GHSA-99GX-MQGQ-5PQG
Vulnerability from github – Published: 2026-08-10 15:33 – Updated: 2026-08-19 18:32In the Linux kernel, the following vulnerability has been resolved:
KVM: SVM: Bump asid_generation on CPU online to avoid ASID collision after hotplug
If a vCPU stays scheduled out (or blocked) while the last pCPU it ran on goes through a hotplug cycle (online->offline->online), and the vCPU then resumes execution on the same pCPU, then it is possible for it to run with an ASID that has now been assigned to a different vCPU, resulting in stale TLB translations being used.
svm_enable_virtualization_cpu() resets asid_generation to 1 and sets next_asid to max_asid + 1 on every CPU online event, including hotplug cycles. Because next_asid starts beyond the pool boundary, the first call to new_asid() after an online event always wraps the pool, incrementing asid_generation to 2 and assigning ASIDs starting from min_asid.
Consider two vCPUs from different VMs, vCPU-A pinned to CPU-X holding asid_generation=2 and ASID=N from before the hotplug event:
-
CPU-X goes offline and back online: asid_generation resets to 1, next_asid = max_asid + 1.
-
One or more vCPUs migrate to CPU-X and call new_asid(), wrapping the pool and consuming ASIDs starting from min_asid. Eventually vCPU-B from a different VM is assigned asid_generation=2, ASID=N — the same ASID that vCPU-A held before the hotplug.
-
vCPU-A enters pre_svm_run() on CPU-X: current_vmcb->cpu is unchanged so the migration branch is skipped. Its saved asid_generation=2 matches sd->asid_generation=2, so the generation check silently passes and vCPU-A continues running with ASID=N — the same ASID just freshly assigned to vCPU-B.
Both vCPUs from different VMs now run on CPU-X with the same ASID, causing them to share NPT TLB entries and producing stale translations.
The collision manifests as a KVM internal error (Suberror: 1, emulation failure). The NPT page fault reports a faulting GPA far outside the VM's physical memory range — a sign of stale TLB translations being used. KVM falls back to instruction emulation, which fails on FPU/XSave instructions (XRSTOR, STMXCSR) that the emulator does not implement.
Fix this by incrementing asid_generation instead of resetting it to 1 in svm_enable_virtualization_cpu(). On module load, asid_generation starts at 0 (memset) and the increment produces 1, identical to the old behaviour. On subsequent hotplug cycles the generation advances beyond any value a vCPU previously observed on this CPU, so the generation check in pre_svm_run() reliably forces new_asid() on every vCPU after every hotplug cycle.
{
"affected": [],
"aliases": [
"CVE-2026-68093"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-08-10T13:19:53Z",
"severity": null
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nKVM: SVM: Bump asid_generation on CPU online to avoid ASID collision after hotplug\n\nIf a vCPU stays scheduled out (or blocked) while the last pCPU it ran\non goes through a hotplug cycle (online-\u003eoffline-\u003eonline), and the vCPU\nthen resumes execution on the same pCPU, then it is possible for it to\nrun with an ASID that has now been assigned to a different vCPU,\nresulting in stale TLB translations being used.\n\nsvm_enable_virtualization_cpu() resets asid_generation to 1 and sets\nnext_asid to max_asid + 1 on every CPU online event, including hotplug\ncycles. Because next_asid starts beyond the pool boundary, the first\ncall to new_asid() after an online event always wraps the pool,\nincrementing asid_generation to 2 and assigning ASIDs starting from\nmin_asid.\n\nConsider two vCPUs from different VMs, vCPU-A pinned to CPU-X holding\nasid_generation=2 and ASID=N from before the hotplug event:\n\n 1. CPU-X goes offline and back online: asid_generation resets to 1,\n next_asid = max_asid + 1.\n\n 2. One or more vCPUs migrate to CPU-X and call new_asid(), wrapping\n the pool and consuming ASIDs starting from min_asid. Eventually\n vCPU-B from a different VM is assigned asid_generation=2, ASID=N\n \u2014 the same ASID that vCPU-A held before the hotplug.\n\n 3. vCPU-A enters pre_svm_run() on CPU-X: current_vmcb-\u003ecpu is\n unchanged so the migration branch is skipped. Its saved\n asid_generation=2 matches sd-\u003easid_generation=2, so the generation\n check silently passes and vCPU-A continues running with ASID=N \u2014\n the same ASID just freshly assigned to vCPU-B.\n\nBoth vCPUs from different VMs now run on CPU-X with the same ASID,\ncausing them to share NPT TLB entries and producing stale translations.\n\nThe collision manifests as a KVM internal error (Suberror: 1, emulation\nfailure). The NPT page fault reports a faulting GPA far outside the\nVM\u0027s physical memory range \u2014 a sign of stale TLB translations being\nused. KVM falls back to instruction emulation, which fails on\nFPU/XSave instructions (XRSTOR, STMXCSR) that the emulator does not\nimplement.\n\nFix this by incrementing asid_generation instead of resetting it to 1\nin svm_enable_virtualization_cpu(). On module load, asid_generation\nstarts at 0 (memset) and the increment produces 1, identical to the\nold behaviour. On subsequent hotplug cycles the generation advances\nbeyond any value a vCPU previously observed on this CPU, so the\ngeneration check in pre_svm_run() reliably forces new_asid() on every\nvCPU after every hotplug cycle.",
"id": "GHSA-99gx-mqgq-5pqg",
"modified": "2026-08-19T18:32:04Z",
"published": "2026-08-10T15:33:35Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-68093"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/0f33b1c457c2199ed130b92cc2ff363a3f7b9415"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/2028b81321dc757b6875b99c10d908e349c141e2"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/25f744ffa0c8e799e06250ce2e618367b166b0d4"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/60283726f2845bd78b95efbd0e50b93944780477"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/6b542d116acecb83a1ca34e8eace304cff6a4ec9"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/7508916b4b55d6f5ecc68cd09774dabd3a6b4440"
}
],
"schema_version": "1.4.0",
"severity": []
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.