[patch 0/4] mmu audit update

[patch 0/4] mmu audit update

[Marcelo Tosatti]

Some updates to the MMU audit code.

The third patch is "guessy" because I could not find the notrap spte
documentation, all I can see is the page-fault error code mask and match
fields in the VMCS, but can't see the link of that to sptes. Can someone
point it out please?

-- 

[patch 1/4] KVM: MMU audit: update count_writable_mappings / count_rmaps

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-1 --]
[-- Type: text/plain, Size: 3975 bytes --]

Under testing, count_writable_mappings returns a value that is 2 integers
larger than what count_rmaps returns.

Suspicion is that either of the two functions is counting a duplicate (either
positively or negatively). 

Modifying check_writable_mappings_rmap to check for rmap existance on
all present MMU pages fails to trigger an error, which should keep Avi
happy.

Also introduce mmu_spte_walk to invoke a callback on all present sptes visible
to the current vcpu, might be useful in the future.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3017,6 +3017,55 @@ static gva_t canonicalize(gva_t gva)
 	return gva;
 }
 
+
+typedef void (*inspect_spte_fn) (struct kvm *kvm, struct kvm_mmu_page *sp,
+				 u64 *sptep);
+
+static void __mmu_spte_walk(struct kvm *kvm, struct kvm_mmu_page *sp,
+			    inspect_spte_fn fn)
+{
+	int i;
+
+	for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
+		u64 ent = sp->spt[i];
+
+		if (is_shadow_present_pte(ent)) {
+			if (sp->role.level > 1) {
+				struct kvm_mmu_page *child;
+				child = page_header(ent & PT64_BASE_ADDR_MASK);
+				__mmu_spte_walk(kvm, child, fn);
+			}
+			if (sp->role.level == 1)
+				fn(kvm, sp, &sp->spt[i]);
+		}
+	}
+}
+
+static void mmu_spte_walk(struct kvm_vcpu *vcpu, inspect_spte_fn fn)
+{
+	int i;
+	struct kvm_mmu_page *sp;
+
+	if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
+		return;
+	if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
+		hpa_t root = vcpu->arch.mmu.root_hpa;
+		sp = page_header(root);
+		__mmu_spte_walk(vcpu->kvm, sp, fn);
+		return;
+	}
+	for (i = 0; i < 4; ++i) {
+		hpa_t root = vcpu->arch.mmu.pae_root[i];
+
+		if (root && VALID_PAGE(root)) {
+			root &= PT64_BASE_ADDR_MASK;
+			sp = page_header(root);
+			__mmu_spte_walk(vcpu->kvm, sp, fn);
+		}
+	}
+	return;
+}
+
 static void audit_mappings_page(struct kvm_vcpu *vcpu, u64 page_pte,
 				gva_t va, int level)
 {
@@ -3109,9 +3158,43 @@ static int count_rmaps(struct kvm_vcpu *
 	return nmaps;
 }
 
-static int count_writable_mappings(struct kvm_vcpu *vcpu)
+void inspect_spte_has_rmap(struct kvm *kvm, struct kvm_mmu_page *sp, u64 *sptep)
+{
+	unsigned long *rmapp;
+	struct kvm_mmu_page *rev_sp;
+	gfn_t gfn;
+
+	if (*sptep & PT_WRITABLE_MASK) {
+		rev_sp = page_header(__pa(sptep));
+		gfn = rev_sp->gfns[sptep - rev_sp->spt];
+
+		if (!gfn_to_memslot(kvm, gfn)) {
+			printk(KERN_ERR "%s: no memslot for gfn %ld\n",
+					 audit_msg, gfn);
+			printk(KERN_ERR "%s: index %ld of sp (gfn=%lx)\n",
+					audit_msg, sptep - rev_sp->spt,
+					rev_sp->gfn);
+			dump_stack();
+			return;
+		}
+
+		rmapp = gfn_to_rmap(kvm, rev_sp->gfns[sptep - rev_sp->spt], 0);
+		if (!*rmapp) {
+			printk(KERN_ERR "%s: no rmap for writable spte %llx\n",
+					 audit_msg, *sptep);
+			dump_stack();
+		}
+	}
+
+}
+
+void audit_writable_sptes_have_rmaps(struct kvm_vcpu *vcpu)
+{
+	mmu_spte_walk(vcpu, inspect_spte_has_rmap);
+}
+
+static void check_writable_mappings_rmap(struct kvm_vcpu *vcpu)
 {
-	int nmaps = 0;
 	struct kvm_mmu_page *sp;
 	int i;
 
@@ -3128,20 +3211,16 @@ static int count_writable_mappings(struc
 				continue;
 			if (!(ent & PT_WRITABLE_MASK))
 				continue;
-			++nmaps;
+			inspect_spte_has_rmap(vcpu->kvm, sp, &pt[i]);
 		}
 	}
-	return nmaps;
+	return;
 }
 
 static void audit_rmap(struct kvm_vcpu *vcpu)
 {
-	int n_rmap = count_rmaps(vcpu);
-	int n_actual = count_writable_mappings(vcpu);
-
-	if (n_rmap != n_actual)
-		printk(KERN_ERR "%s: (%s) rmap %d actual %d\n",
-		       __func__, audit_msg, n_rmap, n_actual);
+	check_writable_mappings_rmap(vcpu);
+	count_rmaps(vcpu);
 }
 
 static void audit_write_protection(struct kvm_vcpu *vcpu)
@@ -3175,6 +3254,7 @@ static void kvm_mmu_audit(struct kvm_vcp
 	audit_rmap(vcpu);
 	audit_write_protection(vcpu);
 	audit_mappings(vcpu);
+	audit_writable_sptes_have_rmaps(vcpu);
 	dbg = olddbg;
 }
 

-- 

[patch 2/4] KVM: MMU audit: update audit_write_protection

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-2 --]
[-- Type: text/plain, Size: 1225 bytes --]

- Unsync pages contain writable sptes in the rmap.
- rmaps do not exclusively contain writable sptes anymore.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3228,20 +3228,28 @@ static void audit_write_protection(struc
 	struct kvm_mmu_page *sp;
 	struct kvm_memory_slot *slot;
 	unsigned long *rmapp;
+	u64 *spte;
 	gfn_t gfn;
 
 	list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) {
 		if (sp->role.direct)
 			continue;
+		if (sp->unsync)
+			continue;
 
 		gfn = unalias_gfn(vcpu->kvm, sp->gfn);
 		slot = gfn_to_memslot_unaliased(vcpu->kvm, sp->gfn);
 		rmapp = &slot->rmap[gfn - slot->base_gfn];
-		if (*rmapp)
-			printk(KERN_ERR "%s: (%s) shadow page has writable"
-			       " mappings: gfn %lx role %x\n",
+
+		spte = rmap_next(vcpu->kvm, rmapp, NULL);
+		while (spte) {
+			if (*spte & PT_WRITABLE_MASK)
+				printk(KERN_ERR "%s: (%s) shadow page has "
+				"writable mappings: gfn %lx role %x\n",
 			       __func__, audit_msg, sp->gfn,
 			       sp->role.word);
+			spte = rmap_next(vcpu->kvm, rmapp, spte);
+		}
 	}
 }
 

-- 

[patch 3/4] KVM: MMU audit: nontrapping ptes in nonleaf level

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-3 --]
[-- Type: text/plain, Size: 849 bytes --]

update_pte sets nonleaf sptes as notrap, so either it or the audit code
is broken.

Choose the audit code.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3081,12 +3081,7 @@ static void audit_mappings_page(struct k
 
 		va = canonicalize(va);
 		if (level > 1) {
-			if (ent == shadow_notrap_nonpresent_pte)
-				printk(KERN_ERR "audit: (%s) nontrapping pte"
-				       " in nonleaf level: levels %d gva %lx"
-				       " level %d pte %llx\n", audit_msg,
-				       vcpu->arch.mmu.root_level, va, level, ent);
-			else
+			if (is_shadow_present_pte(ent))
 				audit_mappings_page(vcpu, ent, va, level - 1);
 		} else {
 			gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, va);

-- 

[patch 4/4] KVM: MMU audit: audit_mappings tweaks

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-4 --]
[-- Type: text/plain, Size: 1103 bytes --]

- Fail early in case gfn_to_pfn returns is_error_pfn.
- For the pre pte write case, avoid spurious "gva is valid but spte is notrap" 
  messages (the emulation code does the guest write first, so this particular
  case is OK).

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3089,6 +3089,11 @@ static void audit_mappings_page(struct k
 			pfn_t pfn = gfn_to_pfn(vcpu->kvm, gfn);
 			hpa_t hpa = (hpa_t)pfn << PAGE_SHIFT;
 
+			if (is_error_pfn(pfn)) {
+				kvm_release_pfn_clean(pfn);
+				continue;
+			}
+
 			if (is_shadow_present_pte(ent)
 			    && (ent & PT64_BASE_ADDR_MASK) != hpa)
 				printk(KERN_ERR "xx audit error: (%s) levels %d"
@@ -3256,7 +3261,8 @@ static void kvm_mmu_audit(struct kvm_vcp
 	audit_msg = msg;
 	audit_rmap(vcpu);
 	audit_write_protection(vcpu);
-	audit_mappings(vcpu);
+	if (strcmp("pre pte write", audit_msg))
+		audit_mappings(vcpu);
 	audit_writable_sptes_have_rmaps(vcpu);
 	dbg = olddbg;
 }

-- 

Re: [patch 0/4] mmu audit update

[Avi Kivity]

Marcelo Tosatti wrote:
> Some updates to the MMU audit code.
>
> The third patch is "guessy" because I could not find the notrap spte
> documentation, all I can see is the page-fault error code mask and match
> fields in the VMCS, but can't see the link of that to sptes. Can someone
> point it out please?
>   

When bypass_guest_pf is set, we tell vmx not to trap if the fault is due 
to page-not-present.  So if we know gpte.p == 0, we set spte.p = 0 and 
allow not-present page faults to go directly to the guest without trapping.

Of course, we still need to trap cases where gpte.p = 1 but we haven't 
mapped the page yet.  So we set a reserved bit in the spte and trap on that.

-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.

Re: [patch 1/4] KVM: MMU audit: update count_writable_mappings / count_rmaps

[Avi Kivity]

Marcelo Tosatti wrote:
> Under testing, count_writable_mappings returns a value that is 2 integers
> larger than what count_rmaps returns.
>
> Suspicion is that either of the two functions is counting a duplicate (either
> positively or negatively). 
>
> Modifying check_writable_mappings_rmap to check for rmap existance on
> all present MMU pages fails to trigger an error, which should keep Avi
> happy.
>
> Also introduce mmu_spte_walk to invoke a callback on all present sptes visible
> to the current vcpu, might be useful in the future.
>
> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
>
> Index: kvm/arch/x86/kvm/mmu.c
> ===================================================================
> --- kvm.orig/arch/x86/kvm/mmu.c
> +++ kvm/arch/x86/kvm/mmu.c
> @@ -3017,6 +3017,55 @@ static gva_t canonicalize(gva_t gva)
>  	return gva;
>  }
>  
> +
> +typedef void (*inspect_spte_fn) (struct kvm *kvm, struct kvm_mmu_page *sp,
> +				 u64 *sptep);
> +
> +static void __mmu_spte_walk(struct kvm *kvm, struct kvm_mmu_page *sp,
> +			    inspect_spte_fn fn)
> +{
> +	int i;
> +
> +	for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
> +		u64 ent = sp->spt[i];
> +
> +		if (is_shadow_present_pte(ent)) {
> +			if (sp->role.level > 1) {
>   

I think this is broken wrt large pages.  We should recurse if role.level 
 > 1 or the G bit is set.

> +	if (*sptep & PT_WRITABLE_MASK) {
> +		rev_sp = page_header(__pa(sptep));
> +		gfn = rev_sp->gfns[sptep - rev_sp->spt];
> +
> +		if (!gfn_to_memslot(kvm, gfn)) {
> +			printk(KERN_ERR "%s: no memslot for gfn %ld\n",
> +					 audit_msg, gfn);
> +			printk(KERN_ERR "%s: index %ld of sp (gfn=%lx)\n",
> +					audit_msg, sptep - rev_sp->spt,
> +					rev_sp->gfn);
> +			dump_stack();
> +			return;
> +		}
> +
> +		rmapp = gfn_to_rmap(kvm, rev_sp->gfns[sptep - rev_sp->spt], 0);
> +		if (!*rmapp) {
> +			printk(KERN_ERR "%s: no rmap for writable spte %llx\n",
> +					 audit_msg, *sptep);
> +			dump_stack();
> +		}
> +	}
>   

Semi-related: we should set up a new exit code to halt the VM so it can 
be inspected, otherwise all those printks and dump_stack()s will quickly 
overwhelm the logging facilities.


-- 
error compiling committee.c: too many arguments to function

Re: [patch 4/4] KVM: MMU audit: audit_mappings tweaks

[Avi Kivity]

Marcelo Tosatti wrote:
> @@ -3256,7 +3261,8 @@ static void kvm_mmu_audit(struct kvm_vcp
>  	audit_msg = msg;
>  	audit_rmap(vcpu);
>  	audit_write_protection(vcpu);
> -	audit_mappings(vcpu);
> +	if (strcmp("pre pte write", audit_msg))
> +		audit_mappings(vcpu);
>  	audit_writable_sptes_have_rmaps(vcpu);
>  	dbg = olddbg;
>  }
>   

strcmp() doesn't return a truth value, please add != 0.

-- 
error compiling committee.c: too many arguments to function

Re: [patch 1/4] KVM: MMU audit: update count_writable_mappings / count_rmaps

[Marcelo Tosatti]

On Mon, Jun 08, 2009 at 12:24:08PM +0300, Avi Kivity wrote:
>> +static void __mmu_spte_walk(struct kvm *kvm, struct kvm_mmu_page *sp,
>> +			    inspect_spte_fn fn)
>> +{
>> +	int i;
>> +
>> +	for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
>> +		u64 ent = sp->spt[i];
>> +
>> +		if (is_shadow_present_pte(ent)) {
>> +			if (sp->role.level > 1) {
>>   
>
> I think this is broken wrt large pages.  We should recurse if role.level  
> > 1 or the G bit is set.

Yes, fixed. Plan to add largepages validity checks later.

> Semi-related: we should set up a new exit code to halt the VM so it can  
> be inspected, otherwise all those printks and dump_stack()s will quickly  
> overwhelm the logging facilities.

Can you clarify on the halt exit code?

Because for other exit codes which similar behaviour is wanted, say,
unhandled vm exit, the policy can be handled in userspace (and the
decision to halt or not seems better suited to happen there). So perhaps
KVM_EXIT_MMU_AUDIT_FAILED?

I wondered before whether it would be good to stop auditing on the first
error, but gave up on the idea.

Re: [patch 1/4] KVM: MMU audit: update count_writable_mappings / count_rmaps

[Avi Kivity]

Marcelo Tosatti wrote:
>> Semi-related: we should set up a new exit code to halt the VM so it can  
>> be inspected, otherwise all those printks and dump_stack()s will quickly  
>> overwhelm the logging facilities.
>>     
>
> Can you clarify on the halt exit code?
>   

set a bit that tells KVM_RUN to quit (like KVM_EXIT_UNKNOWN), when 
userspace sees it it vm_stop()s, waiting for someone to dig in.

Clean logs, clean state.

> Because for other exit codes which similar behaviour is wanted, say,
> unhandled vm exit, the policy can be handled in userspace (and the
> decision to halt or not seems better suited to happen there). So perhaps
> KVM_EXIT_MMU_AUDIT_FAILED?
>   

Yes, the name was bad.  We can do KVM_EXIT_INTERNAL_ERROR and have a 
maintainer_name field in the union to choose who will handle it.

> I wondered before whether it would be good to stop auditing on the first
> error, but gave up on the idea.
>   

It's harder without exceptions.

-- 
error compiling committee.c: too many arguments to function

[patch 0/4] mmu audit update v2

[Marcelo Tosatti]

Addressing comments.


-- 

[patch 1/4] KVM: MMU audit: update count_writable_mappings / count_rmaps

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-1 --]
[-- Type: text/plain, Size: 3997 bytes --]

Under testing, count_writable_mappings returns a value that is 2 integers
larger than what count_rmaps returns.

Suspicion is that either of the two functions is counting a duplicate (either
positively or negatively). 

Modifying check_writable_mappings_rmap to check for rmap existance on
all present MMU pages fails to trigger an error, which should keep Avi
happy.

Also introduce mmu_spte_walk to invoke a callback on all present sptes visible
to the current vcpu, might be useful in the future.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3020,6 +3020,55 @@ static gva_t canonicalize(gva_t gva)
 	return gva;
 }
 
+
+typedef void (*inspect_spte_fn) (struct kvm *kvm, struct kvm_mmu_page *sp,
+				 u64 *sptep);
+
+static void __mmu_spte_walk(struct kvm *kvm, struct kvm_mmu_page *sp,
+			    inspect_spte_fn fn)
+{
+	int i;
+
+	for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
+		u64 ent = sp->spt[i];
+
+		if (is_shadow_present_pte(ent)) {
+			if (sp->role.level > 1 && !is_large_pte(ent)) {
+				struct kvm_mmu_page *child;
+				child = page_header(ent & PT64_BASE_ADDR_MASK);
+				__mmu_spte_walk(kvm, child, fn);
+			}
+			if (sp->role.level == 1)
+				fn(kvm, sp, &sp->spt[i]);
+		}
+	}
+}
+
+static void mmu_spte_walk(struct kvm_vcpu *vcpu, inspect_spte_fn fn)
+{
+	int i;
+	struct kvm_mmu_page *sp;
+
+	if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
+		return;
+	if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
+		hpa_t root = vcpu->arch.mmu.root_hpa;
+		sp = page_header(root);
+		__mmu_spte_walk(vcpu->kvm, sp, fn);
+		return;
+	}
+	for (i = 0; i < 4; ++i) {
+		hpa_t root = vcpu->arch.mmu.pae_root[i];
+
+		if (root && VALID_PAGE(root)) {
+			root &= PT64_BASE_ADDR_MASK;
+			sp = page_header(root);
+			__mmu_spte_walk(vcpu->kvm, sp, fn);
+		}
+	}
+	return;
+}
+
 static void audit_mappings_page(struct kvm_vcpu *vcpu, u64 page_pte,
 				gva_t va, int level)
 {
@@ -3112,9 +3161,43 @@ static int count_rmaps(struct kvm_vcpu *
 	return nmaps;
 }
 
-static int count_writable_mappings(struct kvm_vcpu *vcpu)
+void inspect_spte_has_rmap(struct kvm *kvm, struct kvm_mmu_page *sp, u64 *sptep)
+{
+	unsigned long *rmapp;
+	struct kvm_mmu_page *rev_sp;
+	gfn_t gfn;
+
+	if (*sptep & PT_WRITABLE_MASK) {
+		rev_sp = page_header(__pa(sptep));
+		gfn = rev_sp->gfns[sptep - rev_sp->spt];
+
+		if (!gfn_to_memslot(kvm, gfn)) {
+			printk(KERN_ERR "%s: no memslot for gfn %ld\n",
+					 audit_msg, gfn);
+			printk(KERN_ERR "%s: index %ld of sp (gfn=%lx)\n",
+					audit_msg, sptep - rev_sp->spt,
+					rev_sp->gfn);
+			dump_stack();
+			return;
+		}
+
+		rmapp = gfn_to_rmap(kvm, rev_sp->gfns[sptep - rev_sp->spt], 0);
+		if (!*rmapp) {
+			printk(KERN_ERR "%s: no rmap for writable spte %llx\n",
+					 audit_msg, *sptep);
+			dump_stack();
+		}
+	}
+
+}
+
+void audit_writable_sptes_have_rmaps(struct kvm_vcpu *vcpu)
+{
+	mmu_spte_walk(vcpu, inspect_spte_has_rmap);
+}
+
+static void check_writable_mappings_rmap(struct kvm_vcpu *vcpu)
 {
-	int nmaps = 0;
 	struct kvm_mmu_page *sp;
 	int i;
 
@@ -3131,20 +3214,16 @@ static int count_writable_mappings(struc
 				continue;
 			if (!(ent & PT_WRITABLE_MASK))
 				continue;
-			++nmaps;
+			inspect_spte_has_rmap(vcpu->kvm, sp, &pt[i]);
 		}
 	}
-	return nmaps;
+	return;
 }
 
 static void audit_rmap(struct kvm_vcpu *vcpu)
 {
-	int n_rmap = count_rmaps(vcpu);
-	int n_actual = count_writable_mappings(vcpu);
-
-	if (n_rmap != n_actual)
-		printk(KERN_ERR "%s: (%s) rmap %d actual %d\n",
-		       __func__, audit_msg, n_rmap, n_actual);
+	check_writable_mappings_rmap(vcpu);
+	count_rmaps(vcpu);
 }
 
 static void audit_write_protection(struct kvm_vcpu *vcpu)
@@ -3178,6 +3257,7 @@ static void kvm_mmu_audit(struct kvm_vcp
 	audit_rmap(vcpu);
 	audit_write_protection(vcpu);
 	audit_mappings(vcpu);
+	audit_writable_sptes_have_rmaps(vcpu);
 	dbg = olddbg;
 }
 

-- 

[patch 2/4] KVM: MMU audit: update audit_write_protection

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-2 --]
[-- Type: text/plain, Size: 1225 bytes --]

- Unsync pages contain writable sptes in the rmap.
- rmaps do not exclusively contain writable sptes anymore.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3231,20 +3231,28 @@ static void audit_write_protection(struc
 	struct kvm_mmu_page *sp;
 	struct kvm_memory_slot *slot;
 	unsigned long *rmapp;
+	u64 *spte;
 	gfn_t gfn;
 
 	list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) {
 		if (sp->role.direct)
 			continue;
+		if (sp->unsync)
+			continue;
 
 		gfn = unalias_gfn(vcpu->kvm, sp->gfn);
 		slot = gfn_to_memslot_unaliased(vcpu->kvm, sp->gfn);
 		rmapp = &slot->rmap[gfn - slot->base_gfn];
-		if (*rmapp)
-			printk(KERN_ERR "%s: (%s) shadow page has writable"
-			       " mappings: gfn %lx role %x\n",
+
+		spte = rmap_next(vcpu->kvm, rmapp, NULL);
+		while (spte) {
+			if (*spte & PT_WRITABLE_MASK)
+				printk(KERN_ERR "%s: (%s) shadow page has "
+				"writable mappings: gfn %lx role %x\n",
 			       __func__, audit_msg, sp->gfn,
 			       sp->role.word);
+			spte = rmap_next(vcpu->kvm, rmapp, spte);
+		}
 	}
 }
 

-- 

[patch 3/4] KVM: MMU audit: nontrapping ptes in nonleaf level

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-3 --]
[-- Type: text/plain, Size: 849 bytes --]

update_pte sets nonleaf sptes as notrap, so either it or the audit code
is broken.

Choose the audit code.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3084,12 +3084,7 @@ static void audit_mappings_page(struct k
 
 		va = canonicalize(va);
 		if (level > 1) {
-			if (ent == shadow_notrap_nonpresent_pte)
-				printk(KERN_ERR "audit: (%s) nontrapping pte"
-				       " in nonleaf level: levels %d gva %lx"
-				       " level %d pte %llx\n", audit_msg,
-				       vcpu->arch.mmu.root_level, va, level, ent);
-			else
+			if (is_shadow_present_pte(ent))
 				audit_mappings_page(vcpu, ent, va, level - 1);
 		} else {
 			gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, va);

-- 

[patch 4/4] KVM: MMU audit: audit_mappings tweaks

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-4 --]
[-- Type: text/plain, Size: 1108 bytes --]

- Fail early in case gfn_to_pfn returns is_error_pfn.
- For the pre pte write case, avoid spurious "gva is valid but spte is notrap" 
  messages (the emulation code does the guest write first, so this particular
  case is OK).

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3092,6 +3092,11 @@ static void audit_mappings_page(struct k
 			pfn_t pfn = gfn_to_pfn(vcpu->kvm, gfn);
 			hpa_t hpa = (hpa_t)pfn << PAGE_SHIFT;
 
+			if (is_error_pfn(pfn)) {
+				kvm_release_pfn_clean(pfn);
+				continue;
+			}
+
 			if (is_shadow_present_pte(ent)
 			    && (ent & PT64_BASE_ADDR_MASK) != hpa)
 				printk(KERN_ERR "xx audit error: (%s) levels %d"
@@ -3259,7 +3264,8 @@ static void kvm_mmu_audit(struct kvm_vcp
 	audit_msg = msg;
 	audit_rmap(vcpu);
 	audit_write_protection(vcpu);
-	audit_mappings(vcpu);
+	if (strcmp("pre pte write", audit_msg) != 0)
+		audit_mappings(vcpu);
 	audit_writable_sptes_have_rmaps(vcpu);
 	dbg = olddbg;
 }

-- 

[patch 0/6] mmu audit update v4

[Marcelo Tosatti]

Addressing comments, introducing a new helper, handling largepages.

-- 

[patch 1/6] KVM: MMU: introduce is_last_spte helper

[Marcelo Tosatti]

[-- Attachment #1: last-spte --]
[-- Type: text/plain, Size: 1816 bytes --]

Hiding some of the last largepage / level interaction (which is useful
for gbpages and for zero based levels).

Also merge the PT_PAGE_TABLE_LEVEL clearing loop in unlink_children.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -250,6 +250,15 @@ static int is_rmap_spte(u64 pte)
 	return is_shadow_present_pte(pte);
 }
 
+static int is_last_spte(u64 pte, int level)
+{
+	if (level == PT_PAGE_TABLE_LEVEL)
+		return 1;
+	if (level == PT_DIRECTORY_LEVEL && is_large_pte(pte))
+		return 1;
+	return 0;
+}
+
 static pfn_t spte_to_pfn(u64 pte)
 {
 	return (pte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT;
@@ -1293,25 +1302,17 @@ static void kvm_mmu_page_unlink_children
 
 	pt = sp->spt;
 
-	if (sp->role.level == PT_PAGE_TABLE_LEVEL) {
-		for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
-			if (is_shadow_present_pte(pt[i]))
-				rmap_remove(kvm, &pt[i]);
-			pt[i] = shadow_trap_nonpresent_pte;
-		}
-		return;
-	}
-
 	for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
 		ent = pt[i];
 
 		if (is_shadow_present_pte(ent)) {
-			if (!is_large_pte(ent)) {
+			if (!is_last_spte(ent, sp->role.level)) {
 				ent &= PT64_BASE_ADDR_MASK;
 				mmu_page_remove_parent_pte(page_header(ent),
 							   &pt[i]);
 			} else {
-				--kvm->stat.lpages;
+				if (is_large_pte(ent))
+					--kvm->stat.lpages;
 				rmap_remove(kvm, &pt[i]);
 			}
 		}
@@ -2357,8 +2358,7 @@ static void mmu_pte_write_zap_pte(struct
 
 	pte = *spte;
 	if (is_shadow_present_pte(pte)) {
-		if (sp->role.level == PT_PAGE_TABLE_LEVEL ||
-		    is_large_pte(pte))
+		if (is_last_spte(pte, sp->role.level))
 			rmap_remove(vcpu->kvm, spte);
 		else {
 			child = page_header(pte & PT64_BASE_ADDR_MASK);

-- 

[patch 2/6] KVM: MMU audit: update count_writable_mappings / count_rmaps

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-1 --]
[-- Type: text/plain, Size: 4081 bytes --]

Under testing, count_writable_mappings returns a value that is 2 integers
larger than what count_rmaps returns.

Suspicion is that either of the two functions is counting a duplicate (either
positively or negatively). 

Modifying check_writable_mappings_rmap to check for rmap existance on
all present MMU pages fails to trigger an error, which should keep Avi
happy.

Also introduce mmu_spte_walk to invoke a callback on all present sptes visible
to the current vcpu, might be useful in the future.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3020,6 +3020,55 @@ static gva_t canonicalize(gva_t gva)
 	return gva;
 }
 
+
+typedef void (*inspect_spte_fn) (struct kvm *kvm, struct kvm_mmu_page *sp,
+				 u64 *sptep);
+
+static void __mmu_spte_walk(struct kvm *kvm, struct kvm_mmu_page *sp,
+			    inspect_spte_fn fn)
+{
+	int i;
+
+	for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
+		u64 ent = sp->spt[i];
+
+		if (is_shadow_present_pte(ent)) {
+			if (sp->role.level > 1 && !is_large_pte(ent)) {
+				struct kvm_mmu_page *child;
+				child = page_header(ent & PT64_BASE_ADDR_MASK);
+				__mmu_spte_walk(kvm, child, fn);
+			}
+			if (sp->role.level == 1)
+				fn(kvm, sp, &sp->spt[i]);
+		}
+	}
+}
+
+static void mmu_spte_walk(struct kvm_vcpu *vcpu, inspect_spte_fn fn)
+{
+	int i;
+	struct kvm_mmu_page *sp;
+
+	if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
+		return;
+	if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
+		hpa_t root = vcpu->arch.mmu.root_hpa;
+		sp = page_header(root);
+		__mmu_spte_walk(vcpu->kvm, sp, fn);
+		return;
+	}
+	for (i = 0; i < 4; ++i) {
+		hpa_t root = vcpu->arch.mmu.pae_root[i];
+
+		if (root && VALID_PAGE(root)) {
+			root &= PT64_BASE_ADDR_MASK;
+			sp = page_header(root);
+			__mmu_spte_walk(vcpu->kvm, sp, fn);
+		}
+	}
+	return;
+}
+
 static void audit_mappings_page(struct kvm_vcpu *vcpu, u64 page_pte,
 				gva_t va, int level)
 {
@@ -3112,9 +3161,47 @@ static int count_rmaps(struct kvm_vcpu *
 	return nmaps;
 }
 
-static int count_writable_mappings(struct kvm_vcpu *vcpu)
+void inspect_spte_has_rmap(struct kvm *kvm, struct kvm_mmu_page *sp, u64 *sptep)
+{
+	unsigned long *rmapp;
+	struct kvm_mmu_page *rev_sp;
+	gfn_t gfn;
+
+	if (*sptep & PT_WRITABLE_MASK) {
+		rev_sp = page_header(__pa(sptep));
+		gfn = rev_sp->gfns[sptep - rev_sp->spt];
+
+		if (!gfn_to_memslot(kvm, gfn)) {
+			if (!printk_ratelimit())
+				return;
+			printk(KERN_ERR "%s: no memslot for gfn %ld\n",
+					 audit_msg, gfn);
+			printk(KERN_ERR "%s: index %ld of sp (gfn=%lx)\n",
+					audit_msg, sptep - rev_sp->spt,
+					rev_sp->gfn);
+			dump_stack();
+			return;
+		}
+
+		rmapp = gfn_to_rmap(kvm, rev_sp->gfns[sptep - rev_sp->spt], 0);
+		if (!*rmapp) {
+			if (!printk_ratelimit())
+				return;
+			printk(KERN_ERR "%s: no rmap for writable spte %llx\n",
+					 audit_msg, *sptep);
+			dump_stack();
+		}
+	}
+
+}
+
+void audit_writable_sptes_have_rmaps(struct kvm_vcpu *vcpu)
+{
+	mmu_spte_walk(vcpu, inspect_spte_has_rmap);
+}
+
+static void check_writable_mappings_rmap(struct kvm_vcpu *vcpu)
 {
-	int nmaps = 0;
 	struct kvm_mmu_page *sp;
 	int i;
 
@@ -3131,20 +3218,16 @@ static int count_writable_mappings(struc
 				continue;
 			if (!(ent & PT_WRITABLE_MASK))
 				continue;
-			++nmaps;
+			inspect_spte_has_rmap(vcpu->kvm, sp, &pt[i]);
 		}
 	}
-	return nmaps;
+	return;
 }
 
 static void audit_rmap(struct kvm_vcpu *vcpu)
 {
-	int n_rmap = count_rmaps(vcpu);
-	int n_actual = count_writable_mappings(vcpu);
-
-	if (n_rmap != n_actual)
-		printk(KERN_ERR "%s: (%s) rmap %d actual %d\n",
-		       __func__, audit_msg, n_rmap, n_actual);
+	check_writable_mappings_rmap(vcpu);
+	count_rmaps(vcpu);
 }
 
 static void audit_write_protection(struct kvm_vcpu *vcpu)
@@ -3178,6 +3261,7 @@ static void kvm_mmu_audit(struct kvm_vcp
 	audit_rmap(vcpu);
 	audit_write_protection(vcpu);
 	audit_mappings(vcpu);
+	audit_writable_sptes_have_rmaps(vcpu);
 	dbg = olddbg;
 }
 

-- 

[patch 3/6] KVM: MMU audit: update audit_write_protection

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-2 --]
[-- Type: text/plain, Size: 1225 bytes --]

- Unsync pages contain writable sptes in the rmap.
- rmaps do not exclusively contain writable sptes anymore.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3235,20 +3235,28 @@ static void audit_write_protection(struc
 	struct kvm_mmu_page *sp;
 	struct kvm_memory_slot *slot;
 	unsigned long *rmapp;
+	u64 *spte;
 	gfn_t gfn;
 
 	list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) {
 		if (sp->role.direct)
 			continue;
+		if (sp->unsync)
+			continue;
 
 		gfn = unalias_gfn(vcpu->kvm, sp->gfn);
 		slot = gfn_to_memslot_unaliased(vcpu->kvm, sp->gfn);
 		rmapp = &slot->rmap[gfn - slot->base_gfn];
-		if (*rmapp)
-			printk(KERN_ERR "%s: (%s) shadow page has writable"
-			       " mappings: gfn %lx role %x\n",
+
+		spte = rmap_next(vcpu->kvm, rmapp, NULL);
+		while (spte) {
+			if (*spte & PT_WRITABLE_MASK)
+				printk(KERN_ERR "%s: (%s) shadow page has "
+				"writable mappings: gfn %lx role %x\n",
 			       __func__, audit_msg, sp->gfn,
 			       sp->role.word);
+			spte = rmap_next(vcpu->kvm, rmapp, spte);
+		}
 	}
 }
 

-- 

[patch 4/6] KVM: MMU audit: nontrapping ptes in nonleaf level

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-3 --]
[-- Type: text/plain, Size: 787 bytes --]

It is valid to set non leaf sptes as notrap.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3084,12 +3084,7 @@ static void audit_mappings_page(struct k
 
 		va = canonicalize(va);
 		if (level > 1) {
-			if (ent == shadow_notrap_nonpresent_pte)
-				printk(KERN_ERR "audit: (%s) nontrapping pte"
-				       " in nonleaf level: levels %d gva %lx"
-				       " level %d pte %llx\n", audit_msg,
-				       vcpu->arch.mmu.root_level, va, level, ent);
-			else
+			if (is_shadow_present_pte(ent))
 				audit_mappings_page(vcpu, ent, va, level - 1);
 		} else {
 			gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, va);

-- 

[patch 5/6] KVM: MMU audit: audit_mappings tweaks

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-4 --]
[-- Type: text/plain, Size: 1108 bytes --]

- Fail early in case gfn_to_pfn returns is_error_pfn.
- For the pre pte write case, avoid spurious "gva is valid but spte is notrap" 
  messages (the emulation code does the guest write first, so this particular
  case is OK).

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3092,6 +3092,11 @@ static void audit_mappings_page(struct k
 			pfn_t pfn = gfn_to_pfn(vcpu->kvm, gfn);
 			hpa_t hpa = (hpa_t)pfn << PAGE_SHIFT;
 
+			if (is_error_pfn(pfn)) {
+				kvm_release_pfn_clean(pfn);
+				continue;
+			}
+
 			if (is_shadow_present_pte(ent)
 			    && (ent & PT64_BASE_ADDR_MASK) != hpa)
 				printk(KERN_ERR "xx audit error: (%s) levels %d"
@@ -3263,7 +3268,8 @@ static void kvm_mmu_audit(struct kvm_vcp
 	audit_msg = msg;
 	audit_rmap(vcpu);
 	audit_write_protection(vcpu);
-	audit_mappings(vcpu);
+	if (strcmp("pre pte write", audit_msg) != 0)
+		audit_mappings(vcpu);
 	audit_writable_sptes_have_rmaps(vcpu);
 	dbg = olddbg;
 }

-- 

[patch 6/6] KVM: MMU audit: largepage handling

[Marcelo Tosatti]

[-- Attachment #1: mmu-audit-last-pte --]
[-- Type: text/plain, Size: 1483 bytes --]

Make the audit code aware of largepages.

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

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -3033,12 +3033,11 @@ static void __mmu_spte_walk(struct kvm *
 		u64 ent = sp->spt[i];
 
 		if (is_shadow_present_pte(ent)) {
-			if (sp->role.level > 1 && !is_large_pte(ent)) {
+			if (!is_last_spte(ent, sp->role.level)) {
 				struct kvm_mmu_page *child;
 				child = page_header(ent & PT64_BASE_ADDR_MASK);
 				__mmu_spte_walk(kvm, child, fn);
-			}
-			if (sp->role.level == 1)
+			} else
 				fn(kvm, sp, &sp->spt[i]);
 		}
 	}
@@ -3083,10 +3082,9 @@ static void audit_mappings_page(struct k
 			continue;
 
 		va = canonicalize(va);
-		if (level > 1) {
-			if (is_shadow_present_pte(ent))
-				audit_mappings_page(vcpu, ent, va, level - 1);
-		} else {
+		if (is_shadow_present_pte(ent) && !is_last_spte(ent, level))
+			audit_mappings_page(vcpu, ent, va, level - 1);
+		else {
 			gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, va);
 			gfn_t gfn = gpa >> PAGE_SHIFT;
 			pfn_t pfn = gfn_to_pfn(vcpu->kvm, gfn);
@@ -3183,7 +3181,8 @@ void inspect_spte_has_rmap(struct kvm *k
 			return;
 		}
 
-		rmapp = gfn_to_rmap(kvm, rev_sp->gfns[sptep - rev_sp->spt], 0);
+		rmapp = gfn_to_rmap(kvm, rev_sp->gfns[sptep - rev_sp->spt],
+				    is_large_pte(*sptep));
 		if (!*rmapp) {
 			if (!printk_ratelimit())
 				return;

-- 

Re: [patch 0/6] mmu audit update v4

[Avi Kivity]

Marcelo Tosatti wrote:
> Addressing comments, introducing a new helper, handling largepages.
>
>   

Applied, thanks.

-- 
error compiling committee.c: too many arguments to function