Re: [PATCH v2] HID: playstation: Support DualSense player LED brightness control

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [Critical] Missing cancel_work_sync() in the probe error paths of dualsense_create() and dualshock4_create() causes a use-after-free vulnerability.
--

--- Patch [1]: [PATCH v2] HID: playstation: Support DualSense player LED brightness control ---
commit c4af8f1ee7d50867512c691be4fbde7636ea1510
Author: Kateřina Medvědová <k8ie@mcld.eu>

HID: playstation: Support DualSense player LED brightness control

This patch wires up the controls for the brightness byte on the DualSense
player indicator LEDs. It increases the max_brightness of each player's LED
classdev from 1 to 3, and updates the brightness values before sending
the output reports.

> diff --git a/drivers/hid/hid-playstation.c b/drivers/hid/hid-playstation.c
> index f9dc9378cf777..58786117ec016 100644
> --- a/drivers/hid/hid-playstation.c
> +++ b/drivers/hid/hid-playstation.c
[ ... ]
> @@ -1742,15 +1776,15 @@ static struct ps_device *dualsense_create(struct hid_device *hdev)
>  	int i, ret;
>  
>  	static const struct ps_led_info player_leds_info[] = {
> -		{ LED_FUNCTION_PLAYER1, "white", 1, dualsense_player_led_get_brightness,
> +		{ LED_FUNCTION_PLAYER1, "white", 3, dualsense_player_led_get_brightness,
>  				dualsense_player_led_set_brightness },
> -		{ LED_FUNCTION_PLAYER2, "white", 1, dualsense_player_led_get_brightness,
> +		{ LED_FUNCTION_PLAYER2, "white", 3, dualsense_player_led_get_brightness,
>  				dualsense_player_led_set_brightness },

[Severity: Critical]
This is a pre-existing issue, but the error paths for dualsense_create() and
dualshock4_create() appear to be missing a cancel_work_sync() call.

If initialization fails after the work is scheduled (for example, if
ps_led_register() fails in dualsense_create()), the code jumps to the err
path:

drivers/hid/hid-playstation.c:dualsense_create() {
    ...
err:
	ps_devices_list_remove(ps_dev);
	return ERR_PTR(ret);
}

Since ds->output_worker is not canceled before the devm routines free the
ds/ds4 structure, could this result in a use-after-free when the global
workqueue eventually executes the worker function?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260714184225.84646-1-k8ie@mcld.eu?part=1