Re: [PATCH 2/2 v4] SELinux: per-command whitelisting of ioctls

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Jun 9, 2015 at 1:36 PM, Nick Kralevich <nnk@xxxxxxxxxx> wrote:
> Thanks again Paul for taking the time to review these patches. I'm
> quite excited about getting these patches into the mainline kernel,
> since reducing ioctl attack surface is a big goal of Android's.
>
> Just FYI: the original version of the patches are now shipping with
> the Android M preview release
> (https://developer.android.com/preview/download.html), so if you're
> interested in trying it out, please do. We're doing ioctl filtering on
> tcp/udp socket file descriptors, and so far we haven't seen any bugs
> from this code.

Hi Nick,

Just an update that I haven't forgotten about the latest revision,
it's on my list of things for as soon as the merge window closes.

-Paul

> On Wed, Jun 3, 2015 at 2:01 PM, Paul Moore <paul@xxxxxxxxxxxxxx> wrote:
>> On Wed, Jun 3, 2015 at 2:40 PM, Jeffrey Vander Stoep <jeffv@xxxxxxxxxx> wrote:
>>> Ok, I have finished a revision that addresses your comments (will
>>> email out shortly). You'll be happy to hear that I am using "extended
>>> permissions" instead of "operations."
>>>
>>> I tried to focus on:
>>>
>>> -Creating a stable binary policy format that will work for ioctls and
>>> netlink (and others) such that the policy version XPERMS_IOCTL will
>>> remain valid if/when version XPERMS_NETLINK is added
>>
>> Great, thank you.  I realize we may need to change it when we get
>> there, but I appreciate the effort.
>>
>>> -Keeping the current version simple, don’t partially add logic for
>>> selecting between ioctl/netlink in the AVC. I originally started
>>> adding additional components to the avc structures, but a few
>>> questions came up that Stephen and I did not know the answer to. It
>>> makes sense to punt these decisions to if/when the netlink extended
>>> permissions capability is actually added (saves memory in the
>>> meantime). The internals can change as long as the binary policy is
>>> stable.
>>
>> That sounds fine to me.
>>
>> --
>> paul moore
>> www.paul-moore.com
>
>
>
> --
> Nick Kralevich | Android Security | nnk@xxxxxxxxxx | 650.214.4037



-- 
paul moore
www.paul-moore.com

_______________________________________________
Selinux mailing list
Selinux@xxxxxxxxxxxxx
To unsubscribe, send email to Selinux-leave@xxxxxxxxxxxxx.
To get help, send an email containing "help" to Selinux-request@xxxxxxxxxxxxx.





[Index of Archives]     [Selinux Refpolicy]     [Linux SGX]     [Fedora Users]     [Fedora Desktop]     [Yosemite Photos]     [Yosemite Camping]     [Yosemite Campsites]     [KDE Users]     [Gnome Users]

  Powered by Linux