On 31/01/2018 16:05, Jim Mattson wrote: > On Wed, Jan 31, 2018 at 1:00 PM, Paolo Bonzini <pbonzini@xxxxxxxxxx> wrote: > >> Yes, but how would moving the field into struct loaded_vmcs do anything? >> Only vmon/vmoff would change anything in vmx->nested.vmcs02. > > My suggestion was that nested_vmx_merge_msr_bitmap should set the > vmcs02 version of save_spec_ctrl_on_exit based on the calculated value > of the write permission bit for IA32_SPEC_CTRL in the vmcs02 MSR > permission bitmap. > >> Even then, L1 vmexits will also be penalized because L1 has probably >> done an RDMSR/WRMSR on L2->L1 vmexit. So I don't think it's an issue? > > Yes, it sucks to be L1 in this situation. Can we just say it sucks to be L2 too? :) Because in the end as long as no one ever writes to spec_ctrl, everybody is happy. Paolo
