Re: [PATCH v2 09/17] KVM: x86/mmu: Use '0' as the one and only value for an invalid PAE root

2021-03-05 Thread Paolo Bonzini
On 05/03/21 19:22, Sean Christopherson wrote: On Fri, Mar 05, 2021, Paolo Bonzini wrote: On 05/03/21 02:10, Sean Christopherson wrote: Use '0' to denote an invalid pae_root instead of '0' or INVALID_PAGE. Unlike root_hpa, the pae_roots hold permission bits and thus are guaranteed to be non-zero

Re: [PATCH v2 09/17] KVM: x86/mmu: Use '0' as the one and only value for an invalid PAE root

2021-03-05 Thread Sean Christopherson
On Fri, Mar 05, 2021, Paolo Bonzini wrote: > On 05/03/21 02:10, Sean Christopherson wrote: > > Use '0' to denote an invalid pae_root instead of '0' or INVALID_PAGE. > > Unlike root_hpa, the pae_roots hold permission bits and thus are > > guaranteed to be non-zero. Having to deal with both values l

Re: [PATCH v2 09/17] KVM: x86/mmu: Use '0' as the one and only value for an invalid PAE root

2021-03-05 Thread Paolo Bonzini
On 05/03/21 02:10, Sean Christopherson wrote: Use '0' to denote an invalid pae_root instead of '0' or INVALID_PAGE. Unlike root_hpa, the pae_roots hold permission bits and thus are guaranteed to be non-zero. Having to deal with both values leads to bugs, e.g. failing to set back to INVALID_PAGE,

[PATCH v2 09/17] KVM: x86/mmu: Use '0' as the one and only value for an invalid PAE root

2021-03-04 Thread Sean Christopherson
Use '0' to denote an invalid pae_root instead of '0' or INVALID_PAGE. Unlike root_hpa, the pae_roots hold permission bits and thus are guaranteed to be non-zero. Having to deal with both values leads to bugs, e.g. failing to set back to INVALID_PAGE, warning on the wrong value, etc... Signed-off-