Re: [RFC PATCH 6/12] thermal: introduce active cooling algorithm

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 一, 2012-06-11 at 10:21 +0530, Amit Kachhap wrote:
> On 11 June 2012 08:50, Zhang Rui <rui.zhang@xxxxxxxxx> wrote:
> >
> > Introduce thermal_zone_trip_update() to update the cooling state
> > of all the cooling devices that are binded to an active trip point.
> >
> > This will be used for passive cooling as well, in the future patches.
> > as both active and passive cooling can share the same algorithm,
> > which is
> >
> > 1. if the temperature is higher than a trip point,
> >   a. if the trend is THERMAL_TREND_RAISING, use higher cooling
> >      state for this trip point
> >   b. if the trend is THERMAL_TREND_DROPPING, use lower cooling
> >      state for this trip point
> >
> > 2. if the temperature is lower than a trip point, use lower
> >   cooling state for this trip point.
> >
> > Signed-off-by: Zhang Rui <rui.zhang@xxxxxxxxx>
> > ---
> >  drivers/acpi/thermal.c        |    7 ++-
> >  drivers/thermal/thermal_sys.c |   91 ++++++++++++++++++++++++++++++------------
> >  2 files changed, 71 insertions(+), 27 deletions(-)
> >
> > Index: rtd3/drivers/thermal/thermal_sys.c
> > ===================================================================
> > --- rtd3.orig/drivers/thermal/thermal_sys.c
> > +++ rtd3/drivers/thermal/thermal_sys.c
> > @@ -1035,6 +1035,70 @@ void thermal_cooling_device_unregister(s
> >  }
> >  EXPORT_SYMBOL(thermal_cooling_device_unregister);
> >
> > +/*
> > + * Cooling algorithm for active trip points
> > + *
> > + * 1. if the temperature is higher than a trip point,
> > + *    a. if the trend is THERMAL_TREND_RAISING, use higher cooling
> > + *       state for this trip point
> > + *    b. if the trend is THERMAL_TREND_DROPPING, use lower cooling
> > + *       state for this trip point
> > + *
> > + * 2. if the temperature is lower than a trip point, use lower
> > + *    cooling state for this trip point
> > + *
> > + * Note that this behaves the same as the previous passive cooling
> > + * algorithm.
> > + */
> > +
> > +static void thermal_zone_trip_update(struct thermal_zone_device *tz,
> > +                                    int trip, long temp)
> > +{
> > +       struct thermal_cooling_device_instance *instance;
> > +       struct thermal_cooling_device *cdev = NULL;
> > +       unsigned long cur_state, max_state;
> > +       long trip_temp;
> > +       enum thermal_trend trend;
> > +
> > +       tz->ops->get_trip_temp(tz, trip, &trip_temp);
> > +
> > +       if (temp >= trip_temp) {
> > +               thermal_get_trend(tz, trip, &trend);
> > +
> > +               list_for_each_entry(instance, &tz->cooling_devices, node) {
> > +                       if (instance->trip != trip)
> > +                               continue;
> > +
> > +                       cdev = instance->cdev;
> > +
> > +                       cdev->ops->get_cur_state(cdev, &cur_state);
> > +                       cdev->ops->get_max_state(cdev, &max_state);
> > +
> > +                       if (trend == THERMAL_TREND_RAISING) {
> > +                               cur_state = cur_state < instance->upper ?
> > +                                           (cur_state + 1) : instance->upper;
> > +                       } else if (trend == THERMAL_TREND_DROPPING) {
> > +                               cur_state = cur_state > instance->lower ?
> > +                                   (cur_state - 1) : instance->lower;
> > +                       }
> > +                       cdev->ops->set_cur_state(cdev, cur_state);
> > +               }
> > +       } else {        /* below trip */
> > +               list_for_each_entry(instance, &tz->cooling_devices, node) {
> > +                       if (instance->trip != trip)
> > +                               continue;
> Can we use the trend callback here even if the current temperature is
> less than the trip temperature. Basically according to trend it should
> deactivate the cooling devices as done for activating cooling devices.

yes, this depends on the cooling algorithm we want to use.
When the temperature is lower than the trip point, if we want to stop
the fan immediately instead of spinning down step by step, we can do
this.

Actually, what we are discussing right now should be some cooling
governor stuff. which I'm not sure if we should introduce to the generic
thermal manager or not, right now. :)

thanks,
rui

> > +
> > +                       cdev = instance->cdev;
> > +                       cdev->ops->get_cur_state(cdev, &cur_state);
> > +
> > +                       cur_state = cur_state > instance->lower ?
> > +                                   (cur_state - 1) : instance->lower;
> > +                       cdev->ops->set_cur_state(cdev, cur_state);
> > +               }
> > +       }
> > +
> > +       return;
> > +}
> >  /**
> >  * thermal_zone_device_update - force an update of a thermal zone's state
> >  * @ttz:       the thermal zone to update
> > @@ -1045,9 +1109,6 @@ void thermal_zone_device_update(struct t
> >        int count, ret = 0;
> >        long temp, trip_temp;
> >        enum thermal_trip_type trip_type;
> > -       struct thermal_cooling_device_instance *instance;
> > -       struct thermal_cooling_device *cdev;
> > -       unsigned long cur_state, max_state;
> >
> >        mutex_lock(&tz->lock);
> >
> > @@ -1083,29 +1144,7 @@ void thermal_zone_device_update(struct t
> >                                        tz->ops->notify(tz, count, trip_type);
> >                        break;
> >                case THERMAL_TRIP_ACTIVE:
> > -                       list_for_each_entry(instance, &tz->cooling_devices,
> > -                                           node) {
> > -                               if (instance->trip != count)
> > -                                       continue;
> > -
> > -                               cdev = instance->cdev;
> > -
> > -                               cdev->ops->get_cur_state(cdev, &cur_state);
> > -                               cdev->ops->get_max_state(cdev, &max_state);
> > -
> > -                               if (temp >= trip_temp)
> > -                                       cur_state =
> > -                                               cur_state < instance->upper ?
> > -                                               (cur_state + 1) :
> > -                                               instance->upper;
> > -                               else
> > -                                       cur_state =
> > -                                               cur_state > instance->lower ?
> > -                                               (cur_state - 1) :
> > -                                               instance->lower;
> > -
> > -                               cdev->ops->set_cur_state(cdev, cur_state);
> > -                       }
> > +                       thermal_zone_trip_update(tz, count, temp);
> >                        break;
> >                case THERMAL_TRIP_PASSIVE:
> >                        if (temp >= trip_temp || tz->passive)
> > Index: rtd3/drivers/acpi/thermal.c
> > ===================================================================
> > --- rtd3.orig/drivers/acpi/thermal.c
> > +++ rtd3/drivers/acpi/thermal.c
> > @@ -717,7 +717,12 @@ static int thermal_get_trend(struct ther
> >        if (thermal_get_trip_type(thermal, trip, &type))
> >                return -EINVAL;
> >
> > -       /* Only PASSIVE trip points need TREND */
> > +       if (type == THERMAL_TRIP_ACTIVE) {
> > +               /* aggressive active cooling */
> > +               *trend = THERMAL_TREND_RAISING;
> > +               return 0;
> > +       }
> > +
> >        if (type != THERMAL_TRIP_PASSIVE)
> >                return -EINVAL;
> >
> >
> >


_______________________________________________
linux-pm mailing list
linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linuxfoundation.org/mailman/listinfo/linux-pm



[Index of Archives]     [Linux ACPI]     [Netdev]     [Ethernet Bridging]     [Linux Wireless]     [CPU Freq]     [Kernel Newbies]     [Fedora Kernel]     [Security]     [Linux for Hams]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux Admin]     [Samba]

  Powered by Linux