[PATCH net v2] ipv6: prevent in6_dev_get() from resurrecting inet6_dev

[PATCH net v2] ipv6: prevent in6_dev_get() from resurrecting inet6_dev

[David Lee]

From: Kyle Zeng <kylebot@openai.com>

in6_dev_get() reads dev->ip6_ptr under RCU and then unconditionally
increments its refcount. Device teardown can clear the pointer and drop
the last reference between these operations. The increment then
resurrects an object whose RCU free has already been queued, so callers
can use it after it is freed.

Use refcount_inc_not_zero() and return NULL when the object has already
reached zero. RCU keeps the memory accessible through the attempted
reference acquisition, and a successful increment pins the object for
the caller.

An independent run on the exact unpatched 6f5156d7a31a (v7.2-rc3)
kernel reproduced the invalid reference acquisition as UID 1000:

  refcount_t: addition on 0; use-after-free.
  ip6_mc_source+0xef4/0x17e0

It was followed by the corresponding reference underflow in
ip6_mc_source(). The supplied trace from the same unpatched revision
additionally shows the access after the RCU read-side section ends:

  BUG: KASAN: slab-use-after-free in mutex_lock+0x76/0xe0
  Write of size 8 at addr ffff888015b50240 by task poc/1219

Bug found and triaged by OpenAI Security Research and
validated by Trail of Bits.

Fixes: 8814c4b53381 ("[IPV6] ADDRCONF: Convert addrconf_lock to RCU.")
Cc: stable@vger.kernel.org
Assisted-by: Codex:gpt-5.6-sol gpt-5.5-cyber
Signed-off-by: Kyle Zeng <kylebot@openai.com>
Co-developed-by: David Lee <david.lee@trailofbits.com>
Signed-off-by: David Lee <david.lee@trailofbits.com>
---
Changes in v2:
- Use the net tree and v2 subject prefixes.
- Preserve Kyle as the author and add David's co-development and
  submission trailers.
- Include the initial refcount warning and identify the unpatched test
  revision.

Ido flagged the corresponding IPv4 issue, which turned out to have a
similar refcount race. in_dev_get() performs a zero-to-one refcount
increment that can be reached through RTM_GETNETCONF.

I reproduced it on an unpatched v7.2-rc5 kernel. It first produced a
refcount warning, followed by a KASAN slab-use-after-free in
inet_netconf_fill_devconf().

I will send a separate patch for IPv4 because this bug was introduced
by a different commit.

Link: https://lore.kernel.org/netdev/20260731135202.566337-1-david.lee@trailofbits.com/

 include/net/addrconf.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 8ced27a82..e67642459 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -405,8 +405,8 @@ static inline struct inet6_dev *in6_dev_get(const struct net_device *dev)
 
 	rcu_read_lock();
 	idev = rcu_dereference(dev->ip6_ptr);
-	if (idev)
-		refcount_inc(&idev->refcnt);
+	if (idev && !refcount_inc_not_zero(&idev->refcnt))
+		idev = NULL;
 	rcu_read_unlock();
 	return idev;
 }

Re: [PATCH net v2] ipv6: prevent in6_dev_get() from resurrecting inet6_dev

[Ido Schimmel]

On Mon, Aug 03, 2026 at 12:27:57PM +0000, David Lee wrote:
> From: Kyle Zeng <kylebot@openai.com>
> 
> in6_dev_get() reads dev->ip6_ptr under RCU and then unconditionally
> increments its refcount. Device teardown can clear the pointer and drop
> the last reference between these operations. The increment then
> resurrects an object whose RCU free has already been queued, so callers
> can use it after it is freed.
> 
> Use refcount_inc_not_zero() and return NULL when the object has already
> reached zero. RCU keeps the memory accessible through the attempted
> reference acquisition, and a successful increment pins the object for
> the caller.
> 
> An independent run on the exact unpatched 6f5156d7a31a (v7.2-rc3)
> kernel reproduced the invalid reference acquisition as UID 1000:
> 
>   refcount_t: addition on 0; use-after-free.
>   ip6_mc_source+0xef4/0x17e0
> 
> It was followed by the corresponding reference underflow in
> ip6_mc_source(). The supplied trace from the same unpatched revision
> additionally shows the access after the RCU read-side section ends:
> 
>   BUG: KASAN: slab-use-after-free in mutex_lock+0x76/0xe0
>   Write of size 8 at addr ffff888015b50240 by task poc/1219
> 
> Bug found and triaged by OpenAI Security Research and
> validated by Trail of Bits.
> 
> Fixes: 8814c4b53381 ("[IPV6] ADDRCONF: Convert addrconf_lock to RCU.")
> Cc: stable@vger.kernel.org
> Assisted-by: Codex:gpt-5.6-sol gpt-5.5-cyber
> Signed-off-by: Kyle Zeng <kylebot@openai.com>
> Co-developed-by: David Lee <david.lee@trailofbits.com>
> Signed-off-by: David Lee <david.lee@trailofbits.com>

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

Re: [PATCH net v2] ipv6: prevent in6_dev_get() from resurrecting inet6_dev

[patchwork-bot+netdevbpf]

Hello:

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

On Mon,  3 Aug 2026 12:27:57 +0000 you wrote:
> From: Kyle Zeng <kylebot@openai.com>
> 
> in6_dev_get() reads dev->ip6_ptr under RCU and then unconditionally
> increments its refcount. Device teardown can clear the pointer and drop
> the last reference between these operations. The increment then
> resurrects an object whose RCU free has already been queued, so callers
> can use it after it is freed.
> 
> [...]

Here is the summary with links:
  - [net,v2] ipv6: prevent in6_dev_get() from resurrecting inet6_dev
    https://git.kernel.org/netdev/net/c/0e243671bc7b

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