linux-kernel - [PATCH] can: mcp251x: fix deadlock in error path of mcp251x

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-Id: <20260209144706.2261954-1-alban.bedel@lht.dlh.de>
Date: Mon,  9 Feb 2026 15:47:05 +0100
From: Alban Bedel <alban.bedel@....dlh.de>
To: linux-can@...r.kernel.org
Cc: Alban Bedel <alban.bedel@....dlh.de>,
	Marc Kleine-Budde <mkl@...gutronix.de>,
	Vincent Mailhol <mailhol@...nel.org>,
	Bartosz Golaszewski <brgl@...nel.org>,
	Jakub Kicinski <kuba@...nel.org>,
	Marco Crivellari <marco.crivellari@...e.com>,
	Maud Spierings <maudspierings@...ontroll.com>,
	linux-kernel@...r.kernel.org
Subject: [PATCH] can: mcp251x: fix deadlock in error path of mcp251x_open

The mcp251x_open() function call free_irq() in its error path with the
mpc_lock mutex held. But if an interrupt already occurred the
interrupt handler will be waiting for the mpc_lock and free_irq() will
deadlock waiting for the handler to finish.

This issue is similar to the one fixed in commit 7dd9c26bd6cf ("can:
mcp251x: fix deadlock if an interrupt occurs during mcp251x_open") but
for the error path.

To solve this issue move the call to free_irq() after the lock is
released. Setting `priv->force_quit = 1` beforehand ensure that the IRQ
handler will exit right away once it acquired the lock.

Signed-off-by: Alban Bedel <alban.bedel@....dlh.de>
---
 drivers/net/can/spi/mcp251x.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c
index fa97adf25b734..bb7782582f401 100644
--- a/drivers/net/can/spi/mcp251x.c
+++ b/drivers/net/can/spi/mcp251x.c
@@ -1214,6 +1214,7 @@ static int mcp251x_open(struct net_device *net)
 {
 	struct mcp251x_priv *priv = netdev_priv(net);
 	struct spi_device *spi = priv->spi;
+	bool release_irq = false;
 	unsigned long flags = 0;
 	int ret;
 
@@ -1257,12 +1258,24 @@ static int mcp251x_open(struct net_device *net)
 	return 0;
 
 out_free_irq:
-	free_irq(spi->irq, priv);
+	/* The IRQ handler might be running, and if so it will be waiting
+	 * for the lock. But free_irq() must wait for the handler to finish
+	 * so calling it here would deadlock.
+	 *
+	 * Setting priv->force_quit will let the handler exit right away
+	 * without any access to the hardware. This make it safe to call
+	 * free_irq() after the lock is released.
+	 */
+	priv->force_quit = 1;
+	release_irq = true;
+
 	mcp251x_hw_sleep(spi);
 out_close:
 	mcp251x_power_enable(priv->transceiver, 0);
 	close_candev(net);
 	mutex_unlock(&priv->mcp_lock);
+	if (release_irq)
+		free_irq(spi->irq, priv);
 	return ret;
 }
 
-- 
2.39.5