lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Wed, 13 Sep 2023 17:45:52 +0200
From:   Sebastian Reichel <sebastian.reichel@...labora.com>
To:     Konrad Dybcio <konrad.dybcio@...aro.org>
Cc:     Rob Herring <robh+dt@...nel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        Conor Dooley <conor+dt@...nel.org>,
        Konrad Dybcio <konradybcio@...nel.org>,
        Marijn Suijten <marijn.suijten@...ainline.org>,
        Rob Herring <robh@...nel.org>, devicetree@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-pm@...r.kernel.org
Subject: Re: [PATCH RESEND v2 3/3] power: supply: Introduce MM8013 fuel gauge
 driver

Hi,

On Wed, Aug 23, 2023 at 04:36:15PM +0200, Konrad Dybcio wrote:
> Add a driver for the Mitsumi MM8013 fuel gauge. The driver is a vastly
> cleaned up and improved version of the one that shipped in some obscure
> Lenovo downstream kernel [1], with some register definitions borrowed from
> ChromeOS EC platform code [2].
> 
> [1] https://github.com/adazem009/kernel_lenovo_bengal/commit/b6b346427a871715709bd22aae449b9383f3b66b
> [2] https://chromium.googlesource.com/chromiumos/platform/ec/+/master/driver/battery/mm8013.h
> Signed-off-by: Konrad Dybcio <konrad.dybcio@...aro.org>
> ---
>  MAINTAINERS                   |   5 +
>  drivers/power/supply/Kconfig  |   9 ++
>  drivers/power/supply/Makefile |   1 +
>  drivers/power/supply/mm8013.c | 283 ++++++++++++++++++++++++++++++++++++++++++
>  4 files changed, 298 insertions(+)
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 5ce188b58eaa..ba5f075a2ca8 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -14296,6 +14296,11 @@ W:	https://linuxtv.org
>  T:	git git://linuxtv.org/media_tree.git
>  F:	drivers/media/radio/radio-miropcm20*
>  
> +MITSUMI MM8013 FG DRIVER
> +M:	Konrad Dybcio <konradybcio@...nel.org>
> +F:	Documentation/devicetree/bindings/power/supply/mitsumi,mm8013.yaml
> +F:	drivers/power/supply/mm8013.c
> +
>  MMP SUPPORT
>  R:	Lubomir Rintel <lkundrak@...sk>
>  L:	linux-arm-kernel@...ts.infradead.org (moderated for non-subscribers)
> diff --git a/drivers/power/supply/Kconfig b/drivers/power/supply/Kconfig
> index 663a1c423806..c19e8287d80f 100644
> --- a/drivers/power/supply/Kconfig
> +++ b/drivers/power/supply/Kconfig
> @@ -951,4 +951,13 @@ config CHARGER_QCOM_SMB2
>  	  adds support for the SMB2 switch mode battery charger found
>  	  in PMI8998 and related PMICs.
>  
> +config FUEL_GAUGE_MM8013
> +	tristate "Mitsumi MM8013 fuel gauge driver"
> +	depends on I2C
> +	help
> +	  Say Y here to enable the Mitsumi MM8013 fuel gauge driver.
> +	  It enables the monitoring of many battery parameters, including
> +	  the state of charge, temperature, cycle count, actual and design
> +	  capacity, etc.
> +
>  endif # POWER_SUPPLY
> diff --git a/drivers/power/supply/Makefile b/drivers/power/supply/Makefile
> index a8a9fa6de1e9..ba2c41f060be 100644
> --- a/drivers/power/supply/Makefile
> +++ b/drivers/power/supply/Makefile
> @@ -111,3 +111,4 @@ obj-$(CONFIG_BATTERY_SURFACE)	+= surface_battery.o
>  obj-$(CONFIG_CHARGER_SURFACE)	+= surface_charger.o
>  obj-$(CONFIG_BATTERY_UG3105)	+= ug3105_battery.o
>  obj-$(CONFIG_CHARGER_QCOM_SMB2)	+= qcom_pmi8998_charger.o
> +obj-$(CONFIG_FUEL_GAUGE_MM8013)	+= mm8013.o
> diff --git a/drivers/power/supply/mm8013.c b/drivers/power/supply/mm8013.c
> new file mode 100644
> index 000000000000..ce20c6078116
> --- /dev/null
> +++ b/drivers/power/supply/mm8013.c
> @@ -0,0 +1,283 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (c) 2016-2019 The Linux Foundation. All rights reserved.
> + * Copyright (c) 2023, Linaro Limited
> + */
> +#include <linux/delay.h>
> +#include <linux/i2c.h>
> +#include <linux/power_supply.h>
> +
> +#define REG_BATID			0x00 /* This one is very unclear */
> + #define BATID_101			0x0101 /* 107kOhm */
> + #define BATID_102			0x0102 /* 10kOhm */
> +#define REG_TEMPERATURE			0x06
> +#define REG_VOLTAGE			0x08
> +#define REG_FLAGS			0x0a
> + #define MM8013_FLAG_OTC		BIT(15)
> + #define MM8013_FLAG_OTD		BIT(14)
> + #define MM8013_FLAG_BATHI		BIT(13)
> + #define MM8013_FLAG_FC			BIT(9)
> + #define MM8013_FLAG_CHG		BIT(8)
> + #define MM8013_FLAG_DSG		BIT(0)
> +#define REG_FULL_CHARGE_CAPACITY	0x0e
> +#define REG_AVERAGE_CURRENT		0x14
> +#define REG_AVERAGE_TIME_TO_EMPTY	0x16
> +#define REG_AVERAGE_TIME_TO_FULL	0x18
> +#define REG_CYCLE_COUNT			0x2a
> +#define REG_STATE_OF_CHARGE		0x2c
> +#define REG_DESIGN_CAPACITY		0x3c
> +/* TODO: 0x62-0x68 seem to contain 'MM8013C' in a length-prefixed, non-terminated string */
> +
> +#define DECIKELVIN_TO_DECIDEGC(t)	(t - 2731)
> +
> +struct mm8013_chip {
> +	struct i2c_client *client;
> +};
> +
> +static int mm8013_write_reg(struct i2c_client *client, u8 reg, u16 value)
> +{
> +	int ret;
> +
> +	ret = i2c_smbus_write_word_data(client, reg, value);
> +	if (ret < 0)
> +		dev_err(&client->dev, "%s: err %d\n", __func__, ret);
> +
> +	usleep_range(4000, 5000);
> +	return ret;
> +}
> +
> +static int mm8013_read_reg(struct i2c_client *client, u8 reg)
> +{
> +	int ret;
> +
> +	ret = i2c_smbus_read_word_data(client, reg);
> +	if (ret < 0)
> +		dev_err(&client->dev, "%s: err %d\n", __func__, ret);
> +
> +	usleep_range(4000, 5000);
> +	return ret;
> +}

Please use regmap.

> +static int mm8013_checkdevice(struct mm8013_chip *chip)
> +{
> +	int battery_id, ret;
> +
> +	ret = mm8013_write_reg(chip->client, REG_BATID, 0x0008);
> +	if (ret < 0)
> +		return ret;
> +
> +	ret = mm8013_read_reg(chip->client, REG_BATID);
> +	if (ret < 0)
> +		return ret;
> +
> +	if (ret == BATID_102)
> +		battery_id = 2;
> +	else if (ret == BATID_101)
> +		battery_id = 1;
> +	else
> +		return -EINVAL;
> +
> +	dev_dbg(&chip->client->dev, "battery_id: %d\n", battery_id);
> +
> +	return 0;
> +}
> +
> +static enum power_supply_property mm8013_battery_props[] = {
> +	POWER_SUPPLY_PROP_CAPACITY,
> +	POWER_SUPPLY_PROP_CHARGE_FULL,
> +	POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN,
> +	POWER_SUPPLY_PROP_CURRENT_NOW,
> +	POWER_SUPPLY_PROP_CYCLE_COUNT,
> +	POWER_SUPPLY_PROP_HEALTH,
> +	POWER_SUPPLY_PROP_PRESENT,
> +	POWER_SUPPLY_PROP_STATUS,
> +	POWER_SUPPLY_PROP_TEMP,
> +	POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG,
> +	POWER_SUPPLY_PROP_TIME_TO_FULL_AVG,
> +	POWER_SUPPLY_PROP_VOLTAGE_NOW,
> +};
> +
> +static int mm8013_get_property(struct power_supply *psy,
> +			       enum power_supply_property psp,
> +			       union power_supply_propval *val)
> +{
> +	struct mm8013_chip *chip = psy->drv_data;
> +	struct i2c_client *client = chip->client;
> +	int ret = 0;
> +
> +	switch (psp) {
> +	case POWER_SUPPLY_PROP_CAPACITY:

this is in %, while the next two are in uAh. So the fuel gauge does
not provide the current capacity in uAh
(POWER_SUPPLY_PROP_CHARGE_NOW)?

> +		ret = mm8013_read_reg(client, REG_STATE_OF_CHARGE);
> +		if (ret < 0)
> +			return ret;
> +
> +		val->intval = ret;
> +		break;
> +	case POWER_SUPPLY_PROP_CHARGE_FULL:
> +		ret = mm8013_read_reg(client, REG_FULL_CHARGE_CAPACITY);
> +		if (ret < 0)
> +			return ret;
> +
> +		val->intval = ret;
> +		break;
> +	case POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN:
> +		ret = mm8013_read_reg(client, REG_DESIGN_CAPACITY);
> +		if (ret < 0)
> +			return ret;
> +
> +		val->intval = ret;
> +		break;
> +	case POWER_SUPPLY_PROP_CURRENT_NOW:
> +		ret = mm8013_read_reg(client, REG_AVERAGE_CURRENT);
> +		if (ret < 0)
> +			return ret;
> +
> +		if (ret > S16_MAX)
> +			val->intval -= (1 << 16);
> +		else
> +			val->intval = ret;

this is just 'val->intval = (s16) ret;'.

> +		val->intval *= -1000;

and this seems to be the only property, that properly scales its
values, assuming the hardware reports data in mA.

> +		break;
> +	case POWER_SUPPLY_PROP_CYCLE_COUNT:
> +		ret = mm8013_read_reg(client, REG_CYCLE_COUNT);
> +		if (ret < 0)
> +			return ret;
> +
> +		val->intval = ret;
> +		break;
> +	case POWER_SUPPLY_PROP_HEALTH:
> +		ret = mm8013_read_reg(client, REG_FLAGS);
> +		if (ret < 0)
> +			return ret;
> +
> +		if (ret & MM8013_FLAG_BATHI)
> +			val->intval = POWER_SUPPLY_HEALTH_OVERVOLTAGE;
> +		else if (ret & (MM8013_FLAG_OTD | MM8013_FLAG_OTC))
> +			val->intval = POWER_SUPPLY_HEALTH_OVERHEAT;
> +		else
> +			val->intval = POWER_SUPPLY_HEALTH_GOOD;
> +		break;
> +	case POWER_SUPPLY_PROP_PRESENT:
> +		val->intval = mm8013_read_reg(client, REG_TEMPERATURE) > 0;
> +		break;
> +	case POWER_SUPPLY_PROP_STATUS:
> +		ret = mm8013_read_reg(client, REG_FLAGS);
> +		if (ret < 0)
> +			return ret;
> +
> +		if (ret & MM8013_FLAG_DSG)
> +			val->intval = POWER_SUPPLY_STATUS_DISCHARGING;
> +		else if (ret & MM8013_FLAG_CHG)
> +			val->intval = POWER_SUPPLY_STATUS_CHARGING;
> +		else if (ret & MM8013_FLAG_FC)
> +			val->intval = POWER_SUPPLY_STATUS_FULL;
> +		else
> +			val->intval = POWER_SUPPLY_STATUS_UNKNOWN;
> +		break;
> +	case POWER_SUPPLY_PROP_TEMP:
> +		ret = mm8013_read_reg(client, REG_TEMPERATURE);
> +		if (ret < 0)
> +			return ret;
> +
> +		val->intval = DECIKELVIN_TO_DECIDEGC(ret);
> +		break;
> +	case POWER_SUPPLY_PROP_TIME_TO_EMPTY_AVG:
> +		ret = mm8013_read_reg(client, REG_AVERAGE_TIME_TO_EMPTY);
> +		if (ret < 0)
> +			return ret;
> +
> +		/* The estimation is not yet ready */
> +		if (ret == U16_MAX)
> +			return -ENODATA;
> +
> +		val->intval = ret;
> +		break;
> +	case POWER_SUPPLY_PROP_TIME_TO_FULL_AVG:
> +		ret = mm8013_read_reg(client, REG_AVERAGE_TIME_TO_FULL);
> +		if (ret < 0)
> +			return ret;
> +
> +		/* The estimation is not yet ready */
> +		if (ret == U16_MAX)
> +			return -ENODATA;
> +
> +		val->intval = ret;
> +		break;
> +	case POWER_SUPPLY_PROP_VOLTAGE_NOW:

that's in **micro** volts

> +		ret = mm8013_read_reg(client, REG_VOLTAGE);

this effectively does i2c_smbus_read_word_data() and thus the
maximum is is 65536. 65536uV = 65mV. I have serious doubts, that
this code does what you want. The same is true for a couple of
the other properties.

> +		if (ret < 0)
> +			return ret;
> +
> +		val->intval = ret;
> +		break;
> +	default:
> +		return -EINVAL;
> +	}
> +
> +	return 0;
> +}
> +
> +static const struct power_supply_desc mm8013_desc = {
> +	.name			= "mm8013",
> +	.type			= POWER_SUPPLY_TYPE_BATTERY,
> +	.properties		= mm8013_battery_props,
> +	.num_properties		= ARRAY_SIZE(mm8013_battery_props),
> +	.get_property		= mm8013_get_property,
> +};
> +
> +static int mm8013_probe(struct i2c_client *client)
> +{
> +	struct power_supply_config psy_cfg = {};
> +	struct device *dev = &client->dev;
> +	struct power_supply *psy;
> +	struct mm8013_chip *chip;
> +	int ret = 0;
> +
> +	if (!i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_WORD_DATA))
> +		return dev_err_probe(dev, -EIO,
> +				     "I2C_FUNC_SMBUS_WORD_DATA not supported\n");
> +
> +	chip = devm_kzalloc(dev, sizeof(struct mm8013_chip), GFP_KERNEL);
> +	if (!chip)
> +		return -ENOMEM;
> +
> +	chip->client = client;
> +
> +	ret = mm8013_checkdevice(chip);
> +	if (ret)
> +		return dev_err_probe(dev, -ENODEV, "MM8013 not found\n");

dev_err_probe(dev, ret, ... ?

> +
> +	psy_cfg.drv_data = chip;
> +	psy_cfg.of_node = dev->of_node;
> +
> +	psy = devm_power_supply_register(dev, &mm8013_desc, &psy_cfg);
> +	if (IS_ERR(psy))
> +		return PTR_ERR(psy);
> +
> +	return 0;
> +}
> +
> +static const struct i2c_device_id mm8013_id_table[] = {
> +	{ "mm8013", 0 },
> +	{},

please remove the last ,

> +};
> +MODULE_DEVICE_TABLE(i2c, mm8013_id_table);
> +
> +static const struct of_device_id mm8013_match_table[] = {
> +	{ .compatible = "mitsumi,mm8013" },
> +	{ },

extra space in the Terminator, also terminator should not have a
comma at the end.

> +};
> +
> +static struct i2c_driver mm8013_i2c_driver = {
> +	.probe = mm8013_probe,
> +	.id_table = mm8013_id_table,
> +	.driver = {
> +		.name = "mm8013",
> +		.of_match_table = mm8013_match_table,
> +	},
> +};
> +module_i2c_driver(mm8013_i2c_driver);
> +
> +MODULE_DESCRIPTION("MM8013 fuel gauge driver");
> +MODULE_LICENSE("GPL");

With the next submission please include a dump of the uevent
in sysfs in the cover letter or below the fold, so that its
easy to validty check if the reported values look sensible.

Thanks and sorry for the slow processing,

-- Sebastian

Download attachment "signature.asc" of type "application/pgp-signature" (834 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ