On 17/08/21 18:13, David Matlack wrote:
On Tue, Aug 17, 2021 at 6:00 AM Paolo Bonzini <pbonzini@xxxxxxxxxx> wrote:
On 13/08/21 22:35, David Matlack wrote:
- if (is_writable_pte(new_spte) && !is_writable_pte(old_spte)) {
- /*
- * The gfn of direct spte is stable since it is
- * calculated by sp->gfn.
- */
- gfn = kvm_mmu_page_get_gfn(sp, sptep - sp->spt);
- kvm_vcpu_mark_page_dirty(vcpu, gfn);
- }
+ if (is_writable_pte(new_spte) && !is_writable_pte(old_spte))
+ mark_page_dirty_in_slot(vcpu->kvm, fault->slot, fault->gfn);
Oops, this actually needs kvm_vcpu_mark_page_dirty to receive the slot.
What do you mean? kvm_vcpu_mark_page_dirty ultimately just calls
mark_page_dirty_in_slot.
Yeah, I was thinking of some very old version of the dirty page ring
buffer patches. What I wrote makes no sense. :)
Paolo
