by Lee Jones

[permalink] [raw]

Subject: Re: [PATCH v5 1/6] mfd: rt4831: Adds support for Richtek RT4831 core

On Thu, 25 Mar 2021, ChiYuan Huang wrote:

> HI, Lee:
>
> ChiYuan Huang <[email protected]> 於 2021年1月13日週三下午10:09寫道：
> >
> > Lee Jones <[email protected]> 於 2021年1月13日週三下午8:21寫道：
> > >
> > > On Thu, 17 Dec 2020, cy_huang wrote:
> > >
> > > > From: ChiYuan Huang <[email protected]>
> > > >
> > > > This adds support Richtek RT4831 core. It includes four channel WLED driver
> > > > and Display Bias Voltage outputs.
> > > >
> > > > Signed-off-by: ChiYuan Huang <[email protected]>
> > > > ---
> > > > since v5
> > > > - Rename file name from rt4831-core.c to rt4831.c
> > > > - Change RICHTEK_VID to RICHTEK_VENDOR_ID.
> > > > - Change gpio_desc nameing from 'enable' to 'enable_gpio' in probe.
> > > > - Change variable 'val' to the meaningful name 'chip_id'.
> > > > - Refine the error log when vendor id is not matched.
> > > > - Remove of_match_ptr.
> > > >
> > > > since v2
> > > > - Refine Kconfig descriptions.
> > > > - Add copyright.
> > > > - Refine error logs in probe.
> > > > - Refine comment lines in remove and shutdown.
> > > > ---
> > > > drivers/mfd/Kconfig | 10 +++++
> > > > drivers/mfd/Makefile | 1 +
> > > > drivers/mfd/rt4831.c | 124 +++++++++++++++++++++++++++++++++++++++++++++++++++
> > > > 3 files changed, 135 insertions(+)
> > > > create mode 100644 drivers/mfd/rt4831.c
> > > >
> > > > diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
> > > > index 8b99a13..dfb2640 100644
> > > > --- a/drivers/mfd/Kconfig
> > > > +++ b/drivers/mfd/Kconfig
> > > > @@ -1088,6 +1088,16 @@ config MFD_RDC321X
> > > > southbridge which provides access to GPIOs and Watchdog using the
> > > > southbridge PCI device configuration space.
> > > >
> > > > +config MFD_RT4831
> > > > + tristate "Richtek RT4831 four channel WLED and Display Bias Voltage"
> > > > + depends on I2C
> > > > + select MFD_CORE
> > > > + select REGMAP_I2C
> > > > + help
> > > > + This enables support for the Richtek RT4831 that includes 4 channel
> > > > + WLED driving and Display Bias Voltage. It's commonly used to provide
> > > > + power to the LCD display and LCD backlight.
> > > > +
> > > > config MFD_RT5033
> > > > tristate "Richtek RT5033 Power Management IC"
> > > > depends on I2C
> > > > diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
> > > > index 1780019..28d247b 100644
> > > > --- a/drivers/mfd/Makefile
> > > > +++ b/drivers/mfd/Makefile
> > > > @@ -235,6 +235,7 @@ obj-$(CONFIG_MFD_MENF21BMC) += menf21bmc.o
> > > > obj-$(CONFIG_MFD_HI6421_PMIC) += hi6421-pmic-core.o
> > > > obj-$(CONFIG_MFD_HI655X_PMIC) += hi655x-pmic.o
> > > > obj-$(CONFIG_MFD_DLN2) += dln2.o
> > > > +obj-$(CONFIG_MFD_RT4831) += rt4831.o
> > > > obj-$(CONFIG_MFD_RT5033) += rt5033.o
> > > > obj-$(CONFIG_MFD_SKY81452) += sky81452.o
> > > >
> > > > diff --git a/drivers/mfd/rt4831.c b/drivers/mfd/rt4831.c
> > > > new file mode 100644
> > > > index 00000000..2bf8364
> > > > --- /dev/null
> > > > +++ b/drivers/mfd/rt4831.c
> > > > @@ -0,0 +1,124 @@
> > > > +// SPDX-License-Identifier: GPL-2.0+
> > > > +/*
> > > > + * Copyright (c) 2020 Richtek Technology Corp.
> > >
> > > Nit: If you respin this, please bump the date.
> > >
> > Okay.
> > > > + * Author: ChiYuan Huang <[email protected]>
> > > > + */
> > > > +
> > > > +#include <linux/gpio/consumer.h>
> > > > +#include <linux/i2c.h>
> > > > +#include <linux/kernel.h>
> > > > +#include <linux/mfd/core.h>
> > > > +#include <linux/module.h>
> > > > +#include <linux/regmap.h>
> > > > +
> > > > +#define RT4831_REG_REVISION 0x01
> > > > +#define RT4831_REG_ENABLE 0x08
> > > > +#define RT4831_REG_I2CPROT 0x15
> > > > +
> > > > +#define RICHTEK_VENDOR_ID 0x03
> > > > +#define RT4831_VID_MASK GENMASK(1, 0)
> > > > +#define RT4831_RESET_MASK BIT(7)
> > > > +#define RT4831_I2CSAFETMR_MASK BIT(0)
> > > > +
> > > > +static const struct mfd_cell rt4831_subdevs[] = {
> > > > + OF_MFD_CELL("rt4831-backlight", NULL, NULL, 0, 0, "richtek,rt4831-backlight"),
> > > > + MFD_CELL_NAME("rt4831-regulator")
> > > > +};
> > > > +
> > > > +static bool rt4831_is_accessible_reg(struct device *dev, unsigned int reg)
> > > > +{
> > > > + if (reg >= RT4831_REG_REVISION && reg <= RT4831_REG_I2CPROT)
> > > > + return true;
> > > > + return false;
> > > > +}
> > > > +
> > > > +static const struct regmap_config rt4831_regmap_config = {
> > > > + .reg_bits = 8,
> > > > + .val_bits = 8,
> > > > + .max_register = RT4831_REG_I2CPROT,
> > > > +
> > > > + .readable_reg = rt4831_is_accessible_reg,
> > > > + .writeable_reg = rt4831_is_accessible_reg,
> > > > +};
> > > > +
> > > > +static int rt4831_probe(struct i2c_client *client)
> > > > +{
> > > > + struct gpio_desc *enable_gpio;
> > > > + struct regmap *regmap;
> > > > + unsigned int chip_id;
> > > > + int ret;
> > > > +
> > > > + enable_gpio = devm_gpiod_get_optional(&client->dev, "enable", GPIOD_OUT_HIGH);
> > > > + if (IS_ERR(enable_gpio)) {
> > > > + dev_err(&client->dev, "Failed to get 'enable' GPIO\n");
> > > > + return PTR_ERR(enable_gpio);
> > > > + }
> > > > +
> > > > + regmap = devm_regmap_init_i2c(client, &rt4831_regmap_config);
> > > > + if (IS_ERR(regmap)) {
> > > > + dev_err(&client->dev, "Failed to initialize regmap\n");
> > > > + return PTR_ERR(regmap);
> > > > + }
> > > > +
> > > > + ret = regmap_read(regmap, RT4831_REG_REVISION, &chip_id);
> > > > + if (ret) {
> > > > + dev_err(&client->dev, "Failed to get H/W revision\n");
> > > > + return ret;
> > > > + }
> > > > +
> > > > + if ((chip_id & RT4831_VID_MASK) != RICHTEK_VENDOR_ID) {
> > > > + dev_err(&client->dev, "Chip vendor ID 0x%02x not matched\n", chip_id);
> > > > + return -ENODEV;
> > > > + }
> > > > +
> > > > + /*
> > > > + * Used to prevent the abnormal shutdown.
> > > > + * If SCL/SDA both keep low for one second to reset HW.
> > > > + */
> > > > + ret = regmap_update_bits(regmap, RT4831_REG_I2CPROT, RT4831_I2CSAFETMR_MASK,
> > > > + RT4831_I2CSAFETMR_MASK);
> > > > + if (ret) {
> > > > + dev_err(&client->dev, "Failed to enable I2C safety timer\n");
> > > > + return ret;
> > > > + }
> > > > +
> > > > + return devm_mfd_add_devices(&client->dev, PLATFORM_DEVID_AUTO, rt4831_subdevs,
> > > > + ARRAY_SIZE(rt4831_subdevs), NULL, 0, NULL);
> > > > +}
> > > > +
> > > > +static int rt4831_remove(struct i2c_client *client)
> > > > +{
> > > > + struct regmap *regmap = dev_get_regmap(&client->dev, NULL);
> > > > +
> > > > + /* Disable WLED and DSV outputs */
> > > > + return regmap_update_bits(regmap, RT4831_REG_ENABLE, RT4831_RESET_MASK, RT4831_RESET_MASK);
> > > > +}
> > > > +
> > > > +static void rt4831_shutdown(struct i2c_client *client)
> > > > +{
> > > > + struct regmap *regmap = dev_get_regmap(&client->dev, NULL);
> > > > +
> > > > + /* Disable WLED and DSV outputs */
> > > > + regmap_update_bits(regmap, RT4831_REG_ENABLE, RT4831_RESET_MASK, RT4831_RESET_MASK);
> > > > +}
> > >
> > > What is your reason for providing a .shutdown() routine?
> > >
> > Just like as remove routine to make sure all output are off for the safety.
> > This chip only have one 'enable' pin and I2C as the control signal.
> > As normal shutdown, it can be make sure 'enable' will be pull low.
> > But for some case, if 'enable' always tied to high, like as ARM reset,
> > chip reset only triggered during next booting phase.
> > The period from arm reset to next probe, if user doesn't call DSV and
> > WLED off before machine shutdown/reboot, the WLED/DSV voltage boost
> > circuit will be kept as on.
> > That's why I also put shutdown routine in the driver to reset the whole chip.
>
> If the shutdown routine is not suitable, I think it can be removed.
> There's the HWEN pin inside this IC to make sure all function will be disabled.
>
> There're changed part in my note
> 1. respin the header date.
> 2. Remove the shutdown routine
>
> Anything else?

Just respin and resend.

I will do a full review once it's back on the list.

--
Lee Jones [李琼斯]
Senior Technical Lead - Developer Services
Linaro.org │ Open source software for Arm SoCs
Follow Linaro: Facebook | Twitter | Blog