From: sashiko-bot@kernel.org
To: "Peng Fan (OSS)" <peng.fan@oss.nxp.com>
Cc: linux-gpio@vger.kernel.org, Frank.Li@kernel.org, imx@lists.linux.dev
Subject: Re: [PATCH v6 10/13] gpio: mxc: use devres functions simplify probe code
Date: Sun, 11 Oct 2026 12:05:31 +0000 [thread overview]
Message-ID: <sashiko-outbox-167212@kernel.org> (raw)
In-Reply-To: <20261011-gpio-mxc-cleanup-v6-10-cb47ce2ad758@nxp.com>
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] gpio: use-after-free and premature exposure of irq_domain
--
commit e70e13ad3287f8fd0b2357d428d23270ccfc9522
Author: Peng Fan <peng.fan@nxp.com>
gpio: mxc: use devres functions simplify probe code
This commit replaces legacy IRQ domain creation and manual PM runtime error
handling with devres equivalents in the MXC GPIO driver probe. It removes
error path labels by relying on device-managed allocations to unwind resources.
> diff --git a/drivers/gpio/gpio-mxc.c b/drivers/gpio/gpio-mxc.c
> index 8dcfd05e658a6..09acb186c84c1 100644
> --- a/drivers/gpio/gpio-mxc.c
> +++ b/drivers/gpio/gpio-mxc.c
[ ... ]
> @@ -534,27 +536,31 @@ static int mxc_gpio_probe(struct platform_device *pdev)
>
> err = devm_gpiochip_add_data(dev, &port->gen_gc.gc, port);
> if (err)
> - goto out_bgio;
> + return err;
[Severity: High]
Does this premature exposure of the GPIO chip cause a use-after-free risk?
By calling devm_gpiochip_add_data() before devm_irq_domain_instantiate(), the
GPIO chip is registered and immediately exposed to kernel consumers and
userspace.
If a consumer requests an IRQ during this window, mxc_gpio_to_irq() will
execute:
drivers/gpio/gpio-mxc.c:mxc_gpio_to_irq() {
struct mxc_gpio_port *port = gpiochip_get_data(gc);
return irq_find_mapping(port->domain, offset);
}
This passes a NULL port->domain to irq_find_mapping(), failing to map the IRQ.
Additionally, if a subsequent probe step like mxc_gpio_init_gc() fails, devres
unwinds allocations in reverse order (LIFO). Devres will first destroy the IRQ
domain, leaving port->domain as a dangling pointer to freed memory.
The GPIO chip remains fully registered and exposed until its own devres release
action runs later. A concurrent gpiod_to_irq() call during this unwinding
window will dereference the freed port->domain, causing a use-after-free.
Should devm_irq_domain_instantiate() be called before
devm_gpiochip_add_data() to ensure the domain is fully prepared before exposure
and destroyed after the chip is unregistered?
>
> irq_base = devm_irq_alloc_descs(dev, -1, 0, 32, numa_node_id());
> - if (irq_base < 0) {
> - err = irq_base;
> - goto out_bgio;
> - }
> + if (irq_base < 0)
> + return irq_base;
> +
> + d_info = (struct irq_domain_info) {
> + .fwnode = dev_fwnode(dev),
> + .size = 32,
> + .hwirq_max = 32,
> + .virq_base = irq_base,
> + .ops = &irq_domain_simple_ops,
> + .dev = dev,
> + };
>
> - port->domain = irq_domain_create_legacy(dev_fwnode(dev), 32, irq_base, 0,
> - &irq_domain_simple_ops, NULL);
> - if (!port->domain) {
> - err = -ENODEV;
> - goto out_bgio;
> - }
> + port->domain = devm_irq_domain_instantiate(dev, &d_info);
> + if (IS_ERR(port->domain))
> + return PTR_ERR(port->domain);
>
> irq_domain_set_pm_device(port->domain, dev);
>
> /* gpio-mxc can be a generic irq chip */
> err = mxc_gpio_init_gc(port, irq_base);
> if (err < 0)
> - goto out_irqdomain_remove;
> + return err;
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261011-gpio-mxc-cleanup-v6-0-cb47ce2ad758@nxp.com?part=10
next prev parent reply other threads:[~2026-10-11 12:05 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-11 11:51 [PATCH v6 00/13] gpio: mxc: bug fixes and probe cleanup Peng Fan (OSS)
2026-10-11 11:51 ` [PATCH v6 01/13] gpio: mxc: fix race between chained IRQ handler install and probe completion Peng Fan (OSS)
2026-10-11 12:05 ` sashiko-bot
2026-10-12 4:14 ` Peng Fan
2026-10-11 11:51 ` [PATCH v6 02/13] gpio: mxc: fix both_edges bit operations Peng Fan (OSS)
2026-10-12 9:18 ` Andy Shevchenko
2026-10-12 9:52 ` Peng Fan
2026-10-11 11:51 ` [PATCH v6 03/13] gpio: mxc: fix wakeup_pads " Peng Fan (OSS)
2026-10-12 9:15 ` Andy Shevchenko
2026-10-12 9:51 ` Peng Fan
2026-10-12 10:30 ` Andy Shevchenko
2026-10-12 11:41 ` Peng Fan
2026-10-12 18:25 ` Andy Shevchenko
2026-10-11 11:51 ` [PATCH v6 04/13] gpio: mxc: use irq_set_irq_wake() to simplify code Peng Fan (OSS)
2026-10-11 11:51 ` [PATCH v6 05/13] gpio: mxc: use for_each_set_bit() to iterate wakeup pads Peng Fan (OSS)
2026-10-12 9:19 ` Andy Shevchenko
2026-10-12 9:53 ` Peng Fan
2026-10-11 11:51 ` [PATCH v6 06/13] gpio: mxc: replace of_device_is_compatible() with hwdata flags Peng Fan (OSS)
2026-10-12 9:25 ` Andy Shevchenko
2026-10-12 9:59 ` Peng Fan
2026-10-12 10:32 ` Andy Shevchenko
2026-10-11 11:51 ` [PATCH v6 07/13] gpio: mxc: convert pad wakeup compatible checks to " Peng Fan (OSS)
2026-10-12 9:27 ` Andy Shevchenko
2026-10-12 10:01 ` Peng Fan
2026-10-12 14:31 ` Frank Li
2026-10-11 11:51 ` [PATCH v6 08/13] gpio: mxc: use local dev variable Peng Fan (OSS)
2026-10-12 9:30 ` Andy Shevchenko
2026-10-12 10:05 ` Peng Fan
2026-10-12 10:39 ` Andy Shevchenko
2026-10-12 11:52 ` Peng Fan
2026-10-11 11:51 ` [PATCH v6 09/13] gpio: mxc: use cleanup guard for pm_runtime_get_noresume() balance Peng Fan (OSS)
2026-10-12 9:38 ` Andy Shevchenko
2026-10-12 10:10 ` Peng Fan
2026-10-12 10:41 ` Andy Shevchenko
2026-10-11 11:51 ` [PATCH v6 10/13] gpio: mxc: use devres functions simplify probe code Peng Fan (OSS)
2026-10-11 12:05 ` sashiko-bot [this message]
2026-10-12 4:12 ` Peng Fan
2026-10-11 11:51 ` [PATCH v6 11/13] gpio: mxc: add dev_err_probe() for error paths to help diagnostics Peng Fan (OSS)
2026-10-11 12:04 ` sashiko-bot
2026-10-12 4:13 ` Peng Fan
2026-10-11 11:51 ` [PATCH v6 12/13] gpio: mxc: introduce MXC_ICR macros and use field_prep/field_get Peng Fan (OSS)
2026-10-11 11:51 ` [PATCH v6 13/13] gpio: mxc: use BIT() macro for single-bit operations Peng Fan (OSS)
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-167212@kernel.org \
--to=sashiko-bot@kernel.org \
--cc=Frank.Li@kernel.org \
--cc=imx@lists.linux.dev \
--cc=linux-gpio@vger.kernel.org \
--cc=peng.fan@oss.nxp.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