KVM: VMX: Implement PCID/INVPCID for guests with EPT

[mirror_ubuntu-artful-kernel.git] / arch / x86 / kvm / x86.c

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index be6d54929fa7d661c31f65d076c0daad8086d785..59b59508ff07dc2623887d6be43859215d5bd589 100644 (file)
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -528,6 +528,9 @@ int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
                        return 1;
        }
 
+       if (!(cr0 & X86_CR0_PG) && kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE))
+               return 1;
+
        kvm_x86_ops->set_cr0(vcpu, cr0);
 
        if ((cr0 ^ old_cr0) & X86_CR0_PG) {
@@ -604,10 +607,20 @@ int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
                                   kvm_read_cr3(vcpu)))
                return 1;
 
+       if ((cr4 & X86_CR4_PCIDE) && !(old_cr4 & X86_CR4_PCIDE)) {
+               if (!guest_cpuid_has_pcid(vcpu))
+                       return 1;
+
+               /* PCID can not be enabled when cr3[11:0]!=000H or EFER.LMA=0 */
+               if ((kvm_read_cr3(vcpu) & X86_CR3_PCID_MASK) || !is_long_mode(vcpu))
+                       return 1;
+       }
+
        if (kvm_x86_ops->set_cr4(vcpu, cr4))
                return 1;
 
-       if ((cr4 ^ old_cr4) & pdptr_bits)
+       if (((cr4 ^ old_cr4) & pdptr_bits) ||
+           (!(cr4 & X86_CR4_PCIDE) && (old_cr4 & X86_CR4_PCIDE)))
                kvm_mmu_reset_context(vcpu);
 
        if ((cr4 ^ old_cr4) & X86_CR4_OSXSAVE)
@@ -626,8 +639,12 @@ int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
        }
 
        if (is_long_mode(vcpu)) {
-               if (cr3 & CR3_L_MODE_RESERVED_BITS)
-                       return 1;
+               if (kvm_read_cr4(vcpu) & X86_CR4_PCIDE) {
+                       if (cr3 & CR3_PCID_ENABLED_RESERVED_BITS)
+                               return 1;
+               } else
+                       if (cr3 & CR3_L_MODE_RESERVED_BITS)
+                               return 1;
        } else {
                if (is_pae(vcpu)) {
                        if (cr3 & CR3_PAE_RESERVED_BITS)
@@ -795,6 +812,7 @@ static u32 msrs_to_save[] = {
        MSR_KVM_SYSTEM_TIME_NEW, MSR_KVM_WALL_CLOCK_NEW,
        HV_X64_MSR_GUEST_OS_ID, HV_X64_MSR_HYPERCALL,
        HV_X64_MSR_APIC_ASSIST_PAGE, MSR_KVM_ASYNC_PF_EN, MSR_KVM_STEAL_TIME,
+       MSR_KVM_PV_EOI_EN,
        MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP,
        MSR_STAR,
 #ifdef CONFIG_X86_64
@@ -1437,8 +1455,8 @@ static int set_msr_hyperv_pw(struct kvm_vcpu *vcpu, u32 msr, u64 data)
                break;
        }
        default:
-               pr_unimpl(vcpu, "HYPER-V unimplemented wrmsr: 0x%x "
-                         "data 0x%llx\n", msr, data);
+               vcpu_unimpl(vcpu, "HYPER-V unimplemented wrmsr: 0x%x "
+                           "data 0x%llx\n", msr, data);
                return 1;
        }
        return 0;
@@ -1470,8 +1488,8 @@ static int set_msr_hyperv(struct kvm_vcpu *vcpu, u32 msr, u64 data)
        case HV_X64_MSR_TPR:
                return kvm_hv_vapic_msr_write(vcpu, APIC_TASKPRI, data);
        default:
-               pr_unimpl(vcpu, "HYPER-V unimplemented wrmsr: 0x%x "
-                         "data 0x%llx\n", msr, data);
+               vcpu_unimpl(vcpu, "HYPER-V unimplemented wrmsr: 0x%x "
+                           "data 0x%llx\n", msr, data);
                return 1;
        }
 
@@ -1551,15 +1569,15 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
                data &= ~(u64)0x100;    /* ignore ignne emulation enable */
                data &= ~(u64)0x8;      /* ignore TLB cache disable */
                if (data != 0) {
-                       pr_unimpl(vcpu, "unimplemented HWCR wrmsr: 0x%llx\n",
-                               data);
+                       vcpu_unimpl(vcpu, "unimplemented HWCR wrmsr: 0x%llx\n",
+                                   data);
                        return 1;
                }
                break;
        case MSR_FAM10H_MMIO_CONF_BASE:
                if (data != 0) {
-                       pr_unimpl(vcpu, "unimplemented MMIO_CONF_BASE wrmsr: "
-                               "0x%llx\n", data);
+                       vcpu_unimpl(vcpu, "unimplemented MMIO_CONF_BASE wrmsr: "
+                                   "0x%llx\n", data);
                        return 1;
                }
                break;
@@ -1574,8 +1592,8 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
                           thus reserved and should throw a #GP */
                        return 1;
                }
-               pr_unimpl(vcpu, "%s: MSR_IA32_DEBUGCTLMSR 0x%llx, nop\n",
-                       __func__, data);
+               vcpu_unimpl(vcpu, "%s: MSR_IA32_DEBUGCTLMSR 0x%llx, nop\n",
+                           __func__, data);
                break;
        case MSR_IA32_UCODE_REV:
        case MSR_IA32_UCODE_WRITE:
@@ -1653,6 +1671,10 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
                kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu);
 
                break;
+       case MSR_KVM_PV_EOI_EN:
+               if (kvm_lapic_enable_pv_eoi(vcpu, data))
+                       return 1;
+               break;
 
        case MSR_IA32_MCG_CTL:
        case MSR_IA32_MCG_STATUS:
@@ -1671,8 +1693,8 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
        case MSR_K7_EVNTSEL2:
        case MSR_K7_EVNTSEL3:
                if (data != 0)
-                       pr_unimpl(vcpu, "unimplemented perfctr wrmsr: "
-                               "0x%x data 0x%llx\n", msr, data);
+                       vcpu_unimpl(vcpu, "unimplemented perfctr wrmsr: "
+                                   "0x%x data 0x%llx\n", msr, data);
                break;
        /* at least RHEL 4 unconditionally writes to the perfctr registers,
         * so we ignore writes to make it happy.
@@ -1681,8 +1703,8 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
        case MSR_K7_PERFCTR1:
        case MSR_K7_PERFCTR2:
        case MSR_K7_PERFCTR3:
-               pr_unimpl(vcpu, "unimplemented perfctr wrmsr: "
-                       "0x%x data 0x%llx\n", msr, data);
+               vcpu_unimpl(vcpu, "unimplemented perfctr wrmsr: "
+                           "0x%x data 0x%llx\n", msr, data);
                break;
        case MSR_P6_PERFCTR0:
        case MSR_P6_PERFCTR1:
@@ -1693,8 +1715,8 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
                        return kvm_pmu_set_msr(vcpu, msr, data);
 
                if (pr || data != 0)
-                       pr_unimpl(vcpu, "disabled perfctr wrmsr: "
-                               "0x%x data 0x%llx\n", msr, data);
+                       vcpu_unimpl(vcpu, "disabled perfctr wrmsr: "
+                                   "0x%x data 0x%llx\n", msr, data);
                break;
        case MSR_K7_CLK_CTL:
                /*
@@ -1720,7 +1742,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
                /* Drop writes to this legacy MSR -- see rdmsr
                 * counterpart for further detail.
                 */
-               pr_unimpl(vcpu, "ignored wrmsr: 0x%x data %llx\n", msr, data);
+               vcpu_unimpl(vcpu, "ignored wrmsr: 0x%x data %llx\n", msr, data);
                break;
        case MSR_AMD64_OSVW_ID_LENGTH:
                if (!guest_cpuid_has_osvw(vcpu))
@@ -1738,12 +1760,12 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
                if (kvm_pmu_msr(vcpu, msr))
                        return kvm_pmu_set_msr(vcpu, msr, data);
                if (!ignore_msrs) {
-                       pr_unimpl(vcpu, "unhandled wrmsr: 0x%x data %llx\n",
-                               msr, data);
+                       vcpu_unimpl(vcpu, "unhandled wrmsr: 0x%x data %llx\n",
+                                   msr, data);
                        return 1;
                } else {
-                       pr_unimpl(vcpu, "ignored wrmsr: 0x%x data %llx\n",
-                               msr, data);
+                       vcpu_unimpl(vcpu, "ignored wrmsr: 0x%x data %llx\n",
+                                   msr, data);
                        break;
                }
        }
@@ -1846,7 +1868,7 @@ static int get_msr_hyperv_pw(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
                data = kvm->arch.hv_hypercall;
                break;
        default:
-               pr_unimpl(vcpu, "Hyper-V unhandled rdmsr: 0x%x\n", msr);
+               vcpu_unimpl(vcpu, "Hyper-V unhandled rdmsr: 0x%x\n", msr);
                return 1;
        }
 
@@ -1877,7 +1899,7 @@ static int get_msr_hyperv(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
                data = vcpu->arch.hv_vapic;
                break;
        default:
-               pr_unimpl(vcpu, "Hyper-V unhandled rdmsr: 0x%x\n", msr);
+               vcpu_unimpl(vcpu, "Hyper-V unhandled rdmsr: 0x%x\n", msr);
                return 1;
        }
        *pdata = data;
@@ -2030,10 +2052,10 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
                if (kvm_pmu_msr(vcpu, msr))
                        return kvm_pmu_get_msr(vcpu, msr, pdata);
                if (!ignore_msrs) {
-                       pr_unimpl(vcpu, "unhandled rdmsr: 0x%x\n", msr);
+                       vcpu_unimpl(vcpu, "unhandled rdmsr: 0x%x\n", msr);
                        return 1;
                } else {
-                       pr_unimpl(vcpu, "ignored rdmsr: 0x%x\n", msr);
+                       vcpu_unimpl(vcpu, "ignored rdmsr: 0x%x\n", msr);
                        data = 0;
                }
                break;
@@ -4116,7 +4138,7 @@ static unsigned long emulator_get_cr(struct x86_emulate_ctxt *ctxt, int cr)
                value = kvm_get_cr8(vcpu);
                break;
        default:
-               vcpu_printf(vcpu, "%s: unexpected cr %u\n", __func__, cr);
+               kvm_err("%s: unexpected cr %u\n", __func__, cr);
                return 0;
        }
 
@@ -4145,7 +4167,7 @@ static int emulator_set_cr(struct x86_emulate_ctxt *ctxt, int cr, ulong val)
                res = kvm_set_cr8(vcpu, val);
                break;
        default:
-               vcpu_printf(vcpu, "%s: unexpected cr %u\n", __func__, cr);
+               kvm_err("%s: unexpected cr %u\n", __func__, cr);
                res = -1;
        }
 
@@ -4297,26 +4319,10 @@ static int emulator_intercept(struct x86_emulate_ctxt *ctxt,
        return kvm_x86_ops->check_intercept(emul_to_vcpu(ctxt), info, stage);
 }
 
-static bool emulator_get_cpuid(struct x86_emulate_ctxt *ctxt,
+static void emulator_get_cpuid(struct x86_emulate_ctxt *ctxt,
                               u32 *eax, u32 *ebx, u32 *ecx, u32 *edx)
 {
-       struct kvm_cpuid_entry2 *cpuid = NULL;
-
-       if (eax && ecx)
-               cpuid = kvm_find_cpuid_entry(emul_to_vcpu(ctxt),
-                                           *eax, *ecx);
-
-       if (cpuid) {
-               *eax = cpuid->eax;
-               *ecx = cpuid->ecx;
-               if (ebx)
-                       *ebx = cpuid->ebx;
-               if (edx)
-                       *edx = cpuid->edx;
-               return true;
-       }
-
-       return false;
+       kvm_cpuid(emul_to_vcpu(ctxt), eax, ebx, ecx, edx);
 }
 
 static struct x86_emulate_ops emulate_ops = {
@@ -5296,8 +5302,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 
        r = kvm_mmu_reload(vcpu);
        if (unlikely(r)) {
-               kvm_x86_ops->cancel_injection(vcpu);
-               goto out;
+               goto cancel_injection;
        }
 
        preempt_disable();
@@ -5322,9 +5327,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
                smp_wmb();
                local_irq_enable();
                preempt_enable();
-               kvm_x86_ops->cancel_injection(vcpu);
                r = 1;
-               goto out;
+               goto cancel_injection;
        }
 
        srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
@@ -5388,9 +5392,16 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
        if (unlikely(vcpu->arch.tsc_always_catchup))
                kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
 
-       kvm_lapic_sync_from_vapic(vcpu);
+       if (vcpu->arch.apic_attention)
+               kvm_lapic_sync_from_vapic(vcpu);
 
        r = kvm_x86_ops->handle_exit(vcpu);
+       return r;
+
+cancel_injection:
+       kvm_x86_ops->cancel_injection(vcpu);
+       if (unlikely(vcpu->arch.apic_attention))
+               kvm_lapic_sync_from_vapic(vcpu);
 out:
        return r;
 }
@@ -6304,7 +6315,7 @@ void kvm_arch_free_memslot(struct kvm_memory_slot *free,
 
        for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
                if (!dont || free->arch.lpage_info[i] != dont->arch.lpage_info[i]) {
-                       vfree(free->arch.lpage_info[i]);
+                       kvm_kvfree(free->arch.lpage_info[i]);
                        free->arch.lpage_info[i] = NULL;
                }
        }
@@ -6323,7 +6334,7 @@ int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
                                      slot->base_gfn, level) + 1;
 
                slot->arch.lpage_info[i] =
-                       vzalloc(lpages * sizeof(*slot->arch.lpage_info[i]));
+                       kvm_kvzalloc(lpages * sizeof(*slot->arch.lpage_info[i]));
                if (!slot->arch.lpage_info[i])
                        goto out_free;
 
@@ -6350,7 +6361,7 @@ int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
 
 out_free:
        for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
-               vfree(slot->arch.lpage_info[i]);
+               kvm_kvfree(slot->arch.lpage_info[i]);
                slot->arch.lpage_info[i] = NULL;
        }
        return -ENOMEM;