Re: [PATCH 3/5] thermal: qcom: Add support for MBG Temp monitor

[Satya Priya Kakitapalli <quic_skakitap@quicinc.com>]

On 7/13/2024 5:29 PM, Jonathan Cameron wrote:
> On Fri, 12 Jul 2024 18:13:30 +0530
> Satya Priya Kakitapalli <quic_skakitap@quicinc.com> wrote:
>
>> Add driver for the MBG thermal monitoring device. It monitors
>> the die temperature, and when there is a level 1 upper threshold
>> violation, it receives an interrupt over spmi. The driver reads
>> the fault status register and notifies thermal accordingly.
>>
>> Signed-off-by: Satya Priya Kakitapalli <quic_skakitap@quicinc.com>
> Hi Satya,
>
> Some comments inline.


Thanks for review, I'll address the comments in my V2 patches.


>
> Jonathan
>
>> diff --git a/drivers/thermal/qcom/qcom-spmi-mbg-tm.c b/drivers/thermal/qcom/qcom-spmi-mbg-tm.c
>> new file mode 100644
>> index 000000000000..70964ea5a48d
>> --- /dev/null
>> +++ b/drivers/thermal/qcom/qcom-spmi-mbg-tm.c
>> @@ -0,0 +1,269 @@
>> +// SPDX-License-Identifier: GPL-2.0-only
>> +//Copyright (c) 2024, Qualcomm Innovation Center, Inc. All rights reserved.
>> +
>> +#include <linux/interrupt.h>
>> +#include <linux/irq.h>
>> +#include <linux/module.h>
>> +#include <linux/of.h>
>> +#include <linux/of_device.h>
>> +#include <linux/platform_device.h>
>> +#include <linux/regmap.h>
>> +#include <linux/thermal.h>
>> +#include <linux/iio/consumer.h>
>> +
>> +#include "../thermal_core.h"
>> +
>> +#define MBG_TEMP_MON_MM_MON2_FAULT_STATUS	0x50
>> +
>> +#define MON_FAULT_STATUS_MASK			GENMASK(7, 6)
>> +#define MON_FAULT_STATUS_SHIFT			6
> Use FIELD_GET() and FIELD_PREP() then no need to have shift defined.
>
>> +#define MON2_LVL1_ERR				0x1
>> +
>> +#define MON2_LVL1_UP_THRESH			0x59
>> +
>> +#define MBG_TEMP_MON_MM_MON2_MISC_CFG		0x5f
>> +#define UP_THRESH_EN				BIT(1)
>> +
>> +#define STEP_MV					8
>> +#define MBG_DEFAULT_TEMP_MV			600
>> +#define MBG_TEMP_CONSTANT			1000
>> +#define MIN_TRIP_TEMP				25000
>> +#define MAX_SUPPORTED_TEMP			160000
> these are all device specific but don't sound it. I'd prefix them with MBG_TEMP
>
>
>> +
>> +static int mbg_tm_set_trip_temp(struct thermal_zone_device *tz, int low_temp,
>> +						int temp)
>> +{
>> +	struct mbg_tm_chip *chip = thermal_zone_device_priv(tz);
>> +	int ret = 0, vtemp = 0;
>> +
>> +	mutex_lock(&chip->lock);
> 	guard(mutex)(&chip->lock);
>
>
> As then you don't need to carefully unlock as it will be done at exit of scope.
>
>> +
>> +	/* The HW has a limitation that the trip set must be above 25C */
>> +	if (temp > MIN_TRIP_TEMP && temp < INT_MAX) {
> 		int vtemp;
>
> reduce the scope and it becomes clear no need to init.
>
>> +		mbg_tm_reg_update(chip, MBG_TEMP_MON_MM_MON2_MISC_CFG,
>> +					 UP_THRESH_EN, UP_THRESH_EN);
>> +		vtemp = temp_to_vtemp(temp);
>> +		ret = mbg_tm_write(chip, MON2_LVL1_UP_THRESH, vtemp);
>> +		if (ret < 0) {
>> +			mutex_unlock(&chip->lock);
>> +			return ret;
>> +		}
>> +	} else {
>> +		dev_dbg(chip->dev, "Setting %d failed, set trip between 25C and INT_MAX\n", temp);
>> +		mbg_tm_reg_update(chip, MBG_TEMP_MON_MM_MON2_MISC_CFG,
>> +					UP_THRESH_EN, 0);
>> +	}
>> +
>> +	mutex_unlock(&chip->lock);
>> +
>> +	/*
>> +	 * Configure the last_temp one degree higher, to ensure the
>> +	 * violated temp is returned to thermal framework when it reads
>> +	 * temperature for the first time after the violation happens.
>> +	 * This is needed to account for the inaccuracy in the conversion
>> +	 * formula used which leads to the thermal framework setting back
>> +	 * the same thresholds in case the temperature it reads does not
>> +	 * show violation.
>> +	 */
>> +	chip->last_temp = temp + MBG_TEMP_CONSTANT;
>> +
>> +	return ret;
>> +}
>> +
>> +static const struct thermal_zone_device_ops mbg_tm_ops = {
>> +	.get_temp = mbg_tm_get_temp,
>> +	.set_trips = mbg_tm_set_trip_temp,
>> +};
>> +
>> +static irqreturn_t mbg_tm_isr(int irq, void *data)
>> +{
>> +	struct mbg_tm_chip *chip = data;
>> +	int ret;
>> +	int val = 0;
>> +
>> +	mutex_lock(&chip->lock);
>> +
>> +	ret = mbg_tm_read(chip, MBG_TEMP_MON_MM_MON2_FAULT_STATUS, &val);
>> +
>> +	mutex_unlock(&chip->lock);
> 	scoped_guard(mutex, &chip->lock) {
> 		ret = mbg...
> 		if (ret < 0)
> 			return IRQ_HANDLED;
> 	}
> avoids need to unlock before return and to me at least gives slightly more
> readable code.
>
>> +
>> +	if (ret < 0)
>> +		return IRQ_HANDLED;
>> +
>> +	val &= MON_FAULT_STATUS_MASK;
>> +	if ((val >> MON_FAULT_STATUS_SHIFT) & MON2_LVL1_ERR) {
> as above use FIELD_GET() with the mask - it is more readable and
> does the mask and shift in one call.
>
>
>
>> +		chip->last_temp_set = true;
>> +		thermal_zone_device_update(chip->tz_dev,
>> +					THERMAL_TRIP_VIOLATED);
> I think the above is 80 chars on one line so no need to wrap.
>
>> +		dev_dbg(chip->dev, "Notifying Thermal, fault status=%d\n", val);
>> +	} else {
>> +		dev_dbg(chip->dev, "Lvl 1 upper threshold not violated, ignoring interrupt\n");
>> +	}
>> +
>> +	return IRQ_HANDLED;
>> +}
>> +
>> +static int mbg_tm_probe(struct platform_device *pdev)
>> +{
>> +	struct mbg_tm_chip *chip;
>> +	struct device_node *node = pdev->dev.of_node;
>> +	u32 res;
>> +	int ret = 0;
> It's always set, so don't intiialize here.
>
>> +
>> +	chip = devm_kzalloc(&pdev->dev, sizeof(*chip), GFP_KERNEL);
>> +	if (!chip)
>> +		return -ENOMEM;
>> +
>> +	chip->dev = &pdev->dev;
>> +
>> +	mutex_init(&chip->lock);
>> +
>> +	chip->map = dev_get_regmap(pdev->dev.parent, NULL);
>> +	if (!chip->map)
>> +		return -ENXIO;
>> +
>> +	ret = of_property_read_u32(node, "reg", &res);
> I'm not sure on thermal subsystem opinion on this but I'd use
> the property.h generic firmware property reading stuff rather than of
> specific.
> device_property_read_u32()
>
>> +	if (ret < 0)
>> +		return ret;
>> +
>> +	chip->base = res;
>> +
>> +	chip->irq = platform_get_irq(pdev, 0);
>> +	if (chip->irq < 0)
>> +		return chip->irq;
>> +
>> +	chip->adc = devm_iio_channel_get(&pdev->dev, "thermal");
>> +	if (IS_ERR(chip->adc))
>> +		return PTR_ERR(chip->adc);
>> +
>> +	chip->tz_dev = devm_thermal_of_zone_register(&pdev->dev,
>> +				 0, chip, &mbg_tm_ops);
>> +	if (IS_ERR(chip->tz_dev)) {
>> +		dev_err(&pdev->dev, "failed to register sensor\n");
>> +		return PTR_ERR(chip->tz_dev);
> 		return dev_err_probe(...)
>
>> +	}
>> +
>> +	ret = devm_request_threaded_irq(&pdev->dev, chip->irq, NULL,
>> +			mbg_tm_isr, IRQF_ONESHOT, node->name, chip);
>> +
>> +	return ret;
> 	return devm_request...
>
>> +}