[PATCH 0/2] v2: Bluetooth driver for TI_ST

[PATCH 0/2] v2: Bluetooth driver for TI_ST

[pavan_savoy]

From: Pavan Savoy <pavan_savoy@ti.com>

Marcel,

I have taken care of the changes you've mentioned, Now the
bluetooth driver is also a platform driver and it requires a
platform device and this platform driver's probe will do the
registration to HCI.

other changes include the usage of BT_DBG instead of pr_*
and removal of piece of code which was redundant.

Please review and comment...

Pavan Savoy (2):
  drivers:bluetooth: TI_ST bluetooth driver
  drivers:bluetooth: Kconfig & Makefile for TI BT

 drivers/bluetooth/Kconfig  |   10 +
 drivers/bluetooth/Makefile |    1 +
 drivers/bluetooth/bt_ti.c  |  489 ++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 500 insertions(+), 0 deletions(-)
 create mode 100644 drivers/bluetooth/bt_ti.c

[PATCH 1/2] drivers:bluetooth: TI_ST bluetooth driver

[pavan_savoy]

From: Pavan Savoy <pavan_savoy@ti.com>

This is the bluetooth protocol driver for the TI WiLink7 chipsets.
Texas Instrument's WiLink chipsets combine wireless technologies
like BT, FM, GPS and WLAN onto a single chip.

This Bluetooth driver works on top of the TI_ST shared transport
line discipline driver which also allows other drivers like
FM V4L2 and GPS character driver to make use of the same UART interface.

Signed-off-by: Pavan Savoy <pavan_savoy@ti.com>
---
 drivers/bluetooth/bt_ti.c |  489 +++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 489 insertions(+), 0 deletions(-)
 create mode 100644 drivers/bluetooth/bt_ti.c

diff --git a/drivers/bluetooth/bt_ti.c b/drivers/bluetooth/bt_ti.c
new file mode 100644
index 0000000..dffbb56
--- /dev/null
+++ b/drivers/bluetooth/bt_ti.c
@@ -0,0 +1,489 @@
+/*
+ *  Texas Instrument's Bluetooth Driver For Shared Transport.
+ *
+ *  Bluetooth Driver acts as interface between HCI CORE and
+ *  TI Shared Transport Layer.
+ *
+ *  Copyright (C) 2009-2010 Texas Instruments
+ *  Author: Raja Mani <raja_mani@ti.com>
+ *	Pavan Savoy <pavan_savoy@ti.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <linux/platform_device.h>
+#include <net/bluetooth/bluetooth.h>
+#include <net/bluetooth/hci_core.h>
+
+#include <linux/ti_wilink_st.h>
+
+/* Bluetooth Driver Version */
+#define VERSION               "1.0"
+
+/* Defines number of seconds to wait for reg completion
+ * callback getting called from ST (in case,registration
+ * with ST returns PENDING status)
+ */
+#define BT_REGISTER_TIMEOUT   6000	/* 6 sec */
+
+/* BT driver's local status */
+#define BT_DRV_RUNNING        0
+#define BT_ST_REGISTERED      1
+
+/**
+ * struct hci_st - BT driver operation structure
+ * @hdev: hci device pointer which binds to bt driver
+ * @flags: used locally,to maintain various BT driver status
+ * @streg_cbdata: to hold ST registration callback status
+ * @st_write: write function pointer of ST driver
+ * @wait_for_btdrv_reg_completion - completion sync between hci_st_open
+ *	and hci_st_registration_completion_cb.
+ */
+struct hci_st {
+	struct hci_dev *hdev;
+	unsigned long flags;
+	char streg_cbdata;
+	long (*st_write) (struct sk_buff *);
+	struct completion wait_for_btdrv_reg_completion;
+};
+
+static int reset;
+
+/* Increments HCI counters based on pocket ID (cmd,acl,sco) */
+static inline void hci_st_tx_complete(struct hci_st *hst, int pkt_type)
+{
+	struct hci_dev *hdev;
+	hdev = hst->hdev;
+
+	/* Update HCI stat counters */
+	switch (pkt_type) {
+	case HCI_COMMAND_PKT:
+		hdev->stat.cmd_tx++;
+		break;
+
+	case HCI_ACLDATA_PKT:
+		hdev->stat.acl_tx++;
+		break;
+
+	case HCI_SCODATA_PKT:
+		hdev->stat.cmd_tx++;
+		break;
+	}
+}
+
+/* ------- Interfaces to Shared Transport ------ */
+
+/* Called by ST layer to indicate protocol registration completion
+ * status.hci_st_open() function will wait for signal from this
+ * API when st_register() function returns ST_PENDING.
+ */
+static void hci_st_registration_completion_cb(void *priv_data, char data)
+{
+	struct hci_st *lhst = (struct hci_st *)priv_data;
+	/* hci_st_open() function needs value of 'data' to know
+	 * the registration status(success/fail),So have a back
+	 * up of it.
+	 */
+	lhst->streg_cbdata = data;
+
+	/* Got a feedback from ST for BT driver registration
+	 * request.Wackup hci_st_open() function to continue
+	 * it's open operation.
+	 */
+	complete(&lhst->wait_for_btdrv_reg_completion);
+}
+
+/* Called by Shared Transport layer when receive data is
+ * available */
+static long hci_st_receive(void *priv_data, struct sk_buff *skb)
+{
+	int err;
+	int len;
+	struct hci_st *lhst = (struct hci_st *)priv_data;
+
+	err = 0;
+	len = 0;
+
+	if (skb == NULL) {
+		BT_ERR("Invalid SKB received from ST");
+		return -EFAULT;
+	}
+	if (!lhst) {
+		kfree_skb(skb);
+		BT_ERR("Invalid hci_st memory,freeing SKB");
+		return -EFAULT;
+	}
+	if (!test_bit(BT_DRV_RUNNING, &lhst->flags)) {
+		kfree_skb(skb);
+		BT_ERR("Device is not running,freeing SKB");
+		return -EINVAL;
+	}
+
+	len = skb->len;
+	skb->dev = (struct net_device *)lhst->hdev;
+
+	/* Forward skb to HCI CORE layer */
+	err = hci_recv_frame(skb);
+	if (err) {
+		kfree_skb(skb);
+		BT_ERR("Unable to push skb to HCI CORE(%d),freeing SKB",
+				err);
+		return err;
+	}
+	lhst->hdev->stat.byte_rx += len;
+
+	return 0;
+}
+
+/* ------- Interfaces to HCI layer ------ */
+
+/* Called from HCI core to initialize the device */
+static int hci_st_open(struct hci_dev *hdev)
+{
+	static struct st_proto_s hci_st_proto;
+	unsigned long timeleft;
+	struct hci_st *hst;
+	int err;
+	err = 0;
+
+	BT_DBG("%s %p", hdev->name, hdev);
+	hst = hdev->driver_data;
+
+	/* Populate BT driver info required by ST */
+	memset(&hci_st_proto, 0, sizeof(hci_st_proto));
+
+	/* BT driver ID */
+	hci_st_proto.type = ST_BT;
+
+	/* Receive function which called from ST */
+	hci_st_proto.recv = hci_st_receive;
+
+	/* Packet match function may used in future */
+	hci_st_proto.match_packet = NULL;
+
+	/* Callback to be called when registration is pending */
+	hci_st_proto.reg_complete_cb = hci_st_registration_completion_cb;
+
+	/* This is write function pointer of ST. BT driver will make use of this
+	 * for sending any packets to chip. ST will assign and give to us, so
+	 * make it as NULL */
+	hci_st_proto.write = NULL;
+
+	/* send in the hst to be received at registration complete callback
+	 * and during st's receive
+	 */
+	hci_st_proto.priv_data = hst;
+
+	/* Register with ST layer */
+	err = st_register(&hci_st_proto);
+	if (err == -EINPROGRESS) {
+		/* Prepare wait-for-completion handler data structures.
+		 * Needed to syncronize this and st_registration_completion_cb()
+		 * functions.
+		 */
+		init_completion(&hst->wait_for_btdrv_reg_completion);
+
+		/* Reset ST registration callback status flag , this value
+		 * will be updated in hci_st_registration_completion_cb()
+		 * function whenever it called from ST driver.
+		 */
+		hst->streg_cbdata = -EINPROGRESS;
+
+		/* ST is busy with other protocol registration(may be busy with
+		 * firmware download).So,Wait till the registration callback
+		 * (passed as a argument to st_register() function) getting
+		 * called from ST.
+		 */
+		BT_DBG(" %s waiting for reg completion signal from ST",
+				__func__);
+
+		timeleft =
+			wait_for_completion_timeout
+			(&hst->wait_for_btdrv_reg_completion,
+			 msecs_to_jiffies(BT_REGISTER_TIMEOUT));
+		if (!timeleft) {
+			BT_ERR("Timeout(%d sec),didn't get reg"
+					"completion signal from ST",
+					BT_REGISTER_TIMEOUT / 1000);
+			return -ETIMEDOUT;
+		}
+
+		/* Is ST registration callback called with ERROR value? */
+		if (hst->streg_cbdata != 0) {
+			BT_ERR("ST reg completion CB called with invalid"
+					"status %d", hst->streg_cbdata);
+			return -EAGAIN;
+		}
+		err = 0;
+	} else if (err == -1) {
+		BT_ERR("st_register failed %d", err);
+		return -EAGAIN;
+	}
+
+	/* Do we have proper ST write function? */
+	if (hci_st_proto.write != NULL) {
+		/* We need this pointer for sending any Bluetooth pkts */
+		hst->st_write = hci_st_proto.write;
+	} else {
+		BT_ERR("failed to get ST write func pointer");
+
+		/* Undo registration with ST */
+		err = st_unregister(ST_BT);
+		if (err < 0)
+			BT_ERR("st_unregister failed %d", err);
+
+		hst->st_write = NULL;
+		return -EAGAIN;
+	}
+
+	/* Registration with ST layer is completed successfully,
+	 * now chip is ready to accept commands from HCI CORE.
+	 * Mark HCI Device flag as RUNNING
+	 */
+	set_bit(HCI_RUNNING, &hdev->flags);
+
+	/* Registration with ST successful */
+	set_bit(BT_ST_REGISTERED, &hst->flags);
+
+	return err;
+}
+
+/* Close device */
+static int hci_st_close(struct hci_dev *hdev)
+{
+	int err;
+	struct hci_st *hst;
+	err = 0;
+
+	hst = hdev->driver_data;
+	/* Unregister from ST layer */
+	if (test_and_clear_bit(BT_ST_REGISTERED, &hst->flags)) {
+		err = st_unregister(ST_BT);
+		if (err != 0) {
+			BT_ERR("st_unregister failed %d", err);
+			return -EBUSY;
+		}
+	}
+
+	hst->st_write = NULL;
+
+	/* ST layer would have moved chip to inactive state.
+	 * So,clear HCI device RUNNING flag.
+	 */
+	if (!test_and_clear_bit(HCI_RUNNING, &hdev->flags))
+		return 0;
+
+	return err;
+}
+
+/* Called from HCI CORE , Sends frames to Shared Transport */
+static int hci_st_send_frame(struct sk_buff *skb)
+{
+	struct hci_dev *hdev;
+	struct hci_st *hst;
+	long len;
+
+	if (skb == NULL) {
+		BT_ERR("Invalid skb received from HCI CORE");
+		return -ENOMEM;
+	}
+	hdev = (struct hci_dev *)skb->dev;
+	if (!hdev) {
+		BT_ERR("SKB received for invalid HCI Device (hdev=NULL)");
+		return -ENODEV;
+	}
+	if (!test_bit(HCI_RUNNING, &hdev->flags)) {
+		BT_ERR("Device is not running");
+		return -EBUSY;
+	}
+
+	hst = (struct hci_st *)hdev->driver_data;
+
+	/* Prepend skb with frame type */
+	memcpy(skb_push(skb, 1), &bt_cb(skb)->pkt_type, 1);
+
+	BT_DBG(" %s: type %d len %d", hdev->name, bt_cb(skb)->pkt_type,
+			skb->len);
+
+	/* Insert skb to shared transport layer's transmit queue.
+	 * Freeing skb memory is taken care in shared transport layer,
+	 * so don't free skb memory here.
+	 */
+	if (!hst->st_write) {
+		kfree_skb(skb);
+		BT_ERR(" Can't write to ST, st_write null?");
+		return -EAGAIN;
+	}
+	len = hst->st_write(skb);
+	if (len < 0) {
+		/* Something went wrong in st write , free skb memory */
+		kfree_skb(skb);
+		BT_ERR(" ST write failed (%ld)", len);
+		return -EAGAIN;
+	}
+
+	/* ST accepted our skb. So, Go ahead and do rest */
+	hdev->stat.byte_tx += len;
+	hci_st_tx_complete(hst, bt_cb(skb)->pkt_type);
+
+	return 0;
+}
+
+static void hci_st_destruct(struct hci_dev *hdev)
+{
+	if (!hdev) {
+		BT_ERR("Destruct called with invalid HCI Device"
+				"(hdev=NULL)");
+		return;
+	}
+
+	BT_DBG("%s", hdev->name);
+
+	/* free hci_st memory */
+	if (hdev->driver_data != NULL)
+		kfree(hdev->driver_data);
+
+	return;
+}
+
+/* Creates new HCI device */
+static int hci_st_register_dev(struct hci_st *hst)
+{
+	struct hci_dev *hdev;
+
+	/* Initialize and register HCI device */
+	hdev = hci_alloc_dev();
+	if (!hdev) {
+		BT_ERR("Can't allocate HCI device");
+		return -ENOMEM;
+	}
+	BT_DBG(" HCI device allocated. hdev= %p", hdev);
+
+	hst->hdev = hdev;
+	hdev->bus = HCI_UART;
+	hdev->driver_data = hst;
+	hdev->open = hci_st_open;
+	hdev->close = hci_st_close;
+	hdev->flush = NULL;
+	hdev->send = hci_st_send_frame;
+	hdev->destruct = hci_st_destruct;
+	hdev->owner = THIS_MODULE;
+
+	if (reset)
+		set_bit(HCI_QUIRK_NO_RESET, &hdev->quirks);
+
+	if (hci_register_dev(hdev) < 0) {
+		BT_ERR("Can't register HCI device");
+		hci_free_dev(hdev);
+		return -ENODEV;
+	}
+
+	BT_DBG(" HCI device registered. hdev= %p", hdev);
+	return 0;
+}
+
+
+static int bt_ti_probe(struct platform_device *pdev)
+{
+	int err;
+	static struct hci_st *hst;
+	err = 0;
+
+	BT_DBG(" Bluetooth Driver Version %s", VERSION);
+
+	/* Allocate local resource memory */
+	hst = kzalloc(sizeof(struct hci_st), GFP_KERNEL);
+	if (!hst) {
+		BT_ERR("Can't allocate control structure");
+		return -ENFILE;
+	}
+
+	/* Expose "hciX" device to user space */
+	err = hci_st_register_dev(hst);
+	if (err) {
+		/* Release local resource memory */
+		kfree(hst);
+
+		BT_ERR("Unable to expose hci0 device(%d)", err);
+		return err;
+	}
+	set_bit(BT_DRV_RUNNING, &hst->flags);
+	dev_set_drvdata(&pdev->dev, hst);
+	return err;
+}
+
+static int bt_ti_remove(struct platform_device *pdev)
+{
+	struct hci_st *hst;
+
+	hst = dev_get_drvdata(&pdev->dev);
+	/* Deallocate local resource's memory  */
+	if (hst) {
+		struct hci_dev *hdev = hst->hdev;
+		if (hdev == NULL) {
+			BT_ERR("Invalid hdev memory");
+			kfree(hst);
+		} else {
+			hci_st_close(hdev);
+			if (test_and_clear_bit(BT_DRV_RUNNING, &hst->flags)) {
+				/* Remove HCI device (hciX) created
+				 * in module init.
+				 */
+				hci_unregister_dev(hdev);
+				/* Free HCI device memory */
+				hci_free_dev(hdev);
+			}
+		}
+	}
+	return 0;
+}
+
+static struct platform_driver bt_ti_driver = {
+	.probe = bt_ti_probe,
+	.remove = bt_ti_remove,
+	.driver = {
+		.name = "ti_st_bluetooth",
+		.owner = THIS_MODULE,
+	},
+};
+
+/* ------- Module Init/Exit interfaces ------ */
+static int __init bt_drv_init(void)
+{
+	long ret = 0;
+	ret = platform_driver_register(&bt_ti_driver);
+	if (ret != 0) {
+		BT_ERR("bt_ti platform drv registration failed");
+		return -1;
+	}
+	return 0;
+}
+
+static void __exit bt_drv_exit(void)
+{
+	platform_driver_unregister(&bt_ti_driver);
+}
+
+module_init(bt_drv_init);
+module_exit(bt_drv_exit);
+
+/* ------ Module Info ------ */
+
+module_param(reset, bool, 0644);
+MODULE_PARM_DESC(reset, "Send HCI reset command on initialization");
+MODULE_AUTHOR("Raja Mani <raja_mani@ti.com>");
+MODULE_DESCRIPTION("Bluetooth Driver for TI Shared Transport" VERSION);
+MODULE_VERSION(VERSION);
+MODULE_LICENSE("GPL");
-- 
1.6.5

Re: [PATCH 1/2] drivers:bluetooth: TI_ST bluetooth driver

[Gustavo F. Padovan]

Hi Pavan,

Change the commit subject to "Bluetooth: TI_ST bluetooth driver" 

* pavan_savoy@ti.com <pavan_savoy@ti.com> [2010-10-07 14:47:16 -0400]:

> From: Pavan Savoy <pavan_savoy@ti.com>
> 
> This is the bluetooth protocol driver for the TI WiLink7 chipsets.
> Texas Instrument's WiLink chipsets combine wireless technologies
> like BT, FM, GPS and WLAN onto a single chip.
> 
> This Bluetooth driver works on top of the TI_ST shared transport
> line discipline driver which also allows other drivers like
> FM V4L2 and GPS character driver to make use of the same UART interface.
> 
> Signed-off-by: Pavan Savoy <pavan_savoy@ti.com>
> ---
>  drivers/bluetooth/bt_ti.c |  489 +++++++++++++++++++++++++++++++++++++++++++++
>  1 files changed, 489 insertions(+), 0 deletions(-)
>  create mode 100644 drivers/bluetooth/bt_ti.c

We don't have filename with bt_.. in drivers/bluetooth/. Maybe ti_st.c
should be a better name, or something like that.

> 
> diff --git a/drivers/bluetooth/bt_ti.c b/drivers/bluetooth/bt_ti.c
> new file mode 100644
> index 0000000..dffbb56
> --- /dev/null
> +++ b/drivers/bluetooth/bt_ti.c
> @@ -0,0 +1,489 @@
> +/*
> + *  Texas Instrument's Bluetooth Driver For Shared Transport.
> + *
> + *  Bluetooth Driver acts as interface between HCI CORE and
> + *  TI Shared Transport Layer.
> + *
> + *  Copyright (C) 2009-2010 Texas Instruments
> + *  Author: Raja Mani <raja_mani@ti.com>
> + *	Pavan Savoy <pavan_savoy@ti.com>
> + *
> + *  This program is free software; you can redistribute it and/or modify
> + *  it under the terms of the GNU General Public License version 2 as
> + *  published by the Free Software Foundation.
> + *
> + *  This program is distributed in the hope that it will be useful,
> + *  but WITHOUT ANY WARRANTY; without even the implied warranty of
> + *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + *  GNU General Public License for more details.
> + *
> + *  You should have received a copy of the GNU General Public License
> + *  along with this program; if not, write to the Free Software
> + *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
> + *
> + */
> +
> +#include <linux/platform_device.h>
> +#include <net/bluetooth/bluetooth.h>
> +#include <net/bluetooth/hci_core.h>
> +
> +#include <linux/ti_wilink_st.h>
> +
> +/* Bluetooth Driver Version */
> +#define VERSION               "1.0"
> +
> +/* Defines number of seconds to wait for reg completion
> + * callback getting called from ST (in case,registration
> + * with ST returns PENDING status)
> + */
> +#define BT_REGISTER_TIMEOUT   6000	/* 6 sec */
> +
> +/* BT driver's local status */
> +#define BT_DRV_RUNNING        0
> +#define BT_ST_REGISTERED      1
> +
> +/**
> + * struct hci_st - BT driver operation structure
> + * @hdev: hci device pointer which binds to bt driver
> + * @flags: used locally,to maintain various BT driver status
> + * @streg_cbdata: to hold ST registration callback status
> + * @st_write: write function pointer of ST driver
> + * @wait_for_btdrv_reg_completion - completion sync between hci_st_open
> + *	and hci_st_registration_completion_cb.
> + */
> +struct hci_st {
> +	struct hci_dev *hdev;
> +	unsigned long flags;
> +	char streg_cbdata;
> +	long (*st_write) (struct sk_buff *);
> +	struct completion wait_for_btdrv_reg_completion;
> +};
> +
> +static int reset;
> +
> +/* Increments HCI counters based on pocket ID (cmd,acl,sco) */
> +static inline void hci_st_tx_complete(struct hci_st *hst, int pkt_type)
> +{
> +	struct hci_dev *hdev;
> +	hdev = hst->hdev;
> +
> +	/* Update HCI stat counters */
> +	switch (pkt_type) {
> +	case HCI_COMMAND_PKT:
> +		hdev->stat.cmd_tx++;
> +		break;
> +
> +	case HCI_ACLDATA_PKT:
> +		hdev->stat.acl_tx++;
> +		break;
> +
> +	case HCI_SCODATA_PKT:
> +		hdev->stat.cmd_tx++;

it should be sco_tx here.

> +		break;
> +	}
> +}
> +
> +/* ------- Interfaces to Shared Transport ------ */
> +
> +/* Called by ST layer to indicate protocol registration completion
> + * status.hci_st_open() function will wait for signal from this
> + * API when st_register() function returns ST_PENDING.
> + */
> +static void hci_st_registration_completion_cb(void *priv_data, char data)

That is not the hci layer, so rename this function (and the others) to
something that reflect where they are really doing.

> +{
> +	struct hci_st *lhst = (struct hci_st *)priv_data;
> +	/* hci_st_open() function needs value of 'data' to know
> +	 * the registration status(success/fail),So have a back
> +	 * up of it.
> +	 */
> +	lhst->streg_cbdata = data;
> +
> +	/* Got a feedback from ST for BT driver registration
> +	 * request.Wackup hci_st_open() function to continue
> +	 * it's open operation.
> +	 */
> +	complete(&lhst->wait_for_btdrv_reg_completion);
> +}
> +
> +/* Called by Shared Transport layer when receive data is
> + * available */
> +static long hci_st_receive(void *priv_data, struct sk_buff *skb)
> +{
> +	int err;
> +	int len;

you can put err and len in the same line.

> +	struct hci_st *lhst = (struct hci_st *)priv_data;
> +
> +	err = 0;
> +	len = 0;

and no need to set them to 0 here.

> +
> +	if (skb == NULL) {
> +		BT_ERR("Invalid SKB received from ST");
> +		return -EFAULT;
> +	}

We need a empty line here.

> +	if (!lhst) {
> +		kfree_skb(skb);
> +		BT_ERR("Invalid hci_st memory,freeing SKB");
> +		return -EFAULT;
> +	}

And also here. Check the rest of the code for similar issues.

> +	if (!test_bit(BT_DRV_RUNNING, &lhst->flags)) {
> +		kfree_skb(skb);
> +		BT_ERR("Device is not running,freeing SKB");
> +		return -EINVAL;
> +	}

If you are here, your device is running, right? Or am I missing
something?

> +
> +	len = skb->len;
> +	skb->dev = (struct net_device *)lhst->hdev;
> +
> +	/* Forward skb to HCI CORE layer */
> +	err = hci_recv_frame(skb);
> +	if (err) {
> +		kfree_skb(skb);
> +		BT_ERR("Unable to push skb to HCI CORE(%d),freeing SKB",
> +				err);
> +		return err;
> +	}
> +	lhst->hdev->stat.byte_rx += len;

actually you even don't need len, just use skb->len

> +
> +	return 0;
> +}
> +
> +/* ------- Interfaces to HCI layer ------ */
> +
> +/* Called from HCI core to initialize the device */
> +static int hci_st_open(struct hci_dev *hdev)
> +{
> +	static struct st_proto_s hci_st_proto;
> +	unsigned long timeleft;
> +	struct hci_st *hst;
> +	int err;
> +	err = 0;
> +
> +	BT_DBG("%s %p", hdev->name, hdev);
> +	hst = hdev->driver_data;
> +
> +	/* Populate BT driver info required by ST */
> +	memset(&hci_st_proto, 0, sizeof(hci_st_proto));
> +
> +	/* BT driver ID */
> +	hci_st_proto.type = ST_BT;
> +
> +	/* Receive function which called from ST */
> +	hci_st_proto.recv = hci_st_receive;
> +
> +	/* Packet match function may used in future */
> +	hci_st_proto.match_packet = NULL;

It is already NULL, you dua a memset.

> +
> +	/* Callback to be called when registration is pending */
> +	hci_st_proto.reg_complete_cb = hci_st_registration_completion_cb;
> +
> +	/* This is write function pointer of ST. BT driver will make use of this
> +	 * for sending any packets to chip. ST will assign and give to us, so
> +	 * make it as NULL */
> +	hci_st_proto.write = NULL;

Same here.

> +
> +	/* send in the hst to be received at registration complete callback
> +	 * and during st's receive
> +	 */
> +	hci_st_proto.priv_data = hst;
> +
> +	/* Register with ST layer */
> +	err = st_register(&hci_st_proto);
> +	if (err == -EINPROGRESS) {
> +		/* Prepare wait-for-completion handler data structures.
> +		 * Needed to syncronize this and st_registration_completion_cb()
> +		 * functions.
> +		 */
> +		init_completion(&hst->wait_for_btdrv_reg_completion);

I'm not liking that, but I'll leave for Marcel and others comment.

> +
> +		/* Reset ST registration callback status flag , this value
> +		 * will be updated in hci_st_registration_completion_cb()
> +		 * function whenever it called from ST driver.
> +		 */
> +		hst->streg_cbdata = -EINPROGRESS;
> +
> +		/* ST is busy with other protocol registration(may be busy with
> +		 * firmware download).So,Wait till the registration callback
> +		 * (passed as a argument to st_register() function) getting
> +		 * called from ST.
> +		 */
> +		BT_DBG(" %s waiting for reg completion signal from ST",
> +				__func__);
> +
> +		timeleft =
> +			wait_for_completion_timeout
> +			(&hst->wait_for_btdrv_reg_completion,
> +			 msecs_to_jiffies(BT_REGISTER_TIMEOUT));
> +		if (!timeleft) {
> +			BT_ERR("Timeout(%d sec),didn't get reg"
> +					"completion signal from ST",
> +					BT_REGISTER_TIMEOUT / 1000);
> +			return -ETIMEDOUT;
> +		}
> +
> +		/* Is ST registration callback called with ERROR value? */
> +		if (hst->streg_cbdata != 0) {
> +			BT_ERR("ST reg completion CB called with invalid"
> +					"status %d", hst->streg_cbdata);
> +			return -EAGAIN;
> +		}
> +		err = 0;
> +	} else if (err == -1) {

Use the proper error macro instead "-1" 

> +		BT_ERR("st_register failed %d", err);
> +		return -EAGAIN;
> +	}
> +
> +	/* Do we have proper ST write function? */
> +	if (hci_st_proto.write != NULL) {
> +		/* We need this pointer for sending any Bluetooth pkts */
> +		hst->st_write = hci_st_proto.write;
> +	} else {
> +		BT_ERR("failed to get ST write func pointer");
> +
> +		/* Undo registration with ST */
> +		err = st_unregister(ST_BT);
> +		if (err < 0)
> +			BT_ERR("st_unregister failed %d", err);
> +
> +		hst->st_write = NULL;
> +		return -EAGAIN;
> +	}
> +
> +	/* Registration with ST layer is completed successfully,
> +	 * now chip is ready to accept commands from HCI CORE.
> +	 * Mark HCI Device flag as RUNNING
> +	 */
> +	set_bit(HCI_RUNNING, &hdev->flags);
> +
> +	/* Registration with ST successful */
> +	set_bit(BT_ST_REGISTERED, &hst->flags);
> +
> +	return err;
> +}
> +
> +/* Close device */
> +static int hci_st_close(struct hci_dev *hdev)
> +{
> +	int err;
> +	struct hci_st *hst;

Skip a line after declarations.

> +	err = 0;

you can set err to 0 in the declaration if you really need that.

> +
> +	hst = hdev->driver_data;
> +	/* Unregister from ST layer */
> +	if (test_and_clear_bit(BT_ST_REGISTERED, &hst->flags)) {
> +		err = st_unregister(ST_BT);
> +		if (err != 0) {
> +			BT_ERR("st_unregister failed %d", err);
> +			return -EBUSY;
> +		}
> +	}
> +
> +	hst->st_write = NULL;
> +
> +	/* ST layer would have moved chip to inactive state.
> +	 * So,clear HCI device RUNNING flag.
> +	 */
> +	if (!test_and_clear_bit(HCI_RUNNING, &hdev->flags))
> +		return 0;

Looks you are screwing up the flags here, if it fails on st_unregister()
and returns HCI_RUNNING should keep set?

> +
> +	return err;

Rethink how you are doing error handling here, it should no be
complicated like that.

> +}
> +
> +/* Called from HCI CORE , Sends frames to Shared Transport */
> +static int hci_st_send_frame(struct sk_buff *skb)
> +{
> +	struct hci_dev *hdev;
> +	struct hci_st *hst;
> +	long len;
> +
> +	if (skb == NULL) {
> +		BT_ERR("Invalid skb received from HCI CORE");
> +		return -ENOMEM;
> +	}
> +	hdev = (struct hci_dev *)skb->dev;
> +	if (!hdev) {
> +		BT_ERR("SKB received for invalid HCI Device (hdev=NULL)");
> +		return -ENODEV;
> +	}
> +	if (!test_bit(HCI_RUNNING, &hdev->flags)) {
> +		BT_ERR("Device is not running");
> +		return -EBUSY;
> +	}
> +
> +	hst = (struct hci_st *)hdev->driver_data;
> +
> +	/* Prepend skb with frame type */
> +	memcpy(skb_push(skb, 1), &bt_cb(skb)->pkt_type, 1);
> +
> +	BT_DBG(" %s: type %d len %d", hdev->name, bt_cb(skb)->pkt_type,
> +			skb->len);
> +
> +	/* Insert skb to shared transport layer's transmit queue.
> +	 * Freeing skb memory is taken care in shared transport layer,
> +	 * so don't free skb memory here.
> +	 */
> +	if (!hst->st_write) {
> +		kfree_skb(skb);
> +		BT_ERR(" Can't write to ST, st_write null?");
> +		return -EAGAIN;
> +	}
> +	len = hst->st_write(skb);
> +	if (len < 0) {
> +		/* Something went wrong in st write , free skb memory */

IMHO we don't need comments like that, clearly we now that something
went wrong.

> +		kfree_skb(skb);
> +		BT_ERR(" ST write failed (%ld)", len);
> +		return -EAGAIN;
> +	}
> +
> +	/* ST accepted our skb. So, Go ahead and do rest */
> +	hdev->stat.byte_tx += len;
> +	hci_st_tx_complete(hst, bt_cb(skb)->pkt_type);
> +
> +	return 0;

goto might be better to handle error here.


-- 
Gustavo F. Padovan
ProFUSION embedded systems - http://profusion.mobi

RE: [PATCH 1/2] drivers:bluetooth: TI_ST bluetooth driver

[Savoy, Pavan]

Gustavo / Marcel,
 

> -----Original Message-----
> From: Gustavo F. Padovan [mailto:pao@profusion.mobi] On Behalf Of Gustavo F.
> Padovan
> Sent: Thursday, October 07, 2010 1:46 PM
> To: Savoy, Pavan
> Cc: linux-bluetooth@vger.kernel.org; marcel@holtmann.org; greg@kroah.com;
> linux-kernel@vger.kernel.org
> Subject: Re: [PATCH 1/2] drivers:bluetooth: TI_ST bluetooth driver
> 
> Hi Pavan,
> 
> Change the commit subject to "Bluetooth: TI_ST bluetooth driver"
> 
> * pavan_savoy@ti.com <pavan_savoy@ti.com> [2010-10-07 14:47:16 -0400]:
> 
> > From: Pavan Savoy <pavan_savoy@ti.com>
> >
> > This is the bluetooth protocol driver for the TI WiLink7 chipsets.
> > Texas Instrument's WiLink chipsets combine wireless technologies
> > like BT, FM, GPS and WLAN onto a single chip.
> >
> > This Bluetooth driver works on top of the TI_ST shared transport
> > line discipline driver which also allows other drivers like
> > FM V4L2 and GPS character driver to make use of the same UART interface.
> >
> > Signed-off-by: Pavan Savoy <pavan_savoy@ti.com>
> > ---
> >  drivers/bluetooth/bt_ti.c |  489
> +++++++++++++++++++++++++++++++++++++++++++++
> >  1 files changed, 489 insertions(+), 0 deletions(-)
> >  create mode 100644 drivers/bluetooth/bt_ti.c
> 
> We don't have filename with bt_.. in drivers/bluetooth/. Maybe ti_st.c
> should be a better name, or something like that.
> 
> >
> > diff --git a/drivers/bluetooth/bt_ti.c b/drivers/bluetooth/bt_ti.c
> > new file mode 100644
> > index 0000000..dffbb56
> > --- /dev/null
> > +++ b/drivers/bluetooth/bt_ti.c
> > @@ -0,0 +1,489 @@
> > +/*
> > + *  Texas Instrument's Bluetooth Driver For Shared Transport.
> > + *
> > + *  Bluetooth Driver acts as interface between HCI CORE and
> > + *  TI Shared Transport Layer.
> > + *
> > + *  Copyright (C) 2009-2010 Texas Instruments
> > + *  Author: Raja Mani <raja_mani@ti.com>
> > + *	Pavan Savoy <pavan_savoy@ti.com>
> > + *
> > + *  This program is free software; you can redistribute it and/or modify
> > + *  it under the terms of the GNU General Public License version 2 as
> > + *  published by the Free Software Foundation.
> > + *
> > + *  This program is distributed in the hope that it will be useful,
> > + *  but WITHOUT ANY WARRANTY; without even the implied warranty of
> > + *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> > + *  GNU General Public License for more details.
> > + *
> > + *  You should have received a copy of the GNU General Public License
> > + *  along with this program; if not, write to the Free Software
> > + *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307
> USA
> > + *
> > + */
> > +
> > +#include <linux/platform_device.h>
> > +#include <net/bluetooth/bluetooth.h>
> > +#include <net/bluetooth/hci_core.h>
> > +
> > +#include <linux/ti_wilink_st.h>
> > +
> > +/* Bluetooth Driver Version */
> > +#define VERSION               "1.0"
> > +
> > +/* Defines number of seconds to wait for reg completion
> > + * callback getting called from ST (in case,registration
> > + * with ST returns PENDING status)
> > + */
> > +#define BT_REGISTER_TIMEOUT   6000	/* 6 sec */
> > +
> > +/* BT driver's local status */
> > +#define BT_DRV_RUNNING        0
> > +#define BT_ST_REGISTERED      1
> > +
> > +/**
> > + * struct hci_st - BT driver operation structure
> > + * @hdev: hci device pointer which binds to bt driver
> > + * @flags: used locally,to maintain various BT driver status
> > + * @streg_cbdata: to hold ST registration callback status
> > + * @st_write: write function pointer of ST driver
> > + * @wait_for_btdrv_reg_completion - completion sync between hci_st_open
> > + *	and hci_st_registration_completion_cb.
> > + */
> > +struct hci_st {
> > +	struct hci_dev *hdev;
> > +	unsigned long flags;
> > +	char streg_cbdata;
> > +	long (*st_write) (struct sk_buff *);
> > +	struct completion wait_for_btdrv_reg_completion;
> > +};
> > +
> > +static int reset;
> > +
> > +/* Increments HCI counters based on pocket ID (cmd,acl,sco) */
> > +static inline void hci_st_tx_complete(struct hci_st *hst, int pkt_type)
> > +{
> > +	struct hci_dev *hdev;
> > +	hdev = hst->hdev;
> > +
> > +	/* Update HCI stat counters */
> > +	switch (pkt_type) {
> > +	case HCI_COMMAND_PKT:
> > +		hdev->stat.cmd_tx++;
> > +		break;
> > +
> > +	case HCI_ACLDATA_PKT:
> > +		hdev->stat.acl_tx++;
> > +		break;
> > +
> > +	case HCI_SCODATA_PKT:
> > +		hdev->stat.cmd_tx++;
> 
> it should be sco_tx here.
> 
> > +		break;
> > +	}
> > +}
> > +
> > +/* ------- Interfaces to Shared Transport ------ */
> > +
> > +/* Called by ST layer to indicate protocol registration completion
> > + * status.hci_st_open() function will wait for signal from this
> > + * API when st_register() function returns ST_PENDING.
> > + */
> > +static void hci_st_registration_completion_cb(void *priv_data, char data)
> 
> That is not the hci layer, so rename this function (and the others) to
> something that reflect where they are really doing.
> 
> > +{
> > +	struct hci_st *lhst = (struct hci_st *)priv_data;
> > +	/* hci_st_open() function needs value of 'data' to know
> > +	 * the registration status(success/fail),So have a back
> > +	 * up of it.
> > +	 */
> > +	lhst->streg_cbdata = data;
> > +
> > +	/* Got a feedback from ST for BT driver registration
> > +	 * request.Wackup hci_st_open() function to continue
> > +	 * it's open operation.
> > +	 */
> > +	complete(&lhst->wait_for_btdrv_reg_completion);
> > +}
> > +
> > +/* Called by Shared Transport layer when receive data is
> > + * available */
> > +static long hci_st_receive(void *priv_data, struct sk_buff *skb)
> > +{
> > +	int err;
> > +	int len;
> 
> you can put err and len in the same line.
> 
> > +	struct hci_st *lhst = (struct hci_st *)priv_data;
> > +
> > +	err = 0;
> > +	len = 0;
> 
> and no need to set them to 0 here.
> 
> > +
> > +	if (skb == NULL) {
> > +		BT_ERR("Invalid SKB received from ST");
> > +		return -EFAULT;
> > +	}
> 
> We need a empty line here.
> 
> > +	if (!lhst) {
> > +		kfree_skb(skb);
> > +		BT_ERR("Invalid hci_st memory,freeing SKB");
> > +		return -EFAULT;
> > +	}
> 
> And also here. Check the rest of the code for similar issues.
> 
> > +	if (!test_bit(BT_DRV_RUNNING, &lhst->flags)) {
> > +		kfree_skb(skb);
> > +		BT_ERR("Device is not running,freeing SKB");
> > +		return -EINVAL;
> > +	}
> 
> If you are here, your device is running, right? Or am I missing
> something?
> 
> > +
> > +	len = skb->len;
> > +	skb->dev = (struct net_device *)lhst->hdev;
> > +
> > +	/* Forward skb to HCI CORE layer */
> > +	err = hci_recv_frame(skb);
> > +	if (err) {
> > +		kfree_skb(skb);
> > +		BT_ERR("Unable to push skb to HCI CORE(%d),freeing SKB",
> > +				err);
> > +		return err;
> > +	}
> > +	lhst->hdev->stat.byte_rx += len;
> 
> actually you even don't need len, just use skb->len
> 
> > +
> > +	return 0;
> > +}
> > +
> > +/* ------- Interfaces to HCI layer ------ */
> > +
> > +/* Called from HCI core to initialize the device */
> > +static int hci_st_open(struct hci_dev *hdev)
> > +{
> > +	static struct st_proto_s hci_st_proto;
> > +	unsigned long timeleft;
> > +	struct hci_st *hst;
> > +	int err;
> > +	err = 0;
> > +
> > +	BT_DBG("%s %p", hdev->name, hdev);
> > +	hst = hdev->driver_data;
> > +
> > +	/* Populate BT driver info required by ST */
> > +	memset(&hci_st_proto, 0, sizeof(hci_st_proto));
> > +
> > +	/* BT driver ID */
> > +	hci_st_proto.type = ST_BT;
> > +
> > +	/* Receive function which called from ST */
> > +	hci_st_proto.recv = hci_st_receive;
> > +
> > +	/* Packet match function may used in future */
> > +	hci_st_proto.match_packet = NULL;
> 
> It is already NULL, you dua a memset.
> 
> > +
> > +	/* Callback to be called when registration is pending */
> > +	hci_st_proto.reg_complete_cb = hci_st_registration_completion_cb;
> > +
> > +	/* This is write function pointer of ST. BT driver will make use of
> this
> > +	 * for sending any packets to chip. ST will assign and give to us, so
> > +	 * make it as NULL */
> > +	hci_st_proto.write = NULL;
> 
> Same here.
> 
> > +
> > +	/* send in the hst to be received at registration complete callback
> > +	 * and during st's receive
> > +	 */
> > +	hci_st_proto.priv_data = hst;
> > +
> > +	/* Register with ST layer */
> > +	err = st_register(&hci_st_proto);
> > +	if (err == -EINPROGRESS) {
> > +		/* Prepare wait-for-completion handler data structures.
> > +		 * Needed to syncronize this and st_registration_completion_cb()
> > +		 * functions.
> > +		 */
> > +		init_completion(&hst->wait_for_btdrv_reg_completion);
> 
> I'm not liking that, but I'll leave for Marcel and others comment.

Thanks for the comments, I will take care of them all, but I want to
tell why I do this here.
"st_register" function takes time. It takes about 2 seconds for us
to download the firmware. (or longer based on UART baud-rate).

This is intended to happen in the process context. Say hciconfig hci0 up
takes a while longer than usual or if someone doing an ioctl(HCIDEVUP)
and this is also the reason I have a callback, as and when firmware download
completes, I call the Bluetooth driver's callback.

and I await marcel's comment on this.

> > +
> > +		/* Reset ST registration callback status flag , this value
> > +		 * will be updated in hci_st_registration_completion_cb()
> > +		 * function whenever it called from ST driver.
> > +		 */
> > +		hst->streg_cbdata = -EINPROGRESS;
> > +
> > +		/* ST is busy with other protocol registration(may be busy with
> > +		 * firmware download).So,Wait till the registration callback
> > +		 * (passed as a argument to st_register() function) getting
> > +		 * called from ST.
> > +		 */
> > +		BT_DBG(" %s waiting for reg completion signal from ST",
> > +				__func__);
> > +
> > +		timeleft =
> > +			wait_for_completion_timeout
> > +			(&hst->wait_for_btdrv_reg_completion,
> > +			 msecs_to_jiffies(BT_REGISTER_TIMEOUT));
> > +		if (!timeleft) {
> > +			BT_ERR("Timeout(%d sec),didn't get reg"
> > +					"completion signal from ST",
> > +					BT_REGISTER_TIMEOUT / 1000);
> > +			return -ETIMEDOUT;
> > +		}
> > +
> > +		/* Is ST registration callback called with ERROR value? */
> > +		if (hst->streg_cbdata != 0) {
> > +			BT_ERR("ST reg completion CB called with invalid"
> > +					"status %d", hst->streg_cbdata);
> > +			return -EAGAIN;
> > +		}
> > +		err = 0;
> > +	} else if (err == -1) {
> 
> Use the proper error macro instead "-1"
> 
> > +		BT_ERR("st_register failed %d", err);
> > +		return -EAGAIN;
> > +	}
> > +
> > +	/* Do we have proper ST write function? */
> > +	if (hci_st_proto.write != NULL) {
> > +		/* We need this pointer for sending any Bluetooth pkts */
> > +		hst->st_write = hci_st_proto.write;
> > +	} else {
> > +		BT_ERR("failed to get ST write func pointer");
> > +
> > +		/* Undo registration with ST */
> > +		err = st_unregister(ST_BT);
> > +		if (err < 0)
> > +			BT_ERR("st_unregister failed %d", err);
> > +
> > +		hst->st_write = NULL;
> > +		return -EAGAIN;
> > +	}
> > +
> > +	/* Registration with ST layer is completed successfully,
> > +	 * now chip is ready to accept commands from HCI CORE.
> > +	 * Mark HCI Device flag as RUNNING
> > +	 */
> > +	set_bit(HCI_RUNNING, &hdev->flags);
> > +
> > +	/* Registration with ST successful */
> > +	set_bit(BT_ST_REGISTERED, &hst->flags);
> > +
> > +	return err;
> > +}
> > +
> > +/* Close device */
> > +static int hci_st_close(struct hci_dev *hdev)
> > +{
> > +	int err;
> > +	struct hci_st *hst;
> 
> Skip a line after declarations.
> 
> > +	err = 0;
> 
> you can set err to 0 in the declaration if you really need that.
> 
> > +
> > +	hst = hdev->driver_data;
> > +	/* Unregister from ST layer */
> > +	if (test_and_clear_bit(BT_ST_REGISTERED, &hst->flags)) {
> > +		err = st_unregister(ST_BT);
> > +		if (err != 0) {
> > +			BT_ERR("st_unregister failed %d", err);
> > +			return -EBUSY;
> > +		}
> > +	}
> > +
> > +	hst->st_write = NULL;
> > +
> > +	/* ST layer would have moved chip to inactive state.
> > +	 * So,clear HCI device RUNNING flag.
> > +	 */
> > +	if (!test_and_clear_bit(HCI_RUNNING, &hdev->flags))
> > +		return 0;
> 
> Looks you are screwing up the flags here, if it fails on st_unregister()
> and returns HCI_RUNNING should keep set?
> 
> > +
> > +	return err;
> 
> Rethink how you are doing error handling here, it should no be
> complicated like that.
> 
> > +}
> > +
> > +/* Called from HCI CORE , Sends frames to Shared Transport */
> > +static int hci_st_send_frame(struct sk_buff *skb)
> > +{
> > +	struct hci_dev *hdev;
> > +	struct hci_st *hst;
> > +	long len;
> > +
> > +	if (skb == NULL) {
> > +		BT_ERR("Invalid skb received from HCI CORE");
> > +		return -ENOMEM;
> > +	}
> > +	hdev = (struct hci_dev *)skb->dev;
> > +	if (!hdev) {
> > +		BT_ERR("SKB received for invalid HCI Device (hdev=NULL)");
> > +		return -ENODEV;
> > +	}
> > +	if (!test_bit(HCI_RUNNING, &hdev->flags)) {
> > +		BT_ERR("Device is not running");
> > +		return -EBUSY;
> > +	}
> > +
> > +	hst = (struct hci_st *)hdev->driver_data;
> > +
> > +	/* Prepend skb with frame type */
> > +	memcpy(skb_push(skb, 1), &bt_cb(skb)->pkt_type, 1);
> > +
> > +	BT_DBG(" %s: type %d len %d", hdev->name, bt_cb(skb)->pkt_type,
> > +			skb->len);
> > +
> > +	/* Insert skb to shared transport layer's transmit queue.
> > +	 * Freeing skb memory is taken care in shared transport layer,
> > +	 * so don't free skb memory here.
> > +	 */
> > +	if (!hst->st_write) {
> > +		kfree_skb(skb);
> > +		BT_ERR(" Can't write to ST, st_write null?");
> > +		return -EAGAIN;
> > +	}
> > +	len = hst->st_write(skb);
> > +	if (len < 0) {
> > +		/* Something went wrong in st write , free skb memory */
> 
> IMHO we don't need comments like that, clearly we now that something
> went wrong.
> 
> > +		kfree_skb(skb);
> > +		BT_ERR(" ST write failed (%ld)", len);
> > +		return -EAGAIN;
> > +	}
> > +
> > +	/* ST accepted our skb. So, Go ahead and do rest */
> > +	hdev->stat.byte_tx += len;
> > +	hci_st_tx_complete(hst, bt_cb(skb)->pkt_type);
> > +
> > +	return 0;
> 
> goto might be better to handle error here.
> 
> 
> --
> Gustavo F. Padovan
> ProFUSION embedded systems - http://profusion.mobi

Re: [PATCH 1/2] drivers:bluetooth: TI_ST bluetooth driver

[Marcel Holtmann]

Hi Gustavo,

> Change the commit subject to "Bluetooth: TI_ST bluetooth driver" 
> 
> * pavan_savoy@ti.com <pavan_savoy@ti.com> [2010-10-07 14:47:16 -0400]:
> 
> > From: Pavan Savoy <pavan_savoy@ti.com>
> > 
> > This is the bluetooth protocol driver for the TI WiLink7 chipsets.
> > Texas Instrument's WiLink chipsets combine wireless technologies
> > like BT, FM, GPS and WLAN onto a single chip.
> > 
> > This Bluetooth driver works on top of the TI_ST shared transport
> > line discipline driver which also allows other drivers like
> > FM V4L2 and GPS character driver to make use of the same UART interface.
> > 
> > Signed-off-by: Pavan Savoy <pavan_savoy@ti.com>
> > ---
> >  drivers/bluetooth/bt_ti.c |  489 +++++++++++++++++++++++++++++++++++++++++++++
> >  1 files changed, 489 insertions(+), 0 deletions(-)
> >  create mode 100644 drivers/bluetooth/bt_ti.c
> 
> We don't have filename with bt_.. in drivers/bluetooth/. Maybe ti_st.c
> should be a better name, or something like that.

actually we have the bt prefix for company generic ones where they
didn't wanna use the hardware name.

So I prefer to use the hardware for a driver since it is much more clear
that way. You acronym naming here is bad. It is confusing like hell.

What about just calling this btwilink.c or something. I just spinning
ideas here.

Regards

Marcel

RE: [PATCH 1/2] drivers:bluetooth: TI_ST bluetooth driver

[Savoy, Pavan]

> -----Original Message-----
> From: Marcel Holtmann [mailto:marcel@holtmann.org]
> Sent: Friday, October 08, 2010 3:38 AM
> To: Gustavo F. Padovan
> Cc: Savoy, Pavan; linux-bluetooth@vger.kernel.org; greg@kroah.com; linux-
> kernel@vger.kernel.org
> Subject: Re: [PATCH 1/2] drivers:bluetooth: TI_ST bluetooth driver
> 
> Hi Gustavo,
> 
> > Change the commit subject to "Bluetooth: TI_ST bluetooth driver"
> >
> > * pavan_savoy@ti.com <pavan_savoy@ti.com> [2010-10-07 14:47:16 -0400]:
> >
> > > From: Pavan Savoy <pavan_savoy@ti.com>
> > >
> > > This is the bluetooth protocol driver for the TI WiLink7 chipsets.
> > > Texas Instrument's WiLink chipsets combine wireless technologies
> > > like BT, FM, GPS and WLAN onto a single chip.
> > >
> > > This Bluetooth driver works on top of the TI_ST shared transport
> > > line discipline driver which also allows other drivers like
> > > FM V4L2 and GPS character driver to make use of the same UART interface.
> > >
> > > Signed-off-by: Pavan Savoy <pavan_savoy@ti.com>
> > > ---
> > >  drivers/bluetooth/bt_ti.c |  489
> +++++++++++++++++++++++++++++++++++++++++++++
> > >  1 files changed, 489 insertions(+), 0 deletions(-)
> > >  create mode 100644 drivers/bluetooth/bt_ti.c
> >
> > We don't have filename with bt_.. in drivers/bluetooth/. Maybe ti_st.c
> > should be a better name, or something like that.
> 
> actually we have the bt prefix for company generic ones where they
> didn't wanna use the hardware name.
> 
> So I prefer to use the hardware for a driver since it is much more clear
> that way. You acronym naming here is bad. It is confusing like hell.
> 
> What about just calling this btwilink.c or something. I just spinning
> ideas here.

Ok, I like btwilink.c too. Does bt_wilink.c sound OK?
I have sent a patch with ti_st.c as of now.

Please review, and provide comments. I will incorporate comments and change
Name in the next version of the patch.

Thanks,
Pavan

> Regards
> 
> Marcel
> 

[PATCH 2/2] drivers:bluetooth: Kconfig & Makefile for TI BT

[pavan_savoy]

From: Pavan Savoy <pavan_savoy@ti.com>

Kconfig and Makefile modifications to enable the Bluetooth
driver for Texas Instrument's WiLink 7 chipset.

Signed-off-by: Pavan Savoy <pavan_savoy@ti.com>
---
 drivers/bluetooth/Kconfig  |   10 ++++++++++
 drivers/bluetooth/Makefile |    1 +
 2 files changed, 11 insertions(+), 0 deletions(-)

diff --git a/drivers/bluetooth/Kconfig b/drivers/bluetooth/Kconfig
index 02deef4..936d8ce 100644
--- a/drivers/bluetooth/Kconfig
+++ b/drivers/bluetooth/Kconfig
@@ -219,4 +219,14 @@ config BT_ATH3K
 	  Say Y here to compile support for "Atheros firmware download driver"
 	  into the kernel or say M to compile it as module (ath3k).
 
+config BT_TI
+	tristate "BlueZ bluetooth driver for TI ST"
+	depends on TI_ST
+	help
+	  This enables the Bluetooth driver for Texas Instrument's BT/FM/GPS
+	  combo devices. This makes use of shared transport line discipline
+	  core driver to communicate with the BT core of the combo chip.
+
+	  Say Y here to compile support for Texas Instrument's WiLink7 driver
+	  into the kernel or say M to compile it as module.
 endmenu
diff --git a/drivers/bluetooth/Makefile b/drivers/bluetooth/Makefile
index 71bdf13..63156da 100644
--- a/drivers/bluetooth/Makefile
+++ b/drivers/bluetooth/Makefile
@@ -28,3 +28,4 @@ hci_uart-$(CONFIG_BT_HCIUART_BCSP)	+= hci_bcsp.o
 hci_uart-$(CONFIG_BT_HCIUART_LL)	+= hci_ll.o
 hci_uart-$(CONFIG_BT_HCIUART_ATH3K)	+= hci_ath.o
 hci_uart-objs				:= $(hci_uart-y)
+obj-$(CONFIG_BT_TI)			:= bt_ti.o
-- 
1.6.5

Re: [PATCH 2/2] drivers:bluetooth: Kconfig & Makefile for TI BT

[Gustavo F. Padovan]

* pavan_savoy@ti.com <pavan_savoy@ti.com> [2010-10-07 14:47:17 -0400]:

> From: Pavan Savoy <pavan_savoy@ti.com>
> 
> Kconfig and Makefile modifications to enable the Bluetooth
> driver for Texas Instrument's WiLink 7 chipset.
> 
> Signed-off-by: Pavan Savoy <pavan_savoy@ti.com>
> ---
>  drivers/bluetooth/Kconfig  |   10 ++++++++++
>  drivers/bluetooth/Makefile |    1 +
>  2 files changed, 11 insertions(+), 0 deletions(-)

Just merge this patch in the firt one. No need to keep this separated.

-- 
Gustavo F. Padovan
ProFUSION embedded systems - http://profusion.mobi