Re: [PATCH v4] net: don't call strlen on non-terminated string in dev_set_alias()

[Alexander Potapenko]

On Tue, Jun 6, 2017 at 10:36 PM, David Miller <davem@xxxxxxxxxxxxx> wrote:
> From: Alexander Potapenko <glider@xxxxxxxxxx>
> Date: Tue,  6 Jun 2017 15:56:54 +0200
>
>> KMSAN reported a use of uninitialized memory in dev_set_alias(),
>> which was caused by calling strlcpy() (which in turn called strlen())
>> on the user-supplied non-terminated string.
>>
>> Signed-off-by: Alexander Potapenko <glider@xxxxxxxxxx>
>
> We should not be allowing non-NULL terminated strings for the
> IFLA_IFALIAS attribute.  It's defined as type NLA_STRING in
> the ifla_policy[] array.
Sorry, I couldn't determine from RFC 2233 whether ifAlias is
zero-terminated or not, but looking at validate_nla() I see that
NLA_STRING is supposed to be such.
I'll check what's going on.

> Please figure out why we aren't enforcing the attribute policy
> properly, rather than adding a workaround.
Guess the string has been previously claimed to be non-terminated
here: https://patchwork.ozlabs.org/patch/996/
> Thanks.



-- 
Alexander Potapenko
Software Engineer

Google Germany GmbH
Erika-Mann-StraÃe, 33
80636 MÃnchen

GeschÃftsfÃhrer: Matthew Scott Sucherman, Paul Terence Manicle
Registergericht und -nummer: Hamburg, HRB 86891
Sitz der Gesellschaft: Hamburg