes CPU_PREDOWN.

At the same time it makes the handling of the two states symmetrical.
stop_power_clamp_worker() is called unconditionally and the controversial
error message is removed.

Finally, the hotplug state callbacks are removed after the powerclamping
is stopped to avoid a potential race.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
[pmladek@suse.com: Fixed the possible race in powerclamp_exit()]
Signed-off-by: Petr Mladek <pmladek@suse.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

diff --git a/drivers/thermal/intel_powerclamp.c b/drivers/thermal/intel_powerclamp.c
index a94f7c849a4eefa28b29e835e223203499689fb2..c99af71c0ae78386f18e42536fa6b606b84bfa6d 100644 (file)
--- a/drivers/thermal/intel_powerclamp.c
+++ b/drivers/thermal/intel_powerclamp.c
@@ -43,7 +43,6 @@
 #include <linux/kernel.h>
 #include <linux/delay.h>
 #include <linux/kthread.h>
-#include <linux/freezer.h>
 #include <linux/cpu.h>
 #include <linux/thermal.h>
 #include <linux/slab.h>
@@ -530,8 +529,7 @@ static void start_power_clamp_worker(unsigned long cpu)
        struct powerclamp_worker_data *w_data = per_cpu_ptr(worker_data, cpu);
        struct kthread_worker *worker;
 
-       worker = kthread_create_worker_on_cpu(cpu, KTW_FREEZABLE,
-                                             "kidle_inject/%ld", cpu);
+       worker = kthread_create_worker_on_cpu(cpu, 0, "kidle_inject/%ld", cpu);
        if (IS_ERR(worker))
                return;
 
@@ -622,43 +620,35 @@ static void end_power_clamp(void)
        }
 }
 
-static int powerclamp_cpu_callback(struct notifier_block *nfb,
-                               unsigned long action, void *hcpu)
+static int powerclamp_cpu_online(unsigned int cpu)
 {
-       unsigned long cpu = (unsigned long)hcpu;
+       if (clamping == false)
+               return 0;
+       start_power_clamp_worker(cpu);
+       /* prefer BSP as controlling CPU */
+       if (cpu == 0) {
+               control_cpu = 0;
+               smp_mb();
+       }
+       return 0;
+}
 
-       if (false == clamping)
-               goto exit_ok;
+static int powerclamp_cpu_predown(unsigned int cpu)
+{
+       if (clamping == false)
+               return 0;
 
-       switch (action) {
-       case CPU_ONLINE:
-               start_power_clamp_worker(cpu);
-               /* prefer BSP as controlling CPU */
-               if (cpu == 0) {
-                       control_cpu = 0;
-                       smp_mb();
-               }
-               break;
-       case CPU_DEAD:
-               if (test_bit(cpu, cpu_clamping_mask)) {
-                       pr_err("cpu %lu dead but powerclamping thread is not\n",
-                               cpu);
-                       stop_power_clamp_worker(cpu);
-               }
-               if (cpu == control_cpu) {
-                       control_cpu = smp_processor_id();
-                       smp_mb();
-               }
-       }
+       stop_power_clamp_worker(cpu);
+       if (cpu != control_cpu)
+               return 0;
 
-exit_ok:
-       return NOTIFY_OK;
+       control_cpu = cpumask_first(cpu_online_mask);
+       if (control_cpu == cpu)
+               control_cpu = cpumask_next(cpu, cpu_online_mask);
+       smp_mb();
+       return 0;
 }
 
-static struct notifier_block powerclamp_cpu_notifier = {
-       .notifier_call = powerclamp_cpu_callback,
-};
-
 static int powerclamp_get_max_state(struct thermal_cooling_device *cdev,
                                 unsigned long *state)
 {
@@ -788,6 +778,8 @@ file_error:
        debugfs_remove_recursive(debug_dir);
 }
 
+static enum cpuhp_state hp_state;
+
 static int __init powerclamp_init(void)
 {
        int retval;
@@ -805,7 +797,14 @@ static int __init powerclamp_init(void)
 
        /* set default limit, maybe adjusted during runtime based on feedback */
        window_size = 2;
-       register_hotcpu_notifier(&powerclamp_cpu_notifier);
+       retval = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
+                                          "thermal/intel_powerclamp:online",
+                                          powerclamp_cpu_online,
+                                          powerclamp_cpu_predown);
+       if (retval < 0)
+               goto exit_free;
+
+       hp_state = retval;
 
        worker_data = alloc_percpu(struct powerclamp_worker_data);
        if (!worker_data) {
@@ -830,7 +829,7 @@ static int __init powerclamp_init(void)
 exit_free_thread:
        free_percpu(worker_data);
 exit_unregister:
-       unregister_hotcpu_notifier(&powerclamp_cpu_notifier);
+       cpuhp_remove_state_nocalls(hp_state);
 exit_free:
        kfree(cpu_clamping_mask);
        return retval;
@@ -839,8 +838,8 @@ module_init(powerclamp_init);
 
 static void __exit powerclamp_exit(void)
 {
-       unregister_hotcpu_notifier(&powerclamp_cpu_notifier);
        end_power_clamp();
+       cpuhp_remove_state_nocalls(hp_state);
        free_percpu(worker_data);
        thermal_cooling_device_unregister(cooling_dev);
        kfree(cpu_clamping_mask);