2022-09-01 14:48:30

by Amjad Ouled-Ameur

[permalink] [raw]
Subject: [PATCH v3 4/4] thermal: mediatek: add another get_temp ops for thermal sensors

Provide thermal zone to read thermal sensor in the SoC. We can read all the
thermal sensors value in the SoC by the node /sys/class/thermal/

In mtk_thermal_bank_temperature, return -EAGAIN instead of -EACCESS
on the first read of sensor that often are bogus values.
This can avoid following warning on boot:

thermal thermal_zone6: failed to read out thermal zone (-13)

Signed-off-by: Michael Kao <[email protected]>
Signed-off-by: Hsin-Yi Wang <[email protected]>
Signed-off-by: Amjad Ouled-Ameur <[email protected]>
---
Changes in V3:
- Use proper types.
- Use devm_kmalloc() instead of kmalloc().
- Fix tabs and spaces.

drivers/thermal/mtk_thermal.c | 100 ++++++++++++++++++++++++++--------
1 file changed, 76 insertions(+), 24 deletions(-)

diff --git a/drivers/thermal/mtk_thermal.c b/drivers/thermal/mtk_thermal.c
index 088c388da241..5901787c57f5 100644
--- a/drivers/thermal/mtk_thermal.c
+++ b/drivers/thermal/mtk_thermal.c
@@ -259,6 +259,11 @@ enum mtk_thermal_version {

struct mtk_thermal;

+struct mtk_thermal_zone {
+ struct mtk_thermal *mt;
+ int id;
+};
+
struct thermal_bank_cfg {
unsigned int num_sensors;
const int *sensors;
@@ -709,6 +714,32 @@ static void mtk_thermal_put_bank(struct mtk_thermal_bank *bank)
mutex_unlock(&mt->lock);
}

+static int _get_sensor_temp(struct mtk_thermal *mt, int id)
+{
+ u32 raw;
+ int temp;
+
+ const struct mtk_thermal_data *conf = mt->conf;
+
+ raw = readl(mt->thermal_base + conf->msr[id]);
+
+ if (mt->conf->version == MTK_THERMAL_V1)
+ temp = raw_to_mcelsius_v1(mt, id, raw);
+ else
+ temp = raw_to_mcelsius_v2(mt, id, raw);
+
+ /*
+ * The first read of a sensor often contains very high bogus
+ * temperature value. Filter these out so that the system does
+ * not immediately shut down.
+ */
+
+ if (temp > 200000)
+ return -EAGAIN;
+ else
+ return temp;
+}
+
/**
* mtk_thermal_bank_temperature - get the temperature of a bank
* @bank: The bank
@@ -721,26 +752,9 @@ static int mtk_thermal_bank_temperature(struct mtk_thermal_bank *bank)
struct mtk_thermal *mt = bank->mt;
const struct mtk_thermal_data *conf = mt->conf;
int i, temp = INT_MIN, max = INT_MIN;
- u32 raw;

for (i = 0; i < conf->bank_data[bank->id].num_sensors; i++) {
- raw = readl(mt->thermal_base + conf->msr[i]);
-
- if (mt->conf->version == MTK_THERMAL_V1) {
- temp = raw_to_mcelsius_v1(
- mt, conf->bank_data[bank->id].sensors[i], raw);
- } else {
- temp = raw_to_mcelsius_v2(
- mt, conf->bank_data[bank->id].sensors[i], raw);
- }
-
- /*
- * The first read of a sensor often contains very high bogus
- * temperature value. Filter these out so that the system does
- * not immediately shut down.
- */
- if (temp > 200000)
- temp = 0;
+ temp = _get_sensor_temp(mt, i);

if (temp > max)
max = temp;
@@ -751,7 +765,8 @@ static int mtk_thermal_bank_temperature(struct mtk_thermal_bank *bank)

static int mtk_read_temp(void *data, int *temperature)
{
- struct mtk_thermal *mt = data;
+ struct mtk_thermal_zone *tz = data;
+ struct mtk_thermal *mt = tz->mt;
int i;
int tempmax = INT_MIN;

@@ -770,10 +785,28 @@ static int mtk_read_temp(void *data, int *temperature)
return 0;
}

+static int mtk_read_sensor_temp(void *data, int *temperature)
+{
+ struct mtk_thermal_zone *tz = data;
+ struct mtk_thermal *mt = tz->mt;
+ int id = tz->id - 1;
+
+ if (id < 0)
+ return -EACCES;
+
+ *temperature = _get_sensor_temp(mt, id);
+
+ return 0;
+}
+
static const struct thermal_zone_of_device_ops mtk_thermal_ops = {
.get_temp = mtk_read_temp,
};

+static const struct thermal_zone_of_device_ops mtk_thermal_sensor_ops = {
+ .get_temp = mtk_read_sensor_temp,
+};
+
static void mtk_thermal_init_bank(struct mtk_thermal *mt, int num,
u32 apmixed_phys_base, u32 auxadc_phys_base,
int ctrl_id)
@@ -1072,6 +1105,7 @@ static int mtk_thermal_probe(struct platform_device *pdev)
u64 auxadc_phys_base, apmixed_phys_base;
struct thermal_zone_device *tzdev;
void __iomem *apmixed_base, *auxadc_base;
+ struct mtk_thermal_zone *tz;

mt = devm_kzalloc(&pdev->dev, sizeof(*mt), GFP_KERNEL);
if (!mt)
@@ -1161,11 +1195,29 @@ static int mtk_thermal_probe(struct platform_device *pdev)

platform_set_drvdata(pdev, mt);

- tzdev = devm_thermal_zone_of_sensor_register(&pdev->dev, 0, mt,
- &mtk_thermal_ops);
- if (IS_ERR(tzdev)) {
- ret = PTR_ERR(tzdev);
- goto err_disable_clk_peri_therm;
+ for (i = 0; i < mt->conf->num_sensors + 1; i++) {
+ tz = devm_kmalloc(&pdev->dev, sizeof(*tz), GFP_KERNEL);
+ if (!tz)
+ return -ENOMEM;
+
+ tz->mt = mt;
+ tz->id = i;
+
+ tzdev = devm_thermal_zone_of_sensor_register(&pdev->dev, i, tz, (i == 0) ?
+ &mtk_thermal_ops :
+ &mtk_thermal_sensor_ops);
+
+ if (IS_ERR(tzdev)) {
+ if (PTR_ERR(tzdev) == -ENODEV) {
+ dev_warn(&pdev->dev,
+ "sensor %d not registered in thermal zone in dt\n", i);
+ continue;
+ }
+ if (PTR_ERR(tzdev) == -EACCES) {
+ ret = PTR_ERR(tzdev);
+ goto err_disable_clk_peri_therm;
+ }
+ }
}

ret = devm_thermal_add_hwmon_sysfs(tzdev);
--
2.37.3


2022-09-08 09:52:50

by Daniel Lezcano

[permalink] [raw]
Subject: Re: [PATCH v3 4/4] thermal: mediatek: add another get_temp ops for thermal sensors


Hi Amjad,

On 01/09/2022 15:39, Amjad Ouled-Ameur wrote:
> Provide thermal zone to read thermal sensor in the SoC. We can read all the
> thermal sensors value in the SoC by the node /sys/class/thermal/
>
> In mtk_thermal_bank_temperature, return -EAGAIN instead of -EACCESS
> on the first read of sensor that often are bogus values.
> This can avoid following warning on boot:
>
> thermal thermal_zone6: failed to read out thermal zone (-13)
>
> Signed-off-by: Michael Kao <[email protected]>
> Signed-off-by: Hsin-Yi Wang <[email protected]>
> Signed-off-by: Amjad Ouled-Ameur <[email protected]>
> ---
> Changes in V3:
> - Use proper types.
> - Use devm_kmalloc() instead of kmalloc().
> - Fix tabs and spaces.
>
> drivers/thermal/mtk_thermal.c | 100 ++++++++++++++++++++++++++--------
> 1 file changed, 76 insertions(+), 24 deletions(-)
>
> diff --git a/drivers/thermal/mtk_thermal.c b/drivers/thermal/mtk_thermal.c
> index 088c388da241..5901787c57f5 100644
> --- a/drivers/thermal/mtk_thermal.c
> +++ b/drivers/thermal/mtk_thermal.c
> @@ -259,6 +259,11 @@ enum mtk_thermal_version {
>
> struct mtk_thermal;
>
> +struct mtk_thermal_zone {
> + struct mtk_thermal *mt;
> + int id;
> +};
> +
> struct thermal_bank_cfg {
> unsigned int num_sensors;
> const int *sensors;
> @@ -709,6 +714,32 @@ static void mtk_thermal_put_bank(struct mtk_thermal_bank *bank)
> mutex_unlock(&mt->lock);
> }
>
> +static int _get_sensor_temp(struct mtk_thermal *mt, int id)
> +{
> + u32 raw;
> + int temp;
> +
> + const struct mtk_thermal_data *conf = mt->conf;
> +
> + raw = readl(mt->thermal_base + conf->msr[id]);
> +
> + if (mt->conf->version == MTK_THERMAL_V1)
> + temp = raw_to_mcelsius_v1(mt, id, raw);
> + else
> + temp = raw_to_mcelsius_v2(mt, id, raw);

Can you set a callback at init time instead of checking the version at
each get_sensor_temp() ?

> + /*
> + * The first read of a sensor often contains very high bogus
> + * temperature value. Filter these out so that the system does
> + * not immediately shut down.
> + */
> +
> + if (temp > 200000)
> + return -EAGAIN;
> + else
> + return temp;
> +}
> +
> /**
> * mtk_thermal_bank_temperature - get the temperature of a bank
> * @bank: The bank
> @@ -721,26 +752,9 @@ static int mtk_thermal_bank_temperature(struct mtk_thermal_bank *bank)
> struct mtk_thermal *mt = bank->mt;
> const struct mtk_thermal_data *conf = mt->conf;
> int i, temp = INT_MIN, max = INT_MIN;
> - u32 raw;
>
> for (i = 0; i < conf->bank_data[bank->id].num_sensors; i++) {
> - raw = readl(mt->thermal_base + conf->msr[i]);
> -
> - if (mt->conf->version == MTK_THERMAL_V1) {
> - temp = raw_to_mcelsius_v1(
> - mt, conf->bank_data[bank->id].sensors[i], raw);
> - } else {
> - temp = raw_to_mcelsius_v2(
> - mt, conf->bank_data[bank->id].sensors[i], raw);
> - }
> -
> - /*
> - * The first read of a sensor often contains very high bogus
> - * temperature value. Filter these out so that the system does
> - * not immediately shut down.
> - */
> - if (temp > 200000)
> - temp = 0;
> + temp = _get_sensor_temp(mt, i);
>
> if (temp > max)
> max = temp;
> @@ -751,7 +765,8 @@ static int mtk_thermal_bank_temperature(struct mtk_thermal_bank *bank)
>
> static int mtk_read_temp(void *data, int *temperature)
> {
> - struct mtk_thermal *mt = data;
> + struct mtk_thermal_zone *tz = data;
> + struct mtk_thermal *mt = tz->mt;
> int i;
> int tempmax = INT_MIN;
>
> @@ -770,10 +785,28 @@ static int mtk_read_temp(void *data, int *temperature)
> return 0;
> }
>
> +static int mtk_read_sensor_temp(void *data, int *temperature)
> +{
> + struct mtk_thermal_zone *tz = data;
> + struct mtk_thermal *mt = tz->mt;
> + int id = tz->id - 1;
> +
> + if (id < 0)
> + return -EACCES;
> +
> + *temperature = _get_sensor_temp(mt, id);
> +
> + return 0;
> +}
> +
> static const struct thermal_zone_of_device_ops mtk_thermal_ops = {
> .get_temp = mtk_read_temp,
> };
>
> +static const struct thermal_zone_of_device_ops mtk_thermal_sensor_ops = {
> + .get_temp = mtk_read_sensor_temp,
> +};

Please respin against linux-next, the thermal_zone_of_device_ops
structure does no longer exist. The conversion is trivial, here is a
example:

https://lore.kernel.org/all/[email protected]/


> static void mtk_thermal_init_bank(struct mtk_thermal *mt, int num,
> u32 apmixed_phys_base, u32 auxadc_phys_base,
> int ctrl_id)
> @@ -1072,6 +1105,7 @@ static int mtk_thermal_probe(struct platform_device *pdev)
> u64 auxadc_phys_base, apmixed_phys_base;
> struct thermal_zone_device *tzdev;
> void __iomem *apmixed_base, *auxadc_base;
> + struct mtk_thermal_zone *tz;
>
> mt = devm_kzalloc(&pdev->dev, sizeof(*mt), GFP_KERNEL);
> if (!mt)
> @@ -1161,11 +1195,29 @@ static int mtk_thermal_probe(struct platform_device *pdev)
>
> platform_set_drvdata(pdev, mt);
>
> - tzdev = devm_thermal_zone_of_sensor_register(&pdev->dev, 0, mt,
> - &mtk_thermal_ops);
> - if (IS_ERR(tzdev)) {
> - ret = PTR_ERR(tzdev);
> - goto err_disable_clk_peri_therm;
> + for (i = 0; i < mt->conf->num_sensors + 1; i++) {
> + tz = devm_kmalloc(&pdev->dev, sizeof(*tz), GFP_KERNEL);
> + if (!tz)
> + return -ENOMEM;
> +
> + tz->mt = mt;
> + tz->id = i;
> +
> + tzdev = devm_thermal_zone_of_sensor_register(&pdev->dev, i, tz, (i == 0) ?
> + &mtk_thermal_ops :
> + &mtk_thermal_sensor_ops);
> +
> + if (IS_ERR(tzdev)) {
> + if (PTR_ERR(tzdev) == -ENODEV) {
> + dev_warn(&pdev->dev,
> + "sensor %d not registered in thermal zone in dt\n", i);
> + continue;
> + }
> + if (PTR_ERR(tzdev) == -EACCES) {
> + ret = PTR_ERR(tzdev);
> + goto err_disable_clk_peri_therm;
> + }
> + }
> }
>
> ret = devm_thermal_add_hwmon_sysfs(tzdev);


--
<http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro: <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog

2022-09-08 16:13:44

by Amjad Ouled-Ameur

[permalink] [raw]
Subject: Re: [PATCH v3 4/4] thermal: mediatek: add another get_temp ops for thermal sensors

Hi Daniel,

Thank you for the review.

On 9/8/22 11:07, Daniel Lezcano wrote:
>
> Hi Amjad,
>
> On 01/09/2022 15:39, Amjad Ouled-Ameur wrote:
>> Provide thermal zone to read thermal sensor in the SoC. We can read
>> all the
>> thermal sensors value in the SoC by the node /sys/class/thermal/
>>
>> In mtk_thermal_bank_temperature, return -EAGAIN instead of -EACCESS
>> on the first read of sensor that often are bogus values.
>> This can avoid following warning on boot:
>>
>>    thermal thermal_zone6: failed to read out thermal zone (-13)
>>
>> Signed-off-by: Michael Kao <[email protected]>
>> Signed-off-by: Hsin-Yi Wang <[email protected]>
>> Signed-off-by: Amjad Ouled-Ameur <[email protected]>
>> ---
>> Changes in V3:
>> - Use proper types.
>> - Use devm_kmalloc() instead of kmalloc().
>> - Fix tabs and spaces.
>>
>>   drivers/thermal/mtk_thermal.c | 100 ++++++++++++++++++++++++++--------
>>   1 file changed, 76 insertions(+), 24 deletions(-)
>>
>> diff --git a/drivers/thermal/mtk_thermal.c
>> b/drivers/thermal/mtk_thermal.c
>> index 088c388da241..5901787c57f5 100644
>> --- a/drivers/thermal/mtk_thermal.c
>> +++ b/drivers/thermal/mtk_thermal.c
>> @@ -259,6 +259,11 @@ enum mtk_thermal_version {
>>     struct mtk_thermal;
>>   +struct mtk_thermal_zone {
>> +    struct mtk_thermal *mt;
>> +    int id;
>> +};
>> +
>>   struct thermal_bank_cfg {
>>       unsigned int num_sensors;
>>       const int *sensors;
>> @@ -709,6 +714,32 @@ static void mtk_thermal_put_bank(struct
>> mtk_thermal_bank *bank)
>>           mutex_unlock(&mt->lock);
>>   }
>>   +static int _get_sensor_temp(struct mtk_thermal *mt, int id)
>> +{
>> +    u32 raw;
>> +    int temp;
>> +
>> +    const struct mtk_thermal_data *conf = mt->conf;
>> +
>> +    raw = readl(mt->thermal_base + conf->msr[id]);
>> +
>> +    if (mt->conf->version == MTK_THERMAL_V1)
>> +        temp = raw_to_mcelsius_v1(mt, id, raw);
>> +    else
>> +        temp = raw_to_mcelsius_v2(mt, id, raw);
>
> Can you set a callback at init time instead of checking the version at
> each get_sensor_temp() ?
>
Good idea, will add it in v4.
>> +    /*
>> +     * The first read of a sensor often contains very high bogus
>> +     * temperature value. Filter these out so that the system does
>> +     * not immediately shut down.
>> +     */
>> +
>> +    if (temp > 200000)
>> +        return -EAGAIN;
>> +    else
>> +        return temp;
>> +}
>> +
>>   /**
>>    * mtk_thermal_bank_temperature - get the temperature of a bank
>>    * @bank:    The bank
>> @@ -721,26 +752,9 @@ static int mtk_thermal_bank_temperature(struct
>> mtk_thermal_bank *bank)
>>       struct mtk_thermal *mt = bank->mt;
>>       const struct mtk_thermal_data *conf = mt->conf;
>>       int i, temp = INT_MIN, max = INT_MIN;
>> -    u32 raw;
>>         for (i = 0; i < conf->bank_data[bank->id].num_sensors; i++) {
>> -        raw = readl(mt->thermal_base + conf->msr[i]);
>> -
>> -        if (mt->conf->version == MTK_THERMAL_V1) {
>> -            temp = raw_to_mcelsius_v1(
>> -                mt, conf->bank_data[bank->id].sensors[i], raw);
>> -        } else {
>> -            temp = raw_to_mcelsius_v2(
>> -                mt, conf->bank_data[bank->id].sensors[i], raw);
>> -        }
>> -
>> -        /*
>> -         * The first read of a sensor often contains very high bogus
>> -         * temperature value. Filter these out so that the system does
>> -         * not immediately shut down.
>> -         */
>> -        if (temp > 200000)
>> -            temp = 0;
>> +        temp = _get_sensor_temp(mt, i);
>>             if (temp > max)
>>               max = temp;
>> @@ -751,7 +765,8 @@ static int mtk_thermal_bank_temperature(struct
>> mtk_thermal_bank *bank)
>>     static int mtk_read_temp(void *data, int *temperature)
>>   {
>> -    struct mtk_thermal *mt = data;
>> +    struct mtk_thermal_zone *tz = data;
>> +    struct mtk_thermal *mt = tz->mt;
>>       int i;
>>       int tempmax = INT_MIN;
>>   @@ -770,10 +785,28 @@ static int mtk_read_temp(void *data, int
>> *temperature)
>>       return 0;
>>   }
>>   +static int mtk_read_sensor_temp(void *data, int *temperature)
>> +{
>> +    struct mtk_thermal_zone *tz = data;
>> +    struct mtk_thermal *mt = tz->mt;
>> +    int id = tz->id - 1;
>> +
>> +    if (id < 0)
>> +        return -EACCES;
>> +
>> +    *temperature = _get_sensor_temp(mt, id);
>> +
>> +    return 0;
>> +}
>> +
>>   static const struct thermal_zone_of_device_ops mtk_thermal_ops = {
>>       .get_temp = mtk_read_temp,
>>   };
>>
>> +static const struct thermal_zone_of_device_ops
>> mtk_thermal_sensor_ops = {
>> +    .get_temp = mtk_read_sensor_temp,
>> +};
>
> Please respin against linux-next, the thermal_zone_of_device_ops
> structure does no longer exist. The conversion is trivial, here is a
> example:
>
> https://lore.kernel.org/all/[email protected]/
>
>
Sure, will do.
>
>>   static void mtk_thermal_init_bank(struct mtk_thermal *mt, int num,
>>                     u32 apmixed_phys_base, u32 auxadc_phys_base,
>>                     int ctrl_id)
>> @@ -1072,6 +1105,7 @@ static int mtk_thermal_probe(struct
>> platform_device *pdev)
>>       u64 auxadc_phys_base, apmixed_phys_base;
>>       struct thermal_zone_device *tzdev;
>>       void __iomem *apmixed_base, *auxadc_base;
>> +    struct mtk_thermal_zone *tz;
>>         mt = devm_kzalloc(&pdev->dev, sizeof(*mt), GFP_KERNEL);
>>       if (!mt)
>> @@ -1161,11 +1195,29 @@ static int mtk_thermal_probe(struct
>> platform_device *pdev)
>>         platform_set_drvdata(pdev, mt);
>>   -    tzdev = devm_thermal_zone_of_sensor_register(&pdev->dev, 0, mt,
>> -                             &mtk_thermal_ops);
>> -    if (IS_ERR(tzdev)) {
>> -        ret = PTR_ERR(tzdev);
>> -        goto err_disable_clk_peri_therm;
>> +    for (i = 0; i < mt->conf->num_sensors + 1; i++) {
>> +        tz = devm_kmalloc(&pdev->dev, sizeof(*tz), GFP_KERNEL);
>> +        if (!tz)
>> +            return -ENOMEM;
>> +
>> +        tz->mt = mt;
>> +        tz->id = i;
>> +
>> +        tzdev = devm_thermal_zone_of_sensor_register(&pdev->dev, i,
>> tz, (i == 0) ?
>> +                                 &mtk_thermal_ops :
>> +                                 &mtk_thermal_sensor_ops);
>> +
>> +        if (IS_ERR(tzdev)) {
>> +            if (PTR_ERR(tzdev) == -ENODEV) {
>> +                dev_warn(&pdev->dev,
>> +                     "sensor %d not registered in thermal zone in
>> dt\n", i);
>> +                continue;
>> +            }
>> +            if (PTR_ERR(tzdev) == -EACCES) {
>> +                ret = PTR_ERR(tzdev);
>> +                goto err_disable_clk_peri_therm;
>> +            }
>> +        }
>>       }
>>         ret = devm_thermal_add_hwmon_sysfs(tzdev);
>
>
Regards,

Amjad