On 03/02/21 12:34, Yang Weijiang wrote:
Control-flow Enforcement Technology (CET) provides protection against Return/Jump-Oriented Programming (ROP/JOP) attack. There're two CET subfeatures: Shadow Stack (SHSTK) and Indirect Branch Tracking (IBT). SHSTK is to prevent ROP and IBT is to prevent JOP. Several parts in KVM have been updated to provide guest CET support, including: CPUID/XSAVES settings, MSR passthrough, user-space MSR access interface, vmentry/vmexit config, nested VM etc. These patches are dependent on CET kernel patches for XSAVES support and CET definitions, e.g., MSR and related feature flags. CET kernel patches: refer to [1], [2]. Previous CET KVM patches: refer to [3]. CET QEMU patches: refer to [4]. CET KVM unit-test patch: refer to [5]. [1]: CET Shadow Stack patches v18: https://lkml.kernel.org/linux-api/20210127212524.10188-1-yu-cheng.yu@xxxxxxxxx/ [2]: Indirect Branch Tracking patches v18: https://lkml.kernel.org/linux-api/20210127213028.11362-1-yu-cheng.yu@xxxxxxxxx/ [3]: CET KVM patches v14: https://lkml.kernel.org/kvm/20201106011637.14289-1-weijiang.yang@xxxxxxxxx/ [4]: CET QEMU patches: https://patchwork.ozlabs.org/project/qemu-devel/patch/20201013051935.6052-2-weijiang.yang@xxxxxxxxx/ [5]: CET KVM unit-test patch: https://patchwork.kernel.org/project/kvm/patch/20200506082110.25441-12-weijiang.yang@xxxxxxxxx/ Changes in v15: - Changed patches per Paolo's review feedback on v14. - Added a new patch for GUEST_SSP save/restore in guest SMM case. - Fixed guest call-trace issue due to CET MSR interception. - Removed unnecessary guest CET state cleanup in VMCS. - Rebased patches to 5.11-rc6. Sean Christopherson (2): KVM: x86: Report XSS as an MSR to be saved if there are supported features KVM: x86: Load guest fpu state when accessing MSRs managed by XSAVES Yang Weijiang (12): KVM: x86: Refresh CPUID on writes to MSR_IA32_XSS KVM: x86: Add #CP support in guest exception dispatch KVM: VMX: Introduce CET VMCS fields and flags KVM: x86: Add fault checks for CR4.CET KVM: VMX: Emulate reads and writes to CET MSRs KVM: VMX: Add a synthetic MSR to allow userspace VMM to access GUEST_SSP KVM: x86: Report CET MSRs as to-be-saved if CET is supported KVM: x86: Enable CET virtualization for VMX and advertise CET to userspace KVM: VMX: Pass through CET MSRs to the guest when supported KVM: nVMX: Add helper to check the vmcs01 MSR bitmap for MSR pass-through KVM: nVMX: Enable CET support for nested VMX KVM: x86: Save/Restore GUEST_SSP to/from SMRAM arch/x86/include/asm/kvm_host.h | 4 +- arch/x86/include/asm/vmx.h | 8 ++ arch/x86/include/uapi/asm/kvm.h | 1 + arch/x86/include/uapi/asm/kvm_para.h | 1 + arch/x86/kvm/cpuid.c | 26 +++- arch/x86/kvm/emulate.c | 11 ++ arch/x86/kvm/vmx/capabilities.h | 5 + arch/x86/kvm/vmx/nested.c | 57 ++++++-- arch/x86/kvm/vmx/vmcs12.c | 6 + arch/x86/kvm/vmx/vmcs12.h | 14 +- arch/x86/kvm/vmx/vmx.c | 202 ++++++++++++++++++++++++++- arch/x86/kvm/x86.c | 67 ++++++++- arch/x86/kvm/x86.h | 10 +- 13 files changed, 387 insertions(+), 25 deletions(-)
Queued, though not for 5.12 unless the bare metal support is there too. Paolo
