Re: [PATCH v9 7/7] Bluetooth: hci_qca: Add support for Qualcomm Bluetooth chip wcn3990

[Balakrishna Godavarthi <bgodavar@codeaurora.org>]

Hi Matthias,

On 2018-07-16 21:35, Matthias Kaehlcke wrote:
> On Mon, Jul 16, 2018 at 07:21:56PM +0530, Balakrishna Godavarthi wrote:
>> Hi Matthias,
>> 
>> On 2018-07-07 03:51, Matthias Kaehlcke wrote:
>> > On Thu, Jul 05, 2018 at 10:25:15PM +0530, Balakrishna Godavarthi wrote:
>> > > Add support to set voltage/current of various regulators
>> > > to power up/down Bluetooth chip wcn3990.
>> > >
>> > > Signed-off-by: Balakrishna Godavarthi <bgodavar@codeaurora.org>
>> > > ---
>> > > changes in v9:
>> > >     * moved flow control to vendor and set_baudarte functions.
>> > >     * removed parent regs.
>> > >
>> > > changes in v8:
>> > >     * closing qca buffer, if qca_power_setup fails
>> > >     * chnaged ibs start timer function call location.
>> > >     * updated review comments.
>> > >
>> > > changes in v7:
>> > >     * addressed review comments.
>> > >
>> > > changes in v6:
>> > >     * Hooked up qca_power to qca_serdev.
>> > >     * renamed all the naming inconsistency functions with qca_*
>> > >     * leveraged common code of ROME for wcn3990.
>> > >     * created wrapper functions for re-usable blocks.
>> > >     * updated function of _*regulator_enable and _*regualtor_disable.
>> > >     * removed redundant comments and functions.
>> > >     * addressed review comments.
>> > >
>> > > Changes in v5:
>> > >     * updated regulator vddpa min_uV to 1304000.
>> > >       * addressed review comments.
>> > >
>> > > Changes in v4:
>> > >     * Segregated the changes of btqca from hci_qca
>> > >     * rebased all changes on top of bluetooth-next.
>> > >     * addressed review comments.
>> > >
>> > > ---
>> > >  drivers/bluetooth/btqca.h   |   3 +
>> > >  drivers/bluetooth/hci_qca.c | 387
>> > > +++++++++++++++++++++++++++++++-----
>> > >  2 files changed, 345 insertions(+), 45 deletions(-)
>> > >
>> > > diff --git a/drivers/bluetooth/btqca.h b/drivers/bluetooth/btqca.h
>> > > index a9c2779f3e07..0c01f375fe83 100644
>> > > --- a/drivers/bluetooth/btqca.h
>> > > +++ b/drivers/bluetooth/btqca.h
>> > > @@ -37,6 +37,9 @@
>> > >  #define EDL_TAG_ID_HCI			(17)
>> > >  #define EDL_TAG_ID_DEEP_SLEEP		(27)
>> > >
>> > > +#define QCA_WCN3990_POWERON_PULSE	0xFC
>> > > +#define QCA_WCN3990_POWEROFF_PULSE	0xC0
>> > > +
>> > >  enum qca_bardrate {
>> > >  	QCA_BAUDRATE_115200 	= 0,
>> > >  	QCA_BAUDRATE_57600,
>> > > diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
>> > > index 7ebfaa0edf3f..d62c7785a618 100644
>> > > --- a/drivers/bluetooth/hci_qca.c
>> > > +++ b/drivers/bluetooth/hci_qca.c
>> > > @@ -5,7 +5,7 @@
>> > >   *  protocol extension to H4.
>> > >   *
>> > >   *  Copyright (C) 2007 Texas Instruments, Inc.
>> > > - *  Copyright (c) 2010, 2012 The Linux Foundation. All rights
>> > > reserved.
>> > > + *  Copyright (c) 2010, 2012, 2018 The Linux Foundation. All rights
>> > > reserved.
>> > >   *
>> > >   *  Acknowledgements:
>> > >   *  This file is based on hci_ll.c, which was...
>> > > @@ -31,9 +31,14 @@
>> > >  #include <linux/kernel.h>
>> > >  #include <linux/clk.h>
>> > >  #include <linux/debugfs.h>
>> > > +#include <linux/delay.h>
>> > > +#include <linux/device.h>
>> > >  #include <linux/gpio/consumer.h>
>> > >  #include <linux/mod_devicetable.h>
>> > >  #include <linux/module.h>
>> > > +#include <linux/of_device.h>
>> > > +#include <linux/platform_device.h>
>> > > +#include <linux/regulator/consumer.h>
>> > >  #include <linux/serdev.h>
>> > >
>> > >  #include <net/bluetooth/bluetooth.h>
>> > > @@ -124,12 +129,46 @@ enum qca_speed_type {
>> > >  	QCA_OPER_SPEED
>> > >  };
>> > >
>> > > +/*
>> > > + * Voltage regulator information required for configuring the
>> > > + * QCA Bluetooth chipset
>> > > + */
>> > > +struct qca_vreg {
>> > > +	const char *name;
>> > > +	unsigned int min_uV;
>> > > +	unsigned int max_uV;
>> > > +	unsigned int load_uA;
>> > > +};
>> > > +
>> > > +struct qca_vreg_data {
>> > > +	enum qca_btsoc_type soc_type;
>> > > +	struct qca_vreg *vregs;
>> > > +	size_t num_vregs;
>> > > +};
>> > > +
>> > > +/*
>> > > + * Platform data for the QCA Bluetooth power driver.
>> > > + */
>> > > +struct qca_power {
>> > > +	struct device *dev;
>> > > +	const struct qca_vreg_data *vreg_data;
>> > > +	struct regulator_bulk_data *vreg_bulk;
>> > > +	bool vregs_on;
>> > > +};
>> > > +
>> > >  struct qca_serdev {
>> > >  	struct hci_uart	 serdev_hu;
>> > >  	struct gpio_desc *bt_en;
>> > >  	struct clk	 *susclk;
>> > > +	enum qca_btsoc_type btsoc_type;
>> > > +	struct qca_power *bt_power;
>> > > +	u32 init_speed;
>> > > +	u32 oper_speed;
>> > >  };
>> > >
>> > > +static int qca_power_setup(struct hci_uart *hu, bool on);
>> > > +static void qca_power_shutdown(struct hci_uart *hu);
>> > > +
>> > >  static void __serial_clock_on(struct tty_struct *tty)
>> > >  {
>> > >  	/* TODO: Some chipset requires to enable UART clock on client
>> > > @@ -407,6 +446,7 @@ static int qca_open(struct hci_uart *hu)
>> > >  {
>> > >  	struct qca_serdev *qcadev;
>> > >  	struct qca_data *qca;
>> > > +	int ret;
>> > >
>> > >  	BT_DBG("hu %p qca_open", hu);
>> > >
>> > > @@ -458,19 +498,32 @@ static int qca_open(struct hci_uart *hu)
>> > >
>> > >  	hu->priv = qca;
>> > >
>> > > -	timer_setup(&qca->wake_retrans_timer,
>> > > hci_ibs_wake_retrans_timeout, 0);
>> > > -	qca->wake_retrans = IBS_WAKE_RETRANS_TIMEOUT_MS;
>> > > -
>> > > -	timer_setup(&qca->tx_idle_timer, hci_ibs_tx_idle_timeout, 0);
>> > > -	qca->tx_idle_delay = IBS_TX_IDLE_TIMEOUT_MS;
>> > > -
>> > >  	if (hu->serdev) {
>> > >  		serdev_device_open(hu->serdev);
>> > >
>> > >  		qcadev = serdev_device_get_drvdata(hu->serdev);
>> > > -		gpiod_set_value_cansleep(qcadev->bt_en, 1);
>> > > +		if (qcadev->btsoc_type != QCA_WCN3990) {
>> > > +			gpiod_set_value_cansleep(qcadev->bt_en, 1);
>> > > +		} else {
>> > > +			hu->init_speed = qcadev->init_speed;
>> > > +			hu->oper_speed = qcadev->oper_speed;
>> > > +			ret = qca_power_setup(hu, true);
>> > > +			if (ret) {
>> > > +				destroy_workqueue(qca->workqueue);
>> > > +				kfree_skb(qca->rx_skb);
>> > > +				hu->priv = NULL;
>> > > +				kfree(qca);
>> > > +				return ret;
>> > > +			}
>> > > +		}
>> > >  	}
>> > >
>> > > +	timer_setup(&qca->wake_retrans_timer,
>> > > hci_ibs_wake_retrans_timeout, 0);
>> > > +	qca->wake_retrans = IBS_WAKE_RETRANS_TIMEOUT_MS;
>> > > +
>> > > +	timer_setup(&qca->tx_idle_timer, hci_ibs_tx_idle_timeout, 0);
>> > > +	qca->tx_idle_delay = IBS_TX_IDLE_TIMEOUT_MS;
>> > > +
>> > >  	BT_DBG("HCI_UART_QCA open, tx_idle_delay=%u, wake_retrans=%u",
>> > >  	       qca->tx_idle_delay, qca->wake_retrans);
>> > >
>> > > @@ -554,10 +607,13 @@ static int qca_close(struct hci_uart *hu)
>> > >  	qca->hu = NULL;
>> > >
>> > >  	if (hu->serdev) {
>> > > -		serdev_device_close(hu->serdev);
>> > > -
>> > >  		qcadev = serdev_device_get_drvdata(hu->serdev);
>> > > -		gpiod_set_value_cansleep(qcadev->bt_en, 0);
>> > > +		if (qcadev->btsoc_type == QCA_WCN3990)
>> > > +			qca_power_shutdown(hu);
>> > > +		else
>> > > +			gpiod_set_value_cansleep(qcadev->bt_en, 0);
>> > > +
>> > > +		serdev_device_close(hu->serdev);
>> > >  	}
>> > >
>> > >  	kfree_skb(qca->rx_skb);
>> > > @@ -891,6 +947,7 @@ static int qca_set_baudrate(struct hci_dev
>> > > *hdev, uint8_t baudrate)
>> > >  	struct hci_uart *hu = hci_get_drvdata(hdev);
>> > >  	struct qca_data *qca = hu->priv;
>> > >  	struct sk_buff *skb;
>> > > +	struct qca_serdev *qcadev;
>> > >  	u8 cmd[] = { 0x01, 0x48, 0xFC, 0x01, 0x00 };
>> > >
>> > >  	if (baudrate > QCA_BAUDRATE_3200000)
>> > > @@ -904,6 +961,13 @@ static int qca_set_baudrate(struct hci_dev
>> > > *hdev, uint8_t baudrate)
>> > >  		return -ENOMEM;
>> > >  	}
>> > >
>> > > +	/* Disabling hardware flow control is mandate while
>> >
>> > nit: s/mandate/mandatory|required/
>> >
>> > > +	 * sending change baudrate request to wcn3990 SoC.
>> > > +	 */
>> > > +	qcadev = serdev_device_get_drvdata(hu->serdev);
>> > > +	if (qcadev->btsoc_type == QCA_WCN3990)
>> > > +		hci_uart_set_flow_control(hu, true);
>> > > +
>> > >  	/* Assign commands to change baudrate and packet type. */
>> > >  	skb_put_data(skb, cmd, sizeof(cmd));
>> > >  	hci_skb_pkt_type(skb) = HCI_COMMAND_PKT;
>> > > @@ -919,6 +983,9 @@ static int qca_set_baudrate(struct hci_dev
>> > > *hdev, uint8_t baudrate)
>> > >  	schedule_timeout(msecs_to_jiffies(BAUDRATE_SETTLE_TIMEOUT_MS));
>> > >  	set_current_state(TASK_RUNNING);
>> > >
>> > > +	if (qcadev->btsoc_type == QCA_WCN3990)
>> > > +		hci_uart_set_flow_control(hu, false);
>> > > +
>> > >  	return 0;
>> > >  }
>> > >
>> > > @@ -930,6 +997,36 @@ static inline void host_set_baudrate(struct
>> > > hci_uart *hu, unsigned int speed)
>> > >  		hci_uart_set_baudrate(hu, speed);
>> > >  }
>> > >
>> > > +static int qca_send_vendor_cmd(struct hci_dev *hdev, u8 cmd)
>> > > +{
>> > > +	struct hci_uart *hu = hci_get_drvdata(hdev);
>> > > +	struct qca_data *qca = hu->priv;
>> > > +	struct sk_buff *skb;
>> > > +
>> > > +	bt_dev_dbg(hdev, "sending command %02x to SoC", cmd);
>> > > +
>> > > +	skb = bt_skb_alloc(sizeof(cmd), GFP_KERNEL);
>> > > +	if (!skb)
>> > > +		return -ENOMEM;
>> > > +
>> > > +	/* Disabling hardware flow control is mandate while
>> >
>> > Same nit as above
>> >
>> > > +	 * sending vendor power on and off pulse to SoC.
>> > > +	 */
>> >
>> > The function is called qca_send_vendor_cmd(), the comment about power
>> > on/off pulses seems misplaced here. Are there other 'vendor commands'
>> > that require a different flow control behavior?
>> >
>> > Perhaps the function should have a different name or we need another
>> > wrapper.
>> >
>> > > +	hci_uart_set_flow_control(hu, true);
>> >
>> > Should the changing of the flow control be limited to wcn3990? As of
>> > now the function is only called for wcn3990, however this is not
>> > stated as a requirement and might change in the future.
>> >
>> > > +	skb_put_u8(skb, cmd);
>> > > +	hci_skb_pkt_type(skb) = HCI_COMMAND_PKT;
>> > > +
>> > > +	skb_queue_tail(&qca->txq, skb);
>> > > +	hci_uart_tx_wakeup(hu);
>> > > +
>> > > +	/* Wait for 100 uS for SoC to settle down */
>> > > +	usleep_range(100, 200);
>> >
>> > Is this needed for any 'vendor command' or directly related with the
>> > power on/off pulses?
>> >
>> > > +	hci_uart_set_flow_control(hu, false);
>> > > +
>> > > +	return 0;
>> > > +}
>> > > +
>> > >  static unsigned int qca_get_speed(struct hci_uart *hu,
>> > >  				  enum qca_speed_type speed_type)
>> > >  {
>> > > @@ -952,10 +1049,19 @@ static unsigned int qca_get_speed(struct
>> > > hci_uart *hu,
>> > >
>> > >  static int qca_check_speeds(struct hci_uart *hu)
>> > >  {
>> > > -	/* One or the other speeds should be non zero. */
>> > > -	if (!qca_get_speed(hu, QCA_INIT_SPEED) &&
>> > > -	    !qca_get_speed(hu, QCA_OPER_SPEED))
>> > > +	struct qca_serdev *qcadev;
>> > > +
>> > > +	qcadev = serdev_device_get_drvdata(hu->serdev);
>> > > +	if ((qcadev->btsoc_type == QCA_WCN3990 &&
>> > > +	    !qca_get_speed(hu, QCA_INIT_SPEED)) ||
>> > > +	    !qca_get_speed(hu, QCA_OPER_SPEED)) {
>> > > +		/* Both INIT and OPER speed should be non zero. */
>> > >  		return -EINVAL;
>> > > +	} else if (!qca_get_speed(hu, QCA_INIT_SPEED) &&
>> > > +		    !qca_get_speed(hu, QCA_OPER_SPEED)) {
>> > > +		/* One or the other speeds should be non zero. */
>> > > +		return -EINVAL;
>> > > +	}
>> >
>> > It is questionable if the comments are useful, they are basically
>> > stating the same as the conditions.
>> >
>> > if-else statements with a single statement in all branches shouldn't
>> > use curly braces. Personally I don't dislike them in this case with
>> > the multi-line conditions, but in principle they shoulnd't be there.
>> >
>> > nit: this would be easier to read with nested if statements:
>> >
>> > 	if (qcadev->btsoc_type == QCA_WCN3990)
>> > 		if (!qca_get_speed(hu, QCA_INIT_SPEED)) ||
>> > 		    !qca_get_speed(hu, QCA_OPER_SPEED))
>> > 			return -EINVAL;
>> > 	else
>> > 		if (!qca_get_speed(hu, QCA_INIT_SPEED) &&
>> > 		    !qca_get_speed(hu, QCA_OPER_SPEED)) {
>> > 			return -EINVAL;
>> >
>> > >  	return 0;
>> > >  }
>> > > @@ -986,15 +1092,40 @@ static int qca_set_speed(struct hci_uart *hu,
>> > > enum qca_speed_type speed_type)
>> > >  	return 0;
>> > >  }
>> > >
>> > > +static int qca_wcn3990_init(struct hci_uart *hu)
>> > > +{
>> > > +	int ret;
>> > > +
>> > > +	ret = qca_send_vendor_cmd(hu->hdev, QCA_WCN3990_POWERON_PULSE);
>> > > +	if (ret)
>> > > +		return ret;
>> > > +
>> > > +	/* Wait for 100 ms for SoC to boot up */
>> > > +	msleep(100);
>> > > +	serdev_device_close(hu->serdev);
>> > > +	ret = serdev_device_open(hu->serdev);
>> > > +	if (ret) {
>> > > +		bt_dev_err(hu->hdev, "failed to open port");
>> > > +		return ret;
>> > > +	}
>> > > +
>> > > +	/* Setup initial baudrate */
>> > > +	qca_set_speed(hu, QCA_INIT_SPEED);
>> >
>> > The comment is a bit redundant.
>> >
>> > > +	hci_uart_set_flow_control(hu, false);
>> > > +
>> > > +	return 0;
>> > > +}
>> > > +
>> > >  static int qca_setup(struct hci_uart *hu)
>> > >  {
>> > >  	struct hci_dev *hdev = hu->hdev;
>> > >  	struct qca_data *qca = hu->priv;
>> > >  	unsigned int speed, qca_baudrate = QCA_BAUDRATE_115200;
>> > > +	struct qca_serdev *qcadev;
>> > >  	int ret;
>> > >  	int soc_ver = 0;
>> > >
>> > > -	bt_dev_info(hdev, "ROME setup");
>> > > +	qcadev = serdev_device_get_drvdata(hu->serdev);
>> > >
>> > >  	ret = qca_check_speeds(hu);
>> > >  	if (ret)
>> > > @@ -1006,6 +1137,21 @@ static int qca_setup(struct hci_uart *hu)
>> > >  	/* Setup initial baudrate */
>> > >  	qca_set_speed(hu, QCA_INIT_SPEED);
>> > >
>> > > +	if (qcadev->btsoc_type == QCA_WCN3990) {
>> > > +		bt_dev_dbg(hdev, "setting up wcn3990");
>> >
>> > Seems like this should be bt_dev_info() for consistency with Rome.
>> >
>> > > +		ret = qca_wcn3990_init(hu);
>> > > +		if (ret)
>> > > +			return ret;
>> > > +
>> > > +		ret = qca_read_soc_version(hdev, &soc_ver);
>> > > +		if (ret < 0 || soc_ver == 0) {
>> > > +			bt_dev_err(hdev, "Failed to get version %d", ret);
>> > > +			return ret;
>> > > +		}
>> > > +	} else {
>> > > +		bt_dev_info(hdev, "ROME setup");
>> > > +	}
>> > > +
>> > >  	/* Setup user speed if needed */
>> > >  	speed = qca_get_speed(hu, QCA_OPER_SPEED);
>> > >  	if (speed) {
>> > > @@ -1016,16 +1162,18 @@ static int qca_setup(struct hci_uart *hu)
>> > >  		qca_baudrate = qca_get_baudrate_value(speed);
>> > >  	}
>> > >
>> > > -	/* Get QCA version information */
>> > > -	ret = qca_read_soc_version(hdev, &soc_ver);
>> > > -	if (ret < 0 || soc_ver == 0) {
>> > > -		bt_dev_err(hdev, "QCA Failed to get version (%d)", ret);
>> > > -		return ret;
>> > > +	if (!soc_ver) {
>> >
>> >
>> > The following would probably be clearer:
>> >
>> >   	if (qcadev->btsoc_type != QCA_WCN3990) {
>> >
>> > > +		/* Get QCA version information */
>> > > +		ret = qca_read_soc_version(hdev, &soc_ver);
>> > > +		if (ret < 0 || soc_ver == 0) {
>> > > +			bt_dev_err(hdev, "QCA Failed to get version (%d)", ret);
>> > > +			return ret;
>> > > +		}
>> > > +		bt_dev_info(hdev, "QCA controller version 0x%08x", soc_ver);
>> >
>> > Couldn't we hide some of this in qca_read_soc_version()?
>> >
>> > qca_read_soc_version() could at least do the logging in the error case
>> > (the 'soc_ver == 0' case could be handled there as well), which would
>> > leave us with:
>> >
>> >
>> > 	ret = qca_read_soc_version(hdev, &soc_ver);
>> > 	if (ret)
>> > 		return ret;
>> >
>> > And the same above.
>> >
>> > The controller version can be logged outside of the branch for both
>> > wcn3990 and Rome.
>> >
>> > >  	}
>> > > -	bt_dev_info(hdev, "QCA controller version 0x%08x", soc_ver);
>> > >
>> > >  	/* Setup patch / NVM configurations */
>> > > -	ret = qca_uart_setup(hdev, qca_baudrate, QCA_ROME, soc_ver);
>> > > +	ret = qca_uart_setup(hdev, qca_baudrate, qcadev->btsoc_type,
>> > > soc_ver);
>> > >  	if (!ret) {
>> > >  		set_bit(STATE_IN_BAND_SLEEP_ENABLED, &qca->flags);
>> > >  		qca_debugfs_init(hdev);
>> > > @@ -1061,9 +1209,123 @@ static struct hci_uart_proto qca_proto = {
>> > >  	.dequeue	= qca_dequeue,
>> > >  };
>> > >
>> > > +static const struct qca_vreg_data qca_soc_data = {
>> > > +	.soc_type = QCA_WCN3990,
>> > > +	.vregs = (struct qca_vreg []) {
>> > > +		{ "vddio",   1800000, 1800000,  0 },
>> > > +		{ "vddxo",   1800000, 1800000,  1 },
>> > > +		{ "vddrf",   1304000, 1304000,  1 },
>> > > +		{ "vddch0",  3000000, 3312000,  1 },
>> > > +	},
>> > > +	.num_vregs = 4,
>> > > +};
>> >
>> > I didn't chime in earlier in the discussion with Stephen on the
>> > regulators (https://patchwork.kernel.org/patch/10467911/), however I
>> > agree with him that specifying at load of 1uA doesn't seem to make
>> > much sense. What would happen if the load remained unspecified (or 0)?
>> 
>> [Bala]: On RPMh based designs, calling a regulator_set_load with a 
>> non-zero
>> value moves the regulator to NPM (high power mode) which is required 
>> for BT
>> communication.
>>         and call with 0uA  moves it to low-power-mode (vote from 
>> APPS).
>> Basically whenever BT wants these in NPM  it calls set-load with 1.
> 
> This assumes that the chip is powered by RPMh regulators, which
> depending on the system may or may not be true. On systems with other
> regulators you are telling the regulator that the max load is 1uA, and
> the regulator might comply, going in a low power mode that does not
> provide enough current for the chip to operate properly.
> 
> Doesn't the datasheet of the chip specify actual max currents for
> these consumers that could be used instead of the bogus 1uA? If not
> you could ask the designers of the chip for an approximate number
> (preferably erring on the upper side).

[Bala]: will check and update you on this.

-- 
Regards
Balakrishna.