Re: [PATCH v7 4/4] thermal: mediatek: add another get_temp ops for thermal sensors

[Amjad Ouled-Ameur <aouledameur@baylibre.com>]

Hi Daniel,

On 1/19/23 18:03, Amjad Ouled-Ameur wrote:
> Hi Daniel,
>
> On 12/29/22 16:49, Daniel Lezcano wrote:
>> On 06/12/2022 10:18, Amjad Ouled-Ameur wrote:
>>> Hi Daniel,
>>> On Mon Dec 5, 2022 at 8:39 PM CET, Daniel Lezcano wrote:
>>>>
>>>> Hi Amjad,
>>>>
>>>>
>>>> On 05/12/2022 11:41, Amjad Ouled-Ameur wrote:
>>>>
>>>> [ ... ]
>>>>
>>>>>>> @@ -1161,11 +1197,24 @@ static int mtk_thermal_probe(struct platform_device *pdev)
>>>>>>>             platform_set_drvdata(pdev, mt);
>>>>>>>     -    tzdev = devm_thermal_of_zone_register(&pdev->dev, 0, mt,
>>>>>>> -                          &mtk_thermal_ops);
>>>>>>> -    if (IS_ERR(tzdev)) {
>>>>>>> -        ret = PTR_ERR(tzdev);
>>>>>>> -        goto err_disable_clk_peri_therm;
>>>>>>> +    for (i = 0; i < mt->conf->num_sensors + 1; i++) {
>>>>>>> +        tz = devm_kmalloc(&pdev->dev, sizeof(*tz), GFP_KERNEL);
>>>>>>> +        if (!tz)
>>>>>>> +            return -ENOMEM;
>>>>>>> +
>>>>>>> +        tz->mt = mt;
>>>>>>> +        tz->id = i;
>>>>>>> +
>>>>>>> +        tzdev = devm_thermal_of_zone_register(&pdev->dev, i, tz, (i == 0) ?
>>>>>>> +                                 &mtk_thermal_ops :
>>>>>>> + &mtk_thermal_sensor_ops);
>>>>>>
>>>>>> Here you use again the aggregation
>>>>> I addressed this concern in V6, could you please take a look and let me
>>>>> know what you think [0].
>>>>>
>>>>> [0]: https://lore.kernel.org/all/5eb0cdc2-e9f9-dd42-bf80-b7dcd8bcc196@baylibre.com/
>>>>
>>>> May I misunderstanding but AFAICS, this patch is setting the
>>>> mtk_thermal_ops if the sensor id is zero. The get_temp is computing the
>>>> max temperature in this ops which is what we don't want to do.
>>>
>>> Correct, but I think that is out of scope of this patchset, as the current
>>> driver already uses mtk_thermal_ops for sensor 0. The focus of this patchset
>>> is to add support for the other sensors.
>>>
>>> Besides, what do you suggest as a clean implementation if the current one
>>> no longer meets thermal core requirements ?
>>
>> IIUC, there is a sensor per couple of cores. 1 x 2Bigs, 1 x 2Bigs, 1 x 4 Little, right ?
>
> MT8365 SoC has 4 x A53 CPUs. The SoC has 4 thermal zones per sensor. Thermal zone 0 corresponds
>
> to all 4 x A53 CPUs, the other thermal zones (1, 2 and 3) has nothing to do with CPUs. The cooling device type
>
> used for CPUs is passive. FYI, thermal zones 1, 2 and 3 are present in the SoC for debug-purpose only, they are not supposed
>
> to be used for production.
>
After reconsidering the fact that zones 1, 2 and 3 are only used for dev/debug, it might be best to avoid

aggregation as you suggested, and keep only support for zone 0 in this driver. Thus I suggest I send a V8

where I keep only below fixes for this patch if that's okay with you:

- Define "raw_to_mcelsius" function pointer for "struct thermal_bank_cfg".

- Fix "mtk_thermal" variable in mtk_read_temp().

- Set "mt->raw_to_mcelsius" in probe().


For zones 1, 2 and 3 we can later add a different driver specific for dev/debug to probe them to

avoid confusion.


Regards,

Amjad

>
> Regards,
>
> Amjad
>
>>
>> If it is the case, then a thermal zone per sensor with the trip points and a cooling device for each of them.
>>
>> The two thermal zones for the big will share the same cooling device. The little thermal zone will have its own cooling device.
>>
>> If there is the GPU, then its own cooling device also with devfreq.
>>
>>
>>>> <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs
>>>>
>>>> Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
>>>> <http://twitter.com/#!/linaroorg> Twitter |
>>>> <http://www.linaro.org/linaro-blog/> Blog
>>>
>>