Skip to contentSkip to conversation

fix syzkaller splats reported by Christoph

PATCH In review Florian Westphal 4 participants 10 messages Jan 28, 2020

Subsystems (3)
lore
Share & tools
lore

Florian Westphal

author
#0 lore raw

v4 cover fix syzkaller splats reported by Christoph

This fixes the syzkaller splats with reproducers that Christoph forwarded to the mailing list, including the one wrt. slab oob in syncookie mode.

First patch moves free of the cached extension pointer to the destroy callback to make sure no other call path can allocate a new one. Actually net-next should be fine in this regard (no msk work queue yet), but as Paolo pointed out it looks bogus when 3rd patch calls it with mptcp lock already released.

2nd patch fixes a crash with set/getsockopt interface and resolves a lockdep splat due to calls to rtnl_lock while already holding the msk lock.

I resolved both in the same patch because the fixes need to touch the same spot.

3rd patch fixes another lockdep splat, this time during mptcp_close but again related to rtnl_lock.

Last patch avoid OOB access when syncookies are used.
This should be resolved differently, as mptcp v1 can use 4th ack to recover state, but that will be more intrusive so this fix downgrades to TCP instead.

See patches for more detailed explanation.

 include/linux/tcp.h   |    2 --
 net/ipv4/syncookies.c |    4 ++++
 net/ipv4/tcp_input.c  |    3 +++
 net/ipv6/syncookies.c |    3 +++
 net/mptcp/protocol.c  |   48 ++++++++++++++++++++++++++++--------------------
 net/mptcp/subflow.c   |    5 ++++-
 6 files changed, 42 insertions(+), 23 deletions(-)
author
#1 lore raw

PATCH 1/4 mptcp: protocol: defer freeing of cached ext until last moment

access to msk->cached_ext is only legal if the msk is locked or all concurrent accesses are impossible.

Furthermore, once we start to tear down, we must make sure nothing else can step in and allocate a new cached ext.

So place this code in the destroy callback where it belongs.

Signed-off-by Florian Westphal <fw(a)strlen.de>

---
patch is new in this round.

 net/mptcp/protocol.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
net/mptcp/protocol.c +4 -2
index 39fdca79ce90..f1b1160dbbb4 100644
@@ -654,8 +654,6 @@ static void __mptcp_close(struct sock *sk, long timeout)
 		__mptcp_close_ssk(sk, ssk, subflow, timeout);
 	}
 
-	if (msk->cached_ext)
-		__skb_ext_put(msk->cached_ext);
 	release_sock(sk);
 	sk_common_release(sk);
 }
@@ -776,6 +774,10 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
 
 static void mptcp_destroy(struct sock *sk)
 {
+	struct mptcp_sock *msk = mptcp_sk(sk);
+
+	if (msk->cached_ext)
+		__skb_ext_put(msk->cached_ext);
 }
 
 static int mptcp_setsockopt(struct sock *sk, int level, int optname,
Signature · 1 line
2.24.1
author
#2 lore raw

PATCH v4 2/4 mptcp: protocol: fix panic on user pointer access

In reply to Florian Westphal

Its not possible to call the kernel_(s|g)etsockopt functions here, the address points to user memory:

General protection fault in user access. Non-canonical address?

WARNING: CPU: 1 PID: 5352 at arch/x86/mm/extable.c:77 ex_handler_uaccess+0xba/0xe0 arch/x86/mm/extable.c:77
Kernel panic - not syncing: panic_on_warn set ...
[..]
Call Trace:
 fixup_exception+0x9d/0xcd arch/x86/mm/extable.c:178
 general_protection+0x2d/0x40 arch/x86/entry/entry_64.S:1202
 do_ip_getsockopt+0x1f6/0x1860 net/ipv4/ip_sockglue.c:1323
 ip_getsockopt+0x87/0x1c0 net/ipv4/ip_sockglue.c:1561
 tcp_getsockopt net/ipv4/tcp.c:3691 [inline]
 tcp_getsockopt+0x8c/0xd0 net/ipv4/tcp.c:3685
 kernel_getsockopt+0x121/0x1f0 net/socket.c:3736
 mptcp_getsockopt+0x69/0x90 net/mptcp/protocol.c:830
 __sys_getsockopt+0x13a/0x220 net/socket.c:2175

v2: also fix a circular locking dependency reported by Christoph:
release the mptcp-level lock before calling set/getsockopt on the underlying tcp subflow socket, else this yields:
v4: no need to check for SOL_SOCKET again (Paolo)

WARNING: possible circular locking dependency detected
5.5.0-rc6 #2 Not tainted
------------------------------------------------------
syz-executor.0/16334 is trying to acquire lock:
ffffffff84f7a080 (rtnl_mutex){+.+.}, at: do_ip_setsockopt.isra.0+0x277/0x3820 net/ipv4/ip_sockglue.c:644 but task is already holding lock:
ffff888116503b90 (sk_lock-AF_INET){+.+.}, at: lock_sock include/net/sock.h:1516 [inline] ffff888116503b90 (sk_lock-AF_INET){+.+.}, at: mptcp_setsockopt+0x28/0x90 net/mptcp/protocol.c:1284

which lock already depends on the new lock. the existing dependency chain (in reverse order) is:

-> #1 (sk_lock-AF_INET){+.+.}:

        lock_sock_nested+0xca/0x120 net/core/sock.c:2944
        lock_sock include/net/sock.h:1516 [inline]
        do_ip_setsockopt.isra.0+0x281/0x3820 net/ipv4/ip_sockglue.c:645
        ip_setsockopt+0x44/0xf0 net/ipv4/ip_sockglue.c:1248
        udp_setsockopt+0x5d/0xa0 net/ipv4/udp.c:2639
        __sys_setsockopt+0x152/0x240 net/socket.c:2130
        __do_sys_setsockopt net/socket.c:2146 [inline]
        __se_sys_setsockopt net/socket.c:2143 [inline]
        __x64_sys_setsockopt+0xba/0x150 net/socket.c:2143
        do_syscall_64+0xbd/0x5b0 arch/x86/entry/common.c:294
        entry_SYSCALL_64_after_hwframe+0x49/0xbe

-> #0 (rtnl_mutex){+.+.}:

        check_prev_add kernel/locking/lockdep.c:2475 [inline]
        check_prevs_add kernel/locking/lockdep.c:2580 [inline]
        validate_chain kernel/locking/lockdep.c:2970 [inline]
        __lock_acquire+0x1fb2/0x4680 kernel/locking/lockdep.c:3954
        lock_acquire+0x127/0x330 kernel/locking/lockdep.c:4484
        __mutex_lock_common kernel/locking/mutex.c:956 [inline]
        __mutex_lock+0x158/0x1340 kernel/locking/mutex.c:1103
        do_ip_setsockopt.isra.0+0x277/0x3820 net/ipv4/ip_sockglue.c:644
        ip_setsockopt+0x44/0xf0 net/ipv4/ip_sockglue.c:1248
        tcp_setsockopt net/ipv4/tcp.c:3159 [inline]
        tcp_setsockopt+0x8c/0xd0 net/ipv4/tcp.c:3153
        kernel_setsockopt+0x121/0x1f0 net/socket.c:3767
        mptcp_setsockopt+0x69/0x90 net/mptcp/protocol.c:1288
        __sys_setsockopt+0x152/0x240 net/socket.c:2130
        __do_sys_setsockopt net/socket.c:2146 [inline]
        __se_sys_setsockopt net/socket.c:2143 [inline]
        __x64_sys_setsockopt+0xba/0x150 net/socket.c:2143
        do_syscall_64+0xbd/0x5b0 arch/x86/entry/common.c:294
        entry_SYSCALL_64_after_hwframe+0x49/0xbe

other info that might help us debug this:

  Possible unsafe locking scenario:
        CPU0                    CPU1
        ----                    ----
   lock(sk_lock-AF_INET);
                                lock(rtnl_mutex);
                                lock(sk_lock-AF_INET);

lock(rtnl_mutex);

v4: no need to handle SOL_SOCKET, sys_s/getsockopt handles this already.

Fixes 717e79c867ca5 ("mptcp: Add setsockopt()/getsockopt() socket operations")
Reported-by Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by Florian Westphal <fw(a)strlen.de>
---
 net/mptcp/protocol.c | 34 ++++++++++++++++++----------------
 1 file changed, 18 insertions(+), 16 deletions(-)
net/mptcp/protocol.c +18 -16
index f1b1160dbbb4..432265f84a65 100644
@@ -780,17 +780,19 @@ static void mptcp_destroy(struct sock *sk)
 		__skb_ext_put(msk->cached_ext);
 }
 
+static int subflow_setsockopt(struct socket *sock, int level, int optname,
+			       char __user *optval, unsigned int optlen)
+{
+	return sock->ops->setsockopt(sock, level, optname, optval, optlen);
+}
+
 static int mptcp_setsockopt(struct sock *sk, int level, int optname,
-			    char __user *uoptval, unsigned int optlen)
+			    char __user *optval, unsigned int optlen)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
-	char __kernel *optval;
 	int ret = -EOPNOTSUPP;
 	struct socket *ssock;
 
-	/* will be treated as __user in tcp_setsockopt */
-	optval = (char __kernel __force *)uoptval;
-
 	pr_debug("msk=%p", msk);
 
 	/* @@ the meaning of setsockopt() when the socket is connected and
@@ -798,28 +800,28 @@ static int mptcp_setsockopt(struct sock *sk, int level, int optname,
 	 */
 	lock_sock(sk);
 	ssock = __mptcp_socket_create(msk, MPTCP_SAME_STATE);
+	release_sock(sk);
 	if (!IS_ERR(ssock)) {
 		pr_debug("subflow=%p", ssock->sk);
-		ret = kernel_setsockopt(ssock, level, optname, optval, optlen);
+		ret = subflow_setsockopt(ssock, level, optname, optval, optlen);
 	}
-	release_sock(sk);
 
 	return ret;
 }
 
+static int subflow_getsockopt(struct socket *sock, int level, int optname,
+			       char __user *optval, int __user *optlen)
+{
+	return sock->ops->getsockopt(sock, level, optname, optval, optlen);
+}
+
 static int mptcp_getsockopt(struct sock *sk, int level, int optname,
-			    char __user *uoptval, int __user *uoption)
+			    char __user *optval, int __user *option)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
-	char __kernel *optval;
 	int ret = -EOPNOTSUPP;
-	int __kernel *option;
 	struct socket *ssock;
 
-	/* will be treated as __user in tcp_getsockopt */
-	optval = (char __kernel __force *)uoptval;
-	option = (int __kernel __force *)uoption;
-
 	pr_debug("msk=%p", msk);
 
 	/* @@ the meaning of getsockopt() when the socket is connected and
@@ -827,11 +829,11 @@ static int mptcp_getsockopt(struct sock *sk, int level, int optname,
 	 */
 	lock_sock(sk);
 	ssock = __mptcp_socket_create(msk, MPTCP_SAME_STATE);
+	release_sock(sk);
 	if (!IS_ERR(ssock)) {
 		pr_debug("subflow=%p", ssock->sk);
-		ret = kernel_getsockopt(ssock, level, optname, optval, option);
+		ret = subflow_getsockopt(ssock, level, optname, optval, option);
 	}
-	release_sock(sk);
 
 	return ret;
 }
Signature · 1 line
2.24.1
#3 lore raw

[MPTCP] Re: [PATCH net-next v4 2/4] mptcp: protocol: fix panic on user pointer access

Hi,

Quoted 119 lines · Florian Westphal

Its not possible to call the kernel_(s|g)etsockopt functions here, the address points to user memory:

General protection fault in user access. Non-canonical address?

WARNING: CPU: 1 PID: 5352 at arch/x86/mm/extable.c:77 ex_handler_uaccess+0xba/0xe0 arch/x86/mm/extable.c:77
Kernel panic - not syncing: panic_on_warn set ...
[..]
Call Trace:
 fixup_exception+0x9d/0xcd arch/x86/mm/extable.c:178
 general_protection+0x2d/0x40 arch/x86/entry/entry_64.S:1202
 do_ip_getsockopt+0x1f6/0x1860 net/ipv4/ip_sockglue.c:1323
 ip_getsockopt+0x87/0x1c0 net/ipv4/ip_sockglue.c:1561
 tcp_getsockopt net/ipv4/tcp.c:3691 [inline]
 tcp_getsockopt+0x8c/0xd0 net/ipv4/tcp.c:3685
 kernel_getsockopt+0x121/0x1f0 net/socket.c:3736
 mptcp_getsockopt+0x69/0x90 net/mptcp/protocol.c:830
 __sys_getsockopt+0x13a/0x220 net/socket.c:2175

v2: also fix a circular locking dependency reported by Christoph:
release the mptcp-level lock before calling set/getsockopt on the underlying tcp subflow socket, else this yields:
v4: no need to check for SOL_SOCKET again (Paolo)

WARNING: possible circular locking dependency detected
5.5.0-rc6 #2 Not tainted
------------------------------------------------------
syz-executor.0/16334 is trying to acquire lock:
ffffffff84f7a080 (rtnl_mutex){+.+.}, at: do_ip_setsockopt.isra.0+0x277/0x3820 net/ipv4/ip_sockglue.c:644 but task is already holding lock:
ffff888116503b90 (sk_lock-AF_INET){+.+.}, at: lock_sock include/net/sock.h:1516 [inline] ffff888116503b90 (sk_lock-AF_INET){+.+.}, at: mptcp_setsockopt+0x28/0x90 net/mptcp/protocol.c:1284

which lock already depends on the new lock. the existing dependency chain (in reverse order) is:

-> #1 (sk_lock-AF_INET){+.+.}:

        lock_sock_nested+0xca/0x120 net/core/sock.c:2944
        lock_sock include/net/sock.h:1516 [inline]
        do_ip_setsockopt.isra.0+0x281/0x3820 net/ipv4/ip_sockglue.c:645
        ip_setsockopt+0x44/0xf0 net/ipv4/ip_sockglue.c:1248
        udp_setsockopt+0x5d/0xa0 net/ipv4/udp.c:2639
        __sys_setsockopt+0x152/0x240 net/socket.c:2130
        __do_sys_setsockopt net/socket.c:2146 [inline]
        __se_sys_setsockopt net/socket.c:2143 [inline]
        __x64_sys_setsockopt+0xba/0x150 net/socket.c:2143
        do_syscall_64+0xbd/0x5b0 arch/x86/entry/common.c:294
        entry_SYSCALL_64_after_hwframe+0x49/0xbe

-> #0 (rtnl_mutex){+.+.}:

        check_prev_add kernel/locking/lockdep.c:2475 [inline]
        check_prevs_add kernel/locking/lockdep.c:2580 [inline]
        validate_chain kernel/locking/lockdep.c:2970 [inline]
        __lock_acquire+0x1fb2/0x4680 kernel/locking/lockdep.c:3954
        lock_acquire+0x127/0x330 kernel/locking/lockdep.c:4484
        __mutex_lock_common kernel/locking/mutex.c:956 [inline]
        __mutex_lock+0x158/0x1340 kernel/locking/mutex.c:1103
        do_ip_setsockopt.isra.0+0x277/0x3820 net/ipv4/ip_sockglue.c:644
        ip_setsockopt+0x44/0xf0 net/ipv4/ip_sockglue.c:1248
        tcp_setsockopt net/ipv4/tcp.c:3159 [inline]
        tcp_setsockopt+0x8c/0xd0 net/ipv4/tcp.c:3153
        kernel_setsockopt+0x121/0x1f0 net/socket.c:3767
        mptcp_setsockopt+0x69/0x90 net/mptcp/protocol.c:1288
        __sys_setsockopt+0x152/0x240 net/socket.c:2130
        __do_sys_setsockopt net/socket.c:2146 [inline]
        __se_sys_setsockopt net/socket.c:2143 [inline]
        __x64_sys_setsockopt+0xba/0x150 net/socket.c:2143
        do_syscall_64+0xbd/0x5b0 arch/x86/entry/common.c:294
        entry_SYSCALL_64_after_hwframe+0x49/0xbe

other info that might help us debug this:

  Possible unsafe locking scenario:
        CPU0                    CPU1
        ----                    ----
   lock(sk_lock-AF_INET);
                                lock(rtnl_mutex);
                                lock(sk_lock-AF_INET);

lock(rtnl_mutex);

v4: no need to handle SOL_SOCKET, sys_s/getsockopt handles this already.

Fixes 717e79c867ca5 ("mptcp: Add setsockopt()/getsockopt() socket operations")
Reported-by Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by Florian Westphal <fw(a)strlen.de>
---
 net/mptcp/protocol.c | 34 ++++++++++++++++++----------------
 1 file changed, 18 insertions(+), 16 deletions(-)
net/mptcp/protocol.c +7 -5
index f1b1160dbbb4..432265f84a65 100644
@@ -780,17 +780,19 @@ static void mptcp_destroy(struct sock *sk)
 		__skb_ext_put(msk->cached_ext);
 }
 
+static int subflow_setsockopt(struct socket *sock, int level, int optname,
+			       char __user *optval, unsigned int optlen)
+{
+	return sock->ops->setsockopt(sock, level, optname, optval, optlen);
+}
+
 static int mptcp_setsockopt(struct sock *sk, int level, int optname,
-			    char __user *uoptval, unsigned int optlen)
+			    char __user *optval, unsigned int optlen)
 {
 	struct mptcp_sock *msk = mptcp_sk(sk);
-	char __kernel *optval;
 	int ret = -EOPNOTSUPP;
 	struct socket *ssock;
 
-	/* will be treated as __user in tcp_setsockopt */
-	optval = (char __kernel __force *)uoptval;
-
 	pr_debug("msk=%p", msk);
 
 	/* @@ the meaning of setsockopt() when the socket is connected and

net/mptcp/protocol.c static int mptcp_setsockopt(struct sock *sk, int level, int optname, · line 805

Quoted 10 lines
net/mptcp/protocol.c +2 -2
@@ -798,28 +800,28 @@ static int mptcp_setsockopt(struct sock *sk, int level, int optname,
 	 */
 	lock_sock(sk);
 	ssock = __mptcp_socket_create(msk, MPTCP_SAME_STATE);
+	release_sock(sk);
 	if (!IS_ERR(ssock)) {
 		pr_debug("subflow=%p", ssock->sk);
-		ret = kernel_setsockopt(ssock, level, optname, optval, optlen);
+		ret = subflow_setsockopt(ssock, level, optname, optval, optlen);
 	}
-	release_sock(sk);

Can the patched code possibly race with mptcp_close(), so that after releasing the socket lock the subflow is destroyed (by mptcp_close()) and we hit UaF in subflow_setsockopt()?

I fear we need that lookup/get/<op>/put schema here ?!?

Am I missing something here ?!?

Thanks,

Paolo

author
#4 lore raw

PATCH v2 3/4 mptcp: mptcp_close: avoid a lockdep splat when mcast group was joined

In reply to Florian Westphal

syzbot triggered following lockdep splat:

ffffffff82d2cd40 (rtnl_mutex){+.+.}, at: ip_mc_drop_socket+0x52/0x180 but task is already holding lock:
ffff8881187a2310 (sk_lock-AF_INET){+.+.}, at: mptcp_close+0x18/0x30

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:
-> #1 (sk_lock-AF_INET){+.+.}:

       lock_acquire+0xee/0x230
       lock_sock_nested+0x89/0xc0
       do_ip_setsockopt.isra.0+0x335/0x22f0
       ip_setsockopt+0x35/0x60
       tcp_setsockopt+0x5d/0x90
       __sys_setsockopt+0xf3/0x190
       __x64_sys_setsockopt+0x61/0x70
       do_syscall_64+0x72/0x300
       entry_SYSCALL_64_after_hwframe+0x49/0xbe
-> #0 (rtnl_mutex){+.+.}:
       check_prevs_add+0x2b7/0x1210
       __lock_acquire+0x10b6/0x1400
       lock_acquire+0xee/0x230
       __mutex_lock+0x120/0xc70
       ip_mc_drop_socket+0x52/0x180
       inet_release+0x36/0xe0
       __sock_release+0xfd/0x130
       __mptcp_close+0xa8/0x1f0
       inet_release+0x7f/0xe0
       __sock_release+0x69/0x130
       sock_close+0x18/0x20
       __fput+0x179/0x400
       task_work_run+0xd5/0x110
       do_exit+0x685/0x1510
       do_group_exit+0x7e/0x170
       __x64_sys_exit_group+0x28/0x30
       do_syscall_64+0x72/0x300
       entry_SYSCALL_64_after_hwframe+0x49/0xbe

The trigger is:
socket(AF_INET, SOCK_STREAM, 0x106 /* IPPROTO_MPTCP */) = 4
setsockopt(4, SOL_IP, MCAST_JOIN_GROUP, {gr_interface=7, gr_group={sa_family=AF_INET, sin_port=htons(20003), sin_addr=inet_addr("224.0.0.2")}}, 136) = 0 exit(0)

Which results in a call to rtnl_lock while we are holding
the parent mptcp socket lock via
mptcp_close -> lock_sock(msk) -> inet_release -> ip_mc_drop_socket -> rtnl_lock().

From lockdep point of view we thus have both 'rtnl_lock; lock_sock' and 'lock_sock; rtnl_lock'.

Fix this by stealing the msk conn_list and doing the subflow close without holding the msk lock.

Fixes cec37a6e41aae7bf ("mptcp: Handle MP_CAPABLE options for outgoing connections")
Reported-by Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by Florian Westphal <fw(a)strlen.de>

---
v2: rebase to account for new patch 1.

 net/mptcp/protocol.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
net/mptcp/protocol.c +6 -2
index 432265f84a65..6a8f2978b3e5 100644
@@ -644,17 +644,21 @@ static void __mptcp_close(struct sock *sk, long timeout)
 {
 	struct mptcp_subflow_context *subflow, *tmp;
 	struct mptcp_sock *msk = mptcp_sk(sk);
+	LIST_HEAD(conn_list);
 
 	mptcp_token_destroy(msk->token);
 	inet_sk_state_store(sk, TCP_CLOSE);
 
-	list_for_each_entry_safe(subflow, tmp, &msk->conn_list, node) {
+	list_splice_init(&msk->conn_list, &conn_list);
+
+	release_sock(sk);
+
+	list_for_each_entry_safe(subflow, tmp, &conn_list, node) {
 		struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
 
 		__mptcp_close_ssk(sk, ssk, subflow, timeout);
 	}
 
-	release_sock(sk);
 	sk_common_release(sk);
 }
 
Signature · 1 line
2.24.1
author
#5 lore raw

PATCH 4/4 mptcp: handle tcp fallback when using syn cookies

In reply to Florian Westphal

We can't deal with syncookie mode yet, the syncookie rx path will create tcp rsk instead of mptcp rsk, i.e. we get OOB access because we treat tcp rsk as mptcp rsk:

TCP: request_sock_subflow: Possible SYN flooding on port 20002. Sending cookies. Check SNMP counters.

BUG: KASAN: slab-out-of-bounds in subflow_syn_recv_sock+0x451/0x4d0 net/mptcp/subflow.c:191
Read of size 1 at addr ffff8881167bc148 by task syz-executor099/2120
 subflow_syn_recv_sock+0x451/0x4d0 net/mptcp/subflow.c:191
 tcp_get_cookie_sock+0xcf/0x520 net/ipv4/syncookies.c:209
 cookie_v6_check+0x15a5/0x1e90 net/ipv6/syncookies.c:252
 tcp_v6_cookie_check net/ipv6/tcp_ipv6.c:1123 [inline]
 [..]

Bug can be reproduced via "sysctl net.ipv4.tcp_syncookies=2".

Note that MPTCP should work with syncookies (4th ack would carry needed state), but it appears better to sort that out in -next so do tcp fallback for now.

I decided to remove the ifdef on tcp_rsk "is_mptcp" member because if (IS_ENABLED()) is easier to read than "#ifdef IS_ENABLED()/#endif" pair.

Reported-by Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by Florian Westphal <fw(a)strlen.de>

---
patch is new in this set.

 include/linux/tcp.h   | 2 --
 net/ipv4/syncookies.c | 4 ++++
 net/ipv4/tcp_input.c  | 3 +++
 net/ipv6/syncookies.c | 3 +++
 net/mptcp/subflow.c   | 5 ++++-
 5 files changed, 14 insertions(+), 3 deletions(-)
include/linux/tcp.h +0 -2
index 1cf73e6f85ca..3dc964010fef 100644
@@ -148,9 +148,7 @@ struct tcp_request_sock {
 	const struct tcp_request_sock_ops *af_specific;
 	u64				snt_synack; /* first SYNACK sent time */
 	bool				tfo_listener;
-#if IS_ENABLED(CONFIG_MPTCP)
 	bool				is_mptcp;
-#endif
 	u32				txhash;
 	u32				rcv_isn;
 	u32				snt_isn;
net/ipv4/syncookies.c +4 -0
index 345b2b0ff618..9a4f6b16c9bc 100644
@@ -349,6 +349,10 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	req->ts_recent		= tcp_opt.saw_tstamp ? tcp_opt.rcv_tsval : 0;
 	treq->snt_synack	= 0;
 	treq->tfo_listener	= false;
+
+	if (IS_ENABLED(CONFIG_MPTCP))
+		treq->is_mptcp = 0;
+
 	if (IS_ENABLED(CONFIG_SMC))
 		ireq->smc_ok = 0;
 
net/ipv4/tcp_input.c +3 -0
index e8b840a4767e..e325b4506e25 100644
@@ -6637,6 +6637,9 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 
 	af_ops->init_req(req, sk, skb);
 
+	if (IS_ENABLED(CONFIG_MPTCP) && want_cookie)
+		tcp_rsk(req)->is_mptcp = 0;
+
 	if (security_inet_conn_request(sk, skb, req))
 		goto drop_and_free;
 
net/ipv6/syncookies.c +3 -0
index 30915f6f31e3..13235a012388 100644
@@ -178,6 +178,9 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	treq = tcp_rsk(req);
 	treq->tfo_listener = false;
 
+	if (IS_ENABLED(CONFIG_MPTCP))
+		treq->is_mptcp = 0;
+
 	if (security_inet_conn_request(sk, skb, req))
 		goto out_free;
 
net/mptcp/subflow.c +4 -1
index 1662e1178949..cf9bad97a12a 100644
@@ -186,6 +186,9 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 
 	pr_debug("listener=%p, req=%p, conn=%p", listener, req, listener->conn);
 
+	if (tcp_rsk(req)->is_mptcp == 0)
+		goto create_child;
+
 	/* if the sk is MP_CAPABLE, we try to fetch the client key */
 	subflow_req = mptcp_subflow_rsk(req);
 	if (subflow_req->mp_capable) {
@@ -767,7 +770,7 @@ static void subflow_ulp_clone(const struct request_sock *req,
 	struct mptcp_subflow_context *old_ctx = mptcp_subflow_ctx(newsk);
 	struct mptcp_subflow_context *new_ctx;
 
-	if (!subflow_req->mp_capable) {
+	if (!tcp_rsk(req)->is_mptcp || !subflow_req->mp_capable) {
 		subflow_ulp_fallback(newsk, old_ctx);
 		return;
 	}
Signature · 1 line
2.24.1
#6 lore raw

[MPTCP] Re: [PATCH net-next v1 4/4] mptcp: handle tcp fallback when using syn cookies

Hi Florian,

Quoted 13 lines · Florian Westphal

We can't deal with syncookie mode yet, the syncookie rx path will create tcp rsk instead of mptcp rsk, i.e. we get OOB access because we treat tcp rsk as mptcp rsk:

TCP: request_sock_subflow: Possible SYN flooding on port 20002. Sending cookies. Check SNMP counters.

BUG: KASAN: slab-out-of-bounds in subflow_syn_recv_sock+0x451/0x4d0 net/mptcp/subflow.c:191
Read of size 1 at addr ffff8881167bc148 by task syz-executor099/2120
  subflow_syn_recv_sock+0x451/0x4d0 net/mptcp/subflow.c:191
  tcp_get_cookie_sock+0xcf/0x520 net/ipv4/syncookies.c:209
  cookie_v6_check+0x15a5/0x1e90 net/ipv6/syncookies.c:252
  tcp_v6_cookie_check net/ipv6/tcp_ipv6.c:1123 [inline]
  [..]
Quoted 1 line

Bug can be reproduced via "sysctl net.ipv4.tcp_syncookies=2".

Thank you for looking at that!

Quoted 3 lines

Note that MPTCP should work with syncookies (4th ack would carry needed state), but it appears better to sort that out in -next so do tcp fallback for now.

Indeed, it makes sense.

Quoted 2 lines

I decided to remove the ifdef on tcp_rsk "is_mptcp" member because if (IS_ENABLED()) is easier to read than "#ifdef IS_ENABLED()/#endif" pair.

I agree with you that it is easier to read but will it be OK for upstream? Should we use a bitfield and include 'tfo_listener' in it?

Cheers,
Matt

Signature · 5 lines
Matthieu Baerts | R&D Engineer
matthieu.baerts(a)tessares.net
Tessares SA | Hybrid Access Solutions
www.tessares.net
1 Avenue Jean Monnet, 1348 Louvain-la-Neuve, Belgium
author
#7 lore raw

[MPTCP] Re: [PATCH net-next v1 4/4] mptcp: handle tcp fallback when using syn cookies

Quoted 6 lines · Matthieu Baerts
Quoted 2 lines · Florian Westphal

I decided to remove the ifdef on tcp_rsk "is_mptcp" member because if (IS_ENABLED()) is easier to read than "#ifdef IS_ENABLED()/#endif" pair.

I agree with you that it is easier to read but will it be OK for upstream? Should we use a bitfield and include 'tfo_listener' in it?

It doesn't increase struct size, so I think its OK. bitfield would be better but not for -net, we can do this in 2 weeks when net-next opens again.

#8 lore raw

[MPTCP] Re: [PATCH net-next v1 4/4] mptcp: handle tcp fallback when using syn cookies

In reply to Florian Westphal

I just tested this one as well. Works fine now!

Christoph

Quoted 117 lines · Florian Westphal

We can't deal with syncookie mode yet, the syncookie rx path will create tcp rsk instead of mptcp rsk, i.e. we get OOB access because we treat tcp rsk as mptcp rsk:

TCP: request_sock_subflow: Possible SYN flooding on port 20002. Sending cookies. Check SNMP counters.

BUG: KASAN: slab-out-of-bounds in subflow_syn_recv_sock+0x451/0x4d0 net/mptcp/subflow.c:191
Read of size 1 at addr ffff8881167bc148 by task syz-executor099/2120
 subflow_syn_recv_sock+0x451/0x4d0 net/mptcp/subflow.c:191
 tcp_get_cookie_sock+0xcf/0x520 net/ipv4/syncookies.c:209
 cookie_v6_check+0x15a5/0x1e90 net/ipv6/syncookies.c:252
 tcp_v6_cookie_check net/ipv6/tcp_ipv6.c:1123 [inline]
 [..]

Bug can be reproduced via "sysctl net.ipv4.tcp_syncookies=2".

Note that MPTCP should work with syncookies (4th ack would carry needed state), but it appears better to sort that out in -next so do tcp fallback for now.

I decided to remove the ifdef on tcp_rsk "is_mptcp" member because if (IS_ENABLED()) is easier to read than "#ifdef IS_ENABLED()/#endif" pair.

Reported-by Christoph Paasch <cpaasch(a)apple.com>
Signed-off-by Florian Westphal <fw(a)strlen.de>

---
patch is new in this set.

 include/linux/tcp.h   | 2 --
 net/ipv4/syncookies.c | 4 ++++
 net/ipv4/tcp_input.c  | 3 +++
 net/ipv6/syncookies.c | 3 +++
 net/mptcp/subflow.c   | 5 ++++-
 5 files changed, 14 insertions(+), 3 deletions(-)
include/linux/tcp.h +0 -2
index 1cf73e6f85ca..3dc964010fef 100644
@@ -148,9 +148,7 @@ struct tcp_request_sock {
 	const struct tcp_request_sock_ops *af_specific;
 	u64				snt_synack; /* first SYNACK sent time */
 	bool				tfo_listener;
-#if IS_ENABLED(CONFIG_MPTCP)
 	bool				is_mptcp;
-#endif
 	u32				txhash;
 	u32				rcv_isn;
 	u32				snt_isn;
net/ipv4/syncookies.c +4 -0
index 345b2b0ff618..9a4f6b16c9bc 100644
@@ -349,6 +349,10 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 	req->ts_recent		= tcp_opt.saw_tstamp ? tcp_opt.rcv_tsval : 0;
 	treq->snt_synack	= 0;
 	treq->tfo_listener	= false;
+
+	if (IS_ENABLED(CONFIG_MPTCP))
+		treq->is_mptcp = 0;
+
 	if (IS_ENABLED(CONFIG_SMC))
 		ireq->smc_ok = 0;
 
net/ipv4/tcp_input.c +3 -0
index e8b840a4767e..e325b4506e25 100644
@@ -6637,6 +6637,9 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 
 	af_ops->init_req(req, sk, skb);
 
+	if (IS_ENABLED(CONFIG_MPTCP) && want_cookie)
+		tcp_rsk(req)->is_mptcp = 0;
+
 	if (security_inet_conn_request(sk, skb, req))
 		goto drop_and_free;
 
net/ipv6/syncookies.c +3 -0
index 30915f6f31e3..13235a012388 100644
@@ -178,6 +178,9 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	treq = tcp_rsk(req);
 	treq->tfo_listener = false;
 
+	if (IS_ENABLED(CONFIG_MPTCP))
+		treq->is_mptcp = 0;
+
 	if (security_inet_conn_request(sk, skb, req))
 		goto out_free;
 
net/mptcp/subflow.c +4 -1
index 1662e1178949..cf9bad97a12a 100644
@@ -186,6 +186,9 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 
 	pr_debug("listener=%p, req=%p, conn=%p", listener, req, listener->conn);
 
+	if (tcp_rsk(req)->is_mptcp == 0)
+		goto create_child;
+
 	/* if the sk is MP_CAPABLE, we try to fetch the client key */
 	subflow_req = mptcp_subflow_rsk(req);
 	if (subflow_req->mp_capable) {
@@ -767,7 +770,7 @@ static void subflow_ulp_clone(const struct request_sock *req,
 	struct mptcp_subflow_context *old_ctx = mptcp_subflow_ctx(newsk);
 	struct mptcp_subflow_context *new_ctx;
 
-	if (!subflow_req->mp_capable) {
+	if (!tcp_rsk(req)->is_mptcp || !subflow_req->mp_capable) {
 		subflow_ulp_fallback(newsk, old_ctx);
 		return;
 	}
Signature · 1 line
2.24.1
#9 lore raw
In reply to Florian Westphal
Quoted 25 lines · Florian Westphal

This fixes the syzkaller splats with reproducers that Christoph forwarded to the mailing list, including the one wrt. slab oob in syncookie mode.

First patch moves free of the cached extension pointer to the destroy callback to make sure no other call path can allocate a new one. Actually net-next should be fine in this regard (no msk work queue yet), but as Paolo pointed out it looks bogus when 3rd patch calls it with mptcp lock already released.

2nd patch fixes a crash with set/getsockopt interface and resolves a lockdep splat due to calls to rtnl_lock while already holding the msk lock.

I resolved both in the same patch because the fixes need to touch the same spot.

3rd patch fixes another lockdep splat, this time during mptcp_close but again related to rtnl_lock.

Last patch avoid OOB access when syncookies are used.
This should be resolved differently, as mptcp v1 can use 4th ack to recover state, but that will be more intrusive so this fix downgrades to TCP instead.

See patches for more detailed explanation.

Quoted 7 lines
 include/linux/tcp.h   |    2 --
 net/ipv4/syncookies.c |    4 ++++
 net/ipv4/tcp_input.c  |    3 +++
 net/ipv6/syncookies.c |    3 +++
 net/mptcp/protocol.c  |   48 ++++++++++++++++++++++++++++--------------------
 net/mptcp/subflow.c   |    5 ++++-
 6 files changed, 42 insertions(+), 23 deletions(-)

Except for the already reported note on patch 2/4, I think the others patches are good.

Thanks Florian for fixing our mess ;)

/P