Skip to contentSkip to conversation

KVM: x86/mmu: Calculate and check "full" mmu_role for nested MMU

PATCH Applied Sean Christopherson 3 participants 8 messages Jun 10, 2021 – Jun 11, 2021

Subsystems (3)
lore
Share & tools
lore

Sean Christopherson

author maintainer
#0 lore raw

PATCH 1/1 KVM: x86/mmu: Calculate and check "full" mmu_role for nested MMU

Calculate and check the full mmu_role when initializing the MMU context for the nested MMU, where "full" means the bits and pieces of the role that aren't handled by kvm_calc_mmu_role_common(). While the nested MMU isn't used for shadow paging, things like the number of levels in the guest's page tables are surprisingly important when walking the guest page tables. Failure to reinitialize the nested MMU context if L2's paging mode changes can result in unexpected and/or missed page faults, and likely other explosions.

E.g. if an L1 vCPU is running both a 32-bit PAE L2 and a 64-bit L2, the "common" role calculation will yield the same role for both L2s. If the 64-bit L2 is run after the 32-bit PAE L2, L0 will fail to reinitialize the nested MMU context, ultimately resulting in a bad walk of L2's page tables as the MMU will still have a guest root_level of PT32E_ROOT_LEVEL.

  WARNING: CPU: 4 PID: 167334 at arch/x86/kvm/vmx/vmx.c:3075 ept_save_pdptrs+0x15/0xe0 [kvm_intel]
  Modules linked in: kvm_intel]
  CPU: 4 PID: 167334 Comm: CPU 3/KVM Not tainted 5.13.0-rc1-d849817d5673-reqs #185
  Hardware name: ASUS Q87M-E/Q87M-E, BIOS 1102 03/03/2014
  RIP: 0010:ept_save_pdptrs+0x15/0xe0 [kvm_intel]
  Code: <0f> 0b c3 f6 87 d8 02 00f
  RSP: 0018:ffffbba702dbba00 EFLAGS: 00010202
  RAX: 0000000000000011 RBX: 0000000000000002 RCX: ffffffff810a2c08
  RDX: ffff91d7bc30acc0 RSI: 0000000000000011 RDI: ffff91d7bc30a600
  RBP: ffff91d7bc30a600 R08: 0000000000000010 R09: 0000000000000007
  R10: 0000000000000000 R11: 0000000000000000 R12: ffff91d7bc30a600
  R13: ffff91d7bc30acc0 R14: ffff91d67c123460 R15: 0000000115d7e005
  FS:  00007fe8e9ffb700(0000) GS:ffff91d90fb00000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000000 CR3: 000000029f15a001 CR4: 00000000001726e0
  Call Trace:
   kvm_pdptr_read+0x3a/0x40 [kvm]
   paging64_walk_addr_generic+0x327/0x6a0 [kvm]
   paging64_gva_to_gpa_nested+0x3f/0xb0 [kvm]
   kvm_fetch_guest_virt+0x4c/0xb0 [kvm]
   __do_insn_fetch_bytes+0x11a/0x1f0 [kvm]
   x86_decode_insn+0x787/0x1490 [kvm]
   x86_decode_emulated_instruction+0x58/0x1e0 [kvm]
   x86_emulate_instruction+0x122/0x4f0 [kvm]
   vmx_handle_exit+0x120/0x660 [kvm_intel]
   kvm_arch_vcpu_ioctl_run+0xe25/0x1cb0 [kvm]
   kvm_vcpu_ioctl+0x211/0x5a0 [kvm]
   __x64_sys_ioctl+0x83/0xb0
   do_syscall_64+0x40/0xb0
   entry_SYSCALL_64_after_hwframe+0x44/0xae
Cc Vitaly Kuznetsov <[email protected]>
Fixes bf627a928837 ("x86/kvm/mmu: check if MMU reconfiguration is needed in init_kvm_nested_mmu()")
---
 arch/x86/kvm/mmu/mmu.c | 26 +++++++++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)
arch/x86/kvm/mmu/mmu.c +25 -1
index 0144c40d09c7..8d5876dfc6b7 100644
@@ -4739,9 +4739,33 @@ static void init_kvm_softmmu(struct kvm_vcpu *vcpu)
 	context->inject_page_fault = kvm_inject_page_fault;
 }
 
+static union kvm_mmu_role kvm_calc_nested_mmu_role(struct kvm_vcpu *vcpu)
+{
+	union kvm_mmu_role role = kvm_calc_shadow_root_page_role_common(vcpu, false);
+
+	/*
+	 * Nested MMUs are used only for walking L2's gva->gpa, they never have
+	 * shadow pages of their own and so "direct" has no meaning.   Set it
+	 * to "true" to try to detect bogus usage of the nested MMU.
+	 */
+	role.base.direct = true;
+
+	if (!is_paging(vcpu))
+		role.base.level = 0;
+	else if (is_long_mode(vcpu))
+		role.base.level = is_la57_mode(vcpu) ? PT64_ROOT_5LEVEL :
+						       PT64_ROOT_4LEVEL;
+	else if (is_pae(vcpu))
+		role.base.level = PT32E_ROOT_LEVEL;
+	else
+		role.base.level = PT32_ROOT_LEVEL;
+
+	return role;
+}
+
 static void init_kvm_nested_mmu(struct kvm_vcpu *vcpu)
 {
-	union kvm_mmu_role new_role = kvm_calc_mmu_role_common(vcpu, false);
+	union kvm_mmu_role new_role = kvm_calc_nested_mmu_role(vcpu);
 	struct kvm_mmu *g_context = &vcpu->arch.nested_mmu;
 
 	if (new_role.as_u64 == g_context->mmu_role.as_u64)
Signature · 1 line
2.32.0.272.g935e593368-goog
Reviewed-by on v1 1/1
#1 lore raw
Quoted 83 lines · Sean Christopherson

Calculate and check the full mmu_role when initializing the MMU context for the nested MMU, where "full" means the bits and pieces of the role that aren't handled by kvm_calc_mmu_role_common(). While the nested MMU isn't used for shadow paging, things like the number of levels in the guest's page tables are surprisingly important when walking the guest page tables. Failure to reinitialize the nested MMU context if L2's paging mode changes can result in unexpected and/or missed page faults, and likely other explosions.

E.g. if an L1 vCPU is running both a 32-bit PAE L2 and a 64-bit L2, the "common" role calculation will yield the same role for both L2s. If the 64-bit L2 is run after the 32-bit PAE L2, L0 will fail to reinitialize the nested MMU context, ultimately resulting in a bad walk of L2's page tables as the MMU will still have a guest root_level of PT32E_ROOT_LEVEL.

  WARNING: CPU: 4 PID: 167334 at arch/x86/kvm/vmx/vmx.c:3075 ept_save_pdptrs+0x15/0xe0 [kvm_intel]
  Modules linked in: kvm_intel]
  CPU: 4 PID: 167334 Comm: CPU 3/KVM Not tainted 5.13.0-rc1-d849817d5673-reqs #185
  Hardware name: ASUS Q87M-E/Q87M-E, BIOS 1102 03/03/2014
  RIP: 0010:ept_save_pdptrs+0x15/0xe0 [kvm_intel]
  Code: <0f> 0b c3 f6 87 d8 02 00f
  RSP: 0018:ffffbba702dbba00 EFLAGS: 00010202
  RAX: 0000000000000011 RBX: 0000000000000002 RCX: ffffffff810a2c08
  RDX: ffff91d7bc30acc0 RSI: 0000000000000011 RDI: ffff91d7bc30a600
  RBP: ffff91d7bc30a600 R08: 0000000000000010 R09: 0000000000000007
  R10: 0000000000000000 R11: 0000000000000000 R12: ffff91d7bc30a600
  R13: ffff91d7bc30acc0 R14: ffff91d67c123460 R15: 0000000115d7e005
  FS:  00007fe8e9ffb700(0000) GS:ffff91d90fb00000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000000 CR3: 000000029f15a001 CR4: 00000000001726e0
  Call Trace:
   kvm_pdptr_read+0x3a/0x40 [kvm]
   paging64_walk_addr_generic+0x327/0x6a0 [kvm]
   paging64_gva_to_gpa_nested+0x3f/0xb0 [kvm]
   kvm_fetch_guest_virt+0x4c/0xb0 [kvm]
   __do_insn_fetch_bytes+0x11a/0x1f0 [kvm]
   x86_decode_insn+0x787/0x1490 [kvm]
   x86_decode_emulated_instruction+0x58/0x1e0 [kvm]
   x86_emulate_instruction+0x122/0x4f0 [kvm]
   vmx_handle_exit+0x120/0x660 [kvm_intel]
   kvm_arch_vcpu_ioctl_run+0xe25/0x1cb0 [kvm]
   kvm_vcpu_ioctl+0x211/0x5a0 [kvm]
   __x64_sys_ioctl+0x83/0xb0
   do_syscall_64+0x40/0xb0
   entry_SYSCALL_64_after_hwframe+0x44/0xae
Cc Vitaly Kuznetsov <[email protected]>
Fixes bf627a928837 ("x86/kvm/mmu: check if MMU reconfiguration is needed in init_kvm_nested_mmu()")
---
 arch/x86/kvm/mmu/mmu.c | 26 +++++++++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)
arch/x86/kvm/mmu/mmu.c +21 -0
index 0144c40d09c7..8d5876dfc6b7 100644
@@ -4739,9 +4739,33 @@ static void init_kvm_softmmu(struct kvm_vcpu *vcpu)
 	context->inject_page_fault = kvm_inject_page_fault;
 }
 
+static union kvm_mmu_role kvm_calc_nested_mmu_role(struct kvm_vcpu *vcpu)
+{
+	union kvm_mmu_role role = kvm_calc_shadow_root_page_role_common(vcpu, false);
+
+	/*
+	 * Nested MMUs are used only for walking L2's gva->gpa, they never have
+	 * shadow pages of their own and so "direct" has no meaning.   Set it
+	 * to "true" to try to detect bogus usage of the nested MMU.
+	 */
+	role.base.direct = true;
+
+	if (!is_paging(vcpu))
+		role.base.level = 0;
+	else if (is_long_mode(vcpu))
+		role.base.level = is_la57_mode(vcpu) ? PT64_ROOT_5LEVEL :
+						       PT64_ROOT_4LEVEL;
+	else if (is_pae(vcpu))
+		role.base.level = PT32E_ROOT_LEVEL;
+	else
+		role.base.level = PT32_ROOT_LEVEL;
+

arch/x86/kvm/mmu/mmu.c static void init_kvm_softmmu(struct kvm_vcpu *vcpu) · line 4771

Quoted 10 lines
arch/x86/kvm/mmu/mmu.c +4 -1
+	return role;
+}
+
 static void init_kvm_nested_mmu(struct kvm_vcpu *vcpu)
 {
-	union kvm_mmu_role new_role = kvm_calc_mmu_role_common(vcpu, false);
+	union kvm_mmu_role new_role = kvm_calc_nested_mmu_role(vcpu);
 	struct kvm_mmu *g_context = &vcpu->arch.nested_mmu;
 
 	if (new_role.as_u64 == g_context->mmu_role.as_u64)

Oh, I see. Cautiously:

What I don't quite like (besides the fact that this 'nested_mmu' exists but I don't see an elegant way to get rid of it) is the fact that we now have the same logic to compute 'level' both in kvm_calc_nested_mmu_role() and init_kvm_nested_mmu(). We could've avoided that by re-aranging code in init_kvm_nested_mmu() I guess. Something like (untested):

arch/x86/kvm/mmu/mmu.c +39 -31
index 0144c40d09c7..3b1d1c5016c9 100644
@@ -4741,22 +4741,12 @@ static void init_kvm_softmmu(struct kvm_vcpu *vcpu)
 
 static void init_kvm_nested_mmu(struct kvm_vcpu *vcpu)
 {
-	union kvm_mmu_role new_role = kvm_calc_mmu_role_common(vcpu, false);
+	union kvm_mmu_role new_role = kvm_calc_shadow_root_page_role_common(vcpu, false);
 	struct kvm_mmu *g_context = &vcpu->arch.nested_mmu;
-
-	if (new_role.as_u64 == g_context->mmu_role.as_u64)
-		return;
-
-	g_context->mmu_role.as_u64 = new_role.as_u64;
-	g_context->get_guest_pgd     = get_cr3;
-	g_context->get_pdptr         = kvm_pdptr_read;
-	g_context->inject_page_fault = kvm_inject_page_fault;
-
-	/*
-	 * L2 page tables are never shadowed, so there is no need to sync
-	 * SPTEs.
-	 */
-	g_context->invlpg            = NULL;
+	gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gpa_t gva_or_gpa,
+                            u32 access, struct x86_exception *exception);
+	int root_level;
+	bool nx;
 
 	/*
 	 * Note that arch.mmu->gva_to_gpa translates l2_gpa to l1_gpa using
@@ -4767,27 +4757,45 @@ static void init_kvm_nested_mmu(struct kvm_vcpu *vcpu)
 	 * the gva_to_gpa functions between mmu and nested_mmu are swapped.
 	 */
 	if (!is_paging(vcpu)) {
-		g_context->nx = false;
-		g_context->root_level = 0;
-		g_context->gva_to_gpa = nonpaging_gva_to_gpa_nested;
+		nx = false;
+		root_level = 0;
+		gva_to_gpa = nonpaging_gva_to_gpa_nested;
 	} else if (is_long_mode(vcpu)) {
-		g_context->nx = is_nx(vcpu);
-		g_context->root_level = is_la57_mode(vcpu) ?
-					PT64_ROOT_5LEVEL : PT64_ROOT_4LEVEL;
-		reset_rsvds_bits_mask(vcpu, g_context);
-		g_context->gva_to_gpa = paging64_gva_to_gpa_nested;
+		nx = is_nx(vcpu);
+		root_level = is_la57_mode(vcpu) ?
+			PT64_ROOT_5LEVEL : PT64_ROOT_4LEVEL;
+		gva_to_gpa = paging64_gva_to_gpa_nested;
 	} else if (is_pae(vcpu)) {
-		g_context->nx = is_nx(vcpu);
-		g_context->root_level = PT32E_ROOT_LEVEL;
-		reset_rsvds_bits_mask(vcpu, g_context);
-		g_context->gva_to_gpa = paging64_gva_to_gpa_nested;
+		nx = is_nx(vcpu);
+		root_level = PT32E_ROOT_LEVEL;
+		gva_to_gpa = paging64_gva_to_gpa_nested;
 	} else {
-		g_context->nx = false;
-		g_context->root_level = PT32_ROOT_LEVEL;
-		reset_rsvds_bits_mask(vcpu, g_context);
-		g_context->gva_to_gpa = paging32_gva_to_gpa_nested;
+		nx = false;
+		root_level = PT32_ROOT_LEVEL;
+		gva_to_gpa = paging32_gva_to_gpa_nested;
 	}
 
+	if ((new_role.as_u64 == g_context->mmu_role.as_u64) &&
+	    (root_level == g_context->root_level))
+		return;
+
+	g_context->nx = nx;
+	g_context->root_level = root_level;
+	g_context->gva_to_gpa = gva_to_gpa;
+	if (root_level)
+		reset_rsvds_bits_mask(vcpu, g_context);
+
+	g_context->mmu_role.as_u64 = new_role.as_u64;
+	g_context->get_guest_pgd     = get_cr3;
+	g_context->get_pdptr         = kvm_pdptr_read;
+	g_context->inject_page_fault = kvm_inject_page_fault;
+
+	/*
+	 * L2 page tables are never shadowed, so there is no need to sync
+	 * SPTEs.
+	 */
+	g_context->invlpg            = NULL;
+
 	update_permission_bitmask(vcpu, g_context, false);
 	update_pkru_bitmask(vcpu, g_context, false);
 	update_last_nonleaf_level(vcpu, g_context);
Signature · 1 line
Vitaly
author maintainer
#2 lore raw
Quoted 6 lines · Vitaly Kuznetsov

What I don't quite like (besides the fact that this 'nested_mmu' exists but I don't see an elegant way to get rid of it) is the fact that we now have the same logic to compute 'level' both in kvm_calc_nested_mmu_role() and init_kvm_nested_mmu(). We could've avoided that by re-aranging code in init_kvm_nested_mmu() I guess. Something like (untested):

Yep, cleaning all that up is on my todo list, but there are some hurdles to clear first.

My thought is to either (a) initialize the context from the role, or (b) drop the duplicate context information altogether. For (a), the NX bit is calculated incorrectly in the role stuff, e.g. if paging is disabled then NX is effectively 0, and I need that fix for the vCPU RESET/INIT series. It's benign for the role, but not for the context. And (b) will require auditing for all flavors of MMUs; I wouldn't be the least bit surprised to discover there's a corner case (or just a regular case) that I'm overlooking.

author maintainer
#3 lore raw
Quoted 11 lines · Sean Christopherson
Quoted 6 lines · Vitaly Kuznetsov

What I don't quite like (besides the fact that this 'nested_mmu' exists but I don't see an elegant way to get rid of it) is the fact that we now have the same logic to compute 'level' both in kvm_calc_nested_mmu_role() and init_kvm_nested_mmu(). We could've avoided that by re-aranging code in init_kvm_nested_mmu() I guess. Something like (untested):

Yep, cleaning all that up is on my todo list, but there are some hurdles to clear first.

Quoted 7 lines

My thought is to either (a) initialize the context from the role, or (b) drop the duplicate context information altogether. For (a), the NX bit is calculated incorrectly in the role stuff, e.g. if paging is disabled then NX is effectively 0, and I need that fix for the vCPU RESET/INIT series. It's benign for the role, but not for the context. And (b) will require auditing for all flavors of MMUs; I wouldn't be the least bit surprised to discover there's a corner case (or just a regular case) that I'm overlooking.

Ugh, nested NPT is completely fubar. Except for the "core" mode, all of the role and context calculations are done using L2 state instead of L1 host state. The APM explicitly states that CR0.WP is ignored, and SMEP/SMAP are implicitly ignored by virtue of the NPT walks always being tagged "user", but KVM botches the NX behavior and would mess up LA57 if it were supported.

I sort out the mess, though I'm not sure how it will interact with the reset series...

maintainer Applied
#4 lore raw
In reply to Sean Christopherson
Quoted 84 lines · Sean Christopherson

Calculate and check the full mmu_role when initializing the MMU context for the nested MMU, where "full" means the bits and pieces of the role that aren't handled by kvm_calc_mmu_role_common(). While the nested MMU isn't used for shadow paging, things like the number of levels in the guest's page tables are surprisingly important when walking the guest page tables. Failure to reinitialize the nested MMU context if L2's paging mode changes can result in unexpected and/or missed page faults, and likely other explosions.

E.g. if an L1 vCPU is running both a 32-bit PAE L2 and a 64-bit L2, the "common" role calculation will yield the same role for both L2s. If the 64-bit L2 is run after the 32-bit PAE L2, L0 will fail to reinitialize the nested MMU context, ultimately resulting in a bad walk of L2's page tables as the MMU will still have a guest root_level of PT32E_ROOT_LEVEL.

   WARNING: CPU: 4 PID: 167334 at arch/x86/kvm/vmx/vmx.c:3075 ept_save_pdptrs+0x15/0xe0 [kvm_intel]
   Modules linked in: kvm_intel]
   CPU: 4 PID: 167334 Comm: CPU 3/KVM Not tainted 5.13.0-rc1-d849817d5673-reqs #185
   Hardware name: ASUS Q87M-E/Q87M-E, BIOS 1102 03/03/2014
   RIP: 0010:ept_save_pdptrs+0x15/0xe0 [kvm_intel]
   Code: <0f> 0b c3 f6 87 d8 02 00f
   RSP: 0018:ffffbba702dbba00 EFLAGS: 00010202
   RAX: 0000000000000011 RBX: 0000000000000002 RCX: ffffffff810a2c08
   RDX: ffff91d7bc30acc0 RSI: 0000000000000011 RDI: ffff91d7bc30a600
   RBP: ffff91d7bc30a600 R08: 0000000000000010 R09: 0000000000000007
   R10: 0000000000000000 R11: 0000000000000000 R12: ffff91d7bc30a600
   R13: ffff91d7bc30acc0 R14: ffff91d67c123460 R15: 0000000115d7e005
   FS:  00007fe8e9ffb700(0000) GS:ffff91d90fb00000(0000) knlGS:0000000000000000
   CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
   CR2: 0000000000000000 CR3: 000000029f15a001 CR4: 00000000001726e0
   Call Trace:
    kvm_pdptr_read+0x3a/0x40 [kvm]
    paging64_walk_addr_generic+0x327/0x6a0 [kvm]
    paging64_gva_to_gpa_nested+0x3f/0xb0 [kvm]
    kvm_fetch_guest_virt+0x4c/0xb0 [kvm]
    __do_insn_fetch_bytes+0x11a/0x1f0 [kvm]
    x86_decode_insn+0x787/0x1490 [kvm]
    x86_decode_emulated_instruction+0x58/0x1e0 [kvm]
    x86_emulate_instruction+0x122/0x4f0 [kvm]
    vmx_handle_exit+0x120/0x660 [kvm_intel]
    kvm_arch_vcpu_ioctl_run+0xe25/0x1cb0 [kvm]
    kvm_vcpu_ioctl+0x211/0x5a0 [kvm]
    __x64_sys_ioctl+0x83/0xb0
    do_syscall_64+0x40/0xb0
    entry_SYSCALL_64_after_hwframe+0x44/0xae
Cc Vitaly Kuznetsov <[email protected]>
Fixes bf627a928837 ("x86/kvm/mmu: check if MMU reconfiguration is needed in init_kvm_nested_mmu()")
---
  arch/x86/kvm/mmu/mmu.c | 26 +++++++++++++++++++++++++-
  1 file changed, 25 insertions(+), 1 deletion(-)
arch/x86/kvm/mmu/mmu.c +21 -0
index 0144c40d09c7..8d5876dfc6b7 100644
@@ -4739,9 +4739,33 @@ static void init_kvm_softmmu(struct kvm_vcpu *vcpu)
  	context->inject_page_fault = kvm_inject_page_fault;
  }
 
+static union kvm_mmu_role kvm_calc_nested_mmu_role(struct kvm_vcpu *vcpu)
+{
+	union kvm_mmu_role role = kvm_calc_shadow_root_page_role_common(vcpu, false);
+
+	/*
+	 * Nested MMUs are used only for walking L2's gva->gpa, they never have
+	 * shadow pages of their own and so "direct" has no meaning.   Set it
+	 * to "true" to try to detect bogus usage of the nested MMU.
+	 */
+	role.base.direct = true;
+
+	if (!is_paging(vcpu))
+		role.base.level = 0;
+	else if (is_long_mode(vcpu))
+		role.base.level = is_la57_mode(vcpu) ? PT64_ROOT_5LEVEL :
+						       PT64_ROOT_4LEVEL;
+	else if (is_pae(vcpu))
+		role.base.level = PT32E_ROOT_LEVEL;
+	else
+		role.base.level = PT32_ROOT_LEVEL;
+

arch/x86/kvm/mmu/mmu.c static void init_kvm_softmmu(struct kvm_vcpu *vcpu) · line 4771

Quoted 10 lines
arch/x86/kvm/mmu/mmu.c +4 -1
+	return role;
+}
+
  static void init_kvm_nested_mmu(struct kvm_vcpu *vcpu)
  {
-	union kvm_mmu_role new_role = kvm_calc_mmu_role_common(vcpu, false);
+	union kvm_mmu_role new_role = kvm_calc_nested_mmu_role(vcpu);
  	struct kvm_mmu *g_context = &vcpu->arch.nested_mmu;
 
  	if (new_role.as_u64 == g_context->mmu_role.as_u64)

Queued, thanks.

Paolo

maintainer
#5 lore raw
In reply to Sean Christopherson
Quoted 2 lines · Sean Christopherson

things like the number of levels in the guest's page tables are surprisingly important when walking the guest page tables

Along which path though? I would have naively expected those to be driven only by the context->root_level.

Paolo

author maintainer
#6 lore raw
Quoted 6 lines · Paolo Bonzini
Quoted 2 lines · Sean Christopherson

things like the number of levels in the guest's page tables are surprisingly important when walking the guest page tables

Along which path though? I would have naively expected those to be driven only by the context->root_level.

The functional code is driven by context->root_level, but if KVM doesn't include the level in the mmu_role then it will fail to update context->root_level when L2 changes from 32-bit PAE to 64-bit. If all the CR0/CR4/EFER bits remain the same, only the level will differ. Without this patch, role.level is always '0' for the nested MMU.

maintainer
#7 lore raw
Quoted 8 lines · Sean Christopherson
Quoted 2 lines

Along which path though? I would have naively expected those to be driven only by the context->root_level.

The functional code is driven by context->root_level, but if KVM doesn't include the level in the mmu_role then it will fail to update context->root_level when L2 changes from 32-bit PAE to 64-bit. If all the CR0/CR4/EFER bits remain the same, only the level will differ. Without this patch, role.level is always '0' for the nested MMU.

Yes the problem is the

        if (new_role.as_u64 == g_context->mmu_role.as_u64)
                return;

and the patch lets you preserve the optimization instead of dropping it.
I was wondering if I was missing something else because of the "when
walking the guest page tables" remark.

Paolo