On Thursday, July 19, 2012, Zhang Rui wrote:
> so that the thermal manager can easily find out
> the deepest cooling state request for a cooling device.
>
> cdev->lock is also introduced in this patch to protect this plist.
>
> Signed-off-by: Zhang Rui <rui.zhang@xxxxxxxxx>
I'm not sure if this is actually better than without the plist.
The plist manipulations in thermal_zone_trip_update() look like they
can add quite some overhead in some situations.
Do you have any quantitative justification for this change?
Rafael
> ---
> drivers/thermal/thermal_sys.c | 84 ++++++++++++++++++++++++++---------------
> include/linux/thermal.h | 4 +-
> 2 files changed, 57 insertions(+), 31 deletions(-)
>
> diff --git a/drivers/thermal/thermal_sys.c b/drivers/thermal/thermal_sys.c
> index a40dda1..9f56250 100644
> --- a/drivers/thermal/thermal_sys.c
> +++ b/drivers/thermal/thermal_sys.c
> @@ -54,11 +54,10 @@ struct thermal_instance {
> int trip;
> unsigned long upper; /* Highest cooling state for this trip point */
> unsigned long lower; /* Lowest cooling state for this trip point */
> - unsigned long target; /* expected cooling state */
> char attr_name[THERMAL_NAME_LENGTH];
> struct device_attribute attr;
> struct list_head tz_node; /* node in tz->instances */
> - struct list_head cdev_node; /* node in cdev->instances */
> + struct plist_node cdev_node; /* node in cdev->instances */
> };
>
> static DEFINE_IDR(thermal_tz_idr);
> @@ -787,7 +786,7 @@ int thermal_zone_bind_cooling_device(struct thermal_zone_device *tz,
> dev->trip = trip;
> dev->upper = upper;
> dev->lower = lower;
> - dev->target = -1;
> + plist_node_init(&dev->cdev_node, -1);
>
> result = get_idr(&tz->idr, &tz->lock, &dev->id);
> if (result)
> @@ -809,6 +808,7 @@ int thermal_zone_bind_cooling_device(struct thermal_zone_device *tz,
> goto remove_symbol_link;
>
> mutex_lock(&tz->lock);
> + mutex_lock(&cdev->lock);
> list_for_each_entry(pos, &tz->instances, tz_node)
> if (pos->tz == tz && pos->trip == trip && pos->cdev == cdev) {
> result = -EEXIST;
> @@ -816,8 +816,9 @@ int thermal_zone_bind_cooling_device(struct thermal_zone_device *tz,
> }
> if (!result) {
> list_add_tail(&dev->tz_node, &tz->instances);
> - list_add_tail(&dev->cdev_node, &cdev->instances);
> + plist_add(&dev->cdev_node, &cdev->instances);
> }
> + mutex_unlock(&cdev->lock);
> mutex_unlock(&tz->lock);
>
> if (!result)
> @@ -850,14 +851,17 @@ int thermal_zone_unbind_cooling_device(struct thermal_zone_device *tz,
> struct thermal_instance *pos, *next;
>
> mutex_lock(&tz->lock);
> + mutex_lock(&cdev->lock);
> list_for_each_entry_safe(pos, next, &tz->instances, tz_node) {
> if (pos->tz == tz && pos->trip == trip && pos->cdev == cdev) {
> list_del(&pos->tz_node);
> - list_del(&pos->cdev_node);
> + plist_del(&pos->cdev_node, &pos->cdev->instances);
> + mutex_unlock(&cdev->lock);
> mutex_unlock(&tz->lock);
> goto unbind;
> }
> }
> + mutex_unlock(&cdev->lock);
> mutex_unlock(&tz->lock);
>
> return -ENODEV;
> @@ -923,7 +927,8 @@ thermal_cooling_device_register(char *type, void *devdata,
> }
>
> strcpy(cdev->type, type);
> - INIT_LIST_HEAD(&cdev->instances);
> + plist_head_init(&cdev->instances);
> + mutex_init(&cdev->lock);
> cdev->ops = ops;
> cdev->updated = 1;
> cdev->device.class = &thermal_class;
> @@ -1019,26 +1024,21 @@ EXPORT_SYMBOL(thermal_cooling_device_unregister);
>
> static void thermal_zone_do_update(struct thermal_zone_device *tz)
> {
> - struct thermal_instance *instance1, *instance2;
> + struct thermal_instance *instance;
> + struct plist_node *node;
> struct thermal_cooling_device *cdev;
> - int target;
>
> - list_for_each_entry(instance1, &tz->instances, tz_node) {
> - cdev = instance1->cdev;
> + list_for_each_entry(instance, &tz->instances, tz_node) {
> + cdev = instance->cdev;
>
> /* cooling device has already been updated*/
> if (cdev->updated)
> continue;
>
> - target = 0;
> - /* Make sure cdev enters the deepest cooling state */
> - list_for_each_entry(instance2, &cdev->instances, cdev_node) {
> - if (instance2->target == -1)
> - continue;
> - if (instance2->target > target)
> - target = instance2->target;
> - }
> - cdev->ops->set_cur_state(cdev, target);
> + /* Deepest cooling state required */
> + node = plist_last(&cdev->instances);
> +
> + cdev->ops->set_cur_state(cdev, node->prio < 0 ? 0 : node->prio);
> cdev->updated = 1;
> }
> }
> @@ -1064,7 +1064,7 @@ static void thermal_zone_trip_update(struct thermal_zone_device *tz,
> {
> struct thermal_instance *instance;
> struct thermal_cooling_device *cdev = NULL;
> - unsigned long cur_state, max_state;
> + unsigned long cur_state, max_state, target_state;
> long trip_temp;
> enum thermal_trip_type trip_type;
> enum thermal_trend trend;
> @@ -1088,21 +1088,29 @@ static void thermal_zone_trip_update(struct thermal_zone_device *tz,
>
> cdev->ops->get_cur_state(cdev, &cur_state);
> cdev->ops->get_max_state(cdev, &max_state);
> -
> + target_state = cur_state;
> if (trend == THERMAL_TREND_RAISING) {
> - cur_state = cur_state < instance->upper ?
> + target_state = cur_state < instance->upper ?
> (cur_state + 1) : instance->upper;
> } else if (trend == THERMAL_TREND_DROPPING) {
> - cur_state = cur_state > instance->lower ?
> + target_state = cur_state > instance->lower ?
> (cur_state - 1) : instance->lower;
> }
>
> + if (target_state == cur_state)
> + continue;
> +
> /* activate a passive thermal instance */
> if (trip_type == THERMAL_TRIP_PASSIVE &&
> - instance->target == -1)
> + instance->cdev_node.prio == -1)
> tz->passive++;
>
> - instance->target = cur_state;
> + mutex_lock(&cdev->lock);
> + plist_del(&instance->cdev_node, &cdev->instances);
> + plist_node_init(&instance->cdev_node, target_state);
> + plist_add(&instance->cdev_node, &cdev->instances);
> + mutex_unlock(&cdev->lock);
> +
> cdev->updated = 0; /* cooling device needs update */
> }
> } else { /* below trip */
> @@ -1111,20 +1119,36 @@ static void thermal_zone_trip_update(struct thermal_zone_device *tz,
> continue;
>
> /* Do not use the deacitve thermal instance */
> - if (instance->target == -1)
> + if (instance->cdev_node.prio == -1)
> continue;
> cdev = instance->cdev;
> cdev->ops->get_cur_state(cdev, &cur_state);
>
> - cur_state = cur_state > instance->lower ?
> + target_state = cur_state > instance->lower ?
> (cur_state - 1) : -1;
>
> + if (target_state == cur_state)
> + continue;
> +
> /* deactivate a passive thermal instance */
> if (trip_type == THERMAL_TRIP_PASSIVE &&
> - cur_state == -1)
> + target_state == -1)
> tz->passive--;
> - instance->target = cur_state;
> - cdev->updated = 0; /* cooling device needs update */
> +
> + mutex_lock(&cdev->lock);
> + plist_del(&instance->cdev_node, &cdev->instances);
> + plist_node_init(&instance->cdev_node, target_state);
> + plist_add(&instance->cdev_node, &cdev->instances);
> + mutex_unlock(&cdev->lock);
> +
> + /*
> + * cooling device needs update.
> + * But if the target_state is -1, it means that
> + * the cooling device is already in cooling state 0,
> + * thus we do not need update
> + */
> + if (target_state != -1)
> + cdev->updated = 0;
> }
> }
>
> diff --git a/include/linux/thermal.h b/include/linux/thermal.h
> index f725eb9..376a59e 100644
> --- a/include/linux/thermal.h
> +++ b/include/linux/thermal.h
> @@ -28,6 +28,7 @@
> #include <linux/idr.h>
> #include <linux/device.h>
> #include <linux/workqueue.h>
> +#include <linux/plist.h>
>
> struct thermal_zone_device;
> struct thermal_cooling_device;
> @@ -93,7 +94,8 @@ struct thermal_cooling_device {
> void *devdata;
> const struct thermal_cooling_device_ops *ops;
> int updated; /* 1 if the cooling device does not need update */
> - struct list_head instances;
> + struct mutex lock;
> + struct plist_head instances;
> struct list_head node;
> };
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
