CVE-2026-99236

A flaw was found in the Linux kernel. A local user can trigger out-of-bounds memory access to achieve privilege escalation or arbitrary code execution by submitting an incomplete event filter expression containing trailing logical operators. Alternatively, exploiting this flaw can expose sensitive kernel memory or crash the operating system, resulting in a Denial of Service (DoS).

This vulnerability is rated as an Important severity impact for Red Hat Enterprise Linux because unprivileged local attackers can exploit it to gain root privileges or force a kernel crash. Under default system configurations, trace event filtering can be reached by non-root accounts via the performance event subsystem without administrative capabilities. While the flaw cannot be triggered across the network, the ability of local users to corrupt kernel memory and hijack control flow warrants an Important rating rather than Moderate.

The CVSS score(s) listed for this vulnerability do not reflect the associated product's status, and are included for informational purposes to better understand the severity of this vulnerability.

--- CME Mitigation Guidance (2026-11-04) --- Disclaimer: The following mitigation solutions are provided as a courtesy, pending the application of an official patch. This content was generated utilizing AI tools in conjunction with data from https://cmetaxonomy.org. Customers are advised to independently evaluate and test these temporary mitigations within their own infrastructure.

The highest-value action is to block unprivileged access to perf_event_open and tracing event filter ioctls to prevent incomplete expressions from reaching predicate_parse(): upgrade to the corresponding fixed Red Hat builds, which reject outer-loop fallthrough with -EINVAL when predicates have already been parsed, closing all 2 injection vectors (PERF_EVENT_IOC_SET_FILTER ioctl on self-targeted perf counting events, tracing event filter configuration interface). Until patched, this is only a concern on Linux systems where local users can invoke perf_event_open and ioctls or access trace event filters — Non-Linux platforms are unaffected — so Restrict and audit local user access to the perf_event_open subsystem and trace event filter configuration interfaces. As a compensating measure, Enable KASAN or HWASan runtime memory error detection to detect out-of-bounds reads in kernel slab memory, Enable Supervisor Mode Execution Prevention to prevent execution of userspace payload code during privilege escalation, Enforce Control Flow Integrity or Shadow Call Stack to prevent unauthorized redirection via corrupted branch targets, Enable kernel memory zero-initialization via init_on_alloc to mitigate reuse of stale slab data in branch targets, Maintain SELinux in enforcing mode to restrict access to tracing interfaces and confine privilege escalation blast radius, and Map local users to confined SELinux domains to limit access to tracing interfaces and post-escalation capabilities. These controls limit blast radius and contain exploitation, but they do not fix the underlying incomplete filter expression out-of-bounds read defect — apply the upstream patch to fully remediate.

Highest-impact CME: CME-601 Kernel-Level Syscall Filtering (seccomp) (high)