KVM: nVMX: Disallow access to vmcs12 fields that aren't supported by "hardware"

Disallow access (VMREAD/VMWRITE), both emulated and via a shadow VMCS, to
VMCS fields that the loaded incarnation of KVM doesn't support, e.g. due
to lack of hardware support, as a middle ground between allowing access to
any vmcs12 field defined by KVM (current behavior) and gating access based
on the userspace-defined vCPU model (the most functionally correct, but
very costly, implementation).

Disallowing access to unsupported fields helps a tiny bit in terms of
closing the virtualization hole (see below), but the main motivation is to
avoid having to weed out unsupported fields when synchronizing between
vmcs12 and a shadow VMCS.  Because shadow VMCS accesses are done via
VMREAD and VMWRITE, KVM _must_ filter out unsupported fields (or eat
VMREAD/VMWRITE failures), and filtering out just shadow VMCS fields is
about the same amount of effort, and arguably much more confusing.

As a bonus, this also fixes a KVM-Unit-Test failure bug when running on
_hardware_ without support for TSC Scaling, which fails with the same
signature as the bug fixed by commit ba1f824 ("KVM: nVMX: Dynamically
compute max VMCS index for vmcs12"):

  FAIL: VMX_VMCS_ENUM.MAX_INDEX expected: 19, actual: 17

Dynamically computing the max VMCS index only resolved the issue where KVM
was hardcoding max index, but for CPUs with TSC Scaling, that was "good
enough".

Reviewed-by: Chao Gao <chao.gao@intel.com>
Reviewed-by: Xin Li <xin@zytor.com>
Cc: Xiaoyao Li <xiaoyao.li@intel.com>
Cc: Yosry Ahmed <yosry.ahmed@linux.dev>
Link: https://lore.kernel.org/all/20251026201911.505204-22-xin@zytor.com
Link: https://lore.kernel.org/all/YR2Tf9WPNEzrE7Xg@google.com
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Link: https://patch.msgid.link/20260115173427.716021-4-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>

Author: sean-jc

arch/x86/kvm/vmx/nested.c

@@ -86,6 +86,9 @@ static void init_vmcs_shadow_fields(void)
 			pr_err("Missing field from shadow_read_only_field %x\n",
 			       field + 1);
 
+		if (get_vmcs12_field_offset(field) < 0)
+			continue;
+
 		clear_bit(field, vmx_vmread_bitmap);
 		if (field & 1)
 #ifdef CONFIG_X86_64
@@ -111,6 +114,9 @@ static void init_vmcs_shadow_fields(void)
 			  field <= GUEST_TR_AR_BYTES,
 			  "Update vmcs12_write_any() to drop reserved bits from AR_BYTES");
 
+		if (get_vmcs12_field_offset(field) < 0)
+			continue;
+
 		/*
 		 * PML and the preemption timer can be emulated, but the
 		 * processor cannot vmwrite to fields that don't exist
@@ -7074,12 +7080,6 @@ void nested_vmx_set_vmcs_shadowing_bitmap(void)
 	}
 }
 
-/*
- * Indexing into the vmcs12 uses the VMCS encoding rotated left by 6.  Undo
- * that madness to get the encoding for comparison.
- */
-#define VMCS12_IDX_TO_ENC(idx) ((u16)(((u16)(idx) >> 6) | ((u16)(idx) << 10)))
-
 static u64 nested_vmx_calc_vmcs_enum_msr(void)
 {
 	/*
@@ -7407,6 +7407,12 @@ __init int nested_vmx_hardware_setup(int (*exit_handlers[])(struct kvm_vcpu *))
 {
 	int i;
 
+	/*
+	 * Note!  The set of supported vmcs12 fields is consumed by both VMX
+	 * MSR and shadow VMCS setup.
+	 */
+	nested_vmx_setup_vmcs12_fields();
+
 	nested_vmx_setup_ctls_msrs(&vmcs_config, vmx_capability.ept);
 
 	if (!cpu_has_vmx_shadow_vmcs())

arch/x86/kvm/vmx/vmcs.h

@@ -11,7 +11,15 @@
 
 #include "capabilities.h"
 
+/*
+ * Indexing into the vmcs12 uses the VMCS encoding rotated left by 6 as a very
+ * rudimentary compression of the range of indices.  The compression ratio is
+ * good enough to allow KVM to use a (very sparsely populated) array without
+ * wasting too much memory, while the "algorithm" is fast enough to be used to
+ * lookup vmcs12 fields on-demand, e.g. for emulation.
+ */
 #define ROL16(val, n) ((u16)(((u16)(val) << (n)) | ((u16)(val) >> (16 - (n)))))
+#define VMCS12_IDX_TO_ENC(idx) ROL16(idx, 10)
 #define ENC_TO_VMCS12_IDX(enc) ROL16(enc, 6)
 
 struct vmcs_hdr {

arch/x86/kvm/vmx/vmcs12.c

@@ -9,7 +9,7 @@
 	FIELD(number, name),						\
 	[ENC_TO_VMCS12_IDX(number##_HIGH)] = VMCS12_OFFSET(name) + sizeof(u32)
 
-const unsigned short vmcs12_field_offsets[] = {
+static const u16 kvm_supported_vmcs12_field_offsets[] __initconst = {
 	FIELD(VIRTUAL_PROCESSOR_ID, virtual_processor_id),
 	FIELD(POSTED_INTR_NV, posted_intr_nv),
 	FIELD(GUEST_ES_SELECTOR, guest_es_selector),
@@ -158,4 +158,70 @@ const unsigned short vmcs12_field_offsets[] = {
 	FIELD(HOST_SSP, host_ssp),
 	FIELD(HOST_INTR_SSP_TABLE, host_ssp_tbl),
 };
-const unsigned int nr_vmcs12_fields = ARRAY_SIZE(vmcs12_field_offsets);
+
+u16 vmcs12_field_offsets[ARRAY_SIZE(kvm_supported_vmcs12_field_offsets)] __ro_after_init;
+unsigned int nr_vmcs12_fields __ro_after_init;
+
+#define VMCS12_CASE64(enc) case enc##_HIGH: case enc
+
+static __init bool cpu_has_vmcs12_field(unsigned int idx)
+{
+	switch (VMCS12_IDX_TO_ENC(idx)) {
+	case VIRTUAL_PROCESSOR_ID:
+		return cpu_has_vmx_vpid();
+	case POSTED_INTR_NV:
+		return cpu_has_vmx_posted_intr();
+	VMCS12_CASE64(TSC_MULTIPLIER):
+		return cpu_has_vmx_tsc_scaling();
+	case TPR_THRESHOLD:
+	VMCS12_CASE64(VIRTUAL_APIC_PAGE_ADDR):
+		return cpu_has_vmx_tpr_shadow();
+	VMCS12_CASE64(APIC_ACCESS_ADDR):
+		return cpu_has_vmx_virtualize_apic_accesses();
+	VMCS12_CASE64(POSTED_INTR_DESC_ADDR):
+		return cpu_has_vmx_posted_intr();
+	case GUEST_INTR_STATUS:
+		return cpu_has_vmx_virtual_intr_delivery();
+	VMCS12_CASE64(VM_FUNCTION_CONTROL):
+	VMCS12_CASE64(EPTP_LIST_ADDRESS):
+		return cpu_has_vmx_vmfunc();
+	VMCS12_CASE64(EPT_POINTER):
+		return cpu_has_vmx_ept();
+	VMCS12_CASE64(XSS_EXIT_BITMAP):
+		return cpu_has_vmx_xsaves();
+	VMCS12_CASE64(ENCLS_EXITING_BITMAP):
+		return cpu_has_vmx_encls_vmexit();
+	VMCS12_CASE64(GUEST_IA32_PERF_GLOBAL_CTRL):
+	VMCS12_CASE64(HOST_IA32_PERF_GLOBAL_CTRL):
+		return cpu_has_load_perf_global_ctrl();
+	case SECONDARY_VM_EXEC_CONTROL:
+		return cpu_has_secondary_exec_ctrls();
+	case GUEST_S_CET:
+	case GUEST_SSP:
+	case GUEST_INTR_SSP_TABLE:
+	case HOST_S_CET:
+	case HOST_SSP:
+	case HOST_INTR_SSP_TABLE:
+		return cpu_has_load_cet_ctrl();
+
+	/* KVM always emulates PML and the VMX preemption timer in software. */
+	case GUEST_PML_INDEX:
+	case VMX_PREEMPTION_TIMER_VALUE:
+	default:
+		return true;
+	}
+}
+
+void __init nested_vmx_setup_vmcs12_fields(void)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(kvm_supported_vmcs12_field_offsets); i++) {
+		if (!kvm_supported_vmcs12_field_offsets[i] ||
+		    !cpu_has_vmcs12_field(i))
+			continue;
+
+		vmcs12_field_offsets[i] = kvm_supported_vmcs12_field_offsets[i];
+		nr_vmcs12_fields = i + 1;
+	}
+}

arch/x86/kvm/vmx/vmcs12.h

@@ -374,8 +374,10 @@ static inline void vmx_check_vmcs12_offsets(void)
 	CHECK_OFFSET(guest_pml_index, 996);
 }
 
-extern const unsigned short vmcs12_field_offsets[];
-extern const unsigned int nr_vmcs12_fields;
+extern u16 vmcs12_field_offsets[] __ro_after_init;
+extern unsigned int nr_vmcs12_fields __ro_after_init;
+
+void __init nested_vmx_setup_vmcs12_fields(void);
 
 static inline short get_vmcs12_field_offset(unsigned long field)
 {