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>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220331093040.GA20035@wunner.de>
Date:   Thu, 31 Mar 2022 11:30:40 +0200
From:   Lukas Wunner <lukas@...ner.de>
To:     Oliver Neukum <oneukum@...e.com>
Cc:     Oleksij Rempel <o.rempel@...gutronix.de>,
        Andrew Lunn <andrew@...n.ch>,
        Oleksij Rempel <linux@...pel-privat.de>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        Heiner Kallweit <hkallweit1@...il.com>
Subject: Re: ordering of call to unbind() in usbnet_disconnect

On Thu, Mar 31, 2022 at 11:20:50AM +0200, Oliver Neukum wrote:
> I am afraid, putting my
> maintainer's hat on, I have to point on that we have a stable tree for
> which we will need some solution.
> 
> Nor can usbnet exclusively cater to device that expose their PHY
> over MDIO. (or at all really). Intuitively I must say that exactly reversing
> the order of probe() in disconnect() is kind of the default.
> If there is a need to deviate from that, of course we will acomodate that,
> but making this the exclusive order is another matter.
> 
> I really get that you want to discuss this matter exhaustively, but we
> need to
> come to some kind of conclusion.

I propose the below patch.  If you could provide more details on the
regressions you've reported (+ ideally bugzilla links), I'll be happy
to include them in the commit message.  Thanks!

-- >8 --
Subject: [PATCH] usbnet: Run unregister_netdev() before unbind() again

Oliver says he got bug reports that commit 2c9d6c2b871d ("usbnet: run
unbind() before unregister_netdev()") is causing regressions.

The commit made binding and unbinding of USB Ethernet asymmetrical:
Before, usbnet_probe() first invoked the ->bind() callback and then
register_netdev().  usbnet_disconnect() mirrored that by first invoking
unregister_netdev() and then ->unbind().

Since the commit, the order in usbnet_disconnect() is reversed and no
longer mirrors usbnet_probe().

One consequence is that a PHY disconnected (and stopped) in ->unbind()
is afterwards stopped once more by unregister_netdev() as it closes the
netdev before unregistering.  That necessitates a contortion in ->stop()
because the PHY may only be stopped if it hasn't already been
disconnected.

Reverting the commit allows making the call to phy_stop() unconditional
in ->stop() and also fixes the issues reported by Oliver.

Fixes: 2c9d6c2b871d ("usbnet: run unbind() before unregister_netdev()")
Link: https://lore.kernel.org/netdev/62b944a1-0df2-6e81-397c-6bf9dea266ef@suse.com/
Reported-by: Oliver Neukum <oneukum@...e.com>
Signed-off-by: Lukas Wunner <lukas@...ner.de>
Cc: stable@...r.kernel.org # v5.14+
Cc: Oleksij Rempel <o.rempel@...gutronix.de>
Cc: Martyn Welch <martyn.welch@...labora.com>
Cc: Andrew Lunn <andrew@...n.ch>
---
 drivers/net/usb/asix_devices.c | 6 +-----
 drivers/net/usb/smsc95xx.c     | 3 +--
 drivers/net/usb/usbnet.c       | 6 +++---
 3 files changed, 5 insertions(+), 10 deletions(-)

diff --git a/drivers/net/usb/asix_devices.c b/drivers/net/usb/asix_devices.c
index 4514d35..7d569f0 100644
--- a/drivers/net/usb/asix_devices.c
+++ b/drivers/net/usb/asix_devices.c
@@ -794,11 +794,7 @@ static int ax88772_stop(struct usbnet *dev)
 {
 	struct asix_common_private *priv = dev->driver_priv;
 
-	/* On unplugged USB, we will get MDIO communication errors and the
-	 * PHY will be set in to PHY_HALTED state.
-	 */
-	if (priv->phydev->state != PHY_HALTED)
-		phy_stop(priv->phydev);
+	phy_stop(priv->phydev);
 
 	return 0;
 }
diff --git a/drivers/net/usb/smsc95xx.c b/drivers/net/usb/smsc95xx.c
index 193635e..3dc6df6 100644
--- a/drivers/net/usb/smsc95xx.c
+++ b/drivers/net/usb/smsc95xx.c
@@ -1280,8 +1280,7 @@ static int smsc95xx_start_phy(struct usbnet *dev)
 
 static int smsc95xx_stop(struct usbnet *dev)
 {
-	if (dev->net->phydev)
-		phy_stop(dev->net->phydev);
+	phy_stop(dev->net->phydev);
 
 	return 0;
 }
diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index 9a6450f..e8b4a93 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -1616,9 +1616,6 @@ void usbnet_disconnect (struct usb_interface *intf)
 		   xdev->bus->bus_name, xdev->devpath,
 		   dev->driver_info->description);
 
-	if (dev->driver_info->unbind)
-		dev->driver_info->unbind(dev, intf);
-
 	net = dev->net;
 	unregister_netdev (net);
 
@@ -1626,6 +1623,9 @@ void usbnet_disconnect (struct usb_interface *intf)
 
 	usb_scuttle_anchored_urbs(&dev->deferred);
 
+	if (dev->driver_info->unbind)
+		dev->driver_info->unbind (dev, intf);
+
 	usb_kill_urb(dev->interrupt);
 	usb_free_urb(dev->interrupt);
 	kfree(dev->padding_pkt);
-- 
2.34.1

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ