In the Linux kernel, the following vulnerability has been resolved:
sunrpc: wait for in-flight TLS handshake callback when cancel loses race
When wait_for_completion_interruptible_timeout() in
svc_tcp_handshake() returns 0 (timeout) or -ERESTARTSYS (signal) and
tls_handshake_cancel() then returns false, handshake_complete() has
won the cancellation race: it has set HANDSHAKE_F_REQ_COMPLETED and
is about to invoke svc_tcp_handshake_done(), but the callback's
side effects on xpt_flags and on svsk->sk_handshake_done have not
yet committed.
The current code reads xpt_flags immediately to decide whether the
session succeeded. Two races result.
If the callback has executed set_bit(XPT_TLS_SESSION) but not yet
clear_bit(XPT_HANDSHAKE), svc_tcp_handshake() sees a session,
enqueues the transport, and returns. svc_xprt_received() then
clears XPT_BUSY, a worker thread picks the transport up, the
dispatcher in svc_handle_xprt() observes XPT_HANDSHAKE still set,
and xpo_handshake is invoked a second time. That svc_tcp_handshake()
calls init_completion(&svsk->sk_handshake_done) while the original
callback concurrently calls complete_all() on it, corrupting the
embedded swait_queue.
If the callback has set HANDSHAKE_F_REQ_COMPLETED but not yet
entered svc_tcp_handshake_done(), svc_tcp_handshake() reads
XPT_TLS_SESSION as clear and tears the connection down even though
the handshake is about to succeed.
Wait for the callback to commit before inspecting xpt_flags. The
completion is guaranteed to fire because handshake_complete()
invokes svc_tcp_handshake_done() unconditionally once it has set
HANDSHAKE_F_REQ_COMPLETED.
CVSS Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
CVSS Score: 9.8
AV:N - Reachable over TCP by any remote peer of an in-kernel NFS/RPC server (nfsd) via the pre-auth RPC_AUTH_TLS NULL-procedure STARTTLS path that invokes svc_tcp_handshake() on the accepted SUNRPC transport.
AC:L - The attacker controls both sides of the race by timing TLS client handshake messages against the fixed 5-second SVC_HANDSHAKE_TO and opening many parallel connections to reliably hit the cancel-lost window.
PR:N - No credentials or export authorization are required; svcauth_tls_accept() accepts empty AUTH_TLS on RPC NULL from any TCP client, and svcauth_unix_set_client() skips IP export checks when rq_proc==0.
UI:N - Exploitation requires only scripted network connections and TLS client traffic to nfsd; no victim user interaction such as mounting a filesystem or opening a file is needed.
S:U - Memory corruption occurs in nfsd kernel worker context on the same host; impact is kernel privilege compromise within the server security authority, not a VM escape or IOMMU boundary bypass.
C:H - A second svc_tcp_handshake() can call init_completion() while svc_tcp_handshake_done() concurrently calls complete_all(), corrupting the embedded swait_queue in heap-allocated svc_sock and enabling arbitrary kernel memory disclosure.
I:H - Corrupting completion wait-queue metadata in the svc_sock heap object can be leveraged for arbitrary kernel writes and control-flow hijacking to achieve local privilege escalation on the NFS server.
A:H - The swait_queue corruption can trigger kernel oops/panic, and the alternate race path tears down connections whose TLS handshake is about to succeed, enabling repeatable remote denial-of-service against nfsd.
| Attack Vector |
Network |
Scope |
Unchanged |
| Attack Complexity |
Low |
Confidentiality Impact |
High |
| Privileges Required |
None |
Integrity Impact |
High |
| User Interaction |
None |
Availability Impact |
High |
AV:N - Reachable over TCP by any remote peer of an in-kernel NFS/RPC server (nfsd) via the pre-auth RPC_AUTH_TLS NULL-procedure STARTTLS path that invokes svc_tcp_handshake() on the accepted SUNRPC transport.
AC:L - The attacker controls both sides of the race by timing TLS client handshake messages against the fixed 5-second SVC_HANDSHAKE_TO and opening many parallel connections to reliably hit the cancel-lost window.
PR:N - No credentials or export authorization are required; svcauth_tls_accept() accepts empty AUTH_TLS on RPC NULL from any TCP client, and svcauth_unix_set_client() skips IP export checks when rq_proc==0.
UI:N - Exploitation requires only scripted network connections and TLS client traffic to nfsd; no victim user interaction such as mounting a filesystem or opening a file is needed.
S:U - Memory corruption occurs in nfsd kernel worker context on the same host; impact is kernel privilege compromise within the server security authority, not a VM escape or IOMMU boundary bypass.
C:H - A second svc_tcp_handshake() can call init_completion() while svc_tcp_handshake_done() concurrently calls complete_all(), corrupting the embedded swait_queue in heap-allocated svc_sock and enabling arbitrary kernel memory disclosure.
I:H - Corrupting completion wait-queue metadata in the svc_sock heap object can be leveraged for arbitrary kernel writes and control-flow hijacking to achieve local privilege escalation on the NFS server.
A:H - The swait_queue corruption can trigger kernel oops/panic, and the alternate race path tears down connections whose TLS handshake is about to succeed, enabling repeatable remote denial-of-service against nfsd.
CVSS 3.1