key: No support for family zero

[PATCH] key: No support for family zero

Posted by Edward Adam Davis 3 months, 3 weeks ago

When setting the extended skb data for sadb_x_ipsecrequest, the requested
extended data size exceeds the allocated skb data length, triggering the
reported bug.

Because family only supports AF_INET and AF_INET6, other values will cause
pfkey_sockaddr_fill() to fail, which in turn causes set_ipsecrequest() to
fail.

Therefore, a workaround is available here: using a family value of 0 to
resolve the issue of excessively large extended data length.

syzbot reported:
kernel BUG at net/core/skbuff.c:212!
Call Trace:
 skb_over_panic net/core/skbuff.c:217 [inline]
 skb_put+0x159/0x210 net/core/skbuff.c:2583
 skb_put_zero include/linux/skbuff.h:2788 [inline]
 set_ipsecrequest+0x73/0x680 net/key/af_key.c:3532

Fixes: 08de61beab8a ("[PFKEYV2]: Extension for dynamic update of endpoint address(es)")
Reported-by: syzbot+be97dd4da14ae88b6ba4@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=be97dd4da14ae88b6ba4
Signed-off-by: Edward Adam Davis <eadavis@qq.com>
---
 net/key/af_key.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/key/af_key.c b/net/key/af_key.c
index 2ebde0352245..e658c129b38f 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -3526,6 +3526,9 @@ static int set_ipsecrequest(struct sk_buff *skb,
 	int socklen = pfkey_sockaddr_len(family);
 	int size_req;
 
+	if (!family)
+		return -EINVAL;
+
 	size_req = sizeof(struct sadb_x_ipsecrequest) +
 		   pfkey_sockaddr_pair_size(family);
 
-- 
2.43.0

Re: [PATCH] key: No support for family zero

Posted by Simon Horman 3 months, 1 week ago

+ Shaurya Rane, clingfei

On Sun, Oct 19, 2025 at 10:34:11AM +0800, Edward Adam Davis wrote:
> When setting the extended skb data for sadb_x_ipsecrequest, the requested
> extended data size exceeds the allocated skb data length, triggering the
> reported bug.
> 
> Because family only supports AF_INET and AF_INET6, other values will cause
> pfkey_sockaddr_fill() to fail, which in turn causes set_ipsecrequest() to
> fail.
> 
> Therefore, a workaround is available here: using a family value of 0 to
> resolve the issue of excessively large extended data length.
> 
> syzbot reported:
> kernel BUG at net/core/skbuff.c:212!
> Call Trace:
>  skb_over_panic net/core/skbuff.c:217 [inline]
>  skb_put+0x159/0x210 net/core/skbuff.c:2583
>  skb_put_zero include/linux/skbuff.h:2788 [inline]
>  set_ipsecrequest+0x73/0x680 net/key/af_key.c:3532
> 
> Fixes: 08de61beab8a ("[PFKEYV2]: Extension for dynamic update of endpoint address(es)")
> Reported-by: syzbot+be97dd4da14ae88b6ba4@syzkaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?extid=be97dd4da14ae88b6ba4
> Signed-off-by: Edward Adam Davis <eadavis@qq.com>

Hi,

There are several patches relating to this issue. And they seem
to take one of two approaches.

1. As with this patch [a] and [b]: check the return value of
   pfkey_sockaddr_len()

2. As in [c]: correct the type of the family argument to set_ipsecrequest()


[a] key: No support for family zero
    https://lore.kernel.org/all/tencent_57525DE2DDF41911CFDB8DF525A08D9D9207@qq.com/

[b] net: key: Validate address family in set_ipsecrequest()
  https://lore.kernel.org/all/CANNWa05pX3ratdawb2A6AUBocUgYo+EKZeHBZohQWuBC6_W1AA@mail.gmail.com/

[c] fix integer overflow in set_ipsecrequest
    https://lore.kernel.org/all/20251021030035.1424912-1-1599101385@qq.com/

I would appreciate it if the patch authors could coordinate creating a
patch(set) to address this issue. And look over the more detailed response
I provided to [c].

Thanks!

-- 
pw-bot: changes-requested