On Wed, May 27, 2020 at 06:56:02PM +0200, Paolo Bonzini wrote: > On 27/05/20 18:23, Sean Christopherson wrote: > > Hmm, one option would be to make .get_tdp_level() pure function by passing > > in vcpu->arch.maxphyaddr. That should make the comment redundant. I don't > > love bleeding VMX's implementation into the prototype, but that ship has > > kinda already sailed. > > Well, it's not bleeding the implementation that much, guest MAXPHYADDR > is pretty much the only reason why it's a function and not a constant. > > Another possibility BTW is to make the callback get_max_tdp_level and > make get_tdp_level a function in mmu.c.
I like that idea. We could even avoid get_max_tdp_level() by passing that info into kvm_configure_mmu(), though that may not actually be cleaner. I'll play around with it.
