[PATCH v2 1/2] Bluetooth: Disconnect if E0 is used for Level 4

[PATCH v2 1/2] Bluetooth: Disconnect if E0 is used for Level 4

[Luiz Augusto von Dentz]

From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

E0 is not allowed with Level 4:

BLUETOOTH CORE SPECIFICATION Version 5.2 | Vol 3, Part C page 1319:

  '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'

SC enabled:

> HCI Event: Read Remote Extended Features (0x23) plen 13
        Status: Success (0x00)
        Handle: 256
        Page: 1/2
        Features: 0x0b 0x00 0x00 0x00 0x00 0x00 0x00 0x00
          Secure Simple Pairing (Host Support)
          LE Supported (Host)
          Secure Connections (Host Support)
> HCI Event: Encryption Change (0x08) plen 4
        Status: Success (0x00)
        Handle: 256
        Encryption: Enabled with AES-CCM (0x02)

SC disabled:

> HCI Event: Read Remote Extended Features (0x23) plen 13
        Status: Success (0x00)
        Handle: 256
        Page: 1/2
        Features: 0x03 0x00 0x00 0x00 0x00 0x00 0x00 0x00
          Secure Simple Pairing (Host Support)
          LE Supported (Host)
> HCI Event: Encryption Change (0x08) plen 4
        Status: Success (0x00)
        Handle: 256
        Encryption: Enabled with E0 (0x01)
[May 8 20:23] Bluetooth: hci0: Invalid security: expect AES but E0 was used
< HCI Command: Disconnect (0x01|0x0006) plen 3
        Handle: 256
        Reason: Authentication Failure (0x05)

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_core.h | 10 ++++++----
 net/bluetooth/hci_conn.c         | 17 +++++++++++++++++
 net/bluetooth/hci_event.c        | 20 ++++++++------------
 3 files changed, 31 insertions(+), 16 deletions(-)

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index cdd4f1db8670..c69309ffd40a 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -1402,11 +1402,13 @@ static inline void hci_encrypt_cfm(struct hci_conn *conn, __u8 status)
 	else
 		encrypt = 0x01;
 
-	if (conn->sec_level == BT_SECURITY_SDP)
-		conn->sec_level = BT_SECURITY_LOW;
+	if (!status) {
+		if (conn->sec_level == BT_SECURITY_SDP)
+			conn->sec_level = BT_SECURITY_LOW;
 
-	if (conn->pending_sec_level > conn->sec_level)
-		conn->sec_level = conn->pending_sec_level;
+		if (conn->pending_sec_level > conn->sec_level)
+			conn->sec_level = conn->pending_sec_level;
+	}
 
 	mutex_lock(&hci_cb_list_lock);
 	list_for_each_entry(cb, &hci_cb_list, list) {
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 307800fd18e6..b99b5c6de55a 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -1323,6 +1323,23 @@ int hci_conn_check_link_mode(struct hci_conn *conn)
 			return 0;
 	}
 
+	 /* AES encryption is required for Level 4:
+	  *
+	  * BLUETOOTH CORE SPECIFICATION Version 5.2 | Vol 3, Part C
+	  * page 1319:
+	  *
+	  * 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)
+	  */
+	if (conn->sec_level == BT_SECURITY_FIPS &&
+	    !test_bit(HCI_CONN_AES_CCM, &conn->flags)) {
+		bt_dev_err(conn->hdev,
+			   "Invalid security: Missing AES-CCM usage");
+		return 0;
+	}
+
 	if (hci_conn_ssp_enabled(conn) &&
 	    !test_bit(HCI_CONN_ENCRYPT, &conn->flags))
 		return 0;
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index cfeaee347db3..d6a1e2450696 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -3065,27 +3065,23 @@ static void hci_encrypt_change_evt(struct hci_dev *hdev, struct sk_buff *skb)
 
 	clear_bit(HCI_CONN_ENCRYPT_PEND, &conn->flags);
 
+	/* Check link security requirements are met */
+	if (!hci_conn_check_link_mode(conn))
+		ev->status = HCI_ERROR_AUTH_FAILURE;
+
 	if (ev->status && conn->state == BT_CONNECTED) {
 		if (ev->status == HCI_ERROR_PIN_OR_KEY_MISSING)
 			set_bit(HCI_CONN_AUTH_FAILURE, &conn->flags);
 
+		/* Notify upper layers so they can cleanup before
+		 * disconnecting.
+		 */
+		hci_encrypt_cfm(conn, ev->status);
 		hci_disconnect(conn, HCI_ERROR_AUTH_FAILURE);
 		hci_conn_drop(conn);
 		goto unlock;
 	}
 
-	/* In Secure Connections Only mode, do not allow any connections
-	 * that are not encrypted with AES-CCM using a P-256 authenticated
-	 * combination key.
-	 */
-	if (hci_dev_test_flag(hdev, HCI_SC_ONLY) &&
-	    (!test_bit(HCI_CONN_AES_CCM, &conn->flags) ||
-	     conn->key_type != HCI_LK_AUTH_COMBINATION_P256)) {
-		hci_connect_cfm(conn, HCI_ERROR_AUTH_FAILURE);
-		hci_conn_drop(conn);
-		goto unlock;
-	}
-
 	/* Try reading the encryption key size for encrypted ACL links */
 	if (!ev->status && ev->encrypt && conn->type == ACL_LINK) {
 		struct hci_cp_read_enc_key_size cp;
-- 
2.25.3

[PATCH v2 2/2] Bluetooth: Fix bogus check for re-auth no supported with non-ssp

[Luiz Augusto von Dentz]

From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

This reverts 19f8def031bfa50c579149b200bfeeb919727b27
"Bluetooth: Fix auth_complete_evt for legacy units" which seems to be
working around a bug on a broken controller rather then any limitation
imposed by the Bluetooth spec, in fact if there ws not possible to
re-auth the command shall fail not succeed.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_core.h |  1 -
 net/bluetooth/hci_conn.c         |  8 +++-----
 net/bluetooth/hci_event.c        | 11 ++---------
 3 files changed, 5 insertions(+), 15 deletions(-)

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index c69309ffd40a..13b90b5fdd3b 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -732,7 +732,6 @@ void hci_inquiry_cache_flush(struct hci_dev *hdev);
 /* ----- HCI Connections ----- */
 enum {
 	HCI_CONN_AUTH_PEND,
-	HCI_CONN_REAUTH_PEND,
 	HCI_CONN_ENCRYPT_PEND,
 	HCI_CONN_RSWITCH_PEND,
 	HCI_CONN_MODE_CHANGE_PEND,
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index b99b5c6de55a..f595acf0032b 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -1372,12 +1372,10 @@ static int hci_conn_auth(struct hci_conn *conn, __u8 sec_level, __u8 auth_type)
 		hci_send_cmd(conn->hdev, HCI_OP_AUTH_REQUESTED,
 			     sizeof(cp), &cp);
 
-		/* If we're already encrypted set the REAUTH_PEND flag,
-		 * otherwise set the ENCRYPT_PEND.
+		/* Set the ENCRYPT_PEND to trigger encryption after
+		 * authentication.
 		 */
-		if (test_bit(HCI_CONN_ENCRYPT, &conn->flags))
-			set_bit(HCI_CONN_REAUTH_PEND, &conn->flags);
-		else
+		if (!test_bit(HCI_CONN_ENCRYPT, &conn->flags))
 			set_bit(HCI_CONN_ENCRYPT_PEND, &conn->flags);
 	}
 
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index d6a1e2450696..cfc7edcce3cd 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -2884,14 +2884,8 @@ static void hci_auth_complete_evt(struct hci_dev *hdev, struct sk_buff *skb)
 
 	if (!ev->status) {
 		clear_bit(HCI_CONN_AUTH_FAILURE, &conn->flags);
-
-		if (!hci_conn_ssp_enabled(conn) &&
-		    test_bit(HCI_CONN_REAUTH_PEND, &conn->flags)) {
-			bt_dev_info(hdev, "re-auth of legacy device is not possible.");
-		} else {
-			set_bit(HCI_CONN_AUTH, &conn->flags);
-			conn->sec_level = conn->pending_sec_level;
-		}
+		set_bit(HCI_CONN_AUTH, &conn->flags);
+		conn->sec_level = conn->pending_sec_level;
 	} else {
 		if (ev->status == HCI_ERROR_PIN_OR_KEY_MISSING)
 			set_bit(HCI_CONN_AUTH_FAILURE, &conn->flags);
@@ -2900,7 +2894,6 @@ static void hci_auth_complete_evt(struct hci_dev *hdev, struct sk_buff *skb)
 	}
 
 	clear_bit(HCI_CONN_AUTH_PEND, &conn->flags);
-	clear_bit(HCI_CONN_REAUTH_PEND, &conn->flags);
 
 	if (conn->state == BT_CONFIG) {
 		if (!ev->status && hci_conn_ssp_enabled(conn)) {
-- 
2.25.3

Re: [PATCH v2 1/2] Bluetooth: Disconnect if E0 is used for Level 4

[Luiz Augusto von Dentz]

Hi Marcel,

On Wed, May 20, 2020 at 2:20 PM Luiz Augusto von Dentz
<luiz.dentz@gmail.com> wrote:
>
> From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
>
> E0 is not allowed with Level 4:
>
> BLUETOOTH CORE SPECIFICATION Version 5.2 | Vol 3, Part C page 1319:
>
>   '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'
>
> SC enabled:
>
> > HCI Event: Read Remote Extended Features (0x23) plen 13
>         Status: Success (0x00)
>         Handle: 256
>         Page: 1/2
>         Features: 0x0b 0x00 0x00 0x00 0x00 0x00 0x00 0x00
>           Secure Simple Pairing (Host Support)
>           LE Supported (Host)
>           Secure Connections (Host Support)
> > HCI Event: Encryption Change (0x08) plen 4
>         Status: Success (0x00)
>         Handle: 256
>         Encryption: Enabled with AES-CCM (0x02)
>
> SC disabled:
>
> > HCI Event: Read Remote Extended Features (0x23) plen 13
>         Status: Success (0x00)
>         Handle: 256
>         Page: 1/2
>         Features: 0x03 0x00 0x00 0x00 0x00 0x00 0x00 0x00
>           Secure Simple Pairing (Host Support)
>           LE Supported (Host)
> > HCI Event: Encryption Change (0x08) plen 4
>         Status: Success (0x00)
>         Handle: 256
>         Encryption: Enabled with E0 (0x01)
> [May 8 20:23] Bluetooth: hci0: Invalid security: expect AES but E0 was used
> < HCI Command: Disconnect (0x01|0x0006) plen 3
>         Handle: 256
>         Reason: Authentication Failure (0x05)
>
> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
> ---
>  include/net/bluetooth/hci_core.h | 10 ++++++----
>  net/bluetooth/hci_conn.c         | 17 +++++++++++++++++
>  net/bluetooth/hci_event.c        | 20 ++++++++------------
>  3 files changed, 31 insertions(+), 16 deletions(-)
>
> diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
> index cdd4f1db8670..c69309ffd40a 100644
> --- a/include/net/bluetooth/hci_core.h
> +++ b/include/net/bluetooth/hci_core.h
> @@ -1402,11 +1402,13 @@ static inline void hci_encrypt_cfm(struct hci_conn *conn, __u8 status)
>         else
>                 encrypt = 0x01;
>
> -       if (conn->sec_level == BT_SECURITY_SDP)
> -               conn->sec_level = BT_SECURITY_LOW;
> +       if (!status) {
> +               if (conn->sec_level == BT_SECURITY_SDP)
> +                       conn->sec_level = BT_SECURITY_LOW;
>
> -       if (conn->pending_sec_level > conn->sec_level)
> -               conn->sec_level = conn->pending_sec_level;
> +               if (conn->pending_sec_level > conn->sec_level)
> +                       conn->sec_level = conn->pending_sec_level;
> +       }
>
>         mutex_lock(&hci_cb_list_lock);
>         list_for_each_entry(cb, &hci_cb_list, list) {
> diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
> index 307800fd18e6..b99b5c6de55a 100644
> --- a/net/bluetooth/hci_conn.c
> +++ b/net/bluetooth/hci_conn.c
> @@ -1323,6 +1323,23 @@ int hci_conn_check_link_mode(struct hci_conn *conn)
>                         return 0;
>         }
>
> +        /* AES encryption is required for Level 4:
> +         *
> +         * BLUETOOTH CORE SPECIFICATION Version 5.2 | Vol 3, Part C
> +         * page 1319:
> +         *
> +         * 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)
> +         */
> +       if (conn->sec_level == BT_SECURITY_FIPS &&
> +           !test_bit(HCI_CONN_AES_CCM, &conn->flags)) {
> +               bt_dev_err(conn->hdev,
> +                          "Invalid security: Missing AES-CCM usage");
> +               return 0;
> +       }
> +
>         if (hci_conn_ssp_enabled(conn) &&
>             !test_bit(HCI_CONN_ENCRYPT, &conn->flags))
>                 return 0;
> diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
> index cfeaee347db3..d6a1e2450696 100644
> --- a/net/bluetooth/hci_event.c
> +++ b/net/bluetooth/hci_event.c
> @@ -3065,27 +3065,23 @@ static void hci_encrypt_change_evt(struct hci_dev *hdev, struct sk_buff *skb)
>
>         clear_bit(HCI_CONN_ENCRYPT_PEND, &conn->flags);
>
> +       /* Check link security requirements are met */
> +       if (!hci_conn_check_link_mode(conn))
> +               ev->status = HCI_ERROR_AUTH_FAILURE;
> +
>         if (ev->status && conn->state == BT_CONNECTED) {
>                 if (ev->status == HCI_ERROR_PIN_OR_KEY_MISSING)
>                         set_bit(HCI_CONN_AUTH_FAILURE, &conn->flags);
>
> +               /* Notify upper layers so they can cleanup before
> +                * disconnecting.
> +                */
> +               hci_encrypt_cfm(conn, ev->status);
>                 hci_disconnect(conn, HCI_ERROR_AUTH_FAILURE);
>                 hci_conn_drop(conn);
>                 goto unlock;
>         }
>
> -       /* In Secure Connections Only mode, do not allow any connections
> -        * that are not encrypted with AES-CCM using a P-256 authenticated
> -        * combination key.
> -        */
> -       if (hci_dev_test_flag(hdev, HCI_SC_ONLY) &&
> -           (!test_bit(HCI_CONN_AES_CCM, &conn->flags) ||
> -            conn->key_type != HCI_LK_AUTH_COMBINATION_P256)) {
> -               hci_connect_cfm(conn, HCI_ERROR_AUTH_FAILURE);
> -               hci_conn_drop(conn);
> -               goto unlock;
> -       }
> -
>         /* Try reading the encryption key size for encrypted ACL links */
>         if (!ev->status && ev->encrypt && conn->type == ACL_LINK) {
>                 struct hci_cp_read_enc_key_size cp;
> --
> 2.25.3

Looks like we had left this behind.

-- 
Luiz Augusto von Dentz

Re: [PATCH v2 1/2] Bluetooth: Disconnect if E0 is used for Level 4

[Marcel Holtmann]

Hi Luiz,

> E0 is not allowed with Level 4:
> 
> BLUETOOTH CORE SPECIFICATION Version 5.2 | Vol 3, Part C page 1319:
> 
>  '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'
> 
> SC enabled:
> 
>> HCI Event: Read Remote Extended Features (0x23) plen 13
>        Status: Success (0x00)
>        Handle: 256
>        Page: 1/2
>        Features: 0x0b 0x00 0x00 0x00 0x00 0x00 0x00 0x00
>          Secure Simple Pairing (Host Support)
>          LE Supported (Host)
>          Secure Connections (Host Support)
>> HCI Event: Encryption Change (0x08) plen 4
>        Status: Success (0x00)
>        Handle: 256
>        Encryption: Enabled with AES-CCM (0x02)
> 
> SC disabled:
> 
>> HCI Event: Read Remote Extended Features (0x23) plen 13
>        Status: Success (0x00)
>        Handle: 256
>        Page: 1/2
>        Features: 0x03 0x00 0x00 0x00 0x00 0x00 0x00 0x00
>          Secure Simple Pairing (Host Support)
>          LE Supported (Host)
>> HCI Event: Encryption Change (0x08) plen 4
>        Status: Success (0x00)
>        Handle: 256
>        Encryption: Enabled with E0 (0x01)
> [May 8 20:23] Bluetooth: hci0: Invalid security: expect AES but E0 was used
> < HCI Command: Disconnect (0x01|0x0006) plen 3
>        Handle: 256
>        Reason: Authentication Failure (0x05)
> 
> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
> ---
> include/net/bluetooth/hci_core.h | 10 ++++++----
> net/bluetooth/hci_conn.c         | 17 +++++++++++++++++
> net/bluetooth/hci_event.c        | 20 ++++++++------------
> 3 files changed, 31 insertions(+), 16 deletions(-)

patch has been applied to bluetooth-next tree.

Regards

Marcel

Re: [PATCH v2 2/2] Bluetooth: Fix bogus check for re-auth no supported with non-ssp

[Bastien Nocera]

On Wed, 2020-05-20 at 14:20 -0700, Luiz Augusto von Dentz wrote:
> From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
> 
> This reverts 19f8def031bfa50c579149b200bfeeb919727b27
> "Bluetooth: Fix auth_complete_evt for legacy units" which seems to be
> working around a bug on a broken controller rather then any
> limitation
> imposed by the Bluetooth spec, in fact if there ws not possible to
> re-auth the command shall fail not succeed.
> 
> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Looks like this fell through the cracks, do you want me to rebase and
re-send it?

Re: [PATCH v2 2/2] Bluetooth: Fix bogus check for re-auth no supported with non-ssp

[Luiz Augusto von Dentz]

Hi Bastien,

On Thu, Nov 30, 2023 at 7:02 AM Bastien Nocera <hadess@hadess.net> wrote:
>
> On Wed, 2020-05-20 at 14:20 -0700, Luiz Augusto von Dentz wrote:
> > From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
> >
> > This reverts 19f8def031bfa50c579149b200bfeeb919727b27
> > "Bluetooth: Fix auth_complete_evt for legacy units" which seems to be
> > working around a bug on a broken controller rather then any
> > limitation
> > imposed by the Bluetooth spec, in fact if there ws not possible to
> > re-auth the command shall fail not succeed.
> >
> > Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
>
> Looks like this fell through the cracks, do you want me to rebase and
> re-send it?

Yeah, please resend so we get the CI to test it doesn't break anything.

-- 
Luiz Augusto von Dentz

Re: [PATCH v2 2/2] Bluetooth: Fix bogus check for re-auth no supported with non-ssp

[Bastien Nocera]

On Thu, 2023-11-30 at 08:43 -0500, Luiz Augusto von Dentz wrote:
> Hi Bastien,
> 
> On Thu, Nov 30, 2023 at 7:02 AM Bastien Nocera <hadess@hadess.net>
> wrote:
> > 
> > On Wed, 2020-05-20 at 14:20 -0700, Luiz Augusto von Dentz wrote:
> > > From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
> > > 
> > > This reverts 19f8def031bfa50c579149b200bfeeb919727b27
> > > "Bluetooth: Fix auth_complete_evt for legacy units" which seems
> > > to be
> > > working around a bug on a broken controller rather then any
> > > limitation
> > > imposed by the Bluetooth spec, in fact if there ws not possible
> > > to
> > > re-auth the command shall fail not succeed.
> > > 
> > > Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
> > 
> > Looks like this fell through the cracks, do you want me to rebase
> > and
> > re-send it?
> 
> Yeah, please resend so we get the CI to test it doesn't break
> anything.

Done.

Cheers