linux-kernel - Re: [PATCH 1/2] extcon: Use BIT() macro for the left-shift operation

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:   Thu, 30 Mar 2017 14:12:04 +0300
From:   Andy Shevchenko <andy.shevchenko@...il.com>
To:     Chanwoo Choi <cw00.choi@...sung.com>
Cc:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Hans de Goede <hdegoede@...hat.com>, chanwoo@...nel.org,
        MyungJoo Ham <myungjoo.ham@...sung.com>
Subject: Re: [PATCH 1/2] extcon: Use BIT() macro for the left-shift operation

On Thu, Mar 30, 2017 at 1:52 PM, Chanwoo Choi <cw00.choi@...sung.com> wrote:
> On 2017년 03월 30일 19:38, Andy Shevchenko wrote:
>> On Thu, Mar 30, 2017 at 12:15 PM, Chanwoo Choi <cw00.choi@...sung.com> wrote:
>>> On 2017년 03월 30일 17:59, Andy Shevchenko wrote:

>>>>>         for (i = 0; i < edev->max_supported; i++) {
>>>>>                 count += sprintf(buf + count, "%s=%d\n",
>>>>>                                 extcon_info[edev->supported_cable[i]].name,
>>>>> -                                !!(edev->state & (1 << i)));
>>>>> +                                !!(edev->state & BIT(i)));
>>>>>         }
>>>>
>>>> While change is okay, the above code is fragile. There is a potential
>>>> buffer overflow.
>>>
>>> When extcon device is registered, extcon_dev_register() check a number of
>>> supported external connectors. The maximum number of supported connectors
>>> is 32. There is no buffer overflow.
>>
>> Is there any limit for name? No, there is not (const char *name).
>> Though for now it is quite unlikely to have the issue.
>
> If there is problem, I want to fix it.

For now it seems no problem.

> I hope your more comment because I don't understand what you point out.
> Did you mention the length of connector name? or buf?

name as a field of sturct __extcon_info.
Means, *potentially* even one record would be enough to break the boundaries.

-- 
With Best Regards,
Andy Shevchenko