All of lore.kernel.org
 help / color / mirror / Atom feed
From: Marcel Holtmann <marcel@holtmann.org>
To: Alfonso Acosta <fons@spotify.com>
Cc: linux-bluetooth@vger.kernel.org
Subject: Re: [PATCH] Bluetooth: Include ADV_IND report in Device Connected event
Date: Fri, 3 Oct 2014 11:47:31 +0200	[thread overview]
Message-ID: <7F79038B-2FA8-4DF5-B25A-D7E9AAE922A8@holtmann.org> (raw)
In-Reply-To: <1412325293-6651-1-git-send-email-fons@spotify.com>

Hi Alfonso,

> There are scenarios when autoconnecting to a device after the
> reception of an ADV_IND report (action 0x02), in which userland
> might want to examine the report's contents.
> 
> For instance, the Service Data might have changed and it would be
> useful to know ahead of time before starting any GATT procedures.
> Also, the ADV_IND may contain Manufacturer Specific data which would
> be lost if not propagated to userland. In fact, this patch results
> from the need to rebond with a device lacking persistent storage which
> notifies about losing its LTK in ADV_IND reports.
> 
> This patch appends the ADV_IND report which triggered the
> autoconnection to the EIR Data in the Device Connected event.
> 
> Signed-off-by: Alfonso Acosta <fons@spotify.com>
> ---
> include/net/bluetooth/hci_core.h |  4 ++-
> net/bluetooth/hci_event.c        | 64 +++++++++++++++++++++++++++++-----------
> net/bluetooth/l2cap_core.c       |  2 +-
> net/bluetooth/mgmt.c             |  5 +++-
> 4 files changed, 54 insertions(+), 21 deletions(-)
> 
> diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
> index 37ff1ae..618c82d 100644
> --- a/include/net/bluetooth/hci_core.h
> +++ b/include/net/bluetooth/hci_core.h
> @@ -398,6 +398,8 @@ struct hci_conn {
> 	__u16		le_conn_interval;
> 	__u16		le_conn_latency;
> 	__u16		le_supv_timeout;
> +	__u8		le_pend_adv_data[HCI_MAX_AD_LENGTH];
> +	__u8		le_pend_adv_len;

I would just name it le_adv_data and le_adv_data_len.

> 	__s8		rssi;
> 	__s8		tx_power;
> 	__s8		max_tx_power;
> @@ -1312,7 +1314,7 @@ void mgmt_new_link_key(struct hci_dev *hdev, struct link_key *key,
> 		       bool persistent);
> void mgmt_device_connected(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 link_type,
> 			   u8 addr_type, u32 flags, u8 *name, u8 name_len,
> -			   u8 *dev_class);
> +			   u8 *dev_class, u8 *adv, u8 adv_len);

I keep wondering if we might just handover hci_conn instead. Something Johan needs to comment on here.

> void mgmt_device_disconnected(struct hci_dev *hdev, bdaddr_t *bdaddr,
> 			      u8 link_type, u8 addr_type, u8 reason,
> 			      bool mgmt_connected);
> diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
> index 8b0a2a6..9a58d03 100644
> --- a/net/bluetooth/hci_event.c
> +++ b/net/bluetooth/hci_event.c
> @@ -1576,9 +1576,13 @@ static void hci_check_pending_name(struct hci_dev *hdev, struct hci_conn *conn,
> 	struct discovery_state *discov = &hdev->discovery;
> 	struct inquiry_entry *e;
> 
> -	if (conn && !test_and_set_bit(HCI_CONN_MGMT_CONNECTED, &conn->flags))
> +	if (conn && !test_and_set_bit(HCI_CONN_MGMT_CONNECTED, &conn->flags)) {
> 		mgmt_device_connected(hdev, bdaddr, ACL_LINK, 0x00, 0, name,
> -				      name_len, conn->dev_class);
> +				      name_len, conn->dev_class,
> +				      conn->le_pend_adv_data,
> +				      conn->le_pend_adv_len);
> +		conn->le_pend_adv_len = 0;

Why are you bothering resetting this to 0? In addition why not just have it set to NULL, 0 here in the first place since this is BR/EDR case.

> +	}
> 
> 	if (discov->state == DISCOVERY_STOPPED)
> 		return;
> @@ -2535,10 +2539,14 @@ static void hci_remote_features_evt(struct hci_dev *hdev,
> 		bacpy(&cp.bdaddr, &conn->dst);
> 		cp.pscan_rep_mode = 0x02;
> 		hci_send_cmd(hdev, HCI_OP_REMOTE_NAME_REQ, sizeof(cp), &cp);
> -	} else if (!test_and_set_bit(HCI_CONN_MGMT_CONNECTED, &conn->flags))
> +	} else if (!test_and_set_bit(HCI_CONN_MGMT_CONNECTED, &conn->flags)) {
> 		mgmt_device_connected(hdev, &conn->dst, conn->type,
> 				      conn->dst_type, 0, NULL, 0,
> -				      conn->dev_class);
> +				      conn->dev_class,
> +				      conn->le_pend_adv_data,
> +				      conn->le_pend_adv_len);
> +		conn->le_pend_adv_len = 0;

Same here. This is BR/EDR case.

> +	}
> 
> 	if (!hci_outgoing_auth_needed(hdev, conn)) {
> 		conn->state = BT_CONNECTED;
> @@ -3433,10 +3441,14 @@ static void hci_remote_ext_features_evt(struct hci_dev *hdev,
> 		bacpy(&cp.bdaddr, &conn->dst);
> 		cp.pscan_rep_mode = 0x02;
> 		hci_send_cmd(hdev, HCI_OP_REMOTE_NAME_REQ, sizeof(cp), &cp);
> -	} else if (!test_and_set_bit(HCI_CONN_MGMT_CONNECTED, &conn->flags))
> +	} else if (!test_and_set_bit(HCI_CONN_MGMT_CONNECTED, &conn->flags)) {
> 		mgmt_device_connected(hdev, &conn->dst, conn->type,
> 				      conn->dst_type, 0, NULL, 0,
> -				      conn->dev_class);
> +				      conn->dev_class,
> +				      conn->le_pend_adv_data,
> +				      conn->le_pend_adv_len);
> +		conn->le_pend_adv_len = 0;

This as well. BR/EDR only. They should just go with adv = NULL and adv_len = 0.

> +	}
> 
> 	if (!hci_outgoing_auth_needed(hdev, conn)) {
> 		conn->state = BT_CONNECTED;
> @@ -4213,9 +4225,13 @@ static void hci_le_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *skb)
> 		goto unlock;
> 	}
> 
> -	if (!test_and_set_bit(HCI_CONN_MGMT_CONNECTED, &conn->flags))
> +	if (!test_and_set_bit(HCI_CONN_MGMT_CONNECTED, &conn->flags)) {
> 		mgmt_device_connected(hdev, &conn->dst, conn->type,
> -				      conn->dst_type, 0, NULL, 0, NULL);
> +				      conn->dst_type, 0, NULL, 0, NULL,
> +				      conn->le_pend_adv_data,
> +				      conn->le_pend_adv_len);
> +		conn->le_pend_adv_len = 0;

The reset to 0 seems like not a good idea. What is this trying to fix.

> +	}
> 
> 	conn->sec_level = BT_SECURITY_LOW;
> 	conn->handle = __le16_to_cpu(ev->handle);
> @@ -4269,25 +4285,26 @@ static void hci_le_conn_update_complete_evt(struct hci_dev *hdev,
> }
> 
> /* This function requires the caller holds hdev->lock */
> -static void check_pending_le_conn(struct hci_dev *hdev, bdaddr_t *addr,
> -				  u8 addr_type, u8 adv_type)
> +static struct hci_conn *check_pending_le_conn(struct hci_dev *hdev,
> +					      bdaddr_t *addr,
> +					      u8 addr_type, u8 adv_type)
> {
> 	struct hci_conn *conn;
> 	struct hci_conn_params *params;
> 
> 	/* If the event is not connectable don't proceed further */
> 	if (adv_type != LE_ADV_IND && adv_type != LE_ADV_DIRECT_IND)
> -		return;
> +		return NULL;
> 
> 	/* Ignore if the device is blocked */
> 	if (hci_bdaddr_list_lookup(&hdev->blacklist, addr, addr_type))
> -		return;
> +		return NULL;
> 
> 	/* Most controller will fail if we try to create new connections
> 	 * while we have an existing one in slave role.
> 	 */
> 	if (hdev->conn_hash.le_num_slave > 0)
> -		return;
> +		return NULL;
> 
> 	/* If we're not connectable only connect devices that we have in
> 	 * our pend_le_conns list.
> @@ -4295,7 +4312,7 @@ static void check_pending_le_conn(struct hci_dev *hdev, bdaddr_t *addr,
> 	params = hci_pend_le_action_lookup(&hdev->pend_le_conns,
> 					   addr, addr_type);
> 	if (!params)
> -		return;
> +		return NULL;
> 
> 	switch (params->auto_connect) {
> 	case HCI_AUTO_CONN_DIRECT:
> @@ -4304,7 +4321,7 @@ static void check_pending_le_conn(struct hci_dev *hdev, bdaddr_t *addr,
> 		 * incoming connections from slave devices.
> 		 */
> 		if (adv_type != LE_ADV_DIRECT_IND)
> -			return;
> +			return NULL;
> 		break;
> 	case HCI_AUTO_CONN_ALWAYS:
> 		/* Devices advertising with ADV_IND or ADV_DIRECT_IND
> @@ -4315,7 +4332,7 @@ static void check_pending_le_conn(struct hci_dev *hdev, bdaddr_t *addr,
> 		 */
> 		break;
> 	default:
> -		return;
> +		return NULL;
> 	}
> 
> 	conn = hci_connect_le(hdev, addr, addr_type, BT_SECURITY_LOW,
> @@ -4328,7 +4345,7 @@ static void check_pending_le_conn(struct hci_dev *hdev, bdaddr_t *addr,
> 		 * count consistent once the connection is established.
> 		 */
> 		params->conn = hci_conn_get(conn);
> -		return;
> +		return conn;
> 	}
> 
> 	switch (PTR_ERR(conn)) {
> @@ -4341,7 +4358,10 @@ static void check_pending_le_conn(struct hci_dev *hdev, bdaddr_t *addr,
> 		break;
> 	default:
> 		BT_DBG("Failed to connect: err %ld", PTR_ERR(conn));
> +		return NULL;
> 	}
> +
> +	return conn;
> }
> 
> static void process_adv_report(struct hci_dev *hdev, u8 type, bdaddr_t *bdaddr,
> @@ -4351,6 +4371,7 @@ static void process_adv_report(struct hci_dev *hdev, u8 type, bdaddr_t *bdaddr,
> 	struct smp_irk *irk;
> 	bool match;
> 	u32 flags;
> +	struct hci_conn *conn;

Move this one up after struct smp_irk.

> 
> 	/* Check if we need to convert to identity address */
> 	irk = hci_get_irk(hdev, bdaddr, bdaddr_type);
> @@ -4360,7 +4381,14 @@ static void process_adv_report(struct hci_dev *hdev, u8 type, bdaddr_t *bdaddr,
> 	}
> 
> 	/* Check if we have been requested to connect to this device */
> -	check_pending_le_conn(hdev, bdaddr, bdaddr_type, type);
> +	conn = check_pending_le_conn(hdev, bdaddr, bdaddr_type, type);
> +	if (conn && type == LE_ADV_IND) {

> +		/* Store report for later inclusion by
> +		 * mgmt_device_connected
> +		 */
> +		memcpy(conn->le_pend_adv_data, data, len);
> +		conn->le_pend_adv_len = len;
> +	}
> 
> 	/* Passive scanning shouldn't trigger any device found events,
> 	 * except for devices marked as CONN_REPORT for which we do send
> diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
> index b6f9777..079834c 100644
> --- a/net/bluetooth/l2cap_core.c
> +++ b/net/bluetooth/l2cap_core.c
> @@ -3875,7 +3875,7 @@ static int l2cap_connect_req(struct l2cap_conn *conn,
> 	    !test_and_set_bit(HCI_CONN_MGMT_CONNECTED, &hcon->flags))
> 		mgmt_device_connected(hdev, &hcon->dst, hcon->type,
> 				      hcon->dst_type, 0, NULL, 0,
> -				      hcon->dev_class);
> +				      hcon->dev_class, NULL, 0);
> 	hci_dev_unlock(hdev);

so this reminds me that for BR/EDR connections we sometimes delay the Device Connected event, while for LE we never do. Maybe that also needs to be brought in sync. This is something that goes along with that we should most likely scan for LE connections first and not just directly connect so we can include the advertising data when LE connections are triggered via L2CAP.

> 
> 	l2cap_connect(conn, cmd, data, L2CAP_CONN_RSP, 0);
> diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
> index efb71b0..6f533fbb 100644
> --- a/net/bluetooth/mgmt.c
> +++ b/net/bluetooth/mgmt.c
> @@ -6173,7 +6173,7 @@ static inline u16 eir_append_data(u8 *eir, u16 eir_len, u8 type, u8 *data,
> 
> void mgmt_device_connected(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 link_type,
> 			   u8 addr_type, u32 flags, u8 *name, u8 name_len,
> -			   u8 *dev_class)
> +			   u8 *dev_class, u8 *adv, u8 adv_len)
> {
> 	char buf[512];
> 	struct mgmt_ev_device_connected *ev = (void *) buf;
> @@ -6192,6 +6192,9 @@ void mgmt_device_connected(struct hci_dev *hdev, bdaddr_t *bdaddr, u8 link_type,
> 		eir_len = eir_append_data(ev->eir, eir_len,
> 					  EIR_CLASS_OF_DEV, dev_class, 3);
> 
> +	memcpy(&ev->eir[eir_len], adv, adv_len);
> +	eir_len += adv_len;
> +
> 	ev->eir_len = cpu_to_le16(eir_len);

To not create a mess, this should actually be either BR/EDR data or LE data. So if adv is set, then only include the LE data and otherwise create the BR/EDR data from name and class.

The reason is that we want the EIR_Data field to be sorted and so to avoid any confusion here with the name data potentially be duplicated here, just for now make it simple. And add a comment that we did so.

> 
> 	mgmt_event(MGMT_EV_DEVICE_CONNECTED, hdev, buf,

Regards

Marcel


  reply	other threads:[~2014-10-03  9:47 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-10-03  8:34 [PATCH] Bluetooth: Include ADV_IND report in Device Connected event Alfonso Acosta
2014-10-03  9:47 ` Marcel Holtmann [this message]
2014-10-03 10:20   ` Johan Hedberg
2014-10-03 12:16     ` Alfonso Acosta
2014-10-03 12:13   ` Alfonso Acosta

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=7F79038B-2FA8-4DF5-B25A-D7E9AAE922A8@holtmann.org \
    --to=marcel@holtmann.org \
    --cc=fons@spotify.com \
    --cc=linux-bluetooth@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.