On Thu, 14 Aug 2025, Nuno Sá via B4 Relay wrote:
> From: Nuno Sá <nuno.sa@xxxxxxxxxx>
>
> The LTC4283 is a negative voltage hot swap controller that drives an
> external N-channel MOSFET to allow a board to be safely inserted and
> removed from a live backplane.
>
> Main usage is as an Hardware Monitoring device. However, it has up to 8
> pins that can be configured and used as GPIOs and hence, the device can
> also be a GPIO controller (and so being added as MFD device).
>
> Signed-off-by: Nuno Sá <nuno.sa@xxxxxxxxxx>
> ---
> MAINTAINERS | 2 +
> drivers/mfd/Kconfig | 11 ++++
> drivers/mfd/Makefile | 1 +
> drivers/mfd/ltc4283.c | 140 ++++++++++++++++++++++++++++++++++++++++++++
> include/linux/mfd/ltc4283.h | 33 +++++++++++
> 5 files changed, 187 insertions(+)
>
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 413bb77d5eebe2b51aa9c3af86e7cfd5ab142044..b5f4f1c41c64b738d57c1fb5552a60b4c6b9985c 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -14506,6 +14506,8 @@ L: linux-hwmon@xxxxxxxxxxxxxxx
> L: linux-gpio@xxxxxxxxxxxxxxx
> S: Supported
> F: Documentation/devicetree/bindings/mfd/adi,ltc4283.yaml
> +F: drivers/mfd/ltc4283.c
> +F: include/linux/mfd/ltc4283.h
>
> LTC4286 HARDWARE MONITOR DRIVER
> M: Delphine CC Chiu <Delphine_CC_Chiu@xxxxxxxxxx>
> diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
> index 425c5fba6cb1e7848dcea05bd77c729a71d48e2c..ec3e02d40fd17a0bba29e3157723055feedebd11 100644
> --- a/drivers/mfd/Kconfig
> +++ b/drivers/mfd/Kconfig
> @@ -900,6 +900,17 @@ config MFD_MAX14577
> additional drivers must be enabled in order to use the functionality
> of the device.
>
> +config MFD_LTC4283
> + tristate "LTC4283 Hot Swap Controller"
> + depends on I2C
> + select MFD_CORE
> + select REGMAP_I2C
> + help
> + This enables support for the LTC4283 Negative Voltage Hot Swap
> + Controller. This driver provides common support for accessing the
> + device; additional drivers must be enabled in order to use the
> + functionality of the device.
> +
> config MFD_MAX77541
> tristate "Analog Devices MAX77541/77540 PMIC Support"
> depends on I2C=y
> diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
> index f7bdedd5a66d16bf8ccee0da1236a441e6f085b0..4a3bcd30ab7e12aed4bc5e48294500c77cb61aa7 100644
> --- a/drivers/mfd/Makefile
> +++ b/drivers/mfd/Makefile
> @@ -23,6 +23,7 @@ obj-$(CONFIG_MFD_EXYNOS_LPASS) += exynos-lpass.o
> obj-$(CONFIG_MFD_GATEWORKS_GSC) += gateworks-gsc.o
> obj-$(CONFIG_MFD_MACSMC) += macsmc.o
>
> +obj-$(CONFIG_MFD_LTC4283) += ltc4283.o
> obj-$(CONFIG_MFD_TI_LP873X) += lp873x.o
> obj-$(CONFIG_MFD_TI_LP87565) += lp87565.o
> obj-$(CONFIG_MFD_TI_AM335X_TSCADC) += ti_am335x_tscadc.o
> diff --git a/drivers/mfd/ltc4283.c b/drivers/mfd/ltc4283.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..a45c01278f884630984790c922c057c4f2db9c82
> --- /dev/null
> +++ b/drivers/mfd/ltc4283.c
> @@ -0,0 +1,140 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/*
> + * Analog Devices LTC4283 I2C Negative Voltage Hot Swap Controller
> + *
> + * Copyright 2025 Analog Devices Inc.
> + */
> +#include <linux/bitmap.h>
> +#include <linux/bitops.h>
> +#include <linux/device.h>
> +#include <linux/i2c.h>
> +#include <linux/mfd/core.h>
> +#include <linux/module.h>
> +#include <linux/mod_devicetable.h>
> +#include <linux/regmap.h>
> +#include <linux/property.h>
Alphabetical.
> +
> +#include <linux/mfd/ltc4283.h>
> +
> +static const struct mfd_cell ltc4283_cells[] = {
> + MFD_CELL_OF("ltc4283-hwmon", NULL, NULL, 0, 0, "adi,ltc4283-hwmon"),
> + MFD_CELL_OF("ltc4283-gpio", NULL, NULL, 0, 0, "adi,ltc4283-gpio"),
> +};
> +
> +static bool ltc4283_writable_reg(struct device *dev, unsigned int reg)
> +{
> + switch (reg) {
> + case 0x00 ... 0x03:
Define these magic numbers.
> + return false;
> + case 0x3c:
> + return false;
> + case 0x86 ... 0x8f:
> + return false;
> + case 0x91 ... 0xa1:
> + return false;
> + case 0xa3:
> + return false;
> + case 0xac:
> + return false;
> + case 0xf1 ... 0xff:
> + return false;
> + default:
> + return true;
> + }
> +}
> +
> +static const struct regmap_config ltc4283_regmap_config = {
> + .reg_bits = 8,
> + .val_bits = 8,
> + .max_register = 0xff,
> + .writeable_reg = ltc4283_writable_reg,
> +};
> +
> +static int ltc4283_get_gpio_pins(struct i2c_client *client, u32 *n_cells)
Why not do this in the GPIO driver?
> +{
> + struct device *dev = &client->dev;
> + u32 pins[LTC4283_GPIO_MAX], pin;
> + unsigned long *gpio_mask;
> + int n_pins, ret;
> +
> + /*
> + * The device has up to 8 pins that can be configured either as GPIOS or
> + * for monitoring purposes. Both gpio and hwmon devices need to have
> + * this information in order to do proper validations and
> + * configurations. Hence, this property needs to be in the top level
> + * device.
> + */
> + n_pins = device_property_count_u32(dev, "adi,gpio-pins");
> + if (n_pins < 0)
> + return 0;
> + if (n_pins >= LTC4283_GPIO_MAX)
> + return dev_err_probe(dev, -EINVAL, "Too many GPIO pins specified (%d), max is %d\n",
> + n_pins, LTC4283_GPIO_MAX);
> +
> + ret = device_property_read_u32_array(dev, "adi,gpio-pins", pins, n_pins);
> + if (ret)
> + return dev_err_probe(dev, ret, "Failed to read GPIO pins\n");
> +
> + gpio_mask = devm_bitmap_zalloc(dev, LTC4283_GPIO_MAX, GFP_KERNEL);
> + if (!gpio_mask)
> + return -ENOMEM;
> +
> + for (pin = 0; pin < n_pins; pin++) {
> + if (pins[pin] >= LTC4283_GPIO_MAX)
> + return dev_err_probe(dev, -EINVAL,
> + "Invalid GPIO pin specified (%u), max is %d\n",
> + pins[pin], LTC4283_GPIO_MAX);
> +
> + __set_bit(pins[pin], gpio_mask);
> + }
> +
> + /* Add the GPIO cell */
> + *n_cells += 1;
Just register it anyway and have the GPIO driver error out if it doesn't
have the right cells / properties.
> + i2c_set_clientdata(client, gpio_mask);
> +
> + return 0;
> +}
> +
> +static int ltc4283_probe(struct i2c_client *client)
> +{
> + u32 n_cells = ARRAY_SIZE(ltc4283_cells) - 1;
> + struct regmap *regmap;
> + int ret;
> +
> + regmap = devm_regmap_init_i2c(client, <c4283_regmap_config);
> + if (IS_ERR(regmap))
> + return PTR_ERR(regmap);
> +
> + ret = ltc4283_get_gpio_pins(client, &n_cells);
> + if (ret)
> + return ret;
> +
> + return devm_mfd_add_devices(&client->dev, PLATFORM_DEVID_AUTO,
> + ltc4283_cells, n_cells, NULL, 0, NULL);
> +}
> +
> +static const struct of_device_id ltc4283_of_match[] = {
> + { .compatible = "adi,ltc4283" },
> + { }
> +};
> +MODULE_DEVICE_TABLE(of, ltc4283_of_match);
> +
> +static const struct i2c_device_id ltc4283_i2c_id[] = {
> + { "ltc4283" },
> + { }
> +};
> +MODULE_DEVICE_TABLE(i2c, ltc4283_i2c_id);
> +
> +static struct i2c_driver ltc4283_driver = {
> + .driver = {
> + .name = "ltc4283",
> + .of_match_table = ltc4283_of_match,
> + },
> + .probe = ltc4283_probe,
> + .id_table = ltc4283_i2c_id,
> +};
> +module_i2c_driver(ltc4283_driver);
> +
> +MODULE_AUTHOR("Nuno Sá <nuno.sa@xxxxxxxxxx>");
> +MODULE_DESCRIPTION("LTC4283 MFD I2C driver");
It's not an MFD anything!
And the communication method is meaningless.
You can use "Core driver" instead.
> +MODULE_LICENSE("GPL");
> diff --git a/include/linux/mfd/ltc4283.h b/include/linux/mfd/ltc4283.h
> new file mode 100644
> index 0000000000000000000000000000000000000000..153594009c07b296ce5743e5e817e96464d81cb3
> --- /dev/null
> +++ b/include/linux/mfd/ltc4283.h
> @@ -0,0 +1,33 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +/*
> + * Analog Devices LTC4283 I2C Negative Voltage Hot Swap Controller
> + *
> + * Copyright 2025 Analog Devices Inc.
> + */
> +
> +#ifndef __MFD_LTC4283_H_
> +#define __MFD_LTC4283_H_
> +
> +#include <linux/bitops.h>
> +#include <linux/bits.h>
> +
> +/*
> + * We can have up to 8 gpios. 4 PGIOs and 4 ADIOs. PGIOs start at index 4 in the
> + * gpios mask.
> + */
> +#define LTC4283_PGIOX_START_NR 4
> +
> +#define LTC4283_PGIO_CONFIG 0x10
> +#define LTC4283_PGIO_CFG_MASK(pin) \
> + GENMASK(((pin) - LTC4283_PGIOX_START_NR) * 2 + 1, (((pin) - LTC4283_PGIOX_START_NR) * 2))
What do all of these numbers mean? Can you define them?
> +#define LTC4283_PGIO_CONFIG_2 0x11
> +#define LTC4283_ADC_MASK GENMASK(2, 0)
> +#define LTC4283_PGIO_OUT_MASK(pin) BIT(4 + (pin))
> +
> +#define LTC4283_GPIO_MAX 8
> +
> +/* Non-constant mask variant of FIELD_GET() and FIELD_PREP() */
Why?
> +#define field_get(_mask, _reg) (((_reg) & (_mask)) >> (ffs(_mask) - 1))
> +#define field_prep(_mask, _val) (((_val) << (ffs(_mask) - 1)) & (_mask))
> +
> +#endif
>
> --
> 2.50.1
>
>
--
Lee Jones [李琼斯]
