diff options
Diffstat (limited to 'arch/arm/kvm')
| -rw-r--r-- | arch/arm/kvm/arm.c | 281 | ||||
| -rw-r--r-- | arch/arm/kvm/emulate.c | 74 | ||||
| -rw-r--r-- | arch/arm/kvm/mmu.c | 18 |
3 files changed, 256 insertions, 117 deletions
diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index 66fde745238f..5475e684a6b4 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -16,7 +16,6 @@ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ -#include <linux/cpu.h> #include <linux/cpu_pm.h> #include <linux/errno.h> #include <linux/err.h> @@ -44,6 +43,7 @@ #include <asm/kvm_emulate.h> #include <asm/kvm_coproc.h> #include <asm/kvm_psci.h> +#include <asm/sections.h> #ifdef REQUIRES_VIRT __asm__(".arch_extension virt"); @@ -58,9 +58,14 @@ static DEFINE_PER_CPU(struct kvm_vcpu *, kvm_arm_running_vcpu); /* The VMID used in the VTTBR */ static atomic64_t kvm_vmid_gen = ATOMIC64_INIT(1); -static u8 kvm_next_vmid; +static u32 kvm_next_vmid; +static unsigned int kvm_vmid_bits __read_mostly; static DEFINE_SPINLOCK(kvm_vmid_lock); +static bool vgic_present; + +static DEFINE_PER_CPU(unsigned char, kvm_arm_hardware_enabled); + static void kvm_arm_set_running_vcpu(struct kvm_vcpu *vcpu) { BUG_ON(preemptible()); @@ -85,11 +90,6 @@ struct kvm_vcpu * __percpu *kvm_get_running_vcpus(void) return &kvm_arm_running_vcpu; } -int kvm_arch_hardware_enable(void) -{ - return 0; -} - int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu) { return kvm_vcpu_exiting_guest_mode(vcpu) == IN_GUEST_MODE; @@ -132,7 +132,8 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) kvm->arch.vmid_gen = 0; /* The maximum number of VCPUs is limited by the host's GIC model */ - kvm->arch.max_vcpus = kvm_vgic_get_max_vcpus(); + kvm->arch.max_vcpus = vgic_present ? + kvm_vgic_get_max_vcpus() : KVM_MAX_VCPUS; return ret; out_free_stage2_pgd: @@ -170,6 +171,8 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) int r; switch (ext) { case KVM_CAP_IRQCHIP: + r = vgic_present; + break; case KVM_CAP_IOEVENTFD: case KVM_CAP_DEVICE_CTRL: case KVM_CAP_USER_MEMORY: @@ -431,11 +434,12 @@ static void update_vttbr(struct kvm *kvm) kvm->arch.vmid_gen = atomic64_read(&kvm_vmid_gen); kvm->arch.vmid = kvm_next_vmid; kvm_next_vmid++; + kvm_next_vmid &= (1 << kvm_vmid_bits) - 1; /* update vttbr to be used with the new vmid */ pgd_phys = virt_to_phys(kvm_get_hwpgd(kvm)); BUG_ON(pgd_phys & ~VTTBR_BADDR_MASK); - vmid = ((u64)(kvm->arch.vmid) << VTTBR_VMID_SHIFT) & VTTBR_VMID_MASK; + vmid = ((u64)(kvm->arch.vmid) << VTTBR_VMID_SHIFT) & VTTBR_VMID_MASK(kvm_vmid_bits); kvm->arch.vttbr = pgd_phys | vmid; spin_unlock(&kvm_vmid_lock); @@ -916,6 +920,8 @@ static int kvm_vm_ioctl_set_device_addr(struct kvm *kvm, switch (dev_id) { case KVM_ARM_DEVICE_VGIC_V2: + if (!vgic_present) + return -ENXIO; return kvm_vgic_addr(kvm, type, &dev_addr->addr, true); default: return -ENODEV; @@ -930,6 +936,8 @@ long kvm_arch_vm_ioctl(struct file *filp, switch (ioctl) { case KVM_CREATE_IRQCHIP: { + if (!vgic_present) + return -ENXIO; return kvm_vgic_create(kvm, KVM_DEV_TYPE_ARM_VGIC_V2); } case KVM_ARM_SET_DEVICE_ADDR: { @@ -975,40 +983,96 @@ static void cpu_init_hyp_mode(void *dummy) vector_ptr = (unsigned long)kvm_ksym_ref(__kvm_hyp_vector); __cpu_init_hyp_mode(boot_pgd_ptr, pgd_ptr, hyp_stack_ptr, vector_ptr); + __cpu_init_stage2(); kvm_arm_init_debug(); } -static int hyp_init_cpu_notify(struct notifier_block *self, - unsigned long action, void *cpu) +static void cpu_hyp_reinit(void) { - switch (action) { - case CPU_STARTING: - case CPU_STARTING_FROZEN: + if (is_kernel_in_hyp_mode()) { + /* + * __cpu_init_stage2() is safe to call even if the PM + * event was cancelled before the CPU was reset. + */ + __cpu_init_stage2(); + } else { if (__hyp_get_vectors() == hyp_default_vectors) cpu_init_hyp_mode(NULL); - break; } +} + +static void cpu_hyp_reset(void) +{ + phys_addr_t boot_pgd_ptr; + phys_addr_t phys_idmap_start; + + if (!is_kernel_in_hyp_mode()) { + boot_pgd_ptr = kvm_mmu_get_boot_httbr(); + phys_idmap_start = kvm_get_idmap_start(); + + __cpu_reset_hyp_mode(boot_pgd_ptr, phys_idmap_start); + } +} - return NOTIFY_OK; +static void _kvm_arch_hardware_enable(void *discard) +{ + if (!__this_cpu_read(kvm_arm_hardware_enabled)) { + cpu_hyp_reinit(); + __this_cpu_write(kvm_arm_hardware_enabled, 1); + } } -static struct notifier_block hyp_init_cpu_nb = { - .notifier_call = hyp_init_cpu_notify, -}; +int kvm_arch_hardware_enable(void) +{ + _kvm_arch_hardware_enable(NULL); + return 0; +} + +static void _kvm_arch_hardware_disable(void *discard) +{ + if (__this_cpu_read(kvm_arm_hardware_enabled)) { + cpu_hyp_reset(); + __this_cpu_write(kvm_arm_hardware_enabled, 0); + } +} + +void kvm_arch_hardware_disable(void) +{ + _kvm_arch_hardware_disable(NULL); +} #ifdef CONFIG_CPU_PM static int hyp_init_cpu_pm_notifier(struct notifier_block *self, unsigned long cmd, void *v) { - if (cmd == CPU_PM_EXIT && - __hyp_get_vectors() == hyp_default_vectors) { - cpu_init_hyp_mode(NULL); + /* + * kvm_arm_hardware_enabled is left with its old value over + * PM_ENTER->PM_EXIT. It is used to indicate PM_EXIT should + * re-enable hyp. + */ + switch (cmd) { + case CPU_PM_ENTER: + if (__this_cpu_read(kvm_arm_hardware_enabled)) + /* + * don't update kvm_arm_hardware_enabled here + * so that the hardware will be re-enabled + * when we resume. See below. + */ + cpu_hyp_reset(); + + return NOTIFY_OK; + case CPU_PM_EXIT: + if (__this_cpu_read(kvm_arm_hardware_enabled)) + /* The hardware was enabled before suspend. */ + cpu_hyp_reinit(); + return NOTIFY_OK; - } - return NOTIFY_DONE; + default: + return NOTIFY_DONE; + } } static struct notifier_block hyp_init_cpu_pm_nb = { @@ -1025,6 +1089,91 @@ static inline void hyp_cpu_pm_init(void) } #endif +static void teardown_common_resources(void) +{ + free_percpu(kvm_host_cpu_state); +} + +static int init_common_resources(void) +{ + kvm_host_cpu_state = alloc_percpu(kvm_cpu_context_t); + if (!kvm_host_cpu_state) { + kvm_err("Cannot allocate host CPU state\n"); + return -ENOMEM; + } + + return 0; +} + +static int init_subsystems(void) +{ + int err = 0; + + /* + * Enable hardware so that subsystem initialisation can access EL2. + */ + on_each_cpu(_kvm_arch_hardware_enable, NULL, 1); + + /* + * Register CPU lower-power notifier + */ + hyp_cpu_pm_init(); + + /* + * Init HYP view of VGIC + */ + err = kvm_vgic_hyp_init(); + switch (err) { + case 0: + vgic_present = true; + break; + case -ENODEV: + case -ENXIO: + vgic_present = false; + err = 0; + break; + default: + goto out; + } + + /* + * Init HYP architected timer support + */ + err = kvm_timer_hyp_init(); + if (err) + goto out; + + kvm_perf_init(); + kvm_coproc_table_init(); + +out: + on_each_cpu(_kvm_arch_hardware_disable, NULL, 1); + + return err; +} + +static void teardown_hyp_mode(void) +{ + int cpu; + + if (is_kernel_in_hyp_mode()) + return; + + free_hyp_pgds(); + for_each_possible_cpu(cpu) + free_page(per_cpu(kvm_arm_hyp_stack_page, cpu)); +} + +static int init_vhe_mode(void) +{ + /* set size of VMID supported by CPU */ + kvm_vmid_bits = kvm_get_vmid_bits(); + kvm_info("%d-bit VMID\n", kvm_vmid_bits); + + kvm_info("VHE mode initialized successfully\n"); + return 0; +} + /** * Inits Hyp-mode on all online CPUs */ @@ -1055,7 +1204,7 @@ static int init_hyp_mode(void) stack_page = __get_free_page(GFP_KERNEL); if (!stack_page) { err = -ENOMEM; - goto out_free_stack_pages; + goto out_err; } per_cpu(kvm_arm_hyp_stack_page, cpu) = stack_page; @@ -1068,7 +1217,14 @@ static int init_hyp_mode(void) kvm_ksym_ref(__kvm_hyp_code_end)); if (err) { kvm_err("Cannot map world-switch code\n"); - goto out_free_mappings; + goto out_err; + } + + err = create_hyp_mappings(kvm_ksym_ref(__start_rodata), + kvm_ksym_ref(__end_rodata)); + if (err) { + kvm_err("Cannot map rodata section\n"); + goto out_err; } /* @@ -1080,20 +1236,10 @@ static int init_hyp_mode(void) if (err) { kvm_err("Cannot map hyp stack\n"); - goto out_free_mappings; + goto out_err; } } - /* - * Map the host CPU structures - */ - kvm_host_cpu_state = alloc_percpu(kvm_cpu_context_t); - if (!kvm_host_cpu_state) { - err = -ENOMEM; - kvm_err("Cannot allocate host CPU state\n"); - goto out_free_mappings; - } - for_each_possible_cpu(cpu) { kvm_cpu_context_t *cpu_ctxt; @@ -1102,46 +1248,24 @@ static int init_hyp_mode(void) if (err) { kvm_err("Cannot map host CPU state: %d\n", err); - goto out_free_context; + goto out_err; } } - /* - * Execute the init code on each CPU. - */ - on_each_cpu(cpu_init_hyp_mode, NULL, 1); - - /* - * Init HYP view of VGIC - */ - err = kvm_vgic_hyp_init(); - if (err) - goto out_free_context; - - /* - * Init HYP architected timer support - */ - err = kvm_timer_hyp_init(); - if (err) - goto out_free_context; - #ifndef CONFIG_HOTPLUG_CPU free_boot_hyp_pgd(); #endif - kvm_perf_init(); + /* set size of VMID supported by CPU */ + kvm_vmid_bits = kvm_get_vmid_bits(); + kvm_info("%d-bit VMID\n", kvm_vmid_bits); kvm_info("Hyp mode initialized successfully\n"); return 0; -out_free_context: - free_percpu(kvm_host_cpu_state); -out_free_mappings: - free_hyp_pgds(); -out_free_stack_pages: - for_each_possible_cpu(cpu) - free_page(per_cpu(kvm_arm_hyp_stack_page, cpu)); + out_err: + teardown_hyp_mode(); kvm_err("error initializing Hyp mode: %d\n", err); return err; } @@ -1185,26 +1309,27 @@ int kvm_arch_init(void *opaque) } } - cpu_notifier_register_begin(); - - err = init_hyp_mode(); + err = init_common_resources(); if (err) - goto out_err; + return err; - err = __register_cpu_notifier(&hyp_init_cpu_nb); - if (err) { - kvm_err("Cannot register HYP init CPU notifier (%d)\n", err); + if (is_kernel_in_hyp_mode()) + err = init_vhe_mode(); + else + err = init_hyp_mode(); + if (err) goto out_err; - } - - cpu_notifier_register_done(); - hyp_cpu_pm_init(); + err = init_subsystems(); + if (err) + goto out_hyp; - kvm_coproc_table_init(); return 0; + +out_hyp: + teardown_hyp_mode(); out_err: - cpu_notifier_register_done(); + teardown_common_resources(); return err; } diff --git a/arch/arm/kvm/emulate.c b/arch/arm/kvm/emulate.c index d6c005283678..dc99159857b4 100644 --- a/arch/arm/kvm/emulate.c +++ b/arch/arm/kvm/emulate.c @@ -275,6 +275,40 @@ static u32 exc_vector_base(struct kvm_vcpu *vcpu) return vbar; } +/* + * Switch to an exception mode, updating both CPSR and SPSR. Follow + * the logic described in AArch32.EnterMode() from the ARMv8 ARM. + */ +static void kvm_update_psr(struct kvm_vcpu *vcpu, unsigned long mode) +{ + unsigned long cpsr = *vcpu_cpsr(vcpu); + u32 sctlr = vcpu->arch.cp15[c1_SCTLR]; + + *vcpu_cpsr(vcpu) = (cpsr & ~MODE_MASK) | mode; + + switch (mode) { + case FIQ_MODE: + *vcpu_cpsr(vcpu) |= PSR_F_BIT; + /* Fall through */ + case ABT_MODE: + case IRQ_MODE: + *vcpu_cpsr(vcpu) |= PSR_A_BIT; + /* Fall through */ + default: + *vcpu_cpsr(vcpu) |= PSR_I_BIT; + } + + *vcpu_cpsr(vcpu) &= ~(PSR_IT_MASK | PSR_J_BIT | PSR_E_BIT | PSR_T_BIT); + + if (sctlr & SCTLR_TE) + *vcpu_cpsr(vcpu) |= PSR_T_BIT; + if (sctlr & SCTLR_EE) + *vcpu_cpsr(vcpu) |= PSR_E_BIT; + + /* Note: These now point to the mode banked copies */ + *vcpu_spsr(vcpu) = cpsr; +} + /** * kvm_inject_undefined - inject an undefined exception into the guest * @vcpu: The VCPU to receive the undefined exception @@ -286,29 +320,13 @@ static u32 exc_vector_base(struct kvm_vcpu *vcpu) */ void kvm_inject_undefined(struct kvm_vcpu *vcpu) { - unsigned long new_lr_value; - unsigned long new_spsr_value; unsigned long cpsr = *vcpu_cpsr(vcpu); - u32 sctlr = vcpu->arch.cp15[c1_SCTLR]; bool is_thumb = (cpsr & PSR_T_BIT); u32 vect_offset = 4; u32 return_offset = (is_thumb) ? 2 : 4; - new_spsr_value = cpsr; - new_lr_value = *vcpu_pc(vcpu) - return_offset; - - *vcpu_cpsr(vcpu) = (cpsr & ~MODE_MASK) | UND_MODE; - *vcpu_cpsr(vcpu) |= PSR_I_BIT; - *vcpu_cpsr(vcpu) &= ~(PSR_IT_MASK | PSR_J_BIT | PSR_E_BIT | PSR_T_BIT); - - if (sctlr & SCTLR_TE) - *vcpu_cpsr(vcpu) |= PSR_T_BIT; - if (sctlr & SCTLR_EE) - *vcpu_cpsr(vcpu) |= PSR_E_BIT; - - /* Note: These now point to UND banked copies */ - *vcpu_spsr(vcpu) = cpsr; - *vcpu_reg(vcpu, 14) = new_lr_value; + kvm_update_psr(vcpu, UND_MODE); + *vcpu_reg(vcpu, 14) = *vcpu_pc(vcpu) - return_offset; /* Branch to exception vector */ *vcpu_pc(vcpu) = exc_vector_base(vcpu) + vect_offset; @@ -320,30 +338,14 @@ void kvm_inject_undefined(struct kvm_vcpu *vcpu) */ static void inject_abt(struct kvm_vcpu *vcpu, bool is_pabt, unsigned long addr) { - unsigned long new_lr_value; - unsigned long new_spsr_value; unsigned long cpsr = *vcpu_cpsr(vcpu); - u32 sctlr = vcpu->arch.cp15[c1_SCTLR]; bool is_thumb = (cpsr & PSR_T_BIT); u32 vect_offset; u32 return_offset = (is_thumb) ? 4 : 0; bool is_lpae; - new_spsr_value = cpsr; - new_lr_value = *vcpu_pc(vcpu) + return_offset; - - *vcpu_cpsr(vcpu) = (cpsr & ~MODE_MASK) | ABT_MODE; - *vcpu_cpsr(vcpu) |= PSR_I_BIT | PSR_A_BIT; - *vcpu_cpsr(vcpu) &= ~(PSR_IT_MASK | PSR_J_BIT | PSR_E_BIT | PSR_T_BIT); - - if (sctlr & SCTLR_TE) - *vcpu_cpsr(vcpu) |= PSR_T_BIT; - if (sctlr & SCTLR_EE) - *vcpu_cpsr(vcpu) |= PSR_E_BIT; - - /* Note: These now point to ABT banked copies */ - *vcpu_spsr(vcpu) = cpsr; - *vcpu_reg(vcpu, 14) = new_lr_value; + kvm_update_psr(vcpu, ABT_MODE); + *vcpu_reg(vcpu, 14) = *vcpu_pc(vcpu) + return_offset; if (is_pabt) vect_offset = 12; diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c index 52dd70702501..28c758bafb43 100644 --- a/arch/arm/kvm/mmu.c +++ b/arch/arm/kvm/mmu.c @@ -28,6 +28,7 @@ #include <asm/kvm_mmio.h> #include <asm/kvm_asm.h> #include <asm/kvm_emulate.h> +#include <asm/virt.h> #include "trace.h" @@ -598,6 +599,9 @@ int create_hyp_mappings(void *from, void *to) unsigned long start = KERN_TO_HYP((unsigned long)from); unsigned long end = KERN_TO_HYP((unsigned long)to); + if (is_kernel_in_hyp_mode()) + return 0; + start = start & PAGE_MASK; end = PAGE_ALIGN(end); @@ -630,6 +634,9 @@ int create_hyp_io_mappings(void *from, void *to, phys_addr_t phys_addr) unsigned long start = KERN_TO_HYP((unsigned long)from); unsigned long end = KERN_TO_HYP((unsigned long)to); + if (is_kernel_in_hyp_mode()) + return 0; + /* Check for a valid kernel IO mapping */ if (!is_vmalloc_addr(from) || !is_vmalloc_addr(to - 1)) return -EINVAL; @@ -656,9 +663,9 @@ static void *kvm_alloc_hwpgd(void) * kvm_alloc_stage2_pgd - allocate level-1 table for stage-2 translation. * @kvm: The KVM struct pointer for the VM. * - * Allocates the 1st level table only of size defined by S2_PGD_ORDER (can - * support either full 40-bit input addresses or limited to 32-bit input - * addresses). Clears the allocated pages. + * Allocates only the stage-2 HW PGD level table(s) (can support either full + * 40-bit input addresses or limited to 32-bit input addresses). Clears the + * allocated pages. * * Note we don't need locking here as this is only called when the VM is * created, which can only be done once. @@ -1682,6 +1689,11 @@ phys_addr_t kvm_get_idmap_vector(void) return hyp_idmap_vector; } +phys_addr_t kvm_get_idmap_start(void) +{ + return hyp_idmap_start; +} + int kvm_mmu_init(void) { int err; |