[PATCH v4 4/8] thermal: qcom: tsens-8960: Add support for 8960 family of SoCs

Rajendra Nayak rnayak at codeaurora.org
Thu Nov 5 00:05:54 PST 2015


On 11/03/2015 02:21 AM, Eduardo Valentin wrote:
> On Fri, Oct 09, 2015 at 03:11:06PM +0530, Rajendra Nayak wrote:
>> 8960 family of SoCs have the TSENS device as part of GCC, hence
>> the driver probes the virtual child device created by GCC and
>> uses the parent to extract all DT properties and reuses the GCC
>> regmap.
>>
>> Also GCC/TSENS are part of a  domain thats not always ON.
>> Hence add .suspend and .resume hooks to save and restore some of
>> the inited register context.
>>
>> Also 8960 family have some of the TSENS init sequence thats
>> required to be done by the HLOS driver (some later versions of TSENS
>> do not export these registers to non-secure world, and hence need
>> these initializations to be done by secure bootloaders)
>>
>> 8660 from the same family has just one sensor and hence some register
>> offset/layout differences which need special handling in the driver.
>>
>> Based on the original code from Siddartha Mohanadoss, Stephen Boyd and
>> Narendran Rajan.
>>
>> Signed-off-by: Rajendra Nayak <rnayak at codeaurora.org>
>> ---
>>  drivers/thermal/qcom/Makefile     |   2 +-
>>  drivers/thermal/qcom/tsens-8960.c | 291 ++++++++++++++++++++++++++++++++++++++
>>  drivers/thermal/qcom/tsens.c      |  13 +-
>>  drivers/thermal/qcom/tsens.h      |   2 +-
>>  4 files changed, 302 insertions(+), 6 deletions(-)
>>  create mode 100644 drivers/thermal/qcom/tsens-8960.c
>>
>> diff --git a/drivers/thermal/qcom/Makefile b/drivers/thermal/qcom/Makefile
>> index a471100..f3cefd1 100644
>> --- a/drivers/thermal/qcom/Makefile
>> +++ b/drivers/thermal/qcom/Makefile
>> @@ -1,2 +1,2 @@
>>  obj-$(CONFIG_QCOM_TSENS)	+= qcom_tsens.o
>> -qcom_tsens-y			+= tsens.o tsens-common.o tsens-8916.o tsens-8974.o
>> +qcom_tsens-y			+= tsens.o tsens-common.o tsens-8916.o tsens-8974.o tsens-8960.o
>> diff --git a/drivers/thermal/qcom/tsens-8960.c b/drivers/thermal/qcom/tsens-8960.c
>> new file mode 100644
>> index 0000000..00f45e7
>> --- /dev/null
>> +++ b/drivers/thermal/qcom/tsens-8960.c
>> @@ -0,0 +1,291 @@
>> +/*
>> + * Copyright (c) 2015, The Linux Foundation. All rights reserved.
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License version 2 and
>> + * only 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.
>> + *
>> + */
>> +
>> +#include <linux/platform_device.h>
>> +#include <linux/delay.h>
>> +#include <linux/bitops.h>
>> +#include <linux/regmap.h>
>> +#include "tsens.h"
>> +
>> +#define CAL_MDEGC		30000
>> +
>> +#define CONFIG_ADDR		0x3640
>> +#define CONFIG_ADDR_8660	0x3620
>> +/* CONFIG_ADDR bitmasks */
>> +#define CONFIG			0x9b
>> +#define CONFIG_MASK		0xf
>> +#define CONFIG_8660		1
>> +#define CONFIG_SHIFT_8660	28
>> +#define CONFIG_MASK_8660	(3 << CONFIG_SHIFT_8660)
>> +
>> +#define STATUS_CNTL_ADDR_8064	0x3660
>> +#define CNTL_ADDR		0x3620
>> +/* CNTL_ADDR bitmasks */
>> +#define EN			BIT(0)
>> +#define SW_RST			BIT(1)
>> +#define SENSOR0_EN		BIT(3)
>> +#define SLP_CLK_ENA		BIT(26)
>> +#define SLP_CLK_ENA_8660	BIT(24)
>> +#define MEASURE_PERIOD		1
>> +#define SENSOR0_SHIFT		3
>> +
>> +/* INT_STATUS_ADDR bitmasks */
>> +#define MIN_STATUS_MASK		BIT(0)
>> +#define LOWER_STATUS_CLR	BIT(1)
>> +#define UPPER_STATUS_CLR	BIT(2)
>> +#define MAX_STATUS_MASK		BIT(3)
>> +
>> +#define THRESHOLD_ADDR		0x3624
>> +/* THRESHOLD_ADDR bitmasks */
>> +#define THRESHOLD_MAX_LIMIT_SHIFT	24
>> +#define THRESHOLD_MIN_LIMIT_SHIFT	16
>> +#define THRESHOLD_UPPER_LIMIT_SHIFT	8
>> +#define THRESHOLD_LOWER_LIMIT_SHIFT	0
>> +
>> +/* Initial temperature threshold values */
>> +#define LOWER_LIMIT_TH		0x50
>> +#define UPPER_LIMIT_TH		0xdf
>> +#define MIN_LIMIT_TH		0x0
>> +#define MAX_LIMIT_TH		0xff
>> +
>> +#define S0_STATUS_ADDR		0x3628
>> +#define INT_STATUS_ADDR		0x363c
>> +#define TRDY_MASK		BIT(7)
>> +
>> +static int suspend_8960(struct tsens_device *tmdev)
>> +{
>> +	int ret;
>> +	unsigned int mask;
>> +	struct regmap *map = tmdev->map;
>> +
>> +	ret = regmap_read(map, THRESHOLD_ADDR, &tmdev->ctx.threshold);
>> +	if (ret)
>> +		return ret;
>> +
>> +	ret = regmap_read(map, CNTL_ADDR, &tmdev->ctx.control);
>> +	if (ret)
>> +		return ret;
>> +
>> +	if (tmdev->num_sensors > 1)
>> +		mask = SLP_CLK_ENA | EN;
>> +	else
>> +		mask = SLP_CLK_ENA_8660 | EN;
>> +
>> +	ret = regmap_update_bits(map, CNTL_ADDR, mask, 0);
>> +	if (ret)
>> +		return ret;
>> +
>> +	tmdev->trdy = false;
>> +
>> +	return 0;
>> +}
>> +
>> +static int resume_8960(struct tsens_device *tmdev)
>> +{
>> +	int ret;
>> +	unsigned long reg_cntl;
>> +	struct regmap *map = tmdev->map;
>> +
>> +	ret = regmap_update_bits(map, CNTL_ADDR, SW_RST, SW_RST);
>> +	if (ret)
>> +		return ret;
>> +
>> +	/*
>> +	 * Separate CONFIG restore is not needed only for 8660 as
>> +	 * config is part of CTRL Addr and its restored as such
>> +	 */
>> +	if (tmdev->num_sensors > 1) {
>> +		ret = regmap_update_bits(map, CONFIG_ADDR, CONFIG_MASK, CONFIG);
>> +		if (ret)
>> +			return ret;
>> +	}
>> +
>> +	ret = regmap_write(map, THRESHOLD_ADDR, tmdev->ctx.threshold);
>> +	if (ret)
>> +		return ret;
>> +
>> +	ret = regmap_write(map, CNTL_ADDR, tmdev->ctx.control);
>> +	if (ret)
>> +		return ret;
>> +
>> +	reg_cntl = tmdev->ctx.control;
>> +	reg_cntl >>= SENSOR0_SHIFT;
> 
> What the above two lines are doing? are they no-ops?

Looks like some leftover bits which I should have removed.
Will remove it.

> 
>> +
>> +	return 0;
>> +}
>> +
>> +static int enable_8960(struct tsens_device *tmdev, int id)
>> +{
>> +	int ret;
>> +	u32 reg, mask;
>> +
>> +	ret = regmap_read(tmdev->map, CNTL_ADDR, &reg);
>> +	if (ret)
>> +		return ret;
>> +
>> +	mask = BIT(id + SENSOR0_SHIFT);
>> +	ret = regmap_write(tmdev->map, CNTL_ADDR, reg | SW_RST);
>> +	if (ret)
>> +		return ret;
>> +
>> +	if (tmdev->num_sensors > 1)
>> +		reg |= mask | SLP_CLK_ENA | EN;
>> +	else
>> +		reg |= mask | SLP_CLK_ENA_8660 | EN;
>> +
>> +	tmdev->trdy = false;
>> +	ret = regmap_write(tmdev->map, CNTL_ADDR, reg);
>> +	if (ret)
>> +		return ret;
>> +
>> +	return 0;
>> +}
>> +
>> +static void disable_8960(struct tsens_device *tmdev)
>> +{
>> +	int ret;
>> +	u32 reg_cntl;
>> +	u32 mask;
>> +
>> +	mask = GENMASK(tmdev->num_sensors - 1, 0);
>> +	mask <<= SENSOR0_SHIFT;
>> +	mask |= EN;
>> +
>> +	ret = regmap_read(tmdev->map, CNTL_ADDR, &reg_cntl);
>> +	if (ret)
>> +		return;
>> +
>> +	reg_cntl &= ~mask;
>> +
>> +	if (tmdev->num_sensors > 1)
>> +		reg_cntl &= ~SLP_CLK_ENA;
>> +	else
>> +		reg_cntl &= ~SLP_CLK_ENA_8660;
>> +
>> +	regmap_write(tmdev->map, CNTL_ADDR, reg_cntl);
>> +}
>> +
>> +static int init_8960(struct tsens_device *tmdev)
>> +{
>> +	int ret, i;
>> +	u32 reg_cntl;
>> +
>> +	tmdev->map = dev_get_regmap(tmdev->dev, NULL);
>> +	if (!tmdev->map)
>> +		return -ENODEV;
>> +
>> +	/*
>> +	 * The status registers for each sensor are discontiguous
>> +	 * because some SoCs have 5 sensors while others have more
>> +	 * but the control registers stay in the same place, i.e
>> +	 * directly after the first 5 status registers.
>> +	 */
>> +	for (i = 0; i < tmdev->num_sensors; i++) {
>> +		if (i >= 5)
>> +			tmdev->sensor[i].status = S0_STATUS_ADDR + 40;
>> +		tmdev->sensor[i].status += i * 4;
>> +	}
>> +
>> +	reg_cntl = SW_RST;
>> +	ret = regmap_update_bits(tmdev->map, CNTL_ADDR, SW_RST, reg_cntl);
>> +	if (ret)
>> +		return ret;
>> +
>> +	if (tmdev->num_sensors > 1) {
>> +		reg_cntl |= SLP_CLK_ENA | (MEASURE_PERIOD << 18);
>> +		reg_cntl &= ~SW_RST;
>> +		ret = regmap_update_bits(tmdev->map, CONFIG_ADDR,
>> +					 CONFIG_MASK, CONFIG);
>> +	} else {
>> +		reg_cntl |= SLP_CLK_ENA_8660 | (MEASURE_PERIOD << 16);
>> +		reg_cntl &= ~CONFIG_MASK_8660;
>> +		reg_cntl |= CONFIG_8660 << CONFIG_SHIFT_8660;
>> +	}
>> +
>> +	reg_cntl |= GENMASK(tmdev->num_sensors - 1, 0) << SENSOR0_SHIFT;
>> +	ret = regmap_write(tmdev->map, CNTL_ADDR, reg_cntl);
>> +	if (ret)
>> +		return ret;
>> +
>> +	reg_cntl |= EN;
>> +	ret = regmap_write(tmdev->map, CNTL_ADDR, reg_cntl);
>> +	if (ret)
>> +		return ret;
>> +
>> +	return 0;
>> +}
>> +
>> +static int calibrate_8960(struct tsens_device *tmdev)
>> +{
>> +	int i;
>> +	char *data;
>> +
>> +	ssize_t num_read = tmdev->num_sensors;
>> +	struct tsens_sensor *s = tmdev->sensor;
>> +
>> +	data = qfprom_read(tmdev->dev, "calib");
>> +	if (IS_ERR(data))
>> +		data = qfprom_read(tmdev->dev, "calib_backup");
>> +	if (IS_ERR(data))
>> +		return PTR_ERR(data);
> 
> Oh.. so calibration can fail on prom reads...
> 
>> +
>> +	for (i = 0; i < num_read; i++, s++)
>> +		s->offset = CAL_MDEGC - s->slope * data[i];
>> +
>> +	return 0;
>> +}
>> +
>> +/* Temperature on y axis and ADC-code on x-axis */
>> +static inline int code_to_mdegC(u32 adc_code, const struct tsens_sensor *s)
>> +{
>> +	return adc_code * s->slope + s->offset;
>> +}
>> +
>> +static int get_temp_8960(struct tsens_device *tmdev, int id, int *temp)
>> +{
>> +	int ret;
>> +	u32 code, trdy;
>> +	const struct tsens_sensor *s = &tmdev->sensor[id];
>> +
>> +	if (!tmdev->trdy) {
>> +		ret = regmap_read(tmdev->map, INT_STATUS_ADDR, &trdy);
>> +		if (ret)
>> +			return ret;
>> +		while (!(trdy & TRDY_MASK)) {
>> +			usleep_range(1000, 1100);
>> +			regmap_read(tmdev->map, INT_STATUS_ADDR, &trdy);
> 		Is there a maximum amount of tries?
> 
> 		Would that be any case in which this would be stuck busy
> 		looping ?

right, I'll put some sensible timeout in.

>> +		}
>> +		tmdev->trdy = true;
>> +	}
>> +
>> +	ret = regmap_read(tmdev->map, s->status, &code);
>> +	if (ret)
>> +		return ret;
>> +
>> +	*temp = code_to_mdegC(code, s);
>> +
>> +	dev_dbg(tmdev->dev, "Sensor%d temp is: %d", id, *temp);
> 
> I suppose the thermal tracing could be probably better than a dev_dbg?

Okay, I'll take a look at thermal tracing.

> 
>> +
>> +	return 0;
>> +}
>> +
>> +const struct tsens_ops ops_8960 = {
>> +	.init		= init_8960,
>> +	.calibrate	= calibrate_8960,
>> +	.get_temp	= get_temp_8960,
>> +	.enable		= enable_8960,
>> +	.disable	= disable_8960,
>> +	.suspend	= suspend_8960,
>> +	.resume		= resume_8960,
>> +};
>> diff --git a/drivers/thermal/qcom/tsens.c b/drivers/thermal/qcom/tsens.c
>> index 06ed0c6..d640e8f 100644
>> --- a/drivers/thermal/qcom/tsens.c
>> +++ b/drivers/thermal/qcom/tsens.c
>> @@ -119,7 +119,11 @@ static int tsens_probe(struct platform_device *pdev)
>>  	struct tsens_device *tmdev;
>>  	const struct of_device_id *id;
>>  
>> -	dev = &pdev->dev;
>> +	if (pdev->dev.of_node)
>> +		dev = &pdev->dev;
>> +	else
>> +		dev = pdev->dev.parent;
>> +
> 
> Should this be part of the patch that adds tsens.c?

This is needed only because tsens is part of gcc in 8960.
So I kept the change here so its easier to co-relate why
its needed.

regards,
Rajendra
> 
>>  	np = dev->of_node;
>>  
>>  	num = of_property_count_u32_elems(np, "qcom,tsens-slopes");
>> @@ -141,10 +145,11 @@ static int tsens_probe(struct platform_device *pdev)
>>  					   &s->slope);
>>  
>>  	id = of_match_node(tsens_table, np);
>> -	if (!id)
>> -		return -ENODEV;
>> +	if (id)
>> +		tmdev->ops = id->data;
>> +	else
>> +		tmdev->ops = &ops_8960;
>>  
>> -	tmdev->ops = id->data;
>>  	if (!tmdev->ops || !tmdev->ops->init || !tmdev->ops->calibrate ||
>>  	    !tmdev->ops->get_temp)
>>  		return -EINVAL;
>> diff --git a/drivers/thermal/qcom/tsens.h b/drivers/thermal/qcom/tsens.h
>> index 00183db..10e2e84 100644
>> --- a/drivers/thermal/qcom/tsens.h
>> +++ b/drivers/thermal/qcom/tsens.h
>> @@ -64,6 +64,6 @@ void compute_intercept_slope(struct tsens_device *, u32 *, u32 *, u32);
>>  int init_common(struct tsens_device *);
>>  int get_temp_common(struct tsens_device *, int, int *);
>>  
>> -extern const struct tsens_ops ops_8916, ops_8974;
>> +extern const struct tsens_ops ops_8960, ops_8916, ops_8974;
>>  
>>  #endif /* __QCOM_TSENS_H__ */
>> -- 
>> QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
>> of Code Aurora Forum, hosted by The Linux Foundation
>>

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by The Linux Foundation



More information about the linux-arm-kernel mailing list