Bluetooth: ignore NULL link key and reject connection

with the device which has same BD_ADDR

This patch set is used to relieve CVE-2020-26555. The description of the CVE:

Bluetooth legacy BR/EDR PIN code pairing in Bluetooth Core Specification 1.0B through 5.2 may permit an unauthenticated nearby device to spoof the BD_ADDR of the peer device to complete pairing without knowledge of the PIN. [1]

The detail of this attack is in IEEE paper:
BlueMirror: Reflections on Bluetooth Pairing and Provisioning Protocols
[2]

It's a reflection attack. The paper mentioned that attacker can induce the attacked target to generate null link key (zero key) without PIN code. In BR/EDR, the key generation is actually handled in the controller which is below HCI.

Thus, we can ignore null link key in the handler of "Link Key Notification event" to relieve the attack. And, a condition of this attack is that attacker should change the BR_ADDR of his hacking device (Host B) to equal to the BR_ADDR with the target device being attacked (Host A). So we reject the connection with device which has same BD_ADDR both on HCI_Create_Connection and HCI_Connection_Request to prevent the attack.

Similar implementations also show in btstack project. [3][4][5]

Lee, Chun-Yi (2):
Bluetooth: hci_event: Ignore NULL link key
Bluetooth: Reject connection with the device which has same BD_ADDR

 net/bluetooth/hci_conn.c  |  7 +++++++
 net/bluetooth/hci_event.c | 16 ++++++++++++++++
 2 files changed, 23 insertions(+)

2.35.3

From 2c6cd3f353d21086a3163a9ad461789d203a7ee4 Mon Sep 17 00:00:00 2001
From: "Lee, Chun-Yi" <[email protected]>
Date: Sat, 30 Sep 2023 16:56:56 +0800
Subject: [PATCH 0/2] Bluetooth: ignore NULL link key and reject connection 
with the device which has same BD_ADDR
 
This patch set is used to relieve CVE-2020-26555. The description of the
CVE:

Bluetooth legacy BR/EDR PIN code pairing in Bluetooth Core Specification
1.0B through 5.2 may permit an unauthenticated nearby device to spoof
the BD_ADDR of the peer device to complete pairing without knowledge
of the PIN. [1]

The detail of this attack is in IEEE paper:
BlueMirror: Reflections on Bluetooth Pairing and Provisioning Protocols
[2]

It's a reflection attack. The paper mentioned that attacker can induce
the attacked target to generate null link key (zero key) without PIN
code. In BR/EDR, the key generation is actually handled in the controller
which is below HCI.

Thus, we can ignore null link key in the handler of "Link Key Notification
event" to relieve the attack. And, a condition of this attack is that
attacker should change the BR_ADDR of his hacking device (Host B) to equal
to the BR_ADDR with the target device being attacked (Host A). So we reject
the connection with device which has same BD_ADDR both on HCI_Create_Connection
and HCI_Connection_Request to prevent the attack.

Similar implementations also show in btstack project. [3][4][5]

Link: https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2020-26555 [1]
Link: https://ieeexplore.ieee.org/abstract/document/9474325/authors#authors [2]
Link: https://github.com/bluekitchen/btstack/blob/master/src/hci.c#L3722 [3]
Link: https://github.com/bluekitchen/btstack/blob/master/src/hci.c#L3523 [4]
Link: https://github.com/bluekitchen/btstack/blob/master/src/hci.c#L7297 [5]

Lee, Chun-Yi (2):
  Bluetooth: hci_event: Ignore NULL link key
  Bluetooth: Reject connection with the device which has same BD_ADDR

 net/bluetooth/hci_conn.c  |  7 +++++++
 net/bluetooth/hci_event.c | 16 ++++++++++++++++
 2 files changed, 23 insertions(+)

-- 
2.35.3

This change is used to relieve CVE-2020-26555. The description of the CVE:

Bluetooth legacy BR/EDR PIN code pairing in Bluetooth Core Specification 1.0B through 5.2 may permit an unauthenticated nearby device to spoof the BD_ADDR of the peer device to complete pairing without knowledge of the PIN. [1]

The detail of this attack is in IEEE paper:
BlueMirror: Reflections on Bluetooth Pairing and Provisioning Protocols
[2]

It's a reflection attack. The paper mentioned that attacker can induce the attacked target to generate null link key (zero key) without PIN code. In BR/EDR, the key generation is actually handled in the controller which is below HCI.

Thus, we can ignore null link key in the handler of "Link Key Notification event" to relieve the attack. A similar implementation also shows in btstack project. [3]

v3: Drop the connection when null link key be detected.

v2:
- Used Link: tag instead of Closes:
- Used bt_dev_dbg instead of BT_DBG
- Added Fixes: tag

---
 net/bluetooth/hci_event.c | 8 ++++++++
 1 file changed, 8 insertions(+)

index 95816a938cea..a20a94e85b1a 100644
@@ -4692,6 +4692,14 @@ static void hci_link_key_notify_evt(struct hci_dev *hdev, void *data,
 	if (!conn)
 		goto unlock;
 
+	/* Ignore NULL link key against CVE-2020-26555 */
+	if (!memcmp(ev->link_key, ZERO_KEY, HCI_LINK_KEY_SIZE)) {
+		bt_dev_dbg(hdev, "Ignore NULL link key (ZERO KEY) for %pMR", &ev->bdaddr);
+		hci_disconnect(conn, HCI_ERROR_AUTH_FAILURE);
+		hci_conn_drop(conn);
+		goto unlock;
+	}
+
 	hci_conn_hold(conn);
 	conn->disc_timeout = HCI_DISCONN_TIMEOUT;
 	hci_conn_drop(conn);

2.35.3

From 5fed42878147cffa22e695732d87a85c218d217f Mon Sep 17 00:00:00 2001
From: "Lee, Chun-Yi" <[email protected]>
Date: Fri, 14 Jul 2023 21:14:01 +0800
Subject: [PATCH 1/2] Bluetooth: hci_event: Ignore NULL link key
In-Reply-To: <20231001074524.GA14361@linux-691t>
References: <20231001074524.GA14361@linux-691t>

This change is used to relieve CVE-2020-26555. The description of the
CVE:

Bluetooth legacy BR/EDR PIN code pairing in Bluetooth Core Specification
1.0B through 5.2 may permit an unauthenticated nearby device to spoof
the BD_ADDR of the peer device to complete pairing without knowledge
of the PIN. [1]

The detail of this attack is in IEEE paper:
BlueMirror: Reflections on Bluetooth Pairing and Provisioning Protocols
[2]

It's a reflection attack. The paper mentioned that attacker can induce
the attacked target to generate null link key (zero key) without PIN
code. In BR/EDR, the key generation is actually handled in the controller
which is below HCI.

Thus, we can ignore null link key in the handler of "Link Key Notification
event" to relieve the attack. A similar implementation also shows in
btstack project. [3]

v3: Drop the connection when null link key be detected.

v2:
- Used Link: tag instead of Closes:
- Used bt_dev_dbg instead of BT_DBG
- Added Fixes: tag

Fixes: 55ed8ca10f35 ("Bluetooth: Implement link key handling for the management interface")
Link: https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2020-26555 [1]
Link: https://ieeexplore.ieee.org/abstract/document/9474325/authors#authors [2]
Link: https://github.com/bluekitchen/btstack/blob/master/src/hci.c#L3722 [3]
Signed-off-by: Lee, Chun-Yi <[email protected]>
---
 net/bluetooth/hci_event.c | 8 ++++++++
 1 file changed, 8 insertions(+)

index 95816a938cea..a20a94e85b1a 100644
@@ -4692,6 +4692,14 @@ static void hci_link_key_notify_evt(struct hci_dev *hdev, void *data,
 	if (!conn)
 		goto unlock;
 
+	/* Ignore NULL link key against CVE-2020-26555 */
+	if (!memcmp(ev->link_key, ZERO_KEY, HCI_LINK_KEY_SIZE)) {
+		bt_dev_dbg(hdev, "Ignore NULL link key (ZERO KEY) for %pMR", &ev->bdaddr);
+		hci_disconnect(conn, HCI_ERROR_AUTH_FAILURE);
+		hci_conn_drop(conn);
+		goto unlock;
+	}
+
 	hci_conn_hold(conn);
 	conn->disc_timeout = HCI_DISCONN_TIMEOUT;
 	hci_conn_drop(conn);

2.35.3

This change is used to relieve CVE-2020-26555. The description of the CVE:

Bluetooth legacy BR/EDR PIN code pairing in Bluetooth Core Specification 1.0B through 5.2 may permit an unauthenticated nearby device to spoof the BD_ADDR of the peer device to complete pairing without knowledge of the PIN. [1]

The detail of this attack is in IEEE paper:
BlueMirror: Reflections on Bluetooth Pairing and Provisioning Protocols
[2]

It's a reflection attack. The paper mentioned that attacker can induce the attacked target to generate null link key (zero key) without PIN code. In BR/EDR, the key generation is actually handled in the controller which is below HCI.

A condition of this attack is that attacker should change the BR_ADDR of his hacking device (Host B) to equal to the BR_ADDR with the target device being attacked (Host A).

Thus, we reject the connection with device which has same BD_ADDR both on HCI_Create_Connection and HCI_Connection_Request to prevent the attack. A similar implementation also shows in btstack project. [3][4]

---
 net/bluetooth/hci_conn.c  | 7 +++++++
 net/bluetooth/hci_event.c | 8 ++++++++
 2 files changed, 15 insertions(+)

index 056f9516e46d..583d2e18314e 100644
@@ -1611,6 +1611,13 @@ struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst,
 		return ERR_PTR(-EOPNOTSUPP);
 	}
 
+	/* Reject outgoing connection to device with same BD ADDR against CVE-2020-26555 */
+	if (!bacmp(&hdev->bdaddr, dst))
+	{
+		bt_dev_dbg(hdev, "Reject connection to the device with same BD_ADDR %pMR\n", dst);
+		return ERR_PTR(-ECONNREFUSED);
+	}
+
 	acl = hci_conn_hash_lookup_ba(hdev, ACL_LINK, dst);
 	if (!acl) {
 		acl = hci_conn_add(hdev, ACL_LINK, dst, HCI_ROLE_MASTER);

index a20a94e85b1a..d66718190dc5 100644
@@ -3268,6 +3268,14 @@ static void hci_conn_request_evt(struct hci_dev *hdev, void *data,
 
 	bt_dev_dbg(hdev, "bdaddr %pMR type 0x%x", &ev->bdaddr, ev->link_type);
 
+	/* Reject incoming connection from device with same BD ADDR against CVE-2020-26555 */
+	if (!bacmp(&hdev->bdaddr, &ev->bdaddr))
+	{
+		bt_dev_dbg(hdev, "Reject connection from the device with same BD_ADDR %pMR\n", &ev->bdaddr);
+		hci_reject_conn(hdev, &ev->bdaddr);
+		return;
+	}
+
 	mask |= hci_proto_connect_ind(hdev, &ev->bdaddr, ev->link_type,
 				      &flags);
 

2.35.3

From 2c6cd3f353d21086a3163a9ad461789d203a7ee4 Mon Sep 17 00:00:00 2001
From: "Lee, Chun-Yi" <[email protected]>
Date: Wed, 16 Aug 2023 21:03:06 +0800
Subject: [PATCH 2/2] Bluetooth: Reject connection with the device which has
 same BD_ADDR
In-Reply-To: <20231001074524.GA14361@linux-691t>
References: <20231001074524.GA14361@linux-691t>

This change is used to relieve CVE-2020-26555. The description of
the CVE:

Bluetooth legacy BR/EDR PIN code pairing in Bluetooth Core Specification
1.0B through 5.2 may permit an unauthenticated nearby device to spoof
the BD_ADDR of the peer device to complete pairing without knowledge
of the PIN. [1]

The detail of this attack is in IEEE paper:
BlueMirror: Reflections on Bluetooth Pairing and Provisioning Protocols
[2]

It's a reflection attack. The paper mentioned that attacker can induce
the attacked target to generate null link key (zero key) without PIN
code. In BR/EDR, the key generation is actually handled in the controller
which is below HCI.

A condition of this attack is that attacker should change the
BR_ADDR of his hacking device (Host B) to equal to the BR_ADDR with
the target device being attacked (Host A).

Thus, we reject the connection with device which has same BD_ADDR
both on HCI_Create_Connection and HCI_Connection_Request to prevent
the attack. A similar implementation also shows in btstack project.
[3][4]

Link: https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2020-26555 [1]
Link: https://ieeexplore.ieee.org/abstract/document/9474325/authors#authors [2]
Link: https://github.com/bluekitchen/btstack/blob/master/src/hci.c#L3523 [3]
Link: https://github.com/bluekitchen/btstack/blob/master/src/hci.c#L7297 [4]
Signed-off-by: Lee, Chun-Yi <[email protected]>
---
 net/bluetooth/hci_conn.c  | 7 +++++++
 net/bluetooth/hci_event.c | 8 ++++++++
 2 files changed, 15 insertions(+)

index 056f9516e46d..583d2e18314e 100644
@@ -1611,6 +1611,13 @@ struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst,
 		return ERR_PTR(-EOPNOTSUPP);
 	}
 
+	/* Reject outgoing connection to device with same BD ADDR against CVE-2020-26555 */
+	if (!bacmp(&hdev->bdaddr, dst))
+	{
+		bt_dev_dbg(hdev, "Reject connection to the device with same BD_ADDR %pMR\n", dst);
+		return ERR_PTR(-ECONNREFUSED);
+	}
+
 	acl = hci_conn_hash_lookup_ba(hdev, ACL_LINK, dst);
 	if (!acl) {
 		acl = hci_conn_add(hdev, ACL_LINK, dst, HCI_ROLE_MASTER);

index a20a94e85b1a..d66718190dc5 100644
@@ -3268,6 +3268,14 @@ static void hci_conn_request_evt(struct hci_dev *hdev, void *data,
 
 	bt_dev_dbg(hdev, "bdaddr %pMR type 0x%x", &ev->bdaddr, ev->link_type);
 
+	/* Reject incoming connection from device with same BD ADDR against CVE-2020-26555 */
+	if (!bacmp(&hdev->bdaddr, &ev->bdaddr))
+	{
+		bt_dev_dbg(hdev, "Reject connection from the device with same BD_ADDR %pMR\n", &ev->bdaddr);
+		hci_reject_conn(hdev, &ev->bdaddr);
+		return;
+	}
+
 	mask |= hci_proto_connect_ind(hdev, &ev->bdaddr, ev->link_type,
 				      &flags);
 

2.35.3

Hi experts,

Please ignore this patch set because I used wrong mutt command to send out patch. It causes that the mail has duplicate contents. I will send out a new series.

Sorry for any inconvenience caused!

Joey Lee

This change is used to relieve CVE-2020-26555. The description of the CVE:

Bluetooth legacy BR/EDR PIN code pairing in Bluetooth Core Specification 1.0B through 5.2 may permit an unauthenticated nearby device to spoof the BD_ADDR of the peer device to complete pairing without knowledge of the PIN. [1]

The detail of this attack is in IEEE paper:
BlueMirror: Reflections on Bluetooth Pairing and Provisioning Protocols
[2]

It's a reflection attack. The paper mentioned that attacker can induce the attacked target to generate null link key (zero key) without PIN code. In BR/EDR, the key generation is actually handled in the controller which is below HCI.

Thus, we can ignore null link key in the handler of "Link Key Notification event" to relieve the attack. A similar implementation also shows in btstack project. [3]

v3: Drop the connection when null link key be detected.

v2:
- Used Link: tag instead of Closes:
- Used bt_dev_dbg instead of BT_DBG
- Added Fixes: tag

---
 net/bluetooth/hci_event.c | 8 ++++++++
 1 file changed, 8 insertions(+)

index 95816a938cea..a20a94e85b1a 100644
@@ -4692,6 +4692,14 @@ static void hci_link_key_notify_evt(struct hci_dev *hdev, void *data,
 	if (!conn)
 		goto unlock;
 
+	/* Ignore NULL link key against CVE-2020-26555 */
+	if (!memcmp(ev->link_key, ZERO_KEY, HCI_LINK_KEY_SIZE)) {
+		bt_dev_dbg(hdev, "Ignore NULL link key (ZERO KEY) for %pMR", &ev->bdaddr);
+		hci_disconnect(conn, HCI_ERROR_AUTH_FAILURE);
+		hci_conn_drop(conn);
+		goto unlock;
+	}
+
 	hci_conn_hold(conn);
 	conn->disc_timeout = HCI_DISCONN_TIMEOUT;
 	hci_conn_drop(conn);

2.35.3

This change is used to relieve CVE-2020-26555. The description of the CVE:

Bluetooth legacy BR/EDR PIN code pairing in Bluetooth Core Specification 1.0B through 5.2 may permit an unauthenticated nearby device to spoof the BD_ADDR of the peer device to complete pairing without knowledge of the PIN. [1]

The detail of this attack is in IEEE paper:
BlueMirror: Reflections on Bluetooth Pairing and Provisioning Protocols
[2]

It's a reflection attack. The paper mentioned that attacker can induce the attacked target to generate null link key (zero key) without PIN code. In BR/EDR, the key generation is actually handled in the controller which is below HCI.

A condition of this attack is that attacker should change the BR_ADDR of his hacking device (Host B) to equal to the BR_ADDR with the target device being attacked (Host A).

Thus, we reject the connection with device which has same BD_ADDR both on HCI_Create_Connection and HCI_Connection_Request to prevent the attack. A similar implementation also shows in btstack project. [3][4]

---
 net/bluetooth/hci_conn.c  | 7 +++++++
 net/bluetooth/hci_event.c | 8 ++++++++
 2 files changed, 15 insertions(+)

index 056f9516e46d..583d2e18314e 100644
@@ -1611,6 +1611,13 @@ struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst,
 		return ERR_PTR(-EOPNOTSUPP);
 	}
 
+	/* Reject outgoing connection to device with same BD ADDR against CVE-2020-26555 */
+	if (!bacmp(&hdev->bdaddr, dst))
+	{
+		bt_dev_dbg(hdev, "Reject connection to the device with same BD_ADDR %pMR\n", dst);
+		return ERR_PTR(-ECONNREFUSED);
+	}
+
 	acl = hci_conn_hash_lookup_ba(hdev, ACL_LINK, dst);
 	if (!acl) {
 		acl = hci_conn_add(hdev, ACL_LINK, dst, HCI_ROLE_MASTER);

index a20a94e85b1a..d66718190dc5 100644
@@ -3268,6 +3268,14 @@ static void hci_conn_request_evt(struct hci_dev *hdev, void *data,
 
 	bt_dev_dbg(hdev, "bdaddr %pMR type 0x%x", &ev->bdaddr, ev->link_type);
 
+	/* Reject incoming connection from device with same BD ADDR against CVE-2020-26555 */
+	if (!bacmp(&hdev->bdaddr, &ev->bdaddr))
+	{
+		bt_dev_dbg(hdev, "Reject connection from the device with same BD_ADDR %pMR\n", &ev->bdaddr);
+		hci_reject_conn(hdev, &ev->bdaddr);
+		return;
+	}
+
 	mask |= hci_proto_connect_ind(hdev, &ev->bdaddr, ev->link_type,
 				      &flags);
 

2.35.3

This is an automated email and please do not reply to this email.

Dear Submitter,

Thank you for submitting the patches to the linux bluetooth mailing list.
While preparing the CI tests, the patches you submitted couldn't be applied to the current HEAD of the repository.

----- Output -----

error: patch failed: net/bluetooth/hci_event.c:4692
error: net/bluetooth/hci_event.c: patch does not apply
hint: Use 'git am --show-current-patch' to see the failed patch

Please resolve the issue and submit the patches again.

---
Regards,
Linux Bluetooth

Hi experts,

hm... I will port my patch to newest mainline kernel and send out series again.

Sorry for any inconvenience caused!

Joey Lee

This is automated email and please do not reply to this email!

Dear submitter,

Thank you for submitting the patches to the linux bluetooth mailing list. This is a CI test results with your patch series:
PW Link:https://patchwork.kernel.org/project/bluetooth/list/?series=789096

---Test result---

Test Summary:

CheckPatch                    PASS      0.77 seconds
GitLint                       PASS      0.32 seconds
SubjectPrefix                 PASS      0.10 seconds
BuildKernel                   PASS      43.13 seconds
CheckAllWarning               PASS      46.91 seconds
CheckSparse                   WARNING   53.60 seconds
CheckSmatch                   WARNING   140.37 seconds
BuildKernel32                 PASS      41.66 seconds
TestRunnerSetup               PASS      622.54 seconds
TestRunner_l2cap-tester       PASS      38.78 seconds
TestRunner_iso-tester         PASS      82.16 seconds
TestRunner_bnep-tester        PASS      13.74 seconds
TestRunner_mgmt-tester        FAIL      251.73 seconds
TestRunner_rfcomm-tester      PASS      21.30 seconds
TestRunner_sco-tester         PASS      24.23 seconds
TestRunner_ioctl-tester       PASS      23.16 seconds
TestRunner_mesh-tester        PASS      16.89 seconds
TestRunner_smp-tester         PASS      20.09 seconds
TestRunner_userchan-tester    PASS      15.21 seconds
IncrementalBuild              PASS      40.23 seconds

Details
##############################
Test: CheckSparse - WARNING
Desc: Run sparse tool with linux kernel
Output:
net/bluetooth/hci_event.c: note: in included file (through include/net/bluetooth/hci_core.h):
##############################
Test: CheckSmatch - WARNING
Desc: Run smatch tool with source
Output:
net/bluetooth/hci_event.c: note: in included file (through include/net/bluetooth/hci_core.h):
##############################
Test: TestRunner_mgmt-tester - FAIL
Desc: Run mgmt-tester with test-runner
Output:
Total: 497, Passed: 496 (99.8%), Failed: 1, Not Run: 0

Failed Test Cases
LL Privacy - Unpair 1 Timed out 2.016 seconds

---
Regards,
Linux Bluetooth