[PATCH V2 6/7] coresight: stm: Move ACPI support from AMBA driver to platform driver

[Suzuki K Poulose]

On 05/12/2023 05:20, Anshuman Khandual wrote:
> 
> 
> On 12/4/23 18:47, James Clark wrote:
>>
>>
>> On 04/12/2023 11:50, Anshuman Khandual wrote:
>>>
>>>
>>> On 12/4/23 15:53, James Clark wrote:
>>>>
>>>>
>>>> On 01/12/2023 06:20, Anshuman Khandual wrote:
>>>>> Add support for the stm devices in the platform driver, which can then be
>>>>> used on ACPI based platforms. This change would now allow runtime power
>>>>> management for ACPI based systems. The driver would try to enable the APB
>>>>> clock if available.
>>>>>
>>>>> Cc: Lorenzo Pieralisi <lpieralisi at kernel.org>
>>>>> Cc: Sudeep Holla <sudeep.holla at arm.com>
>>>>> Cc: Suzuki K Poulose <suzuki.poulose at arm.com>
>>>>> Cc: Mike Leach <mike.leach at linaro.org>
>>>>> Cc: James Clark <james.clark at arm.com>
>>>>> Cc: Maxime Coquelin <mcoquelin.stm32 at gmail.com>
>>>>> Cc: Alexandre Torgue <alexandre.torgue at foss.st.com>
>>>>> Cc: linux-acpi at vger.kernel.org
>>>>> Cc: linux-arm-kernel at lists.infradead.org
>>>>> Cc: linux-kernel at vger.kernel.org
>>>>> Cc: coresight at lists.linaro.org
>>>>> Cc: linux-stm32 at st-md-mailman.stormreply.com
>>>>> Signed-off-by: Anshuman Khandual <anshuman.khandual at arm.com>
>>>>> ---
>>>> [...]
>>>>>   
>>>>> -module_amba_driver(stm_driver);
>>>>> +static int stm_platform_probe(struct platform_device *pdev)
>>>>> +{
>>>>> +	struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
>>>>> +	int ret = 0;
>>>>> +
>>>>> +	pm_runtime_get_noresume(&pdev->dev);
>>>>> +	pm_runtime_set_active(&pdev->dev);
>>>>> +	pm_runtime_enable(&pdev->dev);
>>>>> +
>>>>> +	ret = __stm_probe(&pdev->dev, res, NULL);
>>>>
>>>> Very minor nit, but this used to print this:
>>>>
>>>>    coresight stm0: STM500 initialized
>>>>
>>>> And now it prints this:
>>>>
>>>>    coresight stm0: (null) initialized
>>>>
>>>> (null) kind of makes it look a little bit like something has gone wrong.
>>>> Maybe we could just put "initialised" if you don't have a string from ACPI?
>>>
>>> __stm_probe() gets called from both AMBA and platform driver paths. Even though
>>> a NULL check inside dev_info(..."%s initialized\n",...) could be added, but how
>>> to differentiate it from a scenario when coresight_get_uci_data() returns NULL ?
>>
>> Sudeep's suggestion seems ok, just add a hard coded string instead of
>> the NULL. And keep the coresight_get_uci_data() the same.
> 
> Something like this works ?
> 
> diff --git a/drivers/hwtracing/coresight/coresight-stm.c b/drivers/hwtracing/coresight/coresight-stm.c
> index 3387ebc9d8ab..2b6834c4cac6 100644
> --- a/drivers/hwtracing/coresight/coresight-stm.c
> +++ b/drivers/hwtracing/coresight/coresight-stm.c
> @@ -906,7 +906,7 @@ static int __stm_probe(struct device *dev, struct resource *res, void *dev_caps)
>          pm_runtime_put(dev);
>   
>          dev_info(&drvdata->csdev->dev, "%s initialized\n",
> -                (char *)dev_caps);
> +                dev_caps ? (char *)dev_caps: "STM");
>          return 0;
>   

Please could stop abusing the private data for storing the name ? 
Instead, we could read the PID of the device and use a table to map
it to the name and use that instead ?

Suzuki


>   cs_unregister:
>