Merge branch kvm-arm64/misc-6.14 into kvmarm-master/next

Marc Zyngier · Marc Zyngier · commit fa5e4043e9b1 · 2025-01-17T11:06:50.000Z
* kvm-arm64/misc-6.14:
  : .
  : Misc KVM/arm64 changes for 6.14
  :
  : - Don't expose AArch32 EL0 capability when NV is enabled
  :
  : - Update documentation to reflect the full gamut of kvm-arm.mode
  :   behaviours
  :
  : - Use the hypervisor VA bit width when dumping stacktraces
  :
  : - Decouple the hypervisor stack size from PAGE_SIZE, at least
  :   on the surface...
  :
  : - Make use of str_enabled_disabled() when advertising GICv4.1 support
  :
  : - Explicitly handle BRBE traps as UNDEFINED
  : .
  KVM: arm64: Explicitly handle BRBE traps as UNDEFINED
  KVM: arm64: vgic: Use str_enabled_disabled() in vgic_v3_probe()
  arm64: kvm: Introduce nvhe stack size constants
  KVM: arm64: Fix nVHE stacktrace VA bits mask
  Documentation: Update the behaviour of "kvm-arm.mode"
  KVM: arm64: nv: Advertise the lack of AArch32 EL0 support

Signed-off-by: Marc Zyngier &lt;maz@kernel.org&gt;
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
@@ -2748,17 +2748,21 @@
 			nvhe: Standard nVHE-based mode, without support for
 			      protected guests.
 
-			protected: nVHE-based mode with support for guests whose
-				   state is kept private from the host.
+			protected: Mode with support for guests whose state is
+				   kept private from the host, using VHE or
+				   nVHE depending on HW support.
 
 			nested: VHE-based mode with support for nested
-				virtualization. Requires at least ARMv8.3
-				hardware.
+				virtualization. Requires at least ARMv8.4
+				hardware (with FEAT_NV2).
 
 			Defaults to VHE/nVHE based on hardware support. Setting
 			mode to "protected" will disable kexec and hibernation
-			for the host. "nested" is experimental and should be
-			used with extreme caution.
+			for the host. To force nVHE on VHE hardware, add
+			"arm64_sw.hvhe=0 id_aa64mmfr1.vh=0" to the
+			command-line.
+			"nested" is experimental and should be used with
+			extreme caution.
 
 	kvm-arm.vgic_v3_group0_trap=
 			[KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
@@ -139,6 +139,8 @@ static __always_inline unsigned long __kern_hyp_va(unsigned long v)
 
 #define kern_hyp_va(v) 	((typeof(v))(__kern_hyp_va((unsigned long)(v))))
 
+extern u32 __hyp_va_bits;
+
 /*
  * We currently support using a VM-specified IPA size. For backward
  * compatibility, the default IPA size is fixed to 40bits.
diff --git a/arch/arm64/include/asm/memory.h b/arch/arm64/include/asm/memory.h
@@ -145,13 +145,16 @@
 
 #define OVERFLOW_STACK_SIZE	SZ_4K
 
+#define NVHE_STACK_SHIFT       PAGE_SHIFT
+#define NVHE_STACK_SIZE        (UL(1) << NVHE_STACK_SHIFT)
+
 /*
  * With the minimum frame size of [x29, x30], exactly half the combined
  * sizes of the hyp and overflow stacks is the maximum size needed to
  * save the unwinded stacktrace; plus an additional entry to delimit the
  * end.
  */
-#define NVHE_STACKTRACE_SIZE	((OVERFLOW_STACK_SIZE + PAGE_SIZE) / 2 + sizeof(long))
+#define NVHE_STACKTRACE_SIZE	((OVERFLOW_STACK_SIZE + NVHE_STACK_SIZE) / 2 + sizeof(long))
 
 /*
  * Alignment of kernel segments (e.g. .text, .data).
diff --git a/arch/arm64/include/asm/stacktrace/nvhe.h b/arch/arm64/include/asm/stacktrace/nvhe.h
@@ -47,7 +47,7 @@ static inline void kvm_nvhe_unwind_init(struct unwind_state *state,
 
 DECLARE_KVM_NVHE_PER_CPU(unsigned long [OVERFLOW_STACK_SIZE/sizeof(long)], overflow_stack);
 DECLARE_KVM_NVHE_PER_CPU(struct kvm_nvhe_stacktrace_info, kvm_stacktrace_info);
-DECLARE_PER_CPU(unsigned long, kvm_arm_hyp_stack_page);
+DECLARE_PER_CPU(unsigned long, kvm_arm_hyp_stack_base);
 
 void kvm_nvhe_dump_backtrace(unsigned long hyp_offset);
 
diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
@@ -61,7 +61,7 @@ static enum kvm_wfx_trap_policy kvm_wfe_trap_policy __read_mostly = KVM_WFX_NOTR
 
 DECLARE_KVM_HYP_PER_CPU(unsigned long, kvm_hyp_vector);
 
-DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_page);
+DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_base);
 DECLARE_KVM_NVHE_PER_CPU(struct kvm_nvhe_init_params, kvm_init_params);
 
 DECLARE_KVM_NVHE_PER_CPU(struct kvm_cpu_context, kvm_hyp_ctxt);
@@ -2329,7 +2329,7 @@ static void __init teardown_hyp_mode(void)
 
 	free_hyp_pgds();
 	for_each_possible_cpu(cpu) {
-		free_page(per_cpu(kvm_arm_hyp_stack_page, cpu));
+		free_pages(per_cpu(kvm_arm_hyp_stack_base, cpu), NVHE_STACK_SHIFT - PAGE_SHIFT);
 		free_pages(kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu], nvhe_percpu_order());
 
 		if (free_sve) {
@@ -2517,15 +2517,15 @@ static int __init init_hyp_mode(void)
 	 * Allocate stack pages for Hypervisor-mode
 	 */
 	for_each_possible_cpu(cpu) {
-		unsigned long stack_page;
+		unsigned long stack_base;
 
-		stack_page = __get_free_page(GFP_KERNEL);
-		if (!stack_page) {
+		stack_base = __get_free_pages(GFP_KERNEL, NVHE_STACK_SHIFT - PAGE_SHIFT);
+		if (!stack_base) {
 			err = -ENOMEM;
 			goto out_err;
 		}
 
-		per_cpu(kvm_arm_hyp_stack_page, cpu) = stack_page;
+		per_cpu(kvm_arm_hyp_stack_base, cpu) = stack_base;
 	}
 
 	/*
@@ -2594,9 +2594,9 @@ static int __init init_hyp_mode(void)
 	 */
 	for_each_possible_cpu(cpu) {
 		struct kvm_nvhe_init_params *params = per_cpu_ptr_nvhe_sym(kvm_init_params, cpu);
-		char *stack_page = (char *)per_cpu(kvm_arm_hyp_stack_page, cpu);
+		char *stack_base = (char *)per_cpu(kvm_arm_hyp_stack_base, cpu);
 
-		err = create_hyp_stack(__pa(stack_page), &params->stack_hyp_va);
+		err = create_hyp_stack(__pa(stack_base), &params->stack_hyp_va);
 		if (err) {
 			kvm_err("Cannot map hyp stack\n");
 			goto out_err;
@@ -2608,7 +2608,7 @@ static int __init init_hyp_mode(void)
 		 * __hyp_pa() won't do the right thing there, since the stack
 		 * has been mapped in the flexible private VA space.
 		 */
-		params->stack_pa = __pa(stack_page);
+		params->stack_pa = __pa(stack_base);
 	}
 
 	for_each_possible_cpu(cpu) {
diff --git a/arch/arm64/kvm/hyp/nvhe/host.S b/arch/arm64/kvm/hyp/nvhe/host.S
@@ -188,12 +188,12 @@ SYM_FUNC_END(__host_hvc)
 
 	/*
 	 * Test whether the SP has overflowed, without corrupting a GPR.
-	 * nVHE hypervisor stacks are aligned so that the PAGE_SHIFT bit
+	 * nVHE hypervisor stacks are aligned so that the NVHE_STACK_SHIFT bit
 	 * of SP should always be 1.
 	 */
 	add	sp, sp, x0			// sp' = sp + x0
 	sub	x0, sp, x0			// x0' = sp' - x0 = (sp + x0) - x0 = sp
-	tbz	x0, #PAGE_SHIFT, .L__hyp_sp_overflow\@
+	tbz	x0, #NVHE_STACK_SHIFT, .L__hyp_sp_overflow\@
 	sub	x0, sp, x0			// x0'' = sp' - x0' = (sp + x0) - sp = x0
 	sub	sp, sp, x0			// sp'' = sp' - x0 = (sp + x0) - x0 = sp
 
diff --git a/arch/arm64/kvm/hyp/nvhe/mm.c b/arch/arm64/kvm/hyp/nvhe/mm.c
@@ -360,10 +360,10 @@ int pkvm_create_stack(phys_addr_t phys, unsigned long *haddr)
 
 	prev_base = __io_map_base;
 	/*
-	 * Efficient stack verification using the PAGE_SHIFT bit implies
+	 * Efficient stack verification using the NVHE_STACK_SHIFT bit implies
 	 * an alignment of our allocation on the order of the size.
 	 */
-	size = PAGE_SIZE * 2;
+	size = NVHE_STACK_SIZE * 2;
 	addr = ALIGN(__io_map_base, size);
 
 	ret = __pkvm_alloc_private_va_range(addr, size);
@@ -373,12 +373,12 @@ int pkvm_create_stack(phys_addr_t phys, unsigned long *haddr)
 		 * at the higher address and leave the lower guard page
 		 * unbacked.
 		 *
-		 * Any valid stack address now has the PAGE_SHIFT bit as 1
+		 * Any valid stack address now has the NVHE_STACK_SHIFT bit as 1
 		 * and addresses corresponding to the guard page have the
-		 * PAGE_SHIFT bit as 0 - this is used for overflow detection.
+		 * NVHE_STACK_SHIFT bit as 0 - this is used for overflow detection.
 		 */
-		ret = kvm_pgtable_hyp_map(&pkvm_pgtable, addr + PAGE_SIZE,
-					  PAGE_SIZE, phys, PAGE_HYP);
+		ret = kvm_pgtable_hyp_map(&pkvm_pgtable, addr + NVHE_STACK_SIZE,
+					  NVHE_STACK_SIZE, phys, PAGE_HYP);
 		if (ret)
 			__io_map_base = prev_base;
 	}
diff --git a/arch/arm64/kvm/hyp/nvhe/stacktrace.c b/arch/arm64/kvm/hyp/nvhe/stacktrace.c
@@ -28,7 +28,7 @@ static void hyp_prepare_backtrace(unsigned long fp, unsigned long pc)
 	struct kvm_nvhe_stacktrace_info *stacktrace_info = this_cpu_ptr(&kvm_stacktrace_info);
 	struct kvm_nvhe_init_params *params = this_cpu_ptr(&kvm_init_params);
 
-	stacktrace_info->stack_base = (unsigned long)(params->stack_hyp_va - PAGE_SIZE);
+	stacktrace_info->stack_base = (unsigned long)(params->stack_hyp_va - NVHE_STACK_SIZE);
 	stacktrace_info->overflow_stack_base = (unsigned long)this_cpu_ptr(overflow_stack);
 	stacktrace_info->fp = fp;
 	stacktrace_info->pc = pc;
@@ -54,7 +54,7 @@ static struct stack_info stackinfo_get_hyp(void)
 {
 	struct kvm_nvhe_init_params *params = this_cpu_ptr(&kvm_init_params);
 	unsigned long high = params->stack_hyp_va;
-	unsigned long low = high - PAGE_SIZE;
+	unsigned long low = high - NVHE_STACK_SIZE;
 
 	return (struct stack_info) {
 		.low = low,
diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c
@@ -30,6 +30,8 @@ static unsigned long __ro_after_init hyp_idmap_start;
 static unsigned long __ro_after_init hyp_idmap_end;
 static phys_addr_t __ro_after_init hyp_idmap_vector;
 
+u32 __ro_after_init __hyp_va_bits;
+
 static unsigned long __ro_after_init io_map_base;
 
 #define KVM_PGT_FN(fn)		(!is_protected_kvm_enabled() ? fn : p ## fn)
@@ -715,10 +717,10 @@ int create_hyp_stack(phys_addr_t phys_addr, unsigned long *haddr)
 
 	mutex_lock(&kvm_hyp_pgd_mutex);
 	/*
-	 * Efficient stack verification using the PAGE_SHIFT bit implies
+	 * Efficient stack verification using the NVHE_STACK_SHIFT bit implies
 	 * an alignment of our allocation on the order of the size.
 	 */
-	size = PAGE_SIZE * 2;
+	size = NVHE_STACK_SIZE * 2;
 	base = ALIGN_DOWN(io_map_base - size, size);
 
 	ret = __hyp_alloc_private_va_range(base);
@@ -735,12 +737,12 @@ int create_hyp_stack(phys_addr_t phys_addr, unsigned long *haddr)
 	 * at the higher address and leave the lower guard page
 	 * unbacked.
 	 *
-	 * Any valid stack address now has the PAGE_SHIFT bit as 1
+	 * Any valid stack address now has the NVHE_STACK_SHIFT bit as 1
 	 * and addresses corresponding to the guard page have the
-	 * PAGE_SHIFT bit as 0 - this is used for overflow detection.
+	 * NVHE_STACK_SHIFT bit as 0 - this is used for overflow detection.
 	 */
-	ret = __create_hyp_mappings(base + PAGE_SIZE, PAGE_SIZE, phys_addr,
-				    PAGE_HYP);
+	ret = __create_hyp_mappings(base + NVHE_STACK_SIZE, NVHE_STACK_SIZE,
+				    phys_addr, PAGE_HYP);
 	if (ret)
 		kvm_err("Cannot map hyp stack\n");
 
@@ -2085,6 +2087,7 @@ int __init kvm_mmu_init(u32 *hyp_va_bits)
 		goto out_destroy_pgtable;
 
 	io_map_base = hyp_idmap_start;
+	__hyp_va_bits = *hyp_va_bits;
 	return 0;
 
 out_destroy_pgtable:
diff --git a/arch/arm64/kvm/nested.c b/arch/arm64/kvm/nested.c
@@ -830,8 +830,10 @@ static void limit_nv_id_regs(struct kvm *kvm)
 		 NV_FTR(PFR0, RAS)	|
 		 NV_FTR(PFR0, EL3)	|
 		 NV_FTR(PFR0, EL2)	|
-		 NV_FTR(PFR0, EL1));
-	/* 64bit EL1/EL2/EL3 only */
+		 NV_FTR(PFR0, EL1)	|
+		 NV_FTR(PFR0, EL0));
+	/* 64bit only at any EL */
+	val |= FIELD_PREP(NV_FTR(PFR0, EL0), 0b0001);
 	val |= FIELD_PREP(NV_FTR(PFR0, EL1), 0b0001);
 	val |= FIELD_PREP(NV_FTR(PFR0, EL2), 0b0001);
 	val |= FIELD_PREP(NV_FTR(PFR0, EL3), 0b0001);
diff --git a/arch/arm64/kvm/stacktrace.c b/arch/arm64/kvm/stacktrace.c
@@ -19,6 +19,7 @@
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
 
+#include <asm/kvm_mmu.h>
 #include <asm/stacktrace/nvhe.h>
 
 static struct stack_info stackinfo_get_overflow(void)
@@ -50,7 +51,7 @@ static struct stack_info stackinfo_get_hyp(void)
 	struct kvm_nvhe_stacktrace_info *stacktrace_info
 				= this_cpu_ptr_nvhe_sym(kvm_stacktrace_info);
 	unsigned long low = (unsigned long)stacktrace_info->stack_base;
-	unsigned long high = low + PAGE_SIZE;
+	unsigned long high = low + NVHE_STACK_SIZE;
 
 	return (struct stack_info) {
 		.low = low,
@@ -60,8 +61,8 @@ static struct stack_info stackinfo_get_hyp(void)
 
 static struct stack_info stackinfo_get_hyp_kern_va(void)
 {
-	unsigned long low = (unsigned long)*this_cpu_ptr(&kvm_arm_hyp_stack_page);
-	unsigned long high = low + PAGE_SIZE;
+	unsigned long low = (unsigned long)*this_cpu_ptr(&kvm_arm_hyp_stack_base);
+	unsigned long high = low + NVHE_STACK_SIZE;
 
 	return (struct stack_info) {
 		.low = low,
@@ -145,7 +146,7 @@ static void unwind(struct unwind_state *state,
  */
 static bool kvm_nvhe_dump_backtrace_entry(void *arg, unsigned long where)
 {
-	unsigned long va_mask = GENMASK_ULL(vabits_actual - 1, 0);
+	unsigned long va_mask = GENMASK_ULL(__hyp_va_bits - 1, 0);
 	unsigned long hyp_offset = (unsigned long)arg;
 
 	/* Mask tags and convert to kern addr */
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
@@ -1814,6 +1814,9 @@ static u64 sanitise_id_aa64dfr0_el1(const struct kvm_vcpu *vcpu, u64 val)
 	/* Hide SPE from guests */
 	val &= ~ID_AA64DFR0_EL1_PMSVer_MASK;
 
+	/* Hide BRBE from guests */
+	val &= ~ID_AA64DFR0_EL1_BRBE_MASK;
+
 	return val;
 }
 
@@ -5029,6 +5032,14 @@ void kvm_calculate_traps(struct kvm_vcpu *vcpu)
 		kvm->arch.fgu[HAFGRTR_GROUP] |= ~(HAFGRTR_EL2_RES0 |
 						  HAFGRTR_EL2_RES1);
 
+	if (!kvm_has_feat(kvm, ID_AA64DFR0_EL1, BRBE, IMP)) {
+		kvm->arch.fgu[HDFGRTR_GROUP] |= (HDFGRTR_EL2_nBRBDATA  |
+						 HDFGRTR_EL2_nBRBCTL   |
+						 HDFGRTR_EL2_nBRBIDR);
+		kvm->arch.fgu[HFGITR_GROUP] |= (HFGITR_EL2_nBRBINJ |
+						HFGITR_EL2_nBRBIALL);
+	}
+
 	set_bit(KVM_ARCH_FLAG_FGU_INITIALIZED, &kvm->arch.flags);
 out:
 	mutex_unlock(&kvm->arch.config_lock);
diff --git a/arch/arm64/kvm/vgic/vgic-v3.c b/arch/arm64/kvm/vgic/vgic-v3.c
@@ -6,6 +6,7 @@
 #include <linux/kstrtox.h>
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
+#include <linux/string_choices.h>
 #include <kvm/arm_vgic.h>
 #include <asm/kvm_hyp.h>
 #include <asm/kvm_mmu.h>
@@ -663,9 +664,9 @@ int vgic_v3_probe(const struct gic_kvm_info *info)
 	if (info->has_v4) {
 		kvm_vgic_global_state.has_gicv4 = gicv4_enable;
 		kvm_vgic_global_state.has_gicv4_1 = info->has_v4_1 && gicv4_enable;
-		kvm_info("GICv4%s support %sabled\n",
+		kvm_info("GICv4%s support %s\n",
 			 kvm_vgic_global_state.has_gicv4_1 ? ".1" : "",
-			 gicv4_enable ? "en" : "dis");
+			 str_enabled_disabled(gicv4_enable));
 	}
 
 	kvm_vgic_global_state.vcpu_base = 0;
