From: Martin Kaiser <martin@kaiser.cx>
To: Weigang He <geoffreyhe2@gmail.com>
Cc: Shawn Guo <shawnguo@kernel.org>,
Sascha Hauer <s.hauer@pengutronix.de>,
Pengutronix Kernel Team <kernel@pengutronix.de>,
Fabio Estevam <festevam@gmail.com>,
imx@lists.linux.dev, linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] ARM: imx: avic: fix device_node refcount leaks in mxc_init_irq()
Date: Wed, 10 Jun 2026 11:42:27 +0200 [thread overview]
Message-ID: <aikxg5xtXWccXPO8@nb282.user.codasip.com> (raw)
In-Reply-To: <20260610053115.2263570-1-geoffreyhe2@gmail.com>
Hi,
Thus wrote Weigang He (geoffreyhe2@gmail.com):
> mxc_init_irq() obtains two device_node references via
> of_find_compatible_node() and never releases either one:
> - The "fsl,imx25-ccm" node (looked up to map the CCM low-power
> interrupt mask registers on i.MX25) is stored in np, used by
> of_iomap(), and then the same np variable is overwritten by the
> second of_find_compatible_node() call without an of_node_put().
> On i.MX25 this leaks the node reference on every boot.
> - The "fsl,avic" node is passed via of_fwnode_handle(np) to
> irq_domain_create_legacy(), which takes its own reference on the
> fwnode through fwnode_handle_get(), so the caller's reference is
> not transferred. np is then leaked at function return.
> Both lookups predate the switch to irq_domain_create_*(); the missing
> puts have been there since the code was introduced.
> Drop each reference once the value derived from it is no longer needed:
> after of_iomap() has mapped the CCM registers, and after
> irq_domain_create_legacy() has taken its own fwnode reference.
> of_node_put() is NULL-safe, so platforms without these nodes are
> unaffected.
> Found by static analysis tool CodeQL.
> Fixes: 544496ab5cbd ("ARM: imx: move irq_domain_add_legacy call into avic driver")
> Fixes: 9b454d16e57d ("ARM: imx: avic: set low-power interrupt mask for imx25")
> Signed-off-by: Weigang He <geoffreyhe2@gmail.com>
> ---
> arch/arm/mach-imx/avic.c | 2 ++
> 1 file changed, 2 insertions(+)
> diff --git a/arch/arm/mach-imx/avic.c b/arch/arm/mach-imx/avic.c
> index 3067c06b4b8eb..6873a50bbe2c0 100644
> --- a/arch/arm/mach-imx/avic.c
> +++ b/arch/arm/mach-imx/avic.c
> @@ -173,6 +173,7 @@ static void __init mxc_init_irq(void __iomem *irqbase)
> np = of_find_compatible_node(NULL, NULL, "fsl,imx25-ccm");
> mx25_ccm_base = of_iomap(np, 0);
> + of_node_put(np);
> if (mx25_ccm_base) {
> /*
> @@ -203,6 +204,7 @@ static void __init mxc_init_irq(void __iomem *irqbase)
> np = of_find_compatible_node(NULL, NULL, "fsl,avic");
> domain = irq_domain_create_legacy(of_fwnode_handle(np), AVIC_NUM_IRQS, irq_base, 0,
> &irq_domain_simple_ops, NULL);
> + of_node_put(np);
> WARN_ON(!domain);
> for (i = 0; i < AVIC_NUM_IRQS / 32; i++, irq_base += 32)
> base-commit: 0f61b1860cc3f52aef9036d7235ed1f017632193
> --
> 2.43.0
Reviewed-by: Martin Kaiser <martin@kaiser.cx>
Thanks,
Martin
prev parent reply other threads:[~2026-06-10 9:42 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-10 5:31 [PATCH] ARM: imx: avic: fix device_node refcount leaks in mxc_init_irq() Weigang He
2026-06-10 9:42 ` Martin Kaiser [this message]
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=aikxg5xtXWccXPO8@nb282.user.codasip.com \
--to=martin@kaiser.cx \
--cc=festevam@gmail.com \
--cc=geoffreyhe2@gmail.com \
--cc=imx@lists.linux.dev \
--cc=kernel@pengutronix.de \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=s.hauer@pengutronix.de \
--cc=shawnguo@kernel.org \
/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