On 9/5/19 3:00 PM, Marek Vasut wrote:
> On 8/30/19 12:42 PM, Bastian Krause wrote:
>> On 6/28/19 2:21 AM, Marek Vasut wrote:
>>> The battery backup can be disabled on this RTC, e.g. if populated right
>>> out of production. Force the battery backup bit on to enable it.
>>>
>>> Signed-off-by: Marek Vasut <marex@xxxxxxx>
>>> Cc: Arnaud Ebalard <arno@xxxxxxxxxxxx>
>>> Cc: Alexandre Belloni <alexandre.belloni@xxxxxxxxxxxxxxxxxx>
>>> ---
>>> drivers/rtc/rtc-ds1307.c | 7 +++++++
>>> 1 file changed, 7 insertions(+)
>>>
>>> diff --git a/drivers/rtc/rtc-ds1307.c b/drivers/rtc/rtc-ds1307.c
>>> index 1f7e8aefc1eb..4af00cac1eff 100644
>>> --- a/drivers/rtc/rtc-ds1307.c
>>> +++ b/drivers/rtc/rtc-ds1307.c
>>> @@ -121,6 +121,8 @@ enum ds_type {
>>> #define RX8130_REG_FLAG_AF BIT(3)
>>> #define RX8130_REG_CONTROL0 0x1e
>>> #define RX8130_REG_CONTROL0_AIE BIT(3)
>>> +#define RX8130_REG_CONTROL1 0x0f
>>
>> I think it is surprising that RX8130_REG_CONTROL0 is defined absolute
>> and RX8130_REG_CONTROL1 must be used with chip offset. Having this
>> defined consistently would be nice.
>
> It's because this old patch was rebased from before the regmap changes
> were even in.
>
>>> +#define RX8130_REG_CONTROL1_INIEN BIT(4)
>>>
>>> #define MCP794XX_REG_CONTROL 0x07
>>> # define MCP794XX_BIT_ALM0_EN 0x10
>>> @@ -1750,6 +1752,11 @@ static int ds1307_probe(struct i2c_client *client,
>>> DS1307_REG_HOUR << 4 | 0x08, hour);
>>> }
>>> break;
>>> + case rx_8130:
>>> + /* make sure that the backup battery is enabled */
>>> + regmap_write(ds1307->regmap, chip->offset + RX8130_REG_CONTROL1,
>>> + RX8130_REG_CONTROL1_INIEN);
>>> + break;
>>
>> Quick note: if a rechargeable battery or supercap is used the CHGEN bit
>> must be set also.
>
> By rechargable battery, you mean coin cell one ?
Yes, some rechargeable lithium coin/button cell. Or in my case a ELNA
0.2F 3.3V supercap.
Regards,
Bastian
--
Pengutronix e.K.
Industrial Linux Solutions
http://www.pengutronix.de/
Peiner Str. 6-8, 31137 Hildesheim, Germany
Amtsgericht Hildesheim, HRA 2686
