ACPI: register ACPI Processor as generic thermal cooling device

Register ACPI processor as thermal cooling devices.
A combination of processor T-state and P-state are used for thermal throttling.
the processor will reduce the frequency first and then set the T-state.

we use cpufreq_thermal_reduction_pctg to calculate the cpufreq limit,
and call cpufreq_verify_with_limit to set the cpufreq limit.
if cpufreq driver is loaded, then we have four cooling state for cpufreq control.
cooling state 0: cpufreq limit == max_freq
cooling state 1: cpufreq limit == max_freq * 80%
cooling state 2: cpufreq limit == max_freq * 60%
cooling state 3: cpufreq limit == max_freq * 40%

after the cpufreq limit is set to 40 percentage of the max_freq,
we use T-state for cooling.

eg. a processor has P-state support, and it has 8 T-state (T0-T7),
the max_state of the proceesor is 10:

state	cpufreq-limit  T-state
0:	max_freq	T0
1:	max_freq * 80%	T0
2:	max_freq * 60%	T0
3:	max_freq * 40%	T0
4:	max_freq * 40%	T1
5:	max_freq * 40%	T2
6:	max_freq * 40%	T3
7:	max_freq * 40%	T4
8:	max_freq * 40%	T5
9:	max_freq * 40%	T6
10:	max_freq * 40%	T7

Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Thomas Sujith <sujith.thomas@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>

3 files changed, 155 insertions, 8 deletions

diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index e48ee4f8749f..5668c5e8ae19 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -668,6 +668,24 @@ static int __cpuinit acpi_processor_start(struct acpi_device *device)
 
         acpi_processor_power_init(pr, device);
 
+        pr->cdev = thermal_cooling_device_register("Processor", device,
+                                                &processor_cooling_ops);
+        if (pr->cdev)
+                printk(KERN_INFO PREFIX
+                        "%s is registered as cooling_device%d\n",
+                        device->dev.bus_id, pr->cdev->id);
+        else
+                goto end;
+
+        result = sysfs_create_link(&device->dev.kobj, &pr->cdev->device.kobj,
+                                        "thermal_cooling");
+        if (result)
+                return result;
+        result = sysfs_create_link(&pr->cdev->device.kobj, &device->dev.kobj,
+                                        "device");
+        if (result)
+                return result;
+
         if (pr->flags.throttling) {
                 printk(KERN_INFO PREFIX "%s [%s] (supports",
                        acpi_device_name(device), acpi_device_bid(device));
@@ -791,6 +809,11 @@ static int acpi_processor_remove(struct acpi_device *device, int type)
 
         acpi_processor_remove_fs(device);
 
+        sysfs_remove_link(&device->dev.kobj, "thermal_cooling");
+        sysfs_remove_link(&pr->cdev->device.kobj, "device");
+        thermal_cooling_device_unregister(pr->cdev);
+        pr->cdev = NULL;
+
         processors[pr->id] = NULL;
 
         kfree(pr);
diff --git a/drivers/acpi/processor_thermal.c b/drivers/acpi/processor_thermal.c
index 06e6f3fb8825..9cb43f52f7b6 100644
--- a/drivers/acpi/processor_thermal.c
+++ b/drivers/acpi/processor_thermal.c
@@ -32,6 +32,7 @@
 #include <linux/cpufreq.h>
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
+#include <linux/sysdev.h>
 
 #include <asm/uaccess.h>
 
@@ -93,6 +94,9 @@ static int acpi_processor_apply_limit(struct acpi_processor *pr)
  * _any_ cpufreq driver and not only the acpi-cpufreq driver.
  */
 
+#define CPUFREQ_THERMAL_MIN_STEP 0
+#define CPUFREQ_THERMAL_MAX_STEP 3
+
 static unsigned int cpufreq_thermal_reduction_pctg[NR_CPUS];
 static unsigned int acpi_thermal_cpufreq_is_init = 0;
 
@@ -109,8 +113,9 @@ static int acpi_thermal_cpufreq_increase(unsigned int cpu)
         if (!cpu_has_cpufreq(cpu))
                 return -ENODEV;
 
-        if (cpufreq_thermal_reduction_pctg[cpu] < 60) {
-                cpufreq_thermal_reduction_pctg[cpu] += 20;
+        if (cpufreq_thermal_reduction_pctg[cpu] <
+                CPUFREQ_THERMAL_MAX_STEP) {
+                cpufreq_thermal_reduction_pctg[cpu]++;
                 cpufreq_update_policy(cpu);
                 return 0;
         }
@@ -123,8 +128,9 @@ static int acpi_thermal_cpufreq_decrease(unsigned int cpu)
         if (!cpu_has_cpufreq(cpu))
                 return -ENODEV;
 
-        if (cpufreq_thermal_reduction_pctg[cpu] > 20)
-                cpufreq_thermal_reduction_pctg[cpu] -= 20;
+        if (cpufreq_thermal_reduction_pctg[cpu] >
+                (CPUFREQ_THERMAL_MIN_STEP + 1))
+                cpufreq_thermal_reduction_pctg[cpu]--;
         else
                 cpufreq_thermal_reduction_pctg[cpu] = 0;
         cpufreq_update_policy(cpu);
@@ -143,7 +149,7 @@ static int acpi_thermal_cpufreq_notifier(struct notifier_block *nb,
 
         max_freq =
             (policy->cpuinfo.max_freq *
-             (100 - cpufreq_thermal_reduction_pctg[policy->cpu])) / 100;
+             (100 - cpufreq_thermal_reduction_pctg[policy->cpu] * 20)) / 100;
 
         cpufreq_verify_within_limits(policy, 0, max_freq);
 
@@ -155,6 +161,32 @@ static struct notifier_block acpi_thermal_cpufreq_notifier_block = {
         .notifier_call = acpi_thermal_cpufreq_notifier,
 };
 
+static int cpufreq_get_max_state(unsigned int cpu)
+{
+        if (!cpu_has_cpufreq(cpu))
+                return 0;
+
+        return CPUFREQ_THERMAL_MAX_STEP;
+}
+
+static int cpufreq_get_cur_state(unsigned int cpu)
+{
+        if (!cpu_has_cpufreq(cpu))
+                return 0;
+
+        return cpufreq_thermal_reduction_pctg[cpu];
+}
+
+static int cpufreq_set_cur_state(unsigned int cpu, int state)
+{
+        if (!cpu_has_cpufreq(cpu))
+                return 0;
+
+        cpufreq_thermal_reduction_pctg[cpu] = state;
+        cpufreq_update_policy(cpu);
+        return 0;
+}
+
 void acpi_thermal_cpufreq_init(void)
 {
         int i;
@@ -179,6 +211,20 @@ void acpi_thermal_cpufreq_exit(void)
 }
 
 #else                           /* ! CONFIG_CPU_FREQ */
+static int cpufreq_get_max_state(unsigned int cpu)
+{
+        return 0;
+}
+
+static int cpufreq_get_cur_state(unsigned int cpu)
+{
+        return 0;
+}
+
+static int cpufreq_set_cur_state(unsigned int cpu, int state)
+{
+        return 0;
+}
 
 static int acpi_thermal_cpufreq_increase(unsigned int cpu)
 {
@@ -310,6 +356,84 @@ int acpi_processor_get_limit_info(struct acpi_processor *pr)
         return 0;
 }
 
+/* thermal coolign device callbacks */
+static int acpi_processor_max_state(struct acpi_processor *pr)
+{
+        int max_state = 0;
+
+        /*
+         * There exists four states according to
+         * cpufreq_thermal_reduction_ptg. 0, 1, 2, 3
+         */
+        max_state += cpufreq_get_max_state(pr->id);
+        if (pr->flags.throttling)
+                max_state += (pr->throttling.state_count -1);
+
+        return max_state;
+}
+static int
+processor_get_max_state(struct thermal_cooling_device *cdev, char *buf)
+{
+        struct acpi_device *device = cdev->devdata;
+        struct acpi_processor *pr = acpi_driver_data(device);
+
+        if (!device || !pr)
+                return -EINVAL;
+
+        return sprintf(buf, "%d\n", acpi_processor_max_state(pr));
+}
+
+static int
+processor_get_cur_state(struct thermal_cooling_device *cdev, char *buf)
+{
+        struct acpi_device *device = cdev->devdata;
+        struct acpi_processor *pr = acpi_driver_data(device);
+        int cur_state;
+
+        if (!device || !pr)
+                return -EINVAL;
+
+        cur_state = cpufreq_get_cur_state(pr->id);
+        if (pr->flags.throttling)
+                cur_state += pr->throttling.state;
+
+        return sprintf(buf, "%d\n", cur_state);
+}
+
+static int
+processor_set_cur_state(struct thermal_cooling_device *cdev, unsigned int state)
+{
+        struct acpi_device *device = cdev->devdata;
+        struct acpi_processor *pr = acpi_driver_data(device);
+        int result = 0;
+        int max_pstate;
+
+        if (!device || !pr)
+                return -EINVAL;
+
+        max_pstate = cpufreq_get_max_state(pr->id);
+
+        if (state > acpi_processor_max_state(pr))
+                return -EINVAL;
+
+        if (state <= max_pstate) {
+                if (pr->flags.throttling && pr->throttling.state)
+                        result = acpi_processor_set_throttling(pr, 0);
+                cpufreq_set_cur_state(pr->id, state);
+        } else {
+                cpufreq_set_cur_state(pr->id, max_pstate);
+                result = acpi_processor_set_throttling(pr,
+                                state - max_pstate);
+        }
+        return result;
+}
+
+struct thermal_cooling_device_ops processor_cooling_ops = {
+        .get_max_state = processor_get_max_state,
+        .get_cur_state = processor_get_cur_state,
+        .set_cur_state = processor_set_cur_state,
+};
+
 /* /proc interface */
 
 static int acpi_processor_limit_seq_show(struct seq_file *seq, void *offset)
diff --git a/include/acpi/processor.h b/include/acpi/processor.h
index 76411b1fc4fd..0787635a11aa 100644
--- a/include/acpi/processor.h
+++ b/include/acpi/processor.h
@@ -4,7 +4,7 @@
 #include <linux/kernel.h>
 #include <linux/cpu.h>
 #include <linux/cpuidle.h>
-
+#include <linux/thermal.h>
 #include <asm/acpi.h>
 
 #define ACPI_PROCESSOR_BUSY_METRIC      10
@@ -218,7 +218,7 @@ struct acpi_processor {
         struct acpi_processor_performance *performance;
         struct acpi_processor_throttling throttling;
         struct acpi_processor_limit limit;
-
+        struct thermal_cooling_device *cdev;
         /* the _PDC objects for this processor, if any */
         struct acpi_object_list *pdc;
 };
@@ -330,7 +330,7 @@ extern struct cpuidle_driver acpi_idle_driver;
 /* in processor_thermal.c */
 int acpi_processor_get_limit_info(struct acpi_processor *pr);
 extern struct file_operations acpi_processor_limit_fops;
-
+extern struct thermal_cooling_device_ops processor_cooling_ops;
 #ifdef CONFIG_CPU_FREQ
 void acpi_thermal_cpufreq_init(void);
 void acpi_thermal_cpufreq_exit(void);