[PATCH] ipv6: sr: fix 16-byte kernel heap leak in seg6_genl_set_tunsrc()

Hui Peng posted 1 patch 5 days ago
[PATCH] ipv6: sr: fix 16-byte kernel heap leak in seg6_genl_set_tunsrc()
Posted by Hui Peng 5 days ago
seg6_genl_policy[SEG6_ATTR_DST] uses NLA_BINARY with
.len = sizeof(struct in6_addr), which only caps the maximum attribute
length and allows 0-byte SEG6_ATTR_DST attributes.

seg6_genl_set_tunsrc() then unconditionally copies
sizeof(struct in6_addr) (16 bytes) from
nla_data(info->attrs[SEG6_ATTR_DST]) into sdata->tun_src via kmemdup(),
reading 16 bytes of uninitialized skb->head heap memory past skb->tail
and exposing it to userspace via SEG6_CMD_GET_TUNSRC.

Enforce NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)) in
seg6_genl_policy and validate nla_len(info->attrs[SEG6_ATTR_DST]) in
seg6_genl_set_tunsrc().

Fixes: 915d7e5e5930 ("ipv6: sr: add code base for control plane support of SR-IPv6")
Assisted-by: LLM
Signed-off-by: Hui Peng <benquike@gmail.com>
---
diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c
--- a/net/ipv6/seg6.c
+++ b/net/ipv6/seg6.c
@@ -138,8 +138,8 @@ out:
 static struct genl_family seg6_genl_family;
 
 static const struct nla_policy seg6_genl_policy[SEG6_ATTR_MAX + 1] = {
-	[SEG6_ATTR_DST]				= { .type = NLA_BINARY,
-		.len = sizeof(struct in6_addr) },
+	[SEG6_ATTR_DST]				=
+		NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)),
 	[SEG6_ATTR_DSTLEN]			= { .type = NLA_S32, },
 	[SEG6_ATTR_HMACKEYID]		= { .type = NLA_U32, },
 	[SEG6_ATTR_SECRET]			= { .type = NLA_BINARY, },
@@ -242,7 +242,8 @@ static int seg6_genl_set_tunsrc(struct sk_buff *skb, struct genl_info *info)
 
 	sdata = seg6_pernet(net);
 
-	if (!info->attrs[SEG6_ATTR_DST])
+	if (!info->attrs[SEG6_ATTR_DST] ||
+	    nla_len(info->attrs[SEG6_ATTR_DST]) != sizeof(struct in6_addr))
 		return -EINVAL;
 
 	val = nla_data(info->attrs[SEG6_ATTR_DST]);
-- 
2.43.0
Re: [PATCH] ipv6: sr: fix 16-byte kernel heap leak in seg6_genl_set_tunsrc()
Posted by Hangbin Liu 4 days, 14 hours ago
Hi Hui Peng,
On Sat, Sep 19, 2026 at 08:48:06PM +0000, Hui Peng wrote:
> seg6_genl_policy[SEG6_ATTR_DST] uses NLA_BINARY with
> .len = sizeof(struct in6_addr), which only caps the maximum attribute
> length and allows 0-byte SEG6_ATTR_DST attributes.
> 
> seg6_genl_set_tunsrc() then unconditionally copies
> sizeof(struct in6_addr) (16 bytes) from
> nla_data(info->attrs[SEG6_ATTR_DST]) into sdata->tun_src via kmemdup(),
> reading 16 bytes of uninitialized skb->head heap memory past skb->tail
> and exposing it to userspace via SEG6_CMD_GET_TUNSRC.
> 
> Enforce NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)) in
> seg6_genl_policy and validate nla_len(info->attrs[SEG6_ATTR_DST]) in
> seg6_genl_set_tunsrc().
> 
> Fixes: 915d7e5e5930 ("ipv6: sr: add code base for control plane support of SR-IPv6")
> Assisted-by: LLM
> Signed-off-by: Hui Peng <benquike@gmail.com>
> ---
> diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c
> --- a/net/ipv6/seg6.c
> +++ b/net/ipv6/seg6.c
> @@ -138,8 +138,8 @@ out:
>  static struct genl_family seg6_genl_family;
>  
>  static const struct nla_policy seg6_genl_policy[SEG6_ATTR_MAX + 1] = {
> -	[SEG6_ATTR_DST]				= { .type = NLA_BINARY,
> -		.len = sizeof(struct in6_addr) },
> +	[SEG6_ATTR_DST]				=
> +		NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)),

After you setting the policy here.

>  	[SEG6_ATTR_DSTLEN]			= { .type = NLA_S32, },
>  	[SEG6_ATTR_HMACKEYID]		= { .type = NLA_U32, },
>  	[SEG6_ATTR_SECRET]			= { .type = NLA_BINARY, },
> @@ -242,7 +242,8 @@ static int seg6_genl_set_tunsrc(struct sk_buff *skb, struct genl_info *info)
>  
>  	sdata = seg6_pernet(net);
>  
> -	if (!info->attrs[SEG6_ATTR_DST])
> +	if (!info->attrs[SEG6_ATTR_DST] ||
> +	    nla_len(info->attrs[SEG6_ATTR_DST]) != sizeof(struct in6_addr))
>  		return -EINVAL;

There is not need to re-check the length again.

Thanks
Hangbin
[PATCH net v2] ipv6: sr: enforce exact attribute length for SEG6_ATTR_DST
Posted by Hui Peng 3 days, 17 hours ago
In seg6_genl_policy, SEG6_ATTR_DST is defined with .type = NLA_BINARY and
.len = sizeof(struct in6_addr). For NLA_BINARY, .len only enforces the
maximum payload length and permits shorter payloads (e.g., 0 bytes).
When seg6_genl_set_tunsrc() copies sizeof(struct in6_addr) bytes via
kmemdup(val, sizeof(*val), GFP_KERNEL), a short SEG6_ATTR_DST attribute
triggers a 16-byte out-of-bounds read past skb->tail into uninitialized
skb->head memory, which is stored in sdata->tun_src and leaked back to
userspace via SEG6_CMD_GET_TUNSRC.

Switch SEG6_ATTR_DST in seg6_genl_policy to
NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)) so that generic netlink
validation rejects any attribute whose length is not exactly
sizeof(struct in6_addr) with -ERANGE.

Tested in QEMU against Linux 7.3.0-rc3 by sending a SEG6_CMD_SET_TUNSRC
Generic Netlink message with a 0-byte SEG6_ATTR_DST attribute followed
by SEG6_CMD_GET_TUNSRC. On the unfixed kernel, SEG6_CMD_SET_TUNSRC
succeeds (err = 0) and SEG6_CMD_GET_TUNSRC leaks 16 bytes of
uninitialized kernel heap memory (tun_src =
836a61ecc4d25a1042a8d60411cfb378); with this patch applied,
SEG6_CMD_SET_TUNSRC is rejected by netlink policy validation with
-ERANGE (-34) and tun_src remains zeroed.

Fixes: 915d7e5e5930 ("ipv6: sr: add code base for control plane support of SR-IPv6")
Cc: stable@vger.kernel.org
Assisted-by: LLM
Signed-off-by: Hui Peng <benquike@gmail.com>
---
Changes in v2:
- Drop the redundant nla_len(info->attrs[SEG6_ATTR_DST]) !=
  sizeof(struct in6_addr) check in seg6_genl_set_tunsrc() since
  NLA_POLICY_EXACT_LEN() in seg6_genl_policy already enforces the exact
  length, as pointed out by Hangbin Liu.

 net/ipv6/seg6.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c
index 62a7eb779202..8c2b156c227a 100644
--- a/net/ipv6/seg6.c
+++ b/net/ipv6/seg6.c
@@ -138,8 +138,8 @@ void seg6_icmp_srh(struct sk_buff *skb, struct inet6_skb_parm *opt)
 static struct genl_family seg6_genl_family;
 
 static const struct nla_policy seg6_genl_policy[SEG6_ATTR_MAX + 1] = {
-	[SEG6_ATTR_DST]				= { .type = NLA_BINARY,
-		.len = sizeof(struct in6_addr) },
+	[SEG6_ATTR_DST]				=
+		NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)),
 	[SEG6_ATTR_DSTLEN]			= { .type = NLA_S32, },
 	[SEG6_ATTR_HMACKEYID]		= { .type = NLA_U32, },
 	[SEG6_ATTR_SECRET]			= { .type = NLA_BINARY, },
-- 
2.49.0
Re: [PATCH net v2] ipv6: sr: enforce exact attribute length for SEG6_ATTR_DST
Posted by Jakub Kicinski 20 hours ago
On Mon, 21 Sep 2026 04:40:25 +0000 Hui Peng wrote:
> Subject: [PATCH net v2] ipv6: sr: enforce exact attribute length for SEG6_ATTR_DST

Do not send patches in replies to existing threads.
Re: [PATCH net v2] ipv6: sr: enforce exact attribute length for SEG6_ATTR_DST
Posted by Andrea Mayer 22 hours ago
On Mon, 21 Sep 2026 04:40:25 +0000
Hui Peng <benquike@gmail.com> wrote:

> In seg6_genl_policy, SEG6_ATTR_DST is defined with .type = NLA_BINARY and
> .len = sizeof(struct in6_addr). For NLA_BINARY, .len only enforces the
> maximum payload length and permits shorter payloads (e.g., 0 bytes).
> When seg6_genl_set_tunsrc() copies sizeof(struct in6_addr) bytes via
> kmemdup(val, sizeof(*val), GFP_KERNEL), a short SEG6_ATTR_DST attribute
> triggers a 16-byte out-of-bounds read past skb->tail into uninitialized
> skb->head memory, which is stored in sdata->tun_src and leaked back to
> userspace via SEG6_CMD_GET_TUNSRC.
> 
> Switch SEG6_ATTR_DST in seg6_genl_policy to
> NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)) so that generic netlink
> validation rejects any attribute whose length is not exactly
> sizeof(struct in6_addr) with -ERANGE.
> 
> Tested in QEMU against Linux 7.3.0-rc3 by sending a SEG6_CMD_SET_TUNSRC
> Generic Netlink message with a 0-byte SEG6_ATTR_DST attribute followed
> by SEG6_CMD_GET_TUNSRC. On the unfixed kernel, SEG6_CMD_SET_TUNSRC
> succeeds (err = 0) and SEG6_CMD_GET_TUNSRC leaks 16 bytes of
> uninitialized kernel heap memory (tun_src =
> 836a61ecc4d25a1042a8d60411cfb378); with this patch applied,
> SEG6_CMD_SET_TUNSRC is rejected by netlink policy validation with
> -ERANGE (-34) and tun_src remains zeroed.
> 
> Fixes: 915d7e5e5930 ("ipv6: sr: add code base for control plane support of SR-IPv6")
> Cc: stable@vger.kernel.org
> Assisted-by: LLM
> Signed-off-by: Hui Peng <benquike@gmail.com>
> ---
> Changes in v2:
> - Drop the redundant nla_len(info->attrs[SEG6_ATTR_DST]) !=
>   sizeof(struct in6_addr) check in seg6_genl_set_tunsrc() since
>   NLA_POLICY_EXACT_LEN() in seg6_genl_policy already enforces the exact
>   length, as pointed out by Hangbin Liu.
> 
>  net/ipv6/seg6.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

Thanks,
Andrea

Reviewed-by: Andrea Mayer <andrea.mayer@uniroma2.it>
Re: [PATCH net v2] ipv6: sr: enforce exact attribute length for SEG6_ATTR_DST
Posted by Justin Iurman 2 days, 23 hours ago
On 9/21/26 06:40, Hui Peng wrote:
> In seg6_genl_policy, SEG6_ATTR_DST is defined with .type = NLA_BINARY and
> .len = sizeof(struct in6_addr). For NLA_BINARY, .len only enforces the
> maximum payload length and permits shorter payloads (e.g., 0 bytes).
> When seg6_genl_set_tunsrc() copies sizeof(struct in6_addr) bytes via
> kmemdup(val, sizeof(*val), GFP_KERNEL), a short SEG6_ATTR_DST attribute
> triggers a 16-byte out-of-bounds read past skb->tail into uninitialized
> skb->head memory, which is stored in sdata->tun_src and leaked back to
> userspace via SEG6_CMD_GET_TUNSRC.
> 
> Switch SEG6_ATTR_DST in seg6_genl_policy to
> NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)) so that generic netlink
> validation rejects any attribute whose length is not exactly
> sizeof(struct in6_addr) with -ERANGE.
> 
> Tested in QEMU against Linux 7.3.0-rc3 by sending a SEG6_CMD_SET_TUNSRC
> Generic Netlink message with a 0-byte SEG6_ATTR_DST attribute followed
> by SEG6_CMD_GET_TUNSRC. On the unfixed kernel, SEG6_CMD_SET_TUNSRC
> succeeds (err = 0) and SEG6_CMD_GET_TUNSRC leaks 16 bytes of
> uninitialized kernel heap memory (tun_src =
> 836a61ecc4d25a1042a8d60411cfb378); with this patch applied,
> SEG6_CMD_SET_TUNSRC is rejected by netlink policy validation with
> -ERANGE (-34) and tun_src remains zeroed.
> 
> Fixes: 915d7e5e5930 ("ipv6: sr: add code base for control plane support of SR-IPv6")
> Cc: stable@vger.kernel.org
> Assisted-by: LLM
> Signed-off-by: Hui Peng <benquike@gmail.com>
> ---
> Changes in v2:
> - Drop the redundant nla_len(info->attrs[SEG6_ATTR_DST]) !=
>    sizeof(struct in6_addr) check in seg6_genl_set_tunsrc() since
>    NLA_POLICY_EXACT_LEN() in seg6_genl_policy already enforces the exact
>    length, as pointed out by Hangbin Liu.
> 
>   net/ipv6/seg6.c | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c
> index 62a7eb779202..8c2b156c227a 100644
> --- a/net/ipv6/seg6.c
> +++ b/net/ipv6/seg6.c
> @@ -138,8 +138,8 @@ void seg6_icmp_srh(struct sk_buff *skb, struct inet6_skb_parm *opt)
>   static struct genl_family seg6_genl_family;
>   
>   static const struct nla_policy seg6_genl_policy[SEG6_ATTR_MAX + 1] = {
> -	[SEG6_ATTR_DST]				= { .type = NLA_BINARY,
> -		.len = sizeof(struct in6_addr) },
> +	[SEG6_ATTR_DST]				=
> +		NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)),
>   	[SEG6_ATTR_DSTLEN]			= { .type = NLA_S32, },
>   	[SEG6_ATTR_HMACKEYID]		= { .type = NLA_U32, },
>   	[SEG6_ATTR_SECRET]			= { .type = NLA_BINARY, },

Reviewed-by: Justin Iurman <justin.iurman@gmail.com>
Re: [PATCH net v2] ipv6: sr: enforce exact attribute length for SEG6_ATTR_DST
Posted by Hangbin Liu 3 days, 15 hours ago
On Mon, Sep 21, 2026 at 04:40:25AM +0000, Hui Peng wrote:
> In seg6_genl_policy, SEG6_ATTR_DST is defined with .type = NLA_BINARY and
> .len = sizeof(struct in6_addr). For NLA_BINARY, .len only enforces the
> maximum payload length and permits shorter payloads (e.g., 0 bytes).
> When seg6_genl_set_tunsrc() copies sizeof(struct in6_addr) bytes via
> kmemdup(val, sizeof(*val), GFP_KERNEL), a short SEG6_ATTR_DST attribute
> triggers a 16-byte out-of-bounds read past skb->tail into uninitialized
> skb->head memory, which is stored in sdata->tun_src and leaked back to
> userspace via SEG6_CMD_GET_TUNSRC.
> 
> Switch SEG6_ATTR_DST in seg6_genl_policy to
> NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)) so that generic netlink
> validation rejects any attribute whose length is not exactly
> sizeof(struct in6_addr) with -ERANGE.
> 
> Tested in QEMU against Linux 7.3.0-rc3 by sending a SEG6_CMD_SET_TUNSRC
> Generic Netlink message with a 0-byte SEG6_ATTR_DST attribute followed
> by SEG6_CMD_GET_TUNSRC. On the unfixed kernel, SEG6_CMD_SET_TUNSRC
> succeeds (err = 0) and SEG6_CMD_GET_TUNSRC leaks 16 bytes of
> uninitialized kernel heap memory (tun_src =
> 836a61ecc4d25a1042a8d60411cfb378); with this patch applied,
> SEG6_CMD_SET_TUNSRC is rejected by netlink policy validation with
> -ERANGE (-34) and tun_src remains zeroed.
> 
> Fixes: 915d7e5e5930 ("ipv6: sr: add code base for control plane support of SR-IPv6")
> Cc: stable@vger.kernel.org
> Assisted-by: LLM
> Signed-off-by: Hui Peng <benquike@gmail.com>
> ---
> Changes in v2:
> - Drop the redundant nla_len(info->attrs[SEG6_ATTR_DST]) !=
>   sizeof(struct in6_addr) check in seg6_genl_set_tunsrc() since
>   NLA_POLICY_EXACT_LEN() in seg6_genl_policy already enforces the exact
>   length, as pointed out by Hangbin Liu.
> 
>  net/ipv6/seg6.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c
> index 62a7eb779202..8c2b156c227a 100644
> --- a/net/ipv6/seg6.c
> +++ b/net/ipv6/seg6.c
> @@ -138,8 +138,8 @@ void seg6_icmp_srh(struct sk_buff *skb, struct inet6_skb_parm *opt)
>  static struct genl_family seg6_genl_family;
>  
>  static const struct nla_policy seg6_genl_policy[SEG6_ATTR_MAX + 1] = {
> -	[SEG6_ATTR_DST]				= { .type = NLA_BINARY,
> -		.len = sizeof(struct in6_addr) },
> +	[SEG6_ATTR_DST]				=
> +		NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)),
>  	[SEG6_ATTR_DSTLEN]			= { .type = NLA_S32, },
>  	[SEG6_ATTR_HMACKEYID]		= { .type = NLA_U32, },
>  	[SEG6_ATTR_SECRET]			= { .type = NLA_BINARY, },
> -- 
> 2.49.0

Reviewed-by: Hangbin Liu <liuhangbin@kylinos.cn>