Messages in this thread | | | From | Eric Dumazet <> | Date | Tue, 8 Mar 2022 15:18:25 -0800 | Subject | Re: [PATCH v3 net-next] net-core: add rx_otherhost_dropped counter |
| |
On Tue, Mar 8, 2022 at 2:51 PM David Ahern <dsahern@kernel.org> wrote: > > On 3/8/22 2:25 PM, Jeffrey Ji wrote: > > diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c > > index 95f7bb052784..8b87ea99904b 100644 > > --- a/net/ipv4/ip_input.c > > +++ b/net/ipv4/ip_input.c > > @@ -451,6 +451,7 @@ static struct sk_buff *ip_rcv_core(struct sk_buff *skb, struct net *net) > > * that it receives, do not try to analyse it. > > */ > > if (skb->pkt_type == PACKET_OTHERHOST) { > > + atomic_long_inc(&skb->dev->rx_otherhost_dropped); > > drop_reason = SKB_DROP_REASON_OTHERHOST; > > goto drop; > > } > > diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c > > index 5b5ea35635f9..5624c937f87f 100644 > > --- a/net/ipv6/ip6_input.c > > +++ b/net/ipv6/ip6_input.c > > @@ -150,6 +150,7 @@ static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev, > > struct inet6_dev *idev; > > > > if (skb->pkt_type == PACKET_OTHERHOST) { > > + atomic_long_inc(&skb->dev->rx_otherhost_dropped); > > kfree_skb(skb); > > return NULL; > > } > > that's an expensive packet counter for a common path (e.g., hosting > environments).
This was the reason for the initial patch, using SNMP stat, being per cpu.
Adding per-device per-cpu data for this counter will increase cost of netdevice dismantle phase, and increase time for ndo_get_stats64(), especially on hosts with 256 or 512 cpus.
| |