CVE-2024-49998

In the Linux kernel, the following vulnerability has been resolved:

net: dsa: improve shutdown sequence

Alexander Sverdlin presents 2 problems during shutdown with the lan9303 driver. One is specific to lan9303 and the other just happens to reproduce there.

The first problem is that lan9303 is unique among DSA drivers in that it calls devgetdrvdata() at "arbitrary runtime" (not probe, not shutdown, not remove):

phystatemachine() -> ... -> dsauserphyread() -> ds->ops->phyread() -> lan9303phyread() -> chip->ops->phyread() -> lan9303mdiophyread() -> devgetdrvdata()

But we never stop the phystatemachine(), so it may continue to run after dsaswitchshutdown(). Our common pattern in all DSA drivers is to set drvdata to NULL to suppress the remove() method that may come afterwards. But in this case it will result in an NPD.

The second problem is that the way in which we set dp->conduit->dsaptr = NULL; is concurrent with receive packet processing. dsaswitchrcv() checks once whether dev->dsaptr is NULL, but afterwards, rather than continuing to use that non-NULL value, dev->dsaptr is dereferenced again and again without NULL checks: dsaconduitfinduser() and many other places. In between dereferences, there is no locking to ensure that what was valid once continues to be valid.

Both problems have the common aspect that closing the conduit interface solves them.

In the first case, devclose(conduit) triggers the NETDEVGOINGDOWN event in dsausernetdeviceevent() which closes user ports as well. dsaportdisablert() calls phylinkstop(), which synchronously stops the phylink state machine, and ds->ops->phy_read() will thus no longer call into the driver after this point.

In the second case, dev_close(conduit) should do this, as per Documentation/networking/driver.rst:

| Quiescence | ---------- | | After the ndo_stop routine has been called, the hardware must | not receive or transmit any data. All in flight packets must | be aborted. If necessary, poll or wait for completion of | any reset commands.

So it should be sufficient to ensure that later, when we zeroize conduit->dsaptr, there will be no concurrent dsaswitch_rcv() call on this conduit.

The addition of the netifdevicedetach() function is to ensure that ioctls, rtnetlinks and ethtool requests on the user ports no longer propagate down to the driver - we're no longer prepared to handle them.

The race condition actually did not exist when commit 0650bf52b31f ("net: dsa: be compatible with masters which unregister on shutdown") first introduced dsaswitchshutdown(). It was created later, when we stopped unregistering the user interfaces from a bad spot, and we just replaced that sequence with a racy zeroization of conduit->dsa_ptr (one which doesn't ensure that the interfaces aren't up).