{"id":"CVE-2026-98863","kind":"vulnerability","aliases":["CVE-2026-98863"],"related":[],"title":"Linux kernel KVM: failed nested page mapping can leave stale guest memory mapped","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nKVM: x86: Re-pend GET_NESTED_STATE_PAGES if getting said pages fails\n\nRe-pend GET_NESTED_STATE_PAGES before exiting to userspace if getting the\nnested pages fails in the KVM_RUN path.  If userspace re-runs the vCPU, and\nvmcs02 holds valid PFNs from the *previous* run of L2, then KVM could\nre-enter L2 with stale, unpinned PFNs mapped into e.g. the vAPIC page.\n\nNote, both SVM and VMX (as of commit 11722439fb20 (\"KVM: nVMX: Ensure\nKVM_REQ_GET_NESTED_STATE_PAGES is cleared on VM-Exit\") ensure the request\nis cleared on VM-Exit (including the \"forced\" case), i.e. there is no risk\nof double-mapping due to emulated VMLAUNCH/VMRESUME/VMRUN *and* the request\ntrying to map the nested pages.","summary":"When KVM fails to map a nested guest's pages before returning to userspace, the request to retry is not re-queued, so a rerun of the vCPU can re-enter the guest with stale, unpinned host page frames mapped into its virtual APIC page. The sources do not describe a confirmed exploit path or attacker capability beyond this memory-safety consequence.","severity":"unknown","cvss":null,"epss":null,"exploited":false,"exploited_at":null,"products":[],"ecosystems":[],"tags":["hypervisor-hosts","kvm","linux-kernel","runtime","x86"],"not_affected_if":["nested-virtualization-disabled","no-kvm"],"how_to_check":"- Run `uname -r` and compare against the fixed kernel versions in the upstream commit referenced by the NVD entry\n- Check whether KVM nested virtualization is in use with `cat /sys/module/kvm_intel/parameters/nested` or `cat /sys/module/kvm_amd/parameters/nested`\n- unknown for other checks: the sources do not list affected version ranges","fix":"- Update the Linux kernel to a release that includes the upstream fix (the commit re-pends KVM_REQ_GET_NESTED_STATE_PAGES on failure); exact fixed versions: unknown\n- Distribution kernel advisories: unknown","status":"open","published_at":"2026-10-18T16:17:40+00:00","first_seen_at":"2026-10-23T21:57:17+00:00","triaged_at":"2026-10-25T05:12:52+00:00","updated_at":"2026-10-25T05:12:52+00:00","category":"runtime","confidence":0.4,"triage":{"at":"2026-10-25T05:12:52Z","model":"claude-haiku-5-5","output":{"fix":"- Update the Linux kernel to a release that includes the upstream fix (the commit re-pends KVM_REQ_GET_NESTED_STATE_PAGES on failure); exact fixed versions: unknown\n- Distribution kernel advisories: unknown","reach":"local","title":"Linux kernel KVM: failed nested page mapping can leave stale guest memory mapped","impact":"medium","summary":"When KVM fails to map a nested guest's pages before returning to userspace, the request to retry is not re-queued, so a rerun of the vCPU can re-enter the guest with stale, unpinned host page frames mapped into its virtual APIC page. The sources do not describe a confirmed exploit path or attacker capability beyond this memory-safety consequence.","category":"runtime","nicknames":[],"who_cares":["linux-kernel","kvm","x86","hypervisor-hosts"],"confidence":0.4,"needs_auth":"unknown","worst_case":"unknown; stale host memory mapped into a guest, impact not stated","analyst_note":"Check the NVD entry and the linked upstream kernel commit for affected version ranges, fixed versions, and whether a guest can trigger the stale mapping.","how_to_check":"- Run `uname -r` and compare against the fixed kernel versions in the upstream commit referenced by the NVD entry\n- Check whether KVM nested virtualization is in use with `cat /sys/module/kvm_intel/parameters/nested` or `cat /sys/module/kvm_amd/parameters/nested`\n- unknown for other checks: the sources do not list affected version ranges","needs_analyst":true,"not_affected_if":["no-kvm","nested-virtualization-disabled"],"who_is_affected":"Hosts running Linux with KVM on x86 that use nested virtualization (guests that run their own hypervisor) are the relevant systems. Hosts without KVM or without nested virtualization enabled are unlikely to be exposed, but the sources do not state this explicitly.","user_interaction":"unknown","needs_special_setup":"yes","severity_if_unknown":"unknown"},"cost_usd":0.000721,"input_tokens":3579,"output_tokens":726,"severity_before":"unknown"},"analyst_status":null,"analyst_note":null,"search":"'-2026':15A,18A '-98863':16A,19A '11722439fb20':183C 'apic':74B 'attack':86B 'beyond':88B 'capabl':87B 'case':211C 'clear':194C,203C 'commit':182C 'confirm':82B 'consequ':93B 'could':160C 'cve':14A,17A 'describ':80B 'doubl':219C 'double-map':218C 'due':221C 'e.g':171C 'emul':223C 'ensur':186C,199C 'enter':61B,163C 'exit':126C,198C,207C 'exploit':83B 'fail':4A,31B,117C,134C 'follow':99C 'forc':210C 'frame':69B 'get':109C,114C,121C,130C,189C 'guest':11A,36B,63B 'hold':149C 'host':22B,67B 'hypervisor':21B 'hypervisor-host':20B 'i.e':212C 'includ':208C 'kernel':2A,26B,97C 'kvm':3A,23B,30B,104C,137C,159C,184C,187C 'l2':157C,164C 'leav':9A 'linux':1A,25B,96C 'linux-kernel':24B 'map':7A,13A,33B,70B,169C,220C,230C 'memori':12A,91B 'memory-safeti':90B 'nest':5A,35B,110C,122C,132C,190C,232C 'note':175C 'nvmx':185C 'page':6A,38B,68B,75B,112C,116C,124C,133C,174C,192C,233C 'path':84B,139C 'pend':108C,120C 'pfns':151C,168C 'previous':154C 'queu':51B 're':50B,60B,107C,119C,143C,162C 're-ent':59B,161C 're-pend':106C,118C 're-queu':49B 're-run':142C 'req':188C 'request':44B,201C,227C 'rerun':54B 'resolv':103C 'retri':46B 'return':40B 'risk':216C 'run':138C,144C,155C 'runtim':27B 'safeti':92B 'said':115C 'sourc':77B 'stale':10A,65B,166C 'state':111C,123C,191C 'svm':177C 'tri':228C 'unpin':66B,167C 'userspac':42B,128C,141C 'valid':150C 'vapic':173C 'vcpu':57B,146C 'virtual':73B 'vm':197C,206C 'vm-exit':196C,205C 'vmcs02':148C 'vmlaunch/vmresume/vmrun':224C 'vmx':179C 'vulner':100C 'x86':28B,105C","nicknames":[],"mitigations":[],"who_affected":"Hosts running Linux with KVM on x86 that use nested virtualization (guests that run their own hypervisor) are the relevant systems. Hosts without KVM or without nested virtualization enabled are unlikely to be exposed, but the sources do not state this explicitly.","tier":"C","verified":{},"analyst_revisit_at":null,"first_public_at":"2026-10-18T16:17:40+00:00","kev":null,"cvss_vector":null,"cvss_by":null,"poc_count":0,"sources":[{"source":"nvd","external_id":"CVE-2026-98863","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-98863","severity":null,"published_at":"2026-10-18T16:17:40+00:00","modified_at":"2026-10-19T10:19:50+00:00","first_seen_at":"2026-10-23T21:57:17+00:00","fetched_at":"2026-10-23T22:28:56+00:00","note":null}],"documents":[],"notice":"Free to read at a human pace. An API with keys and an MCP server are in preparation; early access: hello@infrastack.ai"}