From: Paul Blakey <paulb@xxxxxxxxxx>
[ Upstream commit 2dc4e9e88cfcc38454d52b01ed3422238c134003 ]
First fragmented packets (frag offset = 0) byte len is zeroed
when stolen by ip_defrag(). And since act_ct update the stats
only afterwards (at end of execute), bytes aren't correctly
accounted for such packets.
To fix this, move stats update to start of action execute.
Fixes: b57dc7c13ea9 ("net/sched: Introduce action ct")
Signed-off-by: Paul Blakey <paulb@xxxxxxxxxx>
Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
net/sched/act_ct.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/sched/act_ct.c b/net/sched/act_ct.c
index 1b4b3514c94f..07f4dce7b535 100644
--- a/net/sched/act_ct.c
+++ b/net/sched/act_ct.c
@@ -960,6 +960,7 @@ static int tcf_ct_act(struct sk_buff *skb, const struct tc_action *a,
tmpl = p->tmpl;
tcf_lastuse_update(&c->tcf_tm);
+ tcf_action_update_bstats(&c->common, skb);
if (clear) {
qdisc_skb_cb(skb)->post_ct = false;
@@ -1049,7 +1050,6 @@ out_push:
qdisc_skb_cb(skb)->post_ct = true;
out_clear:
- tcf_action_update_bstats(&c->common, skb);
if (defrag)
qdisc_skb_cb(skb)->pkt_len = skb->len;
return retval;
--
2.33.0
