Received-SPF: pass (google.com: domain of linux-kernel+bounces-99705-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Message-ID: <35bc4582-8a03-413b-be0e-4cc419715772@linux.intel.com>
Date: Tue, 12 Mar 2024 08:48:59 +0800
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH 21/21] KVM: x86: Add gmem hook for determining max NPT
 mapping level
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org, seanjc@google.com,
 michael.roth@amd.com, isaku.yamahata@intel.com, thomas.lendacky@amd.com
References: <20240227232100.478238-1-pbonzini@redhat.com>
 <20240227232100.478238-22-pbonzini@redhat.com>
From: Binbin Wu <binbin.wu@linux.intel.com>
In-Reply-To: <20240227232100.478238-22-pbonzini@redhat.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit


On 2/28/2024 7:21 AM, Paolo Bonzini wrote:
> From: Michael Roth <michael.roth@amd.com>
>
> 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
>
> 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.
>
> Signed-off-by: Michael Roth <michael.roth@amd.com>
> Message-Id: <20231230172351.574091-31-michael.roth@amd.com>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> ---
>   arch/x86/include/asm/kvm-x86-ops.h | 1 +
>   arch/x86/include/asm/kvm_host.h    | 1 +
>   arch/x86/kvm/mmu/mmu.c             | 7 +++++++
>   3 files changed, 9 insertions(+)
>
> diff --git a/arch/x86/include/asm/kvm-x86-ops.h b/arch/x86/include/asm/kvm-x86-ops.h
> index 42474acb7375..436e3c157fae 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(vcpu_deliver_sipi_vector)
>   KVM_X86_OP_OPTIONAL_RET0(vcpu_get_apicv_inhibit_reasons);
>   KVM_X86_OP_OPTIONAL(get_untagged_addr)
>   KVM_X86_OP_OPTIONAL_RET0(gmem_prepare)
> +KVM_X86_OP_OPTIONAL_RET0(gmem_validate_fault)
>   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 e523b204697d..259e6bb1e447 100644
> --- a/arch/x86/include/asm/kvm_host.h
> +++ b/arch/x86/include/asm/kvm_host.h
> @@ -1806,6 +1806,7 @@ struct kvm_x86_ops {
>   	gva_t (*get_untagged_addr)(struct kvm_vcpu *vcpu, gva_t gva, unsigned int flags);
>   	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 (*gmem_validate_fault)(struct kvm *kvm, kvm_pfn_t pfn, gfn_t gfn, u8 *max_level);

Since it's named "gmem_validate_fault", can we just passed in the 
"fault" as
an argument to avoid passing in pfn, gfn, max_level individually?
I noticed in Isaku's TDX patch set, fault->private would also need to be
passed in.

>   };
>   
>   struct kvm_x86_nested_ops {
> diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
> index 6b4cb71668df..bcf12ac489f9 100644
> --- a/arch/x86/kvm/mmu/mmu.c
> +++ b/arch/x86/kvm/mmu/mmu.c
> @@ -4339,6 +4339,13 @@ static int kvm_faultin_pfn_private(struct kvm_vcpu *vcpu,
>   			       fault->max_level);
>   	fault->map_writable = !(fault->slot->flags & KVM_MEM_READONLY);
>   
> +	r = static_call(kvm_x86_gmem_validate_fault)(vcpu->kvm, fault->pfn,
> +						     fault->gfn, &fault->max_level);
> +	if (r) {
> +		kvm_release_pfn_clean(fault->pfn);
> +		return r;
> +	}
> +
>   	return RET_PF_CONTINUE;
>   }
>