Received-SPF: pass (google.com: domain of linux-kernel+bounces-139407-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
From: isaku.yamahata@intel.com
To: kvm@vger.kernel.org
Cc: isaku.yamahata@intel.com,
	isaku.yamahata@gmail.com,
	linux-kernel@vger.kernel.org,
	Sean Christopherson <seanjc@google.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Michael Roth <michael.roth@amd.com>,
	David Matlack <dmatlack@google.com>,
	Federico Parola <federico.parola@polito.it>,
	Kai Huang <kai.huang@intel.com>
Subject: [PATCH v2 09/10] KVM: SVM: Implement pre_mmu_map_page() to refuse KVM_MAP_MEMORY
Date: Wed, 10 Apr 2024 15:07:35 -0700
Message-ID: <b884bf12436e69a2f16f86067d6ed17e9b86596b.1712785629.git.isaku.yamahata@intel.com>
In-Reply-To: <cover.1712785629.git.isaku.yamahata@intel.com>
References: <cover.1712785629.git.isaku.yamahata@intel.com>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

From: Isaku Yamahata <isaku.yamahata@intel.com>

Implement vendor callback for KVM_MAP_MEMORY for SEV as EOPNOTSUPP because
it should use SEV-specific ioctl instead.  This patch is only to
demonstrate how to implement the hook.

Compile only tested.  I leave the actual implementation to the SEV folks.

Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
---
v2:
- Newly added
---
 arch/x86/kvm/svm/sev.c | 6 ++++++
 arch/x86/kvm/svm/svm.c | 2 ++
 arch/x86/kvm/svm/svm.h | 9 +++++++++
 3 files changed, 17 insertions(+)

diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c
index 1642d7d49bde..ab17d7c16636 100644
--- a/arch/x86/kvm/svm/sev.c
+++ b/arch/x86/kvm/svm/sev.c
@@ -3322,3 +3322,9 @@ struct page *snp_safe_alloc_page(struct kvm_vcpu *vcpu)
 
 	return p;
 }
+
+int sev_pre_mmu_map_page(struct kvm_vcpu *vcpu,
+			 struct kvm_memory_mapping *mapping, u64 *error_code)
+{
+	return -EOPNOTSUPP;
+}
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index 535018f152a3..a886d4409b00 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -5057,6 +5057,8 @@ static struct kvm_x86_ops svm_x86_ops __initdata = {
 	.vcpu_deliver_sipi_vector = svm_vcpu_deliver_sipi_vector,
 	.vcpu_get_apicv_inhibit_reasons = avic_vcpu_get_apicv_inhibit_reasons,
 	.alloc_apic_backing_page = svm_alloc_apic_backing_page,
+
+	.pre_mmu_map_page = sev_pre_mmu_map_page,
 };
 
 /*
diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h
index 323901782547..c8dafcb0bfc6 100644
--- a/arch/x86/kvm/svm/svm.h
+++ b/arch/x86/kvm/svm/svm.h
@@ -689,6 +689,9 @@ int sev_mem_enc_unregister_region(struct kvm *kvm,
 int sev_vm_copy_enc_context_from(struct kvm *kvm, unsigned int source_fd);
 int sev_vm_move_enc_context_from(struct kvm *kvm, unsigned int source_fd);
 void sev_guest_memory_reclaimed(struct kvm *kvm);
+int sev_pre_mmu_map_page(struct kvm_vcpu *vcpu,
+			 struct kvm_memory_mapping *mapping, u64 *error_code);
+
 int sev_handle_vmgexit(struct kvm_vcpu *vcpu);
 
 /* These symbols are used in common code and are stubbed below.  */
@@ -713,6 +716,12 @@ static inline void __init sev_hardware_setup(void) {}
 static inline void sev_hardware_unsetup(void) {}
 static inline int sev_cpu_init(struct svm_cpu_data *sd) { return 0; }
 static inline int sev_dev_get_attr(u32 group, u64 attr, u64 *val) { return -ENXIO; }
+static inline int sev_pre_mmu_map_page(struct kvm_vcpu *vcpu,
+				       struct kvm_memory_mapping *mapping,
+				       u64 *error_code)
+{
+	return -EOPNOTSUPP;
+}
 #define max_sev_asid 0
 #endif
 
-- 
2.43.2