Re: [PATCH V5 4/6] regulator: Add a regulator driver for the PM8008 PMIC

From: Stephen Boyd
Date: Wed Feb 09 2022 - 21:01:11 EST

Quoting Satya Priya (2022-02-08 06:52:18)
> diff --git a/drivers/regulator/qcom-pm8008-regulator.c b/drivers/regulator/qcom-pm8008-regulator.c
> new file mode 100644
> index 0000000..86043b4
> --- /dev/null
> +++ b/drivers/regulator/qcom-pm8008-regulator.c
> @@ -0,0 +1,234 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/* Copyright (c) 2021, The Linux Foundation. All rights reserved. */
> +
> +#include <linux/device.h>
> +#include <linux/interrupt.h>
> +#include <linux/kernel.h>
> +#include <linux/module.h>
> +#include <linux/of.h>
> +#include <linux/of_device.h>
> +#include <linux/platform_device.h>
> +#include <linux/regmap.h>
> +#include <linux/regulator/driver.h>
> +#include <linux/regulator/machine.h>
> +
> +#define STARTUP_DELAY_USEC 20
> +#define VSET_STEP_MV 8
> +#define VSET_STEP_UV (VSET_STEP_MV * 1000)
> +
> +#define LDO_ENABLE_REG(base) ((base) + 0x46)
> +#define ENABLE_BIT BIT(7)
> +
> +#define LDO_STATUS1_REG(base) ((base) + 0x08)
> +#define VREG_READY_BIT BIT(7)
> +
> +#define LDO_VSET_LB_REG(base) ((base) + 0x40)
> +
> +#define LDO_STEPPER_CTL_REG(base) ((base) + 0x3b)
> +#define DEFAULT_VOLTAGE_STEPPER_RATE 38400
> +#define STEP_RATE_MASK GENMASK(1, 0)
> +
> +#define PM8008_NUM_LDOS 7
> +
> +struct regulator_data {
> + const char *name;
> + const char *supply_name;
> + u16 base;
> + int min_uv;
> + int max_uv;
> + int min_dropout_uv;
> + const struct linear_range *voltage_range;
> +};
> +
> +struct pm8008_regulator {
> + struct device *dev;
> + struct regmap *regmap;
> + struct regulator_desc rdesc;
> + u16 base;
> + int step_rate;
> +};
> +
> +static const struct linear_range nldo_ranges[] = {
> + REGULATOR_LINEAR_RANGE(528000, 0, 122, 8000),
> +};
> +
> +static const struct linear_range pldo_ranges[] = {
> + REGULATOR_LINEAR_RANGE(1504000, 0, 237, 8000),
> +};
> +
> +static const struct regulator_data reg_data[] = {
> + /* name parent base min_uv max_uv headroom_uv voltage_range */
> + { "LDO1", "vdd_l1_l2", 0x4000, 528000, 1504000, 225000, nldo_ranges, },
> + { "LDO2", "vdd_l1_l2", 0x4100, 528000, 1504000, 225000, nldo_ranges, },
> + { "LDO3", "vdd_l3_l4", 0x4200, 1504000, 3400000, 300000, pldo_ranges, },
> + { "LDO4", "vdd_l3_l4", 0x4300, 1504000, 3400000, 300000, pldo_ranges, },
> + { "LDO5", "vdd_l5", 0x4400, 1504000, 3400000, 200000, pldo_ranges, },
> + { "LDO6", "vdd_l6", 0x4500, 1504000, 3400000, 200000, pldo_ranges, },
> + { "LDO7", "vdd_l7", 0x4600, 1504000, 3400000, 200000, pldo_ranges, },
> +};
> +
> +static int pm8008_regulator_get_voltage(struct regulator_dev *rdev)
> +{
> + struct pm8008_regulator *pm8008_reg = rdev_get_drvdata(rdev);
> + __le16 mV;
> + int rc;
> +
> + rc = regmap_bulk_read(pm8008_reg->regmap,
> + LDO_VSET_LB_REG(pm8008_reg->base), (void *)&mV, 2);
> + if (rc < 0) {
> + dev_err(&rdev->dev, "failed to read regulator voltage rc=%d\n", rc);
> + return rc;
> + }
> +
> + return le16_to_cpu(mV) * 1000;
> +}
> +
> +static inline int pm8008_write_voltage(struct pm8008_regulator *pm8008_reg,
> + int mV)
> +{
> + int rc;
> + u16 vset_raw;
> +
> + vset_raw = cpu_to_le16(mV);

sparse should complain here that an le16 is degrading to a u16. Please
make vset_raw an __le16 as well.

> +
> + rc = regmap_bulk_write(pm8008_reg->regmap,
> + LDO_VSET_LB_REG(pm8008_reg->base),
> + (const void *)&vset_raw, sizeof(vset_raw));
> + if (rc < 0) {
> + dev_err(pm8008_reg->dev, "failed to write voltage rc=%d\n", rc);

Do we really need this? It could spam the logs in theory. We have
tracepoints on regmap that could be used to figure out that some
read/write failed. I'd like to see a plain

return regmap_bulk_write(...)

> + return rc;
> + }
> +
> + return 0;
> +}
> +
> +static int pm8008_regulator_set_voltage_time(struct regulator_dev *rdev,
> + int old_uV, int new_uv)
> +{
> + struct pm8008_regulator *pm8008_reg = rdev_get_drvdata(rdev);
> +
> + return DIV_ROUND_UP(abs(new_uv - old_uV), pm8008_reg->step_rate);
> +}
> +
> +static int pm8008_regulator_set_voltage(struct regulator_dev *rdev,
> + unsigned int selector)
> +{
> + struct pm8008_regulator *pm8008_reg = rdev_get_drvdata(rdev);
> + int rc, mV;
> +
> + /* voltage control register is set with voltage in millivolts */
> + mV = DIV_ROUND_UP(regulator_list_voltage_linear_range(rdev, selector),
> + 1000);

Aren't there linear range APIs that can avoid any div roundups here?

> + if (mV < 0)
> + return mV;
> +
> + rc = pm8008_write_voltage(pm8008_reg, mV);
> + if (rc < 0)
> + return rc;
> +
> + dev_dbg(&rdev->dev, "voltage set to %d\n", mV * 1000);
> + return 0;
> +}
> +
> +static const struct regulator_ops pm8008_regulator_ops = {
> + .enable = regulator_enable_regmap,
> + .disable = regulator_disable_regmap,
> + .is_enabled = regulator_is_enabled_regmap,
> + .set_voltage_sel = pm8008_regulator_set_voltage,
> + .get_voltage = pm8008_regulator_get_voltage,
> + .list_voltage = regulator_list_voltage_linear,
> + .set_voltage_time = pm8008_regulator_set_voltage_time,
> +};
> +
> +static int pm8008_regulator_of_parse(struct device_node *node,
> + const struct regulator_desc *desc,
> + struct regulator_config *config)
> +{
> + struct pm8008_regulator *pm8008_reg = config->driver_data;
> + int rc;
> + unsigned int reg;
> +
> + /* get slew rate */
> + rc = regmap_bulk_read(pm8008_reg->regmap,
> + LDO_STEPPER_CTL_REG(pm8008_reg->base), &reg, 1);
> + if (rc < 0) {
> + dev_err(pm8008_reg->dev,
> + "%s: failed to read step rate configuration rc=%d\n",
> + pm8008_reg->rdesc.name, rc);
> + return rc;
> + }
> + reg &= STEP_RATE_MASK;
> + pm8008_reg->step_rate = DEFAULT_VOLTAGE_STEPPER_RATE >> reg;
> +
> + return 0;
> +}
> +
> +static int pm8008_regulator_probe(struct platform_device *pdev)
> +{
> + struct device *dev = &pdev->dev;
> + int id = pdev->id % PM8008_NUM_LDOS;
> + struct regulator_dev *rdev;
> + struct pm8008_regulator *pm8008_reg;
> + struct regmap *regmap;
> + struct regulator_config reg_config = {};
> + int rc;
> +
> + dev_dbg(dev, "DEBUG: Probing LDO%d\n", id + 1);

Why can't we probe one regulators (plural) platform device instead of 8
regulator platform devices? A 'struct device' isn't exactly small and
it would be simpler to probe all the regulators in a single loop.

> +
> + regmap = dev_get_regmap(dev->parent, NULL);
> + if (!regmap) {
> + dev_err(dev, "parent regmap is missing\n");