# CVE-2026-99472: KVM: pfncache: use a dedicated invalidation sequence for cache refresh

Status: published. Published: 2026-11-19. Updated: 2026-11-22.

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

KVM: pfncache: use a dedicated invalidation sequence for cache refresh

The gfn_to_pfn_cache refresh path guards against mmu notifier
invalidations which complete while it has dropped gpc->lock for the
HVA->PFN lookup: hva_to_pfn_retry() samples kvm->mmu_invalidate_seq
and retries if it changed, or if mn_active_invalidate_count is still
elevated.

That is insufficient for HVA-based caches. mmu_invalidate_seq is only
advanced by kvm_mmu_invalidate_end() when the invalidated range
overlaps a memslot, and an HVA-based cache (e.g. the Xen shared_info
page mapped with KVM_XEN_ATTR_TYPE_SHARED_INFO_HVA) need not be backed
by any memslot at all. An invalidation of the cached HVA which starts
and ends entirely within the lookup window is thus invisible to the
retry check: mn_active_invalidate_count is back to zero and the
sequence never moved. The refresh then publishes a mapping of a page
which has already been freed, and the next reader dereferences it:

  BUG: KASAN: use-after-free in kvm_xen_shared_info_init+0x3c6/0x440
  Read of size 4 at addr ffff8880599c2900 by task syz.2.383/7257

Since gfn_to_pfn_cache_invalidate_start() deliberately skips caches
which are not currently valid (including one whose refresh is in
progress, as the refresh clears the valid flag before dropping the
lock), the retry check is the only line of defence, and it must fire
for *any* invalidation, not just those hitting a memslot.

Add a dedicated kvm->gpc_invalidate_seq, incremented by every
kvm_mmu_notifier_invalidate_range_end() under mn_invalidate_lock
before mn_active_invalidate_count is decremented, and check it in
hva_to_pfn_retry() instead of mmu_invalidate_seq. Incrementing in
range_end() in the same critical section as the in-progress count
also closes the variant where the cache is activated with the
contested HVA only after invalidate_range_start() has run.

The same bug is also reachable through the per-vCPU vcpu_info cache
(KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO_HVA), where the stale mapping is
then dereferenced by kvm_setup_guest_pvclock() on the next KVM_RUN:

  BUG: KASAN: use-after-free in kvm_setup_guest_pvclock+0x5bf/0x660

This intentionally makes refresh retry on *unrelated* mmu notifier
events; restoring precision (and reworking the GPC locking more
generally) is left for a subsequent series.

Reproducers: https://david.woodhou.se/xen_shinfo_race.c
             https://david.woodhou.se/vcpu_info_race.c

## Remediation

No remediation reported in collected sources

## Affected products and versions

- Linux / Linux: ≥ b9220d32799a62c634f7b0b3618da52772626952; < fabc002f6396202b5ad7ad4ceec24d27aad90f3f (affected, git); ≥ b9220d32799a62c634f7b0b3618da52772626952; < da260bcb56feeb6489a83f310308453a0a7de25e (affected, git). Source: CNA: Linux
- Linux / Linux: 6.9 (affected); ≥ 0; < 6.9 (unaffected, semver); ≥ 7.3.2; ≤ 7.3.* (unaffected, semver); ≥ 7.4-rc1; ≤ * (unaffected, original_commit_for_fix). Source: CNA: Linux

## Severity and prioritization

- CVSS 3.1: 7.8 (high); CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H; source: CNA: Linux

- CVSS 3.1: 7.8 (high); CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H; source: NVD: 416baaa9-dc9f-4396-8d5f-8c081fb06d67

- CISA KEV: No listing in collected sources

- EPSS: 0.16% as of 2026-11-21; percentile 5.50%

- SSVC: Not available

- Weaknesses: Not reported

## Exploit evidence

No exploit evidence in collected sources. This is not proof of no exploitation.

## References

- https://git.kernel.org/stable/c/fabc002f6396202b5ad7ad4ceec24d27aad90f3f (CNA: Linux)

- https://git.kernel.org/stable/c/da260bcb56feeb6489a83f310308453a0a7de25e (CNA: Linux)

- https://git.kernel.org/stable/c/da260bcb56feeb6489a83f310308453a0a7de25e (416baaa9-dc9f-4396-8d5f-8c081fb06d67)

- https://git.kernel.org/stable/c/fabc002f6396202b5ad7ad4ceec24d27aad90f3f (416baaa9-dc9f-4396-8d5f-8c081fb06d67)

## Provenance

- CNA: Linux: https://www.cve.org/CVERecord?id=CVE-2026-99472; retrieved 2026-11-22; source updated 2026-11-22

- NVD: https://nvd.nist.gov/vuln/detail/CVE-2026-99472; retrieved 2026-11-22; source updated 2026-11-22

- FIRST EPSS: https://api.first.org/data/v1/epss?cve=CVE-2026-99472; retrieved 2026-11-22; source updated Not reported

Revision: 7d4abb60d94e572cd8eb73e00d94396cd4a18353413be9235c1317121d7e1f87

Canonical: https://y2.dev/cve/CVE-2026-99472

## Agent and API access

Public CVE pages, Markdown, and JSON are free to read without an account or API key.

- Public catalog: https://y2.dev/cve/index.json. Follow nextCursor with the same search, filters, sort, and limit (1–100). A null cursor marks the end.

- WebMCP: search_cve_disclosures, get_cve_disclosure, get_cve_page_markdown, get_y2_cve_access. Available in compatible browsers.

- Cyber API: https://api.y2.dev/api/v2/cyber/cves and https://api.y2.dev/api/v2/cyber/cves/{cveId}.

- Y2 Pro or Elite: create a workspace API key with the intel:cyber scope and send Authorization: Bearer <Y2_API_KEY>. Setup: https://y2.dev/docs/api/create-key.

- x402: Pay per request without an account or API key. Omit Authorization, read the HTTP 402 PAYMENT-REQUIRED challenge, validate its current requirements, and retry with PAYMENT-SIGNATURE through an authorized payment client. These browser tools never make payments. Guide: https://y2.dev/docs/api/x402.

- OpenAPI: https://y2.dev/api/openapi.yaml. Agent access metadata: https://y2.dev/cve/agent.json.

Collected public records only. Check source status for coverage and freshness.