| Message ID | 20180115154918.4176669-2-arnd@arndb.de |
|---|---|
| State | Accepted |
| Commit | 41e4b391157ff20aa911d1ef6cf3d6db079e9e57 |
| Headers | show |
| Series | [net-next,1/2] netfilter: nf_defrag: mark xt_table structures 'const' again | expand |
On Mon, Jan 15, 2018 at 04:49:06PM +0100, Arnd Bergmann wrote: > We cannot access the skb->_nfct field when CONFIG_NF_CONNTRACK is > disabled: > > net/ipv4/netfilter/nf_defrag_ipv4.c: In function 'ipv4_conntrack_defrag': > net/ipv4/netfilter/nf_defrag_ipv4.c:83:9: error: 'struct sk_buff' has no member named '_nfct' > net/ipv6/netfilter/nf_defrag_ipv6_hooks.c: In function 'ipv6_defrag': > net/ipv6/netfilter/nf_defrag_ipv6_hooks.c:68:9: error: 'struct sk_buff' has no member named '_nfct' > > Both functions already have an #ifdef for this, so let's move the > check in there. Also applied, thanks.
diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c index cbd987f6b1f8..a0d3ad60a411 100644 --- a/net/ipv4/netfilter/nf_defrag_ipv4.c +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c @@ -78,9 +78,11 @@ static unsigned int ipv4_conntrack_defrag(void *priv, if (skb_nfct(skb) && !nf_ct_is_template((struct nf_conn *)skb_nfct(skb))) return NF_ACCEPT; #endif + if (skb->_nfct == IP_CT_UNTRACKED) + return NF_ACCEPT; #endif /* Gather fragments. */ - if (skb->_nfct != IP_CT_UNTRACKED && ip_is_fragment(ip_hdr(skb))) { + if (ip_is_fragment(ip_hdr(skb))) { enum ip_defrag_users user = nf_ct_defrag_user(state->hook, skb); diff --git a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c index 87b503a8f5ef..c87b48359e8f 100644 --- a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +++ b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c @@ -63,10 +63,10 @@ static unsigned int ipv6_defrag(void *priv, /* Previously seen (loopback)? */ if (skb_nfct(skb) && !nf_ct_is_template((struct nf_conn *)skb_nfct(skb))) return NF_ACCEPT; -#endif if (skb->_nfct == IP_CT_UNTRACKED) return NF_ACCEPT; +#endif err = nf_ct_frag6_gather(state->net, skb, nf_ct6_defrag_user(state->hook, skb));
We cannot access the skb->_nfct field when CONFIG_NF_CONNTRACK is disabled: net/ipv4/netfilter/nf_defrag_ipv4.c: In function 'ipv4_conntrack_defrag': net/ipv4/netfilter/nf_defrag_ipv4.c:83:9: error: 'struct sk_buff' has no member named '_nfct' net/ipv6/netfilter/nf_defrag_ipv6_hooks.c: In function 'ipv6_defrag': net/ipv6/netfilter/nf_defrag_ipv6_hooks.c:68:9: error: 'struct sk_buff' has no member named '_nfct' Both functions already have an #ifdef for this, so let's move the check in there. Fixes: 902d6a4c2a4f ("netfilter: nf_defrag: Skip defrag if NOTRACK is set") Signed-off-by: Arnd Bergmann <arnd@arndb.de> --- Please double-check what the right behavior for !CONFIG_NF_CONNTRACK should be, I was only guessing here. --- net/ipv4/netfilter/nf_defrag_ipv4.c | 4 +++- net/ipv6/netfilter/nf_defrag_ipv6_hooks.c | 2 +- 2 files changed, 4 insertions(+), 2 deletions(-) -- 2.9.0