[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20231114004220.6495-6-daniel@iogearbox.net>
Date: Tue, 14 Nov 2023 01:42:17 +0100
From: Daniel Borkmann <daniel@...earbox.net>
To: martin.lau@...nel.org
Cc: kuba@...nel.org,
razor@...ckwall.org,
sdf@...gle.com,
horms@...nel.org,
netdev@...r.kernel.org,
bpf@...r.kernel.org,
Peilin Ye <peilin.ye@...edance.com>,
Youlun Zhang <zhangyoulun@...edance.com>,
Daniel Borkmann <daniel@...earbox.net>
Subject: [PATCH bpf v3 5/8] bpf: Fix dev's rx stats for bpf_redirect_peer traffic
From: Peilin Ye <peilin.ye@...edance.com>
Traffic redirected by bpf_redirect_peer() (used by recent CNIs like Cilium)
is not accounted for in the RX stats of supported devices (that is, veth
and netkit), confusing user space metrics collectors such as cAdvisor [0],
as reported by Youlun.
Fix it by calling dev_sw_netstats_rx_add() in skb_do_redirect(), to update
RX traffic counters. Devices that support ndo_get_peer_dev _must_ use the
@tstats per-CPU counters (instead of @lstats, or @dstats).
To make this more fool-proof, error out when ndo_get_peer_dev is set but
@tstats are not selected.
[0] Specifically, the "container_network_receive_{byte,packet}s_total"
counters are affected.
Fixes: 9aa1206e8f48 ("bpf: Add redirect_peer helper")
Reported-by: Youlun Zhang <zhangyoulun@...edance.com>
Signed-off-by: Peilin Ye <peilin.ye@...edance.com>
Co-developed-by: Daniel Borkmann <daniel@...earbox.net>
Signed-off-by: Daniel Borkmann <daniel@...earbox.net>
Reviewed-by: Nikolay Aleksandrov <razor@...ckwall.org>
---
net/core/dev.c | 8 ++++++++
net/core/filter.c | 1 +
2 files changed, 9 insertions(+)
diff --git a/net/core/dev.c b/net/core/dev.c
index ac9bd7bfd1af..70c66b3b227b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -10053,6 +10053,14 @@ static int netdev_do_alloc_pcpu_stats(struct net_device *dev)
{
void __percpu *v;
+ /* Drivers implementing ndo_get_peer_dev must support tstat
+ * accounting, so that skb_do_redirect() can bump the dev's
+ * RX stats upon network namespace switch.
+ */
+ if (dev->netdev_ops->ndo_get_peer_dev &&
+ dev->pcpu_stat_type != NETDEV_PCPU_STAT_TSTATS)
+ return -EOPNOTSUPP;
+
switch (dev->pcpu_stat_type) {
case NETDEV_PCPU_STAT_NONE:
return 0;
diff --git a/net/core/filter.c b/net/core/filter.c
index 383f96b0a1c7..cca810987c8d 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -2492,6 +2492,7 @@ int skb_do_redirect(struct sk_buff *skb)
net_eq(net, dev_net(dev))))
goto out_drop;
skb->dev = dev;
+ dev_sw_netstats_rx_add(dev, skb->len);
return -EAGAIN;
}
return flags & BPF_F_NEIGH ?
--
2.34.1
Powered by blists - more mailing lists