net: sched: report if filter is too large to dump
authorRoman Kapl <code@rkapl.cz>
Mon, 19 Feb 2018 20:32:51 +0000 (21:32 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Fri, 9 Mar 2018 06:41:10 +0000 (22:41 -0800)
[ Upstream commit 5ae437ad5a2ed573b1ebb04e0afa70b8869f88dd ]

So far, if the filter was too large to fit in the allocated skb, the
kernel did not return any error and stopped dumping. Modify the dumper
so that it returns -EMSGSIZE when a filter fails to dump and it is the
first filter in the skb. If we are not first, we will get a next chance
with more room.

I understand this is pretty near to being an API change, but the
original design (silent truncation) can be considered a bug.

Note: The error case can happen pretty easily if you create a filter
with 32 actions and have 4kb pages. Also recent versions of iproute try
to be clever with their buffer allocation size, which in turn leads to

Signed-off-by: Roman Kapl <code@rkapl.cz>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/sched/cls_api.c

index 934c239..c2fab4b 100644 (file)
@@ -871,13 +871,18 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
                if (tca[TCA_CHAIN] &&
                    nla_get_u32(tca[TCA_CHAIN]) != chain->index)
                        continue;
-               if (!tcf_chain_dump(chain, skb, cb, index_start, &index))
+               if (!tcf_chain_dump(chain, skb, cb, index_start, &index)) {
+                       err = -EMSGSIZE;
                        break;
+               }
        }
 
        cb->args[0] = index;
 
 out:
+       /* If we did no progress, the error (EMSGSIZE) is real */
+       if (skb->len == 0 && err)
+               return err;
        return skb->len;
 }