Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull KVM updates from Paolo Bonzini:
 "ARM:

   - More progress on the protected VM front, now with the full fixed
     feature set as well as the limitation of some hypercalls after
     initialisation.

   - Cleanup of the RAZ/WI sysreg handling, which was pointlessly
     complicated

   - Fixes for the vgic placement in the IPA space, together with a
     bunch of selftests

   - More memcg accounting of the memory allocated on behalf of a guest

   - Timer and vgic selftests

   - Workarounds for the Apple M1 broken vgic implementation

   - KConfig cleanups

   - New kvmarm.mode=none option, for those who really dislike us

  RISC-V:

   - New KVM port.

  x86:

   - New API to control TSC offset from userspace

   - TSC scaling for nested hypervisors on SVM

   - Switch masterclock protection from raw_spin_lock to seqcount

   - Clean up function prototypes in the page fault code and avoid
     repeated memslot lookups

   - Convey the exit reason to userspace on emulation failure

   - Configure time between NX page recovery iterations

   - Expose Predictive Store Forwarding Disable CPUID leaf

   - Allocate page tracking data structures lazily (if the i915 KVM-GT
     functionality is not compiled in)

   - Cleanups, fixes and optimizations for the shadow MMU code

  s390:

   - SIGP Fixes

   - initial preparations for lazy destroy of secure VMs

   - storage key improvements/fixes

   - Log the guest CPNC

  Starting from this release, KVM-PPC patches will come from Michael
  Ellerman's PPC tree"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (227 commits)
  RISC-V: KVM: fix boolreturn.cocci warnings
  RISC-V: KVM: remove unneeded semicolon
  RISC-V: KVM: Fix GPA passed to __kvm_riscv_hfence_gvma_xyz() functions
  RISC-V: KVM: Factor-out FP virtualization into separate sources
  KVM: s390: add debug statement for diag 318 CPNC data
  KVM: s390: pv: properly handle page flags for protected guests
  KVM: s390: Fix handle_sske page fault handling
  KVM: x86: SGX must obey the KVM_INTERNAL_ERROR_EMULATION protocol
  KVM: x86: On emulation failure, convey the exit reason, etc. to userspace
  KVM: x86: Get exit_reason as part of kvm_x86_ops.get_exit_info
  KVM: x86: Clarify the kvm_run.emulation_failure structure layout
  KVM: s390: Add a routine for setting userspace CPU state
  KVM: s390: Simplify SIGP Set Arch handling
  KVM: s390: pv: avoid stalls when making pages secure
  KVM: s390: pv: avoid stalls for kvm_s390_pv_init_vm
  KVM: s390: pv: avoid double free of sida page
  KVM: s390: pv: add macros for UVC CC values
  s390/mm: optimize reset_guest_reference_bit()
  s390/mm: optimize set_guest_storage_key()
  s390/mm: no need for pte_alloc_map_lock() if we know the pmd is present
  ...

1 files changed, 72 insertions, 30 deletions

diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index fe102cd2e518..f5490afe1ebf 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -291,18 +291,12 @@ long kvm_arch_dev_ioctl(struct file *filp,
 
 struct kvm *kvm_arch_alloc_vm(void)
 {
-	if (!has_vhe())
-		return kzalloc(sizeof(struct kvm), GFP_KERNEL);
-
-	return vzalloc(sizeof(struct kvm));
-}
+	size_t sz = sizeof(struct kvm);
 
-void kvm_arch_free_vm(struct kvm *kvm)
-{
 	if (!has_vhe())
-		kfree(kvm);
-	else
-		vfree(kvm);
+		return kzalloc(sz, GFP_KERNEL_ACCOUNT);
+
+	return __vmalloc(sz, GFP_KERNEL_ACCOUNT | __GFP_HIGHMEM | __GFP_ZERO);
 }
 
 int kvm_arch_vcpu_precreate(struct kvm *kvm, unsigned int id)
@@ -620,6 +614,14 @@ static int kvm_vcpu_first_run_init(struct kvm_vcpu *vcpu)
 
 	ret = kvm_arm_pmu_v3_enable(vcpu);
 
+	/*
+	 * Initialize traps for protected VMs.
+	 * NOTE: Move to run in EL2 directly, rather than via a hypercall, once
+	 * the code is in place for first run initialization at EL2.
+	 */
+	if (kvm_vm_is_protected(kvm))
+		kvm_call_hyp_nvhe(__pkvm_vcpu_init_traps, vcpu);
+
 	return ret;
 }
 
@@ -1579,25 +1581,33 @@ static void cpu_set_hyp_vector(void)
 		kvm_call_hyp_nvhe(__pkvm_cpu_set_vector, data->slot);
 }
 
-static void cpu_hyp_reinit(void)
+static void cpu_hyp_init_context(void)
 {
 	kvm_init_host_cpu_context(&this_cpu_ptr_hyp_sym(kvm_host_data)->host_ctxt);
 
-	cpu_hyp_reset();
-
-	if (is_kernel_in_hyp_mode())
-		kvm_timer_init_vhe();
-	else
+	if (!is_kernel_in_hyp_mode())
 		cpu_init_hyp_mode();
+}
 
+static void cpu_hyp_init_features(void)
+{
 	cpu_set_hyp_vector();
-
 	kvm_arm_init_debug();
 
+	if (is_kernel_in_hyp_mode())
+		kvm_timer_init_vhe();
+
 	if (vgic_present)
 		kvm_vgic_init_cpu_hardware();
 }
 
+static void cpu_hyp_reinit(void)
+{
+	cpu_hyp_reset();
+	cpu_hyp_init_context();
+	cpu_hyp_init_features();
+}
+
 static void _kvm_arch_hardware_enable(void *discard)
 {
 	if (!__this_cpu_read(kvm_arm_hardware_enabled)) {
@@ -1788,10 +1798,17 @@ static int do_pkvm_init(u32 hyp_va_bits)
 	int ret;
 
 	preempt_disable();
-	hyp_install_host_vector();
+	cpu_hyp_init_context();
 	ret = kvm_call_hyp_nvhe(__pkvm_init, hyp_mem_base, hyp_mem_size,
 				num_possible_cpus(), kern_hyp_va(per_cpu_base),
 				hyp_va_bits);
+	cpu_hyp_init_features();
+
+	/*
+	 * The stub hypercalls are now disabled, so set our local flag to
+	 * prevent a later re-init attempt in kvm_arch_hardware_enable().
+	 */
+	__this_cpu_write(kvm_arm_hardware_enabled, 1);
 	preempt_enable();
 
 	return ret;
@@ -1802,8 +1819,13 @@ static int kvm_hyp_init_protection(u32 hyp_va_bits)
 	void *addr = phys_to_virt(hyp_mem_base);
 	int ret;
 
+	kvm_nvhe_sym(id_aa64pfr0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64PFR0_EL1);
+	kvm_nvhe_sym(id_aa64pfr1_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64PFR1_EL1);
+	kvm_nvhe_sym(id_aa64isar0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64ISAR0_EL1);
+	kvm_nvhe_sym(id_aa64isar1_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64ISAR1_EL1);
 	kvm_nvhe_sym(id_aa64mmfr0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64MMFR0_EL1);
 	kvm_nvhe_sym(id_aa64mmfr1_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64MMFR1_EL1);
+	kvm_nvhe_sym(id_aa64mmfr2_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64MMFR2_EL1);
 
 	ret = create_hyp_mappings(addr, addr + hyp_mem_size, PAGE_HYP);
 	if (ret)
@@ -1971,9 +1993,25 @@ out_err:
 	return err;
 }
 
-static void _kvm_host_prot_finalize(void *discard)
+static void _kvm_host_prot_finalize(void *arg)
 {
-	WARN_ON(kvm_call_hyp_nvhe(__pkvm_prot_finalize));
+	int *err = arg;
+
+	if (WARN_ON(kvm_call_hyp_nvhe(__pkvm_prot_finalize)))
+		WRITE_ONCE(*err, -EINVAL);
+}
+
+static int pkvm_drop_host_privileges(void)
+{
+	int ret = 0;
+
+	/*
+	 * Flip the static key upfront as that may no longer be possible
+	 * once the host stage 2 is installed.
+	 */
+	static_branch_enable(&kvm_protected_mode_initialized);
+	on_each_cpu(_kvm_host_prot_finalize, &ret, 1);
+	return ret;
 }
 
 static int finalize_hyp_mode(void)
@@ -1987,15 +2025,7 @@ static int finalize_hyp_mode(void)
 	 * None of other sections should ever be introspected.
 	 */
 	kmemleak_free_part(__hyp_bss_start, __hyp_bss_end - __hyp_bss_start);
-
-	/*
-	 * Flip the static key upfront as that may no longer be possible
-	 * once the host stage 2 is installed.
-	 */
-	static_branch_enable(&kvm_protected_mode_initialized);
-	on_each_cpu(_kvm_host_prot_finalize, NULL, 1);
-
-	return 0;
+	return pkvm_drop_host_privileges();
 }
 
 struct kvm_vcpu *kvm_mpidr_to_vcpu(struct kvm *kvm, unsigned long mpidr)
@@ -2064,6 +2094,11 @@ int kvm_arch_init(void *opaque)
 		return -ENODEV;
 	}
 
+	if (kvm_get_mode() == KVM_MODE_NONE) {
+		kvm_info("KVM disabled from command line\n");
+		return -ENODEV;
+	}
+
 	in_hyp_mode = is_kernel_in_hyp_mode();
 
 	if (cpus_have_final_cap(ARM64_WORKAROUND_DEVICE_LOAD_ACQUIRE) ||
@@ -2137,8 +2172,15 @@ static int __init early_kvm_mode_cfg(char *arg)
 		return 0;
 	}
 
-	if (strcmp(arg, "nvhe") == 0 && !WARN_ON(is_kernel_in_hyp_mode()))
+	if (strcmp(arg, "nvhe") == 0 && !WARN_ON(is_kernel_in_hyp_mode())) {
+		kvm_mode = KVM_MODE_DEFAULT;
 		return 0;
+	}
+
+	if (strcmp(arg, "none") == 0) {
+		kvm_mode = KVM_MODE_NONE;
+		return 0;
+	}
 
 	return -EINVAL;
 }