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

bpf: Allow LPM map access from sleepable BPF programs

trie_lookup_elem() annotates its rcu_dereference_check() walks with
only rcu_read_lock_bh_held(). Because rcu_dereference_check(p, c)
resolves to "c || rcu_read_lock_held()", this passes for XDP/NAPI and
classic RCU readers but fails for sleepable BPF programs, which enter
via __bpf_prog_enter_sleepable() and hold only rcu_read_lock_trace().

trie_update_elem() and trie_delete_elem() have the same problem in a
different form: they walk the trie with plain rcu_dereference(), which
asserts rcu_read_lock_held() unconditionally. Both are reachable from
sleepable BPF programs via the bpf_map_update_elem / bpf_map_delete_elem
helpers, and from the syscall path under classic rcu_read_lock(). In
the writer paths the trie is actually protected by trie->lock (an
rqspinlock taken across the walk); we never relied on the RCU read-side
lock to keep nodes alive there.

A sleepable LSM hook that ends up touching an LPM trie therefore
triggers lockdep on debug kernels:

=============================
WARNING: suspicious RCU usage
7.1.0-... Tainted: G E
-----------------------------
kernel/bpf/lpm_trie.c:249 suspicious rcu_dereference_check() usage!
1 lock held by net_tests/540:
#0: (rcu_tasks_trace_srcu_struct){....}-{0:0},
at: __bpf_prog_enter_sleepable+0x26/0x280
Call Trace:
dump_stack_lvl
lockdep_rcu_suspicious
trie_lookup_elem
bpf_prog_..._enforce_security_socket_connect
bpf_trampoline_...
security_socket_connect
__sys_connect
do_syscall_64

This is lockdep-only -- no UAF, since Tasks Trace RCU does serialize
against the trie's reclaim path -- but it spams the console once per
distinct callsite on every debug kernel running a sleepable BPF LSM
that touches an LPM trie, which is increasingly common.

For the lookup path, switch the rcu_dereference_check() annotation
from rcu_read_lock_bh_held() to bpf_rcu_lock_held(), which accepts all
three contexts (classic, BH, Tasks Trace). Other map types already
follow this convention.

For trie_update_elem() and trie_delete_elem(), annotate the walks as
rcu_dereference_protected(*p, 1) -- matching trie_free() in the same
file -- since trie->lock is held across the walk. rqspinlock has no
lockdep_map, so the predicate degenerates to '1' rather than
lockdep_is_held(&trie->lock); the protection is real but not
machine-verifiable. trie_get_next_key() also uses bare
rcu_dereference() but is reachable only from the BPF syscall, which
holds classic rcu_read_lock() before dispatching, so it is left
untouched.
Published: 2026-07-25
Score: 5.5 Medium
EPSS: < 1% Very Low
KEV: No
Impact: n/a
Action: n/a
AI Analysis

Impact

The kernel bug originates from incorrect RCU locking annotations in the LPM trie functions used by BPF programs. When sleepable BPF programs invoke map lookup, update, or delete helpers, the code erroneously checks for a read lock that is not held, resulting in lockdep warnings. The bug does not cause a use‑after‑free, data corruption, or code execution; it merely generates diagnostic console spam and indicates a potential synchronization issue.

Affected Systems

All Linux kernel versions that contain the buggy BPF LPM trie code and run sleepable BPF programs that access LPM maps are affected. The fix was applied in later kernel revisions; any kernel before the commit introducing the corrected RCU annotations may exhibit the warning.

Risk and Exploitability

The CVSS score of 5.5 indicates moderate severity. With an EPSS score of less than 1 % and no listing in the CISA KEV catalog, the likelihood of exploitation is very low. This correctness bug causes lockdep warnings and excessive console output in debug kernels but does not result in use‑after‑free, data corruption, or code execution; it merely signals a potential synchronization issue. Applying the upstream kernel patch removes the warning and eliminates the potential for a race condition, aligning with CWE‑821.

Generated by OpenCVE AI on August 4, 2026 at 14:38 UTC.

Remediation

No vendor fix or workaround currently provided.

OpenCVE Recommended Actions

  • Upgrade the Linux kernel to a version that includes the patch fixing the RCU annotation in the LPM trie – the merge commit series that added bpf_rcu_lock_held() and rcu_dereference_protected().
  • If an immediate kernel upgrade is not feasible, avoid using sleepable BPF programs that access LPM maps until the patch is deployed; the error is confined to debug kernels where lockdep is enabled.
  • Disable lockdep or console logging of suspicious RCU usage only as a temporary measure to prevent console spam, though this does not fix the underlying synchronization issue.

Generated by OpenCVE AI on August 4, 2026 at 14:38 UTC.

Tracking

Sign in to view the affected projects.

Advisories
Source ID Title
Debian DLA Debian DLA DLA-4720-1 linux security update
Debian DLA Debian DLA DLA-4723-1 linux-6.1 security update
Debian DLA Debian DLA DLA-4724-1 linux-6.12 new package
Debian DSA Debian DSA DSA-6405-1 linux security update
History

Sat, 01 Aug 2026 02:30:00 +0000

Type Values Removed Values Added
Weaknesses CWE-823

Wed, 29 Jul 2026 12:15:00 +0000

Type Values Removed Values Added
Weaknesses CWE-821
References
Metrics threat_severity

None

 cvssV3_1

{'score': 5.5, 'vector': 'CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H'}

threat_severity

Low

Mon, 27 Jul 2026 06:00:00 +0000

Type Values Removed Values Added
Weaknesses CWE-823

Sat, 25 Jul 2026 09:30:00 +0000

Type Values Removed Values Added
Description In the Linux kernel, the following vulnerability has been resolved: bpf: Allow LPM map access from sleepable BPF programs trie_lookup_elem() annotates its rcu_dereference_check() walks with only rcu_read_lock_bh_held(). Because rcu_dereference_check(p, c) resolves to "c || rcu_read_lock_held()", this passes for XDP/NAPI and classic RCU readers but fails for sleepable BPF programs, which enter via __bpf_prog_enter_sleepable() and hold only rcu_read_lock_trace(). trie_update_elem() and trie_delete_elem() have the same problem in a different form: they walk the trie with plain rcu_dereference(), which asserts rcu_read_lock_held() unconditionally. Both are reachable from sleepable BPF programs via the bpf_map_update_elem / bpf_map_delete_elem helpers, and from the syscall path under classic rcu_read_lock(). In the writer paths the trie is actually protected by trie->lock (an rqspinlock taken across the walk); we never relied on the RCU read-side lock to keep nodes alive there. A sleepable LSM hook that ends up touching an LPM trie therefore triggers lockdep on debug kernels: ============================= WARNING: suspicious RCU usage 7.1.0-... Tainted: G E ----------------------------- kernel/bpf/lpm_trie.c:249 suspicious rcu_dereference_check() usage! 1 lock held by net_tests/540: #0: (rcu_tasks_trace_srcu_struct){....}-{0:0}, at: __bpf_prog_enter_sleepable+0x26/0x280 Call Trace: dump_stack_lvl lockdep_rcu_suspicious trie_lookup_elem bpf_prog_..._enforce_security_socket_connect bpf_trampoline_... security_socket_connect __sys_connect do_syscall_64 This is lockdep-only -- no UAF, since Tasks Trace RCU does serialize against the trie's reclaim path -- but it spams the console once per distinct callsite on every debug kernel running a sleepable BPF LSM that touches an LPM trie, which is increasingly common. For the lookup path, switch the rcu_dereference_check() annotation from rcu_read_lock_bh_held() to bpf_rcu_lock_held(), which accepts all three contexts (classic, BH, Tasks Trace). Other map types already follow this convention. For trie_update_elem() and trie_delete_elem(), annotate the walks as rcu_dereference_protected(*p, 1) -- matching trie_free() in the same file -- since trie->lock is held across the walk. rqspinlock has no lockdep_map, so the predicate degenerates to '1' rather than lockdep_is_held(&trie->lock); the protection is real but not machine-verifiable. trie_get_next_key() also uses bare rcu_dereference() but is reachable only from the BPF syscall, which holds classic rcu_read_lock() before dispatching, so it is left untouched.
Title bpf: Allow LPM map access from sleepable BPF programs
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: 2026-07-25T08:50:11.522Z

Reserved: 2026-07-19T15:36:31.782Z

Link: CVE-2026-64352

cve-icon Vulnrichment

No data.

cve-icon NVD

Status : Awaiting Analysis

Published: 2026-07-25T10:17:17.547

Modified: 2026-08-11T15:00:57.447

Link: CVE-2026-64352

cve-icon Redhat

Severity : Low

Publid Date: 2026-07-25T00:00:00Z

Links: CVE-2026-64352 - Bugzilla

cve-icon OpenCVE Enrichment

Updated: 2026-08-04T14:45:03Z

Weaknesses
  • CWE-821

    Incorrect Synchronization