Message-ID: <Z1hLlm75iGMogltc@lizhi-Precision-Tower-5810>
Date: Tue, 10 Dec 2024 09:09:26 -0500
From: Frank Li <Frank.li@....com>
To: Marco Felsch <m.felsch@...gutronix.de>
Cc: "Rafael J. Wysocki" <rafael@...nel.org>,
	Daniel Lezcano <daniel.lezcano@...aro.org>,
	Zhang Rui <rui.zhang@...el.com>, Lukasz Luba <lukasz.luba@....com>,
	Rob Herring <robh@...nel.org>,
	Krzysztof Kozlowski <krzk+dt@...nel.org>,
	Conor Dooley <conor+dt@...nel.org>, Shawn Guo <shawnguo@...nel.org>,
	Sascha Hauer <s.hauer@...gutronix.de>,
	Pengutronix Kernel Team <kernel@...gutronix.de>,
	Fabio Estevam <festevam@...il.com>,
	Pengfei Li <pengfei.li_1@....com>, devicetree@...r.kernel.org,
	Peng Fan <peng.fan@....com>, imx@...ts.linux.dev,
	linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org,
	linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH 2/2] thermal: imx91: Add support for i.MX91 thermal
 monitoring unit

On Tue, Dec 10, 2024 at 11:36:46AM +0100, Marco Felsch wrote:
> Hi,
>
> On 24-12-09, Frank Li wrote:
> > From: Pengfei Li <pengfei.li_1@....com>
> >
> > Introduce support for the i.MX91 thermal monitoring unit, which features a
> > single sensor for the CPU. The register layout differs from other chips,
> > necessitating the creation of a dedicated file for this.
> >
> > Signed-off-by: Pengfei Li <pengfei.li_1@....com>
> > Signed-off-by: Peng Fan <peng.fan@....com>
> > Signed-off-by: Frank Li <Frank.Li@....com>
> > ---
> >  drivers/thermal/Kconfig         |  10 ++
> >  drivers/thermal/Makefile        |   1 +
> >  drivers/thermal/imx91_thermal.c | 281 ++++++++++++++++++++++++++++++++++++++++
> >  3 files changed, 292 insertions(+)
> >
> > diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
> > index d3f9686e26e71..da403ed86aeb1 100644
> > --- a/drivers/thermal/Kconfig
> > +++ b/drivers/thermal/Kconfig
> > @@ -296,6 +296,16 @@ config IMX8MM_THERMAL
> >  	  cpufreq is used as the cooling device to throttle CPUs when the passive
> >  	  trip is crossed.
> >
> > +config IMX91_THERMAL
> > +	tristate "Temperature sensor driver for NXP i.MX91 SoC"
> > +	depends on ARCH_MXC || COMPILE_TEST
> > +	depends on OF
> > +	help
> > +	  Support for Temperature sensor found on NXP i.MX91 SoC.
> > +	  It supports one critical trip point and one passive trip point. The
> > +	  cpufreq is used as the cooling device to throttle CPUs when the passive
> > +	  trip is crossed.
> > +
> >  config K3_THERMAL
> >  	tristate "Texas Instruments K3 thermal support"
> >  	depends on ARCH_K3 || COMPILE_TEST
> > diff --git a/drivers/thermal/Makefile b/drivers/thermal/Makefile
> > index 9abf43a74f2bb..08da241e6a598 100644
> > --- a/drivers/thermal/Makefile
> > +++ b/drivers/thermal/Makefile
> > @@ -50,6 +50,7 @@ obj-$(CONFIG_ARMADA_THERMAL)	+= armada_thermal.o
> >  obj-$(CONFIG_IMX_THERMAL)	+= imx_thermal.o
> >  obj-$(CONFIG_IMX_SC_THERMAL)	+= imx_sc_thermal.o
> >  obj-$(CONFIG_IMX8MM_THERMAL)	+= imx8mm_thermal.o
> > +obj-$(CONFIG_IMX91_THERMAL)	+= imx91_thermal.o
> >  obj-$(CONFIG_MAX77620_THERMAL)	+= max77620_thermal.o
> >  obj-$(CONFIG_QORIQ_THERMAL)	+= qoriq_thermal.o
> >  obj-$(CONFIG_DA9062_THERMAL)	+= da9062-thermal.o
> > diff --git a/drivers/thermal/imx91_thermal.c b/drivers/thermal/imx91_thermal.c
> > new file mode 100644
> > index 0000000000000..19c4dafe9c7a9
> > --- /dev/null
> > +++ b/drivers/thermal/imx91_thermal.c
> > @@ -0,0 +1,281 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * Copyright 2024 NXP.
> > + */
> > +
> > +#include <linux/bitfield.h>
> > +#include <linux/clk.h>
> > +#include <linux/err.h>
> > +#include <linux/iopoll.h>
> > +#include <linux/nvmem-consumer.h>
> > +#include <linux/module.h>
> > +#include <linux/of.h>
> > +#include <linux/of_device.h>
> > +#include <linux/platform_device.h>
> > +#include <linux/thermal.h>
> > +
> > +#define CTRL0			0x0
> > +
> > +#define STAT0			0x10
> > +#define STAT0_DRDY0_IF_MASK	BIT(16)
> > +
> > +#define DATA0			0x20
> > +#define DATA_INT_MASK		GENMASK(14, 7)
> > +#define DATA_FRAC_MASK		GENMASK(6, 0)
> > +
> > +#define THR_CTRL01		0x30
> > +#define THR_CTRL23		0x40
> > +
> > +#define CTRL1			0x200
> > +#define CTRL1_SET		0x204
> > +#define CTRL1_CLR		0x208
> > +#define CTRL1_EN		BIT(31)
> > +#define CTRL1_START		BIT(30)
> > +#define CTRL1_STOP		BIT(29)
> > +#define CTRL1_RES_MASK		GENMASK(19, 18)
> > +#define CTRL1_MEAS_MODE_MASK	GENMASK(25, 24)
> > +
> > +#define PERIOD_CTRL		0x270
> > +#define MEAS_FREQ_MASK		GENMASK(23, 0)
> > +
> > +#define REF_DIV			0x280
> > +#define DIV_EN			BIT(31)
> > +#define DIV_MASK		GENMASK(23, 16)
> > +
> > +#define PUD_ST_CTRL		0x2B0
> > +#define PUDL_MASK		GENMASK(23, 16)
> > +
> > +#define TRIM1			0x2E0
> > +#define TRIM2			0x2F0
> > +
> > +#define TMU_TEMP_LOW_LIMIT	-40000
> > +#define TMU_TEMP_HIGH_LIMIT	125000
> > +
> > +#define DEFAULT_TRIM1_CONFIG 0xB561BC2DU
> > +#define DEFAULT_TRIM2_CONFIG 0x65D4U
>
> please align the defines and use lower-case for the hex values.
>
> > +struct tmu_sensor {
> > +	struct imx91_tmu *priv;
> > +	struct thermal_zone_device *tzd;
> > +};
>
> This can be part of imx91_tmu right?
>
> > +struct imx91_tmu {
> > +	void __iomem *base;
> > +	struct clk *clk;
> > +	struct device *dev;
> > +	struct tmu_sensor sensors;
> > +};
> > +
> > +static void imx91_tmu_start(struct imx91_tmu *tmu, bool start)
> > +{
> > +	if (start)
> > +		writel_relaxed(CTRL1_START, tmu->base + CTRL1_SET);
> > +	else
> > +		writel_relaxed(CTRL1_STOP, tmu->base + CTRL1_SET);
>
> Nit:
> 	unsigned int val = start ? CTRL1_START : CTRL1_STOP;
>
> 	writel_relaxed(val, tmu->base + CTRL1_SET);
>
> Aside the nit, why do you use *_relaxed() accessors in you whole code?
> Is the order not strictly required?
>
> > +}
> > +
> > +static void imx91_tmu_enable(struct imx91_tmu *tmu, bool enable)
> > +{
> > +	if (enable)
> > +		writel_relaxed(CTRL1_EN, tmu->base + CTRL1_SET);
> > +	else
> > +		writel_relaxed(CTRL1_EN, tmu->base + CTRL1_CLR);
>
> Same here:
>
> 	unsigned int reg = enable ? CTRL1_SET : CTRL1_CLR;
>
> 	writel_relaxed(CTRL1_EN, tmu->base + reg);
> > +}
> > +
> > +static int imx91_tmu_get_temp(struct thermal_zone_device *tz, int *temp)
> > +{
> > +	struct tmu_sensor *sensor = thermal_zone_device_priv(tz);
> > +	struct imx91_tmu *tmu = sensor->priv;
>
> With tmu_sensor merged into imx91_tmu you don't need this reverse
> pointer logic.
>
> > +	int val;
> > +	int ret;
> > +
> > +	ret = readl_relaxed_poll_timeout(tmu->base + STAT0, val,
> > +					 val & STAT0_DRDY0_IF_MASK, 1000,
> > +					 40000);
> > +	if (ret)
> > +		return -EAGAIN;
> > +
> > +	val = readl_relaxed(tmu->base + DATA0) & 0xffffU;
> 						   ^
> 				Please use a MASK define for it, also no
> 				need to use the U suffix.
>
> Also according the TRM 58.3.4 this doesn't seem right. According the TRM
> this is a s16 (two's-complement) value. You won't get negative values if
> store it within a int.
>
> > +	*temp = val * 1000L / 64L;
> 			  ^     ^
> You don't need to specify the type here explicit since all params would
> be converted to int and no overflow happens.
>
> > +	if (*temp < TMU_TEMP_LOW_LIMIT || *temp > TMU_TEMP_HIGH_LIMIT)
> > +		return -EAGAIN;
> > +
> > +	return 0;
> > +}
> > +
> > +static struct thermal_zone_device_ops tmu_tz_ops = {
> > +	.get_temp = imx91_tmu_get_temp,
> > +};
> > +
> > +static int imx91_init_from_nvmem_cells(struct imx91_tmu *tmu)
> > +{
> > +	struct device *dev = tmu->dev;
> > +	int ret;
> > +	u32 trim1, trim2;
>
> Please use the reverse christmas tree style.
>
> > +	ret = nvmem_cell_read_u32(dev, "trim1", &trim1);
> > +	if (ret)
> > +		return ret;
> > +
> > +	ret = nvmem_cell_read_u32(dev, "trim2", &trim2);
> > +	if (ret)
> > +		return ret;
> > +
> > +	if (trim1 == 0 || trim2 == 0)
> > +		return -EINVAL;
> > +
> > +	writel_relaxed(trim1, tmu->base + TRIM1);
> > +	writel_relaxed(trim2, tmu->base + TRIM2);
> > +
> > +	return 0;
> > +}
> > +
> > +static int imx91_tmu_probe(struct platform_device *pdev)
> > +{
> > +	struct imx91_tmu *tmu;
> > +	unsigned long rate;
> > +	u32 div;
> > +	int ret;
> > +	int i = 0;
>
> Same here.
>
> > +	tmu = devm_kzalloc(&pdev->dev, sizeof(struct imx91_tmu), GFP_KERNEL);
> > +	if (!tmu)
> > +		return -ENOMEM;
> > +
> > +	tmu->dev = &pdev->dev;
> > +
> > +	tmu->base = devm_platform_ioremap_resource(pdev, 0);
> > +	if (IS_ERR(tmu->base))
> > +		return PTR_ERR(tmu->base);
> > +
> > +	tmu->clk = devm_clk_get_prepared(&pdev->dev, NULL);
> > +	if (IS_ERR(tmu->clk))
> > +		return dev_err_probe(&pdev->dev, PTR_ERR(tmu->clk), "failed to get tmu clock\n");
> > +
> > +	tmu->sensors.priv = tmu;
>
> As said, no need for the sensors container.
>
> > +	tmu->sensors.tzd = devm_thermal_of_zone_register(&pdev->dev, i, &tmu->sensors, &tmu_tz_ops);
> > +	if (IS_ERR(tmu->sensors.tzd))
> > +		return dev_err_probe(&pdev->dev, PTR_ERR(tmu->sensors.tzd),
> > +				     "failed to register thermal zone sensor\n");
> > +
> > +	platform_set_drvdata(pdev, tmu);
> > +
> > +	/* disable the monitor during initialization */
> > +	imx91_tmu_enable(tmu, false);
> > +	imx91_tmu_start(tmu, false);
>
> Make use of pm_runtime?

thermal is always on after probe. continue measure tempature. So pm_runtime
is not suitable for this case.

>
> > +	ret = imx91_init_from_nvmem_cells(tmu);
> > +	if (ret) {
> > +		writel_relaxed(DEFAULT_TRIM1_CONFIG, tmu->base + TRIM1);
> > +		writel_relaxed(DEFAULT_TRIM2_CONFIG, tmu->base + TRIM2);
> > +	}
> > +
> > +	/* The typical conv clk is 4MHz, the output freq is 'rate / (div + 1)' */
> > +	rate = clk_get_rate(tmu->clk);
> > +	div = (rate / 4000000) - 1;
>
> Would be nice to validate the div value before passing to the HW and if
> the target rate of 4MHz can't be reached by the div you you should
> return -EINVAL.
>
> > +
> > +	/* Set divider value and enable divider */
> > +	writel_relaxed(DIV_EN | FIELD_PREP(DIV_MASK, div), tmu->base + REF_DIV);
> > +
> > +	/* Set max power up delay: 'Tpud(ms) = 0xFF * 1000 / 4000000' */
> > +	writel_relaxed(FIELD_PREP(PUDL_MASK, 100U), tmu->base + PUD_ST_CTRL);
> > +
> > +	/*
> > +	 * Set resolution mode
> > +	 * 00b - Conversion time = 0.59325 ms
> > +	 * 01b - Conversion time = 1.10525 ms
> > +	 * 10b - Conversion time = 2.12925 ms
> > +	 * 11b - Conversion time = 4.17725 ms
> > +	 */
> > +	writel_relaxed(FIELD_PREP(CTRL1_RES_MASK, 0x3), tmu->base + CTRL1_CLR);
> > +	writel_relaxed(FIELD_PREP(CTRL1_RES_MASK, 0x1), tmu->base + CTRL1_SET);
> > +
> > +	/*
> > +	 * Set measure mode
> > +	 * 00b - Single oneshot measurement
> > +	 * 01b - Continuous measurement
> > +	 * 10b - Periodic oneshot measurement
> > +	 */
>
> For the resolution it's fine to have the values directly coded without a
> define, but here we can definitly use a define and drop the comment.
>
> > +	writel_relaxed(FIELD_PREP(CTRL1_MEAS_MODE_MASK, 0x3), tmu->base + CTRL1_CLR);
> > +	writel_relaxed(FIELD_PREP(CTRL1_MEAS_MODE_MASK, 0x1), tmu->base + CTRL1_SET);
>
> Why do we set it to periodic mode instead of the single-shot? At the
> moment the device doesn't have IRQ support, and so there is no need to
> run the measurements in background.

It is "continous measurement".  first clean MODE_MASK, then then 01.

_CLR means clean bits according to mask
_SET means set bits according to mask.

Frank

>
> > +
> > +	/*
> > +	 * Set Periodic Measurement Frequency to 25Hz:
> > +	 * tMEAS_FREQ = tCONV_CLK * PERIOD_CTRL[MEAS_FREQ]. ->
> > +	 * PERIOD_CTRL(MEAS_FREQ) = (1000 / 25) / (1000 / 4000000);
> > +	 * Where tMEAS_FREQ = Measurement period and tCONV_CLK = 1/fCONV_CLK.
> > +	 * This field should have value greater than count corresponds
> > +	 * to time greater than summation of conversion time, power up
> > +	 * delay, and six times of conversion clock time.
> > +	 * tMEAS_FREQ > (tCONV + tPUD + 6 * tCONV_CLK).
> > +	 * tCONV is conversion time determined by CTRL1[RESOLUTION].
> > +	 */
> > +	writel_relaxed(FIELD_PREP(MEAS_FREQ_MASK, 0x27100), tmu->base + PERIOD_CTRL);
>
> With the single-shot measurements we could remove this part and..
>
> > +
> > +	/* enable the monitor */
> > +	imx91_tmu_enable(tmu, true);
> > +	imx91_tmu_start(tmu, true);
>
> this part as well.
>
> Regards,
>   Marco
>
> > +
> > +	return 0;
> > +}
> > +
> > +static void imx91_tmu_remove(struct platform_device *pdev)
> > +{
> > +	struct imx91_tmu *tmu = platform_get_drvdata(pdev);
> > +
> > +	/* disable tmu */
> > +	imx91_tmu_start(tmu, false);
> > +	imx91_tmu_enable(tmu, false);
> > +}
> > +
> > +static int __maybe_unused imx91_tmu_suspend(struct device *dev)
> > +{
> > +	struct imx91_tmu *tmu = dev_get_drvdata(dev);
> > +
> > +	/* disable tmu */
> > +	imx91_tmu_start(tmu, false);
> > +	imx91_tmu_enable(tmu, false);
> > +
> > +	clk_disable_unprepare(tmu->clk);
> > +
> > +	return 0;
> > +}
> > +
> > +static int __maybe_unused imx91_tmu_resume(struct device *dev)
> > +{
> > +	struct imx91_tmu *tmu = dev_get_drvdata(dev);
> > +	int ret;
> > +
> > +	ret = clk_prepare_enable(tmu->clk);
> > +	if (ret)
> > +		return ret;
> > +
> > +	imx91_tmu_enable(tmu, true);
> > +	imx91_tmu_start(tmu, true);
> > +
> > +	return 0;
> > +}
> > +
> > +static SIMPLE_DEV_PM_OPS(imx91_tmu_pm_ops,
> > +			 imx91_tmu_suspend, imx91_tmu_resume);
> > +
> > +static const struct of_device_id imx91_tmu_table[] = {
> > +	{ .compatible = "fsl,imx91-tmu", },
> > +	{ },
> > +};
> > +MODULE_DEVICE_TABLE(of, imx91_tmu_table);
> > +
> > +static struct platform_driver imx91_tmu = {
> > +	.driver = {
> > +		.name	= "i.MX91_thermal",
> > +		.pm	= pm_ptr(&imx91_tmu_pm_ops),
> > +		.of_match_table = imx91_tmu_table,
> > +	},
> > +	.probe = imx91_tmu_probe,
> > +	.remove = imx91_tmu_remove,
> > +};
> > +module_platform_driver(imx91_tmu);
> > +
> > +MODULE_AUTHOR("Peng Fan <peng.fan@....com>");
> > +MODULE_DESCRIPTION("i.MX91 Thermal Monitor Unit driver");
> > +MODULE_LICENSE("GPL");
> >
> > --
> > 2.34.1
> >
> >
> >

Powered by blists - more mailing lists