[PATCH net 1/2] nexthop: take nh->lock for f6i_list walks in replace check and notify

[PATCH net 1/2] nexthop: take nh->lock for f6i_list walks in replace check and notify

[Xiang Mei (Microsoft)]

fib6_check_nh_list() and __nexthop_replace_notify() walk nh->f6i_list
during an RTNL-serialized nexthop replace without holding nh->lock. IPv6
RTM_NEWROUTE/RTM_DELROUTE run without RTNL and mutate that list under
nh->lock (fib6_add_rt2node_nh(), fib6_purge_rt()), so both walks race a
concurrent route delete that unlinks and frees a fib6_info:

  BUG: KASAN: slab-use-after-free in rt6_fill_node.isra.0 (net/ipv6/route.c:5799)
  Read of size 4 at addr ffff888014607e64 by task exploit/143
   rt6_fill_node.isra.0 (net/ipv6/route.c:5799)
   fib6_rt_update (net/ipv6/route.c:6412)
   __nexthop_replace_notify (net/ipv4/nexthop.c:2542)
   rtm_new_nexthop (net/ipv4/nexthop.c:2554)
   rtnetlink_rcv_msg (net/core/rtnetlink.c:7076)

  BUG: KASAN: slab-use-after-free in fib6_check_nh_list (net/ipv4/nexthop.c:1605)
  Read of size 8 at addr ffff888014a7d068 by task exploit/142
   fib6_check_nh_list (net/ipv4/nexthop.c:1605)
   rtm_new_nexthop (net/ipv4/nexthop.c:2575)
   rtnetlink_rcv_msg (net/core/rtnetlink.c:7076)

Both walks only read the entries and take no tb6_lock, so protect them
with nh->lock; fib6_rt_update() uses gfp_any(), which returns GFP_ATOMIC
under the lock.

Fixes: 081efd18326e ("ipv6: Protect nh->f6i_list with spinlock and flag.")
Reported-by: AutonomousCodeSecurity@microsoft.com
Signed-off-by: Xiang Mei (Microsoft) <xmei5@asu.edu>
---
 net/ipv4/nexthop.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
index 6205bd57aa85..c1d3b7d7100d 100644
--- a/net/ipv4/nexthop.c
+++ b/net/ipv4/nexthop.c
@@ -1597,14 +1597,21 @@ static int fib6_check_nh_list(struct nexthop *old, struct nexthop *new,
 			      struct netlink_ext_ack *extack)
 {
 	struct fib6_info *f6i;
+	int err = 0;
 
 	if (list_empty(&old->f6i_list))
 		return 0;
 
+	spin_lock_bh(&old->lock);
 	list_for_each_entry(f6i, &old->f6i_list, nh_list) {
-		if (check_src_addr(&f6i->fib6_src.addr, extack) < 0)
-			return -EINVAL;
+		err = check_src_addr(&f6i->fib6_src.addr, extack);
+		if (err)
+			break;
 	}
+	spin_unlock_bh(&old->lock);
+
+	if (err)
+		return err;
 
 	return fib6_check_nexthop(new, NULL, extack);
 }
@@ -2538,8 +2545,10 @@ static void __nexthop_replace_notify(struct net *net, struct nexthop *nh,
 			fi->nh_updated = false;
 	}
 
+	spin_lock_bh(&nh->lock);
 	list_for_each_entry(f6i, &nh->f6i_list, nh_list)
 		fib6_rt_update(net, f6i, info);
+	spin_unlock_bh(&nh->lock);
 }
 
 /* send RTM_NEWROUTE with REPLACE flag set for all FIB entries
-- 
2.43.0

[PATCH net 2/2] nexthop: avoid unlocked f6i_list walk in nh_rt_cache_flush

[Xiang Mei (Microsoft)]

nh_rt_cache_flush() walks nh->f6i_list during an RTNL-serialized nexthop
replace without holding nh->lock, racing the unlocked IPv6 route
add/delete that mutate the list under nh->lock and free fib6_info
entries (nh_rt_cache_flush() is inlined into rtm_new_nexthop()):

  BUG: KASAN: slab-use-after-free in nh_rt_cache_flush (net/ipv4/nexthop.c:2243)
  Read of size 8 at addr ffff888012953e18 by task exploit/146
   nh_rt_cache_flush (net/ipv4/nexthop.c:2243)
   replace_nexthop (net/ipv4/nexthop.c:2610)
   rtm_new_nexthop (net/ipv4/nexthop.c:3323)
   rtnetlink_rcv_msg (net/core/rtnetlink.c:7076)

Unlike the other f6i_list walks, this one bumps each route's sernum via
fib6_update_sernum_upto_root(), which needs tb6_lock; taking nh->lock
around it would invert the established tb6_lock -> nh->lock order and
deadlock. As the only purpose is to invalidate cached dsts, bump the
IPv6 sernum for the whole netns with rt_genid_bump_ipv6() instead,
mirroring the rt_cache_flush() already done for IPv4 just above.

Fixes: 081efd18326e ("ipv6: Protect nh->f6i_list with spinlock and flag.")
Reported-by: AutonomousCodeSecurity@microsoft.com
Signed-off-by: Xiang Mei (Microsoft) <xmei5@asu.edu>
---
 net/ipv4/nexthop.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
index c1d3b7d7100d..1e9ccac6bf0f 100644
--- a/net/ipv4/nexthop.c
+++ b/net/ipv4/nexthop.c
@@ -2240,18 +2240,18 @@ static void remove_one_nexthop(struct net *net, struct nexthop *nh,
 static void nh_rt_cache_flush(struct net *net, struct nexthop *nh,
 			      struct nexthop *replaced_nh)
 {
-	struct fib6_info *f6i;
 	struct nh_group *nhg;
+	bool have_f6i;
 	int i;
 
 	if (!list_empty(&nh->fi_list))
 		rt_cache_flush(net);
 
-	list_for_each_entry(f6i, &nh->f6i_list, nh_list) {
-		spin_lock_bh(&f6i->fib6_table->tb6_lock);
-		fib6_update_sernum_upto_root(net, f6i);
-		spin_unlock_bh(&f6i->fib6_table->tb6_lock);
-	}
+	spin_lock_bh(&nh->lock);
+	have_f6i = !list_empty(&nh->f6i_list);
+	spin_unlock_bh(&nh->lock);
+	if (have_f6i)
+		rt_genid_bump_ipv6(net);
 
 	/* if an IPv6 group was replaced, we have to release all old
 	 * dsts to make sure all refcounts are released
-- 
2.43.0

Re: [PATCH net 1/2] nexthop: take nh->lock for f6i_list walks in replace check and notify

[Ido Schimmel]

On Wed, Jul 22, 2026 at 12:29:50AM +0000, Xiang Mei (Microsoft) wrote:
> fib6_check_nh_list() and __nexthop_replace_notify() walk nh->f6i_list
> during an RTNL-serialized nexthop replace without holding nh->lock. IPv6
> RTM_NEWROUTE/RTM_DELROUTE run without RTNL and mutate that list under
> nh->lock (fib6_add_rt2node_nh(), fib6_purge_rt()), so both walks race a
> concurrent route delete that unlinks and frees a fib6_info:
> 
>   BUG: KASAN: slab-use-after-free in rt6_fill_node.isra.0 (net/ipv6/route.c:5799)
>   Read of size 4 at addr ffff888014607e64 by task exploit/143
>    rt6_fill_node.isra.0 (net/ipv6/route.c:5799)
>    fib6_rt_update (net/ipv6/route.c:6412)
>    __nexthop_replace_notify (net/ipv4/nexthop.c:2542)
>    rtm_new_nexthop (net/ipv4/nexthop.c:2554)
>    rtnetlink_rcv_msg (net/core/rtnetlink.c:7076)
> 
>   BUG: KASAN: slab-use-after-free in fib6_check_nh_list (net/ipv4/nexthop.c:1605)
>   Read of size 8 at addr ffff888014a7d068 by task exploit/142
>    fib6_check_nh_list (net/ipv4/nexthop.c:1605)
>    rtm_new_nexthop (net/ipv4/nexthop.c:2575)
>    rtnetlink_rcv_msg (net/core/rtnetlink.c:7076)
> 
> Both walks only read the entries and take no tb6_lock, so protect them
> with nh->lock; fib6_rt_update() uses gfp_any(), which returns GFP_ATOMIC
> under the lock.
> 
> Fixes: 081efd18326e ("ipv6: Protect nh->f6i_list with spinlock and flag.")
> Reported-by: AutonomousCodeSecurity@microsoft.com
> Signed-off-by: Xiang Mei (Microsoft) <xmei5@asu.edu>
> ---
>  net/ipv4/nexthop.c | 13 +++++++++++--
>  1 file changed, 11 insertions(+), 2 deletions(-)
> 
> diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
> index 6205bd57aa85..c1d3b7d7100d 100644
> --- a/net/ipv4/nexthop.c
> +++ b/net/ipv4/nexthop.c
> @@ -1597,14 +1597,21 @@ static int fib6_check_nh_list(struct nexthop *old, struct nexthop *new,
>  			      struct netlink_ext_ack *extack)
>  {
>  	struct fib6_info *f6i;
> +	int err = 0;
>  
>  	if (list_empty(&old->f6i_list))
>  		return 0;
>  
> +	spin_lock_bh(&old->lock);
>  	list_for_each_entry(f6i, &old->f6i_list, nh_list) {
> -		if (check_src_addr(&f6i->fib6_src.addr, extack) < 0)
> -			return -EINVAL;
> +		err = check_src_addr(&f6i->fib6_src.addr, extack);
> +		if (err)
> +			break;
>  	}
> +	spin_unlock_bh(&old->lock);
> +
> +	if (err)
> +		return err;
>  
>  	return fib6_check_nexthop(new, NULL, extack);
>  }
> @@ -2538,8 +2545,10 @@ static void __nexthop_replace_notify(struct net *net, struct nexthop *nh,
>  			fi->nh_updated = false;
>  	}
>  
> +	spin_lock_bh(&nh->lock);
>  	list_for_each_entry(f6i, &nh->f6i_list, nh_list)
>  		fib6_rt_update(net, f6i, info);
> +	spin_unlock_bh(&nh->lock);

I read the feedback from Sashiko, but even if we convert this list to
RCU we would need GFP_ATOMIC. I don't find __nexthop_replace_notify()
too interesting given that modern systems unset
net.ipv4.nexthop_compat_mode, which means that this function is never
called.

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

>  }
>  
>  /* send RTM_NEWROUTE with REPLACE flag set for all FIB entries
> -- 
> 2.43.0
> 

Re: [PATCH net 2/2] nexthop: avoid unlocked f6i_list walk in nh_rt_cache_flush

[Ido Schimmel]

On Wed, Jul 22, 2026 at 12:29:51AM +0000, Xiang Mei (Microsoft) wrote:
> nh_rt_cache_flush() walks nh->f6i_list during an RTNL-serialized nexthop
> replace without holding nh->lock, racing the unlocked IPv6 route
> add/delete that mutate the list under nh->lock and free fib6_info
> entries (nh_rt_cache_flush() is inlined into rtm_new_nexthop()):
> 
>   BUG: KASAN: slab-use-after-free in nh_rt_cache_flush (net/ipv4/nexthop.c:2243)
>   Read of size 8 at addr ffff888012953e18 by task exploit/146
>    nh_rt_cache_flush (net/ipv4/nexthop.c:2243)
>    replace_nexthop (net/ipv4/nexthop.c:2610)
>    rtm_new_nexthop (net/ipv4/nexthop.c:3323)
>    rtnetlink_rcv_msg (net/core/rtnetlink.c:7076)
> 
> Unlike the other f6i_list walks, this one bumps each route's sernum via
> fib6_update_sernum_upto_root(), which needs tb6_lock; taking nh->lock
> around it would invert the established tb6_lock -> nh->lock order and
> deadlock. As the only purpose is to invalidate cached dsts, bump the
> IPv6 sernum for the whole netns with rt_genid_bump_ipv6() instead,
> mirroring the rt_cache_flush() already done for IPv4 just above.
> 
> Fixes: 081efd18326e ("ipv6: Protect nh->f6i_list with spinlock and flag.")
> Reported-by: AutonomousCodeSecurity@microsoft.com
> Signed-off-by: Xiang Mei (Microsoft) <xmei5@asu.edu>
> ---
>  net/ipv4/nexthop.c | 12 ++++++------
>  1 file changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
> index c1d3b7d7100d..1e9ccac6bf0f 100644
> --- a/net/ipv4/nexthop.c
> +++ b/net/ipv4/nexthop.c
> @@ -2240,18 +2240,18 @@ static void remove_one_nexthop(struct net *net, struct nexthop *nh,
>  static void nh_rt_cache_flush(struct net *net, struct nexthop *nh,
>  			      struct nexthop *replaced_nh)
>  {
> -	struct fib6_info *f6i;
>  	struct nh_group *nhg;
> +	bool have_f6i;
>  	int i;
>  
>  	if (!list_empty(&nh->fi_list))
>  		rt_cache_flush(net);
>  
> -	list_for_each_entry(f6i, &nh->f6i_list, nh_list) {
> -		spin_lock_bh(&f6i->fib6_table->tb6_lock);
> -		fib6_update_sernum_upto_root(net, f6i);
> -		spin_unlock_bh(&f6i->fib6_table->tb6_lock);
> -	}
> +	spin_lock_bh(&nh->lock);
> +	have_f6i = !list_empty(&nh->f6i_list);
> +	spin_unlock_bh(&nh->lock);
> +	if (have_f6i)
> +		rt_genid_bump_ipv6(net);

Can you look into converting this list to RCU in net-next so that we
wouldn't need rt_genid_bump_ipv6() and instead restore
fib6_update_sernum_upto_root()?

Note that rt_genid_bump_ipv6() is much more expensive than
rt_cache_flush().

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

>  
>  	/* if an IPv6 group was replaced, we have to release all old
>  	 * dsts to make sure all refcounts are released
> -- 
> 2.43.0
>