diff mbox

[v2] phy: Fix phy_device_free memory leak

Message ID 20130228110152.GA26262@bordel.klfree.net
State Accepted, archived
Delegated to: David Miller
Headers show

Commit Message

Petr Malat Feb. 28, 2013, 11:01 a.m. UTC 
From: Petr Malat <oss@malat.biz>

Fix memory leak in phy_device_free() for the case when phy_device*
returned by phy_device_create() is not registered in the system.

Bug description:
phy_device_create() sets name of kobject using dev_set_name(), which 
allocates memory using kvasprintf(), but this memory isn't freed if 
the underlying device isn't registered properly, because kobject_cleanup()
is not called in that case. This can happen (and actually is happening on 
our machines) if phy_device_register(), called by mdiobus_scan(), fails. 

Patch description:
Embedded struct device is initialized in phy_device_create() and it 
counterpart phy_device_free() just drops one reference to the device,
which leads to proper deinitialization including releasing the kobject 
name memory.

Signed-off-by: Petr Malat <oss@malat.biz>
---
Updated according to according to David Miller proposal - put_device is
used to release the memory now.

Please put me on CC, I'm not signed into the mailing list.

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

David Miller Feb. 28, 2013, 8:38 p.m. UTC | #1 
From: Petr Malat <oss@malat.biz>
Date: Thu, 28 Feb 2013 12:01:52 +0100

> 
> From: Petr Malat <oss@malat.biz>
> 
> Fix memory leak in phy_device_free() for the case when phy_device*
> returned by phy_device_create() is not registered in the system.
> 
> Bug description:
> phy_device_create() sets name of kobject using dev_set_name(), which 
> allocates memory using kvasprintf(), but this memory isn't freed if 
> the underlying device isn't registered properly, because kobject_cleanup()
> is not called in that case. This can happen (and actually is happening on 
> our machines) if phy_device_register(), called by mdiobus_scan(), fails. 
> 
> Patch description:
> Embedded struct device is initialized in phy_device_create() and it 
> counterpart phy_device_free() just drops one reference to the device,
> which leads to proper deinitialization including releasing the kobject 
> name memory.
> 
> Signed-off-by: Petr Malat <oss@malat.biz>

This looks great, applied, thanks.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

--- linux-3.8/drivers/net/phy/phy_device.c.orig	2013-02-19 00:58:34.000000000 +0100
+++ linux-3.8/drivers/net/phy/phy_device.c	2013-02-28 11:20:51.841528627 +0100
@@ -44,13 +44,13 @@  MODULE_LICENSE("GPL");
 
 void phy_device_free(struct phy_device *phydev)
 {
-	kfree(phydev);
+	put_device(&phydev->dev);
 }
 EXPORT_SYMBOL(phy_device_free);
 
 static void phy_device_release(struct device *dev)
 {
-	phy_device_free(to_phy_device(dev));
+	kfree(to_phy_device(dev));
 }
 
 static struct phy_driver genphy_driver;
@@ -201,6 +201,8 @@  struct phy_device *phy_device_create(str
 	   there's no driver _already_ loaded. */
 	request_module(MDIO_MODULE_PREFIX MDIO_ID_FMT, MDIO_ID_ARGS(phy_id));
 
+	device_initialize(&dev->dev);
+
 	return dev;
 }
 EXPORT_SYMBOL(phy_device_create);
@@ -363,9 +365,9 @@  int phy_device_register(struct phy_devic
 	/* Run all of the fixups for this PHY */
 	phy_scan_fixups(phydev);
 
-	err = device_register(&phydev->dev);
+	err = device_add(&phydev->dev);
 	if (err) {
-		pr_err("phy %d failed to register\n", phydev->addr);
+		pr_err("PHY %d failed to add\n", phydev->addr);
 		goto out;
 	}