Hi Simon, On Sun, Jul 25, 2021 at 1:01 AM Simon Glass <s...@chromium.org> wrote: > > Hi Oleksandr, > > On Wed, 21 Jul 2021 at 06:21, Oleksandr Suvorov > <oleksandr.suvo...@toradex.com> wrote: > > > > Initial support for Fairchild's 8 bit I2C gpio expander FXL6408. > > The CONFIG_FXL6408_GPIO define enables support for such devices. > > > > Based on: https://patchwork.kernel.org/patch/9148419/ > > > > Signed-off-by: Oleksandr Suvorov <oleksandr.suvo...@toradex.com> > > --- > > > > drivers/gpio/Kconfig | 7 + > > drivers/gpio/Makefile | 1 + > > drivers/gpio/gpio-fxl6408.c | 371 ++++++++++++++++++++++++++++++++++++ > > 3 files changed, 379 insertions(+) > > create mode 100644 drivers/gpio/gpio-fxl6408.c > > Reviewed-by: Simon Glass <s...@chromium.org> > > Lots of nits below
Thanks for your detailed and extremely useful review! > > > > diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig > > index 0817b12c5f..5883582a7f 100644 > > --- a/drivers/gpio/Kconfig > > +++ b/drivers/gpio/Kconfig > > @@ -123,6 +123,13 @@ config DA8XX_GPIO > > help > > This driver supports the DA8xx GPIO controller > > > > +config FXL6408_GPIO > > + bool "FXL6408 I2C GPIO driver" > > + depends on DM_GPIO && DM_I2C > > + help > > + Support for Fairchild Semiconductor FXL6408 I2C 8-bit GPIO > > + expander. > > Please add some more details here. Some details are being added to the next version of the patch set. > [..] > > > diff --git a/drivers/gpio/gpio-fxl6408.c b/drivers/gpio/gpio-fxl6408.c > > new file mode 100644 > > index 0000000000..282ec0a69c > > --- /dev/null > > +++ b/drivers/gpio/gpio-fxl6408.c > > @@ -0,0 +1,371 @@ > > +// SPDX-License-Identifier: GPL-2.0 > > +/* > > + * Copyright (C) 2021 Toradex > > + * Copyright (C) 2016 Broadcom > > + */ > > + > > +/** > > + * DOC: FXL6408 I2C to GPIO expander. > > + * > > + * This chip has 8 GPIO lines out of it, and is controlled by an I2C > > + * bus (a pair of lines), providing 4x expansion of GPIO lines. It > > + * also provides an interrupt line out for notifying of state changes. > > + * > > + * Any preconfigured state will be left in place until the GPIO lines > > + * get activated. At power on, everything is treated as an input, > > + * default input is HIGH and pulled-up, all interrupts are masked. > > + * > > + * Documentation can be found at: > > + * https://www.fairchildsemi.com/datasheets/FX/FXL6408.pdf > > + * > > + * This driver bases on: > > + * - the original driver by Eric Anholt <e...@anholt.net>: > > + * https://patchwork.kernel.org/patch/9148419/ > > + * - the Toradex version by Max Krummenacher > > <max.krummenac...@toradex.com>: > > + * > > http://git.toradex.com/cgit/linux-toradex.git/tree/drivers/gpio/gpio-fxl6408.c?h=toradex_5.4-2.3.x-imx > > + * - the U-boot PCA953x driver by Peng Fan <van.free...@gmail.com>: > > + * drivers/gpio/pca953x_gpio.c > > + * > > + * TODO: Add interrupts support > > + */ > > + > > +#include <common.h> > > +#include <dm.h> > > +#include <dm/device_compat.h> > > +#include <i2c.h> > > +#include <log.h> > > +#include <asm-generic/gpio.h> > > +#include <asm/global_data.h> > > +#include <linux/bitops.h> > > +#include <dt-bindings/gpio/gpio.h> > > + > > +#define FXL6408_DEVID_CTRL 0x01 > > Do you need the FXL6408 prefix? Actually, no :) Thanks, I renamed all registers in a straight manner. They'll be introduced in the next patchset version. > > +# define FXL6408_SW_RST BIT(0) > > +# define FXL6408_RST_INT BIT(1) > > + > > +/* 3-bit manufacturer ID */ > > +# define FXL6408_MF_MASK GENMASK(7, 5) > > +# define FXL6408_MF_ID(devid) (((devid) & FXL6408_MF_MASK) >> 5) > > +/* 0b101 is for Fairchild assigned by Nokia */ > > +# define FXL6408_FAIRCHILD_MF 5 > > + > > +/* 3-bit firmware revision */ > > +# define FXL6408_FW_MASK GENMASK(4, 2) > > +# define FXL6408_FW_REV(devid) (((devid) & FXL6408_FW_MASK) >> 2) > > This is only used once, so why not include that code inline below? This way is clearer to read as for me. Moreover, if we include only FW_REV()'s code inline, the shifting and the purpose of such shifting were in different places. Don't you mind leaving it as is? (with some names simplifying) > > + > > +/* > > + * Bits set here indicate that the GPIO is an output. > > single-line comment style is /* ... */ Thanks, fixed! > > + */ > > +#define FXL6408_DIRECTION 0x03 > > Then call it DIR_MASK ? Formally, it is not a mask but a register. All register names will be tuned in the next patchset version. > > + > > +enum { > > + FXL6408_DIRECTION_IN, > > + FXL6408_DIRECTION_OUT, > > +}; > > + > > +/* > > + * Bits set here, when the corresponding bit of IO_DIR is set, drive > > + * the output high instead of low. > > + */ > > +#define FXL6408_OUTPUT 0x05 > > + > > +/* > > + * Bits here make the output High-Z, instead of the OUTPUT value. > > + */ > > +#define FXL6408_OUTPUT_HIGH_Z 0x07 > > + > > +/* > > + * Bits here define the expected input state of the GPIO. > > + * INTERRUPT_STATUS bits will be set when the INPUT transitions away > > + * from this value. > > + */ > > +#define FXL6408_INPUT_DEF_STATE 0x09 > > + > > +/* > > + * Bits here enable either pull up or pull down according to > > + * INPUT_PULL_STATE. > > + */ > > +#define FXL6408_INPUT_PULL_ENABLE 0x0b > > + > > +/* > > + * Bits set here selects a pull-up/pull-down state of pin, which > > + * is configured as Input and the corresponding PULL_ENABLE bit is > > + * set. > > + */ > > +#define FXL6408_INPUT_PULL_STATE 0x0d > > + > > +/* > > + * Returns the current status (1 = HIGH) of the input pins. > > + */ > > +#define FXL6408_INPUT 0x0f > > + > > +/* > > + * Mask of pins which can generate interrupts. > > + */ > > +#define FXL6408_INTERRUPT_MASK 0x11 > > +/* > > + * Mask of pins which have generated an interrupt. > > + * Cleared on read. > > + */ > > +#define FXL6408_INTERRUPT_STATUS 0x13 > > + > > +/* > > + * struct fxl6408_info - Data for fxl6408 > > + * > > + * @dev: udevice structure for the device > > + * @addr: i2c slave address > > Please check; these don't match the struct. Thanks! Fixed. > > + * @reg_output: hold the value of output register > > + * @reg_direction: hold the value of direction register > > + */ > > +struct fxl6408_info { > > + struct udevice *dev; > > + int addr; > > + u8 device_id; > > + u8 reg_io_dir; > > + u8 reg_output; > > +}; > > + > > +static int fxl6408_write(struct udevice *dev, int reg, u8 val) > > +{ > > + int ret; > > + > > + ret = dm_i2c_write(dev, reg, &val, 1); > > + if (ret) > > + dev_err(dev, "%s error: %d\n", __func__, ret); > > + > > + return ret; > > +} > > + > > +static int fxl6408_read(struct udevice *dev, int reg, u8 *val) > > You could return the value, or -ve for error, saving a parameter. Good shot! Reimplemented. > > +{ > > + int ret; > > + u8 tmp; > > + > > + ret = dm_i2c_read(dev, reg, &tmp, 1); > > + if (!ret) > > + *val = tmp; > > + else > > + dev_err(dev, "%s error: %d\n", __func__, ret); > > + > > + return ret; > > +} > > + > > +/* > > + * fxl6408_is_output() - check whether the gpio configures as either > > + * output or input. > > + * Return: 0 - input, 1 - output. > > + */ > > +static int fxl6408_is_output(struct udevice *dev, int offset) > > s/int/bool/ and you can drop the !! here and below Done, thanks! > > +{ > > + struct fxl6408_info *info = dev_get_plat(dev); > > + > > + return !!(info->reg_io_dir & BIT(offset)); > > +} > > + > > +static int fxl6408_get_value(struct udevice *dev, uint offset) > > +{ > > + int ret, reg = fxl6408_is_output(dev, offset) ? FXL6408_OUTPUT : > > FXL6408_INPUT; > > + u8 val = 0; > > + > > + ret = fxl6408_read(dev, reg, &val); > > + if (IS_ERR_VALUE(ret)) > > + return ret; > > + > > + return !!(val & BIT(offset)); > > +} > > + > > +static int fxl6408_set_value(struct udevice *dev, uint offset, int value) > > +{ > > + struct fxl6408_info *info = dev_get_plat(dev); > > + u8 val; > > + int ret; > > + > > + if (value) > > + val = info->reg_output | BIT(offset); > > + else > > + val = info->reg_output & ~BIT(offset); > > + > > + ret = fxl6408_write(dev, FXL6408_OUTPUT, val); > > + if (!IS_ERR_VALUE(ret)) > > if (ret < 0) is good enough here > > We are not going to change that, and the macros obfuscates things IMO. Replaced, thanks. > > + info->reg_output = val; > > + > > + return ret; > > +} > > + > > +static int fxl6408_set_direction(struct udevice *dev, uint offset, int dir) > > +{ > > + struct fxl6408_info *info = dev_get_plat(dev); > > + u8 val; > > + int ret; > > + > > + if (dir == FXL6408_DIRECTION_IN) > > Then shouldn't 'dir' be an enum? Good point! Done. > > + val = info->reg_io_dir & ~BIT(offset); > > + else > > + val = info->reg_io_dir | BIT(offset); > > + > > + ret = fxl6408_write(dev, FXL6408_DIRECTION, val); > > + if (!IS_ERR_VALUE(ret)) > > + info->reg_io_dir = val; > > + > > + return ret; > > +} > > + > > +static int fxl6408_direction_input(struct udevice *dev, uint offset) > > +{ > > + return fxl6408_set_direction(dev, offset, FXL6408_DIRECTION_IN); > > +} > > + > > +static int fxl6408_direction_output(struct udevice *dev, uint offset, int > > value) > > +{ > > + int ret; > > + > > + /* Configure output value. */ > > + ret = fxl6408_set_value(dev, offset, value); > > + if (!IS_ERR_VALUE(ret)) > > + /* Configure direction as output. */ > > + fxl6408_set_direction(dev, offset, FXL6408_DIRECTION_OUT); > > + > > + return ret; > > +} > > + > > +static int fxl6408_get_function(struct udevice *dev, uint offset) > > +{ > > + if (fxl6408_is_output(dev, offset)) > > + return GPIOF_OUTPUT; > > + else > > + return GPIOF_INPUT; > > +} > > + > > +static int fxl6408_xlate(struct udevice *dev, struct gpio_desc *desc, > > + struct ofnode_phandle_args *args) > > +{ > > + desc->offset = args->args[0]; > > + desc->flags = args->args[1] & GPIO_ACTIVE_LOW ? GPIOD_ACTIVE_LOW : > > 0; > > + > > + return 0; > > +} > > + > > +static const struct dm_gpio_ops fxl6408_ops = { > > + .direction_input = fxl6408_direction_input, > > + .direction_output = fxl6408_direction_output, I've just found out these callbacks are deprecated. So I'll add set_flags and get_flags callback's implementations in the next version. > > + .get_value = fxl6408_get_value, > > + .set_value = fxl6408_set_value, > > + .get_function = fxl6408_get_function, > > + .xlate = fxl6408_xlate, > > +}; > > Can you put that down below with the other struct? Done. > > + > > +static int fxl6408_probe(struct udevice *dev) > > +{ > > + struct fxl6408_info *info = dev_get_plat(dev); > > + struct gpio_dev_priv *uc_priv = dev_get_uclass_priv(dev); > > + char name[32], label[32], *str; > > + int addr; > > + int ret; > > + int size; > > + const u8 *tmp; > > + u8 val; > > + u32 val32; > > + > > + addr = dev_read_addr(dev); > > + if (addr == 0) > > + return -ENODEV; > > This has a special meaning in drive model. Try -EINVAL instead for > errors reading the DT. Thanks! Fixed. > > + info->addr = addr; > > + > > + /* Check the device ID register to see if it's responding. > > Multi-line comment style is: Thanks, fixed! > /* > * Check the > * .. > */ > > + * This also clears RST_INT as a side effect, so we won't get > > + * the "we've been power cycled" interrupt once we enable > > + * interrupts. > > + */ > > + ret = fxl6408_read(dev, FXL6408_DEVID_CTRL, &val); > > + if (IS_ERR_VALUE(ret)) { > > + dev_err(dev, "FXL6408 probe returned %d\n", ret); > > + return ret; > > + } > > + > > + if (FXL6408_MF_ID(val) != FXL6408_FAIRCHILD_MF) { > > + dev_err(dev, "FXL6408 probe returned DID: 0x%02x\n", val); > > + return -ENODEV; > > Again you cannot return this, as there is a device. How about -ENXIO? Sounds logical. Fixed. > > + } > > + info->device_id = val; > > + > > + /* > > + * Disable High-Z of outputs, so that the OUTPUT updates > > + * actually take effect. > > + */ > > + ret = fxl6408_write(dev, FXL6408_OUTPUT_HIGH_Z, (u8)0); > > + if (IS_ERR_VALUE(ret)) { > > + dev_err(dev, "Error writing High-Z register\n"); > > + return ret; > > + } > > + > > + /* > > + * If configured, set initial output state and direction, > > + * otherwise read them from the chip. > > + */ > > + if (IS_ERR_VALUE(dev_read_u32(dev, "initial_io_dir", &val32))) { > > + ret = fxl6408_read(dev, FXL6408_DIRECTION, > > &info->reg_io_dir); > > + if (IS_ERR_VALUE(ret)) { > > + dev_err(dev, "Error reading direction register\n"); > > + return ret; > > + } > > + } else { > > + info->reg_io_dir = val32 & 0xFF; > > + ret = fxl6408_write(dev, FXL6408_DIRECTION, > > info->reg_io_dir); > > + if (IS_ERR_VALUE(ret)) { > > + dev_err(dev, "Error setting direction register\n"); > > + return ret; > > + } > > + } > > + > > + if (IS_ERR_VALUE(dev_read_u32(dev, "initial_output", &val32))) { > > + ret = fxl6408_read(dev, FXL6408_OUTPUT, &info->reg_output); > > + if (IS_ERR_VALUE(ret)) { > > + dev_err(dev, "Error reading output register\n"); > > + return ret; > > + } > > + } else { > > + info->reg_output = val32 & 0xFF; > > + ret = fxl6408_write(dev, FXL6408_OUTPUT, info->reg_output); > > + if (IS_ERR_VALUE(ret)) { > > + dev_err(dev, "Error setting output register\n"); > > + return ret; > > + } > > + } > > + > > + tmp = dev_read_prop(dev, "label", &size); > > + if (tmp) { > > + size = min(size, (int)sizeof(label) - 1); > > + memcpy(label, tmp, size); > > + label[size] = '\0'; > > + snprintf(name, sizeof(name), "%s@%x_", label, info->addr); > > + } else { > > + snprintf(name, sizeof(name), "gpio@%x_", info->addr); > > + } > > + > > + str = strdup(name); > > + if (!str) > > + return -ENOMEM; > > + > > + uc_priv->bank_name = str; > > + uc_priv->gpio_count = dev_get_driver_data(dev); > > + uc_priv->gpio_base = -1; > > + > > + dev_dbg(dev, "%s (FW rev. %ld) is ready\n", str, > > + FXL6408_FW_REV(info->device_id)); > > + > > + return 0; > > +} > > + > > +static const struct udevice_id fxl6408_ids[] = { > > + { .compatible = "fcs,fxl6408", .data = 8 }, > > + { } > > +}; > > + > > +U_BOOT_DRIVER(fxl6408_gpio) = { > > + .name = "fxl6408_gpio", > > + .id = UCLASS_GPIO, > > + .ops = &fxl6408_ops, > > + .probe = fxl6408_probe, > > + .of_match = fxl6408_ids, > > + .plat_auto = sizeof(struct fxl6408_info), > > +}; > > -- > > 2.31.1 > > > > Regards, > Simon Best regards Oleksandr Suvorov Toradex AG Ebenaustrasse 10 | 6048 Horw | Switzerland | T: +41 41 500 48 00
