linux-kernel - Re: [PATCH v2] HID: asus: support backlight on USB keyboards

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]

Date:   Thu, 6 Apr 2017 11:22:04 +0200
From:   Carlo Caione <carlo@...lessm.com>
To:     Benjamin Tissoires <benjamin.tissoires@...hat.com>
Cc:     Carlo Caione <carlo@...one.org>, jikos@...nel.org,
        linux-input@...r.kernel.org, linux-kernel@...r.kernel.org,
        Linux Upstreaming Team <linux@...lessm.com>
Subject: Re: [PATCH v2] HID: asus: support backlight on USB keyboards

On Thu, Apr 6, 2017 at 11:11 AM, Benjamin Tissoires
<benjamin.tissoires@...hat.com> wrote:
> Hi Carlo,

Hi Benjamin,

[cut]
>> +static int asus_kbd_get_report(struct hid_device *hdev, u8 *buf, size_t buf_size)
>
> Nitpick: should be asus_kbd_set_report()

right :)

[cut]
>> +static void asus_kbd_backlight_work(struct work_struct *work)
>> +{
>> +     struct asus_kbd_leds *led = container_of(work, struct asus_kbd_leds, work);
>> +     u8 buf[] = { FEATURE_KBD_REPORT_ID, 0xba, 0xc5, 0xc4, 0x00 };
>> +     int ret;
>> +
>
> You should probably protect leds->removed by a mutex here to avoid
> having it set to false right after the test. (or not, see asus_remove())

Yeah, makes sense.

[cut]
>> +     /* Initialize keyboard */
>> +     if (asus_kbd_init(hdev) < 0)
>> +             return -ENODEV;
>
> Don't hide the returned error code please. I know it'll be dropped in
> the end, but better not hiding it with an other one. (same for the next
> 2)

ok for the next one but when checking the bit field in kbd_func I
think we still want to return -ENODEV.

>>       struct input_dev *input = hi->input;
>> @@ -178,7 +326,6 @@ static int asus_input_configured(struct hid_device *hdev, struct hid_input *hi)
>>
>>       if (drvdata->quirks & QUIRK_IS_MULTITOUCH) {
>>               int ret;
>> -
>
> Leftover from the previous patch, but please don't remove this empty
> line, it's the coding style policy.

yeah, this slipped through

>>               input_set_abs_params(input, ABS_MT_POSITION_X, 0, MAX_X, 0, 0);
>>               input_set_abs_params(input, ABS_MT_POSITION_Y, 0, MAX_Y, 0, 0);
>>               input_set_abs_params(input, ABS_TOOL_WIDTH, 0, MAX_TOUCH_MAJOR, 0, 0);
>> @@ -198,6 +345,10 @@ static int asus_input_configured(struct hid_device *hdev, struct hid_input *hi)
>>
>>       drvdata->input = input;
>>
>> +     if (drvdata->enable_backlight)
>> +             if (asus_kbd_register_leds(hdev))
>
> could be changed into:
> if (drvdata->enable_backlight && asus_kbd_register_leds(hdev))

OK

[cut]
>> +static void asus_remove(struct hid_device *hdev)
>> +{
>> +     struct asus_drvdata *drvdata = hid_get_drvdata(hdev);
>> +
>> +     if (drvdata->kbd_backlight)
>> +             drvdata->kbd_backlight->removed = true;
>
> Add a cancel_work_sync() here too to terminate currently working
> workers. Bonus point, you don't need the mutex if you call
> cancel_work_sync().

I'll do.

Thanks for the quick review.

-- 
Carlo Caione  |  +39.340.80.30.096  |  Endless