Re: [PATCH v19 06/10] power: reset: Add psci-reboot-mode driver

[Shivendra Pratap <shivendra.pratap@oss.qualcomm.com>]

On 02-01-2026 17:27, Bartosz Golaszewski wrote:
> On Sun, Dec 28, 2025 at 6:21 PM Shivendra Pratap
> <shivendra.pratap@oss.qualcomm.com> wrote:
>>
> 
> [snip]
> 
>> +
>> +static int psci_reboot_mode_probe(struct faux_device *fdev)
>> +{
>> +       struct reboot_mode_driver *reboot;
>> +       struct device_node *psci_np;
>> +       struct device_node *np;
>> +       int ret;
>> +
>> +       psci_np = of_find_compatible_node(NULL, NULL, "arm,psci-1.0");
>> +       if (!psci_np)
>> +               return -ENODEV;
>> +
>> +       /*
>> +        * Find the psci:reboot-mode node.
>> +        * If NULL, continue to register predefined modes.
>> +        * np refcount to be handled by dev;
>> +        * psci_np refcount is decremented by of_find_node_by_name;
>> +        */
> 
> Can you make this comment into full sentences, I had trouble parsing
> the meaning for a minute.
> 
>> +       np = of_find_node_by_name(psci_np, "reboot-mode");
>> +       fdev->dev.of_node = np;
> 
> The logic of the device assigning its own node is a bit sketchy,
> ideally this should be done before the device probes.

Got one doubt while trying to move it to init.

We used a faux_device_create() in init.
faux_device_create() calls the probe function from within its
implementation, so do not find a way to set the node "fdev->dev.of_node 
= np" before the probe call when using faux device.

thanks,
Shivendra