In the Linux kernel, the following vulnerability has been resolved:
net/rds: acquire the fastpath locks in rds_conn_shutdown()
rds_conn_shutdown() quiesces the transmit and receive-refill paths by
waiting for RDS_IN_XMIT and RDS_RECV_REFILL to be sampled clear, and
then runs the transport shutdown and rds_conn_path_reset(). Sampling
the bits clear is not the same as owning them: the moment after the
wait_event() returns, rds_send_xmit() can re-acquire RDS_IN_XMIT (or
rds_ib_recv_refill() can re-acquire RDS_RECV_REFILL) and run
concurrently with the teardown.
The sender does recheck the connection state after taking the lock,
but that recheck is a classic store-buffering pattern: teardown writes
the state and reads the bit while the sender writes the bit and reads
the state. acquire_in_xmit() is only an acquire operation, so on
weakly ordered architectures both sides can miss each other's write,
and the transmit path then runs while the transport zeroes its rings
(e.g. rds_ib_ring_init()) and rds_send_path_reset() rewrites the
transmit state under it.
Oracle UEK fixed the same class of crashes - a 14-year tail of
BUG_ON()s in rds_ib_sub_signaled(), unexpected op-codes and NULL
dereferences in rds_ib_send_cqe_handler() during failover testing -
by making the teardown path acquire the fastpath bit locks instead
of testing them ("rds: Make sure transmit path and connection
tear-down does not run concurrently"). Ownership of a single word is
decided by RMW atomicity, so no cross-variable ordering is needed.
Do the same here: take both locks before calling the transport
shutdown, hold them across rds_conn_path_reset(), and release them
explicitly with a wake-up afterwards. Both are released with
clear_bit_unlock(), so that the ring re-initialization done by the
transport shutdown and the transmit state rewritten by
rds_send_path_reset() are ordered before either bit is seen clear by
the next acquire_in_xmit() or acquire_refill().
The fastpath users of these bits - rds_send_xmit() and
rds_ib_recv_refill() - are trylock style and back off while teardown
owns the locks, so no new lock dependency is introduced for them.
rds_tcp_reset_callbacks() is different: since the previous patch it
acquires RDS_IN_XMIT as well, and it blocks doing so, so its wait now
spans the teardown instead of at most one send batch. That waiter
runs from rds_tcp_accept_one() on the single-threaded krdsd workqueue
and holds rds_tcp_accept_lock and t_conn_path_lock while it waits, so
a duelling SYN accepted while its path is being torn down parks
accept processing for the duration of the teardown - for TCP bounded
by the (up to 5 s) drain loop in rds_tcp_conn_path_shutdown(). An IB
path's drain in rds_ib_conn_path_shutdown() has no round cap, but no
blocking waiter either: rds_tcp_reset_callbacks() is the only blocking
acquirer of these bits and waits only on its own TCP path, and the
fastpaths are trylock-and-back-off on both transports, so a long IB
drain lengthens only that path's own quiesce. The
window is narrow: the accept-side state check has to pass before the
teardown moves the path to RDS_CONN_DISCONNECTING.
Because krdsd is a single global workqueue, everything else queued
there - accept processing for other connections and network
namespaces, and the flush_workqueue(rds_wq) in rds_tcp_listen_stop()
during namespace teardown - waits behind the parked accept worker for
that time. It cannot deadlock, although the waits do point at each
other: the teardown blocks until the bit's holder releases it, and
the holder may be that krdsd accept worker. The holder finishes
without needing anything the teardown owns: the sync cancels
rds_tcp_reset_callbacks() issues target cp_send_w and cp_recv_w on
the path's ordered cp_wq, whose only execution slot is occupied by
the blocked cp_down_w itself, so they are pending at most and cancel
without flushing - a reliance on cp_wq being ordered that is now
noted next to those cancels (on
---truncated---
CVSS Vector: CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
CVSS Score: 8.1
AV:N - A remote RDS/TCP peer on port 16385 drives both sides of the race: RDS ping messages make rds_recv_incoming() call rds_send_pong() and so rds_send_xmit(), and the peer's own FIN/RST reaches rds_tcp_state_change() and then rds_conn_path_drop(), which queues rds_conn_shutdown(). The damaged state is that connection's transmit state and TCP socket.
AC:H - The attacker can start sends and teardowns at will, but the sender only overlaps the teardown if the test_bit() wait in rds_conn_shutdown() and the state recheck in rds_send_xmit() both miss each other's store. That needs a weakly ordered CPU and a narrow reordering window the attacker cannot control.
PR:N - rds_tcp_accept_one() accepts RDS/TCP connections from any peer address with no authentication, so pings and connection drops need no credentials. A local user can also reach it without privilege, since AF_RDS socket creation has no capability check and SO_RDS_TRANSPORT autoloads rds_tcp.
UI:N - No victim action is needed. The attacker sends the pings and resets the connection themselves.
S:U - The race and its memory corruption stay inside the kernel's own security authority. No guest/host or other boundary is crossed.
C:H - rds_send_xmit()/rds_tcp_xmit() can keep using tc->t_sock after rds_tcp_conn_path_shutdown() has called sock_release() on it, and can keep using cp_xmit_rm while rds_send_path_reset() rewrites it. Both are use-after-free on socket and message objects, which could be groomed into kernel memory disclosure.
I:H - The transmit path writes through a freed struct socket/sock, and through transmit and ring state (IB rings zeroed by rds_ib_ring_init()) that the teardown is rewriting at the same time. That is a heap memory-corruption primitive.
A:H - Losing the race crashes the kernel: the commit cites BUG_ON()s in rds_ib_sub_signaled() and NULL dereferences in rds_ib_send_cqe_handler(), and the TCP path can dereference a released socket.
| Attack Vector |
Network |
Scope |
Unchanged |
| Attack Complexity |
High |
Confidentiality Impact |
High |
| Privileges Required |
None |
Integrity Impact |
High |
| User Interaction |
None |
Availability Impact |
High |
AV:N - A remote RDS/TCP peer on port 16385 drives both sides of the race: RDS ping messages make rds_recv_incoming() call rds_send_pong() and so rds_send_xmit(), and the peer's own FIN/RST reaches rds_tcp_state_change() and then rds_conn_path_drop(), which queues rds_conn_shutdown(). The damaged state is that connection's transmit state and TCP socket.
AC:H - The attacker can start sends and teardowns at will, but the sender only overlaps the teardown if the test_bit() wait in rds_conn_shutdown() and the state recheck in rds_send_xmit() both miss each other's store. That needs a weakly ordered CPU and a narrow reordering window the attacker cannot control.
PR:N - rds_tcp_accept_one() accepts RDS/TCP connections from any peer address with no authentication, so pings and connection drops need no credentials. A local user can also reach it without privilege, since AF_RDS socket creation has no capability check and SO_RDS_TRANSPORT autoloads rds_tcp.
UI:N - No victim action is needed. The attacker sends the pings and resets the connection themselves.
S:U - The race and its memory corruption stay inside the kernel's own security authority. No guest/host or other boundary is crossed.
C:H - rds_send_xmit()/rds_tcp_xmit() can keep using tc->t_sock after rds_tcp_conn_path_shutdown() has called sock_release() on it, and can keep using cp_xmit_rm while rds_send_path_reset() rewrites it. Both are use-after-free on socket and message objects, which could be groomed into kernel memory disclosure.
I:H - The transmit path writes through a freed struct socket/sock, and through transmit and ring state (IB rings zeroed by rds_ib_ring_init()) that the teardown is rewriting at the same time. That is a heap memory-corruption primitive.
A:H - Losing the race crashes the kernel: the commit cites BUG_ON()s in rds_ib_sub_signaled() and NULL dereferences in rds_ib_send_cqe_handler(), and the TCP path can dereference a released socket.
CVSS 3.1