[PATCH net v2] ipv6: do not let ipv6_find_hdr() return an offset past the packet end

[PATCH net v2] ipv6: do not let ipv6_find_hdr() return an offset past the packet end

[Norbert Szetei]

ipv6_find_hdr() walks the extension header chain, skipping each header by
the length that header itself declares.  ipv6_optlen() returns up to 2048,
and the skip is never checked against skb->len, so the offset stored in
*offset can point past the end of the packet.

openvswitch installs that offset as the transport header, and
update_ipv6_checksum() then reads and writes the transport checksum field
out of bounds:

  BUG: KASAN: slab-use-after-free in inet_proto_csum_replace16+0x445/0x470
  Read of size 2 at addr ffff88810b754b06 by task ovs_ipv6_oob/629
  CPU: 4 UID: 1000 PID: 629 Comm: ovs_ipv6_oob Tainted: G N 7.3.0-rc3+ #348
  Call Trace:
   inet_proto_csum_replace16+0x445/0x470
   set_ipv6_addr+0x3dd/0x460
   do_execute_actions+0x6a3d/0x7c40
   ovs_execute_actions+0xfd/0x480
   ovs_packet_cmd_execute+0xc38/0xf20
   genl_rcv_msg+0x59e/0x870
   netlink_rcv_skb+0x18b/0x450
   genl_rcv+0x2d/0x40
   netlink_unicast+0x6bc/0xa20

  The buggy address belongs to the object at ffff88810b754980
   which belongs to the cache skbuff_small_head of size 704
  The buggy address is located 390 bytes inside of
   freed 704-byte region [ffff88810b754980, ffff88810b754c40)

Other callers use that offset too, so bound it here rather than in one
caller.

Reject a header whose declared length does not fit in the packet.
ipv6_find_hdr() already fails with -EBADMSG on a malformed chain, so this
adds no new failure mode.

Fixes: f8f626754ebe ("ipv6: Move ipv6_find_hdr() out of Netfilter code.")
Suggested-by: Ilya Maximets <i.maximets@ovn.org>
Suggested-by: Eric Dumazet <edumazet@google.com>
Cc: stable@vger.kernel.org
Assisted-by: LLM
Signed-off-by: Norbert Szetei <norbert@doyensec.com>
---
v2:
  - fix ipv6_find_hdr() instead of the openvswitch caller, per the review
    of v1
  - dropped the openvswitch change
  - retitled
v1: https://lore.kernel.org/netdev/114FAFBD-F7BC-46AE-AF9F-0D15124F130E@doyensec.com/

Reproducer available on request.

 net/ipv6/exthdrs_core.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c
index 9d06d487e8b1..4a9748338cf4 100644
--- a/net/ipv6/exthdrs_core.c
+++ b/net/ipv6/exthdrs_core.c
@@ -278,6 +278,9 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset,
 			hdrlen = ipv6_optlen(hp);
 
 		if (!found) {
+			if (skb->len - start < hdrlen)
+				return -EBADMSG;
+
 			nexthdr = hp->nexthdr;
 			start += hdrlen;
 		}
-- 
2.55.0

Re: [PATCH net v2] ipv6: do not let ipv6_find_hdr() return an offset past the packet end

[Ido Schimmel]

On Wed, Sep 16, 2026 at 09:57:53PM +0200, Norbert Szetei wrote:
> ipv6_find_hdr() walks the extension header chain, skipping each header by
> the length that header itself declares.  ipv6_optlen() returns up to 2048,
> and the skip is never checked against skb->len, so the offset stored in
> *offset can point past the end of the packet.
> 
> openvswitch installs that offset as the transport header, and
> update_ipv6_checksum() then reads and writes the transport checksum field
> out of bounds:
> 
>   BUG: KASAN: slab-use-after-free in inet_proto_csum_replace16+0x445/0x470
>   Read of size 2 at addr ffff88810b754b06 by task ovs_ipv6_oob/629
>   CPU: 4 UID: 1000 PID: 629 Comm: ovs_ipv6_oob Tainted: G N 7.3.0-rc3+ #348
>   Call Trace:
>    inet_proto_csum_replace16+0x445/0x470
>    set_ipv6_addr+0x3dd/0x460
>    do_execute_actions+0x6a3d/0x7c40
>    ovs_execute_actions+0xfd/0x480
>    ovs_packet_cmd_execute+0xc38/0xf20
>    genl_rcv_msg+0x59e/0x870
>    netlink_rcv_skb+0x18b/0x450
>    genl_rcv+0x2d/0x40
>    netlink_unicast+0x6bc/0xa20
> 
>   The buggy address belongs to the object at ffff88810b754980
>    which belongs to the cache skbuff_small_head of size 704
>   The buggy address is located 390 bytes inside of
>    freed 704-byte region [ffff88810b754980, ffff88810b754c40)
> 
> Other callers use that offset too, so bound it here rather than in one
> caller.
> 
> Reject a header whose declared length does not fit in the packet.
> ipv6_find_hdr() already fails with -EBADMSG on a malformed chain, so this
> adds no new failure mode.
> 
> Fixes: f8f626754ebe ("ipv6: Move ipv6_find_hdr() out of Netfilter code.")

Blaming b777e0ce7437 would be more accurate, but it doesn't matter in
practice given that f8f626754ebe is from 2012.

> Suggested-by: Ilya Maximets <i.maximets@ovn.org>
> Suggested-by: Eric Dumazet <edumazet@google.com>
> Cc: stable@vger.kernel.org
> Assisted-by: LLM
> Signed-off-by: Norbert Szetei <norbert@doyensec.com>

Reviewed-by: Ido Schimmel <idosch@nvidia.com>

Re: [PATCH net v2] ipv6: do not let ipv6_find_hdr() return an offset past the packet end

[Ilya Maximets]

On 9/16/26 9:57 PM, Norbert Szetei wrote:
> ipv6_find_hdr() walks the extension header chain, skipping each header by
> the length that header itself declares.  ipv6_optlen() returns up to 2048,
> and the skip is never checked against skb->len, so the offset stored in
> *offset can point past the end of the packet.
> 
> openvswitch installs that offset as the transport header, and
> update_ipv6_checksum() then reads and writes the transport checksum field
> out of bounds:
> 
>   BUG: KASAN: slab-use-after-free in inet_proto_csum_replace16+0x445/0x470
>   Read of size 2 at addr ffff88810b754b06 by task ovs_ipv6_oob/629
>   CPU: 4 UID: 1000 PID: 629 Comm: ovs_ipv6_oob Tainted: G N 7.3.0-rc3+ #348
>   Call Trace:
>    inet_proto_csum_replace16+0x445/0x470
>    set_ipv6_addr+0x3dd/0x460
>    do_execute_actions+0x6a3d/0x7c40
>    ovs_execute_actions+0xfd/0x480
>    ovs_packet_cmd_execute+0xc38/0xf20
>    genl_rcv_msg+0x59e/0x870
>    netlink_rcv_skb+0x18b/0x450
>    genl_rcv+0x2d/0x40
>    netlink_unicast+0x6bc/0xa20
> 
>   The buggy address belongs to the object at ffff88810b754980
>    which belongs to the cache skbuff_small_head of size 704
>   The buggy address is located 390 bytes inside of
>    freed 704-byte region [ffff88810b754980, ffff88810b754c40)
> 
> Other callers use that offset too, so bound it here rather than in one
> caller.
> 
> Reject a header whose declared length does not fit in the packet.
> ipv6_find_hdr() already fails with -EBADMSG on a malformed chain, so this
> adds no new failure mode.
> 
> Fixes: f8f626754ebe ("ipv6: Move ipv6_find_hdr() out of Netfilter code.")
> Suggested-by: Ilya Maximets <i.maximets@ovn.org>
> Suggested-by: Eric Dumazet <edumazet@google.com>
> Cc: stable@vger.kernel.org
> Assisted-by: LLM
> Signed-off-by: Norbert Szetei <norbert@doyensec.com>
> ---
> v2:
>   - fix ipv6_find_hdr() instead of the openvswitch caller, per the review
>     of v1
>   - dropped the openvswitch change
>   - retitled
> v1: https://lore.kernel.org/netdev/114FAFBD-F7BC-46AE-AF9F-0D15124F130E@doyensec.com/
> 
> Reproducer available on request.
> 
>  net/ipv6/exthdrs_core.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c
> index 9d06d487e8b1..4a9748338cf4 100644
> --- a/net/ipv6/exthdrs_core.c
> +++ b/net/ipv6/exthdrs_core.c
> @@ -278,6 +278,9 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset,
>  			hdrlen = ipv6_optlen(hp);
>  
>  		if (!found) {
> +			if (skb->len - start < hdrlen)
> +				return -EBADMSG;
> +
>  			nexthdr = hp->nexthdr;
>  			start += hdrlen;
>  		}

Beside the inaccurate Fixes tag that Ido already flagged, the change
looks good to me.

Reviewed-by:  Ilya Maximets <i.maximets@ovn.org>

Re: [PATCH net v2] ipv6: do not let ipv6_find_hdr() return an offset past the packet end

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to netdev/net.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Wed, 16 Sep 2026 21:57:53 +0200 you wrote:
> ipv6_find_hdr() walks the extension header chain, skipping each header by
> the length that header itself declares.  ipv6_optlen() returns up to 2048,
> and the skip is never checked against skb->len, so the offset stored in
> *offset can point past the end of the packet.
> 
> openvswitch installs that offset as the transport header, and
> update_ipv6_checksum() then reads and writes the transport checksum field
> out of bounds:
> 
> [...]

Here is the summary with links:
  - [net,v2] ipv6: do not let ipv6_find_hdr() return an offset past the packet end
    https://git.kernel.org/netdev/net/c/ee319bd3a0e9

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html