Re: Issue with imx_get_temp()

[Fabio Estevam]

Hi Marco,

On Tue, Dec 3, 2019 at 7:15 AM Marco Felsch <m.felsch@xxxxxxxxxxxxxx> wrote:

> Mh.. it seems that the irq gets enabled before a irq-handler is
> registered. As your backlog shows the thermal_zone_device_register()
> triggers a imx_get_temp() and during boot the irq_enabled is false and
> it seems that your temperature is below the alarm_temp. So in such a
> case the enable_irq() is executed. I don't know what happens if we
> enable a irq without a irq-handler.

I think your analysis makes sense.

Should we move the ' data->irq_enabled = true' just prior to calling
thermal_zone_device_register()?

--- a/drivers/thermal/imx_thermal.c
+++ b/drivers/thermal/imx_thermal.c
@@ -803,6 +803,7 @@ static int imx_thermal_probe(struct platform_device *pdev)
                goto legacy_cleanup;
        }

+       data->irq_enabled = true;
        data->tz = thermal_zone_device_register("imx_thermal_zone",
                                                IMX_TRIP_NUM,
                                                BIT(IMX_TRIP_PASSIVE), data,
@@ -837,7 +838,6 @@ static int imx_thermal_probe(struct platform_device *pdev)
        regmap_write(map, data->socdata->sensor_ctrl + REG_SET,
                     data->socdata->measure_temp_mask);

-       data->irq_enabled = true;
        data->mode = THERMAL_DEVICE_ENABLED;

        ret = devm_request_threaded_irq(&pdev->dev, data->irq,