Skip to content
NewA calmer, faster ZeroHour

Xen Security Advisory 511 v3 (CVE-2026-79603) - Unconditionally do TLB flushing ahead of page scrubbing

oss-security ()2 · 1 read
Part of a story covered by 4 sourcesXen publishes XSA-510 through XSA-513 (v3): four advisories covering hypervisor crashes, data exposure, oxenstored resource exhaustion, and Tapdisk out-of-bounds flawsMerged summary & timeline
AI summary · glm-5.3-flashSummaries by Vani AI

Xen Project released XSA-511 (CVE-2026-79603) fixing missing TLB flushes before page scrubbing that can leak x86 PV guest data.

Xen Security Advisory 511 v3 publicly discloses CVE-2026-79603, a TLB handling flaw in the Xen hypervisor on x86. x86 PV guests can free memory pages while a stale TLB entry still points to them, and Xen only flushes the TLB when the page is reused, potentially exposing stale data ahead of scrubbing. The advisory changes Xen to unconditionally flush the TLB ahead of page scrubbing. The issue was published in version 3 of the advisory.

Key facts
  • CVE-2026-79603 assigned to the TLB flush/scrubbing flaw
  • Affects x86 paravirtual (PV) guests releasing pages with stale TLB entries
  • Fix makes TLB flushing unconditional before page scrubbing
  • Published as public Xen Security Advisory XSA-511 v3
Analyze CVE-2026-79603 with Vani AIImpact, affected versions, exploitation, mitigations and detection

Vulnerabilities mentioned

All
CVEVulnerabilityCVSSEPSSSignalsPublished
CVE-2026-79603
Stale TLB Entry Flaw in Xen Hypervisor Lets x86 PV Guests Modify Scrubbed Pages
4.3<1%

From the original article

Posted by Xen . org security team on Sep 08 Xen Security Advisory CVE-2026-79603 / XSA-511 version 3 Unconditionally do TLB flushing ahead of page scrubbing UPDATES IN VERSION 3 ==================== Public release. ISSUE DESCRIPTION ================= x86 PV guests can free memory pages while still keeping a stale TLB entry pointing to them.

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.