linux-kernel - Re: [PATCH 3/4] security/selinux: decrement sizeof size in strncmp

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:	Sun, 15 Nov 2009 10:44:33 -0800
From:	Casey Schaufler <casey@...aufler-ca.com>
To:	Raja R Harinath <harinath@...rynot.org>
CC:	linux-kernel@...r.kernel.org,
	Casey Schaufler <casey@...aufler-ca.com>
Subject: Re: [PATCH 3/4] security/selinux: decrement sizeof size in strncmp

Raja R Harinath wrote:
> Hi,
>
> Casey Schaufler <casey@...aufler-ca.com> writes:
>
>   
>> Joe Perches wrote:
>>     
> [snip]
>   
>>> I assert that code should be made as readable
>>> as possible and that the code used fit the
>>> reader's expectations.
>>>
>>> strcmp(foo, "BAR") is natural.
>>> strncmp(foo, "BAR", sizeof("BAR")) is unnatural
>>> and should not be used.
>>>       
>> Oh good gravy. I've been writing C code since the 1970's and
>> have seen enough "unnatural" code to make most people think that
>> PASCAL was a good idea. This is not unnatural code. This is an
>> argument over which side of the head of the pin the odd angel
>> should dance on. Give it up. You're advocating a gratuitous
>> change. Can't y'all go find some questionable casts to expunge?
>> That might actually be useful.
>>     
>
> I think the point is that
>
>     strncmp(foo, "BAR", sizeof("BAR"))
>
> is exceedingly similar to
>
>     strncmp(foo, "BAR", strlen("BAR"))
>
> which mean different things.  The point of this series was the suspicion
> that people who intended the "strlen" variant might have used the
> "sizeof" variant.
>   

The point is that the code is correct as written. The change
suggested, changing the sizeof to strlen, would result in
incorrect behavior in the case where foo is "BAR-BAR-BA-RAN".
The other change suggested, which is to change strncmp to
strcmp, would be functionally correct but offers no value,
might be considered less safe in certain circles, and requires
a change that, like any change, adds a trivial amount of risk.

> And, since this confusion exists, it is probably better to use two
> canonical forms for the two different meanings
>
>    strcmp(foo, "BAR")
>    strncmp(foo, "BAR", strlen("BAR"))
>
> and avoid other equivalent formulations.
>
> - Hari
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>
>
>   

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/