diff options
| -rw-r--r-- | kvm_paging_tmpl.h | 30 |
1 files changed, 28 insertions, 2 deletions
diff --git a/kvm_paging_tmpl.h b/kvm_paging_tmpl.h index 93bd9f9..308a007 100644 --- a/kvm_paging_tmpl.h +++ b/kvm_paging_tmpl.h @@ -316,8 +316,34 @@ static uint64_t *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, break; } - if (is_shadow_present_pte(*sptep) && !is_large_pte(*sptep)) - continue; + if (is_shadow_present_pte(*sptep) && !is_large_pte(*sptep)) { + struct kvm_mmu_page *child; + unsigned direct_access; + + if (level != gw->level) + continue; + + /* + * For the direct sp, if the guest pte's dirty bit + * changed form clean to dirty, it will corrupt the + * sp's access: allow writable in the read-only sp, + * so we should update the spte at this point to get + * a new sp with the correct access. + */ + direct_access = gw->pt_access & gw->pte_access; + if (!is_dirty_gpte(gw->ptes[gw->level - 1])) + direct_access &= ~ACC_WRITE_MASK; + + child = page_header(vcpu->kvm, + *sptep & PT64_BASE_ADDR_MASK); + + if (child->role.access == direct_access) + continue; + + mmu_page_remove_parent_pte(child, sptep); + __set_spte(sptep, shadow_trap_nonpresent_pte); + kvm_flush_remote_tlbs(vcpu->kvm); + } if (is_large_pte(*sptep)) { rmap_remove(vcpu->kvm, sptep); |