Linux Sound subsystem development
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Maciej Strozek" <mstrozek@opensource.cirrus.com>
Cc: linux-gpio@vger.kernel.org, linux-sound@vger.kernel.org
Subject: Re: [PATCH] pinctrl: cs42l43: Add GPIO controlled mic/spk mute LEDs
Date: Tue, 10 Nov 2026 14:40:52 +0000	[thread overview]
Message-ID: <sashiko-outbox-207202@kernel.org> (raw)
In-Reply-To: <20261110142058.354518-1-mstrozek@opensource.cirrus.com>

Thank you for your contribution! Sashiko AI review found no new issues
in this patch, but noticed 1 related pre-existing issue it does not
introduce:
- [High] pinctrl: cirrus: cs42l43: unbalanced fwnode put in
  cs42l43_pin_probe() [1]
--

commit 1aa7ff7b2b9b125f7c2e43385c49822080f93214
Author: Maciej Strozek <mstrozek@opensource.cirrus.com>

pinctrl: cs42l43: Add GPIO controlled mic/spk mute LEDs

This commit adds support for driving speaker and microphone mute LEDs via
cs42l43 GPIOs when described by ACPI properties.

> diff --git a/drivers/pinctrl/cirrus/pinctrl-cs42l43.c b/drivers/pinctrl/cirrus/pinctrl-cs42l43.c
> index a1c839b355578..44a9a759e0f65 100644
> --- a/drivers/pinctrl/cirrus/pinctrl-cs42l43.c
> +++ b/drivers/pinctrl/cirrus/pinctrl-cs42l43.c

[ ... ]

> @@ -598,6 +748,12 @@ static int cs42l43_pin_probe(struct platform_device *pdev)
>  		return dev_err_probe(priv->dev, ret,
>  				     "Failed to register gpiochip\n");
>  
> +#if IS_REACHABLE(CONFIG_LEDS_CLASS)
> +	ret = cs42l43_mute_leds_register(priv, dev_fwnode(cs42l43->dev));
> +	if (ret)
> +		return ret;
> +#endif
> +
>  	return 0;
>  }

[Severity: High]
This problem wasn't introduced by this patch, but does assigning fwnode via
device_set_node() in cs42l43_pin_probe() cause an unbalanced
fwnode_handle_put() [1] when the platform device is released?

Earlier in cs42l43_pin_probe(), fwnode is initialized from the parent MFD
device or looked up from the "pinctrl" child node, and then assigned to
priv->dev:

drivers/pinctrl/cirrus/pinctrl-cs42l43.c:cs42l43_pin_probe() {
	...
	if (is_of_node(fwnode)) {
		struct fwnode_handle *child;

		child = fwnode_get_named_child_node(fwnode, "pinctrl");
		if (child) {
			ret = devm_add_action_or_reset(&pdev->dev,
				cs42l43_fwnode_put, child);
			if (ret)
				return ret;

			if (!child->dev)
				child->dev = priv->dev;
			fwnode = child;
		}
	}

	priv->gpio_chip.fwnode = fwnode;

	device_set_node(priv->dev, fwnode);
	...
}

Unlike device_add_software_node(), device_set_node() does not acquire a
reference on fwnode, whereas platform_device_release() unconditionally drops a
reference on non-software nodes when the platform device is freed:

drivers/base/platform.c:platform_device_release() {
	...
	if (!is_software_node(dev_fwnode(dev)))
		fwnode_handle_put(pa->pdev.dev.fwnode);
	...
}

When the "pinctrl" child node is present, its single reference from
fwnode_get_named_child_node() is already released by cs42l43_fwnode_put() via
devres; otherwise fwnode belongs to cs42l43->dev with no extra reference
taken.  When the platform device is later unregistered,
platform_device_release() puts pdev->dev.fwnode an extra time, underflowing
the fwnode reference count and risking a use-after-free.

[1] https://sashiko.dev/#/bug/linux-a633bdf8-b5fe-4770-b14c-d3532525ae12

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261110142058.354518-1-mstrozek@opensource.cirrus.com?part=1

  reply	other threads:[~2026-11-10 14:40 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-10 14:20 [PATCH] pinctrl: cs42l43: Add GPIO controlled mic/spk mute LEDs Maciej Strozek
2026-11-10 14:40 ` sashiko-bot [this message]
2026-11-11 11:01 ` Charles Keepax

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=sashiko-outbox-207202@kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=linux-gpio@vger.kernel.org \
    --cc=linux-sound@vger.kernel.org \
    --cc=mstrozek@opensource.cirrus.com \
    --cc=sashiko-reviews@lists.linux.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox