5.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kyle Zeng <zengyhkyle@xxxxxxxxx>
[ Upstream commit 0113d9c9d1ccc07f5a3710dac4aa24b6d711278c ]
Currently, we assume the skb is associated with a device before calling
__ip_options_compile, which is not always the case if it is re-routed by
ipvs.
When skb->dev is NULL, dev_net(skb->dev) will become null-dereference.
This patch adds a check for the edge case and switch to use the net_device
from the rtable when skb->dev is NULL.
Fixes: ed0de45a1008 ("ipv4: recompile ip options in ipv4_link_failure")
Suggested-by: David Ahern <dsahern@xxxxxxxxxx>
Signed-off-by: Kyle Zeng <zengyhkyle@xxxxxxxxx>
Cc: Stephen Suryaputra <ssuryaextr@xxxxxxxxx>
Cc: Vadim Fedorenko <vfedorenko@xxxxxxxx>
Reviewed-by: David Ahern <dsahern@xxxxxxxxxx>
Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
net/ipv4/route.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index bc6240d327a8f..968cc4aa6e96c 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1220,6 +1220,7 @@ EXPORT_INDIRECT_CALLABLE(ipv4_dst_check);
static void ipv4_send_dest_unreach(struct sk_buff *skb)
{
+ struct net_device *dev;
struct ip_options opt;
int res;
@@ -1237,7 +1238,8 @@ static void ipv4_send_dest_unreach(struct sk_buff *skb)
opt.optlen = ip_hdr(skb)->ihl * 4 - sizeof(struct iphdr);
rcu_read_lock();
- res = __ip_options_compile(dev_net(skb->dev), &opt, skb, NULL);
+ dev = skb->dev ? skb->dev : skb_rtable(skb)->dst.dev;
+ res = __ip_options_compile(dev_net(dev), &opt, skb, NULL);
rcu_read_unlock();
if (res)
--
2.40.1
