diff options
Diffstat (limited to 'target/i386/kvm/kvm.c')
| -rw-r--r-- | target/i386/kvm/kvm.c | 383 |
1 files changed, 249 insertions, 134 deletions
diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c index c8056ef83d..e6f94900f3 100644 --- a/target/i386/kvm/kvm.c +++ b/target/i386/kvm/kvm.c @@ -81,6 +81,16 @@ do { } while (0) #endif +/* + * On older Intel CPUs, KVM uses vm86 mode to emulate 16-bit code directly. + * In order to use vm86 mode, an EPT identity map and a TSS are needed. + * Since these must be part of guest physical memory, we need to allocate + * them, both by setting their start addresses in the kernel and by + * creating a corresponding e820 entry. We need 4 pages before the BIOS, + * so this value allows up to 16M BIOSes. + */ +#define KVM_IDENTITY_BASE 0xfeffc000 + /* From arch/x86/kvm/lapic.h */ #define KVM_APIC_BUS_CYCLE_NS 1 #define KVM_APIC_BUS_FREQUENCY (1000000000ULL / KVM_APIC_BUS_CYCLE_NS) @@ -92,7 +102,17 @@ * 255 kvm_msr_entry structs */ #define MSR_BUF_SIZE 4096 +typedef bool QEMURDMSRHandler(X86CPU *cpu, uint32_t msr, uint64_t *val); +typedef bool QEMUWRMSRHandler(X86CPU *cpu, uint32_t msr, uint64_t val); +typedef struct { + uint32_t msr; + QEMURDMSRHandler *rdmsr; + QEMUWRMSRHandler *wrmsr; +} KVMMSRHandlers; + static void kvm_init_msrs(X86CPU *cpu); +static bool kvm_filter_msr(KVMState *s, uint32_t msr, QEMURDMSRHandler *rdmsr, + QEMUWRMSRHandler *wrmsr); const KVMCapabilityInfo kvm_arch_required_capabilities[] = { KVM_CAP_INFO(SET_TSS_ADDR), @@ -2896,9 +2916,9 @@ static int kvm_msr_energy_thread_init(KVMState *s, MachineState *ms) * 1. Host cpu must be Intel cpu * 2. RAPL must be enabled on the Host */ - if (is_host_cpu_intel()) { - error_report("The RAPL feature can only be enabled on hosts\ - with Intel CPU models"); + if (!is_host_cpu_intel()) { + error_report("The RAPL feature can only be enabled on hosts " + "with Intel CPU models"); ret = 1; goto out; } @@ -2995,10 +3015,174 @@ int kvm_arch_get_default_type(MachineState *ms) return 0; } -int kvm_arch_init(MachineState *ms, KVMState *s) +static int kvm_vm_enable_exception_payload(KVMState *s) +{ + int ret = 0; + has_exception_payload = kvm_check_extension(s, KVM_CAP_EXCEPTION_PAYLOAD); + if (has_exception_payload) { + ret = kvm_vm_enable_cap(s, KVM_CAP_EXCEPTION_PAYLOAD, 0, true); + if (ret < 0) { + error_report("kvm: Failed to enable exception payload cap: %s", + strerror(-ret)); + } + } + + return ret; +} + +static int kvm_vm_enable_triple_fault_event(KVMState *s) +{ + int ret = 0; + has_triple_fault_event = \ + kvm_check_extension(s, + KVM_CAP_X86_TRIPLE_FAULT_EVENT); + if (has_triple_fault_event) { + ret = kvm_vm_enable_cap(s, KVM_CAP_X86_TRIPLE_FAULT_EVENT, 0, true); + if (ret < 0) { + error_report("kvm: Failed to enable triple fault event cap: %s", + strerror(-ret)); + } + } + return ret; +} + +static int kvm_vm_set_identity_map_addr(KVMState *s, uint64_t identity_base) +{ + return kvm_vm_ioctl(s, KVM_SET_IDENTITY_MAP_ADDR, &identity_base); +} + +static int kvm_vm_set_nr_mmu_pages(KVMState *s) { - uint64_t identity_base = 0xfffbc000; uint64_t shadow_mem; + int ret = 0; + shadow_mem = object_property_get_int(OBJECT(s), + "kvm-shadow-mem", + &error_abort); + if (shadow_mem != -1) { + shadow_mem /= 4096; + ret = kvm_vm_ioctl(s, KVM_SET_NR_MMU_PAGES, shadow_mem); + } + return ret; +} + +static int kvm_vm_set_tss_addr(KVMState *s, uint64_t tss_base) +{ + return kvm_vm_ioctl(s, KVM_SET_TSS_ADDR, tss_base); +} + +static int kvm_vm_enable_disable_exits(KVMState *s) +{ + int disable_exits = kvm_check_extension(s, KVM_CAP_X86_DISABLE_EXITS); +/* Work around for kernel header with a typo. TODO: fix header and drop. */ +#if defined(KVM_X86_DISABLE_EXITS_HTL) && !defined(KVM_X86_DISABLE_EXITS_HLT) +#define KVM_X86_DISABLE_EXITS_HLT KVM_X86_DISABLE_EXITS_HTL +#endif + if (disable_exits) { + disable_exits &= (KVM_X86_DISABLE_EXITS_MWAIT | + KVM_X86_DISABLE_EXITS_HLT | + KVM_X86_DISABLE_EXITS_PAUSE | + KVM_X86_DISABLE_EXITS_CSTATE); + } + + return kvm_vm_enable_cap(s, KVM_CAP_X86_DISABLE_EXITS, 0, + disable_exits); +} + +static int kvm_vm_enable_bus_lock_exit(KVMState *s) +{ + int ret = 0; + ret = kvm_check_extension(s, KVM_CAP_X86_BUS_LOCK_EXIT); + if (!(ret & KVM_BUS_LOCK_DETECTION_EXIT)) { + error_report("kvm: bus lock detection unsupported"); + return -ENOTSUP; + } + ret = kvm_vm_enable_cap(s, KVM_CAP_X86_BUS_LOCK_EXIT, 0, + KVM_BUS_LOCK_DETECTION_EXIT); + if (ret < 0) { + error_report("kvm: Failed to enable bus lock detection cap: %s", + strerror(-ret)); + } + + return ret; +} + +static int kvm_vm_enable_notify_vmexit(KVMState *s) +{ + int ret = 0; + if (s->notify_vmexit != NOTIFY_VMEXIT_OPTION_DISABLE) { + uint64_t notify_window_flags = + ((uint64_t)s->notify_window << 32) | + KVM_X86_NOTIFY_VMEXIT_ENABLED | + KVM_X86_NOTIFY_VMEXIT_USER; + ret = kvm_vm_enable_cap(s, KVM_CAP_X86_NOTIFY_VMEXIT, 0, + notify_window_flags); + if (ret < 0) { + error_report("kvm: Failed to enable notify vmexit cap: %s", + strerror(-ret)); + } + } + return ret; +} + +static int kvm_vm_enable_userspace_msr(KVMState *s) +{ + int ret = kvm_vm_enable_cap(s, KVM_CAP_X86_USER_SPACE_MSR, 0, + KVM_MSR_EXIT_REASON_FILTER); + if (ret < 0) { + error_report("Could not enable user space MSRs: %s", + strerror(-ret)); + exit(1); + } + + if (!kvm_filter_msr(s, MSR_CORE_THREAD_COUNT, + kvm_rdmsr_core_thread_count, NULL)) { + error_report("Could not install MSR_CORE_THREAD_COUNT handler!"); + exit(1); + } + + return 0; +} + +static void kvm_vm_enable_energy_msrs(KVMState *s) +{ + bool r; + if (s->msr_energy.enable == true) { + r = kvm_filter_msr(s, MSR_RAPL_POWER_UNIT, + kvm_rdmsr_rapl_power_unit, NULL); + if (!r) { + error_report("Could not install MSR_RAPL_POWER_UNIT \ + handler"); + exit(1); + } + + r = kvm_filter_msr(s, MSR_PKG_POWER_LIMIT, + kvm_rdmsr_pkg_power_limit, NULL); + if (!r) { + error_report("Could not install MSR_PKG_POWER_LIMIT \ + handler"); + exit(1); + } + + r = kvm_filter_msr(s, MSR_PKG_POWER_INFO, + kvm_rdmsr_pkg_power_info, NULL); + if (!r) { + error_report("Could not install MSR_PKG_POWER_INFO \ + handler"); + exit(1); + } + r = kvm_filter_msr(s, MSR_PKG_ENERGY_STATUS, + kvm_rdmsr_pkg_energy_status, NULL); + if (!r) { + error_report("Could not install MSR_PKG_ENERGY_STATUS \ + handler"); + exit(1); + } + } + return; +} + +int kvm_arch_init(MachineState *ms, KVMState *s) +{ int ret; struct utsname utsname; Error *local_err = NULL; @@ -3028,24 +3212,14 @@ int kvm_arch_init(MachineState *ms, KVMState *s) hv_vpindex_settable = kvm_check_extension(s, KVM_CAP_HYPERV_VP_INDEX); - has_exception_payload = kvm_check_extension(s, KVM_CAP_EXCEPTION_PAYLOAD); - if (has_exception_payload) { - ret = kvm_vm_enable_cap(s, KVM_CAP_EXCEPTION_PAYLOAD, 0, true); - if (ret < 0) { - error_report("kvm: Failed to enable exception payload cap: %s", - strerror(-ret)); - return ret; - } + ret = kvm_vm_enable_exception_payload(s); + if (ret < 0) { + return ret; } - has_triple_fault_event = kvm_check_extension(s, KVM_CAP_X86_TRIPLE_FAULT_EVENT); - if (has_triple_fault_event) { - ret = kvm_vm_enable_cap(s, KVM_CAP_X86_TRIPLE_FAULT_EVENT, 0, true); - if (ret < 0) { - error_report("kvm: Failed to enable triple fault event cap: %s", - strerror(-ret)); - return ret; - } + ret = kvm_vm_enable_triple_fault_event(s); + if (ret < 0) { + return ret; } if (s->xen_version) { @@ -3076,36 +3250,23 @@ int kvm_arch_init(MachineState *ms, KVMState *s) uname(&utsname); lm_capable_kernel = strcmp(utsname.machine, "x86_64") == 0; - /* - * On older Intel CPUs, KVM uses vm86 mode to emulate 16-bit code directly. - * In order to use vm86 mode, an EPT identity map and a TSS are needed. - * Since these must be part of guest physical memory, we need to allocate - * them, both by setting their start addresses in the kernel and by - * creating a corresponding e820 entry. We need 4 pages before the BIOS, - * so this value allows up to 16M BIOSes. - */ - identity_base = 0xfeffc000; - ret = kvm_vm_ioctl(s, KVM_SET_IDENTITY_MAP_ADDR, &identity_base); + ret = kvm_vm_set_identity_map_addr(s, KVM_IDENTITY_BASE); if (ret < 0) { return ret; } /* Set TSS base one page after EPT identity map. */ - ret = kvm_vm_ioctl(s, KVM_SET_TSS_ADDR, identity_base + 0x1000); + ret = kvm_vm_set_tss_addr(s, KVM_IDENTITY_BASE + 0x1000); if (ret < 0) { return ret; } /* Tell fw_cfg to notify the BIOS to reserve the range. */ - e820_add_entry(identity_base, 0x4000, E820_RESERVED); + e820_add_entry(KVM_IDENTITY_BASE, 0x4000, E820_RESERVED); - shadow_mem = object_property_get_int(OBJECT(s), "kvm-shadow-mem", &error_abort); - if (shadow_mem != -1) { - shadow_mem /= 4096; - ret = kvm_vm_ioctl(s, KVM_SET_NR_MMU_PAGES, shadow_mem); - if (ret < 0) { - return ret; - } + ret = kvm_vm_set_nr_mmu_pages(s); + if (ret < 0) { + return ret; } if (kvm_check_extension(s, KVM_CAP_X86_SMM) && @@ -3116,20 +3277,7 @@ int kvm_arch_init(MachineState *ms, KVMState *s) } if (enable_cpu_pm) { - int disable_exits = kvm_check_extension(s, KVM_CAP_X86_DISABLE_EXITS); -/* Work around for kernel header with a typo. TODO: fix header and drop. */ -#if defined(KVM_X86_DISABLE_EXITS_HTL) && !defined(KVM_X86_DISABLE_EXITS_HLT) -#define KVM_X86_DISABLE_EXITS_HLT KVM_X86_DISABLE_EXITS_HTL -#endif - if (disable_exits) { - disable_exits &= (KVM_X86_DISABLE_EXITS_MWAIT | - KVM_X86_DISABLE_EXITS_HLT | - KVM_X86_DISABLE_EXITS_PAUSE | - KVM_X86_DISABLE_EXITS_CSTATE); - } - - ret = kvm_vm_enable_cap(s, KVM_CAP_X86_DISABLE_EXITS, 0, - disable_exits); + ret = kvm_vm_enable_disable_exits(s); if (ret < 0) { error_report("kvm: guest stopping CPU not supported: %s", strerror(-ret)); @@ -3140,16 +3288,8 @@ int kvm_arch_init(MachineState *ms, KVMState *s) X86MachineState *x86ms = X86_MACHINE(ms); if (x86ms->bus_lock_ratelimit > 0) { - ret = kvm_check_extension(s, KVM_CAP_X86_BUS_LOCK_EXIT); - if (!(ret & KVM_BUS_LOCK_DETECTION_EXIT)) { - error_report("kvm: bus lock detection unsupported"); - return -ENOTSUP; - } - ret = kvm_vm_enable_cap(s, KVM_CAP_X86_BUS_LOCK_EXIT, 0, - KVM_BUS_LOCK_DETECTION_EXIT); + ret = kvm_vm_enable_bus_lock_exit(s); if (ret < 0) { - error_report("kvm: Failed to enable bus lock detection cap: %s", - strerror(-ret)); return ret; } ratelimit_init(&bus_lock_ratelimit_ctrl); @@ -3158,80 +3298,25 @@ int kvm_arch_init(MachineState *ms, KVMState *s) } } - if (s->notify_vmexit != NOTIFY_VMEXIT_OPTION_DISABLE && - kvm_check_extension(s, KVM_CAP_X86_NOTIFY_VMEXIT)) { - uint64_t notify_window_flags = - ((uint64_t)s->notify_window << 32) | - KVM_X86_NOTIFY_VMEXIT_ENABLED | - KVM_X86_NOTIFY_VMEXIT_USER; - ret = kvm_vm_enable_cap(s, KVM_CAP_X86_NOTIFY_VMEXIT, 0, - notify_window_flags); - if (ret < 0) { - error_report("kvm: Failed to enable notify vmexit cap: %s", - strerror(-ret)); - return ret; - } - } - if (kvm_vm_check_extension(s, KVM_CAP_X86_USER_SPACE_MSR)) { - bool r; - - ret = kvm_vm_enable_cap(s, KVM_CAP_X86_USER_SPACE_MSR, 0, - KVM_MSR_EXIT_REASON_FILTER); - if (ret) { - error_report("Could not enable user space MSRs: %s", - strerror(-ret)); - exit(1); + if (kvm_check_extension(s, KVM_CAP_X86_NOTIFY_VMEXIT)) { + ret = kvm_vm_enable_notify_vmexit(s); + if (ret < 0) { + return ret; } + } - r = kvm_filter_msr(s, MSR_CORE_THREAD_COUNT, - kvm_rdmsr_core_thread_count, NULL); - if (!r) { - error_report("Could not install MSR_CORE_THREAD_COUNT handler: %s", - strerror(-ret)); - exit(1); + if (kvm_vm_check_extension(s, KVM_CAP_X86_USER_SPACE_MSR)) { + ret = kvm_vm_enable_userspace_msr(s); + if (ret < 0) { + return ret; } if (s->msr_energy.enable == true) { - r = kvm_filter_msr(s, MSR_RAPL_POWER_UNIT, - kvm_rdmsr_rapl_power_unit, NULL); - if (!r) { - error_report("Could not install MSR_RAPL_POWER_UNIT \ - handler: %s", - strerror(-ret)); - exit(1); - } - - r = kvm_filter_msr(s, MSR_PKG_POWER_LIMIT, - kvm_rdmsr_pkg_power_limit, NULL); - if (!r) { - error_report("Could not install MSR_PKG_POWER_LIMIT \ - handler: %s", - strerror(-ret)); - exit(1); - } - - r = kvm_filter_msr(s, MSR_PKG_POWER_INFO, - kvm_rdmsr_pkg_power_info, NULL); - if (!r) { - error_report("Could not install MSR_PKG_POWER_INFO \ - handler: %s", - strerror(-ret)); + kvm_vm_enable_energy_msrs(s); + if (kvm_msr_energy_thread_init(s, ms)) { + error_report("kvm : error RAPL feature requirement not met"); exit(1); } - r = kvm_filter_msr(s, MSR_PKG_ENERGY_STATUS, - kvm_rdmsr_pkg_energy_status, NULL); - if (!r) { - error_report("Could not install MSR_PKG_ENERGY_STATUS \ - handler: %s", - strerror(-ret)); - exit(1); - } - r = kvm_msr_energy_thread_init(s, ms); - if (r) { - error_report("kvm : error RAPL feature requirement not meet"); - exit(1); - } - } } @@ -3694,7 +3779,14 @@ static void kvm_msr_entry_add_vmx(X86CPU *cpu, FeatureWordArray f) kvm_msr_entry_add(cpu, MSR_IA32_VMX_CR4_FIXED0, CR4_VMXE_MASK); - if (f[FEAT_VMX_SECONDARY_CTLS] & VMX_SECONDARY_EXEC_TSC_SCALING) { + if (f[FEAT_7_1_EAX] & CPUID_7_1_EAX_FRED) { + /* FRED injected-event data (0x2052). */ + kvm_msr_entry_add(cpu, MSR_IA32_VMX_VMCS_ENUM, 0x52); + } else if (f[FEAT_VMX_EXIT_CTLS] & + VMX_VM_EXIT_ACTIVATE_SECONDARY_CONTROLS) { + /* Secondary VM-exit controls (0x2044). */ + kvm_msr_entry_add(cpu, MSR_IA32_VMX_VMCS_ENUM, 0x44); + } else if (f[FEAT_VMX_SECONDARY_CTLS] & VMX_SECONDARY_EXEC_TSC_SCALING) { /* TSC multiplier (0x2032). */ kvm_msr_entry_add(cpu, MSR_IA32_VMX_VMCS_ENUM, 0x32); } else { @@ -5118,7 +5210,7 @@ static int kvm_get_nested_state(X86CPU *cpu) return ret; } -int kvm_arch_put_registers(CPUState *cpu, int level) +int kvm_arch_put_registers(CPUState *cpu, int level, Error **errp) { X86CPU *x86_cpu = X86_CPU(cpu); int ret; @@ -5133,6 +5225,7 @@ int kvm_arch_put_registers(CPUState *cpu, int level) if (level >= KVM_PUT_RESET_STATE) { ret = kvm_put_msr_feature_control(x86_cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set feature control MSR"); return ret; } } @@ -5140,12 +5233,14 @@ int kvm_arch_put_registers(CPUState *cpu, int level) /* must be before kvm_put_nested_state so that EFER.SVME is set */ ret = has_sregs2 ? kvm_put_sregs2(x86_cpu) : kvm_put_sregs(x86_cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set special registers"); return ret; } if (level >= KVM_PUT_RESET_STATE) { ret = kvm_put_nested_state(x86_cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set nested state"); return ret; } } @@ -5163,6 +5258,7 @@ int kvm_arch_put_registers(CPUState *cpu, int level) if (xen_mode == XEN_EMULATE && level == KVM_PUT_FULL_STATE) { ret = kvm_put_xen_state(cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set Xen state"); return ret; } } @@ -5170,43 +5266,51 @@ int kvm_arch_put_registers(CPUState *cpu, int level) ret = kvm_getput_regs(x86_cpu, 1); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set general purpose registers"); return ret; } ret = kvm_put_xsave(x86_cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set XSAVE"); return ret; } ret = kvm_put_xcrs(x86_cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set XCRs"); return ret; } ret = kvm_put_msrs(x86_cpu, level); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set MSRs"); return ret; } ret = kvm_put_vcpu_events(x86_cpu, level); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set vCPU events"); return ret; } if (level >= KVM_PUT_RESET_STATE) { ret = kvm_put_mp_state(x86_cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set MP state"); return ret; } } ret = kvm_put_tscdeadline_msr(x86_cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set TSC deadline MSR"); return ret; } ret = kvm_put_debugregs(x86_cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to set debug registers"); return ret; } return 0; } -int kvm_arch_get_registers(CPUState *cs) +int kvm_arch_get_registers(CPUState *cs, Error **errp) { X86CPU *cpu = X86_CPU(cs); int ret; @@ -5215,6 +5319,7 @@ int kvm_arch_get_registers(CPUState *cs) ret = kvm_get_vcpu_events(cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get vCPU events"); goto out; } /* @@ -5223,44 +5328,54 @@ int kvm_arch_get_registers(CPUState *cs) */ ret = kvm_get_mp_state(cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get MP state"); goto out; } ret = kvm_getput_regs(cpu, 0); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get general purpose registers"); goto out; } ret = kvm_get_xsave(cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get XSAVE"); goto out; } ret = kvm_get_xcrs(cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get XCRs"); goto out; } ret = has_sregs2 ? kvm_get_sregs2(cpu) : kvm_get_sregs(cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get special registers"); goto out; } ret = kvm_get_msrs(cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get MSRs"); goto out; } ret = kvm_get_apic(cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get APIC"); goto out; } ret = kvm_get_debugregs(cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get debug registers"); goto out; } ret = kvm_get_nested_state(cpu); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get nested state"); goto out; } #ifdef CONFIG_XEN_EMU if (xen_mode == XEN_EMULATE) { ret = kvm_get_xen_state(cs); if (ret < 0) { + error_setg_errno(errp, -ret, "Failed to get Xen state"); goto out; } } @@ -5729,7 +5844,7 @@ static bool kvm_install_msr_filters(KVMState *s) return true; } -bool kvm_filter_msr(KVMState *s, uint32_t msr, QEMURDMSRHandler *rdmsr, +static bool kvm_filter_msr(KVMState *s, uint32_t msr, QEMURDMSRHandler *rdmsr, QEMUWRMSRHandler *wrmsr) { int i; |