openwrt/target/linux/generic/backport-5.4/610-v5.9-net-bridge-clear-bridge-s-private-skb-space-on-xmit.patch
Hauke Mehrtens 2d69d098e0 kernel: bump 5.4 to 5.4.182
The following patch was integrated upstream:
  target/linux/bcm4908/patches-5.4/180-i2c-brcmstb-fix-support-for-DSL-and-CM-variants.patch

All other updated automatically.

The new config option CONFIG_BPF_UNPRIV_DEFAULT_OFF is now handled too.

Compile-tested on: lantiq/xrx200, armvirt/64
Runtime-tested on: lantiq/xrx200, armvirt/64

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2022-03-06 20:55:26 +01:00

29 lines
1.1 KiB
Diff

From: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Date: Fri, 31 Jul 2020 19:26:16 +0300
Subject: [PATCH] net: bridge: clear bridge's private skb space on xmit
We need to clear all of the bridge private skb variables as they can be
stale due to the packet being recirculated through the stack and then
transmitted through the bridge device. Similar memset is already done on
bridge's input. We've seen cases where proxyarp_replied was 1 on routed
multicast packets transmitted through the bridge to ports with neigh
suppress which were getting dropped. Same thing can in theory happen with
the port isolation bit as well.
Fixes: 821f1b21cabb ("bridge: add new BR_NEIGH_SUPPRESS port flag to suppress arp and nd flood")
Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -35,6 +35,8 @@ netdev_tx_t br_dev_xmit(struct sk_buff *
const unsigned char *dest;
u16 vid = 0;
+ memset(skb->cb, 0, sizeof(struct br_input_skb_cb));
+
rcu_read_lock();
nf_ops = rcu_dereference(nf_br_ops);
if (nf_ops && nf_ops->br_dev_xmit_hook(skb)) {