These changes are the raw update to linux-4.4.6-rt14. Kernel sources
[kvmfornfv.git] / kernel / arch / x86 / kvm / hyperv.c
diff --git a/kernel/arch/x86/kvm/hyperv.c b/kernel/arch/x86/kvm/hyperv.c
new file mode 100644 (file)
index 0000000..62cf8c9
--- /dev/null
@@ -0,0 +1,404 @@
+/*
+ * KVM Microsoft Hyper-V emulation
+ *
+ * derived from arch/x86/kvm/x86.c
+ *
+ * Copyright (C) 2006 Qumranet, Inc.
+ * Copyright (C) 2008 Qumranet, Inc.
+ * Copyright IBM Corporation, 2008
+ * Copyright 2010 Red Hat, Inc. and/or its affiliates.
+ * Copyright (C) 2015 Andrey Smetanin <asmetanin@virtuozzo.com>
+ *
+ * Authors:
+ *   Avi Kivity   <avi@qumranet.com>
+ *   Yaniv Kamay  <yaniv@qumranet.com>
+ *   Amit Shah    <amit.shah@qumranet.com>
+ *   Ben-Ami Yassour <benami@il.ibm.com>
+ *   Andrey Smetanin <asmetanin@virtuozzo.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#include "x86.h"
+#include "lapic.h"
+#include "hyperv.h"
+
+#include <linux/kvm_host.h>
+#include <trace/events/kvm.h>
+
+#include "trace.h"
+
+static bool kvm_hv_msr_partition_wide(u32 msr)
+{
+       bool r = false;
+
+       switch (msr) {
+       case HV_X64_MSR_GUEST_OS_ID:
+       case HV_X64_MSR_HYPERCALL:
+       case HV_X64_MSR_REFERENCE_TSC:
+       case HV_X64_MSR_TIME_REF_COUNT:
+       case HV_X64_MSR_CRASH_CTL:
+       case HV_X64_MSR_CRASH_P0 ... HV_X64_MSR_CRASH_P4:
+       case HV_X64_MSR_RESET:
+               r = true;
+               break;
+       }
+
+       return r;
+}
+
+static int kvm_hv_msr_get_crash_data(struct kvm_vcpu *vcpu,
+                                    u32 index, u64 *pdata)
+{
+       struct kvm_hv *hv = &vcpu->kvm->arch.hyperv;
+
+       if (WARN_ON_ONCE(index >= ARRAY_SIZE(hv->hv_crash_param)))
+               return -EINVAL;
+
+       *pdata = hv->hv_crash_param[index];
+       return 0;
+}
+
+static int kvm_hv_msr_get_crash_ctl(struct kvm_vcpu *vcpu, u64 *pdata)
+{
+       struct kvm_hv *hv = &vcpu->kvm->arch.hyperv;
+
+       *pdata = hv->hv_crash_ctl;
+       return 0;
+}
+
+static int kvm_hv_msr_set_crash_ctl(struct kvm_vcpu *vcpu, u64 data, bool host)
+{
+       struct kvm_hv *hv = &vcpu->kvm->arch.hyperv;
+
+       if (host)
+               hv->hv_crash_ctl = data & HV_X64_MSR_CRASH_CTL_NOTIFY;
+
+       if (!host && (data & HV_X64_MSR_CRASH_CTL_NOTIFY)) {
+
+               vcpu_debug(vcpu, "hv crash (0x%llx 0x%llx 0x%llx 0x%llx 0x%llx)\n",
+                         hv->hv_crash_param[0],
+                         hv->hv_crash_param[1],
+                         hv->hv_crash_param[2],
+                         hv->hv_crash_param[3],
+                         hv->hv_crash_param[4]);
+
+               /* Send notification about crash to user space */
+               kvm_make_request(KVM_REQ_HV_CRASH, vcpu);
+       }
+
+       return 0;
+}
+
+static int kvm_hv_msr_set_crash_data(struct kvm_vcpu *vcpu,
+                                    u32 index, u64 data)
+{
+       struct kvm_hv *hv = &vcpu->kvm->arch.hyperv;
+
+       if (WARN_ON_ONCE(index >= ARRAY_SIZE(hv->hv_crash_param)))
+               return -EINVAL;
+
+       hv->hv_crash_param[index] = data;
+       return 0;
+}
+
+static int kvm_hv_set_msr_pw(struct kvm_vcpu *vcpu, u32 msr, u64 data,
+                            bool host)
+{
+       struct kvm *kvm = vcpu->kvm;
+       struct kvm_hv *hv = &kvm->arch.hyperv;
+
+       switch (msr) {
+       case HV_X64_MSR_GUEST_OS_ID:
+               hv->hv_guest_os_id = data;
+               /* setting guest os id to zero disables hypercall page */
+               if (!hv->hv_guest_os_id)
+                       hv->hv_hypercall &= ~HV_X64_MSR_HYPERCALL_ENABLE;
+               break;
+       case HV_X64_MSR_HYPERCALL: {
+               u64 gfn;
+               unsigned long addr;
+               u8 instructions[4];
+
+               /* if guest os id is not set hypercall should remain disabled */
+               if (!hv->hv_guest_os_id)
+                       break;
+               if (!(data & HV_X64_MSR_HYPERCALL_ENABLE)) {
+                       hv->hv_hypercall = data;
+                       break;
+               }
+               gfn = data >> HV_X64_MSR_HYPERCALL_PAGE_ADDRESS_SHIFT;
+               addr = gfn_to_hva(kvm, gfn);
+               if (kvm_is_error_hva(addr))
+                       return 1;
+               kvm_x86_ops->patch_hypercall(vcpu, instructions);
+               ((unsigned char *)instructions)[3] = 0xc3; /* ret */
+               if (__copy_to_user((void __user *)addr, instructions, 4))
+                       return 1;
+               hv->hv_hypercall = data;
+               mark_page_dirty(kvm, gfn);
+               break;
+       }
+       case HV_X64_MSR_REFERENCE_TSC: {
+               u64 gfn;
+               HV_REFERENCE_TSC_PAGE tsc_ref;
+
+               memset(&tsc_ref, 0, sizeof(tsc_ref));
+               hv->hv_tsc_page = data;
+               if (!(data & HV_X64_MSR_TSC_REFERENCE_ENABLE))
+                       break;
+               gfn = data >> HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT;
+               if (kvm_write_guest(
+                               kvm,
+                               gfn << HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT,
+                               &tsc_ref, sizeof(tsc_ref)))
+                       return 1;
+               mark_page_dirty(kvm, gfn);
+               break;
+       }
+       case HV_X64_MSR_CRASH_P0 ... HV_X64_MSR_CRASH_P4:
+               return kvm_hv_msr_set_crash_data(vcpu,
+                                                msr - HV_X64_MSR_CRASH_P0,
+                                                data);
+       case HV_X64_MSR_CRASH_CTL:
+               return kvm_hv_msr_set_crash_ctl(vcpu, data, host);
+       case HV_X64_MSR_RESET:
+               if (data == 1) {
+                       vcpu_debug(vcpu, "hyper-v reset requested\n");
+                       kvm_make_request(KVM_REQ_HV_RESET, vcpu);
+               }
+               break;
+       default:
+               vcpu_unimpl(vcpu, "Hyper-V uhandled wrmsr: 0x%x data 0x%llx\n",
+                           msr, data);
+               return 1;
+       }
+       return 0;
+}
+
+/* Calculate cpu time spent by current task in 100ns units */
+static u64 current_task_runtime_100ns(void)
+{
+       cputime_t utime, stime;
+
+       task_cputime_adjusted(current, &utime, &stime);
+       return div_u64(cputime_to_nsecs(utime + stime), 100);
+}
+
+static int kvm_hv_set_msr(struct kvm_vcpu *vcpu, u32 msr, u64 data, bool host)
+{
+       struct kvm_vcpu_hv *hv = &vcpu->arch.hyperv;
+
+       switch (msr) {
+       case HV_X64_MSR_APIC_ASSIST_PAGE: {
+               u64 gfn;
+               unsigned long addr;
+
+               if (!(data & HV_X64_MSR_APIC_ASSIST_PAGE_ENABLE)) {
+                       hv->hv_vapic = data;
+                       if (kvm_lapic_enable_pv_eoi(vcpu, 0))
+                               return 1;
+                       break;
+               }
+               gfn = data >> HV_X64_MSR_APIC_ASSIST_PAGE_ADDRESS_SHIFT;
+               addr = kvm_vcpu_gfn_to_hva(vcpu, gfn);
+               if (kvm_is_error_hva(addr))
+                       return 1;
+               if (__clear_user((void __user *)addr, PAGE_SIZE))
+                       return 1;
+               hv->hv_vapic = data;
+               kvm_vcpu_mark_page_dirty(vcpu, gfn);
+               if (kvm_lapic_enable_pv_eoi(vcpu,
+                                           gfn_to_gpa(gfn) | KVM_MSR_ENABLED))
+                       return 1;
+               break;
+       }
+       case HV_X64_MSR_EOI:
+               return kvm_hv_vapic_msr_write(vcpu, APIC_EOI, data);
+       case HV_X64_MSR_ICR:
+               return kvm_hv_vapic_msr_write(vcpu, APIC_ICR, data);
+       case HV_X64_MSR_TPR:
+               return kvm_hv_vapic_msr_write(vcpu, APIC_TASKPRI, data);
+       case HV_X64_MSR_VP_RUNTIME:
+               if (!host)
+                       return 1;
+               hv->runtime_offset = data - current_task_runtime_100ns();
+               break;
+       default:
+               vcpu_unimpl(vcpu, "Hyper-V uhandled wrmsr: 0x%x data 0x%llx\n",
+                           msr, data);
+               return 1;
+       }
+
+       return 0;
+}
+
+static int kvm_hv_get_msr_pw(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
+{
+       u64 data = 0;
+       struct kvm *kvm = vcpu->kvm;
+       struct kvm_hv *hv = &kvm->arch.hyperv;
+
+       switch (msr) {
+       case HV_X64_MSR_GUEST_OS_ID:
+               data = hv->hv_guest_os_id;
+               break;
+       case HV_X64_MSR_HYPERCALL:
+               data = hv->hv_hypercall;
+               break;
+       case HV_X64_MSR_TIME_REF_COUNT: {
+               data =
+                    div_u64(get_kernel_ns() + kvm->arch.kvmclock_offset, 100);
+               break;
+       }
+       case HV_X64_MSR_REFERENCE_TSC:
+               data = hv->hv_tsc_page;
+               break;
+       case HV_X64_MSR_CRASH_P0 ... HV_X64_MSR_CRASH_P4:
+               return kvm_hv_msr_get_crash_data(vcpu,
+                                                msr - HV_X64_MSR_CRASH_P0,
+                                                pdata);
+       case HV_X64_MSR_CRASH_CTL:
+               return kvm_hv_msr_get_crash_ctl(vcpu, pdata);
+       case HV_X64_MSR_RESET:
+               data = 0;
+               break;
+       default:
+               vcpu_unimpl(vcpu, "Hyper-V unhandled rdmsr: 0x%x\n", msr);
+               return 1;
+       }
+
+       *pdata = data;
+       return 0;
+}
+
+static int kvm_hv_get_msr(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
+{
+       u64 data = 0;
+       struct kvm_vcpu_hv *hv = &vcpu->arch.hyperv;
+
+       switch (msr) {
+       case HV_X64_MSR_VP_INDEX: {
+               int r;
+               struct kvm_vcpu *v;
+
+               kvm_for_each_vcpu(r, v, vcpu->kvm) {
+                       if (v == vcpu) {
+                               data = r;
+                               break;
+                       }
+               }
+               break;
+       }
+       case HV_X64_MSR_EOI:
+               return kvm_hv_vapic_msr_read(vcpu, APIC_EOI, pdata);
+       case HV_X64_MSR_ICR:
+               return kvm_hv_vapic_msr_read(vcpu, APIC_ICR, pdata);
+       case HV_X64_MSR_TPR:
+               return kvm_hv_vapic_msr_read(vcpu, APIC_TASKPRI, pdata);
+       case HV_X64_MSR_APIC_ASSIST_PAGE:
+               data = hv->hv_vapic;
+               break;
+       case HV_X64_MSR_VP_RUNTIME:
+               data = current_task_runtime_100ns() + hv->runtime_offset;
+               break;
+       default:
+               vcpu_unimpl(vcpu, "Hyper-V unhandled rdmsr: 0x%x\n", msr);
+               return 1;
+       }
+       *pdata = data;
+       return 0;
+}
+
+int kvm_hv_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data, bool host)
+{
+       if (kvm_hv_msr_partition_wide(msr)) {
+               int r;
+
+               mutex_lock(&vcpu->kvm->lock);
+               r = kvm_hv_set_msr_pw(vcpu, msr, data, host);
+               mutex_unlock(&vcpu->kvm->lock);
+               return r;
+       } else
+               return kvm_hv_set_msr(vcpu, msr, data, host);
+}
+
+int kvm_hv_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
+{
+       if (kvm_hv_msr_partition_wide(msr)) {
+               int r;
+
+               mutex_lock(&vcpu->kvm->lock);
+               r = kvm_hv_get_msr_pw(vcpu, msr, pdata);
+               mutex_unlock(&vcpu->kvm->lock);
+               return r;
+       } else
+               return kvm_hv_get_msr(vcpu, msr, pdata);
+}
+
+bool kvm_hv_hypercall_enabled(struct kvm *kvm)
+{
+       return kvm->arch.hyperv.hv_hypercall & HV_X64_MSR_HYPERCALL_ENABLE;
+}
+
+int kvm_hv_hypercall(struct kvm_vcpu *vcpu)
+{
+       u64 param, ingpa, outgpa, ret;
+       uint16_t code, rep_idx, rep_cnt, res = HV_STATUS_SUCCESS, rep_done = 0;
+       bool fast, longmode;
+
+       /*
+        * hypercall generates UD from non zero cpl and real mode
+        * per HYPER-V spec
+        */
+       if (kvm_x86_ops->get_cpl(vcpu) != 0 || !is_protmode(vcpu)) {
+               kvm_queue_exception(vcpu, UD_VECTOR);
+               return 0;
+       }
+
+       longmode = is_64_bit_mode(vcpu);
+
+       if (!longmode) {
+               param = ((u64)kvm_register_read(vcpu, VCPU_REGS_RDX) << 32) |
+                       (kvm_register_read(vcpu, VCPU_REGS_RAX) & 0xffffffff);
+               ingpa = ((u64)kvm_register_read(vcpu, VCPU_REGS_RBX) << 32) |
+                       (kvm_register_read(vcpu, VCPU_REGS_RCX) & 0xffffffff);
+               outgpa = ((u64)kvm_register_read(vcpu, VCPU_REGS_RDI) << 32) |
+                       (kvm_register_read(vcpu, VCPU_REGS_RSI) & 0xffffffff);
+       }
+#ifdef CONFIG_X86_64
+       else {
+               param = kvm_register_read(vcpu, VCPU_REGS_RCX);
+               ingpa = kvm_register_read(vcpu, VCPU_REGS_RDX);
+               outgpa = kvm_register_read(vcpu, VCPU_REGS_R8);
+       }
+#endif
+
+       code = param & 0xffff;
+       fast = (param >> 16) & 0x1;
+       rep_cnt = (param >> 32) & 0xfff;
+       rep_idx = (param >> 48) & 0xfff;
+
+       trace_kvm_hv_hypercall(code, fast, rep_cnt, rep_idx, ingpa, outgpa);
+
+       switch (code) {
+       case HV_X64_HV_NOTIFY_LONG_SPIN_WAIT:
+               kvm_vcpu_on_spin(vcpu);
+               break;
+       default:
+               res = HV_STATUS_INVALID_HYPERCALL_CODE;
+               break;
+       }
+
+       ret = res | (((u64)rep_done & 0xfff) << 32);
+       if (longmode) {
+               kvm_register_write(vcpu, VCPU_REGS_RAX, ret);
+       } else {
+               kvm_register_write(vcpu, VCPU_REGS_RDX, ret >> 32);
+               kvm_register_write(vcpu, VCPU_REGS_RAX, ret & 0xffffffff);
+       }
+
+       return 1;
+}