KVM: SEV: Lock all vCPUs when synchronzing VMSAs for SNP launch finish
Linux
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
| No fix yet |
| Linux Linux |
|
Linux
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
| No fix yet |
| Linux Linux |
|
|
In the Linux kernel, the following vulnerability has been resolved:
KVM: SEV: Lock all vCPUs when synchronzing VMSAs for SNP launch finish
Lock all vCPUs when synchronizing and encrypting VMSAs for SNP guests, as allowing userspace to manipulate and/or run a vCPU while its state is being synchronized would at best corrupt vCPU state, and at worst crash the host kernel.
Opportunistically assert that vcpu->mutex is held when synchronizing its VMSA (the SEV-ES path already locks vCPUs).
From the CVE record · Linux
The Linux kernel's KVM SEV code did not lock all vCPUs while synchronizing VMSAs during SNP guest launch completion. Userspace able to manipulate or run a vCPU during synchronization could corrupt vCPU state or crash the host kernel.
Operators running Linux kernels with KVM SEV for SNP guests should check whether their kernels include the fix.