[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20251127113800-GYA1795104@gentoo.org>
Date: Thu, 27 Nov 2025 19:38:00 +0800
From: Yixun Lan <dlan@...too.org>
To: Shuwei Wu <shuweiwoo@....com>
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>,
Philipp Zabel <p.zabel@...gutronix.de>,
Paul Walmsley <pjw@...nel.org>, Palmer Dabbelt <palmer@...belt.com>,
Albert Ou <aou@...s.berkeley.edu>, Alexandre Ghiti <alex@...ti.fr>,
linux-pm@...r.kernel.org, devicetree@...r.kernel.org,
linux-riscv@...ts.infradead.org, spacemit@...ts.linux.dev,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/3] thermal: K1: Add driver for K1 SoC thermal sensor
Hi Shuwei,
On 02:44 Thu 27 Nov , Shuwei Wu wrote:
> The thermal sensor unit (TSU) on K1 supports monitoring five temperature
> zones. The driver registers these sensors with the thermal framework
> and supports standard operations:
> - Reading temperature (millidegree Celsius)
> - Setting high/low thresholds for interrupts
>
> Signed-off-by: Shuwei Wu <shuweiwoo@....com>
> ---
> drivers/thermal/Kconfig | 14 ++
> drivers/thermal/Makefile | 1 +
> drivers/thermal/k1_thermal.c | 307 +++++++++++++++++++++++++++++++++++++++++++
> 3 files changed, 322 insertions(+)
>
> diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
> index a09c188b9ad11377afe232d89c60504eb7000417..76095d2888980718b39470c09731092a21f7159b 100644
> --- a/drivers/thermal/Kconfig
> +++ b/drivers/thermal/Kconfig
> @@ -495,6 +495,20 @@ config SPRD_THERMAL
> Support for the Spreadtrum thermal sensor driver in the Linux thermal
> framework.
>
> +config K1_THERMAL
please name it as SPACEMIT_K1_THERMAL, having K1 only is too short
> + tristate "SpacemiT K1 thermal sensor driver"
> + depends on ARCH_SPACEMIT || COMPILE_TEST
> + help
> + This driver provides support for the thermal sensor unit (TSU)
> + integrated in the SpacemiT K1 SoC.
> +
> + The TSU monitors temperatures for five thermal zones: soc, package,
> + gpu, cluster0, and cluster1. It supports reporting temperature
> + values and handling high/low threshold interrupts.
> +
> + Say Y here if you want to enable thermal monitoring on SpacemiT K1.
> + If compiled as a module, it will be called k1_thermal.
> +
> config KHADAS_MCU_FAN_THERMAL
> tristate "Khadas MCU controller FAN cooling support"
> depends on OF
> diff --git a/drivers/thermal/Makefile b/drivers/thermal/Makefile
> index d7718978db245faffba98ff95a07c7bcbc776fd2..bf28ffe7a39f916acd608ea6d592c82049b0be17 100644
> --- a/drivers/thermal/Makefile
> +++ b/drivers/thermal/Makefile
> @@ -65,6 +65,7 @@ obj-$(CONFIG_GENERIC_ADC_THERMAL) += thermal-generic-adc.o
> obj-$(CONFIG_UNIPHIER_THERMAL) += uniphier_thermal.o
> obj-$(CONFIG_AMLOGIC_THERMAL) += amlogic_thermal.o
> obj-$(CONFIG_SPRD_THERMAL) += sprd_thermal.o
> +obj-$(CONFIG_K1_THERMAL) += k1_thermal.o
same reason to adjust filename
> obj-$(CONFIG_KHADAS_MCU_FAN_THERMAL) += khadas_mcu_fan.o
> obj-$(CONFIG_LOONGSON2_THERMAL) += loongson2_thermal.o
> obj-$(CONFIG_THERMAL_CORE_TESTING) += testing/
> diff --git a/drivers/thermal/k1_thermal.c b/drivers/thermal/k1_thermal.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..a0e9585cbc5a4e0f7c3a47debb3cfa8e82082d88
> --- /dev/null
> +++ b/drivers/thermal/k1_thermal.c
> @@ -0,0 +1,307 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Thermal sensor driver for SpacemiT K1 SoC
> + *
> + * Copyright (C) 2025 Shuwei Wu <shuweiwoo@....com>
> + */
> +#include <linux/clk.h>
> +#include <linux/err.h>
> +#include <linux/interrupt.h>
> +#include <linux/io.h>
> +#include <linux/kernel.h>
> +#include <linux/module.h>
> +#include <linux/of.h>
> +#include <linux/platform_device.h>
> +#include <linux/reset.h>
> +#include <linux/thermal.h>
> +
> +#include "thermal_hwmon.h"
> +
> +#define MAX_SENSOR_NUMBER 5
> +#define TEMPERATURE_OFFSET 278
this seems a magic number? could you improve the name a bit,
or better put a comment to have a explanation
> +
> +#define K1_TSU_INT_EN 0x14
> +#define K1_TSU_INT_CLR 0x10
> +#define K1_TSU_INT_STA 0x18
> +
> +#define K1_TSU_INT_EN_MASK BIT(0)
> +#define K1_TSU_INT_MASK(x) (GENMASK(2, 1) << ((x) * 2))
> +
> +#define K1_TSU_EN 0x8
> +#define K1_TSU_EN_MASK(x) BIT(x)
> +
> +#define K1_TSU_DATA_BASE 0x20
..
> +#define K1_TSU_DATA(x) (K1_TSU_DATA_BASE + ((x) / 2) * 4)
so this is a register after taking a look at the code..
can you add a 'REG' explicitly? same reason for others
> +#define K1_TSU_DATA_MASK(x) (((x) % 2) ? GENMASK(31, 16) : GENMASK(15, 0))
> +#define K1_TSU_DATA_SHIFT(x) (((x) % 2) ? 16 : 0)
There is only one call, can you just fold it there? instead of using this magic
> +
> +#define K1_TSU_THRSH_BASE 0x40
> +#define K1_TSU_THRSH(x) (K1_TSU_THRSH_BASE + ((x) * 4))
> +#define K1_TSU_THRSH_HIGH_MASK GENMASK(31, 16)
> +#define K1_TSU_THRSH_LOW_MASK GENMASK(15, 0)
> +#define K1_TSU_THRSH_HIGH_SHIFT 16
> +#define K1_TSU_THRSH_LOW_SHIFT 0
> +
> +#define K1_TSU_TIME 0x0C
> +#define K1_TSU_TIME_MASK GENMASK(23, 0)
> +#define K1_TSU_TIME_FILTER_PERIOD GENMASK(21, 20)
> +#define K1_TSU_TIME_ADC_CNT_RST GENMASK(7, 4)
> +#define K1_TSU_TIME_WAIT_REF_CNT GENMASK(3, 0)
> +
> +#define K1_TSU_PCTRL 0x00
> +#define K1_TSU_PCTRL_RAW_SEL BIT(7)
> +#define K1_TSU_PCTRL_TEMP_MODE BIT(3)
> +#define K1_TSU_PCTRL_ENABLE BIT(0)
> +
> +#define K1_TSU_PCTRL_SW_CTRL GENMASK(21, 18)
> +#define K1_TSU_PCTRL_CTUNE GENMASK(11, 8)
> +#define K1_TSU_PCTRL_HW_AUTO_MODE BIT(23)
> +
> +#define K1_TSU_PCTRL2 0x04
> +#define K1_TSU_PCTRL2_CLK_SEL_MASK GENMASK(15, 14)
> +#define K1_TSU_PCTRL2_CLK_SEL_24M (0 << 14)
> +
> +struct k1_thermal_sensor {
> + struct k1_thermal_priv *priv;
> + struct thermal_zone_device *tzd;
> + int id;
> +};
> +
> +struct k1_thermal_priv {
> + void __iomem *base;
> + struct device *dev;
> + struct clk *clk;
> + struct clk *bus_clk;
> + struct reset_control *reset;
> + struct k1_thermal_sensor sensors[MAX_SENSOR_NUMBER];
> +};
> +
> +static int k1_init_sensors(struct platform_device *pdev)
> +{
> + struct k1_thermal_priv *priv = platform_get_drvdata(pdev);
> + unsigned int temp;
so is 'temp' short for temperature? if not, for intermediate register value,
please simply use 'val' for less confusion, sometimes people prefer 'tmp',
but I'd avoid here
> + int i;
> +
> + /* Disable all the interrupts */
> + writel(0xffffffff, priv->base + K1_TSU_INT_EN);
> +
> + /* Configure ADC sampling time and filter period */
> + temp = readl(priv->base + K1_TSU_TIME);
> + temp &= ~K1_TSU_TIME_MASK;
> + temp |= K1_TSU_TIME_FILTER_PERIOD |
> + K1_TSU_TIME_ADC_CNT_RST |
> + K1_TSU_TIME_WAIT_REF_CNT;
> + writel(temp, priv->base + K1_TSU_TIME);
> +
> + /*
> + * Enable all sensors' auto mode, enable dither control,
> + * consecutive mode, and power up sensor.
> + */
> + temp = readl(priv->base + K1_TSU_PCTRL);
> + temp |= K1_TSU_PCTRL_RAW_SEL |
> + K1_TSU_PCTRL_TEMP_MODE |
> + K1_TSU_PCTRL_HW_AUTO_MODE |
> + K1_TSU_PCTRL_ENABLE;
> + temp &= ~K1_TSU_PCTRL_SW_CTRL;
> + temp &= ~K1_TSU_PCTRL_CTUNE;
> + writel(temp, priv->base + K1_TSU_PCTRL);
> +
> + /* Select 24M clk for high speed mode */
> + temp = readl(priv->base + K1_TSU_PCTRL2);
> + temp &= ~K1_TSU_PCTRL2_CLK_SEL_MASK;
> + temp |= K1_TSU_PCTRL2_CLK_SEL_24M;
> + writel(temp, priv->base + K1_TSU_PCTRL2);
> +
> + /* Enable thermal interrupt */
> + temp = readl(priv->base + K1_TSU_INT_EN);
> + temp |= K1_TSU_INT_EN_MASK;
> + writel(temp, priv->base + K1_TSU_INT_EN);
> +
> + /* Enable each sensor */
> + for (i = 0; i < MAX_SENSOR_NUMBER; ++i) {
> + temp = readl(priv->base + K1_TSU_EN);
> + temp &= ~K1_TSU_EN_MASK(i);
> + temp |= K1_TSU_EN_MASK(i);
> + writel(temp, priv->base + K1_TSU_EN);
> + }
> +
> + return 0;
> +}
> +
> +static void k1_enable_sensor_irq(struct k1_thermal_sensor *sensor)
> +{
> + struct k1_thermal_priv *priv = sensor->priv;
> + unsigned int temp;
> +
> + temp = readl(priv->base + K1_TSU_INT_CLR);
> + temp |= K1_TSU_INT_MASK(sensor->id);
> + writel(temp, priv->base + K1_TSU_INT_CLR);
> +
> + temp = readl(priv->base + K1_TSU_INT_EN);
> + temp &= ~K1_TSU_INT_MASK(sensor->id);
> + writel(temp, priv->base + K1_TSU_INT_EN);
> +}
> +
> +/*
> + * The conversion formula used is:
> + * T(m°C) = (((raw_value & mask) >> shift) - TEMPERATURE_OFFSET) * 1000
> + */
> +static int k1_thermal_get_temp(struct thermal_zone_device *tz, int *temp)
> +{
> + struct k1_thermal_sensor *sensor = thermal_zone_device_priv(tz);
> + struct k1_thermal_priv *priv = sensor->priv;
> +
ditto, suggest to introduce intermediate variable 'val', then assign at last step
> + *temp = readl(priv->base + K1_TSU_DATA(sensor->id));
> + *temp &= K1_TSU_DATA_MASK(sensor->id);
> + *temp >>= K1_TSU_DATA_SHIFT(sensor->id);
> +
> + *temp -= TEMPERATURE_OFFSET;
> +
> + *temp *= 1000;
> +
> + return 0;
> +}
> +
> +/*
> + * For each sensor, the hardware threshold register is 32 bits:
> + * - Lower 16 bits [15:0] configure the low threshold temperature.
> + * - Upper 16 bits [31:16] configure the high threshold temperature.
> + */
> +static int k1_thermal_set_trips(struct thermal_zone_device *tz, int low, int high)
> +{
> + struct k1_thermal_sensor *sensor = thermal_zone_device_priv(tz);
> + struct k1_thermal_priv *priv = sensor->priv;
> + int high_code = high;
> + int low_code = low;
> + unsigned int temp;
> +
> + if (low >= high)
> + return -EINVAL;
> +
> + if (low < 0)
> + low_code = 0;
> +
> + high_code = high_code / 1000 + TEMPERATURE_OFFSET;
> + temp = readl(priv->base + K1_TSU_THRSH(sensor->id));
> + temp &= ~K1_TSU_THRSH_HIGH_MASK;
> + temp |= (high_code << K1_TSU_THRSH_HIGH_SHIFT);
> + writel(temp, priv->base + K1_TSU_THRSH(sensor->id));
> +
> + low_code = low_code / 1000 + TEMPERATURE_OFFSET;
> + temp = readl(priv->base + K1_TSU_THRSH(sensor->id));
> + temp &= ~K1_TSU_THRSH_LOW_MASK;
> + temp |= (low_code << K1_TSU_THRSH_LOW_SHIFT);
> + writel(temp, priv->base + K1_TSU_THRSH(sensor->id));
> +
any reason why not to combine above two readl/writel() into one?
> + return 0;
> +}
> +
> +static const struct thermal_zone_device_ops k1_thermal_ops = {
> + .get_temp = k1_thermal_get_temp,
> + .set_trips = k1_thermal_set_trips,
> +};
> +
> +static irqreturn_t k1_thermal_irq_thread(int irq, void *data)
> +{
> + struct k1_thermal_priv *priv = (struct k1_thermal_priv *)data;
> + int msk, status, i;
s/msk/mask/, for better consistency
> +
> + status = readl(priv->base + K1_TSU_INT_STA);
> +
> + for (i = 0; i < MAX_SENSOR_NUMBER; i++) {
> + if (status & K1_TSU_INT_MASK(i)) {
> + msk = readl(priv->base + K1_TSU_INT_CLR);
> + msk |= K1_TSU_INT_MASK(i);
> + writel(msk, priv->base + K1_TSU_INT_CLR);
> + /* Notify thermal framework to update trips */
> + thermal_zone_device_update(priv->sensors[i].tzd, THERMAL_EVENT_UNSPECIFIED);
> + }
> + }
> +
> + return IRQ_HANDLED;
> +}
> +
> +static int k1_thermal_probe(struct platform_device *pdev)
> +{
> + struct device *dev = &pdev->dev;
> + struct k1_thermal_priv *priv;
> + int i, irq, ret;
> +
> + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
> + if (!priv)
> + return -ENOMEM;
> +
..
> + priv->dev = dev;
> + platform_set_drvdata(pdev, priv);
I'd suggest moving above behind to resource acquisition
> +
> + priv->base = devm_platform_ioremap_resource(pdev, 0);
> + if (IS_ERR(priv->base))
> + return PTR_ERR(priv->base);
> +
> + priv->reset = devm_reset_control_get_exclusive_deasserted(dev, NULL);
> + if (IS_ERR(priv->reset))
> + return dev_err_probe(dev, PTR_ERR(priv->reset),
> + "Failed to get/deassert reset control\n");
> +
> + priv->clk = devm_clk_get_enabled(dev, "core");
> + if (IS_ERR(priv->clk))
> + return dev_err_probe(dev, PTR_ERR(priv->clk),
> + "Failed to get core clock\n");
> +
> + priv->bus_clk = devm_clk_get_enabled(dev, "bus");
> + if (IS_ERR(priv->bus_clk))
> + return dev_err_probe(dev, PTR_ERR(priv->bus_clk),
> + "Failed to get bus clock\n");
> +
> + irq = platform_get_irq(pdev, 0);
> + if (irq < 0)
> + return irq;
> +
> + ret = k1_init_sensors(pdev);
> +
> + for (i = 0; i < MAX_SENSOR_NUMBER; ++i) {
> + priv->sensors[i].id = i;
> + priv->sensors[i].priv = priv;
> + priv->sensors[i].tzd = devm_thermal_of_zone_register(dev,
> + i, priv->sensors + i,
> + &k1_thermal_ops);
> + if (IS_ERR(priv->sensors[i].tzd))
> + return dev_err_probe(dev, PTR_ERR(priv->sensors[i].tzd),
> + "Failed to register thermal zone: %d\n", i);
> +
> + /* Attach sysfs hwmon attributes for userspace monitoring */
> + ret = devm_thermal_add_hwmon_sysfs(dev, priv->sensors[i].tzd);
> + if (ret)
> + dev_warn(dev, "Failed to add hwmon sysfs attributes\n");
> +
> + k1_enable_sensor_irq(priv->sensors + i);
> + }
> +
> + ret = devm_request_threaded_irq(dev, irq, NULL,
> + k1_thermal_irq_thread,
> + IRQF_ONESHOT, "k1_thermal", priv);
> + if (ret < 0)
> + return dev_err_probe(dev, ret, "Failed to request IRQ\n");
> +
> + return 0;
> +}
> +
> +static const struct of_device_id k1_thermal_dt_ids[] = {
> + { .compatible = "spacemit,k1-thermal" },
> + { /* sentinel */ }
> +};
> +
> +MODULE_DEVICE_TABLE(of, k1_thermal_dt_ids);
> +
> +static struct platform_driver k1_thermal_driver = {
> + .driver = {
> + .name = "k1_thermal",
> + .of_match_table = k1_thermal_dt_ids,
> + },
> + .probe = k1_thermal_probe,
> +};
> +module_platform_driver(k1_thermal_driver);
> +
> +MODULE_DESCRIPTION("SpacemiT K1 Thermal Sensor Driver");
> +MODULE_AUTHOR("Shuwei Wu <shuweiwoo@....com>");
> +MODULE_LICENSE("GPL");
>
> --
> 2.51.0
>
--
Yixun Lan (dlan)
Powered by blists - more mailing lists