linux-hwmon.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Guenter Roeck <linux@roeck-us.net>
To: "Cormier, Jonathan" <jcormier@criticallink.com>,
	linux-hwmon@vger.kernel.org
Cc: John Pruitt <jpruitt@criticallink.com>,
	Jean Delvare <jdelvare@suse.com>,
	linux-kernel@vger.kernel.org, devicetree@vger.kernel.org,
	Rob Herring <robh+dt@kernel.org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
	Bob Duke <bduke@criticallink.com>
Subject: Re: [PATCH 2/2] hwmon: ltc2945: Allow setting shunt resistor
Date: Thu, 15 Dec 2022 07:11:21 -0800	[thread overview]
Message-ID: <601ddc9c-66a8-0ab4-69ef-ba4e9c0f9420@roeck-us.net> (raw)
In-Reply-To: <20221214220727.1350784-3-jcormier@criticallink.com>

On 12/14/22 14:07, Cormier, Jonathan wrote:
> From: John Pruitt <jpruitt@criticallink.com>
> 
> Added the ability to specify the value of the shunt resistor in the
> device tree instead of assuming it is 1 milliOhm. The value in the
> device tree has the name shunt-resistor-micro-ohms and the
> default value is 1000 micro-ohms in order to preserve the
> current behavior.
> 

Also needs a devicetree match table.

> Signed-off-by: "Cormier, Jonathan" <jcormier@criticallink.com>
> Signed-off-by: John Pruitt <jpruitt@criticallink.com>
> ---
>   drivers/hwmon/ltc2945.c | 86 +++++++++++++++++++++++++++++++----------
>   1 file changed, 65 insertions(+), 21 deletions(-)
> 
> diff --git a/drivers/hwmon/ltc2945.c b/drivers/hwmon/ltc2945.c
> index 9adebb59f604..4e8b92e53133 100644
> --- a/drivers/hwmon/ltc2945.c
> +++ b/drivers/hwmon/ltc2945.c
> @@ -58,6 +58,11 @@
>   #define CONTROL_MULT_SELECT	(1 << 0)
>   #define CONTROL_TEST_MODE	(1 << 4)
>   
> +struct ltc2945_data {
> +	struct regmap *regmap;
> +	u32 shunt_resistor; // number of microohms

No C++ comments in this code, please. Also, "number of" is not needed.
Something like "in microohms" or just "microohms" would be sufficient.

> +};
> +
>   static inline bool is_power_reg(u8 reg)
>   {
>   	return reg < LTC2945_SENSE_H;
> @@ -66,7 +71,9 @@ static inline bool is_power_reg(u8 reg)
>   /* Return the value from the given register in uW, mV, or mA */
>   static long long ltc2945_reg_to_val(struct device *dev, u8 reg)
>   {
> -	struct regmap *regmap = dev_get_drvdata(dev);
> +	struct ltc2945_data *data = dev_get_drvdata(dev);
> +	struct regmap *regmap = data->regmap;
> +	u32 shunt_resistor = data->shunt_resistor;
>   	unsigned int control;
>   	u8 buf[3];
>   	long long val;
> @@ -93,8 +100,11 @@ static long long ltc2945_reg_to_val(struct device *dev, u8 reg)
>   	case LTC2945_MIN_POWER_THRES_H:
>   		/*
>   		 * Convert to uW by assuming current is measured with
> -		 * an 1mOhm sense resistor, similar to current
> -		 * measurements.
> +		 * a 1000 microOhm sense resistor (aka shunt resistor)
> +		 * or what is specified in DT, similar to current
> +		 * measurements. The shunt_resistor value is in
> +		 * microOhms.

The units are already in the variable declaration. No need to repeat.
There is no need to replace "1 mOhm" with "1000 microOhm".

It might make sense to drop all those comments, add a kerneldoc comment
to struct ltc2945_data, and explain it all there.

> +		 *
>   		 * Control register bit 0 selects if voltage at SENSE+/VDD
>   		 * or voltage at ADIN is used to measure power.
>   		 */
> @@ -108,6 +118,8 @@ static long long ltc2945_reg_to_val(struct device *dev, u8 reg)
>   			/* 0.5 mV * 25 uV = 0.0125 uV resolution. */
>   			val = (val * 25LL) >> 1;
>   		}
> +		val *= 1000;
> +		val = DIV_ROUND_CLOSEST_ULL(val, shunt_resistor);
>   		break;
>   	case LTC2945_VIN_H:
>   	case LTC2945_MAX_VIN_H:
> @@ -132,12 +144,15 @@ static long long ltc2945_reg_to_val(struct device *dev, u8 reg)
>   	case LTC2945_MIN_SENSE_THRES_H:
>   		/*
>   		 * 25 uV resolution. Convert to current as measured with
> -		 * an 1 mOhm sense resistor, in mA. If a different sense
> -		 * resistor is installed, calculate the actual current by
> -		 * dividing the reported current by the sense resistor value
> -		 * in mOhm.
> +		 * an 1000 microOhm sense (or shunt) resistor, in mA.
> +		 * If a different shunt resistor is installed, calculate
> +		 * the actual current by dividing the reported current
> +		 * by the shunt resistor value in microOhms. The actual
> +		 * The actual shunt resistor value can be specified
> +		 * in the device tree.

All but the last sentence just replace 1 mOhm with 1000 microOhm, which
does not add any value. Applies to all similar changes; I won't comment
on it anymore below.

>   		 */
> -		val *= 25;
> +		val *= 25 * 1000;
> +		val = DIV_ROUND_CLOSEST_ULL(val, shunt_resistor);
>   		break;
>   	default:
>   		return -EINVAL;
> @@ -148,7 +163,9 @@ static long long ltc2945_reg_to_val(struct device *dev, u8 reg)
>   static int ltc2945_val_to_reg(struct device *dev, u8 reg,
>   			      unsigned long val)
>   {
> -	struct regmap *regmap = dev_get_drvdata(dev);
> +	struct ltc2945_data *data = dev_get_drvdata(dev);
> +	struct regmap *regmap = data->regmap;
> +	u32 shunt_resistor = data->shunt_resistor;
>   	unsigned int control;
>   	int ret;
>   
> @@ -160,8 +177,10 @@ static int ltc2945_val_to_reg(struct device *dev, u8 reg,
>   	case LTC2945_MIN_POWER_THRES_H:
>   		/*
>   		 * Convert to register value by assuming current is measured
> -		 * with an 1mOhm sense resistor, similar to current
> +		 * with a 1000 microOhm sense resistor, (aka shunt resistor)
> +		 * or what is specified in DT, similar to current
>   		 * measurements.
> +		 *
>   		 * Control register bit 0 selects if voltage at SENSE+/VDD
>   		 * or voltage at ADIN is used to measure power, which in turn
>   		 * determines register calculations.
> @@ -171,14 +190,16 @@ static int ltc2945_val_to_reg(struct device *dev, u8 reg,
>   			return ret;
>   		if (control & CONTROL_MULT_SELECT) {
>   			/* 25 mV * 25 uV = 0.625 uV resolution. */
> -			val = DIV_ROUND_CLOSEST(val, 625);
> +			val *= shunt_resistor;

This can now easily result in an overflow.

> +			val = DIV_ROUND_CLOSEST(val, 625 * 1000);
>   		} else {
>   			/*
>   			 * 0.5 mV * 25 uV = 0.0125 uV resolution.
>   			 * Divide first to avoid overflow;
>   			 * accept loss of accuracy.
>   			 */
> -			val = DIV_ROUND_CLOSEST(val, 25) * 2;
> +			val *= shunt_resistor;
> +			val = DIV_ROUND_CLOSEST(val, 25 * 1000) * 2;
>   		}
>   		break;
>   	case LTC2945_VIN_H:
> @@ -204,12 +225,14 @@ static int ltc2945_val_to_reg(struct device *dev, u8 reg,
>   	case LTC2945_MIN_SENSE_THRES_H:
>   		/*
>   		 * 25 uV resolution. Convert to current as measured with
> -		 * an 1 mOhm sense resistor, in mA. If a different sense
> -		 * resistor is installed, calculate the actual current by
> -		 * dividing the reported current by the sense resistor value
> -		 * in mOhm.
> +		 * a 1000 microOhm sense (or shunt) resistor, in mA. If
> +		 * a different shunt resistor is installed, calculate
> +		 * the actual current by dividing the reported current
> +		 * by the sense resistor value in microOhms. The actual
> +		 * shunt resistor value can be specified in the device tree.
>   		 */
> -		val = DIV_ROUND_CLOSEST(val, 25);
> +		val *= shunt_resistor;

Needs to ensure that there are no overflows. Try to write the maximum value
for unsigned long and see what happens.

> +		val = DIV_ROUND_CLOSEST(val, 25 * 1000);
>   		break;
>   	default:
>   		return -EINVAL;
> @@ -234,7 +257,8 @@ static ssize_t ltc2945_value_store(struct device *dev,
>   				   const char *buf, size_t count)
>   {
>   	struct sensor_device_attribute *attr = to_sensor_dev_attr(da);
> -	struct regmap *regmap = dev_get_drvdata(dev);
> +	struct ltc2945_data *data = dev_get_drvdata(dev);
> +	struct regmap *regmap = data->regmap;
>   	u8 reg = attr->index;
>   	unsigned long val;
>   	u8 regbuf[3];
> @@ -269,7 +293,8 @@ static ssize_t ltc2945_history_store(struct device *dev,
>   				     const char *buf, size_t count)
>   {
>   	struct sensor_device_attribute *attr = to_sensor_dev_attr(da);
> -	struct regmap *regmap = dev_get_drvdata(dev);
> +	struct ltc2945_data *data = dev_get_drvdata(dev);
> +	struct regmap *regmap = data->regmap; >   	u8 reg = attr->index;
>   	int num_regs = is_power_reg(reg) ? 3 : 2;
>   	u8 buf_min[3] = { 0xff, 0xff, 0xff };
> @@ -321,7 +346,8 @@ static ssize_t ltc2945_bool_show(struct device *dev,
>   				 struct device_attribute *da, char *buf)
>   {
>   	struct sensor_device_attribute *attr = to_sensor_dev_attr(da);
> -	struct regmap *regmap = dev_get_drvdata(dev);
> +	struct ltc2945_data *data = dev_get_drvdata(dev);
> +	struct regmap *regmap = data->regmap;
>   	unsigned int fault;
>   	int ret;
>   
> @@ -450,6 +476,14 @@ static int ltc2945_probe(struct i2c_client *client)
>   	struct device *dev = &client->dev;
>   	struct device *hwmon_dev;
>   	struct regmap *regmap;
> +	u32 shunt_resistor;
> +
drop this empty line

> +	struct ltc2945_data *data;
> +
> +	data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL);
> +	if (!data)
> +		return -ENOMEM;
> +	dev_set_drvdata(dev, data);
>   
>   	regmap = devm_regmap_init_i2c(client, &ltc2945_regmap_config);
>   	if (IS_ERR(regmap)) {
> @@ -457,11 +491,21 @@ static int ltc2945_probe(struct i2c_client *client)
>   		return PTR_ERR(regmap);
>   	}
>   
> +	data->regmap = regmap;
> +	if (of_property_read_u32(client->dev.of_node,
> +				 "shunt-resistor-micro-ohms", &shunt_resistor))

Please consider using device_property_read_u32().

> +		shunt_resistor = 1000; /* 1000 micro-Ohm if not set via DT */

"if not set via DT" -> "default value". With device_property_read_u32(),
this can be set by other means, not just DT.

> +
> +	if (shunt_resistor == 0)
> +		return -EINVAL;
> +
> +	data->shunt_resistor = shunt_resistor;
> +
>   	/* Clear faults */
>   	regmap_write(regmap, LTC2945_FAULT, 0x00);
>   
>   	hwmon_dev = devm_hwmon_device_register_with_groups(dev, client->name,
> -							   regmap,
> +							   data,
>   							   ltc2945_groups);
>   	return PTR_ERR_OR_ZERO(hwmon_dev);
>   }


  reply	other threads:[~2022-12-15 15:11 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-12-14 22:07 [PATCH 0/2] hwmon: ltc2945: Add binding and shunt resistor support Cormier, Jonathan
2022-12-14 22:07 ` [PATCH 1/2] dt-bindings: hwmon: adi,ltc2945: Add binding Cormier, Jonathan
2022-12-15  9:42   ` Krzysztof Kozlowski
     [not found]     ` <CADL8D3ZUE5WbV0oS6hEVUNh9asrhTKQeGR4McR6Kh6qykSFw=Q@mail.gmail.com>
2022-12-15 14:37       ` Jon Cormier
2022-12-14 22:07 ` [PATCH 2/2] hwmon: ltc2945: Allow setting shunt resistor Cormier, Jonathan
2022-12-15 15:11   ` Guenter Roeck [this message]
2022-12-15 19:42     ` Jon Cormier
2023-01-09 23:35   ` [PATCH v3 0/2] hwmon: ltc2945: Add binding and shunt resistor support Jonathan Cormier
2023-01-09 23:35     ` [PATCH v3 1/5] dt-bindings: hwmon: adi,ltc2945: Add binding Jonathan Cormier
2023-01-10  9:07       ` Krzysztof Kozlowski
2023-01-09 23:35     ` [PATCH v3 2/5] hwmon: ltc2945: Add devicetree match table Jonathan Cormier
2023-01-09 23:35     ` [PATCH v3 3/5] hwmon: ltc2945: Handle error case in ltc2945_value_store Jonathan Cormier
2023-01-10  0:04       ` Guenter Roeck
2023-01-10 18:19         ` Jon Cormier
2023-01-10 18:22           ` Guenter Roeck
2023-01-10 19:25             ` Jon Cormier
2023-01-12  0:44               ` Guenter Roeck
2023-01-18 18:32                 ` Jon Cormier
2023-01-09 23:35     ` [PATCH v3 4/5] hwmon: ltc2945: Allow setting shunt resistor Jonathan Cormier
2023-01-09 23:35     ` [PATCH v3 5/5] hwmon: ltc2945: Convert division to DIV_ROUND_CLOSEST_ULL Jonathan Cormier
2022-12-20  0:04 ` [PATCH v2 0/2] hwmon: ltc2945: Add binding and shunt resistor support Cormier, Jonathan
2022-12-20  0:04   ` [PATCH v2 1/4] dt-bindings: hwmon: adi,ltc2945: Add binding Cormier, Jonathan
2022-12-20 10:15     ` Krzysztof Kozlowski
2022-12-20 14:35       ` Jon Cormier
2022-12-20 14:46         ` Guenter Roeck
2022-12-20 21:47           ` Jon Cormier
2022-12-20  0:04   ` [PATCH v2 2/4] hwmon: ltc2945: Add devicetree match table Cormier, Jonathan
2022-12-28 16:43     ` Guenter Roeck
2022-12-20  0:04   ` [PATCH v2 3/4] hwmon: ltc2945: Allow setting shunt resistor Cormier, Jonathan
2022-12-28 16:49     ` Guenter Roeck
2022-12-20  0:04   ` [PATCH v2 4/4] hwmon: ltc2945: Fix possible overflows Cormier, Jonathan
2022-12-28 17:01     ` Guenter Roeck

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=601ddc9c-66a8-0ab4-69ef-ba4e9c0f9420@roeck-us.net \
    --to=linux@roeck-us.net \
    --cc=bduke@criticallink.com \
    --cc=devicetree@vger.kernel.org \
    --cc=jcormier@criticallink.com \
    --cc=jdelvare@suse.com \
    --cc=jpruitt@criticallink.com \
    --cc=krzysztof.kozlowski+dt@linaro.org \
    --cc=linux-hwmon@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=robh+dt@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).