Re: [PATCH 3/3] mfd: sec-core: Disable buck voltage reset on watchdog falling edge
From: Bartlomiej Zolnierkiewicz
Date: Mon Oct 05 2015 - 09:37:01 EST
Hi,
On Friday, October 02, 2015 05:25:51 PM Krzysztof Kozlowski wrote:
> The WRSTBI bit (disabled by default but enabled by bootloader), when
> set, is responsible for resetting voltages to default values of
> certain bucks on falling edge of Warm Reset Input pin from AP.
>
> However on some boards (with S2MPS13) the pin is pulled down so any
> suspend will effectively trigger the reset of bucks supplying the power
> to the little and big cores. In the same time when resuming, these bucks
> must provide voltage greater or equal to voltage before suspend to match
> the frequency chosen by cpufreq. If voltage (default value of voltage
> after reset) is lower than one set by cpufreq before suspend, then
> system will hang during resuming.
>
> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@xxxxxxxxxxx>
I tested this (with some additional Exynos5433 patches) and it works
like expected (suspend/resume issue is fixed).
Tested-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@xxxxxxxxxxx>
You could also add:
Reported-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@xxxxxxxxxxx>
(since the issue got narrowed down and reported to you by me :).
Best regards,
--
Bartlomiej Zolnierkiewicz
Samsung R&D Institute Poland
Samsung Electronics
> ---
> drivers/mfd/sec-core.c | 26 ++++++++++++++++++++++++++
> include/linux/mfd/samsung/core.h | 2 ++
> include/linux/mfd/samsung/s2mps13.h | 1 +
> 3 files changed, 29 insertions(+)
>
> diff --git a/drivers/mfd/sec-core.c b/drivers/mfd/sec-core.c
> index 589e5efc2d7f..2626fc0b5b8c 100644
> --- a/drivers/mfd/sec-core.c
> +++ b/drivers/mfd/sec-core.c
> @@ -262,6 +262,29 @@ static void sec_pmic_dump_rev(struct sec_pmic_dev *sec_pmic)
> dev_dbg(sec_pmic->dev, "Revision: 0x%x\n", val);
> }
>
> +static void sec_pmic_configure(struct sec_pmic_dev *sec_pmic)
> +{
> + int err;
> +
> + if (sec_pmic->device_type != S2MPS13X)
> + return;
> +
> + if (sec_pmic->pdata->disable_wrstbi) {
> + /*
> + * If WRSTBI pin is pulled down this feature must be disabled
> + * because each Suspend to RAM will trigger buck voltage reset
> + * to default values.
> + */
> + err = regmap_update_bits(sec_pmic->regmap_pmic,
> + S2MPS13_REG_WRSTBI,
> + S2MPS13_REG_WRSTBI_MASK, 0x0);
> + if (err)
> + dev_warn(sec_pmic->dev,
> + "Cannot initialize WRSTBI config: %d\n",
> + err);
> + }
> +}
> +
> #ifdef CONFIG_OF
> /*
> * Only the common platform data elements for s5m8767 are parsed here from the
> @@ -289,6 +312,8 @@ static struct sec_platform_data *sec_pmic_i2c_parse_dt_pdata(
>
> pd->manual_poweroff = of_property_read_bool(dev->of_node,
> "samsung,s2mps11-acokb-ground");
> + pd->disable_wrstbi = of_property_read_bool(dev->of_node,
> + "samsung,s2mps11-wrstbi-ground");
> return pd;
> }
> #else
> @@ -434,6 +459,7 @@ static int sec_pmic_probe(struct i2c_client *i2c,
> goto err_mfd;
>
> device_init_wakeup(sec_pmic->dev, sec_pmic->wakeup);
> + sec_pmic_configure(sec_pmic);
> sec_pmic_dump_rev(sec_pmic);
>
> return ret;
> diff --git a/include/linux/mfd/samsung/core.h b/include/linux/mfd/samsung/core.h
> index aa78957e092f..a06098639399 100644
> --- a/include/linux/mfd/samsung/core.h
> +++ b/include/linux/mfd/samsung/core.h
> @@ -134,6 +134,8 @@ struct sec_platform_data {
> int buck4_init;
> /* Whether or not manually set PWRHOLD to low during shutdown. */
> bool manual_poweroff;
> + /* Disable the WRSTBI (buck voltage warm reset) when probing? */
> + bool disable_wrstbi;
> };
>
> /**
> diff --git a/include/linux/mfd/samsung/s2mps13.h b/include/linux/mfd/samsung/s2mps13.h
> index b1fd675fa36f..239e977ba45d 100644
> --- a/include/linux/mfd/samsung/s2mps13.h
> +++ b/include/linux/mfd/samsung/s2mps13.h
> @@ -184,5 +184,6 @@ enum s2mps13_regulators {
> * Let's assume that default value will be set.
> */
> #define S2MPS13_BUCK_RAMP_DELAY 12500
> +#define S2MPS13_REG_WRSTBI_MASK BIT(5)
>
> #endif /* __LINUX_MFD_S2MPS13_H */
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/