Pull request: bluetooth-next-2.6 2009-12-03
Subsystems (4)
Marcel Holtmann
maintainerHi Dave,
these are the patches for the 2.6.33 merge window. Sorry for being a bit late, but I got dragged into some other work. Besides the ongoing work on the L2CAP ERTM support it is only bug fixes and cleanups.
I did base this against your net-next tree. If you want me to rebase it, then let me know.
Regards
Marcel
Please pull from
git://git.kernel.org/pub/scm/linux/kernel/git/holtmann/bluetooth-next-2.6.git master
This will update the following files:
drivers/bluetooth/btmrvl_debugfs.c | 1 - drivers/bluetooth/btmrvl_drv.h | 1 + drivers/bluetooth/btmrvl_main.c | 55 +++++++++++------- drivers/bluetooth/btmrvl_sdio.c | 4 + drivers/bluetooth/hci_vhci.c | 20 +------ include/net/bluetooth/hci_core.h | 39 +------------ include/net/bluetooth/l2cap.h | 2 +- net/bluetooth/bnep/core.c | 5 +- net/bluetooth/hci_core.c | 39 +++++++++++-- net/bluetooth/hci_event.c | 6 +- net/bluetooth/hci_sock.c | 11 +++- net/bluetooth/hidp/core.c | 9 +++ net/bluetooth/l2cap.c | 110 +++++++++++++++++++++++++++-------- net/bluetooth/rfcomm/core.c | 6 ++ 14 files changed, 192 insertions(+), 116 deletions(-)
through these ChangeSets:
Amitkumar Karwar (1):
Bluetooth: Enable auto sleep mode for btmrvl driver
Ben Hutchings (1):
Bluetooth: Declare MODULE_FIRMWARE for Marvell SDIO driver
Gustavo F. Padovan (5):
Bluetooth: Initialize variables and timers for both channel's sides
Bluetooth: Fix unset of SrejActioned flag
Bluetooth: Fix sending ReqSeq on I-frames
Bluetooth: Implement RejActioned flag
Bluetooth: Fix 'SendRRorRNR' to send the ReqSeq value
Jiri Kosina (1):
Bluetooth: Implement raw output support for HIDP layer
Marcel Holtmann (7):
Bluetooth: Remove unused global minor variable
Bluetooth: Fix miscdev ops owner for virtual driver
Bluetooth: Return ENETDOWN when interface is down
Bluetooth: Turn hci_recv_frame into an exported function
Bluetooth: Unobfuscate tasklet_schedule usage
Bluetooth: Add L2CAP option for max transmit value
Bluetooth: Add RFCOMM option to use L2CAP ERTM mode
Rahul Tank (1):
Bluetooth: Removal of unused variable in btmrvl driver
Thomas Gleixner (1):
Bluetooth: Remove stub ioctl in hci_vhci
Vikram Kandukuri (1):
Bluetooth: Fix handling of BNEP setup connection requests
Marcel Holtmann
maintainerPATCH 1/18 Bluetooth: Removal of unused variable in btmrvl driver
From: Rahul Tank <[email protected]>
This patch removes unused variable "drvdbg" from btmrvl_debugfs_data structure.
--- drivers/bluetooth/btmrvl_debugfs.c | 1 - 1 files changed, 0 insertions(+), 1 deletions(-)
drivers/bluetooth/btmrvl_debugfs.c +0 -1
@@ -29,7 +29,6 @@ struct btmrvl_debugfs_data {struct dentry *root_dir, *config_dir, *status_dir;/* config */ -struct dentry *drvdbg;struct dentry *psmode;struct dentry *pscmd;struct dentry *hsmode;
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 2/18 Bluetooth: Remove stub ioctl in hci_vhci
From: Thomas Gleixner <[email protected]>
Remove the empty ioctl which just returns -EINVAL. vfs_ioctl() will return -ENOTTY instead, but I doubt that any application will notice the difference :)
--- drivers/bluetooth/hci_vhci.c | 7 ------- 1 files changed, 0 insertions(+), 7 deletions(-)
drivers/bluetooth/hci_vhci.c +0 -7
@@ -218,12 +218,6 @@ static unsigned int vhci_poll(struct file *file, poll_table *wait)return POLLOUT |POLLWRNORM ; } -static int vhci_ioctl (struct inode *inode,struct file *file, -unsigned int cmd,unsigned long arg) -{ -return -EINVAL ; -} -static int vhci_open (struct inode *inode,struct file *file) {struct vhci_data *data;
@@ -287,7 +281,6 @@ static const struct file_operations vhci_fops = { .read = vhci_read, .write = vhci_write, .poll = vhci_poll, - .ioctl = vhci_ioctl, .open = vhci_open, .release = vhci_release, };
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 3/18 Bluetooth: Remove unused global minor variable
After the removal of the module parameter for setting the minor number, this variable became unused. So just remove it.
--- drivers/bluetooth/hci_vhci.c | 12 ++---------- 1 files changed, 2 insertions(+), 10 deletions(-)
drivers/bluetooth/hci_vhci.c +2 -10
@@ -41,8 +41,6 @@#define VERSION "1.3" -static int minor =MISC_DYNAMIC_MINOR ; -struct vhci_data {struct hci_dev *hdev;
@@ -295,18 +293,12 @@ static int __init vhci_init(void) {BT_INFO ("Virtual HCI driver ver %s" ,VERSION ); -if (misc_register (&vhci_miscdev) <0 ) { -BT_ERR ("Can't register misc device with minor %d" , minor); -return -EIO ; - } - -return 0 ; +return misc_register (&vhci_miscdev); } static void __exit vhci_exit(void) { -if (misc_deregister (&vhci_miscdev) <0 ) -BT_ERR ("Can't unregister misc device with minor %d" , minor); +misc_deregister (&vhci_miscdev); }module_init (vhci_init);
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 4/18 Bluetooth: Fix miscdev ops owner for virtual driver
The /dev/vhci ops don't refer to the module and so it is possible to unload the module while the file descriptor is in use. This was an accidental removal after the cleanup.
--- drivers/bluetooth/hci_vhci.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-)
drivers/bluetooth/hci_vhci.c +1 -0
@@ -276,6 +276,7 @@ static int vhci_release(struct inode *inode, struct file *file) }static const struct file_operations vhci_fops = { + .owner =THIS_MODULE , .read = vhci_read, .write = vhci_write, .poll = vhci_poll,
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 5/18 Bluetooth: Implement raw output support for HIDP layer
From: Jiri Kosina <[email protected]>
Implement raw output callback which is used by hidraw to send raw data to the underlying device.
Without this patch, the userspace hidraw-based applications can't send output reports to HID Bluetooth devices.
--- net/bluetooth/hidp/core.c | 9 +++++++++ 1 files changed, 9 insertions(+), 0 deletions(-)
net/bluetooth/hidp/core.c +9 -0
@@ -280,6 +280,13 @@ static int hidp_send_report(struct hidp_session *session, struct hid_report *repreturn hidp_queue_report (session, buf, rsize); } +static int hidp_output_raw_report (struct hid_device *hid,unsigned char *data,size_t count) +{ +if (hidp_queue_report (hid->driver_data, data, count)) +return -ENOMEM ; +return count; +} +static void hidp_idle_timeout (unsigned long arg) {struct hidp_session *session = (struct hidp_session *) arg;
@@ -785,6 +792,8 @@ static int hidp_setup_hid(struct hidp_session *session, hid->dev.parent =hidp_get_device (session); hid->ll_driver = &hidp_hid_driver; + hid->hid_output_raw_report = hidp_output_raw_report; + err =hid_add_device (hid);if (err <0 ) goto failed;
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 6/18 Bluetooth: Return ENETDOWN when interface is down
Sending commands to a down interface results in a timeout while clearly it should just return ENETDOWN. When using the ioctls this works fine, but not when using the HCI sockets sendmsg interface.
--- net/bluetooth/hci_sock.c | 5 +++++ 1 files changed, 5 insertions(+), 0 deletions(-)
net/bluetooth/hci_sock.c +5 -0
@@ -414,6 +414,11 @@ static int hci_sock_sendmsg(struct kiocb *iocb, struct socket *sock, goto done; } +if (!test_bit (HCI_UP , &hdev->flags)) { + err = -ENETDOWN ; + goto done; + } +if (!(skb =bt_skb_send_alloc (sk, len, msg->msg_flags &MSG_DONTWAIT , &err))) goto done;
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 7/18 Bluetooth: Turn hci_recv_frame into an exported function
For future simplification it is important that the hci_recv_frame function is no longer an inline function. So move it into the module itself and export it.
--- include/net/bluetooth/hci_core.h | 23 +---------------------- net/bluetooth/hci_core.c | 23 +++++++++++++++++++++++ 2 files changed, 24 insertions(+), 22 deletions(-)
include/net/bluetooth/hci_core.h +1 -22
@@ -437,28 +437,7 @@ int hci_inquiry(void __user *arg);void hci_event_packet (struct hci_dev *hdev,struct sk_buff *skb); -/* Receive frame from HCI drivers */ -static inline int hci_recv_frame (struct sk_buff *skb) -{ -struct hci_dev *hdev = (struct hci_dev *) skb->dev; -if (!hdev || (!test_bit (HCI_UP , &hdev->flags) - && !test_bit (HCI_INIT , &hdev->flags))) { -kfree_skb (skb); -return -ENXIO ; - } - -/* Incomming skb */ -bt_cb (skb)->incoming =1 ; - -/* Time stamp */ -__net_timestamp (skb); - -/* Queue frame for rx task */ -skb_queue_tail (&hdev->rx_q, skb); -hci_sched_rx (hdev); -return 0 ; -} - +int hci_recv_frame (struct sk_buff *skb);int hci_recv_fragment (struct hci_dev *hdev,int type,void *data,int count);int hci_register_sysfs (struct hci_dev *hdev);
net/bluetooth/hci_core.c +23 -0
@@ -987,6 +987,29 @@ int hci_resume_dev(struct hci_dev *hdev) }EXPORT_SYMBOL (hci_resume_dev); +/* Receive frame from HCI drivers */ +int hci_recv_frame (struct sk_buff *skb) +{ +struct hci_dev *hdev = (struct hci_dev *) skb->dev; +if (!hdev || (!test_bit (HCI_UP , &hdev->flags) + && !test_bit (HCI_INIT , &hdev->flags))) { +kfree_skb (skb); +return -ENXIO ; + } + +/* Incomming skb */ +bt_cb (skb)->incoming =1 ; + +/* Time stamp */ +__net_timestamp (skb); + +/* Queue frame for rx task */ +skb_queue_tail (&hdev->rx_q, skb); +hci_sched_rx (hdev); +return 0 ; +} +EXPORT_SYMBOL (hci_recv_frame); +/* Receive packet type fragment */ #define __reassembly (hdev, type) ((hdev)->reassembly[(type) - 2])
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 8/18 Bluetooth: Unobfuscate tasklet_schedule usage
The tasklet schedule function helpers are just an obfuscation. So remove them and call the schedule functions directly.
--- include/net/bluetooth/hci_core.h | 16 ---------------- net/bluetooth/hci_core.c | 18 ++++++++++++------ net/bluetooth/hci_event.c | 6 +++--- net/bluetooth/hci_sock.c | 6 +++--- 4 files changed, 18 insertions(+), 28 deletions(-)
include/net/bluetooth/hci_core.h +0 -16
@@ -367,22 +367,6 @@ static inline void hci_conn_put(struct hci_conn *conn) } } -/* ----- HCI tasks ----- */ -static inline void hci_sched_cmd (struct hci_dev *hdev) -{ -tasklet_schedule (&hdev->cmd_task); -} - -static inline void hci_sched_rx (struct hci_dev *hdev) -{ -tasklet_schedule (&hdev->rx_task); -} - -static inline void hci_sched_tx (struct hci_dev *hdev) -{ -tasklet_schedule (&hdev->tx_task); -} -/* ----- HCI Devices ----- */ static inline void __hci_dev_put (struct hci_dev *d) {
net/bluetooth/hci_core.c +12 -6
@@ -193,8 +193,9 @@ static void hci_init_req(struct hci_dev *hdev, unsigned long opt)while ((skb =skb_dequeue (&hdev->driver_init))) {bt_cb (skb)->pkt_type =HCI_COMMAND_PKT ; skb->dev = (void *) hdev; +skb_queue_tail (&hdev->cmd_q, skb); -hci_sched_cmd (hdev); +tasklet_schedule (&hdev->cmd_task); }skb_queue_purge (&hdev->driver_init);
@@ -1005,7 +1006,8 @@ int hci_recv_frame(struct sk_buff *skb)/* Queue frame for rx task */ skb_queue_tail (&hdev->rx_q, skb); -hci_sched_rx (hdev); +tasklet_schedule (&hdev->rx_task); +return 0 ; }EXPORT_SYMBOL (hci_recv_frame);
@@ -1216,8 +1218,9 @@ int hci_send_cmd(struct hci_dev *hdev, __u16 opcode, __u32 plen, void *param)bt_cb (skb)->pkt_type =HCI_COMMAND_PKT ; skb->dev = (void *) hdev; +skb_queue_tail (&hdev->cmd_q, skb); -hci_sched_cmd (hdev); +tasklet_schedule (&hdev->cmd_task);return 0 ; }
@@ -1294,7 +1297,8 @@ int hci_send_acl(struct hci_conn *conn, struct sk_buff *skb, __u16 flags)spin_unlock_bh (&conn->data_q.lock); } -hci_sched_tx (hdev); +tasklet_schedule (&hdev->tx_task); +return 0 ; }EXPORT_SYMBOL (hci_send_acl);
@@ -1321,8 +1325,10 @@ int hci_send_sco(struct hci_conn *conn, struct sk_buff *skb) skb->dev = (void *) hdev;bt_cb (skb)->pkt_type =HCI_SCODATA_PKT ; +skb_queue_tail (&conn->data_q, skb); -hci_sched_tx (hdev); +tasklet_schedule (&hdev->tx_task); +return 0 ; }EXPORT_SYMBOL (hci_send_sco);
@@ -1635,7 +1641,7 @@ static void hci_cmd_task(unsigned long arg) hdev->cmd_last_tx = jiffies; }else {skb_queue_head (&hdev->cmd_q, skb); -hci_sched_cmd (hdev); +tasklet_schedule (&hdev->cmd_task); } } }
net/bluetooth/hci_event.c +3 -3
@@ -1320,7 +1320,7 @@ static inline void hci_cmd_complete_evt(struct hci_dev *hdev, struct sk_buff *skif (ev->ncmd) {atomic_set (&hdev->cmd_cnt,1 );if (!skb_queue_empty (&hdev->cmd_q)) -hci_sched_cmd (hdev); +tasklet_schedule (&hdev->cmd_task); } }
@@ -1386,7 +1386,7 @@ static inline void hci_cmd_status_evt(struct hci_dev *hdev, struct sk_buff *skb)if (ev->ncmd) {atomic_set (&hdev->cmd_cnt,1 );if (!skb_queue_empty (&hdev->cmd_q)) -hci_sched_cmd (hdev); +tasklet_schedule (&hdev->cmd_task); } }
@@ -1454,7 +1454,7 @@ static inline void hci_num_comp_pkts_evt(struct hci_dev *hdev, struct sk_buff *s } } -hci_sched_tx (hdev); +tasklet_schedule (&hdev->tx_task);tasklet_enable (&hdev->tx_task); }
net/bluetooth/hci_sock.c +3 -3
@@ -445,10 +445,10 @@ static int hci_sock_sendmsg(struct kiocb *iocb, struct socket *sock,if (test_bit (HCI_RAW , &hdev->flags) || (ogf ==0x3f )) {skb_queue_tail (&hdev->raw_q, skb); -hci_sched_tx (hdev); +tasklet_schedule (&hdev->tx_task); }else {skb_queue_tail (&hdev->cmd_q, skb); -hci_sched_cmd (hdev); +tasklet_schedule (&hdev->cmd_task); } }else {if (!capable (CAP_NET_RAW )) {
@@ -457,7 +457,7 @@ static int hci_sock_sendmsg(struct kiocb *iocb, struct socket *sock, }skb_queue_tail (&hdev->raw_q, skb); -hci_sched_tx (hdev); +tasklet_schedule (&hdev->tx_task); } err = len;
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 9/18 Bluetooth: Fix handling of BNEP setup connection requests
From: Vikram Kandukuri <[email protected]>
According to BNEP test specification the proper response should be sent for a setup connection request message after the BNEP connection setup has been completed.
--- net/bluetooth/bnep/core.c | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-)
net/bluetooth/bnep/core.c +4 -1
@@ -230,7 +230,6 @@ static int bnep_rx_control(struct bnep_session *s, void *data, int len)switch (cmd) {case BNEP_CMD_NOT_UNDERSTOOD : -case BNEP_SETUP_CONN_REQ :case BNEP_SETUP_CONN_RSP :case BNEP_FILTER_NET_TYPE_RSP :case BNEP_FILTER_MULTI_ADDR_RSP :
@@ -245,6 +244,10 @@ static int bnep_rx_control(struct bnep_session *s, void *data, int len) err =bnep_ctrl_set_mcfilter (s, data, len);break ; +case BNEP_SETUP_CONN_REQ : + err =bnep_send_rsp (s,BNEP_SETUP_CONN_RSP ,BNEP_CONN_NOT_ALLOWED ); +break ; +default : {u8 pkt[3 ]; pkt[0 ] =BNEP_CONTROL ;
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 10/18 Bluetooth: Initialize variables and timers for both channel's sides
From: Gustavo F. Padovan <[email protected]>
Fix ERTM's full-duplex channel to work as specified by ERTM spec. ERTM needs to handle state vars, timers and counters to send and receive I-frames(the data frames), i.e., for both sides of data communication. We initialize all of them to the default values here. Full-duplex channel is a mandatory feature of ERTM spec.
--- net/bluetooth/l2cap.c | 54 ++++++++++++++++++++++++++++++++---------------- 1 files changed, 36 insertions(+), 18 deletions(-)
net/bluetooth/l2cap.c +36 -18
@@ -2173,6 +2173,21 @@ static void l2cap_add_conf_opt(void **ptr, u8 type, u8 len, unsigned long val) *ptr +=L2CAP_CONF_OPT_SIZE + len; } +static inline void l2cap_ertm_init (struct sock *sk) +{ +l2cap_pi (sk)->expected_ack_seq =0 ; +l2cap_pi (sk)->unacked_frames =0 ; +l2cap_pi (sk)->buffer_seq =0 ; +l2cap_pi (sk)->num_to_ack =0 ; + +setup_timer (&l2cap_pi (sk)->retrans_timer, + l2cap_retrans_timeout, (unsigned long ) sk); +setup_timer (&l2cap_pi (sk)->monitor_timer, + l2cap_monitor_timeout, (unsigned long ) sk); + +__skb_queue_head_init (SREJ_QUEUE (sk)); +} +static int l2cap_mode_supported (__u8 mode,__u32 feat_mask) {u32 local_feat_mask = l2cap_feat_mask;
@@ -2761,17 +2776,13 @@ static inline int l2cap_config_req(struct l2cap_conn *conn, struct l2cap_cmd_hdrl2cap_pi (sk)->fcs =L2CAP_FCS_CRC16 ; sk->sk_state =BT_CONNECTED ; -l2cap_pi (sk)->next_tx_seq =0 ; -l2cap_pi (sk)->expected_ack_seq =0 ; -l2cap_pi (sk)->unacked_frames =0 ; - -setup_timer (&l2cap_pi (sk)->retrans_timer, - l2cap_retrans_timeout, (unsigned long ) sk); -setup_timer (&l2cap_pi (sk)->monitor_timer, - l2cap_monitor_timeout, (unsigned long ) sk); +l2cap_pi (sk)->next_tx_seq =0 ; +l2cap_pi (sk)->expected_tx_seq =0 ;__skb_queue_head_init (TX_QUEUE (sk)); -__skb_queue_head_init (SREJ_QUEUE (sk)); +if (l2cap_pi (sk)->mode ==L2CAP_MODE_ERTM ) +l2cap_ertm_init (sk); +l2cap_chan_ready (sk); goto unlock; }
@@ -2850,11 +2861,12 @@ static inline int l2cap_config_rsp(struct l2cap_conn *conn, struct l2cap_cmd_hdrl2cap_pi (sk)->fcs =L2CAP_FCS_CRC16 ; sk->sk_state =BT_CONNECTED ; +l2cap_pi (sk)->next_tx_seq =0 ;l2cap_pi (sk)->expected_tx_seq =0 ; -l2cap_pi (sk)->buffer_seq =0 ; -l2cap_pi (sk)->num_to_ack =0 ;__skb_queue_head_init (TX_QUEUE (sk)); -__skb_queue_head_init (SREJ_QUEUE (sk)); +if (l2cap_pi (sk)->mode ==L2CAP_MODE_ERTM ) +l2cap_ertm_init (sk); +l2cap_chan_ready (sk); }
@@ -2886,9 +2898,12 @@ static inline int l2cap_disconnect_req(struct l2cap_conn *conn, struct l2cap_cmd sk->sk_shutdown =SHUTDOWN_MASK ;skb_queue_purge (TX_QUEUE (sk)); -skb_queue_purge (SREJ_QUEUE (sk)); -del_timer (&l2cap_pi (sk)->retrans_timer); -del_timer (&l2cap_pi (sk)->monitor_timer); + +if (l2cap_pi (sk)->mode ==L2CAP_MODE_ERTM ) { +skb_queue_purge (SREJ_QUEUE (sk)); +del_timer (&l2cap_pi (sk)->retrans_timer); +del_timer (&l2cap_pi (sk)->monitor_timer); + }l2cap_chan_del (sk,ECONNRESET );bh_unlock_sock (sk);
@@ -2913,9 +2928,12 @@ static inline int l2cap_disconnect_rsp(struct l2cap_conn *conn, struct l2cap_cmdreturn 0 ;skb_queue_purge (TX_QUEUE (sk)); -skb_queue_purge (SREJ_QUEUE (sk)); -del_timer (&l2cap_pi (sk)->retrans_timer); -del_timer (&l2cap_pi (sk)->monitor_timer); + +if (l2cap_pi (sk)->mode ==L2CAP_MODE_ERTM ) { +skb_queue_purge (SREJ_QUEUE (sk)); +del_timer (&l2cap_pi (sk)->retrans_timer); +del_timer (&l2cap_pi (sk)->monitor_timer); + }l2cap_chan_del (sk,0 );bh_unlock_sock (sk);
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 11/18 Bluetooth: Fix unset of SrejActioned flag
From: Gustavo F. Padovan <[email protected]>
SrejActioned is a flag that when set prevents local side to retransmit a I-frame(the data frame) already retransmitted. The local entity can retransmit again only when it receives a SREJ frame with the F-bit set. SREJ frame - Selective Reject frame - is sent when an entity wants the retransmission of a specific I-frame that was lost or corrupted. This bug can put ERTM in an unknown state once the entity can't retransmit.
A frame with the Final bit set is expected when the local side sends a frame with the Poll bit set due to a local busy condition or a retransmission timer expired. (Receipt of P-bit shall always be replied by a frame with the F-bit set).pi->conn_state keeps informations about many ERTM flags including SrejActioned.
--- net/bluetooth/l2cap.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
net/bluetooth/l2cap.c +1 -1
@@ -3443,7 +3443,7 @@ static inline int l2cap_data_channel_sframe(struct sock *sk, u16 rx_control, str }else if (rx_control &L2CAP_CTRL_FINAL ) {if ((pi->conn_state &L2CAP_CONN_SREJ_ACT ) && pi->srej_save_reqseq == tx_seq) - pi->srej_save_reqseq &= ~L2CAP_CONN_SREJ_ACT ; + pi->conn_state &= ~L2CAP_CONN_SREJ_ACT ;else l2cap_retransmit_frame (sk, tx_seq); }
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 12/18 Bluetooth: Fix sending ReqSeq on I-frames
From: Gustavo F. Padovan <[email protected]>
As specified by ERTM spec an ERTM channel can acknowledge received I-frames(the data frames) by sending an I-frame with the proper ReqSeq value (i.e. ReqSeq is set to BufferSeq). Until now we aren't setting the ReqSeq value on I-frame control bits. That way we can save sending S-frames(Supervise frames) only to acknowledge receipt of I-frames. It is very helpful to the full-duplex channel.
ReqSeq is the packet sequence number sent in an acknowledgement frame to acknowledge receipt of frames up to (ReqSeq - 1).
BufferSeq controls the receiver buffer, it is used to delay acknowledgement of new frames to not cause buffer overflow. BufferSeq value is not increased until frames are pulled by reassembly function.
--- include/net/bluetooth/l2cap.h | 1 - net/bluetooth/l2cap.c | 8 ++++++-- 2 files changed, 6 insertions(+), 3 deletions(-)
include/net/bluetooth/l2cap.h +0 -1
@@ -324,7 +324,6 @@ struct l2cap_pinfo {__u8 next_tx_seq;__u8 expected_ack_seq; -__u8 req_seq;__u8 expected_tx_seq;__u8 buffer_seq;__u8 buffer_seq_srej;
net/bluetooth/l2cap.c +6 -2
@@ -1333,7 +1333,7 @@ static int l2cap_retransmit_frame(struct sock *sk, u8 tx_seq) tx_skb =skb_clone (skb,GFP_ATOMIC );bt_cb (skb)->retries++; control =get_unaligned_le16 (tx_skb->data +L2CAP_HDR_SIZE ); - control |= (pi->req_seq <<L2CAP_CTRL_REQSEQ_SHIFT ) + control |= (pi->buffer_seq <<L2CAP_CTRL_REQSEQ_SHIFT ) | (tx_seq <<L2CAP_CTRL_TXSEQ_SHIFT );put_unaligned_le16 (control, tx_skb->data +L2CAP_HDR_SIZE );
@@ -1375,7 +1375,7 @@ static int l2cap_ertm_send(struct sock *sk)bt_cb (skb)->retries++; control =get_unaligned_le16 (tx_skb->data +L2CAP_HDR_SIZE ); - control |= (pi->req_seq <<L2CAP_CTRL_REQSEQ_SHIFT ) + control |= (pi->buffer_seq <<L2CAP_CTRL_REQSEQ_SHIFT ) | (pi->next_tx_seq <<L2CAP_CTRL_TXSEQ_SHIFT );put_unaligned_le16 (control, tx_skb->data +L2CAP_HDR_SIZE );
@@ -3298,12 +3298,16 @@ static inline int l2cap_data_channel_iframe(struct sock *sk, u16 rx_control, str {struct l2cap_pinfo *pi =l2cap_pi (sk);u8 tx_seq =__get_txseq (rx_control); +u8 req_seq =__get_reqseq (rx_control);u16 tx_control =0 ;u8 sar = rx_control >>L2CAP_CTRL_SAR_SHIFT ;int err =0 ;BT_DBG ("sk %p rx_control 0x%4.4x len %d" , sk, rx_control, skb->len); + pi->expected_ack_seq = req_seq; +l2cap_drop_acked_frames (sk); +if (tx_seq == pi->expected_tx_seq) goto expected;
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 13/18 Bluetooth: Implement RejActioned flag
From: Gustavo F. Padovan <[email protected]>
RejActioned is used to prevent retransmission when a entity is on the WAIT_F state, i.e., waiting for a frame with F-bit set due local busy condition or a expired retransmission timer. (When these two events raise they send a frame with the Poll bit set and enters in the WAIT_F state to wait for a frame with the Final bit set.)
The local entity doesn't send I-frames(the data frames) until the receipt of a frame with F-bit set. When that happens it also set RejActioned to false. RejActioned is a mandatory feature of ERTM spec.
--- include/net/bluetooth/l2cap.h | 1 + net/bluetooth/l2cap.c | 38 +++++++++++++++++++++++++++++++++++--- 2 files changed, 36 insertions(+), 3 deletions(-)
include/net/bluetooth/l2cap.h +1 -0
@@ -374,6 +374,7 @@ struct l2cap_pinfo {#define L2CAP_CONN_SEND_PBIT 0x10#define L2CAP_CONN_REMOTE_BUSY 0x20#define L2CAP_CONN_LOCAL_BUSY 0x40 +#define L2CAP_CONN_REJ_ACT 0x80#define __mod_retrans_timer () mod_timer(&l2cap_pi(sk)->retrans_timer, \ jiffies + msecs_to_jiffies(L2CAP_DEFAULT_RETRANS_TO));
net/bluetooth/l2cap.c +35 -3
@@ -3362,6 +3362,16 @@ expected:return 0 ; } +if (rx_control &L2CAP_CTRL_FINAL ) { +if (pi->conn_state &L2CAP_CONN_REJ_ACT ) + pi->conn_state &= ~L2CAP_CONN_REJ_ACT ; +else { + sk->sk_send_head =TX_QUEUE (sk)->next; + pi->next_tx_seq = pi->expected_ack_seq; +l2cap_ertm_send (sk); + } + } + pi->buffer_seq = (pi->buffer_seq +1 ) %64 ; err =l2cap_sar_reassembly_sdu (sk, skb, rx_control);
@@ -3398,6 +3408,14 @@ static inline int l2cap_data_channel_sframe(struct sock *sk, u16 rx_control, str pi->expected_ack_seq = tx_seq;l2cap_drop_acked_frames (sk); +if (pi->conn_state &L2CAP_CONN_REJ_ACT ) + pi->conn_state &= ~L2CAP_CONN_REJ_ACT ; +else { + sk->sk_send_head =TX_QUEUE (sk)->next; + pi->next_tx_seq = pi->expected_ack_seq; +l2cap_ertm_send (sk); + } +if (!(pi->conn_state &L2CAP_CONN_WAIT_F ))break ;
@@ -3425,10 +3443,24 @@ static inline int l2cap_data_channel_sframe(struct sock *sk, u16 rx_control, str pi->expected_ack_seq =__get_reqseq (rx_control);l2cap_drop_acked_frames (sk); - sk->sk_send_head =TX_QUEUE (sk)->next; - pi->next_tx_seq = pi->expected_ack_seq; +if (rx_control &L2CAP_CTRL_FINAL ) { +if (pi->conn_state &L2CAP_CONN_REJ_ACT ) + pi->conn_state &= ~L2CAP_CONN_REJ_ACT ; +else { + sk->sk_send_head =TX_QUEUE (sk)->next; + pi->next_tx_seq = pi->expected_ack_seq; +l2cap_ertm_send (sk); + } + }else { + sk->sk_send_head =TX_QUEUE (sk)->next; + pi->next_tx_seq = pi->expected_ack_seq; +l2cap_ertm_send (sk); -l2cap_ertm_send (sk); +if (pi->conn_state &L2CAP_CONN_WAIT_F ) { + pi->srej_save_reqseq = tx_seq; + pi->conn_state |=L2CAP_CONN_REJ_ACT ; + } + }break ;
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 14/18 Bluetooth: Fix 'SendRRorRNR' to send the ReqSeq value
From: Gustavo F. Padovan <[email protected]>
SendRRorRNR needs to acknowledge received I-frames (actually every packet needs to acknowledge received I-frames by sending the proper packet sequence number), so ReqSeq is set to the next I-frame number sequence to be pulled by the reassembly function.
SendRRorRNR tells the remote side about local busy conditions, it sends a Receiver Ready frame if local busy is false or a Receiver Not Ready if local busy is true.
ReqSeq is the packet's field to send the number of the acknowledged packets.
--- net/bluetooth/l2cap.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)
net/bluetooth/l2cap.c +2 -0
@@ -373,6 +373,8 @@ static inline int l2cap_send_rr_or_rnr(struct l2cap_pinfo *pi, u16 control)else control |=L2CAP_SUPER_RCV_READY ; + control |= pi->buffer_seq <<L2CAP_CTRL_REQSEQ_SHIFT ; +return l2cap_send_sframe (pi, control); }
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 15/18 Bluetooth: Enable auto sleep mode for btmrvl driver
From: Amitkumar Karwar <[email protected]>
The auto sleep mode for btmrvl driver is not enabled by default. This patch enables auto sleep mode when card is probed.
--- drivers/bluetooth/btmrvl_drv.h | 1 + drivers/bluetooth/btmrvl_main.c | 55 +++++++++++++++++++++++--------------- drivers/bluetooth/btmrvl_sdio.c | 2 + 3 files changed, 36 insertions(+), 22 deletions(-)
drivers/bluetooth/btmrvl_drv.h +1 -0
@@ -131,6 +131,7 @@ void btmrvl_check_evtpkt(struct btmrvl_private *priv, struct sk_buff *skb);int btmrvl_process_event (struct btmrvl_private *priv,struct sk_buff *skb);int btmrvl_send_module_cfg_cmd (struct btmrvl_private *priv,int subcmd); +int btmrvl_enable_ps (struct btmrvl_private *priv);int btmrvl_prepare_command (struct btmrvl_private *priv);#ifdef CONFIG_DEBUG_FS
drivers/bluetooth/btmrvl_main.c +33 -22
67 lines not on this page · raw message
drivers/bluetooth/btmrvl_sdio.c +2 -0
@@ -930,6 +930,8 @@ static int btmrvl_sdio_probe(struct sdio_func *func, priv->hw_wakeup_firmware = btmrvl_sdio_wakeup_fw;btmrvl_send_module_cfg_cmd (priv,MODULE_BRINGUP_REQ ); + priv->btmrvl_dev.psmode =1 ; +btmrvl_enable_ps (priv);return 0 ;
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 16/18 Bluetooth: Add L2CAP option for max transmit value
For testing purposes it is important to modify the max transmit value.
--- net/bluetooth/l2cap.c | 6 +++++- 1 files changed, 5 insertions(+), 1 deletions(-)
net/bluetooth/l2cap.c +5 -1
@@ -54,6 +54,7 @@#define VERSION "2.14"static int enable_ertm =0 ; +static int max_transmit =L2CAP_DEFAULT_MAX_TX ;static u32 l2cap_feat_mask =L2CAP_FEAT_FIXED_CHAN ;static u8 l2cap_fixed_chan[8 ] = {0x02 , };
@@ -2253,7 +2254,7 @@ done:case L2CAP_MODE_ERTM : rfc.mode =L2CAP_MODE_ERTM ; rfc.txwin_size =L2CAP_DEFAULT_TX_WINDOW ; - rfc.max_transmit =L2CAP_DEFAULT_MAX_TX ; + rfc.max_transmit = max_transmit; rfc.retrans_timeout =0 ; rfc.monitor_timeout =0 ; rfc.max_pdu_size =cpu_to_le16 (L2CAP_DEFAULT_MAX_PDU_SIZE );
@@ -4060,6 +4061,9 @@ module_exit(l2cap_exit);module_param (enable_ertm, bool,0644 );MODULE_PARM_DESC (enable_ertm,"Enable enhanced retransmission mode" ); +module_param (max_transmit, uint,0644 ); +MODULE_PARM_DESC (max_transmit,"Max transmit value (default = 3)" ); +MODULE_AUTHOR ("Marcel Holtmann <[email protected]>" );MODULE_DESCRIPTION ("Bluetooth L2CAP ver " VERSION );MODULE_VERSION (VERSION );
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 17/18 Bluetooth: Add RFCOMM option to use L2CAP ERTM mode
By default the RFCOMM layer would still use L2CAP basic mode. For testing purposes this option enables RFCOMM to select enhanced retransmission mode.
--- net/bluetooth/rfcomm/core.c | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-)
net/bluetooth/rfcomm/core.c +6 -0
@@ -51,6 +51,7 @@static int disable_cfc =0 ;static int channel_mtu =-1 ;static unsigned int l2cap_mtu =RFCOMM_MAX_L2CAP_MTU ; +static int l2cap_ertm =0 ;static struct task_struct *rfcomm_thread;
@@ -702,6 +703,8 @@ static struct rfcomm_session *rfcomm_session_create(bdaddr_t *src, bdaddr_t *dst sk = sock->sk;lock_sock (sk);l2cap_pi (sk)->imtu = l2cap_mtu; +if (l2cap_ertm) +l2cap_pi (sk)->mode =L2CAP_MODE_ERTM ;release_sock (sk); s =rfcomm_session_add (sock,BT_BOUND );
@@ -2185,6 +2188,9 @@ MODULE_PARM_DESC(channel_mtu, "Default MTU for the RFCOMM channel");module_param (l2cap_mtu, uint,0644 );MODULE_PARM_DESC (l2cap_mtu,"Default MTU for the L2CAP connection" ); +module_param (l2cap_ertm, bool,0644 ); +MODULE_PARM_DESC (l2cap_ertm,"Use L2CAP ERTM mode for connection" ); +MODULE_AUTHOR ("Marcel Holtmann <[email protected]>" );MODULE_DESCRIPTION ("Bluetooth RFCOMM ver " VERSION );MODULE_VERSION (VERSION );
Signature · 1 line
1.6.5.2
Marcel Holtmann
maintainerPATCH 18/18 Bluetooth: Declare MODULE_FIRMWARE for Marvell SDIO driver
From: Ben Hutchings <[email protected]>
--- drivers/bluetooth/btmrvl_sdio.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)
drivers/bluetooth/btmrvl_sdio.c +2 -0
@@ -1003,3 +1003,5 @@ MODULE_AUTHOR("Marvell International Ltd.");MODULE_DESCRIPTION ("Marvell BT-over-SDIO driver ver " VERSION );MODULE_VERSION (VERSION );MODULE_LICENSE ("GPL v2" ); +MODULE_FIRMWARE ("sd8688_helper.bin" ); +MODULE_FIRMWARE ("sd8688.bin" );
Signature · 1 line
1.6.5.2
David Miller
maintainerFrom: Marcel Holtmann <[email protected]>
Date: Thu, 3 Dec 2009 19:45:37 +0100
Quoted 6 lines
these are the patches for the 2.6.33 merge window. Sorry for being a bit late, but I got dragged into some other work. Besides the ongoing work on the L2CAP ERTM support it is only bug fixes and cleanups.
I did base this against your net-next tree. If you want me to rebase it, then let me know.
Looks good Marcel, all pulled into net-next-2.6, thanks!