3.4-stable review patch. If anyone has any objections, please let me know. ------------------ From: Andre Guedes <andre.guedes@xxxxxxxxxxxxx> commit f8776218e8546397be64ad2bc0ebf4748522d6e3 upstream. While playing the peripheral role, the host gets a LE Long Term Key Request Event from the controller when a connection is established with a bonded device. The host then informs the LTK which should be used for the connection. Once the link is encrypted, the host gets an Encryption Change Event. Therefore we should set conn->pending_sec_level instead of conn-> sec_level in hci_le_ltk_request_evt. This way, conn->sec_level is properly updated in hci_encrypt_change_evt. Moreover, since we have a LTK associated to the device, we have at least BT_SECURITY_MEDIUM security level. Signed-off-by: Andre Guedes <andre.guedes@xxxxxxxxxxxxx> Signed-off-by: Gustavo Padovan <gustavo.padovan@xxxxxxxxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- net/bluetooth/hci_event.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -3375,7 +3375,9 @@ static inline void hci_le_ltk_request_ev cp.handle = cpu_to_le16(conn->handle); if (ltk->authenticated) - conn->sec_level = BT_SECURITY_HIGH; + conn->pending_sec_level = BT_SECURITY_HIGH; + else + conn->pending_sec_level = BT_SECURITY_MEDIUM; hci_send_cmd(hdev, HCI_OP_LE_LTK_REPLY, sizeof(cp), &cp); -- To unsubscribe from this list: send the line "unsubscribe stable" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html