[RFC 5/9] Modify the page fault path to meet the direct build EPT requirement

[Yulei Zhang <yulei.kernel@xxxxxxxxx>]

From: Yulei Zhang <yuleixzhang@xxxxxxxxxxx>

Refine the fast page fault code so that it can be used in either
normal ept mode or direct build EPT mode.

Signed-off-by: Yulei Zhang <yuleixzhang@xxxxxxxxxxx>
---
 arch/x86/kvm/mmu/mmu.c | 30 +++++++++++++++++++++---------
 1 file changed, 21 insertions(+), 9 deletions(-)

diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index df703deac928..b59a4502d1f6 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -3471,12 +3471,13 @@ static bool page_fault_can_be_fast(u32 error_code)
  * someone else modified the SPTE from its original value.
  */
 static bool
-fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
+fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, gpa_t gpa,
 			u64 *sptep, u64 old_spte, u64 new_spte)
 {
 	gfn_t gfn;
 
-	WARN_ON(!sp->role.direct);
+	WARN_ON(!vcpu->arch.direct_build_tdp &&
+		(!page_header(__pa(sptep))->role.direct));
 
 	/*
 	 * Theoretically we could also set dirty bit (and flush TLB) here in
@@ -3498,7 +3499,7 @@ fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
 		 * The gfn of direct spte is stable since it is
 		 * calculated by sp->gfn.
 		 */
-		gfn = kvm_mmu_page_get_gfn(sp, sptep - sp->spt);
+		gfn = gpa >> PAGE_SHIFT;
 		kvm_vcpu_mark_page_dirty(vcpu, gfn);
 	}
 
@@ -3526,10 +3527,10 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
 			    u32 error_code)
 {
 	struct kvm_shadow_walk_iterator iterator;
-	struct kvm_mmu_page *sp;
 	bool fault_handled = false;
 	u64 spte = 0ull;
 	uint retry_count = 0;
+	int pte_level = 0;
 
 	if (!page_fault_can_be_fast(error_code))
 		return false;
@@ -3539,12 +3540,20 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
 	do {
 		u64 new_spte;
 
-		for_each_shadow_entry_lockless(vcpu, cr2_or_gpa, iterator, spte)
+		for_each_shadow_entry_lockless(vcpu, cr2_or_gpa, iterator, spte) {
 			if (!is_shadow_present_pte(spte))
 				break;
+		}
+
+		if (iterator.level < PT_PAGE_TABLE_LEVEL)
+			pte_level  = PT_PAGE_TABLE_LEVEL;
+		else
+			pte_level = iterator.level;
+
+		WARN_ON(!vcpu->arch.direct_build_tdp &&
+			(pte_level != page_header(__pa(iterator.sptep))->role.level));
 
-		sp = page_header(__pa(iterator.sptep));
-		if (!is_last_spte(spte, sp->role.level))
+		if (!is_last_spte(spte, pte_level))
 			break;
 
 		/*
@@ -3587,7 +3596,7 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
 			 *
 			 * See the comments in kvm_arch_commit_memory_region().
 			 */
-			if (sp->role.level > PT_PAGE_TABLE_LEVEL)
+			if (pte_level > PT_PAGE_TABLE_LEVEL)
 				break;
 		}
 
@@ -3601,7 +3610,7 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
 		 * since the gfn is not stable for indirect shadow page. See
 		 * Documentation/virt/kvm/locking.txt to get more detail.
 		 */
-		fault_handled = fast_pf_fix_direct_spte(vcpu, sp,
+		fault_handled = fast_pf_fix_direct_spte(vcpu, cr2_or_gpa,
 							iterator.sptep, spte,
 							new_spte);
 		if (fault_handled)
@@ -4153,6 +4162,9 @@ static int direct_page_fault(struct kvm_vcpu *vcpu, gpa_t gpa, u32 error_code,
 	if (fast_page_fault(vcpu, gpa, error_code))
 		return RET_PF_RETRY;
 
+	if (vcpu->arch.direct_build_tdp)
+		return RET_PF_EMULATE;
+
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-- 
2.17.1