2020-05-14 20:55:01

by Dmitry Osipenko

Subject: [PATCH v1] mfd: tps65910: Correct power-off programming sequence

This patch fixes system shutdown on a devices that use TPS65910 as a
system's power controller. In accordance to the TPS65910 datasheet, the
PMIC's state-machine transitions into the OFF state only when DEV_OFF
bit of DEVCTRL_REG is set. The ON / SLEEP states also should be cleared,
otherwise PMIC won't get into a proper state on shutdown. Devices like
Nexus 7 tablet and Ouya game console are now shutting down properly.

Tested-by: Zack Pearsall <[email protected]>
Signed-off-by: Dmitry Osipenko <[email protected]>
---
drivers/mfd/tps65910.c | 12 ++++++++++--
1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/mfd/tps65910.c b/drivers/mfd/tps65910.c
index 11959021b50a..22116cee411d 100644
--- a/drivers/mfd/tps65910.c
+++ b/drivers/mfd/tps65910.c
@@ -440,8 +440,16 @@ static void tps65910_power_off(void)
DEVCTRL_PWR_OFF_MASK) < 0)
return;

- tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
- DEVCTRL_DEV_ON_MASK);
+ if (tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
+ DEVCTRL_DEV_SLP_MASK) < 0)
+ return;
+
+ if (tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
+ DEVCTRL_DEV_ON_MASK) < 0)
+ return;
+
+ tps65910_reg_set_bits(tps65910, TPS65910_DEVCTRL,
+ DEVCTRL_DEV_OFF_MASK);
}

static int tps65910_i2c_probe(struct i2c_client *i2c,
--
2.26.0

2020-05-15 13:33:37

by Michał Mirosław

Subject: Re: [PATCH v1] mfd: tps65910: Correct power-off programming sequence

On Thu, May 14, 2020 at 11:50:21PM +0300, Dmitry Osipenko wrote:
> This patch fixes system shutdown on a devices that use TPS65910 as a
> system's power controller. In accordance to the TPS65910 datasheet, the
> PMIC's state-machine transitions into the OFF state only when DEV_OFF
> bit of DEVCTRL_REG is set. The ON / SLEEP states also should be cleared,
> otherwise PMIC won't get into a proper state on shutdown. Devices like
> Nexus 7 tablet and Ouya game console are now shutting down properly.
>
> Tested-by: Zack Pearsall <[email protected]>
> Signed-off-by: Dmitry Osipenko <[email protected]>
> ---
> drivers/mfd/tps65910.c | 12 ++++++++++--
> 1 file changed, 10 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/mfd/tps65910.c b/drivers/mfd/tps65910.c
> index 11959021b50a..22116cee411d 100644
> --- a/drivers/mfd/tps65910.c
> +++ b/drivers/mfd/tps65910.c
> @@ -440,8 +440,16 @@ static void tps65910_power_off(void)
> DEVCTRL_PWR_OFF_MASK) < 0)
> return;
>
> - tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
> - DEVCTRL_DEV_ON_MASK);
> + if (tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
> + DEVCTRL_DEV_SLP_MASK) < 0)
> + return;
> +
> + if (tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
> + DEVCTRL_DEV_ON_MASK) < 0)
> + return;
> +
> + tps65910_reg_set_bits(tps65910, TPS65910_DEVCTRL,
> + DEVCTRL_DEV_OFF_MASK);
> }

Isn't it enough to update the DEVCTRL with just one
tps65910_reg_update_bits()?

Best Regards
Micha? Miros?aw

2020-05-15 17:32:48

by Dmitry Osipenko

Subject: Re: [PATCH v1] mfd: tps65910: Correct power-off programming sequence

15.05.2020 16:30, Michał Mirosław пишет:
> On Thu, May 14, 2020 at 11:50:21PM +0300, Dmitry Osipenko wrote:
>> This patch fixes system shutdown on a devices that use TPS65910 as a
>> system's power controller. In accordance to the TPS65910 datasheet, the
>> PMIC's state-machine transitions into the OFF state only when DEV_OFF
>> bit of DEVCTRL_REG is set. The ON / SLEEP states also should be cleared,
>> otherwise PMIC won't get into a proper state on shutdown. Devices like
>> Nexus 7 tablet and Ouya game console are now shutting down properly.
>>
>> Tested-by: Zack Pearsall <[email protected]>
>> Signed-off-by: Dmitry Osipenko <[email protected]>
>> ---
>> drivers/mfd/tps65910.c | 12 ++++++++++--
>> 1 file changed, 10 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/mfd/tps65910.c b/drivers/mfd/tps65910.c
>> index 11959021b50a..22116cee411d 100644
>> --- a/drivers/mfd/tps65910.c
>> +++ b/drivers/mfd/tps65910.c
>> @@ -440,8 +440,16 @@ static void tps65910_power_off(void)
>> DEVCTRL_PWR_OFF_MASK) < 0)
>> return;
>>
>> - tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
>> - DEVCTRL_DEV_ON_MASK);
>> + if (tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
>> + DEVCTRL_DEV_SLP_MASK) < 0)
>> + return;
>> +
>> + if (tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
>> + DEVCTRL_DEV_ON_MASK) < 0)
>> + return;
>> +
>> + tps65910_reg_set_bits(tps65910, TPS65910_DEVCTRL,
>> + DEVCTRL_DEV_OFF_MASK);
>> }
>
> Isn't it enough to update the DEVCTRL with just one
> tps65910_reg_update_bits()?

IIRC, we've tried that variant and it didn't work. Although, maybe it
was affected by some other changes that we were trying simultaneously,
so could be worthwhile to re-check it. Thank you for the comment!

2020-05-24 19:11:18

by Dmitry Osipenko

Subject: Re: [PATCH v1] mfd: tps65910: Correct power-off programming sequence

15.05.2020 20:30, Dmitry Osipenko пишет:
> 15.05.2020 16:30, Michał Mirosław пишет:
>> On Thu, May 14, 2020 at 11:50:21PM +0300, Dmitry Osipenko wrote:
>>> This patch fixes system shutdown on a devices that use TPS65910 as a
>>> system's power controller. In accordance to the TPS65910 datasheet, the
>>> PMIC's state-machine transitions into the OFF state only when DEV_OFF
>>> bit of DEVCTRL_REG is set. The ON / SLEEP states also should be cleared,
>>> otherwise PMIC won't get into a proper state on shutdown. Devices like
>>> Nexus 7 tablet and Ouya game console are now shutting down properly.
>>>
>>> Tested-by: Zack Pearsall <[email protected]>
>>> Signed-off-by: Dmitry Osipenko <[email protected]>
>>> ---
>>> drivers/mfd/tps65910.c | 12 ++++++++++--
>>> 1 file changed, 10 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/mfd/tps65910.c b/drivers/mfd/tps65910.c
>>> index 11959021b50a..22116cee411d 100644
>>> --- a/drivers/mfd/tps65910.c
>>> +++ b/drivers/mfd/tps65910.c
>>> @@ -440,8 +440,16 @@ static void tps65910_power_off(void)
>>> DEVCTRL_PWR_OFF_MASK) < 0)
>>> return;
>>>
>>> - tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
>>> - DEVCTRL_DEV_ON_MASK);
>>> + if (tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
>>> + DEVCTRL_DEV_SLP_MASK) < 0)
>>> + return;
>>> +
>>> + if (tps65910_reg_clear_bits(tps65910, TPS65910_DEVCTRL,
>>> + DEVCTRL_DEV_ON_MASK) < 0)
>>> + return;
>>> +
>>> + tps65910_reg_set_bits(tps65910, TPS65910_DEVCTRL,
>>> + DEVCTRL_DEV_OFF_MASK);
>>> }
>>
>> Isn't it enough to update the DEVCTRL with just one
>> tps65910_reg_update_bits()?
>
> IIRC, we've tried that variant and it didn't work. Although, maybe it
> was affected by some other changes that we were trying simultaneously,
> so could be worthwhile to re-check it. Thank you for the comment!
>

We tried it once again and the variant with a single update_bits() for
all bits doesn't work. The single update_bits() for the ON/OFF_MASK mask
works, it's actually the same what downstream driver does [1].

[1]
https://android.googlesource.com/kernel/tegra/+/refs/heads/android-tegra3-grouper-3.1-lollipop-release/drivers/mfd/tps6591x.c#374

I'll prepare a v2.