linux-kernel - Re: [PATCH] platform/x86: hp_accel: Fix an error handling path in 'lis3lv02d

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

Date:   Mon, 8 Nov 2021 21:09:12 +0100
From:   Christophe JAILLET <christophe.jaillet@...adoo.fr>
To:     Mark Gross <markgross@...nel.org>
Cc:     eric.piel@...mplin-utc.net, hdegoede@...hat.com,
        dmitry.torokhov@...il.com, giedriuswork@...il.com,
        dvhart@...ux.intel.com, akpm@...ux-foundation.org, pavel@...e.cz,
        platform-driver-x86@...r.kernel.org, linux-kernel@...r.kernel.org,
        kernel-janitors@...r.kernel.org
Subject: Re: [PATCH] platform/x86: hp_accel: Fix an error handling path in
 'lis3lv02d_probe()'

Le 08/11/2021 à 20:48, Mark Gross a écrit :
> On Sun, Nov 07, 2021 at 08:57:07PM +0100, Christophe JAILLET wrote:
>> If 'led_classdev_register()' fails, some additional resources should be
>> released.
>>
>> Add the missing 'i8042_remove_filter()' and 'lis3lv02d_remove_fs()' calls
>> that are already in the remove function but are missing here.
>>
>> Fixes: a4c724d0723b ("platform: hp_accel: add a i8042 filter to remove HPQ6000 data from kb bus stream")
>> Fixes: 9e0c79782143 ("lis3lv02d: merge with leds hp disk")
>> Signed-off-by: Christophe JAILLET <christophe.jaillet@...adoo.fr>
>> ---
>>   drivers/platform/x86/hp_accel.c | 2 ++
>>   1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/platform/x86/hp_accel.c b/drivers/platform/x86/hp_accel.c
>> index b183967ecfb7..435a91fe2568 100644
>> --- a/drivers/platform/x86/hp_accel.c
>> +++ b/drivers/platform/x86/hp_accel.c
>> @@ -331,9 +331,11 @@ static int lis3lv02d_probe(struct platform_device *device)
> adding some lines of context:
> 
> 326         /* filter to remove HPQ6000 accelerometer data
> 327          * from keyboard bus stream */
> 328         if (strstr(dev_name(&device->dev), "HPQ6000"))
> 329                 i8042_install_filter(hp_accel_i8042_filter);
> 330
>>   	INIT_WORK(&hpled_led.work, delayed_set_status_worker);
>>   	ret = led_classdev_register(NULL, &hpled_led.led_classdev);
>>   	if (ret) {
>> +		i8042_remove_filter(hp_accel_i8042_filter);
> This filter was added under a conditional.  Should it not be removed under a
> similar conditional?

Agreed that it looks odd, but in the remove function, we already don't 
have the conditional.

Moreover, in, we have 'i8042_remove_filter()':
	if (i8042_platform_filter != filter) {
		ret = -EINVAL;
		goto out;
	}
So, if 'i8042_install_filter(hp_accel_i8042_filter)' is not called, the 
removal will be a no-op.

>>   		lis3lv02d_joystick_disable(&lis3_dev);
>>   		lis3lv02d_poweroff(&lis3_dev);
>>   		flush_work(&hpled_led.work);
>> +		lis3lv02d_remove_fs(&lis3_dev);
> where was the fs ever added?

In 'lis3lv02d_init_device()' (see [1]), like what is undone with 
'lis3lv02d_joystick_disable()' and 'lis3lv02d_poweroff()'.

'lis3lv02d_remove_fs()' is also already part of the remove function.

I guess that having a 'lis3lv02d_uninit_device()' would be much more 
cleaner.


[1]: 
https://elixir.bootlin.com/linux/v5.15.1/source/drivers/misc/lis3lv02d/lis3lv02d.c#L1188

CJ
> 
> --mark
> 
>>   		return ret;
>>   	}
>>   
>> -- 
>> 2.30.2
>>
>