[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20201127111429.GR2455276@dell>
Date: Fri, 27 Nov 2020 11:14:29 +0000
From: Lee Jones <lee.jones@...aro.org>
To: Shihlun Lin <shihlun.lin@...antech.com.tw>
Cc: "David S . Miller" <davem@...emloft.net>,
Rob Herring <robh@...nel.org>,
Jean Delvare <jdelvare@...e.com>,
Guenter Roeck <linux@...ck-us.net>,
Wim Van Sebroeck <wim@...ux-watchdog.org>,
linux-kernel@...r.kernel.org, devicetree@...r.kernel.org,
linux-hwmon@...r.kernel.org, linux-watchdog@...r.kernel.org,
Campion Kang <campion.kang@...antech.com.tw>,
AceLan Kao <chia-lin.kao@...onical.com>
Subject: Re: [RESEND PATCH v4 4/6] mfd: ahc1ec0: Add support for Advantech
embedded controller
On Wed, 25 Nov 2020, Shihlun Lin wrote:
> AHC1EC0 is the embedded controller driver for Advantech industrial
> products. This provides sub-devices such as hwmon and watchdog, and also
> expose functions for sub-devices to read/write the value to embedded
> controller.
>
> Signed-off-by: Shihlun Lin <shihlun.lin@...antech.com.tw>
> ---
> drivers/mfd/Kconfig | 10 +
> drivers/mfd/Makefile | 2 +
> drivers/mfd/ahc1ec0.c | 1405 +++++++++++++++++++++++++++++++++++
> include/linux/mfd/ahc1ec0.h | 338 +++++++++
> 4 files changed, 1755 insertions(+)
> create mode 100644 drivers/mfd/ahc1ec0.c
> create mode 100644 include/linux/mfd/ahc1ec0.h
>
> diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
> index 8b99a13669bf..1cc40217f798 100644
> --- a/drivers/mfd/Kconfig
> +++ b/drivers/mfd/Kconfig
> @@ -2166,5 +2166,15 @@ config MFD_INTEL_M10_BMC
> additional drivers must be enabled in order to use the functionality
> of the device.
>
> +config MFD_AHC1EC0
> + tristate "Advantech Embedded Controller Module"
> + depends on X86
> + select MFD_CORE
> + help
> + This is the core function that for Advantech EC drivers. It
> + includes the sub-devices such as hwmon, watchdog, etc. And also
> + provides expose functions for sub-devices to read/write the value
> + to embedded controller.
> +
> endmenu
> endif
> diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
> index 1780019d2474..80a9a2bdc3ba 100644
> --- a/drivers/mfd/Makefile
> +++ b/drivers/mfd/Makefile
> @@ -267,3 +267,5 @@ obj-$(CONFIG_MFD_KHADAS_MCU) += khadas-mcu.o
> obj-$(CONFIG_SGI_MFD_IOC3) += ioc3.o
> obj-$(CONFIG_MFD_SIMPLE_MFD_I2C) += simple-mfd-i2c.o
> obj-$(CONFIG_MFD_INTEL_M10_BMC) += intel-m10-bmc.o
> +
> +obj-$(CONFIG_MFD_AHC1EC0) += ahc1ec0.o
> diff --git a/drivers/mfd/ahc1ec0.c b/drivers/mfd/ahc1ec0.c
> new file mode 100644
> index 000000000000..768d2063bed1
> --- /dev/null
> +++ b/drivers/mfd/ahc1ec0.c
> @@ -0,0 +1,1405 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/*
> + * Base driver for Advantech controlling EC chip AHC1EC0
> + *
> + * Copyright (C) 2020, Advantech Automation Corp.
> + *
> + * Change Log:
> + * Version 1.00 <10/10/2014> Sun.Lang
> + * - Initial version
> + * Version 1.01 <11/05/2015> Jiangwei.Zhu
> + * - Modify read_ad_value() function.
> + * - Add smbus_read_byte() function.
> + * - Add smbus_write_byte() function.
> + * - Add wait_smbus_protocol_finish() function.
> + * Version 1.02 <03/04/2016> Jiangwei.Zhu
> + * - Add smbus_read_word() function.
> + * Version 1.03 <01/22/2017> Ji.Xu
> + * - Add detect to Advantech porduct name "ECU".
> + * Version 1.04 <09/20/2017> Ji.Xu
> + * - Update to support detect Advantech product name in UEFI
> + * BIOS(DMI).
> + * Version 1.05 <11/02/2017> Ji.Xu
> + * - Fixed issue: Cache coherency error when exec 'ioremap_uncache()'
> + * in kernel-4.10.
> + * Version 2.00 <11/04/2020> Shihlun.Lin
> + * - Update: Replace ioremap_nocache() with ioremap_uc() since
> + * ioremap_uc() was used on the entire PCI BAR.
> + */
This is v1.0.
> +#include <linux/kconfig.h>
> +#include <linux/module.h>
> +#include <linux/errno.h>
> +#include <linux/delay.h>
> +#include <linux/io.h>
> +#include <linux/wait.h>
> +#include <linux/slab.h>
> +#include <linux/mutex.h>
> +#include <linux/mfd/ahc1ec0.h>
> +#include <linux/acpi.h>
> +#include <linux/mod_devicetable.h>
> +#include <linux/mfd/core.h>
> +#include <linux/of_platform.h>
> +#include <linux/platform_device.h>
Alphabetical.
> +#define ADVANTECH_EC_NAME "ahc1ec0"
No need to define names.
Please just use the string where you need it.
> +#define ADVANTECH_EC_MFD_VER "2.0.0"
> +#define ADVANTECH_EC_MFD_DATE "11/04/2020"
No software version numbers or dates please.
This will artificially age the driver.
> +struct mutex lock;
Not global please. Place this in driver data.
> +enum {
> + ADVEC_SUBDEV_BRIGHTNESS = 0,
> + ADVEC_SUBDEV_EEPROM,
> + ADVEC_SUBDEV_GPIO,
> + ADVEC_SUBDEV_HWMON,
> + ADVEC_SUBDEV_LED,
> + ADVEC_SUBDEV_WDT,
> + ADVEC_SUBDEV_MAX,
> +};
Why do you need to enumerate these?
Are they numbered off in the datasheet?
> +static int wait_ibf(void)
ibf?
Please use better/more forthcoming nomenclature.
> +{
> + int i;
> +
> + for (i = 0; i < EC_MAX_TIMEOUT_COUNT; i++) {
> + if ((inb(EC_COMMAND_PORT) & 0x02) == 0)
No magic numbers. Please define all of them.
> + return 0;
> +
> + udelay(EC_UDELAY_TIME);
This is not a good name for a define.
What are you delaying/waiting for?
> + }
> +
> + return -ETIMEDOUT;
> +}
> +
> +/* Wait OBF (Output buffer full) set */
If you name the function better, you won't need to supplement with
comments.
> +static int wait_obf(void)
> +{
> + int i;
> +
> + for (i = 0; i < EC_MAX_TIMEOUT_COUNT; i++) {
> + if ((inb(EC_COMMAND_PORT) & 0x01) != 0)
> + return 0;
> +
> + udelay(EC_UDELAY_TIME);
> + }
> +
> + return -ETIMEDOUT;
> +}
> +
> +/* Read data from EC HW ram */
> +static int read_hw_ram(uchar addr, uchar *data)
> +{
> + int ret;
> +
> + mutex_lock(&lock);
> +
> + /* Step 0. Wait IBF clear */
> + ret = wait_ibf();
> + if (ret)
> + goto error;
> +
> + /* Step 1. Send "read EC HW ram" command to EC Command port */
Not required.
> + outb(EC_HW_RAM_READ, EC_COMMAND_PORT);
> +
> + /* Step 2. Wait IBF clear */
Better function names eliminates the need for this.
Same for below too.
> + ret = wait_ibf();
> + if (ret)
> + goto error;
> +
> + /* Step 3. Send "EC HW ram" address to EC Data port */
> + outb(addr, EC_STATUS_PORT);
> +
> + /* Step 4. Wait OBF set */
> + ret = wait_obf();
> + if (ret)
> + goto error;
> +
> + /* Step 5. Get "EC HW ram" data from EC Data port */
> + *data = inb(EC_STATUS_PORT);
> +
> + mutex_unlock(&lock);
> +
> + return ret;
> +
> +error:
> + mutex_unlock(&lock);
> + pr_err("%s: Wait for IBF or OBF too long. line: %d", __func__, __LINE__);
> + return ret;
> +}
Any reason not to use regmap?
> +/* Write data to EC HW ram */
> +int write_hw_ram(uchar addr, uchar data)
> +{
> + int ret;
> +
> + mutex_lock(&lock);
> + /* Step 0. Wait IBF clear */
> + ret = wait_ibf();
> + if (ret)
> + goto error;
> +
> + /* Step 1. Send "write EC HW ram" command to EC command port */
> + outb(EC_HW_RAM_WRITE, EC_COMMAND_PORT);
> +
> + /* Step 2. Wait IBF clear */
> + ret = wait_ibf();
> + if (ret)
> + goto error;
> +
> + /* Step 3. Send "EC HW ram" address to EC Data port */
> + outb(addr, EC_STATUS_PORT);
> +
> + /* Step 4. Wait IBF clear */
> + ret = wait_ibf();
> + if (ret)
> + goto error;
> +
> + /* Step 5. Send "EC HW ram" data to EC Data port */
> + outb(data, EC_STATUS_PORT);
> +
> + mutex_unlock(&lock);
> +
> + return 0;
> +
> +error:
> + mutex_unlock(&lock);
> + pr_err("%s: Wait for IBF too long. line: %d", __func__, __LINE__);
> + return ret;
> +}
> +EXPORT_SYMBOL_GPL(write_hw_ram);
[Big snip!]
None of this belongs in MFD. See below at the prototypes.
> +static const struct mfd_cell adv_ec_sub_cells[] = {
> + { .name = "adv-ec-brightness", },
> + { .name = "adv-ec-eeprom", },
> + { .name = "adv-ec-gpio", },
> + { .name = "adv-ec-hwmon", },
> + { .name = "adv-ec-led", },
> + { .name = "adv-ec-wdt", },
> +};
> +
> +static int adv_ec_init_ec_data(struct adv_ec_platform_data *pdata)
> +{
> + int ret;
> +
> + pdata->sub_dev_mask = 0;
> + pdata->sub_dev_nb = 0;
> + pdata->dym_tbl = NULL;
> + pdata->bios_product_name = NULL;
> +
> + /* Get product name */
> + pdata->bios_product_name = kmalloc(AMI_ADVANTECH_BOARD_ID_LENGTH, GFP_KERNEL);
> + if (!pdata->bios_product_name)
> + return -ENOMEM;
> +
> + memset(pdata->bios_product_name, 0, AMI_ADVANTECH_BOARD_ID_LENGTH);
> + ret = adv_ec_get_productname(pdata->bios_product_name);
> + if (ret)
> + return ret;
> +
> + /* Get pin table */
> + pdata->dym_tbl = kmalloc(EC_MAX_TBL_NUM*sizeof(struct Dynamic_Tab), GFP_KERNEL);
> + if (!pdata->dym_tbl)
> + return -ENOMEM;
> +
> + ret = adv_get_dynamic_tab(pdata);
> +
> + return 0;
> +}
> +
> +static int adv_ec_parse_prop(struct adv_ec_platform_data *pdata)
> +{
> + int i, ret;
> + u32 nb, sub_dev[ADVEC_SUBDEV_MAX];
> +
> + ret = device_property_read_u32(pdata->dev, "advantech,sub-dev-nb", &nb);
> + if (ret < 0) {
> + dev_err(pdata->dev, "get sub-dev-nb failed! (%d)", ret);
> + return ret;
> + }
> + pdata->sub_dev_nb = nb;
> +
> + ret = device_property_read_u32_array(pdata->dev, "advantech,sub-dev", sub_dev, nb);
> + if (ret < 0) {
> + dev_err(pdata->dev, "get sub-dev failed! (%d)", ret);
> + return ret;
> + }
> +
> + for (i = 0; i < nb; i++) {
> + switch (sub_dev[i]) {
> + case ADVEC_SUBDEV_BRIGHTNESS:
> + case ADVEC_SUBDEV_EEPROM:
> + case ADVEC_SUBDEV_GPIO:
> + case ADVEC_SUBDEV_HWMON:
> + case ADVEC_SUBDEV_LED:
> + case ADVEC_SUBDEV_WDT:
> + pdata->sub_dev_mask |= BIT(sub_dev[i]);
> + break;
> + default:
> + dev_err(pdata->dev, "invalid prop value(%d)!", sub_dev[i]);
> + }
> + }
> + dev_info(pdata->dev, "sub-dev mask = 0x%x", pdata->sub_dev_mask);
> +
> + return 0;
> +}
> +
> +static int adv_ec_probe(struct platform_device *pdev)
> +{
> + int ret, i;
> + struct device *dev = &pdev->dev;
> + struct adv_ec_platform_data *adv_ec_data;
> +
> + adv_ec_data = kmalloc(sizeof(struct adv_ec_platform_data), GFP_KERNEL);
Use devm_* managed resource, then you don't have to worry about
freeing it on error/release.
> + if (!adv_ec_data) {
> + ret = -ENOMEM;
> + goto err_plat_data;
Just return -ENOMEM here.
> + }
> +
> + dev_set_drvdata(dev, (void *)adv_ec_data);
You shouldn't need to cast this.
> + adv_ec_data->dev = dev;
> +
> + mutex_init(&lock);
> +
> + ret = adv_ec_init_ec_data(adv_ec_data);
> + if (ret)
> + goto err_init_data;
> +
> + ret = adv_ec_parse_prop(adv_ec_data);
> + if (ret)
> + goto err_prop;
> +
> + for (i = 0; i < ARRAY_SIZE(adv_ec_sub_cells); i++) {
> + if (adv_ec_data->sub_dev_mask & BIT(i)) {
> + ret = mfd_add_hotplug_devices(dev, &adv_ec_sub_cells[i], 1);
Why are you passing them in individually?
Just pass the whole struct and let the framework do the rest.
> + if (ret)
> + dev_err(dev, "failed to add %s subdevice: %d",
> + adv_ec_sub_cells[i].name, ret);
> + }
> + }
> +
> + dev_info(&pdev->dev, "Ver:%s, Data:%s, probe done",
> + ADVANTECH_EC_MFD_VER, ADVANTECH_EC_MFD_DATE);
Remove this.
> + return 0;
> +
> +err_prop:
> +err_init_data:
> + kfree(adv_ec_data->dym_tbl);
> + kfree(adv_ec_data->bios_product_name);
> + kfree(adv_ec_data);
Use devm_* then you don't have do worry about this.
> +err_plat_data:
> + return ret;
> +}
> +
> +static int adv_ec_remove(struct platform_device *pdev)
> +{
> + struct adv_ec_platform_data *adv_ec_data;
> +
> + adv_ec_data = (struct adv_ec_platform_data *)dev_get_drvdata(&pdev->dev);
You shouldn't need to cast.
> + kfree(adv_ec_data->dym_tbl);
> + kfree(adv_ec_data->bios_product_name);
> + kfree(adv_ec_data);
Not needed.
> + mfd_remove_devices(&pdev->dev);
Use devm_*
> + mutex_destroy(&lock);
> +
> + return 0;
> +}
> +
> +#ifdef CONFIG_OF
> +static const struct of_device_id adv_ec_of_match[] = {
> + { .compatible = "advantech,ahc1ec0", },
> + {}
> +};
> +MODULE_DEVICE_TABLE(of, adv_ec_of_match);
> +#endif
> +
> +#ifdef CONFIG_ACPI
> +static const struct acpi_device_id adv_ec_acpi_match[] = {
> + {"AHC1EC0", },
> + { },
> +};
> +MODULE_DEVICE_TABLE(acpi, adv_ec_acpi_match);
> +#endif
> +
> +static const struct platform_device_id adv_ec_id[] = {
> + { ADVANTECH_EC_NAME, },
> + {}
> +};
> +MODULE_DEVICE_TABLE(platform, adv_ec_id);
Can you tell me what this does?
> +static struct platform_driver adv_ec_driver = {
> + .driver = {
> + .name = ADVANTECH_EC_NAME,
> +#ifdef CONFIG_OF
> + .of_match_table = of_match_ptr(adv_ec_of_match),
> +#endif
> +#ifdef CONFIG_ACPI
> + .acpi_match_table = ACPI_PTR(adv_ec_acpi_match),
> +#endif
Remove all of the #ifery above.
Just mark adv_ec_of_match and adv_ec_acpi_match as __maybe_unused.
> + },
> + .id_table = adv_ec_id,
> + .probe = adv_ec_probe,
> + .remove = adv_ec_remove,
> +};
> +module_platform_driver(adv_ec_driver);
> +
> +MODULE_LICENSE("Dual BSD/GPL");
> +MODULE_AUTHOR("sun.lang");
Contact details please i.e. email address.
> +MODULE_DESCRIPTION("Advantech EC MFD Driver.");
No such thing as an "MFD Driver". Please describe what it does/is.
> diff --git a/include/linux/mfd/ahc1ec0.h b/include/linux/mfd/ahc1ec0.h
> new file mode 100644
> index 000000000000..9c1991695305
> --- /dev/null
> +++ b/include/linux/mfd/ahc1ec0.h
> @@ -0,0 +1,338 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +#ifndef __LINUX_MFD_AHC1EC0_H
> +#define __LINUX_MFD_AHC1EC0_H
> +
> +#include <linux/device.h>
> +
> +#define uchar unsigned int
Nope. Standard types please.
> +#define EC_COMMAND_PORT 0x29A
> +#define EC_STATUS_PORT 0x299
> +
> +#define EC_UDELAY_TIME 200
> +#define EC_MAX_TIMEOUT_COUNT 5000
Where did those values come from?
> +/* AD command */
What is "AD"?
> +#define EC_AD_INDEX_WRITE 0x15
> +#define EC_AD_LSB_READ 0x16
> +#define EC_AD_MSB_READ 0x1F
> +
> +/* voltage device id */
"Voltage device ID"
> +#define EC_DID_CMOSBAT 0x50 /* CMOS coin battery voltage */
> +#define EC_DID_CMOSBAT_X2 0x51 /* CMOS coin battery voltage*2 */
> +#define EC_DID_CMOSBAT_X10 0x52 /* CMOS coin battery voltage*10 */
> +#define EC_DID_5VS0 0x56 /* 5VS0 voltage */
> +#define EC_DID_5VS0_X2 0x57 /* 5VS0 voltage*2 */
> +#define EC_DID_5VS0_X10 0x58 /* 5VS0 voltage*10 */
> +#define EC_DID_5VS5 0x59 /* 5VS5 voltage */
> +#define EC_DID_5VS5_X2 0x5A /* 5VS5 voltage*2 */
> +#define EC_DID_5VS5_X10 0x5B /* 5VS5 voltage*10 */
> +#define EC_DID_12VS0 0x62 /* 12VS0 voltage */
> +#define EC_DID_12VS0_X2 0x63 /* 12VS0 voltage*2 */
> +#define EC_DID_12VS0_X10 0x64 /* 12VS0 voltage*10 */
> +#define EC_DID_VCOREA 0x65 /* CPU A core voltage */
> +#define EC_DID_VCOREA_X2 0x66 /* CPU A core voltage*2 */
> +#define EC_DID_VCOREA_X10 0x67 /* CPU A core voltage*10 */
> +#define EC_DID_VCOREB 0x68 /* CPU B core voltage */
> +#define EC_DID_VCOREB_X2 0x69 /* CPU B core voltage*2 */
> +#define EC_DID_VCOREB_X10 0x6A /* CPU B core voltage*10 */
> +#define EC_DID_DC 0x6B /* ADC. onboard voltage */
> +#define EC_DID_DC_X2 0x6C /* ADC. onboard voltage*2 */
> +#define EC_DID_DC_X10 0x6D /* ADC. onboard voltage*10 */
> +#define EC_DID_SMBOEM0 0x28 /* SMBUS/I2C. Smbus channel 0 */
This should be in order.
> +/* Current device id */
"ID"
> +#define EC_DID_CURRENT 0x74
> +
> +/* ACPI commands */
> +#define EC_ACPI_RAM_READ 0x80
> +#define EC_ACPI_RAM_WRITE 0x81
> +
> +/*
> + * Dynamic control table commands
> + * The table includes HW pin number,Device ID,and Pin polarity
> + */
Spaces after the ','s.
> +#define EC_TBL_WRITE_ITEM 0x20
> +#define EC_TBL_GET_PIN 0x21
> +#define EC_TBL_GET_DEVID 0x22
> +#define EC_MAX_TBL_NUM 32
> +
> +/* LED Device ID table */
> +#define EC_DID_LED_RUN 0xE1
> +#define EC_DID_LED_ERR 0xE2
> +#define EC_DID_LED_SYS_RECOVERY 0xE3
> +#define EC_DID_LED_D105_G 0xE4
> +#define EC_DID_LED_D106_G 0xE5
> +#define EC_DID_LED_D107_G 0xE6
> +
> +/* LED control HW ram address 0xA0-0xAF */
"RAM"
> +#define EC_HWRAM_LED_BASE_ADDR 0xA0
> +#define EC_HWRAM_LED_PIN(N) (EC_HWRAM_LED_BASE_ADDR + (4 * (N))) // N:0-3
> +#define EC_HWRAM_LED_CTRL_HIBYTE(N) (EC_HWRAM_LED_BASE_ADDR + (4 * (N)) + 1)
> +#define EC_HWRAM_LED_CTRL_LOBYTE(N) (EC_HWRAM_LED_BASE_ADDR + (4 * (N)) + 2)
> +#define EC_HWRAM_LED_DEVICE_ID(N) (EC_HWRAM_LED_BASE_ADDR + (4 * (N)) + 3)
> +
> +/* LED control bit */
> +#define LED_CTRL_ENABLE_BIT (1 << 4)
> +#define LED_CTRL_INTCTL_BIT (1 << 5)
BIT()
> +#define LED_CTRL_LEDBIT_MASK (0x03FF << 6)
> +#define LED_CTRL_POLARITY_MASK (0x000F << 0)
> +#define LED_CTRL_INTCTL_EXTERNAL 0
> +#define LED_CTRL_INTCTL_INTERNAL 1
> +
> +#define LED_DISABLE 0x0
> +#define LED_ON 0x1
> +#define LED_FAST 0x3
> +#define LED_NORMAL 0x5
> +#define LED_SLOW 0x7
> +#define LED_MANUAL 0xF
> +
> +#define LED_CTRL_LEDBIT_DISABLE (0x0000)
> +#define LED_CTRL_LEDBIT_ON (0x03FF)
> +#define LED_CTRL_LEDBIT_FAST (0x02AA)
> +#define LED_CTRL_LEDBIT_NORMAL (0x0333)
> +#define LED_CTRL_LEDBIT_SLOW (0x03E0)
Why are some of these in brackets and others aren't?
> +/* Get the device name */
> +#define AMI_BIOS_NAME "AMIBIOS"
Just use the string in-place.
> +#define AMI_BIOS_NAME_ADDRESS 0x000FF400
> +#define AMI_BIOS_NAME_LENGTH strlen(AMI_BIOS_NAME)
> +#define AMI_ADVANTECH_BOARD_ID_ADDRESS 0x000FE840
> +#define AMI_ADVANTECH_BOARD_ID_LENGTH 32
> +#define AMI_ADVANTECH_BOARD_NAME_ADDRESS 0x000FF550
> +#define AMI_ADVANTECH_BOARD_NAME_LENGTH _ADVANTECH_BOARD_NAME_LENGTH
> +#define AMI_UEFI_ADVANTECH_BOARD_NAME_ADDRESS 0x000F0000
> +#define AMI_UEFI_ADVANTECH_BOARD_NAME_LENGTH 0xFFFF
> +
> +/*
> + * EC WDT commands
Expand "EC WDT"
> + * EC can send multistage watchdog event. System can setup watchdog event
"multi-stage"
> + * independently to makeup event sequence.
"make up"
> + */
> +#define EC_WDT_START 0x28
> +#define EC_WDT_STOP 0x29
> +#define EC_WDT_RESET 0x2A
> +#define EC_WDT_BOOTTMEWDT_STOP 0x2B
> +#define EC_HW_RAM 0x89
> +#define EC_EVENT_FLAG 0x57
> +#define EC_RESET_EVENT 0x04
> +#define EC_COMMANS_PORT_IBF_MASK 0x02
> +#define EC_ENABLE_DELAY_L 0x59
> +#define EC_ENABLE_DELAY_H 0x58
> +#define EC_POWER_BTN_TIME_L 0x5B
> +#define EC_POWER_BTN_TIME_H 0x5A
> +#define EC_RESET_DELAY_TIME_L 0x5F
> +#define EC_RESET_DELAY_TIME_H 0x5E
> +#define EC_PIN_DELAY_TIME_L 0x61
> +#define EC_PIN_DELAY_TIME_H 0x60
> +#define EC_SCI_DELAY_TIME_H 0x62
> +#define EC_SCI_DELAY_TIME_L 0x63
In order or separate.
> +/* EC ACPI commands */
> +#define EC_ACPI_DATA_READ 0x80
> +#define EC_ACPI_DATA_WRITE 0x81
> +
> +/* Brightness ACPI Addr */
> +#define BRIGHTNESS_ACPI_ADDR 0x50
> +
> +/* EC HW Ram commands */
> +#define EC_HW_EXTEND_RAM_READ 0x86
> +#define EC_HW_EXTEND_RAM_WRITE 0x87
> +#define EC_HW_RAM_READ 0x88
> +#define EC_HW_RAM_WRITE 0x89
> +
> +/* EC Smbus commands */
> +#define EC_SMBUS_CHANNEL_SET 0x8A /* Set selector number (SMBUS channel) */
> +#define EC_SMBUS_ENABLE_I2C 0x8C /* Enable channel I2C */
> +#define EC_SMBUS_DISABLE_I2C 0x8D /* Disable channel I2C */
> +
> +/* Smbus transmit protocol */
> +#define EC_SMBUS_PROTOCOL 0x00
> +
> +/* SMBUS status */
> +#define EC_SMBUS_STATUS 0x01
> +
> +/* SMBUS device slave address (bit0 must be 0) */
> +#define EC_SMBUS_SLV_ADDR 0x02
> +
> +/* SMBUS device command */
> +#define EC_SMBUS_CMD 0x03
> +
> +/* 0x04-0x24 Data In read process, return data are stored in this address */
> +#define EC_SMBUS_DATA 0x04
> +
> +#define EC_SMBUS_DAT_OFFSET(n) (EC_SMBUS_DATA + (n))
> +
> +/* SMBUS channel selector (0-4) */
> +#define EC_SMBUS_CHANNEL 0x2B
> +
> +/* EC SMBUS transmit Protocol code */
> +#define SMBUS_QUICK_WRITE 0x02 /* Write Quick Command */
> +#define SMBUS_QUICK_READ 0x03 /* Read Quick Command */
> +#define SMBUS_BYTE_SEND 0x04 /* Send Byte */
> +#define SMBUS_BYTE_RECEIVE 0x05 /* Receive Byte */
> +#define SMBUS_BYTE_WRITE 0x06 /* Write Byte */
> +#define SMBUS_BYTE_READ 0x07 /* Read Byte */
> +#define SMBUS_WORD_WRITE 0x08 /* Write Word */
> +#define SMBUS_WORD_READ 0x09 /* Read Word */
> +#define SMBUS_BLOCK_WRITE 0x0A /* Write Block */
> +#define SMBUS_BLOCK_READ 0x0B /* Read Block */
> +#define SMBUS_PROC_CALL 0x0C /* Process Call */
> +#define SMBUS_BLOCK_PROC_CALL 0x0D /* Write Block-Read Block Process Call */
> +#define SMBUS_I2C_READ_WRITE 0x0E /* I2C block Read-Write */
> +#define SMBUS_I2C_WRITE_READ 0x0F /* I2C block Write-Read */
> +
> +/* GPIO control commands */
> +#define EC_GPIO_INDEX_WRITE 0x10
> +#define EC_GPIO_STATUS_READ 0x11
> +#define EC_GPIO_STATUS_WRITE 0x12
> +#define EC_GPIO_DIR_READ 0x1D
> +#define EC_GPIO_DIR_WRITE 0x1E
> +
> +/* One Key Recovery commands */
> +#define EC_ONE_KEY_FLAG 0x9C
> +
> +/* ASG OEM commands */
> +#define EC_ASG_OEM 0xEA
> +#define EC_ASG_OEM_READ 0x00
> +#define EC_ASG_OEM_WRITE 0x01
> +#define EC_OEM_POWER_STATUS_VIN1 0X10
> +#define EC_OEM_POWER_STATUS_VIN2 0X11
> +#define EC_OEM_POWER_STATUS_BAT1 0X12
> +#define EC_OEM_POWER_STATUS_BAT2 0X13
> +
> +/* GPIO DEVICE ID */
> +#define EC_DID_ALTGPIO_0 0x10 /* 0x10 AltGpio0 User define gpio */
> +#define EC_DID_ALTGPIO_1 0x11 /* 0x11 AltGpio1 User define gpio */
> +#define EC_DID_ALTGPIO_2 0x12 /* 0x12 AltGpio2 User define gpio */
> +#define EC_DID_ALTGPIO_3 0x13 /* 0x13 AltGpio3 User define gpio */
> +#define EC_DID_ALTGPIO_4 0x14 /* 0x14 AltGpio4 User define gpio */
> +#define EC_DID_ALTGPIO_5 0x15 /* 0x15 AltGpio5 User define gpio */
> +#define EC_DID_ALTGPIO_6 0x16 /* 0x16 AltGpio6 User define gpio */
> +#define EC_DID_ALTGPIO_7 0x17 /* 0x17 AltGpio7 User define gpio */
> +
> +/* Lmsensor Chip Register */
> +#define NSLM96163_CHANNEL 0x02
> +
> +/* NS_LM96163 address 0x98 */
> +#define NSLM96163_ADDR 0x98
> +
> +/* LM96163 index(0x00) Local Temperature (Signed MSB) */
> +#define NSLM96163_LOC_TEMP 0x00
> +
> +#define F75387_REG_R_MANU_ID 0x5D
> +#define F75387_REG_R_CHIP_ID 0x5A
> +
> +#define LMF75387_MANU_ID_FINTEK 0x1934 //VENDOR ID
> +#define LMF75387_CHIP_ID_F75387 0x0410 //CHIPID
> +
> +/* Lmsensor Chip SMUBS Slave Address */
> +#define LMF75387_SMBUS_SLAVE_ADDRESS_5C 0x5c
> +#define LMF75387_SMBUS_SLAVE_ADDRESS_5A 0x5A
> +#define INA266_SMBUS_SLAVE_ADDRESS_8A 0x8A
> +
> +/* Temperature */
> +#define F75387_REG_R_TEMP0_MSB 0x14 /* 1 degree */
> +#define F75387_REG_R_TEMP0_LSB 0x1A /* 1/256 degree */
> +
> +#define F75387_REG_R_TEMP1_MSB 0x15 /* 1 degree */
> +#define F75387_REG_R_TEMP1_LSB 0x1B /* 1/256 degree */
> +
> +/* LOCAL Temperature */
> +#define F75387_REG_R_TEMP2_MSB 0x1C /* local temp., 1 degree */
> +#define F75387_REG_R_TEMP2_LSB 0x1D /* 1/256 degree */
> +
> +/* Voltage */
> +#define F75387_REG_R_V1 0x11 /* 8mV */
> +#define F75387_REG_R_V2 0x12 /* 8mV */
> +#define F75387_REG_R_V3 0x13 /* 8mV */
> +
> +/* HWMON registers */
> +#define INA266_REG_VOLTAGE 0x02 /* 1.25mV */
> +#define INA266_REG_POWER 0x03 /* 25mW */
> +#define INA266_REG_CURRENT 0x04 /* 1mA */
> +
> +struct HW_PIN_TBL {
> + uchar vbat[2];
> + uchar v5[2];
> + uchar v12[2];
> + uchar vcore[2];
> + uchar vdc[2];
> + uchar ec_current[2];
> + uchar power[2];
> +};
> +
> +struct Dynamic_Tab {
> + uchar DeviceID;
> + uchar HWPinNumber;
> +};
> +
> +struct EC_SMBOEM0 {
> + uchar HWPinNumber;
> +};
> +
> +struct EC_READ_HW_RAM {
> + unsigned int addr;
> + unsigned int data;
> +};
> +
> +struct EC_WRITE_HW_RAM {
> + unsigned int addr;
> + unsigned int data;
> +};
> +
> +struct EC_SMBUS_WORD_DATA {
> + unsigned char Channel;
> + unsigned char Address;
> + unsigned char Register;
> + unsigned short Value;
> +};
> +
> +struct EC_SMBUS_READ_BYTE {
> + unsigned char Channel;
> + unsigned char Address;
> + unsigned char Register;
> + unsigned char Data;
> +};
> +
> +struct EC_SMBUS_WRITE_BYTE {
> + unsigned char Channel;
> + unsigned char Address;
> + unsigned char Register;
> + unsigned char Data;
> +};
> +
> +struct pled_hw_pin_tbl {
> + uchar pled[6];
> +};
> +
> +struct adv_ec_platform_data {
> + char *bios_product_name;
> + int sub_dev_nb;
> + u32 sub_dev_mask;
> +
> + struct device *dev;
> + struct class *adv_ec_class;
> +
> + struct Dynamic_Tab *dym_tbl;
> +};
[...]
> +int read_ad_value(uchar hwpin, uchar multi);
> +int read_acpi_value(uchar addr, uchar *pvalue);
> +int write_hw_ram(uchar addr, uchar data);
> +int write_hwram_command(uchar data);
> +int smbus_read_word(struct EC_SMBUS_WORD_DATA *ptr_ec_smbus_word_data);
> +int smbus_read_byte(struct EC_SMBUS_READ_BYTE *ptr_ec_smbus_read_byte);
> +int write_acpi_value(uchar addr, uchar value);
> +int read_gpio_status(uchar PinNumber, uchar *pvalue);
> +int write_gpio_status(uchar PinNumber, uchar value);
> +int read_gpio_dir(uchar PinNumber, uchar *pvalue);
> +int write_gpio_dir(uchar PinNumber, uchar value);
> +int write_hw_extend_ram(uchar addr, uchar data);
> +int smbus_write_byte(struct EC_SMBUS_WRITE_BYTE *ptr_ec_smbus_write_byte);
> +int read_onekey_status(uchar addr, uchar *pdata);
> +int write_onekey_status(uchar addr);
> +int ec_oem_get_status(uchar addr, uchar *pdata);
> +int ec_oem_set_status(uchar addr, uchar pdata);
This lot shouldn't be in this driver.
I don't know what it is, but it has nothing to do with MFD.
I would consider something like drivers/platform for this.
--
Lee Jones [李琼斯]
Senior Technical Lead - Developer Services
Linaro.org │ Open source software for Arm SoCs
Follow Linaro: Facebook | Twitter | Blog
Powered by blists - more mailing lists