Re: [PATCH 1/3] thermal: hisilicon: add new hisilicon thermal sensor driver
On 03/23/2015 12:46 PM, Leo Yan wrote:
On Fri, Mar 20, 2015 at 03:55:27PM +, Mark Rutland wrote:
That may be the case in the code as it stands today, but per the binding
the trip points are the temperatures at which an action is to be taken.
The thermal-zone has poilling-delay and polling-delay-passive, but
there's no reason you couldn't also use the interrupt to handle the
"hot" trip-point, adn the reset at the "critical" trip-point. All that's
missing is the plumbing in order to do so.
So please co-ordinate with the thermal framework to do that.
Let's dig further more for this point, so that we can get more specific
gudiance and have a good preparation for next version's patch set.
After i reviewed the thermal framework code, currently have one smooth
way to co-ordinate the trip points w/t thermal framework: use the function
*thermal_zone_device_register()* to register sensor, and can use the
callback function .get_trip_temp to tell thermal framework for the
trip points' temperature.
For hisi thermal case, now the driver is using the function
*thermal_zone_of_sensor_register* to register sensor, but use this way
i have not found there have standard APIs which can be used by sensor
driver to get the trip points info from thermal framework.
I may miss something for thermal framework, so if have existed APIs to
get the trip points, could pls point out?
I am only familiar with the binding, not the Linux implementation -- The
latter can change to accomodate your hardware without requiring binding
changes. Please co-ordinate with the thermal maintainers.
Found the functions of_thermal_get_trip_points(tz) and of_thermal_get_ntrips(tz)
will help for this.
+ if (of_property_read_bool(np, "hisilicon,tsensor-bind-irq")) {
+
+ if (data->irq_bind_sensor != -1)
+ dev_warn(&pdev->dev, "irq has bound to index %d\n",
+data->irq_bind_sensor);
+
+ /* bind irq to this sensor */
+ data->irq_bind_sensor = index;
+ }
I don't see why this should be specified in the DT. Why do you believe
it should?
The thermal sensor module has four sensors, but have only one
interrupt signal; This interrupt can only be used by one sensor;
So want to use dts to bind the interrupt with one selected sensor.
That's not all that great, though I'm not exactly sure how the kernel
would select the best sensor to measure with. It would be good if you
could talk to the thermal maintainers w.r.t. this.
This will be decided by the silicon, right? Every soc has different
combination with cpu/gpu/vpu, so which part is hottest, this maybe
highly dependent on individual SoC.
S/W just need provide the flexibility so that later can choose
the interrupt to bind with the sensor within the hottest part.
Then the property you care about is which sensor is closest to what is
likely to be the hottest component. Given that, the kernel can decide
how to use the interrupt.
Will modify the driver to dynamically bind the interrupt to hottest
sensor; Appreciate for good suggestion.
Thanks,
Leo Yan
Hi mark:
Thank you for your comments,please wait the next version to slove your
presenting problem.
Thanks.
Xinwei
--
To unsubscribe from this list: send the line "unsubscribe linux-pm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 1/3] thermal: hisilicon: add new hisilicon thermal sensor driver
On Fri, Mar 20, 2015 at 03:55:27PM +, Mark Rutland wrote:
> > > That may be the case in the code as it stands today, but per the binding
> > > the trip points are the temperatures at which an action is to be taken.
> > >
> > > The thermal-zone has poilling-delay and polling-delay-passive, but
> > > there's no reason you couldn't also use the interrupt to handle the
> > > "hot" trip-point, adn the reset at the "critical" trip-point. All that's
> > > missing is the plumbing in order to do so.
> > >
> > > So please co-ordinate with the thermal framework to do that.
> >
> > Let's dig further more for this point, so that we can get more specific
> > gudiance and have a good preparation for next version's patch set.
> >
> > After i reviewed the thermal framework code, currently have one smooth
> > way to co-ordinate the trip points w/t thermal framework: use the function
> > *thermal_zone_device_register()* to register sensor, and can use the
> > callback function .get_trip_temp to tell thermal framework for the
> > trip points' temperature.
> >
> > For hisi thermal case, now the driver is using the function
> > *thermal_zone_of_sensor_register* to register sensor, but use this way
> > i have not found there have standard APIs which can be used by sensor
> > driver to get the trip points info from thermal framework.
> >
> > I may miss something for thermal framework, so if have existed APIs to
> > get the trip points, could pls point out?
>
> I am only familiar with the binding, not the Linux implementation -- The
> latter can change to accomodate your hardware without requiring binding
> changes. Please co-ordinate with the thermal maintainers.
Found the functions of_thermal_get_trip_points(tz) and of_thermal_get_ntrips(tz)
will help for this.
> > > > > > + if (of_property_read_bool(np,
> > > > > > "hisilicon,tsensor-bind-irq")) {
> > > > > > +
> > > > > > + if (data->irq_bind_sensor != -1)
> > > > > > + dev_warn(&pdev->dev, "irq has bound to
> > > > > > index %d\n",
> > > > > > +data->irq_bind_sensor);
> > > > > > +
> > > > > > + /* bind irq to this sensor */
> > > > > > + data->irq_bind_sensor = index;
> > > > > > + }
> > > > >
> > > > > I don't see why this should be specified in the DT. Why do you believe
> > > > > it should?
> > > >
> > > > The thermal sensor module has four sensors, but have only one
> > > > interrupt signal; This interrupt can only be used by one sensor;
> > > > So want to use dts to bind the interrupt with one selected sensor.
> > >
> > > That's not all that great, though I'm not exactly sure how the kernel
> > > would select the best sensor to measure with. It would be good if you
> > > could talk to the thermal maintainers w.r.t. this.
> >
> > This will be decided by the silicon, right? Every soc has different
> > combination with cpu/gpu/vpu, so which part is hottest, this maybe
> > highly dependent on individual SoC.
> >
> > S/W just need provide the flexibility so that later can choose
> > the interrupt to bind with the sensor within the hottest part.
>
> Then the property you care about is which sensor is closest to what is
> likely to be the hottest component. Given that, the kernel can decide
> how to use the interrupt.
Will modify the driver to dynamically bind the interrupt to hottest
sensor; Appreciate for good suggestion.
Thanks,
Leo Yan
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 1/3] thermal: hisilicon: add new hisilicon thermal sensor driver
> > That may be the case in the code as it stands today, but per the binding
> > the trip points are the temperatures at which an action is to be taken.
> >
> > The thermal-zone has poilling-delay and polling-delay-passive, but
> > there's no reason you couldn't also use the interrupt to handle the
> > "hot" trip-point, adn the reset at the "critical" trip-point. All that's
> > missing is the plumbing in order to do so.
> >
> > So please co-ordinate with the thermal framework to do that.
>
> Let's dig further more for this point, so that we can get more specific
> gudiance and have a good preparation for next version's patch set.
>
> After i reviewed the thermal framework code, currently have one smooth
> way to co-ordinate the trip points w/t thermal framework: use the function
> *thermal_zone_device_register()* to register sensor, and can use the
> callback function .get_trip_temp to tell thermal framework for the
> trip points' temperature.
>
> For hisi thermal case, now the driver is using the function
> *thermal_zone_of_sensor_register* to register sensor, but use this way
> i have not found there have standard APIs which can be used by sensor
> driver to get the trip points info from thermal framework.
>
> I may miss something for thermal framework, so if have existed APIs to
> get the trip points, could pls point out?
I am only familiar with the binding, not the Linux implementation -- The
latter can change to accomodate your hardware without requiring binding
changes. Please co-ordinate with the thermal maintainers.
> > > > > + if (of_property_read_bool(np, "hisilicon,tsensor-bind-irq")) {
> > > > > +
> > > > > + if (data->irq_bind_sensor != -1)
> > > > > + dev_warn(&pdev->dev, "irq has bound to index
> > > > > %d\n",
> > > > > +data->irq_bind_sensor);
> > > > > +
> > > > > + /* bind irq to this sensor */
> > > > > + data->irq_bind_sensor = index;
> > > > > + }
> > > >
> > > > I don't see why this should be specified in the DT. Why do you believe
> > > > it should?
> > >
> > > The thermal sensor module has four sensors, but have only one
> > > interrupt signal; This interrupt can only be used by one sensor;
> > > So want to use dts to bind the interrupt with one selected sensor.
> >
> > That's not all that great, though I'm not exactly sure how the kernel
> > would select the best sensor to measure with. It would be good if you
> > could talk to the thermal maintainers w.r.t. this.
>
> This will be decided by the silicon, right? Every soc has different
> combination with cpu/gpu/vpu, so which part is hottest, this maybe
> highly dependent on individual SoC.
>
> S/W just need provide the flexibility so that later can choose
> the interrupt to bind with the sensor within the hottest part.
Then the property you care about is which sensor is closest to what is
likely to be the hottest component. Given that, the kernel can decide
how to use the interrupt.
Thanks,
Mark.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 1/3] thermal: hisilicon: add new hisilicon thermal sensor driver
On 2015年03月20日 19:24, Mark Rutland wrote:
+ ret = of_property_read_u32(np, "hisilicon,tsensor-thres-temp",
+ &sensor->thres_temp);
+ if (ret) {
+ dev_err(&pdev->dev, "failed to get thres value %d: %d\n",
+ index, ret);
+ return ret;
+ }
+
+ ret = of_property_read_u32(np, "hisilicon,tsensor-reset-temp",
+ &sensor->reset_temp);
+ if (ret) {
+ dev_err(&pdev->dev, "failed to get reset value %d: %d\n",
+ index, ret);
+ return ret;
+ }
I see now that these properties result in the HW being programmed. You
should figure out how to reconcile these with thermal-zone trip points
rather than having parallel properties.
Set "tsensor-thres-temp" to register so that if thermal reaches the
threshold, the sensor will trigger h/w interrupt.
Set "tsensor-reset-temp" to register so that if thermal reaches the
reset value, the sensor will assert SoC reset signal to trigger h/w
reset.
I understand this.
This is different w/t thermal-zone trip points, the trip points are
used for timer polling.
That may be the case in the code as it stands today, but per the binding
the trip points are the temperatures at which an action is to be taken.
The thermal-zone has poilling-delay and polling-delay-passive, but
there's no reason you couldn't also use the interrupt to handle the
"hot" trip-point, adn the reset at the "critical" trip-point. All that's
missing is the plumbing in order to do so.
So please co-ordinate with the thermal framework to do that.
In order to co-ordinate with the thermal framework, We will fix
"tsensor-thres-temp"
and "tsensor-reset-temp" value in the dts. This temperatue will higher than
thermal
zone trip-point value. During polling delay time if SoC temperature is above
thermal
zone trip-point value and below "tsensor-thres-temp" value,the systerm will use
this
thermal framework ways to realize basic function such as cpu cooling device.
Otherwise
it will use interrput mode to cause some functions.
This interrupt mode help thermal framework way to complete function, in high
temperature
situation this interrupt is better than other mode. I think that it can work
well.
Xinwei
Do u think below modification is more reasonable?
- Set "tsensor-thres-temp" = 70, which equal to thermal-zone
passive trip point, so that we can use h/w interrupt to update the
thermal value immediately, rather than using polling method w/t long
delay;
- Set "tsensor-reset-temp" = <10>, which is higher than
thermal-zone critical trip point, so that the s/w reset method has
higher priority than h/w reset; we also can easily know the reset is
caused by thermal framework; "tsensor-reset-temp" is only used to
protect h/w circuit.
As mentioned above, I think that you should co-ordinate with the thermal
framework. You're worknig around limitations inthe code as it stands
today rather than solving the fundamental issue.
+ if (of_property_read_bool(np, "hisilicon,tsensor-bind-irq")) {
+
+ if (data->irq_bind_sensor != -1)
+ dev_warn(&pdev->dev, "irq has bound to index %d\n",
+data->irq_bind_sensor);
+
+ /* bind irq to this sensor */
+ data->irq_bind_sensor = index;
+ }
I don't see why this should be specified in the DT. Why do you believe
it should?
The thermal sensor module has four sensors, but have only one
interrupt signal; This interrupt can only be used by one sensor;
So want to use dts to bind the interrupt with one selected sensor.
That's not all that great, though I'm not exactly sure how the kernel
would select the best sensor to measure with. It would be good if you
could talk to the thermal maintainers w.r.t. this.
+static int hisi_thermal_probe(struct platform_device *pdev)
+{
+ struct hisi_thermal_data *data;
+ struct resource *res;
+ int i;
+ int ret;
+
+ if (!cpufreq_get_current_driver()) {
+ dev_dbg(&pdev->dev, "no cpufreq driver!");
+ return -EPROBE_DEFER;
+ }
Surely we care about not burning out the board even if we don't have
cpufreq?
Is there any ordering guarantee between the probing of this driver and
cpufreq?
Yes, here need binding the thermal sensor w/t cpu cooling device,
and cpu cooling device is based on cpufreq driver.
Sure, but if you don't have a cooling device you still want the critical
temperature reset and so on, no?
Mark.
--
To unsubscribe from this list: send the line "unsubscribe linux-pm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel
Re: [PATCH 1/3] thermal: hisilicon: add new hisilicon thermal sensor driver
On Fri, Mar 20, 2015 at 11:24:14AM +, Mark Rutland wrote:
> > > > + ret = of_property_read_u32(np, "hisilicon,tsensor-thres-temp",
> > > > + &sensor->thres_temp);
> > > > + if (ret) {
> > > > + dev_err(&pdev->dev, "failed to get thres value %d:
> > > > %d\n",
> > > > + index, ret);
> > > > + return ret;
> > > > + }
> > > > +
> > > > + ret = of_property_read_u32(np, "hisilicon,tsensor-reset-temp",
> > > > + &sensor->reset_temp);
> > > > + if (ret) {
> > > > + dev_err(&pdev->dev, "failed to get reset value %d:
> > > > %d\n",
> > > > + index, ret);
> > > > + return ret;
> > > > + }
> > >
> > > I see now that these properties result in the HW being programmed. You
> > > should figure out how to reconcile these with thermal-zone trip points
> > > rather than having parallel properties.
> >
> > Set "tsensor-thres-temp" to register so that if thermal reaches the
> > threshold, the sensor will trigger h/w interrupt.
> >
> > Set "tsensor-reset-temp" to register so that if thermal reaches the
> > reset value, the sensor will assert SoC reset signal to trigger h/w
> > reset.
>
> I understand this.
>
> > This is different w/t thermal-zone trip points, the trip points are
> > used for timer polling.
>
> That may be the case in the code as it stands today, but per the binding
> the trip points are the temperatures at which an action is to be taken.
>
> The thermal-zone has poilling-delay and polling-delay-passive, but
> there's no reason you couldn't also use the interrupt to handle the
> "hot" trip-point, adn the reset at the "critical" trip-point. All that's
> missing is the plumbing in order to do so.
>
> So please co-ordinate with the thermal framework to do that.
Let's dig further more for this point, so that we can get more specific
gudiance and have a good preparation for next version's patch set.
After i reviewed the thermal framework code, currently have one smooth
way to co-ordinate the trip points w/t thermal framework: use the function
*thermal_zone_device_register()* to register sensor, and can use the
callback function .get_trip_temp to tell thermal framework for the
trip points' temperature.
For hisi thermal case, now the driver is using the function
*thermal_zone_of_sensor_register* to register sensor, but use this way
i have not found there have standard APIs which can be used by sensor
driver to get the trip points info from thermal framework.
I may miss something for thermal framework, so if have existed APIs to
get the trip points, could pls point out?
> > Do u think below modification is more reasonable?
> >
> > - Set "tsensor-thres-temp" = 70, which equal to thermal-zone
> > passive trip point, so that we can use h/w interrupt to update the
> > thermal value immediately, rather than using polling method w/t long
> > delay;
> >
> > - Set "tsensor-reset-temp" = <10>, which is higher than
> > thermal-zone critical trip point, so that the s/w reset method has
> > higher priority than h/w reset; we also can easily know the reset is
> > caused by thermal framework; "tsensor-reset-temp" is only used to
> > protect h/w circuit.
>
> As mentioned above, I think that you should co-ordinate with the thermal
> framework. You're worknig around limitations inthe code as it stands
> today rather than solving the fundamental issue.
>
> > > > + if (of_property_read_bool(np, "hisilicon,tsensor-bind-irq")) {
> > > > +
> > > > + if (data->irq_bind_sensor != -1)
> > > > + dev_warn(&pdev->dev, "irq has bound to index
> > > > %d\n",
> > > > +data->irq_bind_sensor);
> > > > +
> > > > + /* bind irq to this sensor */
> > > > + data->irq_bind_sensor = index;
> > > > + }
> > >
> > > I don't see why this should be specified in the DT. Why do you believe
> > > it should?
> >
> > The thermal sensor module has four sensors, but have only one
> > interrupt signal; This interrupt can only be used by one sensor;
> > So want to use dts to bind the interrupt with one selected sensor.
>
> That's not all that great, though I'm not exactly sure how the kernel
> would select the best sensor to measure with. It would be good if you
> could talk to the thermal maintainers w.r.t. this.
This will be decided by the silicon, right? Every soc has different
combination with cpu/gpu/vpu, so which part is hottest, this maybe
highly dependent on individual SoC.
S/W just need provide the flexibility so that later can choose
the interrupt to bind with the sensor within the hottest part.
> > > > +static int hisi_thermal_probe(struct platform_device *pdev)
> > > > +{
> > > > + struct hisi_thermal_data *data;
> > > > + struct resource *res;
> > > > + int i;
> > > > + int ret;
Re: [PATCH 1/3] thermal: hisilicon: add new hisilicon thermal sensor driver
> > > + ret = of_property_read_u32(np, "hisilicon,tsensor-thres-temp",
> > > + &sensor->thres_temp);
> > > + if (ret) {
> > > + dev_err(&pdev->dev, "failed to get thres value %d: %d\n",
> > > + index, ret);
> > > + return ret;
> > > + }
> > > +
> > > + ret = of_property_read_u32(np, "hisilicon,tsensor-reset-temp",
> > > + &sensor->reset_temp);
> > > + if (ret) {
> > > + dev_err(&pdev->dev, "failed to get reset value %d: %d\n",
> > > + index, ret);
> > > + return ret;
> > > + }
> >
> > I see now that these properties result in the HW being programmed. You
> > should figure out how to reconcile these with thermal-zone trip points
> > rather than having parallel properties.
>
> Set "tsensor-thres-temp" to register so that if thermal reaches the
> threshold, the sensor will trigger h/w interrupt.
>
> Set "tsensor-reset-temp" to register so that if thermal reaches the
> reset value, the sensor will assert SoC reset signal to trigger h/w
> reset.
I understand this.
> This is different w/t thermal-zone trip points, the trip points are
> used for timer polling.
That may be the case in the code as it stands today, but per the binding
the trip points are the temperatures at which an action is to be taken.
The thermal-zone has poilling-delay and polling-delay-passive, but
there's no reason you couldn't also use the interrupt to handle the
"hot" trip-point, adn the reset at the "critical" trip-point. All that's
missing is the plumbing in order to do so.
So please co-ordinate with the thermal framework to do that.
> Do u think below modification is more reasonable?
>
> - Set "tsensor-thres-temp" = 70, which equal to thermal-zone
> passive trip point, so that we can use h/w interrupt to update the
> thermal value immediately, rather than using polling method w/t long
> delay;
>
> - Set "tsensor-reset-temp" = <10>, which is higher than
> thermal-zone critical trip point, so that the s/w reset method has
> higher priority than h/w reset; we also can easily know the reset is
> caused by thermal framework; "tsensor-reset-temp" is only used to
> protect h/w circuit.
As mentioned above, I think that you should co-ordinate with the thermal
framework. You're worknig around limitations inthe code as it stands
today rather than solving the fundamental issue.
> > > + if (of_property_read_bool(np, "hisilicon,tsensor-bind-irq")) {
> > > +
> > > + if (data->irq_bind_sensor != -1)
> > > + dev_warn(&pdev->dev, "irq has bound to index
> > > %d\n",
> > > +data->irq_bind_sensor);
> > > +
> > > + /* bind irq to this sensor */
> > > + data->irq_bind_sensor = index;
> > > + }
> >
> > I don't see why this should be specified in the DT. Why do you believe
> > it should?
>
> The thermal sensor module has four sensors, but have only one
> interrupt signal; This interrupt can only be used by one sensor;
> So want to use dts to bind the interrupt with one selected sensor.
That's not all that great, though I'm not exactly sure how the kernel
would select the best sensor to measure with. It would be good if you
could talk to the thermal maintainers w.r.t. this.
> > > +static int hisi_thermal_probe(struct platform_device *pdev)
> > > +{
> > > + struct hisi_thermal_data *data;
> > > + struct resource *res;
> > > + int i;
> > > + int ret;
> > > +
> > > + if (!cpufreq_get_current_driver()) {
> > > + dev_dbg(&pdev->dev, "no cpufreq driver!");
> > > + return -EPROBE_DEFER;
> > > + }
> >
> > Surely we care about not burning out the board even if we don't have
> > cpufreq?
> >
> > Is there any ordering guarantee between the probing of this driver and
> > cpufreq?
>
> Yes, here need binding the thermal sensor w/t cpu cooling device,
> and cpu cooling device is based on cpufreq driver.
Sure, but if you don't have a cooling device you still want the critical
temperature reset and so on, no?
Mark.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 1/3] thermal: hisilicon: add new hisilicon thermal sensor driver
在 2015/3/19 22:17, Mark Rutland 写道:
> On Thu, Mar 19, 2015 at 07:57:27AM +, kongxinwei wrote:
>> This patch adds the support for hisilicon thermal sensor, within
>> hisilicon SoC. there will register sensors for thermal framework
>> and use device tree to bind cooling device.
>>
>> Signed-off-by: Leo Yan
>> Signed-off-by: kongxinwei
>> ---
>> drivers/thermal/Kconfig| 8 +
>> drivers/thermal/Makefile | 1 +
>> drivers/thermal/hisi_thermal.c | 531
>> +
>> 3 files changed, 540 insertions(+)
>> create mode 100644 drivers/thermal/hisi_thermal.c
>
> [...]
>
>> + ret = of_property_read_u32(np, "hisilicon,tsensor-lag-value",
>> + &sensor->lag);
>
> This wasn't in the binding.
>
good comment, delete it.
Xinwei
> [...]
>
>> + ret = of_property_read_u32(np, "hisilicon,tsensor-thres-temp",
>> + &sensor->thres_temp);
>> + if (ret) {
>> + dev_err(&pdev->dev, "failed to get thres value %d: %d\n",
>> + index, ret);
>> + return ret;
>> + }
>> +
>> + ret = of_property_read_u32(np, "hisilicon,tsensor-reset-temp",
>> + &sensor->reset_temp);
>> + if (ret) {
>> + dev_err(&pdev->dev, "failed to get reset value %d: %d\n",
>> + index, ret);
>> + return ret;
>> + }
>
> I see now that these properties result in the HW being programmed. You
> should figure out how to reconcile these with thermal-zone trip points
> rather than having parallel properties.
>
oh,firstly,this "tsensor-thres-temp" property is threshold temperature value
which causes interrupt function by setting thermal value register. "thermal
-zone trip points" applies scanning mode to cause other function such as
cooling freq .., but this "tsensor-thres-temp" property be used by interrupt
mode. when scanning mode don't satisfies systerm demands, the interrut mode
perfectly help scanning mode to complete function. "tsensor-thres-temp" temp-
erature is higher than "thermal-zone trip points" temperature, so this "
tsensor-thres-temp" property is secondary attribute.
secondly, this "tsensor-reset-temp" property is hardware protect temperature
which is close to or is below to burn out the SoC. When SoC temperature is
"tsensor-reset-temp" temperature value, SoC will be force to reboot and ensure
SoC not to burn out. So it don't conflict thermal-zone.
>> +
>> + if (of_property_read_bool(np, "hisilicon,tsensor-bind-irq")) {
>> +
>> + if (data->irq_bind_sensor != -1)
>> + dev_warn(&pdev->dev, "irq has bound to index %d\n",
>> +data->irq_bind_sensor);
>> +
>> + /* bind irq to this sensor */
>> + data->irq_bind_sensor = index;
>> + }
>
> I don't see why this should be specified in the DT. Why do you believe
> it should?
>
SoC include foure thermal sensor modules,every modules is able to cause
interrupt,so binding a module to realize interupt function and i believe
it should be specified.
> [...]
>
>> +static int hisi_thermal_probe(struct platform_device *pdev)
>> +{
>> + struct hisi_thermal_data *data;
>> + struct resource *res;
>> + int i;
>> + int ret;
>> +
>> + if (!cpufreq_get_current_driver()) {
>> + dev_dbg(&pdev->dev, "no cpufreq driver!");
>> + return -EPROBE_DEFER;
>> + }
>
> Surely we care about not burning out the board even if we don't have
> cpufreq?
>
> Is there any ordering guarantee between the probing of this driver and
> cpufreq?
>
>
Yes! It will use thermal framework to realize cpu cooling device function.
> [...]
>
>> + data->clk = devm_clk_get(&pdev->dev, NULL);
>
> You gave this clock a name in the binding. Use it or drop it.
>
Thanks comment,use it.
> Mark.
>
Xinwei
> .
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 1/3] thermal: hisilicon: add new hisilicon thermal sensor driver
hi Mark,
Thanks for reviewing, pls see below questions.
On Thu, Mar 19, 2015 at 02:17:53PM +, Mark Rutland wrote:
> On Thu, Mar 19, 2015 at 07:57:27AM +, kongxinwei wrote:
> > This patch adds the support for hisilicon thermal sensor, within
> > hisilicon SoC. there will register sensors for thermal framework
> > and use device tree to bind cooling device.
> >
> > Signed-off-by: Leo Yan
> > Signed-off-by: kongxinwei
> > ---
> > drivers/thermal/Kconfig| 8 +
> > drivers/thermal/Makefile | 1 +
> > drivers/thermal/hisi_thermal.c | 531
> > +
> > 3 files changed, 540 insertions(+)
> > create mode 100644 drivers/thermal/hisi_thermal.c
[...]
> > + ret = of_property_read_u32(np, "hisilicon,tsensor-thres-temp",
> > + &sensor->thres_temp);
> > + if (ret) {
> > + dev_err(&pdev->dev, "failed to get thres value %d: %d\n",
> > + index, ret);
> > + return ret;
> > + }
> > +
> > + ret = of_property_read_u32(np, "hisilicon,tsensor-reset-temp",
> > + &sensor->reset_temp);
> > + if (ret) {
> > + dev_err(&pdev->dev, "failed to get reset value %d: %d\n",
> > + index, ret);
> > + return ret;
> > + }
>
> I see now that these properties result in the HW being programmed. You
> should figure out how to reconcile these with thermal-zone trip points
> rather than having parallel properties.
Set "tsensor-thres-temp" to register so that if thermal reaches the
threshold, the sensor will trigger h/w interrupt.
Set "tsensor-reset-temp" to register so that if thermal reaches the
reset value, the sensor will assert SoC reset signal to trigger h/w
reset.
This is different w/t thermal-zone trip points, the trip points are
used for timer polling. Do u think below modification is more
reasonable?
- Set "tsensor-thres-temp" = 70, which equal to thermal-zone
passive trip point, so that we can use h/w interrupt to update the
thermal value immediately, rather than using polling method w/t long
delay;
- Set "tsensor-reset-temp" = <10>, which is higher than
thermal-zone critical trip point, so that the s/w reset method has
higher priority than h/w reset; we also can easily know the reset is
caused by thermal framework; "tsensor-reset-temp" is only used to
protect h/w circuit.
tsensor: tsensor@0,f7030700 {
acpu1_sensor {
hisilicon,tsensor-id = <1>;
hisilicon,tsensor-lag-value = <10>;
hisilicon,tsensor-thres-temp = <7>;
hisilicon,tsensor-reset-temp = <10>;
};
};
thermal-zones {
...
cluster1: cluster1 {
polling-delay-passive = <1000>; /* milliseconds */
polling-delay = <5000>; /* milliseconds */
/* sensor ID */
thermal-sensors = <&tsensor 1>;
trips {
cluster1_alert: cluster1_alert {
temperature = <7>; /* millicelsius
*/
hysteresis = <2000>; /* millicelsius */
type = "passive";
};
cluster1_crit: cluster1_crit {
temperature = <9>; /* millicelsius
*/
hysteresis = <2000>; /* millicelsius */
type = "critical";
};
};
};
> > +
> > + if (of_property_read_bool(np, "hisilicon,tsensor-bind-irq")) {
> > +
> > + if (data->irq_bind_sensor != -1)
> > + dev_warn(&pdev->dev, "irq has bound to index %d\n",
> > +data->irq_bind_sensor);
> > +
> > + /* bind irq to this sensor */
> > + data->irq_bind_sensor = index;
> > + }
>
> I don't see why this should be specified in the DT. Why do you believe
> it should?
The thermal sensor module has four sensors, but have only one
interrupt signal; This interrupt can only be used by one sensor;
So want to use dts to bind the interrupt with one selected sensor.
> [...]
>
> > +static int hisi_thermal_probe(struct platform_device *pdev)
> > +{
> > + struct hisi_thermal_data *data;
> > + struct resource *res;
> > + int i;
> > + int ret;
> > +
> > + if (!cpufreq_get_current_driver()) {
> > + dev_dbg(&pdev->dev, "no cpufreq driver!");
> > + return -EPROBE_DEFER;
> > + }
>
> Surely we care about not burning out the board even if we don't have
Re: [PATCH 1/3] thermal: hisilicon: add new hisilicon thermal sensor driver
On Thu, Mar 19, 2015 at 07:57:27AM +, kongxinwei wrote:
> This patch adds the support for hisilicon thermal sensor, within
> hisilicon SoC. there will register sensors for thermal framework
> and use device tree to bind cooling device.
>
> Signed-off-by: Leo Yan
> Signed-off-by: kongxinwei
> ---
> drivers/thermal/Kconfig| 8 +
> drivers/thermal/Makefile | 1 +
> drivers/thermal/hisi_thermal.c | 531
> +
> 3 files changed, 540 insertions(+)
> create mode 100644 drivers/thermal/hisi_thermal.c
[...]
> + ret = of_property_read_u32(np, "hisilicon,tsensor-lag-value",
> + &sensor->lag);
This wasn't in the binding.
[...]
> + ret = of_property_read_u32(np, "hisilicon,tsensor-thres-temp",
> + &sensor->thres_temp);
> + if (ret) {
> + dev_err(&pdev->dev, "failed to get thres value %d: %d\n",
> + index, ret);
> + return ret;
> + }
> +
> + ret = of_property_read_u32(np, "hisilicon,tsensor-reset-temp",
> + &sensor->reset_temp);
> + if (ret) {
> + dev_err(&pdev->dev, "failed to get reset value %d: %d\n",
> + index, ret);
> + return ret;
> + }
I see now that these properties result in the HW being programmed. You
should figure out how to reconcile these with thermal-zone trip points
rather than having parallel properties.
> +
> + if (of_property_read_bool(np, "hisilicon,tsensor-bind-irq")) {
> +
> + if (data->irq_bind_sensor != -1)
> + dev_warn(&pdev->dev, "irq has bound to index %d\n",
> +data->irq_bind_sensor);
> +
> + /* bind irq to this sensor */
> + data->irq_bind_sensor = index;
> + }
I don't see why this should be specified in the DT. Why do you believe
it should?
[...]
> +static int hisi_thermal_probe(struct platform_device *pdev)
> +{
> + struct hisi_thermal_data *data;
> + struct resource *res;
> + int i;
> + int ret;
> +
> + if (!cpufreq_get_current_driver()) {
> + dev_dbg(&pdev->dev, "no cpufreq driver!");
> + return -EPROBE_DEFER;
> + }
Surely we care about not burning out the board even if we don't have
cpufreq?
Is there any ordering guarantee between the probing of this driver and
cpufreq?
[...]
> + data->clk = devm_clk_get(&pdev->dev, NULL);
You gave this clock a name in the binding. Use it or drop it.
Mark.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[PATCH 1/3] thermal: hisilicon: add new hisilicon thermal sensor driver
This patch adds the support for hisilicon thermal sensor, within
hisilicon SoC. there will register sensors for thermal framework
and use device tree to bind cooling device.
Signed-off-by: Leo Yan
Signed-off-by: kongxinwei
---
drivers/thermal/Kconfig| 8 +
drivers/thermal/Makefile | 1 +
drivers/thermal/hisi_thermal.c | 531 +
3 files changed, 540 insertions(+)
create mode 100644 drivers/thermal/hisi_thermal.c
diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
index af40db0..81aee01 100644
--- a/drivers/thermal/Kconfig
+++ b/drivers/thermal/Kconfig
@@ -136,6 +136,14 @@ config THERMAL_EMULATION
because userland can easily disable the thermal policy by simply
flooding this sysfs node with low temperature values.
+config HISI_THERMAL
+ tristate "Hisilicon thermal driver"
+ depends on ARCH_HISI && CPU_THERMAL && OF
+ help
+ Enable this to plug hisilicon's thermal sensor driver into the Linux
+ thermal framework. cpufreq is used as the cooling device to throttle
+ CPUs when the passive trip is crossed.
+
config IMX_THERMAL
tristate "Temperature sensor driver for Freescale i.MX SoCs"
depends on CPU_THERMAL
diff --git a/drivers/thermal/Makefile b/drivers/thermal/Makefile
index fa0dc48..08ae7ac 100644
--- a/drivers/thermal/Makefile
+++ b/drivers/thermal/Makefile
@@ -39,3 +39,4 @@ obj-$(CONFIG_TI_SOC_THERMAL) += ti-soc-thermal/
obj-$(CONFIG_INT340X_THERMAL) += int340x_thermal/
obj-$(CONFIG_ST_THERMAL) += st/
obj-$(CONFIG_TEGRA_SOCTHERM) += tegra_soctherm.o
+obj-$(CONFIG_HISI_THERMAL) += hisi_thermal.o
diff --git a/drivers/thermal/hisi_thermal.c b/drivers/thermal/hisi_thermal.c
new file mode 100644
index 000..438b1d1
--- /dev/null
+++ b/drivers/thermal/hisi_thermal.c
@@ -0,0 +1,531 @@
+/*
+ * Hisilicon thermal sensor driver
+ *
+ * Copyright (c) 2014-2015 Hisilicon Limited.
+ * Copyright (c) 2014-2015 Linaro Limited.
+ *
+ * Xinwei Kong
+ * Leo Yan
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+
+#define TEMP0_LAG (0x0)
+#define TEMP0_TH (0x4)
+#define TEMP0_RST_TH (0x8)
+#define TEMP0_CFG (0xC)
+#define TEMP0_EN (0x10)
+#define TEMP0_INT_EN (0x14)
+#define TEMP0_INT_CLR (0x18)
+#define TEMP0_RST_MSK (0x1C)
+#define TEMP0_RAW_INT (0x20)
+#define TEMP0_MSK_INT (0x24)
+#define TEMP0_VALUE(0x28)
+
+#define HISI_TEMP_BASE (-60)
+#define HISI_TEMP_PASSIVE (85000)
+
+#define HISI_MAX_SENSORS 4
+
+struct hisi_thermal_sensor {
+ struct hisi_thermal_data *thermal;
+ struct thermal_zone_device *tzd;
+
+ uint32_t id;
+ uint32_t lag;
+ uint32_t thres_temp, reset_temp;
+};
+
+struct hisi_thermal_data {
+ struct platform_device *pdev;
+ struct clk *clk;
+
+ int irq, irq_bind_sensor;
+ bool irq_enabled;
+
+ unsigned int sensors_num;
+ struct hisi_thermal_sensor sensors[HISI_MAX_SENSORS];
+
+ void __iomem *regs;
+};
+
+static DEFINE_SPINLOCK(thermal_lock);
+
+/* in millicelsius */
+static inline int _step_to_temp(int step)
+{
+ /*
+* Every step equals (1 * 200) / 255 celsius, and finally
+* need convert to millicelsius.
+*/
+ return (HISI_TEMP_BASE + (step * 200 / 255)) * 1000;
+}
+
+static inline int _temp_to_step(int temp)
+{
+ return ((temp / 1000 - HISI_TEMP_BASE) * 255 / 200);
+}
+
+static long hisi_thermal_get_sensor_temp(struct hisi_thermal_data *data,
+struct hisi_thermal_sensor *sensor)
+{
+ unsigned long flags;
+ int val;
+
+ spin_lock_irqsave(&thermal_lock, flags);
+
+ /* disable module firstly */
+ writel(0x0, data->regs + TEMP0_EN);
+
+ /* select sensor id */
+ writel((sensor->id << 12), data->regs + TEMP0_CFG);
+
+ /* enable module */
+ writel(0x1, data->regs + TEMP0_EN);
+
+ mdelay(5);
+
+ val = readl(data->regs + TEMP0_VALUE);
+ val = _step_to_temp(val);
+
+ /* adjust for negative value */
+ val = (val < 0) ? 0 : val;
+
+ spin_unlock_irqrestore(&thermal_lock, f
