netdev - [PATCH net-next 2/2] net: mii_timestamper: fix static allocation by PHY driver

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 for Android: free password hash cracker in your pocket

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

Date:   Thu, 30 Jan 2020 18:44:51 +0100
From:   Michael Walle <michael@...le.cc>
To:     linux-kernel@...r.kernel.org, netdev@...r.kernel.org
Cc:     Richard Cochran <richardcochran@...il.com>,
        Russell King <linux@...linux.org.uk>,
        Heiner Kallweit <hkallweit1@...il.com>,
        Florian Fainelli <f.fainelli@...il.com>,
        Andrew Lunn <andrew@...n.ch>, Michael Walle <michael@...le.cc>
Subject: [PATCH net-next 2/2] net: mii_timestamper: fix static allocation by PHY driver

If phydev->mii_ts is set by the PHY driver, it will always be
overwritten in of_mdiobus_register_phy(). Fix it. Also make sure, that
the unregister() doesn't do anything if the mii_timestamper was provided by
the PHY driver.

Fixes: 1dca22b18421 ("net: mdio: of: Register discovered MII time stampers.")
Signed-off-by: Michael Walle <michael@...le.cc>
---
 drivers/net/phy/mii_timestamper.c | 7 +++++++
 drivers/of/of_mdio.c              | 8 +++++++-
 2 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/mii_timestamper.c b/drivers/net/phy/mii_timestamper.c
index 2f12c5d901df..b71b7456462d 100644
--- a/drivers/net/phy/mii_timestamper.c
+++ b/drivers/net/phy/mii_timestamper.c
@@ -111,6 +111,13 @@ void unregister_mii_timestamper(struct mii_timestamper *mii_ts)
 	struct mii_timestamping_desc *desc;
 	struct list_head *this;
 
+	/* mii_timestamper statically registered by the PHY driver won't use the
+	 * register_mii_timestamper() and thus don't have ->device set. Don't
+	 * try to unregister these.
+	 */
+	if (!mii_ts->device)
+		return;
+
 	mutex_lock(&tstamping_devices_lock);
 	list_for_each(this, &mii_timestamping_devices) {
 		desc = list_entry(this, struct mii_timestamping_desc, list);
diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
index db0ed5879803..8270bbf505fb 100644
--- a/drivers/of/of_mdio.c
+++ b/drivers/of/of_mdio.c
@@ -124,7 +124,13 @@ static int of_mdiobus_register_phy(struct mii_bus *mdio,
 		of_node_put(child);
 		return rc;
 	}
-	phy->mii_ts = mii_ts;
+
+	/* phy->mii_ts may already be defined by the PHY driver. A
+	 * mii_timestamper probed via the device tree will still have
+	 * precedence.
+	 */
+	if (mii_ts)
+		phy->mii_ts = mii_ts;
 
 	dev_dbg(&mdio->dev, "registered phy %pOFn at address %i\n",
 		child, addr);
-- 
2.20.1