[PATCH v3 3/4] rtc: rk8xx: Add base support for the RK808 PMIC RTC

Kever Yang kever.yang at rock-chips.com
Wed Apr 29 04:39:28 CEST 2020


+Elaine,

Please help to review this patch.


Thanks,

- Kever

On 2020/4/29 上午12:14, sunil at amarulasolutions.com wrote:
> From: Suniel Mahesh <sunil at amarulasolutions.com>
>
> Rockchip RK808 PMIC provides an integrated RTC module. It is
> commonly used with Rockchip SoCs. Add basic support to access
> date and time.
>
> Signed-off-by: Suniel Mahesh <sunil at amarulasolutions.com>
> ---
> Changes for v3:
> - no changes
>
> Changes for v2:
> -  moved corresponding configs which enable rtc into a seperate
>     patch.
> -  changed subject line.
>
> Note:
> 1. The RK808 PMIC RTC has a hardware bug. It counts 31 days
> for november month and the weeks register counts 0 - 7.
>
> 2. This driver does a temporary fix, where as in if date is Nov 31,
> then it resets the date to Dec 1(this happens only if date cmd is queried
> from u-boot command line/script). Similarly for the weeks register, 0(sun)
> - 6(sat). If 7 is encountered then it is reset to zero.
>
> 3. u-boot generally loads linux/other binary. Linux has a full fledged
> driver implemented along with a workaround.
> https://lkml.org/lkml/2015/12/2/1202
>
> 4. Is this change acceptable ? please comment
> ---
>   drivers/rtc/Kconfig     |   8 +++
>   drivers/rtc/Makefile    |   1 +
>   drivers/rtc/rk808-rtc.c | 165 ++++++++++++++++++++++++++++++++++++++++++++++++
>   3 files changed, 174 insertions(+)
>   create mode 100644 drivers/rtc/rk808-rtc.c
>
> diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig
> index 59e2fc4..a754d1b 100644
> --- a/drivers/rtc/Kconfig
> +++ b/drivers/rtc/Kconfig
> @@ -75,6 +75,14 @@ config RTC_ISL1208
>   	  This driver supports reading and writing the RTC/calendar and detects
>   	  total power failures.
>   
> +config RTC_RK808
> +	bool "Enable Rockchip RK8XX RTC driver"
> +        depends on DM_RTC && PMIC_RK8XX
> +	default y
> +	help
> +	  Basic support for Rockchip RK808 PMIC Real Time Clock devices for
> +	  time and date.
> +
>   config RTC_RV3029
>   	bool "Enable RV3029 driver"
>   	depends on DM_RTC
> diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile
> index 12eb449..63e2c34 100644
> --- a/drivers/rtc/Makefile
> +++ b/drivers/rtc/Makefile
> @@ -44,6 +44,7 @@ obj-$(CONFIG_RTC_PCF8563) += pcf8563.o
>   obj-$(CONFIG_RTC_PCF2127) += pcf2127.o
>   obj-$(CONFIG_RTC_PL031) += pl031.o
>   obj-$(CONFIG_RTC_PT7C4338) += pt7c4338.o
> +obj-$(CONFIG_RTC_RK808) += rk808-rtc.o
>   obj-$(CONFIG_RTC_RS5C372A) += rs5c372.o
>   obj-$(CONFIG_RTC_RV3029) += rv3029.o
>   obj-$(CONFIG_RTC_RV8803) += rv8803.o
> diff --git a/drivers/rtc/rk808-rtc.c b/drivers/rtc/rk808-rtc.c
> new file mode 100644
> index 0000000..b63cced
> --- /dev/null
> +++ b/drivers/rtc/rk808-rtc.c
> @@ -0,0 +1,165 @@
> +// SPDX-License-Identifier: GPL-2.0+
> +/*
> + * RTC driver for Rockchip RK808 PMIC.
> + *
> + * Copyright (C) 2020 Amarula Solutions(India).
> + * Suniel Mahesh <sunil at amarulasolutions.com>
> + *
> + * Based on code from Linux kernel:
> + * Copyright (c) 2014, Fuzhou Rockchip Electronics Co., Ltd
> + * Author: Chris Zhong <zyw at rock-chips.com>
> + * Author: Zhang Qing <zhangqing at rock-chips.com>
> + *
> + * Date & Time support (no alarms and interrupts)
> + */
> +
> +#include <command.h>
> +#include <common.h>
> +#include <dm.h>
> +#include <i2c.h>
> +#include <rtc.h>
> +#include <power/rk8xx_pmic.h>
> +#include <power/pmic.h>
> +
> +/* RTC_CTRL_REG bitfields */
> +#define BIT_RTC_CTRL_REG_STOP_RTC_M		BIT(0)
> +
> +/* RK808 has a shadowed register for saving a "frozen" RTC time.
> + * When user setting "GET_TIME" to 1, the time will save in this shadowed
> + * register. If set "READSEL" to 1, user read rtc time register, actually
> + * get the time of that moment. If we need the real time, clr this bit.
> + */
> +
> +#define BIT_RTC_CTRL_REG_RTC_GET_TIME		BIT(6)
> +#define BIT_RTC_CTRL_REG_RTC_READSEL_M		BIT(7)
> +#define RTC_STATUS_MASK				0xFE
> +
> +#define SECONDS_REG_MSK				0x7F
> +#define MINUTES_REG_MAK				0x7F
> +#define HOURS_REG_MSK				0x3F
> +#define DAYS_REG_MSK				0x3F
> +#define MONTHS_REG_MSK				0x1F
> +#define YEARS_REG_MSK				0xFF
> +#define WEEKS_REG_MSK				0x7
> +
> +/* REG_SECONDS_REG through REG_YEARS_REG is how many registers? */
> +
> +#define NUM_TIME_REGS	(REG_WEEKS - REG_SECONDS + 1)
> +
> +static int rk808_rtc_set(struct udevice *dev, const struct rtc_time *tm)
> +{
> +	u8 rtc_data[NUM_TIME_REGS];
> +
> +	debug("RTC date/time %4d-%02d-%02d(%d) %02d:%02d:%02d\n",
> +			tm->tm_year, tm->tm_mon, tm->tm_mday,
> +			tm->tm_wday, tm->tm_hour, tm->tm_min, tm->tm_sec);
> +
> +	rtc_data[0] = bin2bcd(tm->tm_sec);
> +	rtc_data[1] = bin2bcd(tm->tm_min);
> +	rtc_data[2] = bin2bcd(tm->tm_hour);
> +	rtc_data[3] = bin2bcd(tm->tm_mday);
> +	rtc_data[4] = bin2bcd(tm->tm_mon);
> +	rtc_data[5] = bin2bcd(tm->tm_year - 2000);
> +	rtc_data[6] = bin2bcd(tm->tm_wday);
> +
> +/* Stop RTC while updating the RTC registers */
> +	pmic_clrsetbits(dev->parent, REG_RTC_CTRL, 0,
> +						BIT_RTC_CTRL_REG_STOP_RTC_M);
> +	pmic_write(dev->parent, REG_SECONDS, rtc_data, NUM_TIME_REGS);
> +
> +/* Start RTC again */
> +	pmic_clrsetbits(dev->parent, REG_RTC_CTRL,
> +						BIT_RTC_CTRL_REG_STOP_RTC_M, 0);
> +	return 0;
> +}
> +
> +static int rk808_rtc_get(struct udevice *dev, struct rtc_time *tm)
> +{
> +	u8 rtc_data[NUM_TIME_REGS];
> +
> +/* Force an update of the shadowed registers right now */
> +	pmic_clrsetbits(dev->parent, REG_RTC_CTRL, 0,
> +						BIT_RTC_CTRL_REG_RTC_GET_TIME);
> +
> +/*
> + * After we set the GET_TIME bit, the rtc time can't be read
> + * immediately. So we should wait up to 31.25 us, about one cycle of
> + * 32khz. If we clear the GET_TIME bit here, the time of i2c transfer
> + * certainly more than 31.25us: 16 * 2.5us at 400kHz bus frequency.
> + */
> +	pmic_clrsetbits(dev->parent, REG_RTC_CTRL,
> +					BIT_RTC_CTRL_REG_RTC_GET_TIME, 0);
> +	pmic_read(dev->parent, REG_SECONDS, rtc_data, NUM_TIME_REGS);
> +
> +	tm->tm_sec = bcd2bin(rtc_data[0] & SECONDS_REG_MSK);
> +	tm->tm_min = bcd2bin(rtc_data[1] & MINUTES_REG_MAK);
> +	tm->tm_hour = bcd2bin(rtc_data[2] & HOURS_REG_MSK);
> +	tm->tm_mday = bcd2bin(rtc_data[3] & DAYS_REG_MSK);
> +	tm->tm_mon = (bcd2bin(rtc_data[4] & MONTHS_REG_MSK));
> +	tm->tm_year = (bcd2bin(rtc_data[5] & YEARS_REG_MSK)) + 2000;
> +	tm->tm_wday = bcd2bin(rtc_data[6] & WEEKS_REG_MSK);
> +/*
> + * RK808 PMIC RTC h/w counts/has 31 days in november. This is corrected
> + * when date cmd is invoked on prompt. checks for the current day and
> + * if it is 31 November, then adjusts it to 1 December.
> + *
> + * h/w also has weeks register which counts from 0 to 7(0(sun)-6(sat)).
> + * 7 is an unknown state, reset it back to 0(sun).
> + */
> +	if (tm->tm_mon == 11 && tm->tm_mday == 31) {
> +		debug("correcting Nov 31st to Dec 1st (HW bug)\n");
> +		tm->tm_mon += 1;
> +		tm->tm_mday = 1;
> +		if (tm->tm_wday == 7)
> +			tm->tm_wday = 0;
> +		rk808_rtc_set(dev, tm);
> +	}
> +
> +	if (tm->tm_wday == 7) {
> +		tm->tm_wday = 0;
> +		rk808_rtc_set(dev, tm);
> +	}
> +
> +	debug("RTC date/time %4d-%02d-%02d(%d) %02d:%02d:%02d\n",
> +			tm->tm_year, tm->tm_mon, tm->tm_mday,
> +			tm->tm_wday, tm->tm_hour, tm->tm_min, tm->tm_sec);
> +	return 0;
> +}
> +
> +static int rk808_rtc_reset(struct udevice *dev)
> +{
> +/* Not needed */
> +	return 0;
> +}
> +
> +static int rk808_rtc_init(struct udevice *dev)
> +{
> +	struct rtc_time tm;
> +
> +/* start rtc running by default, and use shadowed timer. */
> +	pmic_clrsetbits(dev->parent, REG_RTC_CTRL, 0,
> +					BIT_RTC_CTRL_REG_RTC_READSEL_M);
> +	pmic_reg_write(dev->parent, REG_RTC_STATUS, RTC_STATUS_MASK);
> +/* set init time */
> +	rk808_rtc_get(dev, &tm);
> +	return 0;
> +}
> +
> +static int rk808_rtc_probe(struct udevice *dev)
> +{
> +	rk808_rtc_init(dev);
> +	return 0;
> +}
> +
> +static const struct rtc_ops rk808_rtc_ops = {
> +	.get = rk808_rtc_get,
> +	.set = rk808_rtc_set,
> +	.reset = rk808_rtc_reset,
> +};
> +
> +U_BOOT_DRIVER(rk808_rtc) = {
> +	.name     = "rk808_rtc",
> +	.id       = UCLASS_RTC,
> +	.ops      = &rk808_rtc_ops,
> +	.probe    = rk808_rtc_probe,
> +};




More information about the U-Boot mailing list