Hi Jaehoon, Thanks for the review.
On Mon, 8 Mar 2021 at 07:03, Jaehoon Chung <jh80.ch...@samsung.com> wrote: > Dear Ying-Chun > > On 3/8/21 3:18 AM, Ying-Chun Liu wrote: > > From: "Ying-Chun Liu (PaulLiu)" <paul....@linaro.org> > > > > Anatop is an integrated regulator inside i.MX6 SoC. > > There are 3 digital regulators which controls PU, CORE (ARM), and SOC. > > And 3 analog regulators which controls 1P1, 2P5, 3P0 (USB). > > This patch adds the Anatop regulator driver. > > > > Signed-off-by: Ying-Chun Liu (PaulLiu) <paul....@linaro.org> > > --- > > drivers/power/regulator/Kconfig | 10 + > > drivers/power/regulator/Makefile | 1 + > > drivers/power/regulator/anatop_regulator.c | 289 +++++++++++++++++++++ > > 3 files changed, 300 insertions(+) > > create mode 100644 drivers/power/regulator/anatop_regulator.c > > > > diff --git a/drivers/power/regulator/Kconfig > b/drivers/power/regulator/Kconfig > > index fbbea18c7d..1cd1f3f5ed 100644 > > --- a/drivers/power/regulator/Kconfig > > +++ b/drivers/power/regulator/Kconfig > > @@ -312,6 +312,16 @@ config DM_REGULATOR_STPMIC1 > > by the PMIC device. This driver is controlled by a device tree node > > which includes voltage limits. > > > > +config DM_REGULATOR_ANATOP > > + bool "Enable driver for ANATOP regulators" > > + depends on DM_REGULATOR > > + select REGMAP > > + select SYSCON > > + help > > + Enable support for the Freescale i.MX on-chip ANATOP LDOs > > + regulators. It is recommended that this option be enabled on > > + i.MX6 platform. > > + > > config SPL_DM_REGULATOR_STPMIC1 > > bool "Enable driver for STPMIC1 regulators in SPL" > > depends on SPL_DM_REGULATOR && PMIC_STPMIC1 > > diff --git a/drivers/power/regulator/Makefile > b/drivers/power/regulator/Makefile > > index 9d58112dcb..e7198da911 100644 > > --- a/drivers/power/regulator/Makefile > > +++ b/drivers/power/regulator/Makefile > > @@ -30,3 +30,4 @@ obj-$(CONFIG_DM_REGULATOR_TPS65910) += > tps65910_regulator.o > > obj-$(CONFIG_DM_REGULATOR_TPS62360) += tps62360_regulator.o > > obj-$(CONFIG_$(SPL_)DM_REGULATOR_STPMIC1) += stpmic1.o > > obj-$(CONFIG_DM_REGULATOR_TPS65941) += tps65941_regulator.o > > +obj-$(CONFIG_$(SPL_)DM_REGULATOR_ANATOP) += anatop_regulator.o > > diff --git a/drivers/power/regulator/anatop_regulator.c > b/drivers/power/regulator/anatop_regulator.c > > new file mode 100644 > > index 0000000000..2bb5cdbac5 > > --- /dev/null > > +++ b/drivers/power/regulator/anatop_regulator.c > > @@ -0,0 +1,289 @@ > > +// SPDX-License-Identifier: GPL-2.0+ > > +// > > +// Copyright (C) 2011 Freescale Semiconductor, Inc. All Rights Reserved. > > +// Copyright (C) 2021 Linaro > > + > > +#include <common.h> > > +#include <errno.h> > > +#include <dm.h> > > +#include <log.h> > > +#include <linux/delay.h> > > +#include <linux/err.h> > > +#include <power/pmic.h> > > +#include <power/regulator.h> > > +#include <regmap.h> > > +#include <syscon.h> > > +#include <linux/bitops.h> > > +#include <linux/ioport.h> > > +#include <dm/device-internal.h> > > +#include <dm/device_compat.h> > > +#include <dm/read.h> > > Well, i think that it can be removed more than now. > > Will fix it in v2. > > + > > +#define LDO_RAMP_UP_UNIT_IN_CYCLES 64 /* 64 cycles per step */ > > +#define LDO_RAMP_UP_FREQ_IN_MHZ 24 /* cycle based on 24M OSC */ > > + > > +#define LDO_POWER_GATE 0x00 > > +#define LDO_FET_FULL_ON 0x1f > > + > > +struct anatop_regulator { > > + const char *name; > > + u32 control_reg; > > + u32 vol_bit_shift; > > + u32 vol_bit_width; > > + u32 min_bit_val; > > + u32 min_voltage; > > + u32 max_voltage; > > + u32 delay_reg; > > + u32 delay_bit_shift; > > + u32 delay_bit_width; > > +}; > > + > > +static u32 anatop_get_bits(struct udevice *dev, u32 addr, int bit_shift, > > + int bit_width) > > +{ > > + struct udevice *syscon; > > + struct regmap *regmap; > > + int err; > > + u32 val, mask; > > + > > + syscon = dev_get_parent(dev); > > + if (!syscon) { > > + dev_err(dev, "%s: unable to find syscon device\n", > __func__); > > + return err; > > + } > > + > > + regmap = syscon_get_regmap(syscon); > > + if (IS_ERR(regmap)) { > > + dev_err(dev, "%s: unable to find regmap (%ld)\n", __func__, > > + PTR_ERR(regmap)); > > + return PTR_ERR(regmap); > > + } > > + > > + if (bit_width == 32) > > Use macro instead of 32, plz. > > Will fix it in v2. > > + mask = ~0; > > + else > > + mask = (1 << bit_width) - 1; > > + > > + err = regmap_read(regmap, addr, &val); > > + if (err) > > + return err; > > + > > + val = (val >> bit_shift) & mask; > > + > > + return val; > > +} > > + > > +void anatop_set_bits(struct udevice *dev, u32 addr, int bit_shift, > > + int bit_width, u32 data) > > static? > doesn't it need to return int type? > If there is a somehting failed, it seems that it needs to pass the error > number. > (get_bits is returend to error..but set_bits doesn't return. It's strange.) > > Will fix it in v2. > And How about passing the struct anatop_regulator instead of each values? > anatop_get/set_bits(struct anatop_regulator *regulator) { > .. > } > > Will fix it in v2. > > +{ > > + struct udevice *syscon; > > + struct regmap *regmap; > > + int err; > > + u32 val, mask; > > + > > + syscon = dev_get_parent(dev); > > + if (!syscon) { > > + dev_err(dev, "%s: unable to find syscon device\n", > __func__); > > + return; > > + } > > + > > + regmap = syscon_get_regmap(syscon); > > + if (IS_ERR(regmap)) { > > + dev_err(dev, > > + "%s: unable to find regmap (%ld)\n", __func__, > > + PTR_ERR(regmap)); > > + return; > > + } > > + > > + if (bit_width == 32) > > + mask = ~0; > > + else > > + mask = (1 << bit_width) - 1; > > + > > + err = regmap_read(regmap, addr, &val); > > + if (err) { > > + dev_err(dev, > > + "%s: cannot read reg (%d)\n", __func__, > > + err); > > + return; > > + } > > + val = val & ~(mask << bit_shift); > > + err = regmap_write(regmap, addr, (data << bit_shift) | val); > > + if (err) { > > + dev_err(dev, > > + "%s: cannot wrie reg (%d)\n", __func__, > > + err); > > + return; > > + } > > +} > > + > > +static int anatop_set_voltage(struct udevice *dev, int uV) > > +{ > > + const struct anatop_regulator *anatop_reg = dev_get_plat(dev); > > + u32 val; > > + u32 sel; > > + int delayus = 0; > > + int ret = 0; > > + int uv; > > + > > + uv = uV; > > Not need to assign at here. Assign to above. > int uv = uV; > > Will fix it in v2. > > > + dev_dbg(dev, "%s: uv %d, min %d, max %d\n", __func__, > > + uv, anatop_reg->min_voltage, > > + anatop_reg->max_voltage); > > + > > + if (uv < anatop_reg->min_voltage) > > + return -EINVAL; > > + > > + if (!anatop_reg->control_reg) > > + return -ENOTSUPP; > > + > > + sel = DIV_ROUND_UP(uv - anatop_reg->min_voltage, 25000); > > What is 25000? If it's min value, use MACRO, plz. > > It is "STEP" or "RESOLUTION". Sorry for my bad English. 25000 uV is the anatop regulator's "STEP" or "RESOLUTION". So the uV is register value * 25000 uV + min uV. I'll change it to use MACRO with a proper name to avoid confusion. > > + if (sel * 25000 + anatop_reg->min_voltage > > anatop_reg->max_voltage) > > + return -EINVAL; > > + val = anatop_reg->min_bit_val + sel; > > + dev_dbg(dev, "%s: calculated val %d\n", __func__, val); > > + > > + /* check whether need to care about LDO ramp up speed */ > > + if (anatop_reg->delay_bit_width) { > > + /* > > + * the delay for LDO ramp up time is > > + * based on the register setting, we need > > + * to calculate how many steps LDO need to > > + * ramp up, and how much delay needed. (us) > > + */ > > + u32 old_sel; > > + u32 new_sel = sel; > > + > > + old_sel = anatop_get_bits(dev, > > + anatop_reg->control_reg, > > + anatop_reg->vol_bit_shift, > > + anatop_reg->vol_bit_width); > > + old_sel = old_sel - anatop_reg->min_bit_val; > > + > > + if (new_sel > old_sel) { > > + val = anatop_get_bits(dev, > > + anatop_reg->delay_reg, > > + anatop_reg->delay_bit_shift, > > + anatop_reg->delay_bit_width); > > + delayus = (new_sel - old_sel) * > > + (LDO_RAMP_UP_UNIT_IN_CYCLES << val) / > > + LDO_RAMP_UP_FREQ_IN_MHZ + 1; > > + } > > + } > > + > > + anatop_set_bits(dev, > > + anatop_reg->control_reg, > > + anatop_reg->vol_bit_shift, > > + anatop_reg->vol_bit_width, > > + val); > > + > > + if (delayus) > > + udelay(delayus); > > + > > + return ret; > > +} > > + > > +static int anatop_get_voltage(struct udevice *dev) > > +{ > > + const struct anatop_regulator *anatop_reg = dev_get_plat(dev); > > + u32 val; > > + u32 sel; > > + > > + val = anatop_get_bits(dev, > > + anatop_reg->control_reg, > > + anatop_reg->vol_bit_shift, > > + anatop_reg->vol_bit_width); > > + sel = val - anatop_reg->min_bit_val; > > + > > + return sel * 25000 + anatop_reg->min_voltage; > > +} > > + > > +static const struct dm_regulator_ops anatop_regulator_ops = { > > + .set_value = anatop_set_voltage, > > + .get_value = anatop_get_voltage, > > +}; > > + > > +static int anatop_regulator_probe(struct udevice *dev) > > +{ > > + struct anatop_regulator *sreg; > > + int ret = 0; > > + > > + sreg = dev_get_plat(dev); > > + if (!sreg) { > > + dev_err(dev, "failed to get plat data\n"); > > + return -ENOMEM; > > + } > > + > > + sreg->name = ofnode_read_string(dev_ofnode(dev), "regulator-name"); > > + if (!sreg->name) { > > + dev_err(dev, "failed to get a regulator-name\n"); > > + return -EINVAL; > > + } > > + > > + ret = ofnode_read_u32(dev_ofnode(dev), > > + "anatop-reg-offset", > > + &sreg->control_reg); > > + if (ret) { > > + dev_err(dev, "no anatop-reg-offset property set\n"); > > + return ret; > > + } > > + ret = ofnode_read_u32(dev_ofnode(dev), > > + "anatop-vol-bit-width", > > + &sreg->vol_bit_width); > > + if (ret) { > > + dev_err(dev, "no anatop-vol-bit-width property set\n"); > > + return ret; > > + } > > + ret = ofnode_read_u32(dev_ofnode(dev), > > + "anatop-vol-bit-shift", > > + &sreg->vol_bit_shift); > > + if (ret) { > > + dev_err(dev, "no anatop-vol-bit-shift property set\n"); > > + return ret; > > + } > > + ret = ofnode_read_u32(dev_ofnode(dev), > > + "anatop-min-bit-val", > > + &sreg->min_bit_val); > > + if (ret) { > > + dev_err(dev, "no anatop-min-bit-val property set\n"); > > + return ret; > > + } > > + ret = ofnode_read_u32(dev_ofnode(dev), > > + "anatop-min-voltage", > > + &sreg->min_voltage); > > I don't know why anatop-min-voltage is need? > Doesn't it use "regulator-min-microvolt"? > > It is needed. It is because the anatop-min-voltage is used to calculate the register value of anatop. But regulator-min-microvolt might not be equal to anatop-min-voltage. Because regulator-min-microvolt describes the device that consumes the regulator. So the device might need higher min-microvolt based on the hardware design and will be described in device tree settings. But the value to be written in the anatop register must be based on anatop-min-voltage. > > > + if (ret) { > > + dev_err(dev, "no anatop-min-voltage property set\n"); > > + return ret; > > + } > > + ret = ofnode_read_u32(dev_ofnode(dev), > > + "anatop-max-voltage", > > + &sreg->max_voltage); > > Ditto. > > Best Regards, > Jaehoon Chung > > > + if (ret) { > > + dev_err(dev, "no anatop-max-voltage property set\n"); > > + return ret; > > + } > > + > > + /* read LDO ramp up setting, only for core reg */ > > + ofnode_read_u32(dev_ofnode(dev), "anatop-delay-reg-offset", > > + &sreg->delay_reg); > > + ofnode_read_u32(dev_ofnode(dev), "anatop-delay-bit-width", > > + &sreg->delay_bit_width); > > + ofnode_read_u32(dev_ofnode(dev), "anatop-delay-bit-shift", > > + &sreg->delay_bit_shift); > > + > > + return 0; > > +} > > + > > +static const struct udevice_id of_anatop_regulator_match_tbl[] = { > > + { .compatible = "fsl,anatop-regulator", }, > > + { /* end */ } > > +}; > > + > > +U_BOOT_DRIVER(anatop_regulator) = { > > + .name = "anatop_regulator", > > + .id = UCLASS_REGULATOR, > > + .ops = &anatop_regulator_ops, > > + .of_match = of_anatop_regulator_match_tbl, > > + .plat_auto = sizeof(struct anatop_regulator), > > + .probe = anatop_regulator_probe, > > +}; > > > > Yours, Paul