Description
In the Linux kernel, the following vulnerability has been resolved:

Revert "wireguard: device: enable threaded NAPI"

This reverts commit 933466fc50a8e4eb167acbd0d8ec96a078462e9c which is
commit db9ae3b6b43c79b1ba87eea849fd65efa05b4b2e upstream.

We have had three independent production user reports in combination
with Cilium utilizing WireGuard as encryption underneath that k8s Pod
E/W traffic to certain peer nodes fully stalled. The situation appears
as follows:

- Occurs very rarely but at random times under heavy networking load.
- Once the issue triggers the decryption side stops working completely
for that WireGuard peer, other peers keep working fine. The stall
happens also for newly initiated connections towards that particular
WireGuard peer.
- Only the decryption side is affected, never the encryption side.
- Once it triggers, it never recovers and remains in this state,
the CPU/mem on that node looks normal, no leak, busy loop or crash.
- bpftrace on the affected system shows that wg_prev_queue_enqueue
fails, thus the MAX_QUEUED_PACKETS (1024 skbs!) for the peer's
rx_queue is reached.
- Also, bpftrace shows that wg_packet_rx_poll for that peer is never
called again after reaching this state for that peer. For other
peers wg_packet_rx_poll does get called normally.
- Commit db9ae3b ("wireguard: device: enable threaded NAPI")
switched WireGuard to threaded NAPI by default. The default has
not been changed for triggering the issue, neither did CPU
hotplugging occur (i.e. 5bd8de2 ("wireguard: queueing: always
return valid online CPU in wg_cpumask_choose_online()")).
- The issue has been observed with stable kernels of v5.15 as well as
v6.1. It was reported to us that v5.10 stable is working fine, and
no report on v6.6 stable either (somewhat related discussion in [0]
though).
- In the WireGuard driver the only material difference between v5.10
stable and v5.15 stable is the switch to threaded NAPI by default.

[0] https://lore.kernel.org/netdev/CA+wXwBTT74RErDGAnj98PqS=wvdh8eM1pi4q6tTdExtjnokKqA@mail.gmail.com/

Breakdown of the problem:

1) skbs arriving for decryption are enqueued to the peer->rx_queue in
wg_packet_consume_data via wg_queue_enqueue_per_device_and_peer.
2) The latter only moves the skb into the MPSC peer queue if it does
not surpass MAX_QUEUED_PACKETS (1024) which is kept track in an
atomic counter via wg_prev_queue_enqueue.
3) In case enqueueing was successful, the skb is also queued up
in the device queue, round-robin picks a next online CPU, and
schedules the decryption worker.
4) The wg_packet_decrypt_worker, once scheduled, picks these up
from the queue, decrypts the packets and once done calls into
wg_queue_enqueue_per_peer_rx.
5) The latter updates the state to PACKET_STATE_CRYPTED on success
and calls napi_schedule on the per peer->napi instance.
6) NAPI then polls via wg_packet_rx_poll. wg_prev_queue_peek checks
on the peer->rx_queue. It will wg_prev_queue_dequeue if the
queue->peeked skb was not cached yet, or just return the latter
otherwise. (wg_prev_queue_drop_peeked later clears the cache.)
7) From an ordering perspective, the peer->rx_queue has skbs in order
while the device queue with the per-CPU worker threads from a
global ordering PoV can finish the decryption and signal the skb
PACKET_STATE_CRYPTED out of order.
8) A situation can be observed that the first packet coming in will
be stuck waiting for the decryption worker to be scheduled for
a longer time when the system is under pressure.
9) While this is the case, the other CPUs in the meantime finish
decryption and call into napi_schedule.
10) Now in wg_packet_rx_poll it picks up the first in-order skb
from the peer->rx_queue and sees that its state is still
PACKET_STATE_UNCRYPTED. The NAPI poll routine then exits e
---truncated---
Published: 2026-06-24
Score: 7.5 High
EPSS: < 1% Very Low
KEV: No
Impact: n/a
Action: n/a
AI Analysis

Impact

The vulnerability resides in the Linux kernel’s WireGuard decryption driver. When WireGuard is built with threaded NAPI enabled by default, the decryption worker may become permanently stalled for a particular peer under heavy network load. This stall occurs after the peer’s receive queue reaches its maximum of 1024 skbs, causing the decryption side to cease processing further packets for that peer while other peers continue normally. Although the kernel remains operational, the affected peer experiences persistent packet loss, effectively denying service for that connection without producing a crash or system-wide crash-loop.

Affected Systems

Linux kernel versions 5.15 stable and 6.1 stable are affected when the commit that enables threaded NAPI by default is present. The issue is not observed in 5.10 stable or 6.6 stable, indicating the regression is tied to the threaded NAPI change. Any system running a kernel with threaded NAPI enabled for WireGuard and handling pod‑to‑pod or inter‑node traffic is potentially impacted.

Risk and Exploitability

The CVSS score of 7.5 indicates high severity, while the reported EPSS score of less than 1% suggests a very low likelihood of exploitation at present. The likely attack vector requires an adversary to generate high‑volume WireGuard traffic directed at a specific peer to trigger the stall; this inference is based on the description that the issue occurs rarely under heavy networking load. The vulnerability is not listed in the CISA KEV catalog. Because the stall is limited to the decryption path of WireGuard and does not affect encryption, system stability, or other networking components, the overall risk is confined to disrupted peer communication rather than a full system compromise.

Generated by OpenCVE AI on June 28, 2026 at 14:37 UTC.

Remediation

No vendor fix or workaround currently provided.

OpenCVE Recommended Actions

  • Revert the kernel commit that enabled threaded NAPI for the WireGuard device, restoring the original serial decryption path.
  • If reverting the commit is not immediately feasible, configure the kernel or WireGuard driver to disable threaded NAPI for WireGuard devices, ensuring that decryption is handled in legacy poll mode.
  • Continuously monitor WireGuard decryption metrics and network traffic on nodes to detect sudden drops in packet processing or stalls on specific peers, enabling rapid response before a full service outage occurs.

Generated by OpenCVE AI on June 28, 2026 at 14:37 UTC.

Tracking

Sign in to view the affected projects.

Advisories

No advisories yet.

History

Wed, 01 Jul 2026 12:45:00 +0000

Type Values Removed Values Added
References

Sun, 28 Jun 2026 08:00:00 +0000

Type Values Removed Values Added
Metrics cvssV3_1

{'score': 7.0, 'vector': 'CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H'}

 cvssV3_1

{'score': 7.5, 'vector': 'CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H'}

Fri, 26 Jun 2026 06:30:00 +0000

Type Values Removed Values Added
Weaknesses CWE-400
CWE-665

Fri, 26 Jun 2026 00:15:00 +0000

Type Values Removed Values Added
Weaknesses CWE-367
References
Metrics threat_severity

None

 cvssV3_1

{'score': 7.0, 'vector': 'CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H'}

threat_severity

Moderate

Wed, 24 Jun 2026 18:45:00 +0000

Type Values Removed Values Added
Weaknesses CWE-400
CWE-665

Wed, 24 Jun 2026 17:15:00 +0000

Type Values Removed Values Added
Description In the Linux kernel, the following vulnerability has been resolved: Revert "wireguard: device: enable threaded NAPI" This reverts commit 933466fc50a8e4eb167acbd0d8ec96a078462e9c which is commit db9ae3b6b43c79b1ba87eea849fd65efa05b4b2e upstream. We have had three independent production user reports in combination with Cilium utilizing WireGuard as encryption underneath that k8s Pod E/W traffic to certain peer nodes fully stalled. The situation appears as follows: - Occurs very rarely but at random times under heavy networking load. - Once the issue triggers the decryption side stops working completely for that WireGuard peer, other peers keep working fine. The stall happens also for newly initiated connections towards that particular WireGuard peer. - Only the decryption side is affected, never the encryption side. - Once it triggers, it never recovers and remains in this state, the CPU/mem on that node looks normal, no leak, busy loop or crash. - bpftrace on the affected system shows that wg_prev_queue_enqueue fails, thus the MAX_QUEUED_PACKETS (1024 skbs!) for the peer's rx_queue is reached. - Also, bpftrace shows that wg_packet_rx_poll for that peer is never called again after reaching this state for that peer. For other peers wg_packet_rx_poll does get called normally. - Commit db9ae3b ("wireguard: device: enable threaded NAPI") switched WireGuard to threaded NAPI by default. The default has not been changed for triggering the issue, neither did CPU hotplugging occur (i.e. 5bd8de2 ("wireguard: queueing: always return valid online CPU in wg_cpumask_choose_online()")). - The issue has been observed with stable kernels of v5.15 as well as v6.1. It was reported to us that v5.10 stable is working fine, and no report on v6.6 stable either (somewhat related discussion in [0] though). - In the WireGuard driver the only material difference between v5.10 stable and v5.15 stable is the switch to threaded NAPI by default. [0] https://lore.kernel.org/netdev/CA+wXwBTT74RErDGAnj98PqS=wvdh8eM1pi4q6tTdExtjnokKqA@mail.gmail.com/ Breakdown of the problem: 1) skbs arriving for decryption are enqueued to the peer->rx_queue in wg_packet_consume_data via wg_queue_enqueue_per_device_and_peer. 2) The latter only moves the skb into the MPSC peer queue if it does not surpass MAX_QUEUED_PACKETS (1024) which is kept track in an atomic counter via wg_prev_queue_enqueue. 3) In case enqueueing was successful, the skb is also queued up in the device queue, round-robin picks a next online CPU, and schedules the decryption worker. 4) The wg_packet_decrypt_worker, once scheduled, picks these up from the queue, decrypts the packets and once done calls into wg_queue_enqueue_per_peer_rx. 5) The latter updates the state to PACKET_STATE_CRYPTED on success and calls napi_schedule on the per peer->napi instance. 6) NAPI then polls via wg_packet_rx_poll. wg_prev_queue_peek checks on the peer->rx_queue. It will wg_prev_queue_dequeue if the queue->peeked skb was not cached yet, or just return the latter otherwise. (wg_prev_queue_drop_peeked later clears the cache.) 7) From an ordering perspective, the peer->rx_queue has skbs in order while the device queue with the per-CPU worker threads from a global ordering PoV can finish the decryption and signal the skb PACKET_STATE_CRYPTED out of order. 8) A situation can be observed that the first packet coming in will be stuck waiting for the decryption worker to be scheduled for a longer time when the system is under pressure. 9) While this is the case, the other CPUs in the meantime finish decryption and call into napi_schedule. 10) Now in wg_packet_rx_poll it picks up the first in-order skb from the peer->rx_queue and sees that its state is still PACKET_STATE_UNCRYPTED. The NAPI poll routine then exits e ---truncated---
Title Revert "wireguard: device: enable threaded NAPI"
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: 2026-07-02T05:38:00.920Z

Reserved: 2026-06-09T07:44:35.371Z

Link: CVE-2026-52945

cve-icon Vulnrichment

No data.

cve-icon NVD

No data.

cve-icon Redhat

Severity : Moderate

Publid Date: 2026-06-24T00:00:00Z

Links: CVE-2026-52945 - Bugzilla

cve-icon OpenCVE Enrichment

Updated: 2026-06-28T14:45:17Z

Weaknesses
  • CWE-367

    Time-of-check Time-of-use (TOCTOU) Race Condition