Merge v6.6.29

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

11 files changed, 67 insertions, 17 deletions

diff --git a/arch/x86/include/asm/barrier.h b/arch/x86/include/asm/barrier.h
index 0216f63a366b..d0795b5fab46 100644
--- a/arch/x86/include/asm/barrier.h
+++ b/arch/x86/include/asm/barrier.h
@@ -79,6 +79,9 @@ do {									\
 #define __smp_mb__before_atomic()	do { } while (0)
 #define __smp_mb__after_atomic()	do { } while (0)
 
+/* Writing to CR3 provides a full memory barrier in switch_mm(). */
+#define smp_mb__after_switch_mm()	do { } while (0)
+
 #include <asm-generic/barrier.h>
 
 #endif /* _ASM_X86_BARRIER_H */
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index fb9f5fa96cc9..ccba66da7a5d 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -828,6 +828,7 @@ struct kvm_vcpu_arch {
 	int cpuid_nent;
 	struct kvm_cpuid_entry2 *cpuid_entries;
 	struct kvm_hypervisor_cpuid kvm_cpuid;
+	bool is_amd_compatible;
 
 	/*
 	 * FIXME: Drop this macro and use KVM_NR_GOVERNED_FEATURES directly
diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index 5ff69b1d39b2..c2dc9b7426ac 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -1651,7 +1651,8 @@ static void __init bhi_select_mitigation(void)
 		return;
 
 	/* Retpoline mitigates against BHI unless the CPU has RRSBA behavior */
-	if (cpu_feature_enabled(X86_FEATURE_RETPOLINE)) {
+	if (boot_cpu_has(X86_FEATURE_RETPOLINE) &&
+	    !boot_cpu_has(X86_FEATURE_RETPOLINE_LFENCE)) {
 		spec_ctrl_disable_kernel_rrsba();
 		if (rrsba_disabled)
 			return;
@@ -2803,11 +2804,13 @@ static const char *spectre_bhi_state(void)
 {
 	if (!boot_cpu_has_bug(X86_BUG_BHI))
 		return "; BHI: Not affected";
-	else if  (boot_cpu_has(X86_FEATURE_CLEAR_BHB_HW))
+	else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_HW))
 		return "; BHI: BHI_DIS_S";
-	else if  (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP))
+	else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP))
 		return "; BHI: SW loop, KVM: SW loop";
-	else if (boot_cpu_has(X86_FEATURE_RETPOLINE) && rrsba_disabled)
+	else if (boot_cpu_has(X86_FEATURE_RETPOLINE) &&
+		 !boot_cpu_has(X86_FEATURE_RETPOLINE_LFENCE) &&
+		 rrsba_disabled)
 		return "; BHI: Retpoline";
 	else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP_ON_VMEXIT))
 		return "; BHI: Vulnerable, KVM: SW loop";
diff --git a/arch/x86/kernel/cpu/cpuid-deps.c b/arch/x86/kernel/cpu/cpuid-deps.c
index e462c1d3800a..6fb6d8a57cec 100644
--- a/arch/x86/kernel/cpu/cpuid-deps.c
+++ b/arch/x86/kernel/cpu/cpuid-deps.c
@@ -44,7 +44,10 @@ static const struct cpuid_dep cpuid_deps[] = {
 	{ X86_FEATURE_F16C,			X86_FEATURE_XMM2,     },
 	{ X86_FEATURE_AES,			X86_FEATURE_XMM2      },
 	{ X86_FEATURE_SHA_NI,			X86_FEATURE_XMM2      },
+	{ X86_FEATURE_GFNI,			X86_FEATURE_XMM2      },
 	{ X86_FEATURE_FMA,			X86_FEATURE_AVX       },
+	{ X86_FEATURE_VAES,			X86_FEATURE_AVX       },
+	{ X86_FEATURE_VPCLMULQDQ,		X86_FEATURE_AVX       },
 	{ X86_FEATURE_AVX2,			X86_FEATURE_AVX,      },
 	{ X86_FEATURE_AVX512F,			X86_FEATURE_AVX,      },
 	{ X86_FEATURE_AVX512IFMA,		X86_FEATURE_AVX512F   },
@@ -56,9 +59,6 @@ static const struct cpuid_dep cpuid_deps[] = {
 	{ X86_FEATURE_AVX512VL,			X86_FEATURE_AVX512F   },
 	{ X86_FEATURE_AVX512VBMI,		X86_FEATURE_AVX512F   },
 	{ X86_FEATURE_AVX512_VBMI2,		X86_FEATURE_AVX512VL  },
-	{ X86_FEATURE_GFNI,			X86_FEATURE_AVX512VL  },
-	{ X86_FEATURE_VAES,			X86_FEATURE_AVX512VL  },
-	{ X86_FEATURE_VPCLMULQDQ,		X86_FEATURE_AVX512VL  },
 	{ X86_FEATURE_AVX512_VNNI,		X86_FEATURE_AVX512VL  },
 	{ X86_FEATURE_AVX512_BITALG,		X86_FEATURE_AVX512VL  },
 	{ X86_FEATURE_AVX512_4VNNIW,		X86_FEATURE_AVX512F   },
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index d65f91d0a8bc..77458137cab1 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -362,6 +362,7 @@ static void kvm_vcpu_after_set_cpuid(struct kvm_vcpu *vcpu)
 
 	kvm_update_pv_runtime(vcpu);
 
+	vcpu->arch.is_amd_compatible = guest_cpuid_is_amd_or_hygon(vcpu);
 	vcpu->arch.maxphyaddr = cpuid_query_maxphyaddr(vcpu);
 	vcpu->arch.reserved_gpa_bits = kvm_vcpu_reserved_gpa_bits_raw(vcpu);
 
diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h
index 284fa4704553..57ee789ada14 100644
--- a/arch/x86/kvm/cpuid.h
+++ b/arch/x86/kvm/cpuid.h
@@ -125,6 +125,16 @@ static inline bool guest_cpuid_is_intel(struct kvm_vcpu *vcpu)
 	return best && is_guest_vendor_intel(best->ebx, best->ecx, best->edx);
 }
 
+static inline bool guest_cpuid_is_amd_compatible(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.is_amd_compatible;
+}
+
+static inline bool guest_cpuid_is_intel_compatible(struct kvm_vcpu *vcpu)
+{
+	return !guest_cpuid_is_amd_compatible(vcpu);
+}
+
 static inline int guest_cpuid_family(struct kvm_vcpu *vcpu)
 {
 	struct kvm_cpuid_entry2 *best;
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 23fab75993a5..66a2c4c0ae10 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -2772,7 +2772,8 @@ int kvm_apic_local_deliver(struct kvm_lapic *apic, int lvt_type)
 		trig_mode = reg & APIC_LVT_LEVEL_TRIGGER;
 
 		r = __apic_accept_irq(apic, mode, vector, 1, trig_mode, NULL);
-		if (r && lvt_type == APIC_LVTPC)
+		if (r && lvt_type == APIC_LVTPC &&
+		    guest_cpuid_is_intel_compatible(apic->vcpu))
 			kvm_lapic_set_reg(apic, APIC_LVTPC, reg | APIC_LVT_MASKED);
 		return r;
 	}
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index 11c484d72eab..294775b7383b 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -4788,7 +4788,7 @@ static void reset_guest_rsvds_bits_mask(struct kvm_vcpu *vcpu,
 				context->cpu_role.base.level, is_efer_nx(context),
 				guest_can_use(vcpu, X86_FEATURE_GBPAGES),
 				is_cr4_pse(context),
-				guest_cpuid_is_amd_or_hygon(vcpu));
+				guest_cpuid_is_amd_compatible(vcpu));
 }
 
 static void __reset_rsvds_bits_mask_ept(struct rsvd_bits_validate *rsvd_check,
diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
index 6cd4dd631a2f..8eef3ed5fe04 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.c
+++ b/arch/x86/kvm/mmu/tdp_mmu.c
@@ -1506,6 +1506,16 @@ void kvm_tdp_mmu_try_split_huge_pages(struct kvm *kvm,
 	}
 }
 
+static bool tdp_mmu_need_write_protect(struct kvm_mmu_page *sp)
+{
+	/*
+	 * All TDP MMU shadow pages share the same role as their root, aside
+	 * from level, so it is valid to key off any shadow page to determine if
+	 * write protection is needed for an entire tree.
+	 */
+	return kvm_mmu_page_ad_need_write_protect(sp) || !kvm_ad_enabled();
+}
+
 /*
  * Clear the dirty status of all the SPTEs mapping GFNs in the memslot. If
  * AD bits are enabled, this will involve clearing the dirty bit on each SPTE.
@@ -1516,7 +1526,8 @@ void kvm_tdp_mmu_try_split_huge_pages(struct kvm *kvm,
 static bool clear_dirty_gfn_range(struct kvm *kvm, struct kvm_mmu_page *root,
 			   gfn_t start, gfn_t end)
 {
-	u64 dbit = kvm_ad_enabled() ? shadow_dirty_mask : PT_WRITABLE_MASK;
+	const u64 dbit = tdp_mmu_need_write_protect(root) ? PT_WRITABLE_MASK :
+							    shadow_dirty_mask;
 	struct tdp_iter iter;
 	bool spte_set = false;
 
@@ -1530,7 +1541,7 @@ retry:
 		if (!is_shadow_present_pte(iter.old_spte))
 			continue;
 
-		KVM_MMU_WARN_ON(kvm_ad_enabled() &&
+		KVM_MMU_WARN_ON(dbit == shadow_dirty_mask &&
 				spte_ad_need_write_protect(iter.old_spte));
 
 		if (!(iter.old_spte & dbit))
@@ -1578,8 +1589,8 @@ bool kvm_tdp_mmu_clear_dirty_slot(struct kvm *kvm,
 static void clear_dirty_pt_masked(struct kvm *kvm, struct kvm_mmu_page *root,
 				  gfn_t gfn, unsigned long mask, bool wrprot)
 {
-	u64 dbit = (wrprot || !kvm_ad_enabled()) ? PT_WRITABLE_MASK :
-						   shadow_dirty_mask;
+	const u64 dbit = (wrprot || tdp_mmu_need_write_protect(root)) ? PT_WRITABLE_MASK :
+									shadow_dirty_mask;
 	struct tdp_iter iter;
 
 	lockdep_assert_held_write(&kvm->mmu_lock);
@@ -1591,7 +1602,7 @@ static void clear_dirty_pt_masked(struct kvm *kvm, struct kvm_mmu_page *root,
 		if (!mask)
 			break;
 
-		KVM_MMU_WARN_ON(kvm_ad_enabled() &&
+		KVM_MMU_WARN_ON(dbit == shadow_dirty_mask &&
 				spte_ad_need_write_protect(iter.old_spte));
 
 		if (iter.level > PG_LEVEL_4K ||
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index b2ed051611b0..dae499e2da84 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -7858,8 +7858,28 @@ static u64 vmx_get_perf_capabilities(void)
 
 	if (vmx_pebs_supported()) {
 		perf_cap |= host_perf_cap & PERF_CAP_PEBS_MASK;
-		if ((perf_cap & PERF_CAP_PEBS_FORMAT) < 4)
-			perf_cap &= ~PERF_CAP_PEBS_BASELINE;
+
+		/*
+		 * Disallow adaptive PEBS as it is functionally broken, can be
+		 * used by the guest to read *host* LBRs, and can be used to
+		 * bypass userspace event filters.  To correctly and safely
+		 * support adaptive PEBS, KVM needs to:
+		 *
+		 * 1. Account for the ADAPTIVE flag when (re)programming fixed
+		 *    counters.
+		 *
+		 * 2. Gain support from perf (or take direct control of counter
+		 *    programming) to support events without adaptive PEBS
+		 *    enabled for the hardware counter.
+		 *
+		 * 3. Ensure LBR MSRs cannot hold host data on VM-Entry with
+		 *    adaptive PEBS enabled and MSR_PEBS_DATA_CFG.LBRS=1.
+		 *
+		 * 4. Document which PMU events are effectively exposed to the
+		 *    guest via adaptive PEBS, and make adaptive PEBS mutually
+		 *    exclusive with KVM_SET_PMU_EVENT_FILTER if necessary.
+		 */
+		perf_cap &= ~PERF_CAP_PEBS_BASELINE;
 	}
 
 	return perf_cap;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 4ed8a7dc0536..cab7680f0d68 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3317,7 +3317,7 @@ static bool is_mci_status_msr(u32 msr)
 static bool can_set_mci_status(struct kvm_vcpu *vcpu)
 {
 	/* McStatusWrEn enabled? */
-	if (guest_cpuid_is_amd_or_hygon(vcpu))
+	if (guest_cpuid_is_amd_compatible(vcpu))
 		return !!(vcpu->arch.msr_hwcr & BIT_ULL(18));
 
 	return false;