Re: [PATCH v6 4/4] thermal: mediatek: add another get_temp ops for thermal sensors

From: Daniel Lezcano
Date: Sat Oct 22 2022 - 13:03:29 EST


On 19/10/2022 16:17, 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 <michael.kao@xxxxxxxxxxxx>
Signed-off-by: Hsin-Yi Wang <hsinyi@xxxxxxxxxxxx>
Signed-off-by: Amjad Ouled-Ameur <aouledameur@xxxxxxxxxxxx>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@xxxxxxxxxxxxx>

Overall the series looks good to me, however there is a couple of things to fix. See below

---
drivers/thermal/mtk_thermal.c | 104 ++++++++++++++++++++++++++++++++----------
1 file changed, 79 insertions(+), 25 deletions(-)

diff --git a/drivers/thermal/mtk_thermal.c b/drivers/thermal/mtk_thermal.c
index 3a5df1440822..311ad611fdab 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;
+};

Do you really need to create a new structure for that ?

struct thermal_bank_cfg {
unsigned int num_sensors;
const int *sensors;
@@ -307,6 +312,8 @@ struct mtk_thermal {
const struct mtk_thermal_data *conf;
struct mtk_thermal_bank banks[MAX_NUM_ZONES];
+
+ int (*raw_to_mcelsius)(struct mtk_thermal *mt, int sensno, s32 raw);
};
/* MT8183 thermal sensor data */
@@ -709,6 +716,29 @@ 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]);
+
+ temp = mt->raw_to_mcelsius(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 +751,9 @@ 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;
@@ -749,9 +762,10 @@ static int mtk_thermal_bank_temperature(struct mtk_thermal_bank *bank)
return max;
}
-static int mtk_read_temp(struct thermal_zone_device *tz, int *temperature)
+static int mtk_read_temp(struct thermal_zone_device *tzdev, int *temperature)
{
- struct mtk_thermal *mt = tz->devdata;
+ struct mtk_thermal_zone *tz = tzdev->devdata;
+ struct mtk_thermal *mt = tz->mt;
int i;
int tempmax = INT_MIN;
@@ -770,10 +784,28 @@ static int mtk_read_temp(struct thermal_zone_device *tz, int *temperature)
return 0;
}
+static int mtk_read_sensor_temp(struct thermal_zone_device *tzdev, int *temperature)
+{
+ struct mtk_thermal_zone *tz = tzdev->devdata;
+ 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_device_ops mtk_thermal_ops = {
.get_temp = mtk_read_temp,
};
+static const struct thermal_zone_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 +1104,7 @@ 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)
@@ -1150,6 +1183,9 @@ static int mtk_thermal_probe(struct platform_device *pdev)
mtk_thermal_turn_on_buffer(mt, apmixed_base);
+ mt->raw_to_mcelsius = (mt->conf->version == MTK_THERMAL_V1) ?
+ raw_to_mcelsius_v1 : raw_to_mcelsius_v2;
+
if (mt->conf->version == MTK_THERMAL_V2) {
mtk_thermal_release_periodic_ts(mt, auxadc_base);
}
@@ -1161,11 +1197,29 @@ static int mtk_thermal_probe(struct platform_device *pdev)
platform_set_drvdata(pdev, mt);
- tzdev = devm_thermal_of_zone_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_of_zone_register(&pdev->dev, i, tz, (i == 0) ?
+ &mtk_thermal_ops :
+ &mtk_thermal_sensor_ops);

We want to prevent the aggregation from the different sensors within a driver. I know there is already a function doing that for the previous sensor version but that is something we don't want to continue.

Using mtk_thermal_ops tries to overcome this.

Also, the userspace needs to know to which device a thermal zone is related to. Here all the thermal zones have the same name so the userspace does not know if it is the GPU, the CPU, the chassis, the NPU, ...


+
+ if (IS_ERR(tzdev)) {
+ if (PTR_ERR(tzdev) == -ENODEV) {
+ dev_warn(&pdev->dev,
+ "sensor %d not registered in thermal zone in dt\n", i);

ENODEV is not considered an error, so the warning can be removed

+ continue;
+ }
+ if (PTR_ERR(tzdev) == -EACCES) {

When devm_thermal_of_zone_register() returns -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