On Wed, Feb 12, 2020 at 05:42:33PM +0100, Paolo Bonzini wrote: > On 12/02/20 17:28, Sean Christopherson wrote: > > On Wed, Feb 12, 2020 at 01:00:59PM +0100, Paolo Bonzini wrote: > >> On 07/02/20 18:37, Sean Christopherson wrote: > >>> Rename kvm_mmu->get_cr3() to call out that it is retrieving a guest > >>> value, as opposed to kvm_mmu->set_cr3(), which sets a host value, and to > >>> note that it will return L1's EPTP when nested EPT is in use. Hopefully > >>> the new name will also make it more obvious that L1's nested_cr3 is > >>> returned in SVM's nested NPT case. > >>> > >>> No functional change intended. > >> > >> Should we call it "get_pgd", since that is how Linux calls the top-level > >> directory? I always get confused by PUD/PMD, but as long as we only > >> keep one /p.d/ moniker it should be fine. > > > > Heh, I have the exact same sentiment. get_pgd() works for me. > > Ok, I'll post a patch that uses get_guest_pgd() as soon as I open > kvm/next for 5.7 material. I need to resend the 5-level nested EPT support, I'll include this change. Should I also include patches 4, 5 and 7 when I send v3 of that series? Your earlier mail said they were queued for 5.6, but AFAICT only patches 1 and 2 made it into 5.6 (which is not a big deal at all). On Wed, Feb 12, 2020 at 01:03:03PM +0100, Paolo Bonzini wrote: > On 07/02/20 18:37, Sean Christopherson wrote: > > Sean Christopherson (7): > > KVM: nVMX: Use correct root level for nested EPT shadow page tables > > KVM: x86/mmu: Fix struct guest_walker arrays for 5-level paging > > KVM: nVMX: Allow L1 to use 5-level page walks for nested EPT > > KVM: nVMX: Rename nested_ept_get_cr3() to nested_ept_get_eptp() > > KVM: nVMX: Rename EPTP validity helper and associated variables > > KVM: x86/mmu: Rename kvm_mmu->get_cr3() to ->get_guest_cr3_or_eptp() > > KVM: nVMX: Drop unnecessary check on ept caps for execute-only > > > > Queued 1-2-4-5-7 (for 5.6), thanks!
