Re: [RFC PATCH 2/3] mfd: rk8xx: release the power hold GPIOs before powering off

[Lee Jones <lee@kernel.org>]

On Wed, 09 Sep 2026, sashiko-bot@kernel.org wrote:

> Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
> - [Medium] Acquiring 'power-hold' GPIOs after registering MFD child devices can cause probe deferral thrashing.
> - [Low] The commit message description violates the MFD subsystem formatting rules by using a lowercase letter.

Please review, explain or fix the Sashiko review(s).

If this is a set, this request includes the other patches too.

> --
> 
> commit 99f9b4ec1bb40a89a22b218c05cb3242225a6bc9
> Author: Yaozhong Li <yaozhonguwl@gmail.com>
> 
> mfd: rk8xx: release the power hold GPIOs before powering off
> 
> This commit adds support for "power hold" GPIO lines to the rk8xx driver. It
> acquires these lines during probe and releases them with a delay before the
> PMIC shutdown request is written.
> 
> > mfd: rk8xx: release the power hold GPIOs before powering off
> 
> [Severity: Low]
> This isn't a functional bug, but the commit message subject violates the MFD
> subsystem formatting rules. The MFD guidelines require capitalizing the
> description after the subsystem prefix. Should "release" be updated to
> "Release"?
> 
> > diff --git a/drivers/mfd/rk8xx-core.c b/drivers/mfd/rk8xx-core.c
> > index 3dcf6abfda74f..fd1efbb94008d 100644
> > --- a/drivers/mfd/rk8xx-core.c
> > +++ b/drivers/mfd/rk8xx-core.c
> [ ... ]
> > @@ -874,6 +894,15 @@ int rk8xx_probe(struct device *dev, int variant, unsigned int irq, struct regmap
> >  
> >  	if (device_property_read_bool(dev, "system-power-controller") ||
> >  	    device_property_read_bool(dev, "rockchip,system-power-controller")) {
> > +		rk808->power_hold_gpios = devm_gpiod_get_array_optional(dev,
> > +							"power-hold", GPIOD_OUT_HIGH);
> > +		if (IS_ERR(rk808->power_hold_gpios))
> > +			return dev_err_probe(dev, PTR_ERR(rk808->power_hold_gpios),
> > +					     "failed to get power hold GPIOs\n");
> > +
> 
> [Severity: Medium]
> If the GPIO provider is not yet ready, devm_gpiod_get_array_optional() will
> return -EPROBE_DEFER. Because this acquisition occurs after
> devm_mfd_add_devices() has already run earlier in rk8xx_probe(), won't the
> resulting devres cleanup unregister all newly registered child devices (such
> as regulators, RTC, and clocks) while we wait for the GPIOs?
> 
> Could the GPIO acquisition be moved earlier in the probe function, before the
> child devices are registered, to avoid probe deferral thrashing?
> 
> -- 
> Sashiko AI review · https://sashiko.dev/#/patchset/20260909092728.1859-1-yaozhonguwl@gmail.com?part=2

-- 
Lee Jones