Re: [PATCH 5/5] Input: aw86927 - stop playback over suspend and on shutdown

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Input: aw86927: missing devm_add_action_or_reset() for aw86927_reset()
--

commit becd63da0830df31ce436337adc571d6ea9ae15f
Author: Qianhui Xu <xiaqinghuan@mail.ustc.edu.cn>

Input: aw86927 - stop playback over suspend and on shutdown

This commit stops haptic playback, drains pending work, and parks the
chip in reset on shutdown, while also adding suspend and resume handlers
to prevent playback across system sleep.

> diff --git a/drivers/input/misc/aw86927.c b/drivers/input/misc/aw86927.c
> index e8fd880ac1ded..6d45c310a73fd 100644
> --- a/drivers/input/misc/aw86927.c
> +++ b/drivers/input/misc/aw86927.c

[ ... ]

> @@ -900,6 +909,58 @@ static int aw86927_probe(struct i2c_client *client)
>  	return 0;
>  }
>  
> +static void aw86927_reset(void *data)
> +{
> +	struct aw86927_data *haptics = data;
> +
> +	gpiod_set_value_cansleep(haptics->reset_gpio, 1);
> +}
> +
> +static void aw86927_shutdown(struct i2c_client *client)
> +{
> +	struct aw86927_data *haptics = i2c_get_clientdata(client);
> +
> +	disable_work_sync(&haptics->play_work);
> +	aw86927_reset(haptics);
> +}

[Severity: Medium]
aw86927_reset() has the void (*)(void *) callback signature for
devm_add_action_or_reset(), but it is only called from aw86927_shutdown().

In aw86927_probe(), reset_gpio is acquired in the asserted state and then
deasserted by aw86927_hw_reset():

drivers/input/misc/aw86927.c:aw86927_probe() {
	...
	haptics->reset_gpio = devm_gpiod_get(haptics->dev, "reset", GPIOD_OUT_HIGH);
	if (IS_ERR(haptics->reset_gpio))
		return dev_err_probe(haptics->dev, PTR_ERR(haptics->reset_gpio),
				     "Failed to get reset gpio\n");

	/* Hardware reset */
	aw86927_hw_reset(haptics);
	...
}

Should aw86927_reset() also be registered with devm_add_action_or_reset()
after aw86927_hw_reset() in aw86927_probe() so that the chip is placed back
into hardware reset if a subsequent step in aw86927_probe() fails or when
the driver is unbound?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261027132929.242158-1-xiaqinghuan@mail.ustc.edu.cn?part=5