Date:	Fri, 13 Jun 2014 09:03:01 -0700
From:	Guenter Roeck <linux@...ck-us.net>
To:	Heiko Schocher <hs@...x.de>, lm-sensors@...sensors.org
CC:	Jean Delvare <khali@...ux-fr.org>, linux-kernel@...r.kernel.org,
	devicetree@...r.kernel.org, linux-doc@...r.kernel.org
Subject: Re: [PATCH v2] hwmon: Driver for TI TMP103 temperature sensor

On 06/13/2014 12:10 AM, Heiko Schocher wrote:
> Driver for the TI TMP103.
>
> The TI TMP103 is similar to the TMP102.  It differs from the TMP102
> by having only 8 bit registers.
>
> Signed-off-by: Heiko Schocher <hs@...x.de>
>
Hello Heiko,

Couple of additional comments.

Thanks,
Guenter

> ---
>
> Cc: Jean Delvare <khali@...ux-fr.org>
> Cc: Guenter Roeck <linux@...ck-us.net>
> Cc: linux-kernel@...r.kernel.org
> Cc: devicetree@...r.kernel.org
> Cc: linux-doc@...r.kernel.org
>
> - change for v2:
>    - add comments from GuenterRoeck:
>      - remove Cc from commit subject
>      - add devicetree mailinglist
>      - move Documentation to Documentation/hwmon/tmp103
>      - remove devicetree bindings from Documentation
>      - add compatible string to
>        "Documentation/devicetree/bindings/i2c/trivial-devices.txt"
>      - remove CamelCase
>      - fix Codingstyle issues
>      - use ATTRIBUTE_GROUPS and devm_hwmon_device_register_with_groups()
>      - remove unsused define TMP103_CONFIG_RD_ONLY
>      - restore config register when exit()
>      - use regmap
>
>   .../devicetree/bindings/i2c/trivial-devices.txt    |   1 +
>   Documentation/hwmon/tmp103                         |  28 ++
>   drivers/hwmon/Kconfig                              |  10 +
>   drivers/hwmon/Makefile                             |   1 +
>   drivers/hwmon/tmp103.c                             | 304 +++++++++++++++++++++
>   5 files changed, 344 insertions(+)
>   create mode 100644 Documentation/hwmon/tmp103
>   create mode 100644 drivers/hwmon/tmp103.c
>
> diff --git a/Documentation/devicetree/bindings/i2c/trivial-devices.txt b/Documentation/devicetree/bindings/i2c/trivial-devices.txt
> index bef86e5..fc944e0 100644
> --- a/Documentation/devicetree/bindings/i2c/trivial-devices.txt
> +++ b/Documentation/devicetree/bindings/i2c/trivial-devices.txt
> @@ -83,5 +83,6 @@ stm,m41t80		M41T80 - SERIAL ACCESS RTC WITH ALARMS
>   taos,tsl2550		Ambient Light Sensor with SMBUS/Two Wire Serial Interface
>   ti,tsc2003		I2C Touch-Screen Controller
>   ti,tmp102		Low Power Digital Temperature Sensor with SMBUS/Two Wire Serial Interface
> +ti,tmp103		Low Power Digital Temperature Sensor with SMBUS/Two Wire Serial Interface
>   ti,tmp275		Digital Temperature Sensor
>   winbond,wpct301		i2c trusted platform module (TPM)
> diff --git a/Documentation/hwmon/tmp103 b/Documentation/hwmon/tmp103
> new file mode 100644
> index 0000000..559fea5
> --- /dev/null
> +++ b/Documentation/hwmon/tmp103
> @@ -0,0 +1,28 @@
> +Kernel driver tmp103
> +====================
> +
> +Supported chips:
> +  * Texas Instruments TMP103
> +    Prefix: 'tmp103'
> +    Addresses scanned: none
> +    Product info and datasheet: http://www.ti.com/product/tmp103
> +
> +Author:
> +	Heiko Schocher <hs@...x.de>
> +
> +Description
> +-----------
> +
> +The TMP103 is a digital output temperature sensor in a four-ball
> +wafer chip-scale package (WCSP). The TMP103 is capable of reading
> +temperatures to a resolution of 1°C. The TMP103 is specified for
> +operation over a temperature range of –40°C to +125°C.
> +
> +Resolution: 8 Bits
> +Accuracy: ±1°C Typ (–10°C to +100°C)
> +
> +The driver provides the common sysfs-interface for temperatures (see
> +Documentation/hwmon/sysfs-interface under Temperatures).
> +
> +please refer how to instantiate this driver:

s/please/Please/

> +Documentation/i2c/instantiating-devices
> diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
> index 0034316..0f44dbb 100644
> --- a/drivers/hwmon/Kconfig
> +++ b/drivers/hwmon/Kconfig
> @@ -1381,6 +1381,16 @@ config SENSORS_TMP102
>   	  This driver can also be built as a module.  If so, the module
>   	  will be called tmp102.
>
> +config SENSORS_TMP103
> +	tristate "Texas Instruments TMP103"
> +	depends on I2C
> +	help
> +	  If you say yes here you get support for Texas Instruments TMP103
> +	  sensor chips.
> +
> +	  This driver can also be built as a module.  If so, the module
> +	  will be called tmp103.
> +
>   config SENSORS_TMP401
>   	tristate "Texas Instruments TMP401 and compatibles"
>   	depends on I2C
> diff --git a/drivers/hwmon/Makefile b/drivers/hwmon/Makefile
> index 11798ad..8e2f6a2 100644
> --- a/drivers/hwmon/Makefile
> +++ b/drivers/hwmon/Makefile
> @@ -134,6 +134,7 @@ obj-$(CONFIG_SENSORS_SMSC47M192)+= smsc47m192.o
>   obj-$(CONFIG_SENSORS_AMC6821)	+= amc6821.o
>   obj-$(CONFIG_SENSORS_THMC50)	+= thmc50.o
>   obj-$(CONFIG_SENSORS_TMP102)	+= tmp102.o
> +obj-$(CONFIG_SENSORS_TMP103)	+= tmp103.o
>   obj-$(CONFIG_SENSORS_TMP401)	+= tmp401.o
>   obj-$(CONFIG_SENSORS_TMP421)	+= tmp421.o
>   obj-$(CONFIG_SENSORS_TWL4030_MADC)+= twl4030-madc-hwmon.o
> diff --git a/drivers/hwmon/tmp103.c b/drivers/hwmon/tmp103.c
> new file mode 100644
> index 0000000..01119a6
> --- /dev/null
> +++ b/drivers/hwmon/tmp103.c
> @@ -0,0 +1,304 @@
> +/*
> + * Texas Instruments TMP103 SMBus temperature sensor driver
> + * Copyright (C) 2014 Heiko Schocher <hs@...x.de>
> + *
> + * Based on:
> + * Texas Instruments TMP102 SMBus temperature sensor driver
> + *
> + * Copyright (C) 2010 Steven King <sfking@...dc.com>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * 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/module.h>
> +#include <linux/init.h>
> +#include <linux/slab.h>
> +#include <linux/i2c.h>
> +#include <linux/hwmon.h>
> +#include <linux/hwmon-sysfs.h>
> +#include <linux/err.h>
> +#include <linux/mutex.h>
> +#include <linux/device.h>
> +#include <linux/jiffies.h>
> +#include <linux/regmap.h>
> +
> +#define	DRIVER_NAME "tmp103"
> +
> +#define	TMP103_TEMP_REG			0x00
> +#define	TMP103_CONF_REG			0x01
> +#define	TMP103_TLOW_REG			0x02
> +#define	TMP103_THIGH_REG		0x03
> +
> +#define		TMP103_CONF_M0		0x01
> +#define		TMP103_CONF_M1		0x02
> +#define		TMP103_CONF_LC		0x04
> +#define		TMP103_CONF_FL		0x08
> +#define		TMP103_CONF_FH		0x10
> +#define		TMP103_CONF_CR0		0x20
> +#define		TMP103_CONF_CR1		0x40
> +#define		TMP103_CONF_ID		0x80
> +#define		TMP103_CONF_SD		(TMP103_CONF_M0 | TMP103_CONF_M1)
> +
No tab between #define and the actual definition, please.

> +struct tmp103 {
> +	struct i2c_client *client;

No longer needed. See below.

> +	struct device *hwmon_dev;
> +	struct regmap *regmap;
> +	struct mutex lock;
> +	unsigned int config_orig;
> +	unsigned long last_update;
> +	int temp[3];

Both no longer needed.

Actually, the entire structure is not really needed anymore with the changes suggested
below. The only value still needed would be regmap, and you could store that directly
in drvdata.

> +};
> +
> +static inline int tmp103_reg_to_mc(s8 val)
> +{
> +	return val * 1000;
> +}
> +
> +static inline u8 tmp103_mc_to_reg(int val)
> +{
> +	return DIV_ROUND_CLOSEST(val, 1000);
> +}
> +
> +static const u8 tmp103_reg[] = {
> +	TMP103_TEMP_REG,
> +	TMP103_TLOW_REG,
> +	TMP103_THIGH_REG,
> +};

No longer needed; you can put the register directly into sda->index.

> +
> +#define TMP103_INVALID	0xffff
> +
> +static ssize_t tmp103_show_temp(struct device *dev,
> +				struct device_attribute *attr,
> +				char *buf)
> +{
> +	struct sensor_device_attribute *sda = to_sensor_dev_attr(attr);
> +	struct i2c_client *client = to_i2c_client(dev);
> +	struct tmp103 *tmp103 = i2c_get_clientdata(client);

Too complicated. Attributes are now attached to the hwmon device, and tmp103
can be retrieved from there.

	struct tmp103 *tmp103 = dev_get_drvdata(dev);
or, with all the changes I suggested, even simpler
	struct regmap *regmap = dev_get_drvdata(dev);

> +	unsigned int status;
> +	int ret;
> +
> +	mutex_lock(&tmp103->lock);
> +	if (time_after(jiffies, tmp103->last_update + HZ / 3)) {
> +		int i;
> +
> +		for (i = 0; i < ARRAY_SIZE(tmp103->temp); ++i) {
> +			ret = regmap_read(tmp103->regmap, tmp103_reg[i],
> +					  &status);
> +			if (ret < 0)
> +				tmp103->temp[i] = TMP103_INVALID;

Good idea, but not really what we'd want to see. In this case the error
should be reported to use space, ie return the value of 'ret', not "invalid".
Also, you don't need to read all registers here, and just forget about
local caching (not worth it).

Suggested replacement below.

> +			else
> +				tmp103->temp[i] = tmp103_reg_to_mc(status);
> +		}
> +		tmp103->last_update = jiffies;
> +	}
> +	mutex_unlock(&tmp103->lock);
> +
> +	if (tmp103->temp[sda->index] != TMP103_INVALID)
> +		return sprintf(buf, "%d\n", tmp103->temp[sda->index]);
> +
> +	return sprintf(buf, "invalid\n");
> +}

static ssize_t tmp103_show_temp(struct device *dev,
				struct device_attribute *attr,
				char *buf)
{
	struct sensor_device_attribute *sda = to_sensor_dev_attr(attr);
	struct tmp103 *tmp103 = dev_get_drvdata(dev);	/* or struct regmap *regmap = dev_get_drvdata(dev); */
	unsigned int regval;
	int ret;

	ret = regmap_read(tmp103->regmap, sda->index, &regval);
	if (ret < 0)
		return ret;

	return return sprintf(buf, "%d\n", tmp103_reg_to_mc(regval));
}


> +
> +static ssize_t tmp103_set_temp(struct device *dev,
> +			       struct device_attribute *attr,
> +			       const char *buf, size_t count)
> +{
> +	struct sensor_device_attribute *sda = to_sensor_dev_attr(attr);
> +	struct i2c_client *client = to_i2c_client(dev);
> +	struct tmp103 *tmp103 = i2c_get_clientdata(client);

Same as above.
	struct tmp103 *tmp103 = dev_get_drvdata(dev);


> +	long val;
> +	int ret;
> +
> +	if (kstrtol(buf, 10, &val) < 0)
> +		return -EINVAL;
> +
> +	val = clamp_val(val, -55000, 127000);
> +	mutex_lock(&tmp103->lock);
> +	tmp103->temp[sda->index] = val;
> +	ret = regmap_write(tmp103->regmap, tmp103_reg[sda->index],
> +			   tmp103_mc_to_reg(val));

After dropping tmp103->temp[] you don't need the lock anymore,
which means it can go away completely.

> +	mutex_unlock(&tmp103->lock);
> +	return ret ? : count;
> +}
> +
> +static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, tmp103_show_temp, NULL , 0);
> +
As suggested above, instead of using an indirect index, use TMP103_TEMP_REG directly
to simplify the code.

> +static SENSOR_DEVICE_ATTR(temp1_min, S_IWUSR | S_IRUGO, tmp103_show_temp,
> +			  tmp103_set_temp, 1);
> +
> +static SENSOR_DEVICE_ATTR(temp1_max, S_IWUSR | S_IRUGO, tmp103_show_temp,
> +			  tmp103_set_temp, 2);
> +
> +static struct attribute *tmp103_attrs[] = {
> +	&sensor_dev_attr_temp1_input.dev_attr.attr,
> +	&sensor_dev_attr_temp1_min.dev_attr.attr,
> +	&sensor_dev_attr_temp1_max.dev_attr.attr,
> +	NULL
> +};
> +ATTRIBUTE_GROUPS(tmp103);
> +
> +static struct regmap_config tmp103_regmap_config = {
> +	.reg_bits = 8,
> +	.val_bits = 8,
> +	.max_register = TMP103_THIGH_REG,

You can also add a 'volatile' function to tell regmap which registers are volatile.

	.volatile_reg = tmp103_is_volatile,

with

static bool tmp103_regmap_is_volatile(struct device *dev, unsigned int reg)
{
	return reg == TMP103_TEMP_REG;
}

With this, only the temperature register will be read after the first initialization.
If you add alarm attributes, you'd have to add the configuration register, but not
for now.

> +};
> +
> +#define TMP103_CONFIG		(TMP103_CONF_CR1 | TMP103_CONF_M1)
> +
Can you move this to the other definitions ? TMP103_CONF_SD is already there,
so this would be more consistent.

> +static int tmp103_probe(struct i2c_client *client,
> +			const struct i2c_device_id *id)
> +{
> +	struct device *dev = &client->dev;
> +	struct device *hwmon_dev;
> +	struct tmp103 *tmp103;
> +	int ret;
> +	unsigned int status;
> +
> +	if (!i2c_check_functionality(client->adapter,
> +				     I2C_FUNC_SMBUS_BYTE_DATA)) {
> +		dev_err(&client->dev,
> +			"adapter doesn't support SMBus byte transactions\n");
> +		return -ENODEV;
> +	}
> +
> +	tmp103 = devm_kzalloc(&client->dev, sizeof(*tmp103), GFP_KERNEL);
> +	if (!tmp103)
> +		return -ENOMEM;
> +
> +	i2c_set_clientdata(client, tmp103);
> +	tmp103->client = client;

Both no longer needed.

> +
> +	tmp103->regmap = devm_regmap_init_i2c(client, &tmp103_regmap_config);
> +	if (IS_ERR(tmp103->regmap)) {
> +		dev_err(dev, "failed to allocate register map\n");
> +		return PTR_ERR(tmp103->regmap);
> +	}
> +
> +	ret = regmap_read(tmp103->regmap, TMP103_CONF_REG, &status);
> +	if (ret < 0)
> +		goto fail_init_regmap;
> +
> +	tmp103->config_orig = status;
> +	ret = regmap_write(tmp103->regmap, TMP103_CONF_REG, TMP103_CONFIG);
> +	if (ret < 0) {
> +		dev_err(&client->dev, "error writing config register\n");
> +		goto fail_restore_config;
> +	}
> +
> +	ret = regmap_read(tmp103->regmap, TMP103_CONF_REG, &status);
> +	if (ret < 0) {
> +		dev_err(&client->dev, "error reading config register\n");
> +		goto fail_restore_config;
> +	}
> +	if (status != TMP103_CONFIG) {
> +		dev_err(&client->dev, "config settings did not stick\n");
> +		ret = -ENODEV;
> +		goto fail_restore_config;
> +	}

The above write/read sequence can be made conditional.
If (config_orig & 0x67) == TMP103_CONFIG, there is no need to update the
configuration register.

Also, I just realized that the current code won't work: The FH and FL bits
are not controlled by you, but by the chip. So the verification (as written)
will fail if you have an alarm condition. I would suggest to drop the
verification entirely; I don't think it adds much if any value.
At the very least, you'll have to mask the result against 0x67.

Actually, I would suggest to drop config_orig entirely. Then you can
use regmap_update_bits() to set the configuration. Also, you can then use
devm_hwmon_device_register_with_groups(), and drop the remove function
as well.

> +	tmp103->last_update = jiffies - HZ;

No longer needed.

> +	mutex_init(&tmp103->lock);
> +
> +	hwmon_dev = hwmon_device_register_with_groups(dev, client->name,
> +						      tmp103, tmp103_groups);
> +	if (IS_ERR(hwmon_dev)) {
> +		dev_dbg(dev, "unable to register hwmon device\n");


> +		status = PTR_ERR(hwmon_dev);
> +		goto fail_restore_config;
> +	}
> +	tmp103->hwmon_dev = hwmon_dev;
> +
> +	return 0;
> +
> +fail_restore_config:
> +	regmap_write(tmp103->regmap, TMP103_CONF_REG, tmp103->config_orig);
> +fail_init_regmap:
> +	regmap_exit(tmp103->regmap);

This is not needed since you used the devm_ initialization function above.
In fact, it is likely wrong.

If you drop restoring the configuration register, the entire sequence can become

	hwmon_dev = devm_hwmon_device_register_with_groups(dev, client->name,
						           tmp103, tmp103_groups);
	return PTR_ERR_OR_ZERO(hwmon_dev);

> +	return ret;
> +}
> +
> +static int tmp103_remove(struct i2c_client *client)
> +{
> +	struct tmp103 *tmp103 = i2c_get_clientdata(client);
> +	int ret;
> +
> +	ret = regmap_write(tmp103->regmap, TMP103_CONF_REG,
> +			   tmp103->config_orig);

Only needed if (config_orig & 0x67) != TMP103_CONFIG.

> +	if (ret < 0)
> +		dev_err(&client->dev, "error writing config register\n");
> +
I'd suggest to drop the error message. It does not really matter,
after all, and is quite unlikely.

> +	hwmon_device_unregister(tmp103->hwmon_dev);
> +
> +	return 0;
> +}
> +
> +#ifdef CONFIG_PM
> +static int tmp103_suspend(struct device *dev)
> +{
> +	struct i2c_client *client = to_i2c_client(dev);
> +	struct tmp103 *tmp103 = i2c_get_clientdata(client);
> +	unsigned int config;
> +	int ret;
> +
> +	ret = regmap_read(tmp103->regmap, TMP103_CONF_REG, &config);
> +	if (ret < 0)
> +		return ret;
> +
> +	config &= ~TMP103_CONF_SD;
> +	return regmap_write(tmp103->regmap, TMP103_CONF_REG, config);

You can use regmap_update_bits() here.

> +}
> +
> +static int tmp103_resume(struct device *dev)
> +{
> +	struct i2c_client *client = to_i2c_client(dev);
> +	struct tmp103 *tmp103 = i2c_get_clientdata(client);
> +	unsigned int config;
> +	int ret;
> +
> +	ret = regmap_read(tmp103->regmap, TMP103_CONF_REG, &config);
> +	if (ret < 0)
> +		return ret;
> +
> +	config |= TMP103_CONF_SD;
> +	return regmap_write(tmp103->regmap, TMP103_CONF_REG, config);

You can use regmap_update_bits() here.

> +}
> +
> +static const struct dev_pm_ops tmp103_dev_pm_ops = {
> +	.suspend	= tmp103_suspend,
> +	.resume		= tmp103_resume,
> +};
> +
> +#define TMP103_DEV_PM_OPS (&tmp103_dev_pm_ops)
> +#else
> +#define	TMP103_DEV_PM_OPS NULL
> +#endif /* CONFIG_PM */
> +
> +static const struct i2c_device_id tmp103_id[] = {
> +	{ DRIVER_NAME, 0 },
> +	{ }
> +};
> +MODULE_DEVICE_TABLE(i2c, tmp103_id);
> +
> +static struct i2c_driver tmp103_driver = {
> +	.driver = {
> +		.name	= DRIVER_NAME,
> +		.pm	= TMP103_DEV_PM_OPS,
> +	},
> +	.probe		= tmp103_probe,
> +	.remove		= tmp103_remove,
> +	.id_table	= tmp103_id,
> +};
> +
> +module_i2c_driver(tmp103_driver);
> +
> +MODULE_AUTHOR("Heiko Schocher <hs@...x.de>");
> +MODULE_DESCRIPTION("Texas Instruments TMP103 temperature sensor driver");
> +MODULE_LICENSE("GPL");
>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists