]> git.ipfire.org Git - thirdparty/kernel/stable.git/commitdiff
net: sunrpc: interpret the return value of kstrtou32 correctly
authorj.nixdorf@avm.de <j.nixdorf@avm.de>
Tue, 5 Jan 2021 14:17:01 +0000 (15:17 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sat, 23 Jan 2021 14:36:55 +0000 (15:36 +0100)
commit 86b53fbf08f48d353a86a06aef537e78e82ba721 upstream.

A return value of 0 means success. This is documented in lib/kstrtox.c.

This was found by trying to mount an NFS share from a link-local IPv6
address with the interface specified by its index:

  mount("[fe80::1%1]:/srv/nfs", "/mnt", "nfs", 0, "nolock,addr=fe80::1%1")

Before this commit this failed with EINVAL and also caused the following
message in dmesg:

  [...] NFS: bad IP address specified: addr=fe80::1%1

The syscall using the same address based on the interface name instead
of its index succeeds.

Credits for this patch go to my colleague Christian Speich, who traced
the origin of this bug to this line of code.

Signed-off-by: Johannes Nixdorf <j.nixdorf@avm.de>
Fixes: 00cfaa943ec3 ("replace strict_strto calls")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/sunrpc/addr.c

index 8391c2785550135d270b1039f5a7b7670aa05ef7..7404f02702a1c44a15168fa3ae1c393f73720732 100644 (file)
@@ -184,7 +184,7 @@ static int rpc_parse_scope_id(struct net *net, const char *buf,
                        scope_id = dev->ifindex;
                        dev_put(dev);
                } else {
-                       if (kstrtou32(p, 10, &scope_id) == 0) {
+                       if (kstrtou32(p, 10, &scope_id) != 0) {
                                kfree(p);
                                return 0;
                        }