Re: Driver-Core: devtmpfs - reset inode permissions before unlinking

[Kay Sievers]

On Thu, Jan 14, 2010 at 22:07, Henrique de Moraes Holschuh
<hmh@xxxxxxxxxx> wrote:
> On Thu, 14 Jan 2010, Kay Sievers wrote:
>> On Thu, Jan 14, 2010 at 21:53, Henrique de Moraes Holschuh
>> <hmh@xxxxxxxxxx> wrote:
>> > On Wed, 13 Jan 2010, Kay Sievers wrote:
>> >> across the device lifetime by creating hardlinks, in the unusual case
>> >> that there is a user-writable directory on the same filesystem.
>> >
>> > Does a tmpfs mounted in /dev/shm count as "user-writable directory on the
>> > same filesystem" ?
>>
>> Not if it's a separate tmpfs mount, which is recommended. Only if it's
>> just a plain directory on the /dev filesystem.
>
> Yeah, I noticed the abusurdity of my question when I re-read it, thanks for
> being kind in the reply.
>
> That said, this does fix a possible security problem when a misconfigured
> system is used, and the fix looks rather simple... ÂCan it go to -stable
> eventually, even if it is months in the future, after it gets some testing
> in .34 ? Â Minor problems are still problems...

Sure, we could do that. There is some stuff in the current .33 kernel,
which could go into .32-stable too, if that's useful.

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