diff options
Diffstat (limited to 'arch/arm64/kvm')
24 files changed, 497 insertions, 479 deletions
diff --git a/arch/arm64/kvm/Kconfig b/arch/arm64/kvm/Kconfig index 096e45acadb2..713248f240e0 100644 --- a/arch/arm64/kvm/Kconfig +++ b/arch/arm64/kvm/Kconfig @@ -19,7 +19,6 @@ if VIRTUALIZATION menuconfig KVM bool "Kernel-based Virtual Machine (KVM) support" - depends on AS_HAS_ARMV8_4 select KVM_COMMON select KVM_GENERIC_HARDWARE_ENABLING select KVM_GENERIC_MMU_NOTIFIER diff --git a/arch/arm64/kvm/arch_timer.c b/arch/arm64/kvm/arch_timer.c index 5133dcbfe9f7..701ea10a63f1 100644 --- a/arch/arm64/kvm/arch_timer.c +++ b/arch/arm64/kvm/arch_timer.c @@ -108,16 +108,16 @@ static void timer_set_ctl(struct arch_timer_context *ctxt, u32 ctl) switch(arch_timer_ctx_index(ctxt)) { case TIMER_VTIMER: - __vcpu_sys_reg(vcpu, CNTV_CTL_EL0) = ctl; + __vcpu_assign_sys_reg(vcpu, CNTV_CTL_EL0, ctl); break; case TIMER_PTIMER: - __vcpu_sys_reg(vcpu, CNTP_CTL_EL0) = ctl; + __vcpu_assign_sys_reg(vcpu, CNTP_CTL_EL0, ctl); break; case TIMER_HVTIMER: - __vcpu_sys_reg(vcpu, CNTHV_CTL_EL2) = ctl; + __vcpu_assign_sys_reg(vcpu, CNTHV_CTL_EL2, ctl); break; case TIMER_HPTIMER: - __vcpu_sys_reg(vcpu, CNTHP_CTL_EL2) = ctl; + __vcpu_assign_sys_reg(vcpu, CNTHP_CTL_EL2, ctl); break; default: WARN_ON(1); @@ -130,16 +130,16 @@ static void timer_set_cval(struct arch_timer_context *ctxt, u64 cval) switch(arch_timer_ctx_index(ctxt)) { case TIMER_VTIMER: - __vcpu_sys_reg(vcpu, CNTV_CVAL_EL0) = cval; + __vcpu_assign_sys_reg(vcpu, CNTV_CVAL_EL0, cval); break; case TIMER_PTIMER: - __vcpu_sys_reg(vcpu, CNTP_CVAL_EL0) = cval; + __vcpu_assign_sys_reg(vcpu, CNTP_CVAL_EL0, cval); break; case TIMER_HVTIMER: - __vcpu_sys_reg(vcpu, CNTHV_CVAL_EL2) = cval; + __vcpu_assign_sys_reg(vcpu, CNTHV_CVAL_EL2, cval); break; case TIMER_HPTIMER: - __vcpu_sys_reg(vcpu, CNTHP_CVAL_EL2) = cval; + __vcpu_assign_sys_reg(vcpu, CNTHP_CVAL_EL2, cval); break; default: WARN_ON(1); @@ -1036,7 +1036,7 @@ void kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu) if (vcpu_has_nv(vcpu)) { struct arch_timer_offset *offs = &vcpu_vtimer(vcpu)->offset; - offs->vcpu_offset = &__vcpu_sys_reg(vcpu, CNTVOFF_EL2); + offs->vcpu_offset = __ctxt_sys_reg(&vcpu->arch.ctxt, CNTVOFF_EL2); offs->vm_offset = &vcpu->kvm->arch.timer_data.poffset; } @@ -1766,7 +1766,7 @@ int kvm_vm_ioctl_set_counter_offset(struct kvm *kvm, mutex_lock(&kvm->lock); - if (lock_all_vcpus(kvm)) { + if (!kvm_trylock_all_vcpus(kvm)) { set_bit(KVM_ARCH_FLAG_VM_COUNTER_OFFSET, &kvm->arch.flags); /* @@ -1778,7 +1778,7 @@ int kvm_vm_ioctl_set_counter_offset(struct kvm *kvm, kvm->arch.timer_data.voffset = offset->counter_offset; kvm->arch.timer_data.poffset = offset->counter_offset; - unlock_all_vcpus(kvm); + kvm_unlock_all_vcpus(kvm); } else { ret = -EBUSY; } diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index 36cfcffb40d8..23dd3f3fc3eb 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -825,10 +825,6 @@ int kvm_arch_vcpu_run_pid_change(struct kvm_vcpu *vcpu) if (!kvm_arm_vcpu_is_finalized(vcpu)) return -EPERM; - ret = kvm_arch_vcpu_run_map_fp(vcpu); - if (ret) - return ret; - if (likely(vcpu_has_run_once(vcpu))) return 0; @@ -1924,49 +1920,6 @@ int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) } } -/* unlocks vcpus from @vcpu_lock_idx and smaller */ -static void unlock_vcpus(struct kvm *kvm, int vcpu_lock_idx) -{ - struct kvm_vcpu *tmp_vcpu; - - for (; vcpu_lock_idx >= 0; vcpu_lock_idx--) { - tmp_vcpu = kvm_get_vcpu(kvm, vcpu_lock_idx); - mutex_unlock(&tmp_vcpu->mutex); - } -} - -void unlock_all_vcpus(struct kvm *kvm) -{ - lockdep_assert_held(&kvm->lock); - - unlock_vcpus(kvm, atomic_read(&kvm->online_vcpus) - 1); -} - -/* Returns true if all vcpus were locked, false otherwise */ -bool lock_all_vcpus(struct kvm *kvm) -{ - struct kvm_vcpu *tmp_vcpu; - unsigned long c; - - lockdep_assert_held(&kvm->lock); - - /* - * Any time a vcpu is in an ioctl (including running), the - * core KVM code tries to grab the vcpu->mutex. - * - * By grabbing the vcpu->mutex of all VCPUs we ensure that no - * other VCPUs can fiddle with the state while we access it. - */ - kvm_for_each_vcpu(c, tmp_vcpu, kvm) { - if (!mutex_trylock(&tmp_vcpu->mutex)) { - unlock_vcpus(kvm, c - 1); - return false; - } - } - - return true; -} - static unsigned long nvhe_percpu_size(void) { return (unsigned long)CHOOSE_NVHE_SYM(__per_cpu_end) - @@ -2172,7 +2125,7 @@ static void cpu_hyp_init(void *discard) static void cpu_hyp_uninit(void *discard) { - if (__this_cpu_read(kvm_hyp_initialized)) { + if (!is_protected_kvm_enabled() && __this_cpu_read(kvm_hyp_initialized)) { cpu_hyp_reset(); __this_cpu_write(kvm_hyp_initialized, 0); } @@ -2388,8 +2341,13 @@ static void __init teardown_hyp_mode(void) free_hyp_pgds(); for_each_possible_cpu(cpu) { + if (per_cpu(kvm_hyp_initialized, cpu)) + continue; + 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 (!kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu]) + continue; if (free_sve) { struct cpu_sve_state *sve_state; @@ -2397,6 +2355,9 @@ static void __init teardown_hyp_mode(void) sve_state = per_cpu_ptr_nvhe_sym(kvm_host_data, cpu)->sve_state; free_pages((unsigned long) sve_state, pkvm_host_sve_state_order()); } + + free_pages(kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu], nvhe_percpu_order()); + } } @@ -2790,6 +2751,7 @@ int kvm_arch_irq_bypass_add_producer(struct irq_bypass_consumer *cons, return kvm_vgic_v4_set_forwarding(irqfd->kvm, prod->irq, &irqfd->irq_entry); } + void kvm_arch_irq_bypass_del_producer(struct irq_bypass_consumer *cons, struct irq_bypass_producer *prod) { @@ -2800,8 +2762,30 @@ void kvm_arch_irq_bypass_del_producer(struct irq_bypass_consumer *cons, if (irq_entry->type != KVM_IRQ_ROUTING_MSI) return; - kvm_vgic_v4_unset_forwarding(irqfd->kvm, prod->irq, - &irqfd->irq_entry); + kvm_vgic_v4_unset_forwarding(irqfd->kvm, prod->irq); +} + +bool kvm_arch_irqfd_route_changed(struct kvm_kernel_irq_routing_entry *old, + struct kvm_kernel_irq_routing_entry *new) +{ + if (old->type != KVM_IRQ_ROUTING_MSI || + new->type != KVM_IRQ_ROUTING_MSI) + return true; + + return memcmp(&old->msi, &new->msi, sizeof(new->msi)); +} + +int kvm_arch_update_irqfd_routing(struct kvm *kvm, unsigned int host_irq, + uint32_t guest_irq, bool set) +{ + /* + * Remapping the vLPI requires taking the its_lock mutex to resolve + * the new translation. We're in spinlock land at this point, so no + * chance of resolving the translation. + * + * Unmap the vLPI and fall back to software LPI injection. + */ + return kvm_vgic_v4_unset_forwarding(kvm, host_irq); } void kvm_arch_irq_bypass_stop(struct irq_bypass_consumer *cons) diff --git a/arch/arm64/kvm/debug.c b/arch/arm64/kvm/debug.c index 0e4c805e7e89..1a7dab333f55 100644 --- a/arch/arm64/kvm/debug.c +++ b/arch/arm64/kvm/debug.c @@ -216,9 +216,9 @@ void kvm_debug_set_guest_ownership(struct kvm_vcpu *vcpu) void kvm_debug_handle_oslar(struct kvm_vcpu *vcpu, u64 val) { if (val & OSLAR_EL1_OSLK) - __vcpu_sys_reg(vcpu, OSLSR_EL1) |= OSLSR_EL1_OSLK; + __vcpu_rmw_sys_reg(vcpu, OSLSR_EL1, |=, OSLSR_EL1_OSLK); else - __vcpu_sys_reg(vcpu, OSLSR_EL1) &= ~OSLSR_EL1_OSLK; + __vcpu_rmw_sys_reg(vcpu, OSLSR_EL1, &=, ~OSLSR_EL1_OSLK); preempt_disable(); kvm_arch_vcpu_put(vcpu); diff --git a/arch/arm64/kvm/fpsimd.c b/arch/arm64/kvm/fpsimd.c index 7f6e43d25691..15e17aca1dec 100644 --- a/arch/arm64/kvm/fpsimd.c +++ b/arch/arm64/kvm/fpsimd.c @@ -15,32 +15,6 @@ #include <asm/sysreg.h> /* - * Called on entry to KVM_RUN unless this vcpu previously ran at least - * once and the most recent prior KVM_RUN for this vcpu was called from - * the same task as current (highly likely). - * - * This is guaranteed to execute before kvm_arch_vcpu_load_fp(vcpu), - * such that on entering hyp the relevant parts of current are already - * mapped. - */ -int kvm_arch_vcpu_run_map_fp(struct kvm_vcpu *vcpu) -{ - struct user_fpsimd_state *fpsimd = ¤t->thread.uw.fpsimd_state; - int ret; - - /* pKVM has its own tracking of the host fpsimd state. */ - if (is_protected_kvm_enabled()) - return 0; - - /* Make sure the host task fpsimd state is visible to hyp: */ - ret = kvm_share_hyp(fpsimd, fpsimd + 1); - if (ret) - return ret; - - return 0; -} - -/* * Prepare vcpu for saving the host's FPSIMD state and loading the guest's. * The actual loading is done by the FPSIMD access trap taken to hyp. * @@ -103,8 +77,8 @@ void kvm_arch_vcpu_ctxsync_fp(struct kvm_vcpu *vcpu) fp_state.sve_state = vcpu->arch.sve_state; fp_state.sve_vl = vcpu->arch.sve_max_vl; fp_state.sme_state = NULL; - fp_state.svcr = &__vcpu_sys_reg(vcpu, SVCR); - fp_state.fpmr = &__vcpu_sys_reg(vcpu, FPMR); + fp_state.svcr = __ctxt_sys_reg(&vcpu->arch.ctxt, SVCR); + fp_state.fpmr = __ctxt_sys_reg(&vcpu->arch.ctxt, FPMR); fp_state.fp_type = &vcpu->arch.fp_type; if (vcpu_has_sve(vcpu)) diff --git a/arch/arm64/kvm/hyp/exception.c b/arch/arm64/kvm/hyp/exception.c index 424a5107cddb..6a2a899a344e 100644 --- a/arch/arm64/kvm/hyp/exception.c +++ b/arch/arm64/kvm/hyp/exception.c @@ -37,7 +37,7 @@ static inline void __vcpu_write_sys_reg(struct kvm_vcpu *vcpu, u64 val, int reg) if (unlikely(vcpu_has_nv(vcpu))) vcpu_write_sys_reg(vcpu, val, reg); else if (!__vcpu_write_sys_reg_to_cpu(val, reg)) - __vcpu_sys_reg(vcpu, reg) = val; + __vcpu_assign_sys_reg(vcpu, reg, val); } static void __vcpu_write_spsr(struct kvm_vcpu *vcpu, unsigned long target_mode, @@ -51,7 +51,7 @@ static void __vcpu_write_spsr(struct kvm_vcpu *vcpu, unsigned long target_mode, } else if (has_vhe()) { write_sysreg_el1(val, SYS_SPSR); } else { - __vcpu_sys_reg(vcpu, SPSR_EL1) = val; + __vcpu_assign_sys_reg(vcpu, SPSR_EL1, val); } } diff --git a/arch/arm64/kvm/hyp/include/hyp/switch.h b/arch/arm64/kvm/hyp/include/hyp/switch.h index bb9f2eecfb67..2ad57b117385 100644 --- a/arch/arm64/kvm/hyp/include/hyp/switch.h +++ b/arch/arm64/kvm/hyp/include/hyp/switch.h @@ -45,7 +45,7 @@ static inline void __fpsimd_save_fpexc32(struct kvm_vcpu *vcpu) if (!vcpu_el1_is_32bit(vcpu)) return; - __vcpu_sys_reg(vcpu, FPEXC32_EL2) = read_sysreg(fpexc32_el2); + __vcpu_assign_sys_reg(vcpu, FPEXC32_EL2, read_sysreg(fpexc32_el2)); } static inline void __activate_traps_fpsimd32(struct kvm_vcpu *vcpu) @@ -65,6 +65,136 @@ static inline void __activate_traps_fpsimd32(struct kvm_vcpu *vcpu) } } +static inline void __activate_cptr_traps_nvhe(struct kvm_vcpu *vcpu) +{ + u64 val = CPTR_NVHE_EL2_RES1 | CPTR_EL2_TAM | CPTR_EL2_TTA; + + /* + * Always trap SME since it's not supported in KVM. + * TSM is RES1 if SME isn't implemented. + */ + val |= CPTR_EL2_TSM; + + if (!vcpu_has_sve(vcpu) || !guest_owns_fp_regs()) + val |= CPTR_EL2_TZ; + + if (!guest_owns_fp_regs()) + val |= CPTR_EL2_TFP; + + write_sysreg(val, cptr_el2); +} + +static inline void __activate_cptr_traps_vhe(struct kvm_vcpu *vcpu) +{ + /* + * With VHE (HCR.E2H == 1), accesses to CPACR_EL1 are routed to + * CPTR_EL2. In general, CPACR_EL1 has the same layout as CPTR_EL2, + * except for some missing controls, such as TAM. + * In this case, CPTR_EL2.TAM has the same position with or without + * VHE (HCR.E2H == 1) which allows us to use here the CPTR_EL2.TAM + * shift value for trapping the AMU accesses. + */ + u64 val = CPTR_EL2_TAM | CPACR_EL1_TTA; + u64 cptr; + + if (guest_owns_fp_regs()) { + val |= CPACR_EL1_FPEN; + if (vcpu_has_sve(vcpu)) + val |= CPACR_EL1_ZEN; + } + + if (!vcpu_has_nv(vcpu)) + goto write; + + /* + * The architecture is a bit crap (what a surprise): an EL2 guest + * writing to CPTR_EL2 via CPACR_EL1 can't set any of TCPAC or TTA, + * as they are RES0 in the guest's view. To work around it, trap the + * sucker using the very same bit it can't set... + */ + if (vcpu_el2_e2h_is_set(vcpu) && is_hyp_ctxt(vcpu)) + val |= CPTR_EL2_TCPAC; + + /* + * Layer the guest hypervisor's trap configuration on top of our own if + * we're in a nested context. + */ + if (is_hyp_ctxt(vcpu)) + goto write; + + cptr = vcpu_sanitised_cptr_el2(vcpu); + + /* + * Pay attention, there's some interesting detail here. + * + * The CPTR_EL2.xEN fields are 2 bits wide, although there are only two + * meaningful trap states when HCR_EL2.TGE = 0 (running a nested guest): + * + * - CPTR_EL2.xEN = x0, traps are enabled + * - CPTR_EL2.xEN = x1, traps are disabled + * + * In other words, bit[0] determines if guest accesses trap or not. In + * the interest of simplicity, clear the entire field if the guest + * hypervisor has traps enabled to dispel any illusion of something more + * complicated taking place. + */ + if (!(SYS_FIELD_GET(CPACR_EL1, FPEN, cptr) & BIT(0))) + val &= ~CPACR_EL1_FPEN; + if (!(SYS_FIELD_GET(CPACR_EL1, ZEN, cptr) & BIT(0))) + val &= ~CPACR_EL1_ZEN; + + if (kvm_has_feat(vcpu->kvm, ID_AA64MMFR3_EL1, S2POE, IMP)) + val |= cptr & CPACR_EL1_E0POE; + + val |= cptr & CPTR_EL2_TCPAC; + +write: + write_sysreg(val, cpacr_el1); +} + +static inline void __activate_cptr_traps(struct kvm_vcpu *vcpu) +{ + if (!guest_owns_fp_regs()) + __activate_traps_fpsimd32(vcpu); + + if (has_vhe() || has_hvhe()) + __activate_cptr_traps_vhe(vcpu); + else + __activate_cptr_traps_nvhe(vcpu); +} + +static inline void __deactivate_cptr_traps_nvhe(struct kvm_vcpu *vcpu) +{ + u64 val = CPTR_NVHE_EL2_RES1; + + if (!cpus_have_final_cap(ARM64_SVE)) + val |= CPTR_EL2_TZ; + if (!cpus_have_final_cap(ARM64_SME)) + val |= CPTR_EL2_TSM; + + write_sysreg(val, cptr_el2); +} + +static inline void __deactivate_cptr_traps_vhe(struct kvm_vcpu *vcpu) +{ + u64 val = CPACR_EL1_FPEN; + + if (cpus_have_final_cap(ARM64_SVE)) + val |= CPACR_EL1_ZEN; + if (cpus_have_final_cap(ARM64_SME)) + val |= CPACR_EL1_SMEN; + + write_sysreg(val, cpacr_el1); +} + +static inline void __deactivate_cptr_traps(struct kvm_vcpu *vcpu) +{ + if (has_vhe() || has_hvhe()) + __deactivate_cptr_traps_vhe(vcpu); + else + __deactivate_cptr_traps_nvhe(vcpu); +} + #define reg_to_fgt_masks(reg) \ ({ \ struct fgt_masks *m; \ @@ -456,7 +586,7 @@ static inline void fpsimd_lazy_switch_to_host(struct kvm_vcpu *vcpu) */ if (vcpu_has_sve(vcpu)) { zcr_el1 = read_sysreg_el1(SYS_ZCR); - __vcpu_sys_reg(vcpu, vcpu_sve_zcr_elx(vcpu)) = zcr_el1; + __vcpu_assign_sys_reg(vcpu, vcpu_sve_zcr_elx(vcpu), zcr_el1); /* * The guest's state is always saved using the guest's max VL. @@ -486,11 +616,6 @@ static void kvm_hyp_save_fpsimd_host(struct kvm_vcpu *vcpu) */ if (system_supports_sve()) { __hyp_sve_save_host(); - - /* Re-enable SVE traps if not supported for the guest vcpu. */ - if (!vcpu_has_sve(vcpu)) - cpacr_clear_set(CPACR_EL1_ZEN, 0); - } else { __fpsimd_save_state(host_data_ptr(host_ctxt.fp_regs)); } @@ -541,10 +666,7 @@ static inline bool kvm_hyp_handle_fpsimd(struct kvm_vcpu *vcpu, u64 *exit_code) /* Valid trap. Switch the context: */ /* First disable enough traps to allow us to update the registers */ - if (sve_guest || (is_protected_kvm_enabled() && system_supports_sve())) - cpacr_clear_set(0, CPACR_EL1_FPEN | CPACR_EL1_ZEN); - else - cpacr_clear_set(0, CPACR_EL1_FPEN); + __deactivate_cptr_traps(vcpu); isb(); /* Write out the host state if it's in the registers */ @@ -566,6 +688,13 @@ static inline bool kvm_hyp_handle_fpsimd(struct kvm_vcpu *vcpu, u64 *exit_code) *host_data_ptr(fp_owner) = FP_STATE_GUEST_OWNED; + /* + * Re-enable traps necessary for the current state of the guest, e.g. + * those enabled by a guest hypervisor. The ERET to the guest will + * provide the necessary context synchronization. + */ + __activate_cptr_traps(vcpu); + return true; } diff --git a/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h b/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h index b9cff893bbe0..4d0dbea4c56f 100644 --- a/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h +++ b/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h @@ -307,11 +307,11 @@ static inline void __sysreg32_save_state(struct kvm_vcpu *vcpu) vcpu->arch.ctxt.spsr_irq = read_sysreg(spsr_irq); vcpu->arch.ctxt.spsr_fiq = read_sysreg(spsr_fiq); - __vcpu_sys_reg(vcpu, DACR32_EL2) = read_sysreg(dacr32_el2); - __vcpu_sys_reg(vcpu, IFSR32_EL2) = read_sysreg(ifsr32_el2); + __vcpu_assign_sys_reg(vcpu, DACR32_EL2, read_sysreg(dacr32_el2)); + __vcpu_assign_sys_reg(vcpu, IFSR32_EL2, read_sysreg(ifsr32_el2)); if (has_vhe() || kvm_debug_regs_in_use(vcpu)) - __vcpu_sys_reg(vcpu, DBGVCR32_EL2) = read_sysreg(dbgvcr32_el2); + __vcpu_assign_sys_reg(vcpu, DBGVCR32_EL2, read_sysreg(dbgvcr32_el2)); } static inline void __sysreg32_restore_state(struct kvm_vcpu *vcpu) diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-main.c b/arch/arm64/kvm/hyp/nvhe/hyp-main.c index 8e8848de4d47..3206b2c07f82 100644 --- a/arch/arm64/kvm/hyp/nvhe/hyp-main.c +++ b/arch/arm64/kvm/hyp/nvhe/hyp-main.c @@ -26,7 +26,7 @@ void __kvm_hyp_host_forward_smc(struct kvm_cpu_context *host_ctxt); static void __hyp_sve_save_guest(struct kvm_vcpu *vcpu) { - __vcpu_sys_reg(vcpu, ZCR_EL1) = read_sysreg_el1(SYS_ZCR); + __vcpu_assign_sys_reg(vcpu, ZCR_EL1, read_sysreg_el1(SYS_ZCR)); /* * On saving/restoring guest sve state, always use the maximum VL for * the guest. The layout of the data when saving the sve state depends @@ -69,7 +69,10 @@ static void fpsimd_sve_sync(struct kvm_vcpu *vcpu) if (!guest_owns_fp_regs()) return; - cpacr_clear_set(0, CPACR_EL1_FPEN | CPACR_EL1_ZEN); + /* + * Traps have been disabled by __deactivate_cptr_traps(), but there + * hasn't necessarily been a context synchronization event yet. + */ isb(); if (vcpu_has_sve(vcpu)) @@ -79,7 +82,7 @@ static void fpsimd_sve_sync(struct kvm_vcpu *vcpu) has_fpmr = kvm_has_fpmr(kern_hyp_va(vcpu->kvm)); if (has_fpmr) - __vcpu_sys_reg(vcpu, FPMR) = read_sysreg_s(SYS_FPMR); + __vcpu_assign_sys_reg(vcpu, FPMR, read_sysreg_s(SYS_FPMR)); if (system_supports_sve()) __hyp_sve_restore_host(); diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c index 95d7534c9679..8957734d6183 100644 --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c @@ -479,6 +479,7 @@ static int host_stage2_adjust_range(u64 addr, struct kvm_mem_range *range) { struct kvm_mem_range cur; kvm_pte_t pte; + u64 granule; s8 level; int ret; @@ -496,18 +497,21 @@ static int host_stage2_adjust_range(u64 addr, struct kvm_mem_range *range) return -EPERM; } - do { - u64 granule = kvm_granule_size(level); + for (; level <= KVM_PGTABLE_LAST_LEVEL; level++) { + if (!kvm_level_supports_block_mapping(level)) + continue; + granule = kvm_granule_size(level); cur.start = ALIGN_DOWN(addr, granule); cur.end = cur.start + granule; - level++; - } while ((level <= KVM_PGTABLE_LAST_LEVEL) && - !(kvm_level_supports_block_mapping(level) && - range_included(&cur, range))); + if (!range_included(&cur, range)) + continue; + *range = cur; + return 0; + } - *range = cur; + WARN_ON(1); - return 0; + return -EINVAL; } int host_stage2_idmap_locked(phys_addr_t addr, u64 size, diff --git a/arch/arm64/kvm/hyp/nvhe/switch.c b/arch/arm64/kvm/hyp/nvhe/switch.c index 73affe1333a4..0e752b515d0f 100644 --- a/arch/arm64/kvm/hyp/nvhe/switch.c +++ b/arch/arm64/kvm/hyp/nvhe/switch.c @@ -47,65 +47,6 @@ struct fgt_masks hdfgwtr2_masks; extern void kvm_nvhe_prepare_backtrace(unsigned long fp, unsigned long pc); -static void __activate_cptr_traps(struct kvm_vcpu *vcpu) -{ - u64 val = CPTR_EL2_TAM; /* Same bit irrespective of E2H */ - - if (!guest_owns_fp_regs()) - __activate_traps_fpsimd32(vcpu); - - if (has_hvhe()) { - val |= CPACR_EL1_TTA; - - if (guest_owns_fp_regs()) { - val |= CPACR_EL1_FPEN; - if (vcpu_has_sve(vcpu)) - val |= CPACR_EL1_ZEN; - } - - write_sysreg(val, cpacr_el1); - } else { - val |= CPTR_EL2_TTA | CPTR_NVHE_EL2_RES1; - - /* - * Always trap SME since it's not supported in KVM. - * TSM is RES1 if SME isn't implemented. - */ - val |= CPTR_EL2_TSM; - - if (!vcpu_has_sve(vcpu) || !guest_owns_fp_regs()) - val |= CPTR_EL2_TZ; - - if (!guest_owns_fp_regs()) - val |= CPTR_EL2_TFP; - - write_sysreg(val, cptr_el2); - } -} - -static void __deactivate_cptr_traps(struct kvm_vcpu *vcpu) -{ - if (has_hvhe()) { - u64 val = CPACR_EL1_FPEN; - - if (cpus_have_final_cap(ARM64_SVE)) - val |= CPACR_EL1_ZEN; - if (cpus_have_final_cap(ARM64_SME)) - val |= CPACR_EL1_SMEN; - - write_sysreg(val, cpacr_el1); - } else { - u64 val = CPTR_NVHE_EL2_RES1; - - if (!cpus_have_final_cap(ARM64_SVE)) - val |= CPTR_EL2_TZ; - if (!cpus_have_final_cap(ARM64_SME)) - val |= CPTR_EL2_TSM; - - write_sysreg(val, cptr_el2); - } -} - static void __activate_traps(struct kvm_vcpu *vcpu) { ___activate_traps(vcpu, vcpu->arch.hcr_el2); diff --git a/arch/arm64/kvm/hyp/vhe/switch.c b/arch/arm64/kvm/hyp/vhe/switch.c index c9b330dc2066..477f1580ffea 100644 --- a/arch/arm64/kvm/hyp/vhe/switch.c +++ b/arch/arm64/kvm/hyp/vhe/switch.c @@ -90,87 +90,6 @@ static u64 __compute_hcr(struct kvm_vcpu *vcpu) return hcr | (guest_hcr & ~NV_HCR_GUEST_EXCLUDE); } -static void __activate_cptr_traps(struct kvm_vcpu *vcpu) -{ - u64 cptr; - - /* - * With VHE (HCR.E2H == 1), accesses to CPACR_EL1 are routed to - * CPTR_EL2. In general, CPACR_EL1 has the same layout as CPTR_EL2, - * except for some missing controls, such as TAM. - * In this case, CPTR_EL2.TAM has the same position with or without - * VHE (HCR.E2H == 1) which allows us to use here the CPTR_EL2.TAM - * shift value for trapping the AMU accesses. - */ - u64 val = CPACR_EL1_TTA | CPTR_EL2_TAM; - - if (guest_owns_fp_regs()) { - val |= CPACR_EL1_FPEN; - if (vcpu_has_sve(vcpu)) - val |= CPACR_EL1_ZEN; - } else { - __activate_traps_fpsimd32(vcpu); - } - - if (!vcpu_has_nv(vcpu)) - goto write; - - /* - * The architecture is a bit crap (what a surprise): an EL2 guest - * writing to CPTR_EL2 via CPACR_EL1 can't set any of TCPAC or TTA, - * as they are RES0 in the guest's view. To work around it, trap the - * sucker using the very same bit it can't set... - */ - if (vcpu_el2_e2h_is_set(vcpu) && is_hyp_ctxt(vcpu)) - val |= CPTR_EL2_TCPAC; - - /* - * Layer the guest hypervisor's trap configuration on top of our own if - * we're in a nested context. - */ - if (is_hyp_ctxt(vcpu)) - goto write; - - cptr = vcpu_sanitised_cptr_el2(vcpu); - - /* - * Pay attention, there's some interesting detail here. - * - * The CPTR_EL2.xEN fields are 2 bits wide, although there are only two - * meaningful trap states when HCR_EL2.TGE = 0 (running a nested guest): - * - * - CPTR_EL2.xEN = x0, traps are enabled - * - CPTR_EL2.xEN = x1, traps are disabled - * - * In other words, bit[0] determines if guest accesses trap or not. In - * the interest of simplicity, clear the entire field if the guest - * hypervisor has traps enabled to dispel any illusion of something more - * complicated taking place. - */ - if (!(SYS_FIELD_GET(CPACR_EL1, FPEN, cptr) & BIT(0))) - val &= ~CPACR_EL1_FPEN; - if (!(SYS_FIELD_GET(CPACR_EL1, ZEN, cptr) & BIT(0))) - val &= ~CPACR_EL1_ZEN; - - if (kvm_has_feat(vcpu->kvm, ID_AA64MMFR3_EL1, S2POE, IMP)) - val |= cptr & CPACR_EL1_E0POE; - - val |= cptr & CPTR_EL2_TCPAC; - -write: - write_sysreg(val, cpacr_el1); -} - -static void __deactivate_cptr_traps(struct kvm_vcpu *vcpu) -{ - u64 val = CPACR_EL1_FPEN | CPACR_EL1_ZEN_EL1EN; - - if (cpus_have_final_cap(ARM64_SME)) - val |= CPACR_EL1_SMEN_EL1EN; - - write_sysreg(val, cpacr_el1); -} - static void __activate_traps(struct kvm_vcpu *vcpu) { u64 val; @@ -223,9 +142,9 @@ static void __deactivate_traps(struct kvm_vcpu *vcpu) */ val = read_sysreg_el0(SYS_CNTP_CVAL); if (map.direct_ptimer == vcpu_ptimer(vcpu)) - __vcpu_sys_reg(vcpu, CNTP_CVAL_EL0) = val; + __vcpu_assign_sys_reg(vcpu, CNTP_CVAL_EL0, val); if (map.direct_ptimer == vcpu_hptimer(vcpu)) - __vcpu_sys_reg(vcpu, CNTHP_CVAL_EL2) = val; + __vcpu_assign_sys_reg(vcpu, CNTHP_CVAL_EL2, val); offset = read_sysreg_s(SYS_CNTPOFF_EL2); @@ -639,10 +558,10 @@ static int __kvm_vcpu_run_vhe(struct kvm_vcpu *vcpu) host_ctxt = host_data_ptr(host_ctxt); guest_ctxt = &vcpu->arch.ctxt; - sysreg_save_host_state_vhe(host_ctxt); - fpsimd_lazy_switch_to_guest(vcpu); + sysreg_save_host_state_vhe(host_ctxt); + /* * Note that ARM erratum 1165522 requires us to configure both stage 1 * and stage 2 translation for the guest context before we clear @@ -667,15 +586,23 @@ static int __kvm_vcpu_run_vhe(struct kvm_vcpu *vcpu) __deactivate_traps(vcpu); - fpsimd_lazy_switch_to_host(vcpu); - sysreg_restore_host_state_vhe(host_ctxt); + __debug_switch_to_host(vcpu); + + /* + * Ensure that all system register writes above have taken effect + * before returning to the host. In VHE mode, CPTR traps for + * FPSIMD/SVE/SME also apply to EL2, so FPSIMD/SVE/SME state must be + * manipulated after the ISB. + */ + isb(); + + fpsimd_lazy_switch_to_host(vcpu); + if (guest_owns_fp_regs()) __fpsimd_save_fpexc32(vcpu); - __debug_switch_to_host(vcpu); - return exit_code; } NOKPROBE_SYMBOL(__kvm_vcpu_run_vhe); @@ -705,12 +632,6 @@ int __kvm_vcpu_run(struct kvm_vcpu *vcpu) */ local_daif_restore(DAIF_PROCCTX_NOIRQ); - /* - * When we exit from the guest we change a number of CPU configuration - * parameters, such as traps. We rely on the isb() in kvm_call_hyp*() - * to make sure these changes take effect before running the host or - * additional guests. - */ return ret; } diff --git a/arch/arm64/kvm/hyp/vhe/sysreg-sr.c b/arch/arm64/kvm/hyp/vhe/sysreg-sr.c index 3814b0b2c937..73e4bc7fde9e 100644 --- a/arch/arm64/kvm/hyp/vhe/sysreg-sr.c +++ b/arch/arm64/kvm/hyp/vhe/sysreg-sr.c @@ -18,17 +18,17 @@ static void __sysreg_save_vel2_state(struct kvm_vcpu *vcpu) { /* These registers are common with EL1 */ - __vcpu_sys_reg(vcpu, PAR_EL1) = read_sysreg(par_el1); - __vcpu_sys_reg(vcpu, TPIDR_EL1) = read_sysreg(tpidr_el1); - - __vcpu_sys_reg(vcpu, ESR_EL2) = read_sysreg_el1(SYS_ESR); - __vcpu_sys_reg(vcpu, AFSR0_EL2) = read_sysreg_el1(SYS_AFSR0); - __vcpu_sys_reg(vcpu, AFSR1_EL2) = read_sysreg_el1(SYS_AFSR1); - __vcpu_sys_reg(vcpu, FAR_EL2) = read_sysreg_el1(SYS_FAR); - __vcpu_sys_reg(vcpu, MAIR_EL2) = read_sysreg_el1(SYS_MAIR); - __vcpu_sys_reg(vcpu, VBAR_EL2) = read_sysreg_el1(SYS_VBAR); - __vcpu_sys_reg(vcpu, CONTEXTIDR_EL2) = read_sysreg_el1(SYS_CONTEXTIDR); - __vcpu_sys_reg(vcpu, AMAIR_EL2) = read_sysreg_el1(SYS_AMAIR); + __vcpu_assign_sys_reg(vcpu, PAR_EL1, read_sysreg(par_el1)); + __vcpu_assign_sys_reg(vcpu, TPIDR_EL1, read_sysreg(tpidr_el1)); + + __vcpu_assign_sys_reg(vcpu, ESR_EL2, read_sysreg_el1(SYS_ESR)); + __vcpu_assign_sys_reg(vcpu, AFSR0_EL2, read_sysreg_el1(SYS_AFSR0)); + __vcpu_assign_sys_reg(vcpu, AFSR1_EL2, read_sysreg_el1(SYS_AFSR1)); + __vcpu_assign_sys_reg(vcpu, FAR_EL2, read_sysreg_el1(SYS_FAR)); + __vcpu_assign_sys_reg(vcpu, MAIR_EL2, read_sysreg_el1(SYS_MAIR)); + __vcpu_assign_sys_reg(vcpu, VBAR_EL2, read_sysreg_el1(SYS_VBAR)); + __vcpu_assign_sys_reg(vcpu, CONTEXTIDR_EL2, read_sysreg_el1(SYS_CONTEXTIDR)); + __vcpu_assign_sys_reg(vcpu, AMAIR_EL2, read_sysreg_el1(SYS_AMAIR)); /* * In VHE mode those registers are compatible between EL1 and EL2, @@ -46,21 +46,21 @@ static void __sysreg_save_vel2_state(struct kvm_vcpu *vcpu) * are always trapped, ensuring that the in-memory * copy is always up-to-date. A small blessing... */ - __vcpu_sys_reg(vcpu, SCTLR_EL2) = read_sysreg_el1(SYS_SCTLR); - __vcpu_sys_reg(vcpu, TTBR0_EL2) = read_sysreg_el1(SYS_TTBR0); - __vcpu_sys_reg(vcpu, TTBR1_EL2) = read_sysreg_el1(SYS_TTBR1); - __vcpu_sys_reg(vcpu, TCR_EL2) = read_sysreg_el1(SYS_TCR); + __vcpu_assign_sys_reg(vcpu, SCTLR_EL2, read_sysreg_el1(SYS_SCTLR)); + __vcpu_assign_sys_reg(vcpu, TTBR0_EL2, read_sysreg_el1(SYS_TTBR0)); + __vcpu_assign_sys_reg(vcpu, TTBR1_EL2, read_sysreg_el1(SYS_TTBR1)); + __vcpu_assign_sys_reg(vcpu, TCR_EL2, read_sysreg_el1(SYS_TCR)); if (ctxt_has_tcrx(&vcpu->arch.ctxt)) { - __vcpu_sys_reg(vcpu, TCR2_EL2) = read_sysreg_el1(SYS_TCR2); + __vcpu_assign_sys_reg(vcpu, TCR2_EL2, read_sysreg_el1(SYS_TCR2)); if (ctxt_has_s1pie(&vcpu->arch.ctxt)) { - __vcpu_sys_reg(vcpu, PIRE0_EL2) = read_sysreg_el1(SYS_PIRE0); - __vcpu_sys_reg(vcpu, PIR_EL2) = read_sysreg_el1(SYS_PIR); + __vcpu_assign_sys_reg(vcpu, PIRE0_EL2, read_sysreg_el1(SYS_PIRE0)); + __vcpu_assign_sys_reg(vcpu, PIR_EL2, read_sysreg_el1(SYS_PIR)); } if (ctxt_has_s1poe(&vcpu->arch.ctxt)) - __vcpu_sys_reg(vcpu, POR_EL2) = read_sysreg_el1(SYS_POR); + __vcpu_assign_sys_reg(vcpu, POR_EL2, read_sysreg_el1(SYS_POR)); } /* @@ -70,13 +70,13 @@ static void __sysreg_save_vel2_state(struct kvm_vcpu *vcpu) */ val = read_sysreg_el1(SYS_CNTKCTL); val &= CNTKCTL_VALID_BITS; - __vcpu_sys_reg(vcpu, CNTHCTL_EL2) &= ~CNTKCTL_VALID_BITS; - __vcpu_sys_reg(vcpu, CNTHCTL_EL2) |= val; + __vcpu_rmw_sys_reg(vcpu, CNTHCTL_EL2, &=, ~CNTKCTL_VALID_BITS); + __vcpu_rmw_sys_reg(vcpu, CNTHCTL_EL2, |=, val); } - __vcpu_sys_reg(vcpu, SP_EL2) = read_sysreg(sp_el1); - __vcpu_sys_reg(vcpu, ELR_EL2) = read_sysreg_el1(SYS_ELR); - __vcpu_sys_reg(vcpu, SPSR_EL2) = read_sysreg_el1(SYS_SPSR); + __vcpu_assign_sys_reg(vcpu, SP_EL2, read_sysreg(sp_el1)); + __vcpu_assign_sys_reg(vcpu, ELR_EL2, read_sysreg_el1(SYS_ELR)); + __vcpu_assign_sys_reg(vcpu, SPSR_EL2, read_sysreg_el1(SYS_SPSR)); } static void __sysreg_restore_vel2_state(struct kvm_vcpu *vcpu) diff --git a/arch/arm64/kvm/hypercalls.c b/arch/arm64/kvm/hypercalls.c index 569941eeb3fe..58c5fe7d7572 100644 --- a/arch/arm64/kvm/hypercalls.c +++ b/arch/arm64/kvm/hypercalls.c @@ -270,6 +270,7 @@ int kvm_smccc_call_handler(struct kvm_vcpu *vcpu) u32 feature; u8 action; gpa_t gpa; + uuid_t uuid; action = kvm_smccc_get_action(vcpu, func_id); switch (action) { @@ -355,10 +356,11 @@ int kvm_smccc_call_handler(struct kvm_vcpu *vcpu) val[0] = gpa; break; case ARM_SMCCC_VENDOR_HYP_CALL_UID_FUNC_ID: - val[0] = ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_0; - val[1] = ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_1; - val[2] = ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_2; - val[3] = ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_3; + uuid = ARM_SMCCC_VENDOR_HYP_UID_KVM; + val[0] = smccc_uuid_to_reg(&uuid, 0); + val[1] = smccc_uuid_to_reg(&uuid, 1); + val[2] = smccc_uuid_to_reg(&uuid, 2); + val[3] = smccc_uuid_to_reg(&uuid, 3); break; case ARM_SMCCC_VENDOR_HYP_KVM_FEATURES_FUNC_ID: val[0] = smccc_feat->vendor_hyp_bmap; diff --git a/arch/arm64/kvm/nested.c b/arch/arm64/kvm/nested.c index 291dbe38eb5c..dc1d26559bfa 100644 --- a/arch/arm64/kvm/nested.c +++ b/arch/arm64/kvm/nested.c @@ -918,6 +918,8 @@ static void invalidate_vncr_va(struct kvm *kvm, } } +#define tlbi_va_s1_to_va(v) (u64)sign_extend64((v) << 12, 48) + static void compute_s1_tlbi_range(struct kvm_vcpu *vcpu, u32 inst, u64 val, struct s1e2_tlbi_scope *scope) { @@ -964,7 +966,7 @@ static void compute_s1_tlbi_range(struct kvm_vcpu *vcpu, u32 inst, u64 val, scope->size = ttl_to_size(FIELD_GET(TLBI_TTL_MASK, val)); if (!scope->size) scope->size = SZ_1G; - scope->va = (val << 12) & ~(scope->size - 1); + scope->va = tlbi_va_s1_to_va(val) & ~(scope->size - 1); scope->asid = FIELD_GET(TLBIR_ASID_MASK, val); break; case OP_TLBI_ASIDE1: @@ -992,7 +994,7 @@ static void compute_s1_tlbi_range(struct kvm_vcpu *vcpu, u32 inst, u64 val, scope->size = ttl_to_size(FIELD_GET(TLBI_TTL_MASK, val)); if (!scope->size) scope->size = SZ_1G; - scope->va = (val << 12) & ~(scope->size - 1); + scope->va = tlbi_va_s1_to_va(val) & ~(scope->size - 1); break; case OP_TLBI_RVAE2: case OP_TLBI_RVAE2IS: @@ -1400,6 +1402,21 @@ static void kvm_map_l1_vncr(struct kvm_vcpu *vcpu) } } +#define has_tgran_2(__r, __sz) \ + ({ \ + u64 _s1, _s2, _mmfr0 = __r; \ + \ + _s2 = SYS_FIELD_GET(ID_AA64MMFR0_EL1, \ + TGRAN##__sz##_2, _mmfr0); \ + \ + _s1 = SYS_FIELD_GET(ID_AA64MMFR0_EL1, \ + TGRAN##__sz, _mmfr0); \ + \ + ((_s2 != ID_AA64MMFR0_EL1_TGRAN##__sz##_2_NI && \ + _s2 != ID_AA64MMFR0_EL1_TGRAN##__sz##_2_TGRAN##__sz) || \ + (_s2 == ID_AA64MMFR0_EL1_TGRAN##__sz##_2_TGRAN##__sz && \ + _s1 != ID_AA64MMFR0_EL1_TGRAN##__sz##_NI)); \ + }) /* * Our emulated CPU doesn't support all the possible features. For the * sake of simplicity (and probably mental sanity), wipe out a number @@ -1409,6 +1426,8 @@ static void kvm_map_l1_vncr(struct kvm_vcpu *vcpu) */ u64 limit_nv_id_reg(struct kvm *kvm, u32 reg, u64 val) { + u64 orig_val = val; + switch (reg) { case SYS_ID_AA64ISAR0_EL1: /* Support everything but TME */ @@ -1478,13 +1497,16 @@ u64 limit_nv_id_reg(struct kvm *kvm, u32 reg, u64 val) */ switch (PAGE_SIZE) { case SZ_4K: - val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN4_2, IMP); + if (has_tgran_2(orig_val, 4)) + val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN4_2, IMP); fallthrough; case SZ_16K: - val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN16_2, IMP); + if (has_tgran_2(orig_val, 16)) + val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN16_2, IMP); fallthrough; case SZ_64K: - val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN64_2, IMP); + if (has_tgran_2(orig_val, 64)) + val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN64_2, IMP); break; } @@ -1755,7 +1777,7 @@ int kvm_init_nv_sysregs(struct kvm_vcpu *vcpu) out: for (enum vcpu_sysreg sr = __SANITISED_REG_START__; sr < NR_SYS_REGS; sr++) - (void)__vcpu_sys_reg(vcpu, sr); + __vcpu_rmw_sys_reg(vcpu, sr, |=, 0); return 0; } diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c index 25c29107f13f..b03dbda7f1ab 100644 --- a/arch/arm64/kvm/pmu-emul.c +++ b/arch/arm64/kvm/pmu-emul.c @@ -178,7 +178,7 @@ static void kvm_pmu_set_pmc_value(struct kvm_pmc *pmc, u64 val, bool force) val |= lower_32_bits(val); } - __vcpu_sys_reg(vcpu, reg) = val; + __vcpu_assign_sys_reg(vcpu, reg, val); /* Recreate the perf event to reflect the updated sample_period */ kvm_pmu_create_perf_event(pmc); @@ -204,7 +204,7 @@ void kvm_pmu_set_counter_value(struct kvm_vcpu *vcpu, u64 select_idx, u64 val) void kvm_pmu_set_counter_value_user(struct kvm_vcpu *vcpu, u64 select_idx, u64 val) { kvm_pmu_release_perf_event(kvm_vcpu_idx_to_pmc(vcpu, select_idx)); - __vcpu_sys_reg(vcpu, counter_index_to_reg(select_idx)) = val; + __vcpu_assign_sys_reg(vcpu, counter_index_to_reg(select_idx), val); kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu); } @@ -239,7 +239,7 @@ static void kvm_pmu_stop_counter(struct kvm_pmc *pmc) reg = counter_index_to_reg(pmc->idx); - __vcpu_sys_reg(vcpu, reg) = val; + __vcpu_assign_sys_reg(vcpu, reg, val); kvm_pmu_release_perf_event(pmc); } @@ -503,14 +503,14 @@ static void kvm_pmu_counter_increment(struct kvm_vcpu *vcpu, reg = __vcpu_sys_reg(vcpu, counter_index_to_reg(i)) + 1; if (!kvm_pmc_is_64bit(pmc)) reg = lower_32_bits(reg); - __vcpu_sys_reg(vcpu, counter_index_to_reg(i)) = reg; + __vcpu_assign_sys_reg(vcpu, counter_index_to_reg(i), reg); /* No overflow? move on */ if (kvm_pmc_has_64bit_overflow(pmc) ? reg : lower_32_bits(reg)) continue; /* Mark overflow */ - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) |= BIT(i); + __vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, |=, BIT(i)); if (kvm_pmu_counter_can_chain(pmc)) kvm_pmu_counter_increment(vcpu, BIT(i + 1), @@ -556,7 +556,7 @@ static void kvm_pmu_perf_overflow(struct perf_event *perf_event, perf_event->attr.sample_period = period; perf_event->hw.sample_period = period; - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) |= BIT(idx); + __vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, |=, BIT(idx)); if (kvm_pmu_counter_can_chain(pmc)) kvm_pmu_counter_increment(vcpu, BIT(idx + 1), @@ -602,7 +602,7 @@ void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u64 val) kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu); /* The reset bits don't indicate any state, and shouldn't be saved. */ - __vcpu_sys_reg(vcpu, PMCR_EL0) = val & ~(ARMV8_PMU_PMCR_C | ARMV8_PMU_PMCR_P); + __vcpu_assign_sys_reg(vcpu, PMCR_EL0, (val & ~(ARMV8_PMU_PMCR_C | ARMV8_PMU_PMCR_P))); if (val & ARMV8_PMU_PMCR_C) kvm_pmu_set_counter_value(vcpu, ARMV8_PMU_CYCLE_IDX, 0); @@ -779,7 +779,7 @@ void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data, u64 reg; reg = counter_index_to_evtreg(pmc->idx); - __vcpu_sys_reg(vcpu, reg) = data & kvm_pmu_evtyper_mask(vcpu->kvm); + __vcpu_assign_sys_reg(vcpu, reg, (data & kvm_pmu_evtyper_mask(vcpu->kvm))); kvm_pmu_create_perf_event(pmc); } @@ -914,9 +914,9 @@ void kvm_vcpu_reload_pmu(struct kvm_vcpu *vcpu) { u64 mask = kvm_pmu_implemented_counter_mask(vcpu); - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) &= mask; - __vcpu_sys_reg(vcpu, PMINTENSET_EL1) &= mask; - __vcpu_sys_reg(vcpu, PMCNTENSET_EL0) &= mask; + __vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, &=, mask); + __vcpu_rmw_sys_reg(vcpu, PMINTENSET_EL1, &=, mask); + __vcpu_rmw_sys_reg(vcpu, PMCNTENSET_EL0, &=, mask); kvm_pmu_reprogram_counter_mask(vcpu, mask); } @@ -1038,7 +1038,7 @@ static void kvm_arm_set_nr_counters(struct kvm *kvm, unsigned int nr) u64 val = __vcpu_sys_reg(vcpu, MDCR_EL2); val &= ~MDCR_EL2_HPMN; val |= FIELD_PREP(MDCR_EL2_HPMN, kvm->arch.nr_pmu_counters); - __vcpu_sys_reg(vcpu, MDCR_EL2) = val; + __vcpu_assign_sys_reg(vcpu, MDCR_EL2, val); } } } diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index a6cf2888d150..c20bd6f21e60 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -228,7 +228,7 @@ void vcpu_write_sys_reg(struct kvm_vcpu *vcpu, u64 val, int reg) * to reverse-translate virtual EL2 system registers for a * non-VHE guest hypervisor. */ - __vcpu_sys_reg(vcpu, reg) = val; + __vcpu_assign_sys_reg(vcpu, reg, val); switch (reg) { case CNTHCTL_EL2: @@ -263,7 +263,7 @@ void vcpu_write_sys_reg(struct kvm_vcpu *vcpu, u64 val, int reg) return; memory_write: - __vcpu_sys_reg(vcpu, reg) = val; + __vcpu_assign_sys_reg(vcpu, reg, val); } /* CSSELR values; used to index KVM_REG_ARM_DEMUX_ID_CCSIDR */ @@ -605,7 +605,7 @@ static int set_oslsr_el1(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd, if ((val ^ rd->val) & ~OSLSR_EL1_OSLK) return -EINVAL; - __vcpu_sys_reg(vcpu, rd->reg) = val; + __vcpu_assign_sys_reg(vcpu, rd->reg, val); return 0; } @@ -791,7 +791,7 @@ static u64 reset_pmu_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) mask |= GENMASK(n - 1, 0); reset_unknown(vcpu, r); - __vcpu_sys_reg(vcpu, r->reg) &= mask; + __vcpu_rmw_sys_reg(vcpu, r->reg, &=, mask); return __vcpu_sys_reg(vcpu, r->reg); } @@ -799,7 +799,7 @@ static u64 reset_pmu_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) static u64 reset_pmevcntr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { reset_unknown(vcpu, r); - __vcpu_sys_reg(vcpu, r->reg) &= GENMASK(31, 0); + __vcpu_rmw_sys_reg(vcpu, r->reg, &=, GENMASK(31, 0)); return __vcpu_sys_reg(vcpu, r->reg); } @@ -811,7 +811,7 @@ static u64 reset_pmevtyper(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) return 0; reset_unknown(vcpu, r); - __vcpu_sys_reg(vcpu, r->reg) &= kvm_pmu_evtyper_mask(vcpu->kvm); + __vcpu_rmw_sys_reg(vcpu, r->reg, &=, kvm_pmu_evtyper_mask(vcpu->kvm)); return __vcpu_sys_reg(vcpu, r->reg); } @@ -819,7 +819,7 @@ static u64 reset_pmevtyper(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) static u64 reset_pmselr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { reset_unknown(vcpu, r); - __vcpu_sys_reg(vcpu, r->reg) &= PMSELR_EL0_SEL_MASK; + __vcpu_rmw_sys_reg(vcpu, r->reg, &=, PMSELR_EL0_SEL_MASK); return __vcpu_sys_reg(vcpu, r->reg); } @@ -835,7 +835,7 @@ static u64 reset_pmcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) * The value of PMCR.N field is included when the * vCPU register is read via kvm_vcpu_read_pmcr(). */ - __vcpu_sys_reg(vcpu, r->reg) = pmcr; + __vcpu_assign_sys_reg(vcpu, r->reg, pmcr); return __vcpu_sys_reg(vcpu, r->reg); } @@ -907,7 +907,7 @@ static bool access_pmselr(struct kvm_vcpu *vcpu, struct sys_reg_params *p, return false; if (p->is_write) - __vcpu_sys_reg(vcpu, PMSELR_EL0) = p->regval; + __vcpu_assign_sys_reg(vcpu, PMSELR_EL0, p->regval); else /* return PMSELR.SEL field */ p->regval = __vcpu_sys_reg(vcpu, PMSELR_EL0) @@ -1076,7 +1076,7 @@ static int set_pmreg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r, u64 va { u64 mask = kvm_pmu_accessible_counter_mask(vcpu); - __vcpu_sys_reg(vcpu, r->reg) = val & mask; + __vcpu_assign_sys_reg(vcpu, r->reg, val & mask); kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu); return 0; @@ -1103,10 +1103,10 @@ static bool access_pmcnten(struct kvm_vcpu *vcpu, struct sys_reg_params *p, val = p->regval & mask; if (r->Op2 & 0x1) /* accessing PMCNTENSET_EL0 */ - __vcpu_sys_reg(vcpu, PMCNTENSET_EL0) |= val; + __vcpu_rmw_sys_reg(vcpu, PMCNTENSET_EL0, |=, val); else /* accessing PMCNTENCLR_EL0 */ - __vcpu_sys_reg(vcpu, PMCNTENSET_EL0) &= ~val; + __vcpu_rmw_sys_reg(vcpu, PMCNTENSET_EL0, &=, ~val); kvm_pmu_reprogram_counter_mask(vcpu, val); } else { @@ -1129,10 +1129,10 @@ static bool access_pminten(struct kvm_vcpu *vcpu, struct sys_reg_params *p, if (r->Op2 & 0x1) /* accessing PMINTENSET_EL1 */ - __vcpu_sys_reg(vcpu, PMINTENSET_EL1) |= val; + __vcpu_rmw_sys_reg(vcpu, PMINTENSET_EL1, |=, val); else /* accessing PMINTENCLR_EL1 */ - __vcpu_sys_reg(vcpu, PMINTENSET_EL1) &= ~val; + __vcpu_rmw_sys_reg(vcpu, PMINTENSET_EL1, &=, ~val); } else { p->regval = __vcpu_sys_reg(vcpu, PMINTENSET_EL1); } @@ -1151,10 +1151,10 @@ static bool access_pmovs(struct kvm_vcpu *vcpu, struct sys_reg_params *p, if (p->is_write) { if (r->CRm & 0x2) /* accessing PMOVSSET_EL0 */ - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) |= (p->regval & mask); + __vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, |=, (p->regval & mask)); else /* accessing PMOVSCLR_EL0 */ - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) &= ~(p->regval & mask); + __vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, &=, ~(p->regval & mask)); } else { p->regval = __vcpu_sys_reg(vcpu, PMOVSSET_EL0); } @@ -1185,8 +1185,8 @@ static bool access_pmuserenr(struct kvm_vcpu *vcpu, struct sys_reg_params *p, if (!vcpu_mode_priv(vcpu)) return undef_access(vcpu, p, r); - __vcpu_sys_reg(vcpu, PMUSERENR_EL0) = - p->regval & ARMV8_PMU_USERENR_MASK; + __vcpu_assign_sys_reg(vcpu, PMUSERENR_EL0, + (p->regval & ARMV8_PMU_USERENR_MASK)); } else { p->regval = __vcpu_sys_reg(vcpu, PMUSERENR_EL0) & ARMV8_PMU_USERENR_MASK; @@ -1237,7 +1237,7 @@ static int set_pmcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r, if (!kvm_supports_32bit_el0()) val |= ARMV8_PMU_PMCR_LC; - __vcpu_sys_reg(vcpu, r->reg) = val; + __vcpu_assign_sys_reg(vcpu, r->reg, val); kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu); return 0; @@ -2213,7 +2213,7 @@ static u64 reset_clidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) if (kvm_has_mte(vcpu->kvm)) clidr |= 2ULL << CLIDR_TTYPE_SHIFT(loc); - __vcpu_sys_reg(vcpu, r->reg) = clidr; + __vcpu_assign_sys_reg(vcpu, r->reg, clidr); return __vcpu_sys_reg(vcpu, r->reg); } @@ -2227,7 +2227,7 @@ static int set_clidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd, if ((val & CLIDR_EL1_RES0) || (!(ctr_el0 & CTR_EL0_IDC) && idc)) return -EINVAL; - __vcpu_sys_reg(vcpu, rd->reg) = val; + __vcpu_assign_sys_reg(vcpu, rd->reg, val); return 0; } @@ -2404,7 +2404,7 @@ static bool access_sp_el1(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { if (p->is_write) - __vcpu_sys_reg(vcpu, SP_EL1) = p->regval; + __vcpu_assign_sys_reg(vcpu, SP_EL1, p->regval); else p->regval = __vcpu_sys_reg(vcpu, SP_EL1); @@ -2428,7 +2428,7 @@ static bool access_spsr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { if (p->is_write) - __vcpu_sys_reg(vcpu, SPSR_EL1) = p->regval; + __vcpu_assign_sys_reg(vcpu, SPSR_EL1, p->regval); else p->regval = __vcpu_sys_reg(vcpu, SPSR_EL1); @@ -2440,7 +2440,7 @@ static bool access_cntkctl_el12(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { if (p->is_write) - __vcpu_sys_reg(vcpu, CNTKCTL_EL1) = p->regval; + __vcpu_assign_sys_reg(vcpu, CNTKCTL_EL1, p->regval); else p->regval = __vcpu_sys_reg(vcpu, CNTKCTL_EL1); @@ -2454,7 +2454,9 @@ static u64 reset_hcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) if (!cpus_have_final_cap(ARM64_HAS_HCR_NV1)) val |= HCR_E2H; - return __vcpu_sys_reg(vcpu, r->reg) = val; + __vcpu_assign_sys_reg(vcpu, r->reg, val); + + return __vcpu_sys_reg(vcpu, r->reg); } static unsigned int __el2_visibility(const struct kvm_vcpu *vcpu, @@ -2622,10 +2624,10 @@ static bool access_mdcr(struct kvm_vcpu *vcpu, */ if (hpmn > vcpu->kvm->arch.nr_pmu_counters) { hpmn = vcpu->kvm->arch.nr_pmu_counters; - u64_replace_bits(val, hpmn, MDCR_EL2_HPMN); + u64p_replace_bits(&val, hpmn, MDCR_EL2_HPMN); } - __vcpu_sys_reg(vcpu, MDCR_EL2) = val; + __vcpu_assign_sys_reg(vcpu, MDCR_EL2, val); /* * Request a reload of the PMU to enable/disable the counters @@ -2754,7 +2756,7 @@ static int set_imp_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r, static u64 reset_mdcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { - __vcpu_sys_reg(vcpu, r->reg) = vcpu->kvm->arch.nr_pmu_counters; + __vcpu_assign_sys_reg(vcpu, r->reg, vcpu->kvm->arch.nr_pmu_counters); return vcpu->kvm->arch.nr_pmu_counters; } @@ -4790,7 +4792,7 @@ void kvm_reset_sys_regs(struct kvm_vcpu *vcpu) r->reset(vcpu, r); if (r->reg >= __SANITISED_REG_START__ && r->reg < NR_SYS_REGS) - (void)__vcpu_sys_reg(vcpu, r->reg); + __vcpu_rmw_sys_reg(vcpu, r->reg, |=, 0); } set_bit(KVM_ARCH_FLAG_ID_REGS_INITIALIZED, &kvm->arch.flags); @@ -5012,7 +5014,7 @@ int kvm_sys_reg_set_user(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg, if (r->set_user) { ret = (r->set_user)(vcpu, r, val); } else { - __vcpu_sys_reg(vcpu, r->reg) = val; + __vcpu_assign_sys_reg(vcpu, r->reg, val); ret = 0; } diff --git a/arch/arm64/kvm/sys_regs.h b/arch/arm64/kvm/sys_regs.h index cc6338d38766..ef97d9fc67cc 100644 --- a/arch/arm64/kvm/sys_regs.h +++ b/arch/arm64/kvm/sys_regs.h @@ -137,7 +137,7 @@ static inline u64 reset_unknown(struct kvm_vcpu *vcpu, { BUG_ON(!r->reg); BUG_ON(r->reg >= NR_SYS_REGS); - __vcpu_sys_reg(vcpu, r->reg) = 0x1de7ec7edbadc0deULL; + __vcpu_assign_sys_reg(vcpu, r->reg, 0x1de7ec7edbadc0deULL); return __vcpu_sys_reg(vcpu, r->reg); } @@ -145,7 +145,7 @@ static inline u64 reset_val(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { BUG_ON(!r->reg); BUG_ON(r->reg >= NR_SYS_REGS); - __vcpu_sys_reg(vcpu, r->reg) = r->val; + __vcpu_assign_sys_reg(vcpu, r->reg, r->val); return __vcpu_sys_reg(vcpu, r->reg); } diff --git a/arch/arm64/kvm/vgic/vgic-debug.c b/arch/arm64/kvm/vgic/vgic-debug.c index f8425f381de9..2684f273d9e1 100644 --- a/arch/arm64/kvm/vgic/vgic-debug.c +++ b/arch/arm64/kvm/vgic/vgic-debug.c @@ -490,6 +490,9 @@ static int vgic_its_debug_show(struct seq_file *s, void *v) struct its_device *dev = iter->dev; struct its_ite *ite = iter->ite; + if (!ite) + return 0; + if (list_is_first(&ite->ite_list, &dev->itt_head)) { seq_printf(s, "\n"); seq_printf(s, "Device ID: 0x%x, Event ID Range: [0 - %llu]\n", @@ -498,7 +501,7 @@ static int vgic_its_debug_show(struct seq_file *s, void *v) seq_printf(s, "-----------------------------------------------\n"); } - if (ite && ite->irq && ite->collection) { + if (ite->irq && ite->collection) { seq_printf(s, "%8u %8u %8u %8u %8u %2d\n", ite->event_id, ite->irq->intid, ite->irq->hwintid, ite->collection->target_addr, diff --git a/arch/arm64/kvm/vgic/vgic-init.c b/arch/arm64/kvm/vgic/vgic-init.c index 1f33e71c2a73..eb1205654ac8 100644 --- a/arch/arm64/kvm/vgic/vgic-init.c +++ b/arch/arm64/kvm/vgic/vgic-init.c @@ -84,15 +84,40 @@ int kvm_vgic_create(struct kvm *kvm, u32 type) !kvm_vgic_global_state.can_emulate_gicv2) return -ENODEV; - /* Must be held to avoid race with vCPU creation */ + /* + * Ensure mutual exclusion with vCPU creation and any vCPU ioctls by: + * + * - Holding kvm->lock to prevent KVM_CREATE_VCPU from reaching + * kvm_arch_vcpu_precreate() and ensuring created_vcpus is stable. + * This alone is insufficient, as kvm_vm_ioctl_create_vcpu() drops + * the kvm->lock before completing the vCPU creation. + */ lockdep_assert_held(&kvm->lock); + /* + * - Acquiring the vCPU mutex for every *online* vCPU to prevent + * concurrent vCPU ioctls for vCPUs already visible to userspace. + */ ret = -EBUSY; - if (!lock_all_vcpus(kvm)) + if (kvm_trylock_all_vcpus(kvm)) return ret; + /* + * - Taking the config_lock which protects VGIC data structures such + * as the per-vCPU arrays of private IRQs (SGIs, PPIs). + */ mutex_lock(&kvm->arch.config_lock); + /* + * - Bailing on the entire thing if a vCPU is in the middle of creation, + * dropped the kvm->lock, but hasn't reached kvm_arch_vcpu_create(). + * + * The whole combination of this guarantees that no vCPU can get into + * KVM with a VGIC configuration inconsistent with the VM's VGIC. + */ + if (kvm->created_vcpus != atomic_read(&kvm->online_vcpus)) + goto out_unlock; + if (irqchip_in_kernel(kvm)) { ret = -EEXIST; goto out_unlock; @@ -142,7 +167,7 @@ int kvm_vgic_create(struct kvm *kvm, u32 type) out_unlock: mutex_unlock(&kvm->arch.config_lock); - unlock_all_vcpus(kvm); + kvm_unlock_all_vcpus(kvm); return ret; } diff --git a/arch/arm64/kvm/vgic/vgic-its.c b/arch/arm64/kvm/vgic/vgic-its.c index 569f9da9049f..534049c7c94b 100644 --- a/arch/arm64/kvm/vgic/vgic-its.c +++ b/arch/arm64/kvm/vgic/vgic-its.c @@ -306,39 +306,34 @@ static int update_lpi_config(struct kvm *kvm, struct vgic_irq *irq, } } - raw_spin_unlock_irqrestore(&irq->irq_lock, flags); - if (irq->hw) - return its_prop_update_vlpi(irq->host_irq, prop, needs_inv); + ret = its_prop_update_vlpi(irq->host_irq, prop, needs_inv); - return 0; + raw_spin_unlock_irqrestore(&irq->irq_lock, flags); + return ret; } static int update_affinity(struct vgic_irq *irq, struct kvm_vcpu *vcpu) { - int ret = 0; - unsigned long flags; + struct its_vlpi_map map; + int ret; - raw_spin_lock_irqsave(&irq->irq_lock, flags); + guard(raw_spinlock_irqsave)(&irq->irq_lock); irq->target_vcpu = vcpu; - raw_spin_unlock_irqrestore(&irq->irq_lock, flags); - if (irq->hw) { - struct its_vlpi_map map; - - ret = its_get_vlpi(irq->host_irq, &map); - if (ret) - return ret; + if (!irq->hw) + return 0; - if (map.vpe) - atomic_dec(&map.vpe->vlpi_count); - map.vpe = &vcpu->arch.vgic_cpu.vgic_v3.its_vpe; - atomic_inc(&map.vpe->vlpi_count); + ret = its_get_vlpi(irq->host_irq, &map); + if (ret) + return ret; - ret = its_map_vlpi(irq->host_irq, &map); - } + if (map.vpe) + atomic_dec(&map.vpe->vlpi_count); - return ret; + map.vpe = &vcpu->arch.vgic_cpu.vgic_v3.its_vpe; + atomic_inc(&map.vpe->vlpi_count); + return its_map_vlpi(irq->host_irq, &map); } static struct kvm_vcpu *collection_to_vcpu(struct kvm *kvm, @@ -756,12 +751,17 @@ int vgic_its_inject_msi(struct kvm *kvm, struct kvm_msi *msi) /* Requires the its_lock to be held. */ static void its_free_ite(struct kvm *kvm, struct its_ite *ite) { + struct vgic_irq *irq = ite->irq; list_del(&ite->ite_list); /* This put matches the get in vgic_add_lpi. */ - if (ite->irq) { - if (ite->irq->hw) - WARN_ON(its_unmap_vlpi(ite->irq->host_irq)); + if (irq) { + scoped_guard(raw_spinlock_irqsave, &irq->irq_lock) { + if (irq->hw) + WARN_ON(its_unmap_vlpi(ite->irq->host_irq)); + + irq->hw = false; + } vgic_put_irq(kvm, ite->irq); } @@ -1971,7 +1971,7 @@ static int vgic_its_attr_regs_access(struct kvm_device *dev, mutex_lock(&dev->kvm->lock); - if (!lock_all_vcpus(dev->kvm)) { + if (kvm_trylock_all_vcpus(dev->kvm)) { mutex_unlock(&dev->kvm->lock); return -EBUSY; } @@ -2006,7 +2006,7 @@ static int vgic_its_attr_regs_access(struct kvm_device *dev, } out: mutex_unlock(&dev->kvm->arch.config_lock); - unlock_all_vcpus(dev->kvm); + kvm_unlock_all_vcpus(dev->kvm); mutex_unlock(&dev->kvm->lock); return ret; } @@ -2676,7 +2676,7 @@ static int vgic_its_ctrl(struct kvm *kvm, struct vgic_its *its, u64 attr) mutex_lock(&kvm->lock); - if (!lock_all_vcpus(kvm)) { + if (kvm_trylock_all_vcpus(kvm)) { mutex_unlock(&kvm->lock); return -EBUSY; } @@ -2698,7 +2698,7 @@ static int vgic_its_ctrl(struct kvm *kvm, struct vgic_its *its, u64 attr) mutex_unlock(&its->its_lock); mutex_unlock(&kvm->arch.config_lock); - unlock_all_vcpus(kvm); + kvm_unlock_all_vcpus(kvm); mutex_unlock(&kvm->lock); return ret; } diff --git a/arch/arm64/kvm/vgic/vgic-kvm-device.c b/arch/arm64/kvm/vgic/vgic-kvm-device.c index 359094f68c23..f9ae790163fb 100644 --- a/arch/arm64/kvm/vgic/vgic-kvm-device.c +++ b/arch/arm64/kvm/vgic/vgic-kvm-device.c @@ -268,7 +268,7 @@ static int vgic_set_common_attr(struct kvm_device *dev, return -ENXIO; mutex_lock(&dev->kvm->lock); - if (!lock_all_vcpus(dev->kvm)) { + if (kvm_trylock_all_vcpus(dev->kvm)) { mutex_unlock(&dev->kvm->lock); return -EBUSY; } @@ -276,7 +276,7 @@ static int vgic_set_common_attr(struct kvm_device *dev, mutex_lock(&dev->kvm->arch.config_lock); r = vgic_v3_save_pending_tables(dev->kvm); mutex_unlock(&dev->kvm->arch.config_lock); - unlock_all_vcpus(dev->kvm); + kvm_unlock_all_vcpus(dev->kvm); mutex_unlock(&dev->kvm->lock); return r; } @@ -390,7 +390,7 @@ static int vgic_v2_attr_regs_access(struct kvm_device *dev, mutex_lock(&dev->kvm->lock); - if (!lock_all_vcpus(dev->kvm)) { + if (kvm_trylock_all_vcpus(dev->kvm)) { mutex_unlock(&dev->kvm->lock); return -EBUSY; } @@ -415,7 +415,7 @@ static int vgic_v2_attr_regs_access(struct kvm_device *dev, out: mutex_unlock(&dev->kvm->arch.config_lock); - unlock_all_vcpus(dev->kvm); + kvm_unlock_all_vcpus(dev->kvm); mutex_unlock(&dev->kvm->lock); if (!ret && !is_write) @@ -554,7 +554,7 @@ static int vgic_v3_attr_regs_access(struct kvm_device *dev, mutex_lock(&dev->kvm->lock); - if (!lock_all_vcpus(dev->kvm)) { + if (kvm_trylock_all_vcpus(dev->kvm)) { mutex_unlock(&dev->kvm->lock); return -EBUSY; } @@ -611,7 +611,7 @@ static int vgic_v3_attr_regs_access(struct kvm_device *dev, out: mutex_unlock(&dev->kvm->arch.config_lock); - unlock_all_vcpus(dev->kvm); + kvm_unlock_all_vcpus(dev->kvm); mutex_unlock(&dev->kvm->lock); if (!ret && uaccess && !is_write) { diff --git a/arch/arm64/kvm/vgic/vgic-v3-nested.c b/arch/arm64/kvm/vgic/vgic-v3-nested.c index 4f6954c30674..679aafe77de2 100644 --- a/arch/arm64/kvm/vgic/vgic-v3-nested.c +++ b/arch/arm64/kvm/vgic/vgic-v3-nested.c @@ -36,6 +36,11 @@ struct shadow_if { static DEFINE_PER_CPU(struct shadow_if, shadow_if); +static int lr_map_idx_to_shadow_idx(struct shadow_if *shadow_if, int idx) +{ + return hweight16(shadow_if->lr_map & (BIT(idx) - 1)); +} + /* * Nesting GICv3 support * @@ -209,6 +214,29 @@ u64 vgic_v3_get_misr(struct kvm_vcpu *vcpu) return reg; } +static u64 translate_lr_pintid(struct kvm_vcpu *vcpu, u64 lr) +{ + struct vgic_irq *irq; + + if (!(lr & ICH_LR_HW)) + return lr; + + /* We have the HW bit set, check for validity of pINTID */ + irq = vgic_get_vcpu_irq(vcpu, FIELD_GET(ICH_LR_PHYS_ID_MASK, lr)); + /* If there was no real mapping, nuke the HW bit */ + if (!irq || !irq->hw || irq->intid > VGIC_MAX_SPI) + lr &= ~ICH_LR_HW; + + /* Translate the virtual mapping to the real one, even if invalid */ + if (irq) { + lr &= ~ICH_LR_PHYS_ID_MASK; + lr |= FIELD_PREP(ICH_LR_PHYS_ID_MASK, (u64)irq->hwintid); + vgic_put_irq(vcpu->kvm, irq); + } + + return lr; +} + /* * For LRs which have HW bit set such as timer interrupts, we modify them to * have the host hardware interrupt number instead of the virtual one programmed @@ -217,58 +245,37 @@ u64 vgic_v3_get_misr(struct kvm_vcpu *vcpu) static void vgic_v3_create_shadow_lr(struct kvm_vcpu *vcpu, struct vgic_v3_cpu_if *s_cpu_if) { - unsigned long lr_map = 0; - int index = 0; + struct shadow_if *shadow_if; + + shadow_if = container_of(s_cpu_if, struct shadow_if, cpuif); + shadow_if->lr_map = 0; for (int i = 0; i < kvm_vgic_global_state.nr_lr; i++) { u64 lr = __vcpu_sys_reg(vcpu, ICH_LRN(i)); - struct vgic_irq *irq; if (!(lr & ICH_LR_STATE)) - lr = 0; - - if (!(lr & ICH_LR_HW)) - goto next; - - /* We have the HW bit set, check for validity of pINTID */ - irq = vgic_get_vcpu_irq(vcpu, FIELD_GET(ICH_LR_PHYS_ID_MASK, lr)); - if (!irq || !irq->hw || irq->intid > VGIC_MAX_SPI ) { - /* There was no real mapping, so nuke the HW bit */ - lr &= ~ICH_LR_HW; - if (irq) - vgic_put_irq(vcpu->kvm, irq); - goto next; - } - - /* Translate the virtual mapping to the real one */ - lr &= ~ICH_LR_PHYS_ID_MASK; - lr |= FIELD_PREP(ICH_LR_PHYS_ID_MASK, (u64)irq->hwintid); + continue; - vgic_put_irq(vcpu->kvm, irq); + lr = translate_lr_pintid(vcpu, lr); -next: - s_cpu_if->vgic_lr[index] = lr; - if (lr) { - lr_map |= BIT(i); - index++; - } + s_cpu_if->vgic_lr[hweight16(shadow_if->lr_map)] = lr; + shadow_if->lr_map |= BIT(i); } - container_of(s_cpu_if, struct shadow_if, cpuif)->lr_map = lr_map; - s_cpu_if->used_lrs = index; + s_cpu_if->used_lrs = hweight16(shadow_if->lr_map); } void vgic_v3_sync_nested(struct kvm_vcpu *vcpu) { struct shadow_if *shadow_if = get_shadow_if(); - int i, index = 0; + int i; for_each_set_bit(i, &shadow_if->lr_map, kvm_vgic_global_state.nr_lr) { u64 lr = __vcpu_sys_reg(vcpu, ICH_LRN(i)); struct vgic_irq *irq; if (!(lr & ICH_LR_HW) || !(lr & ICH_LR_STATE)) - goto next; + continue; /* * If we had a HW lr programmed by the guest hypervisor, we @@ -277,15 +284,13 @@ void vgic_v3_sync_nested(struct kvm_vcpu *vcpu) */ irq = vgic_get_vcpu_irq(vcpu, FIELD_GET(ICH_LR_PHYS_ID_MASK, lr)); if (WARN_ON(!irq)) /* Shouldn't happen as we check on load */ - goto next; + continue; - lr = __gic_v3_get_lr(index); + lr = __gic_v3_get_lr(lr_map_idx_to_shadow_idx(shadow_if, i)); if (!(lr & ICH_LR_STATE)) irq->active = false; vgic_put_irq(vcpu->kvm, irq); - next: - index++; } } @@ -356,25 +361,23 @@ void vgic_v3_put_nested(struct kvm_vcpu *vcpu) val = __vcpu_sys_reg(vcpu, ICH_HCR_EL2); val &= ~ICH_HCR_EL2_EOIcount_MASK; val |= (s_cpu_if->vgic_hcr & ICH_HCR_EL2_EOIcount_MASK); - __vcpu_sys_reg(vcpu, ICH_HCR_EL2) = val; - __vcpu_sys_reg(vcpu, ICH_VMCR_EL2) = s_cpu_if->vgic_vmcr; + __vcpu_assign_sys_reg(vcpu, ICH_HCR_EL2, val); + __vcpu_assign_sys_reg(vcpu, ICH_VMCR_EL2, s_cpu_if->vgic_vmcr); for (i = 0; i < 4; i++) { - __vcpu_sys_reg(vcpu, ICH_AP0RN(i)) = s_cpu_if->vgic_ap0r[i]; - __vcpu_sys_reg(vcpu, ICH_AP1RN(i)) = s_cpu_if->vgic_ap1r[i]; + __vcpu_assign_sys_reg(vcpu, ICH_AP0RN(i), s_cpu_if->vgic_ap0r[i]); + __vcpu_assign_sys_reg(vcpu, ICH_AP1RN(i), s_cpu_if->vgic_ap1r[i]); } for_each_set_bit(i, &shadow_if->lr_map, kvm_vgic_global_state.nr_lr) { val = __vcpu_sys_reg(vcpu, ICH_LRN(i)); val &= ~ICH_LR_STATE; - val |= s_cpu_if->vgic_lr[i] & ICH_LR_STATE; + val |= s_cpu_if->vgic_lr[lr_map_idx_to_shadow_idx(shadow_if, i)] & ICH_LR_STATE; - __vcpu_sys_reg(vcpu, ICH_LRN(i)) = val; - s_cpu_if->vgic_lr[i] = 0; + __vcpu_assign_sys_reg(vcpu, ICH_LRN(i), val); } - shadow_if->lr_map = 0; vcpu->arch.vgic_cpu.vgic_v3.used_lrs = 0; } @@ -398,9 +401,7 @@ void vgic_v3_nested_update_mi(struct kvm_vcpu *vcpu) { bool level; - level = __vcpu_sys_reg(vcpu, ICH_HCR_EL2) & ICH_HCR_EL2_En; - if (level) - level &= vgic_v3_get_misr(vcpu); + level = (__vcpu_sys_reg(vcpu, ICH_HCR_EL2) & ICH_HCR_EL2_En) && vgic_v3_get_misr(vcpu); kvm_vgic_inject_irq(vcpu->kvm, vcpu, vcpu->kvm->arch.vgic.mi_intid, level, vcpu); } diff --git a/arch/arm64/kvm/vgic/vgic-v4.c b/arch/arm64/kvm/vgic/vgic-v4.c index c7de6154627c..193946108192 100644 --- a/arch/arm64/kvm/vgic/vgic-v4.c +++ b/arch/arm64/kvm/vgic/vgic-v4.c @@ -444,7 +444,7 @@ int kvm_vgic_v4_set_forwarding(struct kvm *kvm, int virq, if (IS_ERR(its)) return 0; - mutex_lock(&its->its_lock); + guard(mutex)(&its->its_lock); /* * Perform the actual DevID/EventID -> LPI translation. @@ -455,11 +455,13 @@ int kvm_vgic_v4_set_forwarding(struct kvm *kvm, int virq, */ if (vgic_its_resolve_lpi(kvm, its, irq_entry->msi.devid, irq_entry->msi.data, &irq)) - goto out; + return 0; + + raw_spin_lock_irqsave(&irq->irq_lock, flags); /* Silently exit if the vLPI is already mapped */ if (irq->hw) - goto out; + goto out_unlock_irq; /* * Emit the mapping request. If it fails, the ITS probably @@ -479,68 +481,74 @@ int kvm_vgic_v4_set_forwarding(struct kvm *kvm, int virq, ret = its_map_vlpi(virq, &map); if (ret) - goto out; + goto out_unlock_irq; irq->hw = true; irq->host_irq = virq; atomic_inc(&map.vpe->vlpi_count); /* Transfer pending state */ - raw_spin_lock_irqsave(&irq->irq_lock, flags); - if (irq->pending_latch) { - ret = irq_set_irqchip_state(irq->host_irq, - IRQCHIP_STATE_PENDING, - irq->pending_latch); - WARN_RATELIMIT(ret, "IRQ %d", irq->host_irq); + if (!irq->pending_latch) + goto out_unlock_irq; - /* - * Clear pending_latch and communicate this state - * change via vgic_queue_irq_unlock. - */ - irq->pending_latch = false; - vgic_queue_irq_unlock(kvm, irq, flags); - } else { - raw_spin_unlock_irqrestore(&irq->irq_lock, flags); - } + ret = irq_set_irqchip_state(irq->host_irq, IRQCHIP_STATE_PENDING, + irq->pending_latch); + WARN_RATELIMIT(ret, "IRQ %d", irq->host_irq); -out: - mutex_unlock(&its->its_lock); + /* + * Clear pending_latch and communicate this state + * change via vgic_queue_irq_unlock. + */ + irq->pending_latch = false; + vgic_queue_irq_unlock(kvm, irq, flags); + return ret; + +out_unlock_irq: + raw_spin_unlock_irqrestore(&irq->irq_lock, flags); return ret; } -int kvm_vgic_v4_unset_forwarding(struct kvm *kvm, int virq, - struct kvm_kernel_irq_routing_entry *irq_entry) +static struct vgic_irq *__vgic_host_irq_get_vlpi(struct kvm *kvm, int host_irq) { - struct vgic_its *its; struct vgic_irq *irq; - int ret; + unsigned long idx; + + guard(rcu)(); + xa_for_each(&kvm->arch.vgic.lpi_xa, idx, irq) { + if (!irq->hw || irq->host_irq != host_irq) + continue; + + if (!vgic_try_get_irq_kref(irq)) + return NULL; + + return irq; + } + + return NULL; +} + +int kvm_vgic_v4_unset_forwarding(struct kvm *kvm, int host_irq) +{ + struct vgic_irq *irq; + unsigned long flags; + int ret = 0; if (!vgic_supports_direct_msis(kvm)) return 0; - /* - * Get the ITS, and escape early on error (not a valid - * doorbell for any of our vITSs). - */ - its = vgic_get_its(kvm, irq_entry); - if (IS_ERR(its)) + irq = __vgic_host_irq_get_vlpi(kvm, host_irq); + if (!irq) return 0; - mutex_lock(&its->its_lock); - - ret = vgic_its_resolve_lpi(kvm, its, irq_entry->msi.devid, - irq_entry->msi.data, &irq); - if (ret) - goto out; - - WARN_ON(irq->hw && irq->host_irq != virq); + raw_spin_lock_irqsave(&irq->irq_lock, flags); + WARN_ON(irq->hw && irq->host_irq != host_irq); if (irq->hw) { atomic_dec(&irq->target_vcpu->arch.vgic_cpu.vgic_v3.its_vpe.vlpi_count); irq->hw = false; - ret = its_unmap_vlpi(virq); + ret = its_unmap_vlpi(host_irq); } -out: - mutex_unlock(&its->its_lock); + raw_spin_unlock_irqrestore(&irq->irq_lock, flags); + vgic_put_irq(kvm, irq); return ret; } |