Add dedicated helpers to emulate RESET instead of having the relevant code scattered through vcpu_create() and vcpu_reset(). Paolo, I think this is what you meant by "have init_vmcb/svm_vcpu_reset look more like the VMX code"[*]. Patch 01 is a bit odd; it's essentially an explicit acknowledgement that KVM's emulation is far from complete. It caught my eye when auditing the "create" flows to ensure they didn't touch guest state, which should be handled by "reset". I waffled between deleting it outright and moving it to the new __vmx_vcpu_reset(), and opted to delete outright to discourage ad hoc clearing of MSRs during RESET, which isn't a maintainable approach. [*] https://lore.kernel.org/all/c3563870-62c3-897d-3148-e48bb755310c@xxxxxxxxxx/ Sean Christopherson (3): KVM: VMX: Drop explicit zeroing of MSR guest values at vCPU creation KVM: VMX: Move RESET emulation to vmx_vcpu_reset() KVM: SVM: Move RESET emulation to svm_vcpu_reset() arch/x86/kvm/svm/sev.c | 6 ++-- arch/x86/kvm/svm/svm.c | 29 ++++++++++-------- arch/x86/kvm/svm/svm.h | 2 +- arch/x86/kvm/vmx/vmx.c | 67 ++++++++++++++++++++---------------------- 4 files changed, 53 insertions(+), 51 deletions(-) -- 2.33.0.309.g3052b89438-goog
