Skip to contentSkip to conversation

kvm: x86/mmu: Fix is_tdp_mmu_check when using PAE

PATCH In review Ben Gardon 4 participants 5 messages Nov 11, 2020 – Nov 16, 2020

Subsystems (3)
lore
Share & tools
lore

Ben Gardon

author
#0 lore raw

PATCH 1/1 kvm: x86/mmu: Fix is_tdp_mmu_check when using PAE

When PAE is in use, the root_hpa will not have a shadow page assoicated with it. In this case the kernel will crash with a NULL pointer dereference. Add checks to ensure is_tdp_mmu_root works as intended even when using PAE.

Tested: compiles

Fixes 02c00b3a2f7e ("kvm: x86/mmu: Allocate and free TDP MMU roots")
---
 arch/x86/kvm/mmu/tdp_mmu.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
arch/x86/kvm/mmu/tdp_mmu.c +10 -0
index 27e381c9da6c..13013f4d98ad 100644
@@ -49,8 +49,18 @@ bool is_tdp_mmu_root(struct kvm *kvm, hpa_t hpa)
 {
 	struct kvm_mmu_page *sp;
 
+	if (WARN_ON(!VALID_PAGE(hpa)))
+		return false;
+
 	sp = to_shadow_page(hpa);
 
+	/*
+	 * If this VM is being run with PAE, the TDP MMU will not be enabled
+	 * and the root HPA will not have a shadow page associated with it.
+	 */
+	if (!sp)
+		return false;
+
 	return sp->tdp_mmu_page && sp->root_count;
 }
 
Signature · 1 line
2.29.2.222.g5d2a92d10f8-goog
#1 lore raw
Quoted 4 lines · Ben Gardon

When PAE is in use, the root_hpa will not have a shadow page assoicated with it. In this case the kernel will crash with a NULL pointer dereference. Add checks to ensure is_tdp_mmu_root works as intended even when using PAE.

This seems to work in my amd64 case as well.
(https://marc.info/?l=linux-kernel&m=160494962201032&w=2)

Quoted 35 lines

Tested: compiles

Fixes 02c00b3a2f7e ("kvm: x86/mmu: Allocate and free TDP MMU roots")
---
 arch/x86/kvm/mmu/tdp_mmu.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
arch/x86/kvm/mmu/tdp_mmu.c +10 -0
index 27e381c9da6c..13013f4d98ad 100644
@@ -49,8 +49,18 @@ bool is_tdp_mmu_root(struct kvm *kvm, hpa_t hpa)
 {
 	struct kvm_mmu_page *sp;
 
+	if (WARN_ON(!VALID_PAGE(hpa)))
+		return false;
+
 	sp = to_shadow_page(hpa);
 
+	/*
+	 * If this VM is being run with PAE, the TDP MMU will not be enabled
+	 * and the root HPA will not have a shadow page associated with it.
+	 */
+	if (!sp)
+		return false;
+
 	return sp->tdp_mmu_page && sp->root_count;
 }
 
Signature · 1 line
2.29.2.222.g5d2a92d10f8-goog
Signature · 1 line
Jamie Heilman                     http://audible.transient.net/~jamie/
Tested-by on v1 1/1
#2 lore raw
In reply to Ben Gardon
Quoted 33 lines · On Wed, 11 Nov 2020 10:53:37 -0800 Ben Gardon <[email protected]> wrote:

When PAE is in use, the root_hpa will not have a shadow page assoicated with it. In this case the kernel will crash with a NULL pointer dereference. Add checks to ensure is_tdp_mmu_root works as intended even when using PAE.

Tested: compiles

Fixes 02c00b3a2f7e ("kvm: x86/mmu: Allocate and free TDP MMU roots")
---
 arch/x86/kvm/mmu/tdp_mmu.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
arch/x86/kvm/mmu/tdp_mmu.c +5 -0
index 27e381c9da6c..13013f4d98ad 100644
@@ -49,8 +49,18 @@ bool is_tdp_mmu_root(struct kvm *kvm, hpa_t hpa)
 {
 	struct kvm_mmu_page *sp;
 
+	if (WARN_ON(!VALID_PAGE(hpa)))
+		return false;
+
 	sp = to_shadow_page(hpa);
 
+	/*
+	 * If this VM is being run with PAE, the TDP MMU will not be

enabled

arch/x86/kvm/mmu/tdp_mmu.c +1 -0
+	 * and the root HPA will not have a shadow page associated

with it.

arch/x86/kvm/mmu/tdp_mmu.c bool is_tdp_mmu_root(struct kvm *kvm, hpa_t hpa) · line 65

Quoted 6 lines
arch/x86/kvm/mmu/tdp_mmu.c +4 -0
+	 */
+	if (!sp)
+		return false;
+
 	return sp->tdp_mmu_page && sp->root_count;
 }

Fixes is_tdp_mmu_root NULL pointer dereference
Tested on: Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz

maintainer
#3 lore raw
In reply to Ben Gardon
Quoted 28 lines · Ben Gardon

When PAE is in use, the root_hpa will not have a shadow page assoicated with it. In this case the kernel will crash with a NULL pointer dereference. Add checks to ensure is_tdp_mmu_root works as intended even when using PAE.

Tested: compiles

Fixes 02c00b3a2f7e ("kvm: x86/mmu: Allocate and free TDP MMU roots")
---
  arch/x86/kvm/mmu/tdp_mmu.c | 10 ++++++++++
  1 file changed, 10 insertions(+)
arch/x86/kvm/mmu/tdp_mmu.c +3 -0
index 27e381c9da6c..13013f4d98ad 100644
@@ -49,8 +49,18 @@ bool is_tdp_mmu_root(struct kvm *kvm, hpa_t hpa)
  {
  	struct kvm_mmu_page *sp;
 
+	if (WARN_ON(!VALID_PAGE(hpa)))
+		return false;
+
  	sp = to_shadow_page(hpa);
 

arch/x86/kvm/mmu/tdp_mmu.c bool is_tdp_mmu_root(struct kvm *kvm, hpa_t hpa) · line 66

Quoted 10 lines
arch/x86/kvm/mmu/tdp_mmu.c +7 -0
+	/*
+	 * If this VM is being run with PAE, the TDP MMU will not be enabled
+	 * and the root HPA will not have a shadow page associated with it.
+	 */
+	if (!sp)
+		return false;
+
  	return sp->tdp_mmu_page && sp->root_count;
  }
 

If this was just PAE, it would be easier to test "if (shadow_root_level >= PT64_ROOT_4LEVEL)"---and more correct too, because using the page_private of __pa(vcpu->arch.mmu->pae_root) is a bit untidy; we should only use page_private for pages that we know have a shadow page.

In Jamie's case however, it is x86_64 (so kvm_mmu_get_tdp_level(vcpu) == 4 and therefore the "if (shadow_root_level >= PT64_ROOT_4LEVEL)" would be true) but without EPT. In that case we go through

	vcpu->arch.mmu->root_hpa = __pa(vcpu->arch.mmu->lm_root);

but lm_root is allocated with get_zeroed_page and therefore to_shadow_page is NULL.

I am thinking of testing simply "if (tdp_enabled)" so that we can see if there are other cases with to_shadow_page(hpa) == NULL and we don't sweep them under the rug. Or test "if (tdp_enabled)" and also WARN if !sp. What do you think?

Paolo

author
#4 lore raw
Quoted 52 lines · Paolo Bonzini
Quoted 28 lines · Ben Gardon

When PAE is in use, the root_hpa will not have a shadow page assoicated with it. In this case the kernel will crash with a NULL pointer dereference. Add checks to ensure is_tdp_mmu_root works as intended even when using PAE.

Tested: compiles

Fixes 02c00b3a2f7e ("kvm: x86/mmu: Allocate and free TDP MMU roots")
---
  arch/x86/kvm/mmu/tdp_mmu.c | 10 ++++++++++
  1 file changed, 10 insertions(+)
arch/x86/kvm/mmu/tdp_mmu.c +3 -0
index 27e381c9da6c..13013f4d98ad 100644
@@ -49,8 +49,18 @@ bool is_tdp_mmu_root(struct kvm *kvm, hpa_t hpa)
  {
      struct kvm_mmu_page *sp;
 
+     if (WARN_ON(!VALID_PAGE(hpa)))
+             return false;
+
      sp = to_shadow_page(hpa);
 

arch/x86/kvm/mmu/tdp_mmu.c bool is_tdp_mmu_root(struct kvm *kvm, hpa_t hpa) · line 66

Quoted 10 lines
arch/x86/kvm/mmu/tdp_mmu.c +7 -0
+     /*
+      * If this VM is being run with PAE, the TDP MMU will not be enabled
+      * and the root HPA will not have a shadow page associated with it.
+      */
+     if (!sp)
+             return false;
+
      return sp->tdp_mmu_page && sp->root_count;
  }
 

If this was just PAE, it would be easier to test "if (shadow_root_level

Quoted 1 line

= PT64_ROOT_4LEVEL)"---and more correct too, because using the

page_private of __pa(vcpu->arch.mmu->pae_root) is a bit untidy; we should only use page_private for pages that we know have a shadow page.

In Jamie's case however, it is x86_64 (so kvm_mmu_get_tdp_level(vcpu) == 4 and therefore the "if (shadow_root_level >= PT64_ROOT_4LEVEL)" would be true) but without EPT. In that case we go through

        vcpu->arch.mmu->root_hpa = __pa(vcpu->arch.mmu->lm_root);
Quoted 7 lines

but lm_root is allocated with get_zeroed_page and therefore to_shadow_page is NULL.

I am thinking of testing simply "if (tdp_enabled)" so that we can see if there are other cases with to_shadow_page(hpa) == NULL and we don't sweep them under the rug. Or test "if (tdp_enabled)" and also WARN if !sp. What do you think?

That sounds good to me. I see you already mailed out a modified version of this change, so I'll go review it.
Thanks!

Quoted 3 lines

Paolo