{"api_version":"1","generated_at":"2026-09-25T22:41:42+00:00","cve":"CVE-2026-89775","urls":{"html":"https://cve.report/CVE-2026-89775","api":"https://cve.report/api/cve/CVE-2026-89775.json","docs":"https://cve.report/api","cve_org":"https://www.cve.org/CVERecord?id=CVE-2026-89775","nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-89775"},"summary":{"title":"KVM: arm64: Handle negative S1 walk levels in VNCR TLB size evaluation","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nKVM: arm64: Handle negative S1 walk levels in VNCR TLB size evaluation\n\nComputing the effects of a TLB invalidation involves looking at\nthe size of the mapping cached by the TLB. For S1 mappings such as\nVNCR, this is deducted from the combination of the base granule size\nand the mapping level.\n\nHowever, this implies that the S1 MMU is *on*. When the MMU is off,\nwe indicate this with the level being set to a \"creative\" value of\n-127 (S1_MMU_DISABLED).\n\nThis ends-up being misinterpreted by pgshift_level_to_ttl() as it\ndoesn't handle negative levels at all (the level is immediately cast\nto a u8 and only the bottom two bits considered), leading to an\ninvalidation size of 0. Not helpful.\n\nTidy-up pgshift_level_to_ttl() to handle these negative levels, and\nttl_to_size() to always return SZ_1G when no valid TTL is present.\nThis allows the removal of open-coded checks for similar situations.\n\nNote that the check for a negative value not explicitely checking for\nS1_MMU_DISABLED is deliberate, so that actual negative levels introduced\nwith LVA2 and D128 can take the same path if we ever support them.","state":"PUBLISHED","assigner":"Linux","published_at":"2026-09-16 09:17:07","updated_at":"2026-09-16 18:17:19"},"problem_types":[],"metrics":[{"version":"3.1","source":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","type":"Secondary","score":"9.3","severity":"CRITICAL","vector":"CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H","data":{"version":"3.1","vectorString":"CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H","baseScore":9.3,"baseSeverity":"CRITICAL","attackVector":"LOCAL","attackComplexity":"LOW","privilegesRequired":"NONE","userInteraction":"NONE","scope":"CHANGED","confidentialityImpact":"HIGH","integrityImpact":"HIGH","availabilityImpact":"HIGH"}},{"version":"3.1","source":"CNA","type":"DECLARED","score":"9.3","severity":"CRITICAL","vector":"CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H","data":{"baseScore":9.3,"baseSeverity":"CRITICAL","vectorString":"CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H","version":"3.1"}}],"references":[{"url":"http://www.openwall.com/lists/oss-security/2026/09/16/14","name":"http://www.openwall.com/lists/oss-security/2026/09/16/14","refsource":"af854a3a-2127-422b-91ae-364da2661108","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/be54a70067c7dbe1381157e8a238fa4b16ea9c05","name":"https://git.kernel.org/stable/c/be54a70067c7dbe1381157e8a238fa4b16ea9c05","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/1c9fca34b9625a67a7f1a03c8604f3df760c6c49","name":"https://git.kernel.org/stable/c/1c9fca34b9625a67a7f1a03c8604f3df760c6c49","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/8053393680d4fa3eb962667d2be95dd39f0940e5","name":"https://git.kernel.org/stable/c/8053393680d4fa3eb962667d2be95dd39f0940e5","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-89775","name":"CVE Program record","refsource":"CVE.ORG","tags":["canonical"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-89775","name":"NVD vulnerability detail","refsource":"NVD","tags":["canonical","analysis"]}],"affected":[{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 7270cc9157f474dfc46750a34c9d7defc686b2eb be54a70067c7dbe1381157e8a238fa4b16ea9c05 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 7270cc9157f474dfc46750a34c9d7defc686b2eb 1c9fca34b9625a67a7f1a03c8604f3df760c6c49 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 7270cc9157f474dfc46750a34c9d7defc686b2eb 8053393680d4fa3eb962667d2be95dd39f0940e5 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 6.16","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.16 semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.18.51 6.18.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.2.5 7.2.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.3-rc1 * original_commit_for_fix","platforms":[]}],"timeline":[],"solutions":[],"workarounds":[],"exploits":[],"credits":[],"nvd_cpes":[],"vendor_comments":[],"enrichments":{"kev":null,"epss":{"cve_year":"2026","cve_id":"89775","cve":"CVE-2026-89775","epss":"0.001840000","percentile":"0.082350000","score_date":"2026-09-16","updated_at":"2026-09-17 00:07:17"},"legacy_qids":[]},"source_records":{"cve_program":{"containers":{"adp":[{"providerMetadata":{"dateUpdated":"2026-09-16T17:07:09.850Z","orgId":"af854a3a-2127-422b-91ae-364da2661108","shortName":"CVE"},"references":[{"url":"http://www.openwall.com/lists/oss-security/2026/09/16/14"}],"title":"CVE Program Container"}],"cna":{"affected":[{"defaultStatus":"unaffected","product":"Linux","programFiles":["arch/arm64/kvm/nested.c"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"lessThan":"be54a70067c7dbe1381157e8a238fa4b16ea9c05","status":"affected","version":"7270cc9157f474dfc46750a34c9d7defc686b2eb","versionType":"git"},{"lessThan":"1c9fca34b9625a67a7f1a03c8604f3df760c6c49","status":"affected","version":"7270cc9157f474dfc46750a34c9d7defc686b2eb","versionType":"git"},{"lessThan":"8053393680d4fa3eb962667d2be95dd39f0940e5","status":"affected","version":"7270cc9157f474dfc46750a34c9d7defc686b2eb","versionType":"git"}]},{"defaultStatus":"affected","product":"Linux","programFiles":["arch/arm64/kvm/nested.c"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"status":"affected","version":"6.16"},{"lessThan":"6.16","status":"unaffected","version":"0","versionType":"semver"},{"lessThanOrEqual":"6.18.*","status":"unaffected","version":"6.18.51","versionType":"semver"},{"lessThanOrEqual":"7.2.*","status":"unaffected","version":"7.2.5","versionType":"semver"},{"lessThanOrEqual":"*","status":"unaffected","version":"7.3-rc1","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"cpeMatch":[{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"6.18.51","versionStartIncluding":"6.16","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.2.5","versionStartIncluding":"6.16","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.3-rc1","versionStartIncluding":"6.16","vulnerable":true}],"negate":false,"operator":"OR"}]}],"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nKVM: arm64: Handle negative S1 walk levels in VNCR TLB size evaluation\n\nComputing the effects of a TLB invalidation involves looking at\nthe size of the mapping cached by the TLB. For S1 mappings such as\nVNCR, this is deducted from the combination of the base granule size\nand the mapping level.\n\nHowever, this implies that the S1 MMU is *on*. When the MMU is off,\nwe indicate this with the level being set to a \"creative\" value of\n-127 (S1_MMU_DISABLED).\n\nThis ends-up being misinterpreted by pgshift_level_to_ttl() as it\ndoesn't handle negative levels at all (the level is immediately cast\nto a u8 and only the bottom two bits considered), leading to an\ninvalidation size of 0. Not helpful.\n\nTidy-up pgshift_level_to_ttl() to handle these negative levels, and\nttl_to_size() to always return SZ_1G when no valid TTL is present.\nThis allows the removal of open-coded checks for similar situations.\n\nNote that the check for a negative value not explicitely checking for\nS1_MMU_DISABLED is deliberate, so that actual negative levels introduced\nwith LVA2 and D128 can take the same path if we ever support them."}],"metrics":[{"cvssV3_1":{"baseScore":9.3,"baseSeverity":"CRITICAL","vectorString":"CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H","version":"3.1"},"scenarios":[{"lang":"en","value":"AV:L - A nested arm64 L1 hypervisor reaches the bug via KVM_RUN when host KVM handles VNCR_EL2 TLB invalidation (MMU notifiers, memslot unmap, or guest EL2 TLBI into kvm_invalidate_vncr_ipa/invalidate_vncr_va). This is local KVM ioctl/guest execution, not a network, adjacent-radio, or physical-device path.\nAC:L - An L1 guest deterministically clears SCTLR_EL2.M and programs TCR_EL2.TG0 to 16KB or 64KB so S1_MMU_DISABLED (-127) is mis-cast as walk level 1 and ttl_to_size() returns 0, skipping all VNCR invalidation. Nested virt on arm64 cloud hosts is an attacker-selectable deployment, not an uncontrollable race.\nPR:N - Exploitation requires no host root or init-namespace capabilities beyond running hypervisor code in a tenant KVM guest on an arm64 host with nested virtualization enabled; the attacker operates entirely from within their assigned VM, consistent with other arm64 KVM nested-virt CVE scoring.\nUI:N - No victim user or administrator action is required beyond the attacker operating their own nested-virtualization workload; SCTLR/TCR setup, VNCR faults, guest TLBI, and balloon or memslot-unmap MMU notifiers are fully attacker-driven.\nS:C - Missed VNCR invalidation leaves a writable host per-CPU fixmap covering a guest IPA after the backing page is unmapped, so host KVM keeps reading and writing reused host physical pages while emulating EL2 register state, crossing the guest-to-hypervisor security boundary.\nC:H - The stale VNCR pseudo-TLB retains the old HPA and is remapped with PAGE_KERNEL; after MMU-notifier unmap the host reads reused pages through the fixmap during VNCR/sysreg emulation, yielding a use-after-free disclosure primitive across guest and host memory.\nI:H - Hardware NV2 and kvm_map_l1_vncr() keep a writable kernel mapping of the stale HPA, so nested EL2 register writes land in pages the host already freed and reused, giving a use-after-free write primitive against host kernel or co-resident memory.\nA:H - Accessing an unmapped or reused HPA through the VNCR fixmap can oops or panic the host kernel, and corrupting reused host pages can hang or crash the hypervisor and co-resident VMs; the attacker can retrigger the missed-invalidation sequence."}]}],"providerMetadata":{"dateUpdated":"2026-09-16T14:38:27.335Z","orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux"},"references":[{"url":"https://git.kernel.org/stable/c/be54a70067c7dbe1381157e8a238fa4b16ea9c05"},{"url":"https://git.kernel.org/stable/c/1c9fca34b9625a67a7f1a03c8604f3df760c6c49"},{"url":"https://git.kernel.org/stable/c/8053393680d4fa3eb962667d2be95dd39f0940e5"}],"title":"KVM: arm64: Handle negative S1 walk levels in VNCR TLB size evaluation","x_generator":{"engine":"bippy-1.2.0"}}},"cveMetadata":{"assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","assignerShortName":"Linux","cveId":"CVE-2026-89775","datePublished":"2026-09-16T08:39:16.109Z","dateReserved":"2026-09-11T19:38:34.765Z","dateUpdated":"2026-09-16T17:07:09.850Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"},"nvd":{"publishedDate":"2026-09-16 09:17:07","lastModifiedDate":"2026-09-16 18:17:19","problem_types":[],"metrics":{"cvssMetricV31":[{"source":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","type":"Secondary","cvssData":{"version":"3.1","vectorString":"CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H","baseScore":9.3,"baseSeverity":"CRITICAL","attackVector":"LOCAL","attackComplexity":"LOW","privilegesRequired":"NONE","userInteraction":"NONE","scope":"CHANGED","confidentialityImpact":"HIGH","integrityImpact":"HIGH","availabilityImpact":"HIGH"},"exploitabilityScore":2.5,"impactScore":6}]},"configurations":[]},"legacy_mitre":{"record":{"CveYear":"2026","CveId":"89775","Ordinal":"1","Title":"KVM: arm64: Handle negative S1 walk levels in VNCR TLB size eval","CVE":"CVE-2026-89775","Year":"2026"},"notes":[{"CveYear":"2026","CveId":"89775","Ordinal":"1","NoteData":"In the Linux kernel, the following vulnerability has been resolved:\n\nKVM: arm64: Handle negative S1 walk levels in VNCR TLB size evaluation\n\nComputing the effects of a TLB invalidation involves looking at\nthe size of the mapping cached by the TLB. For S1 mappings such as\nVNCR, this is deducted from the combination of the base granule size\nand the mapping level.\n\nHowever, this implies that the S1 MMU is *on*. When the MMU is off,\nwe indicate this with the level being set to a \"creative\" value of\n-127 (S1_MMU_DISABLED).\n\nThis ends-up being misinterpreted by pgshift_level_to_ttl() as it\ndoesn't handle negative levels at all (the level is immediately cast\nto a u8 and only the bottom two bits considered), leading to an\ninvalidation size of 0. Not helpful.\n\nTidy-up pgshift_level_to_ttl() to handle these negative levels, and\nttl_to_size() to always return SZ_1G when no valid TTL is present.\nThis allows the removal of open-coded checks for similar situations.\n\nNote that the check for a negative value not explicitely checking for\nS1_MMU_DISABLED is deliberate, so that actual negative levels introduced\nwith LVA2 and D128 can take the same path if we ever support them.","Type":"Description","Title":"KVM: arm64: Handle negative S1 walk levels in VNCR TLB size eval"}]}}}