On 5/1/24 10:51, Michael Roth wrote:
In the case of SEV-SNP, whether or not a 2MB page can be mapped via a
2MB mapping in the guest's nested page table depends on whether or not
any subpages within the range have already been initialized as private
in the RMP table. The existing mixed-attribute tracking in KVM is
insufficient here, for instance:
- gmem allocates 2MB page
- guest issues PVALIDATE on 2MB page
- guest later converts a subpage to shared
- SNP host code issues PSMASH to split 2MB RMP mapping to 4K
- KVM MMU splits NPT mapping to 4K
- guest later converts that shared page back to private
At this point there are no mixed attributes, and KVM would normally
allow for 2MB NPT mappings again, but this is actually not allowed
because the RMP table mappings are 4K and cannot be promoted on the
hypervisor side, so the NPT mappings must still be limited to 4K to
match this.
Add a hook to determine the max NPT mapping size in situations like
this.
Suggested-by: Sean Christopherson <seanjc@xxxxxxxxxx>
Signed-off-by: Michael Roth <michael.roth@xxxxxxx>
---
arch/x86/include/asm/kvm-x86-ops.h | 1 +
arch/x86/include/asm/kvm_host.h | 1 +
arch/x86/kvm/mmu/mmu.c | 18 ++++++++++++++++--
3 files changed, 18 insertions(+), 2 deletions(-)
diff --git a/arch/x86/include/asm/kvm-x86-ops.h b/arch/x86/include/asm/kvm-x86-ops.h
index c81990937ab4..566d19b02483 100644
--- a/arch/x86/include/asm/kvm-x86-ops.h
+++ b/arch/x86/include/asm/kvm-x86-ops.h
@@ -140,6 +140,7 @@ KVM_X86_OP_OPTIONAL_RET0(vcpu_get_apicv_inhibit_reasons);
KVM_X86_OP_OPTIONAL(get_untagged_addr)
KVM_X86_OP_OPTIONAL(alloc_apic_backing_page)
KVM_X86_OP_OPTIONAL_RET0(gmem_prepare)
+KVM_X86_OP_OPTIONAL_RET0(private_max_mapping_level)
KVM_X86_OP_OPTIONAL(gmem_invalidate)
#undef KVM_X86_OP
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index c6c5018376be..87265b73906a 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -1816,6 +1816,7 @@ struct kvm_x86_ops {
void *(*alloc_apic_backing_page)(struct kvm_vcpu *vcpu);
int (*gmem_prepare)(struct kvm *kvm, kvm_pfn_t pfn, gfn_t gfn, int max_order);
void (*gmem_invalidate)(kvm_pfn_t start, kvm_pfn_t end);
+ int (*private_max_mapping_level)(struct kvm *kvm, kvm_pfn_t pfn);
};
struct kvm_x86_nested_ops {
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index 510eb1117012..0d556da052f6 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -4271,6 +4271,20 @@ static inline u8 kvm_max_level_for_order(int order)
return PG_LEVEL_4K;
}
+static u8 kvm_max_private_mapping_level(struct kvm *kvm, kvm_pfn_t pfn,
+ u8 max_level, int gmem_order)
+{
+ if (max_level == PG_LEVEL_4K)
+ return PG_LEVEL_4K;
+
+ max_level = min(kvm_max_level_for_order(gmem_order), max_level);
+ if (max_level == PG_LEVEL_4K)
+ return PG_LEVEL_4K;
+
+ return min(max_level,
+ static_call(kvm_x86_private_max_mapping_level)(kvm, pfn));
+}
Since you're returning 0 both as a default and, later in the series, for non-SNP guests, you need to treat 0 as "don't care":
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index de35dee25bf6..62ad38b2a8c9 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -4274,6 +4274,8 @@ static inline u8 kvm_max_level_for_order(int order)
static u8 kvm_max_private_mapping_level(struct kvm *kvm, kvm_pfn_t pfn,
u8 max_level, int gmem_order)
{
+ u8 req_max_level;
+
if (max_level == PG_LEVEL_4K)
return PG_LEVEL_4K;
@@ -4281,8 +4283,11 @@ static u8 kvm_max_private_mapping_level(struct kvm *kvm, kvm_pfn_t pfn,
if (max_level == PG_LEVEL_4K)
return PG_LEVEL_4K;
- return min(max_level,
- static_call(kvm_x86_private_max_mapping_level)(kvm, pfn));
+ req_max_level = static_call(kvm_x86_private_max_mapping_level)(kvm, pfn);
+ if (req_max_level)
+ max_level = min(max_level, req_max_level);
+
+ return req_max_level;
}
static int kvm_faultin_pfn_private(struct kvm_vcpu *vcpu,
Not beautiful but it does the job I guess.
Paolo
