[v3] thermal: imx8mm: Allow reboot after critical temperature

Commit Message

Fabio Estevam Aug. 24, 2023, 12:21 p.m. UTC

From: Fabio Estevam <[e-mail]>

Currently, after the SoC reaches the critical temperature, the board
goes through a poweroff mechanism.

In some cases, such behavior does not suit well, as the board may be
unattended in the field and rebooting may be a better approach.

The bootloader may also check the temperature and only allow the boot to
proceed when the temperature is below a certain threshold.

Introduce a reboot_on_critical parameter to indicate that the board
will go through a reboot after the critical temperature is reached.

When this parameter is not passed, the default behavior of forcing a
shutdown is preserved.

Tested on a imx8mm-evk board by passing 'imx8mm_thermal.reboot_on_critical'
via kernel command line.

Signed-off-by: Fabio Estevam <[e-mail]>
---
Changes since v2:
- Set the .critical ops at init time and get rid of the
thermal_zone_device_critical(tz) call. (Daniel)

 drivers/thermal/imx8mm_thermal.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

Patch

diff --git a/drivers/thermal/imx8mm_thermal.c b/drivers/thermal/imx8mm_thermal.c
index e89b11b3f2b9..f156e6e81abc 100644
--- a/drivers/thermal/imx8mm_thermal.c
+++ b/drivers/thermal/imx8mm_thermal.c
@@ -10,9 +10,11 @@ 
 #include <linux/err.h>
 #include <linux/io.h>
 #include <linux/module.h>
+#include <linux/moduleparam.h>
 #include <linux/nvmem-consumer.h>
 #include <linux/of.h>
 #include <linux/platform_device.h>
+#include <linux/reboot.h>
 #include <linux/slab.h>
 #include <linux/thermal.h>
 
@@ -75,6 +77,11 @@ 
 #define TMU_VER1		0x1
 #define TMU_VER2		0x2
 
+static bool reboot_on_critical;
+module_param(reboot_on_critical, bool, 0444);
+MODULE_PARM_DESC(reboot_on_critical,
+		 "Reboot the system after the critical temperature is reached.");
+
 struct thermal_soc_data {
 	u32 num_sensors;
 	u32 version;
@@ -146,7 +153,7 @@  static int tmu_get_temp(struct thermal_zone_device *tz, int *temp)
 	return tmu->socdata->get_temp(sensor, temp);
 }
 
-static const struct thermal_zone_device_ops tmu_tz_ops = {
+static struct thermal_zone_device_ops tmu_tz_ops = {
 	.get_temp = tmu_get_temp,
 };
 
@@ -293,6 +300,13 @@  static int imx8mm_tmu_probe_set_calib(struct platform_device *pdev,
 	return imx8mm_tmu_probe_set_calib_v2(pdev, tmu);
 }
 
+static void tmu_critical(struct thermal_zone_device *tz)
+{
+	dev_emerg(thermal_zone_device(tz), "%s: critical temperature reached\n",
+		  thermal_zone_device_type(tz));
+	kernel_restart(NULL);
+}
+
 static int imx8mm_tmu_probe(struct platform_device *pdev)
 {
 	const struct thermal_soc_data *data;
@@ -327,6 +341,9 @@  static int imx8mm_tmu_probe(struct platform_device *pdev)
 	/* disable the monitor during initialization */
 	imx8mm_tmu_enable(tmu, false);
 
+	if (reboot_on_critical)
+		tmu_tz_ops.critical = tmu_critical;
+
 	for (i = 0; i < data->num_sensors; i++) {
 		tmu->sensors[i].priv = tmu;
 		tmu->sensors[i].tzd =