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

udp: clear skb->dev before running a sockmap verdict

On the UDP receive path skb->dev is repurposed as dev_scratch (the
truesize/state cache set by udp_set_dev_scratch()), through the
union { struct net_device *dev; unsigned long dev_scratch; } in sk_buff.

When a UDP socket is in a sockmap, sk_data_ready is
sk_psock_verdict_data_ready(), which calls udp_read_skb() -> recv_actor()
(sk_psock_verdict_recv) to run the attached SK_SKB verdict program in softirq.
If that program calls a socket-lookup helper (bpf_sk_lookup_tcp/udp,
bpf_skc_lookup_tcp), bpf_skc_lookup() does:

if (skb->dev)
caller_net = dev_net(skb->dev);

skb->dev still holds the dev_scratch value (a non-NULL integer), so dev_net()
dereferences it as a struct net_device * and the kernel takes a general
protection fault on a non-canonical address in softirq:

Oops: general protection fault, probably for non-canonical address 0x1010000800004a0
CPU: 1 UID: 0 PID: 1406 Comm: syz.2.19 Not tainted 7.1.0-rc6 #1 PREEMPT(full)
RIP: 0010:bpf_skc_lookup net/core/filter.c:7033 [inline]
RIP: 0010:bpf_sk_lookup+0x45/0x160 net/core/filter.c:7047
Call Trace:
<IRQ>
bpf_prog_4675cb904b7071f8+0x12e/0x14e
bpf_prog_run_pin_on_cpu+0xc6/0x1f0
sk_psock_verdict_recv+0x1ba/0x350
udp_read_skb+0x31a/0x370
sk_psock_verdict_data_ready+0x2e3/0x600
__udp_enqueue_schedule_skb+0x4c8/0x650
udpv6_queue_rcv_one_skb+0x3ec/0x740
udp6_unicast_rcv_skb+0x11d/0x140
ip6_protocol_deliver_rcu+0x61e/0x950
ip6_input_finish+0xa9/0x150
NF_HOOK+0x286/0x2f0
ip6_input+0x117/0x220
NF_HOOK+0x286/0x2f0
__netif_receive_skb+0x85/0x200
process_backlog+0x374/0x9a0
__napi_poll+0x4f/0x1c0
net_rx_action+0x3b0/0x770
handle_softirqs+0x15a/0x460
do_softirq+0x57/0x80
</IRQ>

The rmem charge that dev_scratch accounted for is released by skb_recv_udp() on
dequeue, just above, so the scratch is dead by the time recv_actor() runs. Clear
skb->dev so bpf_skc_lookup() falls back to sock_net(skb->sk), which
skb_set_owner_sk_safe() set just above.
Published: 2026-06-25
Score: n/a
EPSS: n/a
KEV: No
Impact: n/a
Action: n/a
AI Analysis

Impact

In the Linux kernel, an improper handling of the skb->dev field in the UDP receive path can cause a general protection fault. The field, repurposed as dev_scratch during an in‑flight UDP packet, remains non‑NULL when a sockmap verdict program performs a socket lookup. The lookup function interprets this value as a net_device pointer, dereferencing an invalid memory address and crashing the kernel with an oops. The crash degrades system availability and requires a reboot, providing an attack surface for denial of service.

Affected Systems

The issue affects all Linux kernel implementations that incorporate the buggy udp code before the patch that clears skb->dev before running a sockmap verdict. The commit references in the advisory (e.g., 1b585673a2249f13678e7ac443ac683ba767e0b6) represent the fixes applied. No specific version numbers are listed, so any supported kernel that has not yet applied this change is vulnerable.

Risk and Exploitability

The vulnerability is exploitable by an attacker who can craft a UDP packet that triggers a sockmap verdict program using socket‑lookup helpers, or by injecting a malicious BPF program that exercises the same code path. The lack of an EPSS score and absence from the KEV catalog suggest no publicly documented exploits, but the code path is reachable through normal networking operations and BPF interfaces. Because the fault occurs in softirq, it is straightforward to trigger a panic once the conditions are satisfied, and the crash provides a deterministic denial of service vector.

Generated by OpenCVE AI on June 25, 2026 at 11:01 UTC.

Remediation

No vendor fix or workaround currently provided.

OpenCVE Recommended Actions

  • Update the Linux kernel to a release that includes the patch clearing skb->dev before running a sockmap verdict (commit 1b585673a2249f13678e7ac443ac683ba767e0b6 and associated updates).
  • If an immediate kernel upgrade is not possible, disable or remove any BPF sockmap verdict programs that invoke socket‑lookup helpers (bpf_sk_lookup_tcp/udp), as the failure occurs only when those helpers are used.
  • Continuously monitor kernel logs (dmesg, /var/log/kern.log) for general protection fault or oops messages that indicate the crash has occurred, and trigger alerts if such events are detected.

Generated by OpenCVE AI on June 25, 2026 at 11:01 UTC.

Tracking

Sign in to view the affected projects.

Advisories

No advisories yet.

History

Thu, 25 Jun 2026 11:30:00 +0000

Type Values Removed Values Added
Weaknesses CWE-476

Thu, 25 Jun 2026 09:15:00 +0000

Type Values Removed Values Added
Description In the Linux kernel, the following vulnerability has been resolved: udp: clear skb->dev before running a sockmap verdict On the UDP receive path skb->dev is repurposed as dev_scratch (the truesize/state cache set by udp_set_dev_scratch()), through the union { struct net_device *dev; unsigned long dev_scratch; } in sk_buff. When a UDP socket is in a sockmap, sk_data_ready is sk_psock_verdict_data_ready(), which calls udp_read_skb() -> recv_actor() (sk_psock_verdict_recv) to run the attached SK_SKB verdict program in softirq. If that program calls a socket-lookup helper (bpf_sk_lookup_tcp/udp, bpf_skc_lookup_tcp), bpf_skc_lookup() does: if (skb->dev) caller_net = dev_net(skb->dev); skb->dev still holds the dev_scratch value (a non-NULL integer), so dev_net() dereferences it as a struct net_device * and the kernel takes a general protection fault on a non-canonical address in softirq: Oops: general protection fault, probably for non-canonical address 0x1010000800004a0 CPU: 1 UID: 0 PID: 1406 Comm: syz.2.19 Not tainted 7.1.0-rc6 #1 PREEMPT(full) RIP: 0010:bpf_skc_lookup net/core/filter.c:7033 [inline] RIP: 0010:bpf_sk_lookup+0x45/0x160 net/core/filter.c:7047 Call Trace: <IRQ> bpf_prog_4675cb904b7071f8+0x12e/0x14e bpf_prog_run_pin_on_cpu+0xc6/0x1f0 sk_psock_verdict_recv+0x1ba/0x350 udp_read_skb+0x31a/0x370 sk_psock_verdict_data_ready+0x2e3/0x600 __udp_enqueue_schedule_skb+0x4c8/0x650 udpv6_queue_rcv_one_skb+0x3ec/0x740 udp6_unicast_rcv_skb+0x11d/0x140 ip6_protocol_deliver_rcu+0x61e/0x950 ip6_input_finish+0xa9/0x150 NF_HOOK+0x286/0x2f0 ip6_input+0x117/0x220 NF_HOOK+0x286/0x2f0 __netif_receive_skb+0x85/0x200 process_backlog+0x374/0x9a0 __napi_poll+0x4f/0x1c0 net_rx_action+0x3b0/0x770 handle_softirqs+0x15a/0x460 do_softirq+0x57/0x80 </IRQ> The rmem charge that dev_scratch accounted for is released by skb_recv_udp() on dequeue, just above, so the scratch is dead by the time recv_actor() runs. Clear skb->dev so bpf_skc_lookup() falls back to sock_net(skb->sk), which skb_set_owner_sk_safe() set just above.
Title udp: clear skb->dev before running a sockmap verdict
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: 2026-06-25T08:38:58.189Z

Reserved: 2026-06-09T07:44:35.390Z

Link: CVE-2026-53184

cve-icon Vulnrichment

No data.

cve-icon NVD

No data.

cve-icon Redhat

No data.

cve-icon OpenCVE Enrichment

Updated: 2026-06-25T11:15:10Z

Weaknesses