[PATCH net 0/2] tcp: fix congestion control UAFs in do_tcp_getsockopt()

[PATCH net 0/2] tcp: fix congestion control UAFs in do_tcp_getsockopt()

[Cen Zhang (Microsoft)]

do_tcp_getsockopt() has two lockless reads of icsk_ca_ops. Since BPF
struct_ops congestion control made icsk_ca_ops point to dynamically
allocated memory, a concurrent setsockopt(TCP_CONGESTION) can replace
the pointer and free the old object while either reader is using it.

Patch 1 fixes the TCP_CONGESTION path by copying ca_ops->name to a
stack buffer while holding rcu_read_lock().

Patch 2 fixes the TCP_CC_INFO path by keeping the ca_ops->get_info
lookup and call inside an RCU read-side critical section.

Eric Dumazet's related TCP data-race annotation work [1] adds
READ_ONCE()/WRITE_ONCE() annotations for lockless TCP socket field
accesses. icsk_ca_ops would similarly benefit from such annotations,
but that is an independent data-race concern. This series focuses
solely on the use-after-free fixes via rcu_read_lock().

[1]
https://lore.kernel.org/all/20260416200319.3608680-1-edumazet@google.com/

Cen Zhang (Microsoft) (2):
  tcp: fix use-after-free in do_tcp_getsockopt(TCP_CONGESTION)
  tcp: fix use-after-free in do_tcp_getsockopt(TCP_CC_INFO)

 net/ipv4/tcp.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)


base-commit: 746fc0787f616da418ffc04a110296fe95d53491
-- 
2.55.0

[PATCH net 1/2] tcp: fix use-after-free in do_tcp_getsockopt(TCP_CONGESTION)

[Cen Zhang (Microsoft)]

do_tcp_getsockopt() reads icsk->icsk_ca_ops->name without holding
rcu_read_lock(). Since commit 0baf26b0fcd7 ("bpf: tcp: Support
tcp_congestion_ops in bpf"), icsk_ca_ops can point to dynamically
allocated BPF struct_ops memory that may be freed concurrently via
setsockopt(TCP_CONGESTION), leading to a use-after-free.

  BUG: KASAN: slab-use-after-free in _copy_to_user+0x37/0x60
  Read of size 16 at addr ffff888013505260 by task exploit/149
   _copy_to_user+0x37/0x60
   do_tcp_getsockopt+0x158a/0x2460 (net/ipv4/tcp.c:4585)
   tcp_getsockopt+0x91/0xf0
   __sys_getsockopt+0xf7/0x170

Fix this by holding rcu_read_lock() around the ca_ops->name access and
copying the name to a stack buffer before releasing the lock.

Fixes: 0baf26b0fcd7 ("bpf: tcp: Support tcp_congestion_ops in bpf")
Reported-by: AutonomousCodeSecurity@microsoft.com
Reported-by: Xiang Mei (Microsoft) <xmei5@asu.edu>
Reported-by: Cen Zhang (Microsoft) <blbllhy@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Cen Zhang (Microsoft) <blbllhy@gmail.com>
---
The unsynchronized icsk_ca_ops load also constitutes a data race.
READ_ONCE()/WRITE_ONCE() annotations are intentionally left to a
separate change; related TCP annotation work is available at:
https://lore.kernel.org/all/20260416200319.3608680-1-edumazet@google.com/

 net/ipv4/tcp.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index b4237d0e994d..7360ff718f1d 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -4577,16 +4577,23 @@ int do_tcp_getsockopt(struct sock *sk, int level,
 		val = !inet_csk_in_pingpong_mode(sk);
 		break;
 
-	case TCP_CONGESTION:
+	case TCP_CONGESTION: {
+		char ca_name[TCP_CA_NAME_MAX] = {};
+
 		if (copy_from_sockptr(&len, optlen, sizeof(int)))
 			return -EFAULT;
 		len = min_t(unsigned int, len, TCP_CA_NAME_MAX);
 		if (copy_to_sockptr(optlen, &len, sizeof(int)))
 			return -EFAULT;
-		if (copy_to_sockptr(optval, icsk->icsk_ca_ops->name, len))
+
+		rcu_read_lock();
+		memcpy(ca_name, icsk->icsk_ca_ops->name, sizeof(ca_name));
+		rcu_read_unlock();
+
+		if (copy_to_sockptr(optval, ca_name, len))
 			return -EFAULT;
 		return 0;
-
+	}
 	case TCP_ULP:
 		if (copy_from_sockptr(&len, optlen, sizeof(int)))
 			return -EFAULT;
-- 
2.55.0

[PATCH net 2/2] tcp: fix use-after-free in do_tcp_getsockopt(TCP_CC_INFO)

[Cen Zhang (Microsoft)]

do_tcp_getsockopt() reads icsk->icsk_ca_ops and dereferences the
get_info function pointer without rcu_read_lock(). With BPF struct_ops
congestion control, ca_ops can point to dynamically allocated memory
that is freed concurrently, resulting in a use-after-free when the
kernel dereferences or calls through the stale pointer.

  BUG: KASAN: slab-use-after-free in do_tcp_getsockopt+0x2037/0x23e0
  Read of size 8 at addr ffff888013701258 by task exploit/149
   do_tcp_getsockopt+0x2037/0x23e0 (net/ipv4/tcp.c:4564)
   tcp_getsockopt+0x91/0xf0
   __sys_getsockopt+0xf7/0x170

Fix this by wrapping the ca_ops load and get_info call within
rcu_read_lock()/rcu_read_unlock().

Fixes: 0baf26b0fcd7 ("bpf: tcp: Support tcp_congestion_ops in bpf")
Reported-by: AutonomousCodeSecurity@microsoft.com
Reported-by: Cen Zhang (Microsoft) <blbllhy@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Cen Zhang (Microsoft) <blbllhy@gmail.com>
---
The unsynchronized icsk_ca_ops load also constitutes a data race.
READ_ONCE()/WRITE_ONCE() annotations are intentionally left to a
separate change; related TCP annotation work is available at:
https://lore.kernel.org/all/20260416200319.3608680-1-edumazet@google.com/

 net/ipv4/tcp.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 7360ff718f1d..43ccd2d37f9c 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -4562,9 +4562,11 @@ int do_tcp_getsockopt(struct sock *sk, int level,
 		if (copy_from_sockptr(&len, optlen, sizeof(int)))
 			return -EFAULT;
 
+		rcu_read_lock();
 		ca_ops = icsk->icsk_ca_ops;
 		if (ca_ops && ca_ops->get_info)
 			sz = ca_ops->get_info(sk, ~0U, &attr, &info);
+		rcu_read_unlock();
 
 		len = min_t(unsigned int, len, sz);
 		if (copy_to_sockptr(optlen, &len, sizeof(int)))
-- 
2.55.0

Re: [PATCH net 2/2] tcp: fix use-after-free in do_tcp_getsockopt(TCP_CC_INFO)

[Eric Dumazet]

On Fri, Aug 21, 2026 at 8:25 PM Cen Zhang (Microsoft) <blbllhy@gmail.com> wrote:
>
> do_tcp_getsockopt() reads icsk->icsk_ca_ops and dereferences the
> get_info function pointer without rcu_read_lock(). With BPF struct_ops
> congestion control, ca_ops can point to dynamically allocated memory
> that is freed concurrently, resulting in a use-after-free when the
> kernel dereferences or calls through the stale pointer.
>
>   BUG: KASAN: slab-use-after-free in do_tcp_getsockopt+0x2037/0x23e0
>   Read of size 8 at addr ffff888013701258 by task exploit/149
>    do_tcp_getsockopt+0x2037/0x23e0 (net/ipv4/tcp.c:4564)
>    tcp_getsockopt+0x91/0xf0
>    __sys_getsockopt+0xf7/0x170
>
> Fix this by wrapping the ca_ops load and get_info call within
> rcu_read_lock()/rcu_read_unlock().
>
> Fixes: 0baf26b0fcd7 ("bpf: tcp: Support tcp_congestion_ops in bpf")
> Reported-by: AutonomousCodeSecurity@microsoft.com
> Reported-by: Cen Zhang (Microsoft) <blbllhy@gmail.com>
> Cc: stable@vger.kernel.org
> Signed-off-by: Cen Zhang (Microsoft) <blbllhy@gmail.com>
> ---
> The unsynchronized icsk_ca_ops load also constitutes a data race.
> READ_ONCE()/WRITE_ONCE() annotations are intentionally left to a
> separate change; related TCP annotation work is available at:
> https://lore.kernel.org/all/20260416200319.3608680-1-edumazet@google.com/


I don't understand.

This patch is incomplete. Please fix the remaining issues in a single
patch or series.

We have limited time; we don't want to revisit the same area  in one month.

pw-bot: cr

Re: [PATCH net 2/2] tcp: fix use-after-free in do_tcp_getsockopt(TCP_CC_INFO)

[Cen Zhang (Microsoft)]

Thanks for the review, Eric. I will send v2 with the data races fixed as
well.

Thanks,
Cen