so just to sum up, there is basically no permission revocation or denial in
SEAndroid?

On Tue, Jun 23, 2015 at 7:09 PM, Stephen Smalley <[email protected]> wrote:

> On 06/23/2015 11:42 AM, Tal Palant wrote:
> > so basically install time mac isn't relevant in Android M?
> > And the first part of labeling is used to block applications with a
> > large variety of permissions from being installed all together?
>
> As rpcraig explained, only the part of install-time MAC that supported
> labeling of apps based on signature and optionally package name was ever
> upstreamed into AOSP.  And even in our own branches, we dropped the part
> of install-time MAC that was enforcing restrictions over Android
> permissions long ago; only the support for enforcing a whitelist of what
> apps could be installed remained.  Regardless, that was always an
> install-time permissions check not a runtime check.
>
> We also had experimental runtime permission revocation support in our
> branches for a while back in the 4.2 and earlier days, but dropped that
> when Android 4.3 was released with App Ops, and switched over to working
> on that.  Which led to our Enterprise Ops mechanism for enforcing
> enterprise restrictions over App Ops.  As the new runtime permissions
> mechanism seems similar to or based upon App Ops, that seems like a
> possible area for investigation but we can't do that until we have
> source for M.
>
> None of our middleware MAC enforcement mechanisms were ever accepted
> into AOSP, only our SELinux contributions.  M does include several
> advances to SELinux in Android, and there will be a couple of talks at
> the upcoming Linux Security Summit that will be discussing those advances.
>
>


-- 
טל פולו פלנט
כי שם כזה יש רק אחד
_______________________________________________
Seandroid-list mailing list
[email protected]
To unsubscribe, send email to [email protected].
To get help, send an email containing "help" to 
[email protected].

Reply via email to