CVE-2026-31528: perf: Make sure to use pmu_ctx->pmu for groups
In the Linux kernel, the following vulnerability has been resolved:
perf: Make sure to use pmu_ctx->pmu for groups
Oliver reported that x86_pmu_del() ended up doing an out-of-bound memory access
when group_sched_in() fails and needs to roll back.
This *should* be handled by the transaction callbacks, but he found that when
the group leader is a software event, the transaction handlers of the wrong PMU
are used. Despite the move_group case in perf_event_open() and group_sched_in()
using pmu_ctx->pmu.
Turns out, inherit uses event->pmu to clone the events, effectively undoing the
move_group case for all inherited contexts. Fix this by also making inherit use
pmu_ctx->pmu, ensuring all inherited counters end up in the same pmu context.
Similarly, __perf_event_read() should use equally use pmu_ctx->pmu for the
group case.
Security readout for executives and security teams
Plain-English summary
A flaw in Linux performance monitoring can trigger an out-of-bounds kernel memory access during failed event-group scheduling and rollback. A local, low-privileged user may be able to compromise confidentiality, integrity, or availability. The supplied CVSS score is 7.8. No active exploitation is established by the provided sources.
Executive priority
Treat this as a high-priority kernel maintenance issue, especially for shared or multi-user Linux infrastructure. It is not supported as an emergency internet-wide threat because exploitation requires local access and no active exploitation is documented. Accelerate vendor-supported updates where untrusted users have local access.
Technical view
Inherited perf event groups can retain event->pmu instead of pmu_ctx->pmu. When a software event leads the group, rollback may invoke transaction handlers for the wrong PMU, causing x86_pmu_del() to access memory out of bounds. The fix consistently uses pmu_ctx->pmu for inherited group events and grouped reads.
Likely exposure
Exposure is limited to affected Linux kernels and requires local, low-privileged access without user interaction. Risk is higher on multi-user or shared systems where untrusted users can reach kernel performance-monitoring interfaces. The bundle lists affected versions but does not provide a clear fixed-version mapping for each maintained branch.
Exploitation context
The CVSS vector indicates a local, low-complexity, low-privilege attack with potential high impact across confidentiality, integrity, and availability. CISA KEV status is false, and the supplied sources provide no evidence of public exploitation, exploit availability, or attacks in the wild.
Researcher notes
The reported condition is an x86 perf rollback-path out-of-bounds access involving mixed-PMU groups and inherited contexts. The sources describe the faulty PMU selection and corrective commits, but do not establish exploitability details, reliable impact outcomes, public proof-of-concept availability, or complete branch-specific fixed-version boundaries.
Mitigation direction
Consult Linux distribution advisories for fixed packages covering CVE-2026-31528.
Upgrade to a vendor-supported kernel containing the applicable cited stable fix.
Prioritize shared systems allowing untrusted local users to access performance-monitoring facilities.
Apply vendor-recommended temporary restrictions if immediate kernel replacement is unavailable.
Validation and detection
Inventory kernel versions across physical hosts, virtual machines, and container hosts.
Compare installed versions against authoritative distribution advisories for CVE-2026-31528.
Confirm the running kernel includes the applicable cited stable commit after remediation.
Identify systems where untrusted local users can access performance-monitoring interfaces.
Reboot where required, then verify the fixed kernel is actually running.
Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.
Potential ATT&CK relevance
Conservative CVE-to-ATT&CK context
These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.
ATT&CK lookup starting points
Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.
cve · low confidence lookup
CVE-2026-31528 mapping review
Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.
These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.
1CVSS vectors
3Timeline events
0ADP providers
6Source links
CVSS vector scores
1 official score
We collect every scored CVSS vector available in the official CNA and ADP containers. When more than one version is present, the table keeps the source vectors side by side instead of collapsing them into the highest score.