From: Debin Zhu <mowenroot@163.com>
To: linux-security-module@vger.kernel.org
Cc: netdev@vger.kernel.org, selinux@vger.kernel.org,
paul@paul-moore.com, linux-kernel@vger.kernel.org,
kuba@kernel.org, Debin Zhu <mowenroot@163.com>,
Bitao Ouyang <1985755126@qq.com>
Subject: [PATCH v2] netlabel: Fix NULL pointer exception caused by CALIPSO on IPv4 sockets
Date: Sun, 30 Mar 2025 18:33:39 +0800 [thread overview]
Message-ID: <20250330103339.30794-1-mowenroot@163.com> (raw)
Vulnerability Description:
From Linux Kernel v4.0 to the latest version,
a type confusion issue exists in the `netlbl_conn_setattr`
function (`net/netlabel/netlabel_kapi.c`) within SELinux,
which can lead to a local DoS attack.
When calling `netlbl_conn_setattr`,
`addr->sa_family` is used to determine the function behavior.
If `sk` is an IPv4 socket,
but the `connect` function is called with an IPv6 address,
the function `calipso_sock_setattr()` is triggered.
Inside this function, the following code is executed:
sk_fullsock(__sk) ? inet_sk(__sk)->pinet6 : NULL;
Since `sk` is an IPv4 socket, `pinet6` is `NULL`,
leading to a null pointer dereference and triggering a DoS attack.
<TASK>
calipso_sock_setattr+0x4f/0x80 net/netlabel/netlabel_calipso.c:557
netlbl_conn_setattr+0x12a/0x390 net/netlabel/netlabel_kapi.c:1152
selinux_netlbl_socket_connect_helper
selinux_netlbl_socket_connect_locked+0xf5/0x1d0
selinux_netlbl_socket_connect+0x22/0x40 security/selinux/netlabel.c:611
selinux_socket_connect+0x60/0x80 security/selinux/hooks.c:4923
security_socket_connect+0x71/0xb0 security/security.c:2260
__sys_connect_file+0xa4/0x190 net/socket.c:2007
__sys_connect+0x145/0x170 net/socket.c:2028
__do_sys_connect net/socket.c:2038 [inline]
__se_sys_connect net/socket.c:2035 [inline]
__x64_sys_connect+0x6e/0xb0 net/socket.c:2035
do_syscall_x64 arch/x86/entry/common.c:51
Affected Versions:
- Linux 4.0 - Latest Linux Kernel version
Reproduction Steps:
Use the `netlabelctl` tool and
run the following commands to trigger the vulnerability:
netlabelctl map del default
netlabelctl cipsov4 add pass doi:8 tags:1
netlabelctl map add default address:192.168.1.0/24 protocol:cipsov4,8
netlabelctl calipso add pass doi:7
netlabelctl map add default address:2001:db8::1/32 protocol:calipso,7
Then, execute the following PoC code:
int sockfd = socket(AF_INET, SOCK_STREAM, 0);
struct sockaddr_in6 server_addr = {0};
server_addr.sin6_family = AF_INET6;
server_addr.sin6_port = htons(8080);
const char *ipv6_str = "2001:db8::1";
inet_pton(AF_INET6, ipv6_str, &server_addr.sin6_addr);
connect(sockfd, (struct sockaddr*)&server_addr, sizeof(server_addr));
Suggested Fix:
When using an IPv4 address on an IPv6 UDP/datagram socket,
the operation will invoke the IPv4 datagram code through
the IPv6 datagram code and execute successfully.
It is necessary to check whether the `pinet6` pointer
returned by `inet6_sk()` is NULL; otherwise,
unexpected issues may occur.
Signed-off-by: Debin Zhu <mowenroot@163.com>
Signed-off-by: Bitao Ouyang <1985755126@qq.com>
---
net/ipv6/calipso.c | 23 +++++++++++++++++++----
1 file changed, 19 insertions(+), 4 deletions(-)
diff --git a/net/ipv6/calipso.c b/net/ipv6/calipso.c
index dbcea9fee..a8a8736df 100644
--- a/net/ipv6/calipso.c
+++ b/net/ipv6/calipso.c
@@ -1072,8 +1072,13 @@ static int calipso_sock_getattr(struct sock *sk,
struct ipv6_opt_hdr *hop;
int opt_len, len, ret_val = -ENOMSG, offset;
unsigned char *opt;
- struct ipv6_txoptions *txopts = txopt_get(inet6_sk(sk));
+ struct ipv6_pinfo *pinfo = inet6_sk(sk);
+ struct ipv6_txoptions *txopts;
+ if (!pinfo)
+ return -EAFNOSUPPORT;
+
+ txopts = txopt_get(pinfo);
if (!txopts || !txopts->hopopt)
goto done;
@@ -1125,8 +1130,13 @@ static int calipso_sock_setattr(struct sock *sk,
{
int ret_val;
struct ipv6_opt_hdr *old, *new;
- struct ipv6_txoptions *txopts = txopt_get(inet6_sk(sk));
-
+ struct ipv6_pinfo *pinfo = inet6_sk(sk);
+ struct ipv6_txoptions *txopts;
+
+ if (!pinfo)
+ return -EAFNOSUPPORT;
+
+ txopts = txopt_get(pinfo);
old = NULL;
if (txopts)
old = txopts->hopopt;
@@ -1153,8 +1163,13 @@ static int calipso_sock_setattr(struct sock *sk,
static void calipso_sock_delattr(struct sock *sk)
{
struct ipv6_opt_hdr *new_hop;
- struct ipv6_txoptions *txopts = txopt_get(inet6_sk(sk));
+ struct ipv6_pinfo *pinfo = inet6_sk(sk);
+ struct ipv6_txoptions *txopts;
+ if (!pinfo)
+ return;
+
+ txopts = txopt_get(pinfo);
if (!txopts || !txopts->hopopt)
goto done;
--
mowenroot@163.com
next reply other threads:[~2025-03-30 10:34 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-30 10:33 Debin Zhu [this message]
-- strict thread matches above, loose matches on Subject: below --
2025-03-28 16:02 [PATCH] " Paul Moore
2025-03-30 10:40 ` [PATCH v2] " Debin Zhu
2025-04-01 10:17 ` Paolo Abeni
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250330103339.30794-1-mowenroot@163.com \
--to=mowenroot@163.com \
--cc=1985755126@qq.com \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-security-module@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=paul@paul-moore.com \
--cc=selinux@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®