net: act_bpf: fix size mismatch on filter preparation
authorDaniel Borkmann <dborkman@redhat.com>
Thu, 22 Jan 2015 09:58:19 +0000 (10:58 +0100)
committerDavid S. Miller <davem@davemloft.net>
Tue, 27 Jan 2015 00:08:55 +0000 (16:08 -0800)
Similarly as in cls_bpf, also this code needs to reject mismatches.

Reference: http://article.gmane.org/gmane.linux.network/347406
Fixes: d23b8ad8ab23 ("tc: add BPF based action")
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Acked-by: Jiri Pirko <jiri@resnulli.us>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/sched/act_bpf.c

index 1bd257e473a9a378d00da3593ab199e5ae87b827..82c5d7fc19881577d888fad3d5b0369e4698172b 100644 (file)
@@ -122,6 +122,9 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla,
                return -EINVAL;
 
        bpf_size = bpf_num_ops * sizeof(*bpf_ops);
+       if (bpf_size != nla_len(tb[TCA_ACT_BPF_OPS]))
+               return -EINVAL;
+
        bpf_ops = kzalloc(bpf_size, GFP_KERNEL);
        if (!bpf_ops)
                return -ENOMEM;