PCI: Prevent out of bounds access in numa_node override
authorSasha Levin <sasha.levin@oracle.com>
Wed, 7 Oct 2015 16:03:28 +0000 (11:03 -0500)
committerBjorn Helgaas <bhelgaas@google.com>
Wed, 7 Oct 2015 16:03:28 +0000 (11:03 -0500)
63692df103e9 ("PCI: Allow numa_node override via sysfs") didn't check that
the numa node provided by userspace is valid.  Passing a node number too
high would attempt to access invalid memory and trigger a kernel panic.

Fixes: 63692df103e9 ("PCI: Allow numa_node override via sysfs")
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
CC: stable@vger.kernel.org # v3.19+
drivers/pci/pci-sysfs.c

index 312f23a8429cd9331b45afaf72a278e84bd41d83..92618686604cb9d314aa1e6bf833363cfbaaa1b5 100644 (file)
@@ -216,7 +216,7 @@ static ssize_t numa_node_store(struct device *dev,
        if (ret)
                return ret;
 
-       if (!node_online(node))
+       if (node >= MAX_NUMNODES || !node_online(node))
                return -EINVAL;
 
        add_taint(TAINT_FIRMWARE_WORKAROUND, LOCKDEP_STILL_OK);