| Message ID | 20230830011128.1415752-3-iii@linux.ibm.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 6bd5bcb18f9467a655cf99c03ba5eeb64d896e41 |
| Delegated to: | BPF |
| Headers | show |
| Series | Implement cpuv4 support for s390x | expand |
[ +Florian ] On 8/30/23 3:07 AM, Ilya Leoshkevich wrote: > bpf_nf testcase fails on s390x: bpf_skb_ct_lookup() cannot find the > entry that was added by bpf_ct_insert_entry() within the same BPF > function. > > The reason is that this entry is deleted by nf_ct_gc_expired(). > > The CT timeout starts ticking after the CT confirmation; therefore > nf_conn.timeout is initially set to the timeout value, and > __nf_conntrack_confirm() sets it to the deadline value. > bpf_ct_insert_entry() sets IPS_CONFIRMED_BIT, but does not adjust the > timeout, making its value meaningless and causing false positives. > > Fix the problem by making bpf_ct_insert_entry() adjust the timeout, > like __nf_conntrack_confirm(). > > Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com> Should we route this fix via bpf tree instead? Also, could you reply with a Fixes tag? > --- > net/netfilter/nf_conntrack_bpf.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/net/netfilter/nf_conntrack_bpf.c b/net/netfilter/nf_conntrack_bpf.c > index c7a6114091ae..b21799d468d2 100644 > --- a/net/netfilter/nf_conntrack_bpf.c > +++ b/net/netfilter/nf_conntrack_bpf.c > @@ -381,6 +381,8 @@ __bpf_kfunc struct nf_conn *bpf_ct_insert_entry(struct nf_conn___init *nfct_i) > struct nf_conn *nfct = (struct nf_conn *)nfct_i; > int err; > > + if (!nf_ct_is_confirmed(nfct)) > + nfct->timeout += nfct_time_stamp; > nfct->status |= IPS_CONFIRMED; > err = nf_conntrack_hash_check_insert(nfct); > if (err < 0) { > Thanks, Daniel
On Thu, 2023-08-31 at 17:30 +0200, Daniel Borkmann wrote: > [ +Florian ] > > On 8/30/23 3:07 AM, Ilya Leoshkevich wrote: > > bpf_nf testcase fails on s390x: bpf_skb_ct_lookup() cannot find the > > entry that was added by bpf_ct_insert_entry() within the same BPF > > function. > > > > The reason is that this entry is deleted by nf_ct_gc_expired(). > > > > The CT timeout starts ticking after the CT confirmation; therefore > > nf_conn.timeout is initially set to the timeout value, and > > __nf_conntrack_confirm() sets it to the deadline value. > > bpf_ct_insert_entry() sets IPS_CONFIRMED_BIT, but does not adjust > > the > > timeout, making its value meaningless and causing false positives. > > > > Fix the problem by making bpf_ct_insert_entry() adjust the timeout, > > like __nf_conntrack_confirm(). > > > > Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com> > > Should we route this fix via bpf tree instead? Also, could you reply > with > a Fixes tag? Yes, putting this into the bpf tree makes sense to me. Should I resend with a different subject-prefix? Fixes: 2cdaa3eefed8 ("netfilter: conntrack: restore IPS_CONFIRMED out of nf_conntrack_hash_check_insert()") [...]
diff --git a/net/netfilter/nf_conntrack_bpf.c b/net/netfilter/nf_conntrack_bpf.c index c7a6114091ae..b21799d468d2 100644 --- a/net/netfilter/nf_conntrack_bpf.c +++ b/net/netfilter/nf_conntrack_bpf.c @@ -381,6 +381,8 @@ __bpf_kfunc struct nf_conn *bpf_ct_insert_entry(struct nf_conn___init *nfct_i) struct nf_conn *nfct = (struct nf_conn *)nfct_i; int err; + if (!nf_ct_is_confirmed(nfct)) + nfct->timeout += nfct_time_stamp; nfct->status |= IPS_CONFIRMED; err = nf_conntrack_hash_check_insert(nfct); if (err < 0) {
bpf_nf testcase fails on s390x: bpf_skb_ct_lookup() cannot find the entry that was added by bpf_ct_insert_entry() within the same BPF function. The reason is that this entry is deleted by nf_ct_gc_expired(). The CT timeout starts ticking after the CT confirmation; therefore nf_conn.timeout is initially set to the timeout value, and __nf_conntrack_confirm() sets it to the deadline value. bpf_ct_insert_entry() sets IPS_CONFIRMED_BIT, but does not adjust the timeout, making its value meaningless and causing false positives. Fix the problem by making bpf_ct_insert_entry() adjust the timeout, like __nf_conntrack_confirm(). Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com> --- net/netfilter/nf_conntrack_bpf.c | 2 ++ 1 file changed, 2 insertions(+)