linux-kernel - Re: [PATCH v11] platform/x86: add lenovo wmi camera button driver

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <31bc6c0a-c0e7-4055-82ec-33da51368e0f@gmx.de>
Date: Tue, 26 Mar 2024 21:59:58 +0100
From: Armin Wolf <W_Armin@....de>
To: Andy Shevchenko <andriy.shevchenko@...el.com>, Ai Chao <aichao@...inos.cn>
Cc: hdegoede@...hat.com, ilpo.jarvinen@...ux.intel.com,
 linux-kernel@...r.kernel.org, platform-driver-x86@...r.kernel.org
Subject: Re: [PATCH v11] platform/x86: add lenovo wmi camera button driver

Am 25.03.24 um 17:29 schrieb Andy Shevchenko:

> On Fri, Mar 22, 2024 at 02:47:50PM +0800, Ai Chao wrote:
>> Add lenovo generic wmi driver to support camera button.
> WMI
>
>> The Camera button is a GPIO device. This driver receives ACPI notifyi
>> when the camera button is switched on/off. This driver is used in
>> Lenovo A70, it is a Computer integrated machine.
>> +config LENOVO_WMI_CAMERA
>> +	tristate "Lenovo WMI Camera Button driver"
>> +	depends on ACPI_WMI
>> +	depends on INPUT
> No COMPILE_TEST?
>
>> +	help
>> +	  This driver provides support for Lenovo camera button. The Camera
>> +	  button is a GPIO device. This driver receives ACPI notify when the
>> +	  camera button is switched on/off.
>> +
>> +	  To compile this driver as a module, choose M here: the module
>> +	  will be called lenovo-wmi-camera.
> ...
>
>> +#include <linux/acpi.h>
>> +#include <linux/device.h>
>> +#include <linux/input.h>
>> +#include <linux/module.h>
>> +#include <linux/mutex.h>
> + types.h
>
>> +#include <linux/wmi.h>
> ...
>
>> +struct lenovo_wmi_priv {
>> +	struct input_dev *idev;
>> +	struct mutex notify_lock;	/* lenovo wmi camera button notify lock */
> WMI
>
>> +};
> ...
>
>> +	/* obj->buffer.pointer[0] is camera mode:
>> +	 *      0 camera close
>> +	 *      1 camera open
>> +	 */
> /*
>   * The correct multi-line comment style
>   * is depicted here.
>   */
>
> ...
>
>> +	keycode = (camera_mode == SW_CAMERA_ON ?
>> +		   KEY_CAMERA_ACCESS_ENABLE : KEY_CAMERA_ACCESS_DISABLE);
> Useless parentheses.
>
> ...
>
>> +	ret = input_register_device(priv->idev);
>> +	if (ret)
>> +		return ret;
>> +	mutex_init(&priv->notify_lock);
> Your mutex should be initialized before use. Have you tested that?

Hi,

i suggested that the mutex be initialized after calling input_register_device().
The reason for this is that the mutex is only used inside the WMI notify callback,
and the WMI driver core will only call it after probe() has returned.

So imho it should be safe.

Thanks,
Armin Wolf

>
> ...
>
>> +static struct wmi_driver lenovo_wmi_driver = {
>> +	.driver = {
>> +		.name = "lenovo-wmi-camera",
>> +		.probe_type = PROBE_PREFER_ASYNCHRONOUS,
>> +	},
>> +	.id_table = lenovo_wmi_id_table,
>> +	.no_singleton = true,
>> +	.probe = lenovo_wmi_probe,
>> +	.notify = lenovo_wmi_notify,
>> +	.remove = lenovo_wmi_remove,
>> +};
>> +
> Unneeded blank line.
>
>> +module_wmi_driver(lenovo_wmi_driver);
> ...
>
>> +MODULE_DEVICE_TABLE(wmi, lenovo_wmi_id_table);
> Please, move it closer to the respective table.
>