[patch 2/3] KVM: dont require read-only host ptes

[Marcelo Tosatti <mtosatti@redhat.com>]

[-- Attachment #1: gfn-to-pfn-ronly --]
[-- Type: text/plain, Size: 7088 bytes --]

gfn_to_pfn requires a writable host pte, failing otherwise.

Change it to fallback to read-only "acquision', informing the callers. 

Hopefully the ptes are cache-hot so the overhead is minimal.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

Index: kvm/arch/ia64/kvm/kvm-ia64.c
===================================================================
--- kvm.orig/arch/ia64/kvm/kvm-ia64.c
+++ kvm/arch/ia64/kvm/kvm-ia64.c
@@ -1589,7 +1589,7 @@ int kvm_arch_prepare_memory_region(struc
 		return -ENOMEM;
 
 	for (i = 0; i < npages; i++) {
-		pfn = gfn_to_pfn(kvm, base_gfn + i);
+		pfn = gfn_to_pfn(kvm, base_gfn + i, NULL);
 		if (!kvm_is_mmio_pfn(pfn)) {
 			kvm_set_pmt_entry(kvm, base_gfn + i,
 					pfn << PAGE_SHIFT,
Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -2273,6 +2273,7 @@ static int nonpaging_map(struct kvm_vcpu
 {
 	int r;
 	int level;
+	int writable;
 	pfn_t pfn;
 	unsigned long mmu_seq;
 
@@ -2289,10 +2290,10 @@ static int nonpaging_map(struct kvm_vcpu
 
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
-	pfn = gfn_to_pfn(vcpu->kvm, gfn);
+	pfn = gfn_to_pfn(vcpu->kvm, gfn, &writable);
 
 	/* mmio */
-	if (is_error_pfn(pfn))
+	if (is_error_pfn(pfn) || !writable)
 		return kvm_handle_bad_page(vcpu->kvm, gfn, pfn);
 
 	spin_lock(&vcpu->kvm->mmu_lock);
@@ -2581,6 +2582,8 @@ static int tdp_page_fault(struct kvm_vcp
 	pfn_t pfn;
 	int r;
 	int level;
+	int writable;
+	int write = error_code & PFERR_WRITE_MASK;
 	gfn_t gfn = gpa >> PAGE_SHIFT;
 	unsigned long mmu_seq;
 
@@ -2597,15 +2600,14 @@ static int tdp_page_fault(struct kvm_vcp
 
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
-	pfn = gfn_to_pfn(vcpu->kvm, gfn);
-	if (is_error_pfn(pfn))
+	pfn = gfn_to_pfn(vcpu->kvm, gfn, &writable);
+	if (is_error_pfn(pfn) || !writable)
 		return kvm_handle_bad_page(vcpu->kvm, gfn, pfn);
 	spin_lock(&vcpu->kvm->mmu_lock);
 	if (mmu_notifier_retry(vcpu, mmu_seq))
 		goto out_unlock;
 	kvm_mmu_free_some_pages(vcpu);
-	r = __direct_map(vcpu, gpa, error_code & PFERR_WRITE_MASK,
-			 level, gfn, pfn);
+	r = __direct_map(vcpu, gpa, write, level, gfn, pfn);
 	spin_unlock(&vcpu->kvm->mmu_lock);
 
 	return r;
@@ -3043,6 +3045,7 @@ static void mmu_guess_page_from_pte_writ
 {
 	gfn_t gfn;
 	pfn_t pfn;
+	int writable;
 
 	if (!is_present_gpte(gpte))
 		return;
@@ -3050,9 +3053,9 @@ static void mmu_guess_page_from_pte_writ
 
 	vcpu->arch.update_pte.mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
-	pfn = gfn_to_pfn(vcpu->kvm, gfn);
+	pfn = gfn_to_pfn(vcpu->kvm, gfn, &writable);
 
-	if (is_error_pfn(pfn)) {
+	if (is_error_pfn(pfn) || !writable) {
 		kvm_release_pfn_clean(pfn);
 		return;
 	}
Index: kvm/arch/x86/kvm/paging_tmpl.h
===================================================================
--- kvm.orig/arch/x86/kvm/paging_tmpl.h
+++ kvm/arch/x86/kvm/paging_tmpl.h
@@ -536,6 +536,7 @@ static int FNAME(page_fault)(struct kvm_
 	int write_fault = error_code & PFERR_WRITE_MASK;
 	int user_fault = error_code & PFERR_USER_MASK;
 	int fetch_fault = error_code & PFERR_FETCH_MASK;
+	int writable;
 	struct guest_walker walker;
 	u64 *sptep;
 	int write_pt = 0;
@@ -573,10 +574,10 @@ static int FNAME(page_fault)(struct kvm_
 
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
-	pfn = gfn_to_pfn(vcpu->kvm, walker.gfn);
+	pfn = gfn_to_pfn(vcpu->kvm, walker.gfn, &writable);
 
 	/* mmio */
-	if (is_error_pfn(pfn))
+	if (is_error_pfn(pfn) || !writable)
 		return kvm_handle_bad_page(vcpu->kvm, walker.gfn, pfn);
 
 	spin_lock(&vcpu->kvm->mmu_lock);
Index: kvm/include/linux/kvm_host.h
===================================================================
--- kvm.orig/include/linux/kvm_host.h
+++ kvm/include/linux/kvm_host.h
@@ -302,7 +302,7 @@ void kvm_set_page_accessed(struct page *
 
 pfn_t hva_to_pfn_atomic(struct kvm *kvm, unsigned long addr);
 pfn_t gfn_to_pfn_atomic(struct kvm *kvm, gfn_t gfn);
-pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn);
+pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn, int *writable);
 pfn_t gfn_to_pfn_memslot(struct kvm *kvm,
 			 struct kvm_memory_slot *slot, gfn_t gfn);
 int memslot_id(struct kvm *kvm, gfn_t gfn);
Index: kvm/virt/kvm/kvm_main.c
===================================================================
--- kvm.orig/virt/kvm/kvm_main.c
+++ kvm/virt/kvm/kvm_main.c
@@ -948,17 +948,28 @@ unsigned long gfn_to_hva(struct kvm *kvm
 }
 EXPORT_SYMBOL_GPL(gfn_to_hva);
 
-static pfn_t hva_to_pfn(struct kvm *kvm, unsigned long addr, bool atomic)
+static pfn_t hva_to_pfn(struct kvm *kvm, unsigned long addr, bool atomic,
+			int *writable)
 {
 	struct page *page[1];
 	int npages;
 	pfn_t pfn;
 
+	if (writable)
+		*writable = 1;
+
 	if (atomic)
 		npages = __get_user_pages_fast(addr, 1, 1, page);
 	else {
 		might_sleep();
 		npages = get_user_pages_fast(addr, 1, 1, page);
+
+		/* attempt to map read-only */
+		if (unlikely(npages != 1) && writable) {
+			npages = get_user_pages_fast(addr, 1, 0, page);
+			if (npages == 1)
+				*writable = 0;
+		}
 	}
 
 	if (unlikely(npages != 1)) {
@@ -995,11 +1006,11 @@ return_fault_page:
 
 pfn_t hva_to_pfn_atomic(struct kvm *kvm, unsigned long addr)
 {
-	return hva_to_pfn(kvm, addr, true);
+	return hva_to_pfn(kvm, addr, true, NULL);
 }
 EXPORT_SYMBOL_GPL(hva_to_pfn_atomic);
 
-static pfn_t __gfn_to_pfn(struct kvm *kvm, gfn_t gfn, bool atomic)
+static pfn_t __gfn_to_pfn(struct kvm *kvm, gfn_t gfn, bool atomic, int *writable)
 {
 	unsigned long addr;
 
@@ -1009,18 +1020,18 @@ static pfn_t __gfn_to_pfn(struct kvm *kv
 		return page_to_pfn(bad_page);
 	}
 
-	return hva_to_pfn(kvm, addr, atomic);
+	return hva_to_pfn(kvm, addr, atomic, writable);
 }
 
 pfn_t gfn_to_pfn_atomic(struct kvm *kvm, gfn_t gfn)
 {
-	return __gfn_to_pfn(kvm, gfn, true);
+	return __gfn_to_pfn(kvm, gfn, true, NULL);
 }
 EXPORT_SYMBOL_GPL(gfn_to_pfn_atomic);
 
-pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn)
+pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn, int *writable)
 {
-	return __gfn_to_pfn(kvm, gfn, false);
+	return __gfn_to_pfn(kvm, gfn, false, writable);
 }
 EXPORT_SYMBOL_GPL(gfn_to_pfn);
 
@@ -1028,7 +1039,7 @@ pfn_t gfn_to_pfn_memslot(struct kvm *kvm
 			 struct kvm_memory_slot *slot, gfn_t gfn)
 {
 	unsigned long addr = gfn_to_hva_memslot(slot, gfn);
-	return hva_to_pfn(kvm, addr, false);
+	return hva_to_pfn(kvm, addr, false, NULL);
 }
 
 int gfn_to_page_many_atomic(struct kvm *kvm, gfn_t gfn, struct page **pages,
@@ -1052,7 +1063,7 @@ struct page *gfn_to_page(struct kvm *kvm
 {
 	pfn_t pfn;
 
-	pfn = gfn_to_pfn(kvm, gfn);
+	pfn = gfn_to_pfn(kvm, gfn, NULL);
 	if (!kvm_is_mmio_pfn(pfn))
 		return pfn_to_page(pfn);
 
Index: kvm/arch/powerpc/kvm/book3s.c
===================================================================
--- kvm.orig/arch/powerpc/kvm/book3s.c
+++ kvm/arch/powerpc/kvm/book3s.c
@@ -468,7 +468,7 @@ pfn_t kvmppc_gfn_to_pfn(struct kvm_vcpu 
 		return pfn;
 	}
 
-	return gfn_to_pfn(vcpu->kvm, gfn);
+	return gfn_to_pfn(vcpu->kvm, gfn, NULL);
 }
 
 /* Book3s_32 CPUs always have 32 bytes cache line size, which Linux assumes. To