Description
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---
Published: 2026-09-25
Score: 8.1 High
EPSS: < 1% Very Low
KEV: No
Impact: Kernel crash leading to denial of service
Action: Immediate Patch
AI Analysis

Impact

The flaw exists in the Linux kernel’s RDS networking subsystem; during a connection shutdown the routine fails to acquire the fastpath locks before performing the transport teardown. This race allows the transmit and receive paths to run concurrently with the shutdown, which can induce memory corruption, zeroing of rings, and NULL dereferences that lead to a kernel panic. The result is a loss of system availability, as a crash in core networking code brings the host down.

Affected Systems

All Linux kernel releases that include the RDS subsystem are affected until the patch that acquires the fastpath locks before teardown is applied. No specific version range is enumerated in the advisory; the change is made in a kernel commit and applies to kernels incorporating that commit.

Risk and Exploitability

The CVSS score of 8.1 indicates high severity. The EPSS score of less than 1% shows that exploitation is presently unlikely, and the vulnerability is not listed in CISA KEV. The race requires triggering concurrent RDS operations, which typically means local or privileged access to the system. A successful exploit would crash the kernel, resulting in a denial-of-service condition.

Generated by OpenCVE AI on September 30, 2026 at 11:44 UTC.

Remediation

No solution or workaround provided in the CVE record.

OpenCVE Recommended Actions

  • Apply a kernel update that contains the RDS lock fix—upgrade to a kernel version that includes the commit adding acquire operations before shutdown.
  • If a patch is not yet available, disable RDS traffic by blocking the protocol at the firewall or setting sysctl parameters to prevent use of RDS until the update can be applied.
  • Monitor hosts for kernel crashes or panics and maintain a rapid response plan to reload the kernel or roll back if a crash occurs.

Generated by OpenCVE AI on September 30, 2026 at 11:44 UTC.

Tracking

Sign in to view the affected projects.

Advisories
Source ID Title
Debian DLA Debian DLA DLA-4817-1 linux-6.12 security update
Debian DLA Debian DLA DLA-4841-1 linux security update
Debian DSA Debian DSA DSA-6528-1 linux security update
Ubuntu USN Ubuntu USN USN-8961-1 Linux kernel vulnerabilities
History

Sat, 03 Oct 2026 11:15:00 +0000

Type Values Removed Values Added
References

Tue, 29 Sep 2026 20:45:00 +0000

Type Values Removed Values Added
Weaknesses CWE-362

Tue, 29 Sep 2026 13:15:00 +0000

Type Values Removed Values Added
Weaknesses CWE-366
References
Metrics threat_severity

None

 threat_severity

Moderate

Fri, 25 Sep 2026 20:30:00 +0000

Type Values Removed Values Added
Weaknesses CWE-362

Fri, 25 Sep 2026 15:00:00 +0000

Type Values Removed Values Added
Metrics cvssV3_1

{'score': 8.1, 'vector': 'CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H'}

Fri, 25 Sep 2026 10:30:00 +0000

Type Values Removed Values Added
Description 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---
Title net/rds: acquire the fastpath locks in rds_conn_shutdown()
First Time appeared Linux
Linux linux Kernel
CPEs cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*
Vendors & Products Linux
Linux linux Kernel
References

Subscriptions

Linux Linux Kernel
cve-icon MITRE

Status: PUBLISHED

Assigner: Linux

Published:

Updated: 2026-10-19T09:13:18.938Z

Reserved: 2026-09-25T10:19:56.074Z

Link: CVE-2026-98069

cve-icon Vulnrichment

No data.

cve-icon NVD

Status : Undergoing Analysis

Published: 2026-09-25T11:17:36.043

Modified: 2026-10-19T10:18:01.887

Link: CVE-2026-98069

cve-icon Redhat

Severity : Moderate

Publid Date: 2026-09-25T00:00:00Z

Links: CVE-2026-98069 - Bugzilla

cve-icon OpenCVE Enrichment

Updated: 2026-09-30T11:45:16Z

Weaknesses
  • CWE-366

    Race Condition within a Thread