Re: [PATCH v1] Bluetooth: Enforce key size of 16 bytes on FIPS level

[Luiz Augusto von Dentz <luiz.dentz@gmail.com>]

Hi Archie,

On Tue, Sep 22, 2020 at 12:37 AM Archie Pusaka <apusaka@google.com> wrote:
>
> Hi Luiz,
>
> On Tue, 22 Sep 2020 at 01:13, Luiz Augusto von Dentz
> <luiz.dentz@gmail.com> wrote:
> >
> > Hi Archie,
> >
> > On Mon, Sep 21, 2020 at 1:31 AM Archie Pusaka <apusaka@google.com> wrote:
> > >
> > > From: Archie Pusaka <apusaka@chromium.org>
> > >
> > > According to the spec Ver 5.2, Vol 3, Part C, Sec 5.2.2.8:
> > > Device in security mode 4 level 4 shall enforce:
> > > 128-bit equivalent strength for link and encryption keys required
> > > using FIPS approved algorithms (E0 not allowed, SAFER+ not allowed,
> > > and P-192 not allowed; encryption key not shortened)
> > >
> > > This patch rejects connection with key size below 16 for FIPS level
> > > services.
> > >
> > > Signed-off-by: Archie Pusaka <apusaka@chromium.org>
> > > Reviewed-by: Alain Michaud <alainm@chromium.org>
> > >
> > > ---
> > >
> > >  net/bluetooth/l2cap_core.c | 7 ++++++-
> > >  1 file changed, 6 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
> > > index ade83e224567..306616ec26e6 100644
> > > --- a/net/bluetooth/l2cap_core.c
> > > +++ b/net/bluetooth/l2cap_core.c
> > > @@ -1515,8 +1515,13 @@ static bool l2cap_check_enc_key_size(struct hci_conn *hcon)
> > >          * that have no key size requirements. Ensure that the link is
> > >          * actually encrypted before enforcing a key size.
> > >          */
> > > +       int min_key_size = hcon->hdev->min_enc_key_size;
> > > +
> > > +       if (hcon->sec_level == BT_SECURITY_FIPS)
> > > +               min_key_size = 16;
> > > +
> > >         return (!test_bit(HCI_CONN_ENCRYPT, &hcon->flags) ||
> > > -               hcon->enc_key_size >= hcon->hdev->min_enc_key_size);
> > > +               hcon->enc_key_size >= min_key_size);
> >
> > While this looks fine to me, it looks like this should be placed
> > elsewhere since it takes an hci_conn and it is not L2CAP specific.
>
> From what I understood, it is permissible to use AES-CCM P-256
> encryption with key length < 16 when encrypting the link, but such a
> connection does not satisfy security level 4, and therefore must not
> be given access to level 4 services. However, I think it is
> permissible to give them access to level 3 services or below.
>
> Should I use l2cap chan->sec_level for this purpose? I'm kind of lost
> on the difference between hcon->sec_level and chan->sec_level.

The chan->sec_level is L2CAP channel required sec_level while
hcon->sec_level is the current secure level in effect, at some point I
guess we assign the hcon->sec_level with chan->sec_level but Im not
sure if that has already happened here or not.

> >
> > >  }
> > >
> > >  static void l2cap_do_start(struct l2cap_chan *chan)
> > > --
> > > 2.28.0.681.g6f77f65b4e-goog
> > >
> >
> >
> > --
> > Luiz Augusto von Dentz



-- 
Luiz Augusto von Dentz