prev_nr = vcpu->arch.exception.nr;
if (prev_nr == DF_VECTOR) {
/* triple fault -> shutdown */
- set_bit(KVM_REQ_TRIPLE_FAULT, &vcpu->requests);
+ kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu);
return;
}
class1 = exception_class(prev_nr);
HV_X64_MSR_GUEST_OS_ID, HV_X64_MSR_HYPERCALL,
HV_X64_MSR_APIC_ASSIST_PAGE,
MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP,
- MSR_K6_STAR,
+ MSR_STAR,
#ifdef CONFIG_X86_64
MSR_CSTAR, MSR_KERNEL_GS_BASE, MSR_SYSCALL_MASK, MSR_LSTAR,
#endif
static u32 emulated_msrs[] = {
MSR_IA32_MISC_ENABLE,
+ MSR_IA32_MCG_STATUS,
+ MSR_IA32_MCG_CTL,
};
static int set_efer(struct kvm_vcpu *vcpu, u64 efer)
if (!vcpu->time_page)
return 0;
- set_bit(KVM_REQ_KVMCLOCK_UPDATE, &v->requests);
+ kvm_make_request(KVM_REQ_KVMCLOCK_UPDATE, v);
return 1;
}
return r;
}
+static void wbinvd_ipi(void *garbage)
+{
+ wbinvd();
+}
+
+static bool need_emulate_wbinvd(struct kvm_vcpu *vcpu)
+{
+ return vcpu->kvm->arch.iommu_domain &&
+ !(vcpu->kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY);
+}
+
void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
+ /* Address WBINVD may be executed by guest */
+ if (need_emulate_wbinvd(vcpu)) {
+ if (kvm_x86_ops->has_wbinvd_exit())
+ cpumask_set_cpu(cpu, vcpu->arch.wbinvd_dirty_mask);
+ else if (vcpu->cpu != -1 && vcpu->cpu != cpu)
+ smp_call_function_single(vcpu->cpu,
+ wbinvd_ipi, NULL, 1);
+ }
+
kvm_x86_ops->vcpu_load(vcpu, cpu);
if (unlikely(per_cpu(cpu_tsc_khz, cpu) == 0)) {
unsigned long khz = cpufreq_quick_get(cpu);
0 /* Reserved */ | f_lm | F(3DNOWEXT) | F(3DNOW);
/* cpuid 1.ecx */
const u32 kvm_supported_word4_x86_features =
- F(XMM3) | 0 /* Reserved, DTES64, MONITOR */ |
+ F(XMM3) | F(PCLMULQDQ) | 0 /* DTES64, MONITOR */ |
0 /* DS-CPL, VMX, SMX, EST */ |
0 /* TM2 */ | F(SSSE3) | 0 /* CNXT-ID */ | 0 /* Reserved */ |
0 /* Reserved */ | F(CX16) | 0 /* xTPR Update, PDCM */ |
0 /* Reserved, DCA */ | F(XMM4_1) |
F(XMM4_2) | F(X2APIC) | F(MOVBE) | F(POPCNT) |
- 0 /* Reserved, AES */ | F(XSAVE) | 0 /* OSXSAVE */;
+ 0 /* Reserved, AES */ | F(XSAVE) | 0 /* OSXSAVE */ | F(AVX);
/* cpuid 0x80000001.ecx */
const u32 kvm_supported_word6_x86_features =
F(LAHF_LM) | F(CMP_LEGACY) | F(SVM) | 0 /* ExtApicSpace */ |
printk(KERN_DEBUG "kvm: set_mce: "
"injects mce exception while "
"previous one is in progress!\n");
- set_bit(KVM_REQ_TRIPLE_FAULT, &vcpu->requests);
+ kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu);
return 0;
}
if (banks[1] & MCI_STATUS_VAL)
struct kvm_vcpu *vcpu = filp->private_data;
void __user *argp = (void __user *)arg;
int r;
- struct kvm_lapic_state *lapic = NULL;
- struct kvm_xsave *xsave = NULL;
- struct kvm_xcrs *xcrs = NULL;
+ union {
+ struct kvm_lapic_state *lapic;
+ struct kvm_xsave *xsave;
+ struct kvm_xcrs *xcrs;
+ void *buffer;
+ } u;
+ u.buffer = NULL;
switch (ioctl) {
case KVM_GET_LAPIC: {
r = -EINVAL;
if (!vcpu->arch.apic)
goto out;
- lapic = kzalloc(sizeof(struct kvm_lapic_state), GFP_KERNEL);
+ u.lapic = kzalloc(sizeof(struct kvm_lapic_state), GFP_KERNEL);
r = -ENOMEM;
- if (!lapic)
+ if (!u.lapic)
goto out;
- r = kvm_vcpu_ioctl_get_lapic(vcpu, lapic);
+ r = kvm_vcpu_ioctl_get_lapic(vcpu, u.lapic);
if (r)
goto out;
r = -EFAULT;
- if (copy_to_user(argp, lapic, sizeof(struct kvm_lapic_state)))
+ if (copy_to_user(argp, u.lapic, sizeof(struct kvm_lapic_state)))
goto out;
r = 0;
break;
r = -EINVAL;
if (!vcpu->arch.apic)
goto out;
- lapic = kmalloc(sizeof(struct kvm_lapic_state), GFP_KERNEL);
+ u.lapic = kmalloc(sizeof(struct kvm_lapic_state), GFP_KERNEL);
r = -ENOMEM;
- if (!lapic)
+ if (!u.lapic)
goto out;
r = -EFAULT;
- if (copy_from_user(lapic, argp, sizeof(struct kvm_lapic_state)))
+ if (copy_from_user(u.lapic, argp, sizeof(struct kvm_lapic_state)))
goto out;
- r = kvm_vcpu_ioctl_set_lapic(vcpu, lapic);
+ r = kvm_vcpu_ioctl_set_lapic(vcpu, u.lapic);
if (r)
goto out;
r = 0;
break;
}
case KVM_GET_XSAVE: {
- xsave = kzalloc(sizeof(struct kvm_xsave), GFP_KERNEL);
+ u.xsave = kzalloc(sizeof(struct kvm_xsave), GFP_KERNEL);
r = -ENOMEM;
- if (!xsave)
+ if (!u.xsave)
break;
- kvm_vcpu_ioctl_x86_get_xsave(vcpu, xsave);
+ kvm_vcpu_ioctl_x86_get_xsave(vcpu, u.xsave);
r = -EFAULT;
- if (copy_to_user(argp, xsave, sizeof(struct kvm_xsave)))
+ if (copy_to_user(argp, u.xsave, sizeof(struct kvm_xsave)))
break;
r = 0;
break;
}
case KVM_SET_XSAVE: {
- xsave = kzalloc(sizeof(struct kvm_xsave), GFP_KERNEL);
+ u.xsave = kzalloc(sizeof(struct kvm_xsave), GFP_KERNEL);
r = -ENOMEM;
- if (!xsave)
+ if (!u.xsave)
break;
r = -EFAULT;
- if (copy_from_user(xsave, argp, sizeof(struct kvm_xsave)))
+ if (copy_from_user(u.xsave, argp, sizeof(struct kvm_xsave)))
break;
- r = kvm_vcpu_ioctl_x86_set_xsave(vcpu, xsave);
+ r = kvm_vcpu_ioctl_x86_set_xsave(vcpu, u.xsave);
break;
}
case KVM_GET_XCRS: {
- xcrs = kzalloc(sizeof(struct kvm_xcrs), GFP_KERNEL);
+ u.xcrs = kzalloc(sizeof(struct kvm_xcrs), GFP_KERNEL);
r = -ENOMEM;
- if (!xcrs)
+ if (!u.xcrs)
break;
- kvm_vcpu_ioctl_x86_get_xcrs(vcpu, xcrs);
+ kvm_vcpu_ioctl_x86_get_xcrs(vcpu, u.xcrs);
r = -EFAULT;
- if (copy_to_user(argp, xcrs,
+ if (copy_to_user(argp, u.xcrs,
sizeof(struct kvm_xcrs)))
break;
r = 0;
break;
}
case KVM_SET_XCRS: {
- xcrs = kzalloc(sizeof(struct kvm_xcrs), GFP_KERNEL);
+ u.xcrs = kzalloc(sizeof(struct kvm_xcrs), GFP_KERNEL);
r = -ENOMEM;
- if (!xcrs)
+ if (!u.xcrs)
break;
r = -EFAULT;
- if (copy_from_user(xcrs, argp,
+ if (copy_from_user(u.xcrs, argp,
sizeof(struct kvm_xcrs)))
break;
- r = kvm_vcpu_ioctl_x86_set_xcrs(vcpu, xcrs);
+ r = kvm_vcpu_ioctl_x86_set_xcrs(vcpu, u.xcrs);
break;
}
default:
r = -EINVAL;
}
out:
- kfree(lapic);
- kfree(xsave);
- kfree(xcrs);
+ kfree(u.buffer);
return r;
}
return kvm->arch.n_alloc_mmu_pages;
}
-gfn_t unalias_gfn_instantiation(struct kvm *kvm, gfn_t gfn)
-{
- int i;
- struct kvm_mem_alias *alias;
- struct kvm_mem_aliases *aliases;
-
- aliases = kvm_aliases(kvm);
-
- for (i = 0; i < aliases->naliases; ++i) {
- alias = &aliases->aliases[i];
- if (alias->flags & KVM_ALIAS_INVALID)
- continue;
- if (gfn >= alias->base_gfn
- && gfn < alias->base_gfn + alias->npages)
- return alias->target_gfn + gfn - alias->base_gfn;
- }
- return gfn;
-}
-
-gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn)
-{
- int i;
- struct kvm_mem_alias *alias;
- struct kvm_mem_aliases *aliases;
-
- aliases = kvm_aliases(kvm);
-
- for (i = 0; i < aliases->naliases; ++i) {
- alias = &aliases->aliases[i];
- if (gfn >= alias->base_gfn
- && gfn < alias->base_gfn + alias->npages)
- return alias->target_gfn + gfn - alias->base_gfn;
- }
- return gfn;
-}
-
-/*
- * Set a new alias region. Aliases map a portion of physical memory into
- * another portion. This is useful for memory windows, for example the PC
- * VGA region.
- */
-static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm,
- struct kvm_memory_alias *alias)
-{
- int r, n;
- struct kvm_mem_alias *p;
- struct kvm_mem_aliases *aliases, *old_aliases;
-
- r = -EINVAL;
- /* General sanity checks */
- if (alias->memory_size & (PAGE_SIZE - 1))
- goto out;
- if (alias->guest_phys_addr & (PAGE_SIZE - 1))
- goto out;
- if (alias->slot >= KVM_ALIAS_SLOTS)
- goto out;
- if (alias->guest_phys_addr + alias->memory_size
- < alias->guest_phys_addr)
- goto out;
- if (alias->target_phys_addr + alias->memory_size
- < alias->target_phys_addr)
- goto out;
-
- r = -ENOMEM;
- aliases = kzalloc(sizeof(struct kvm_mem_aliases), GFP_KERNEL);
- if (!aliases)
- goto out;
-
- mutex_lock(&kvm->slots_lock);
-
- /* invalidate any gfn reference in case of deletion/shrinking */
- memcpy(aliases, kvm->arch.aliases, sizeof(struct kvm_mem_aliases));
- aliases->aliases[alias->slot].flags |= KVM_ALIAS_INVALID;
- old_aliases = kvm->arch.aliases;
- rcu_assign_pointer(kvm->arch.aliases, aliases);
- synchronize_srcu_expedited(&kvm->srcu);
- kvm_mmu_zap_all(kvm);
- kfree(old_aliases);
-
- r = -ENOMEM;
- aliases = kzalloc(sizeof(struct kvm_mem_aliases), GFP_KERNEL);
- if (!aliases)
- goto out_unlock;
-
- memcpy(aliases, kvm->arch.aliases, sizeof(struct kvm_mem_aliases));
-
- p = &aliases->aliases[alias->slot];
- p->base_gfn = alias->guest_phys_addr >> PAGE_SHIFT;
- p->npages = alias->memory_size >> PAGE_SHIFT;
- p->target_gfn = alias->target_phys_addr >> PAGE_SHIFT;
- p->flags &= ~(KVM_ALIAS_INVALID);
-
- for (n = KVM_ALIAS_SLOTS; n > 0; --n)
- if (aliases->aliases[n - 1].npages)
- break;
- aliases->naliases = n;
-
- old_aliases = kvm->arch.aliases;
- rcu_assign_pointer(kvm->arch.aliases, aliases);
- synchronize_srcu_expedited(&kvm->srcu);
- kfree(old_aliases);
- r = 0;
-
-out_unlock:
- mutex_unlock(&kvm->slots_lock);
-out:
- return r;
-}
-
static int kvm_vm_ioctl_get_irqchip(struct kvm *kvm, struct kvm_irqchip *chip)
{
int r;
union {
struct kvm_pit_state ps;
struct kvm_pit_state2 ps2;
- struct kvm_memory_alias alias;
struct kvm_pit_config pit_config;
} u;
goto out;
break;
}
- case KVM_SET_MEMORY_REGION: {
- struct kvm_memory_region kvm_mem;
- struct kvm_userspace_memory_region kvm_userspace_mem;
-
- r = -EFAULT;
- if (copy_from_user(&kvm_mem, argp, sizeof kvm_mem))
- goto out;
- kvm_userspace_mem.slot = kvm_mem.slot;
- kvm_userspace_mem.flags = kvm_mem.flags;
- kvm_userspace_mem.guest_phys_addr = kvm_mem.guest_phys_addr;
- kvm_userspace_mem.memory_size = kvm_mem.memory_size;
- r = kvm_vm_ioctl_set_memory_region(kvm, &kvm_userspace_mem, 0);
- if (r)
- goto out;
- break;
- }
case KVM_SET_NR_MMU_PAGES:
r = kvm_vm_ioctl_set_nr_mmu_pages(kvm, arg);
if (r)
case KVM_GET_NR_MMU_PAGES:
r = kvm_vm_ioctl_get_nr_mmu_pages(kvm);
break;
- case KVM_SET_MEMORY_ALIAS:
- r = -EFAULT;
- if (copy_from_user(&u.alias, argp, sizeof(struct kvm_memory_alias)))
- goto out;
- r = kvm_vm_ioctl_set_memory_alias(kvm, &u.alias);
- if (r)
- goto out;
- break;
case KVM_CREATE_IRQCHIP: {
struct kvm_pic *vpic;
goto emul_write;
page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
+ if (is_error_page(page)) {
+ kvm_release_page_clean(page);
+ goto emul_write;
+ }
kaddr = kmap_atomic(page, KM_USER0);
kaddr += offset_in_page(gpa);
return X86EMUL_CONTINUE;
}
+int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu)
+{
+ if (!need_emulate_wbinvd(vcpu))
+ return X86EMUL_CONTINUE;
+
+ if (kvm_x86_ops->has_wbinvd_exit()) {
+ smp_call_function_many(vcpu->arch.wbinvd_dirty_mask,
+ wbinvd_ipi, NULL, 1);
+ cpumask_clear(vcpu->arch.wbinvd_dirty_mask);
+ }
+ wbinvd();
+ return X86EMUL_CONTINUE;
+}
+EXPORT_SYMBOL_GPL(kvm_emulate_wbinvd);
+
int emulate_clts(struct kvm_vcpu *vcpu)
{
kvm_x86_ops->set_cr0(vcpu, kvm_read_cr0_bits(vcpu, ~X86_CR0_TS));
return EMULATE_FAIL;
}
+static bool reexecute_instruction(struct kvm_vcpu *vcpu, gva_t gva)
+{
+ gpa_t gpa;
+
+ if (tdp_enabled)
+ return false;
+
+ /*
+ * if emulation was due to access to shadowed page table
+ * and it failed try to unshadow page and re-entetr the
+ * guest to let CPU execute the instruction.
+ */
+ if (kvm_mmu_unprotect_page_virt(vcpu, gva))
+ return true;
+
+ gpa = kvm_mmu_gva_to_gpa_system(vcpu, gva, NULL);
+
+ if (gpa == UNMAPPED_GVA)
+ return true; /* let cpu generate fault */
+
+ if (!kvm_is_error_hva(gfn_to_hva(vcpu->kvm, gpa >> PAGE_SHIFT)))
+ return true;
+
+ return false;
+}
+
int emulate_instruction(struct kvm_vcpu *vcpu,
unsigned long cr2,
u16 error_code,
++vcpu->stat.insn_emulation;
if (r) {
- if (kvm_mmu_unprotect_page_virt(vcpu, cr2))
+ if (reexecute_instruction(vcpu, cr2))
return EMULATE_DONE;
if (emulation_type & EMULTYPE_SKIP)
return EMULATE_FAIL;
r = x86_emulate_insn(&vcpu->arch.emulate_ctxt, &emulate_ops);
if (r) { /* emulation failed */
- /*
- * if emulation was due to access to shadowed page table
- * and it failed try to unshadow page and re-entetr the
- * guest to let CPU execute the instruction.
- */
- if (kvm_mmu_unprotect_page_virt(vcpu, cr2))
+ if (reexecute_instruction(vcpu, cr2))
return EMULATE_DONE;
return handle_emulation_failure(vcpu);
bool req_int_win = !irqchip_in_kernel(vcpu->kvm) &&
vcpu->run->request_interrupt_window;
- if (vcpu->requests)
- if (test_and_clear_bit(KVM_REQ_MMU_RELOAD, &vcpu->requests))
- kvm_mmu_unload(vcpu);
-
- r = kvm_mmu_reload(vcpu);
- if (unlikely(r))
- goto out;
-
if (vcpu->requests) {
- if (test_and_clear_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests))
+ if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu))
+ kvm_mmu_unload(vcpu);
+ if (kvm_check_request(KVM_REQ_MIGRATE_TIMER, vcpu))
__kvm_migrate_timers(vcpu);
- if (test_and_clear_bit(KVM_REQ_KVMCLOCK_UPDATE, &vcpu->requests))
+ if (kvm_check_request(KVM_REQ_KVMCLOCK_UPDATE, vcpu))
kvm_write_guest_time(vcpu);
- if (test_and_clear_bit(KVM_REQ_MMU_SYNC, &vcpu->requests))
+ if (kvm_check_request(KVM_REQ_MMU_SYNC, vcpu))
kvm_mmu_sync_roots(vcpu);
- if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
+ if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu))
kvm_x86_ops->tlb_flush(vcpu);
- if (test_and_clear_bit(KVM_REQ_REPORT_TPR_ACCESS,
- &vcpu->requests)) {
+ if (kvm_check_request(KVM_REQ_REPORT_TPR_ACCESS, vcpu)) {
vcpu->run->exit_reason = KVM_EXIT_TPR_ACCESS;
r = 0;
goto out;
}
- if (test_and_clear_bit(KVM_REQ_TRIPLE_FAULT, &vcpu->requests)) {
+ if (kvm_check_request(KVM_REQ_TRIPLE_FAULT, vcpu)) {
vcpu->run->exit_reason = KVM_EXIT_SHUTDOWN;
r = 0;
goto out;
}
- if (test_and_clear_bit(KVM_REQ_DEACTIVATE_FPU, &vcpu->requests)) {
+ if (kvm_check_request(KVM_REQ_DEACTIVATE_FPU, vcpu)) {
vcpu->fpu_active = 0;
kvm_x86_ops->fpu_deactivate(vcpu);
}
}
+ r = kvm_mmu_reload(vcpu);
+ if (unlikely(r))
+ goto out;
+
preempt_disable();
kvm_x86_ops->prepare_guest_switch(vcpu);
srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx);
kvm_vcpu_block(vcpu);
vcpu->srcu_idx = srcu_read_lock(&kvm->srcu);
- if (test_and_clear_bit(KVM_REQ_UNHALT, &vcpu->requests))
+ if (kvm_check_request(KVM_REQ_UNHALT, vcpu))
{
switch(vcpu->arch.mp_state) {
case KVM_MP_STATE_HALTED:
vcpu->guest_fpu_loaded = 0;
fpu_save_init(&vcpu->arch.guest_fpu);
++vcpu->stat.fpu_reload;
- set_bit(KVM_REQ_DEACTIVATE_FPU, &vcpu->requests);
+ kvm_make_request(KVM_REQ_DEACTIVATE_FPU, vcpu);
trace_kvm_fpu(0);
}
vcpu->arch.time_page = NULL;
}
+ free_cpumask_var(vcpu->arch.wbinvd_dirty_mask);
fx_free(vcpu);
kvm_x86_ops->vcpu_free(vcpu);
}
}
vcpu->arch.mcg_cap = KVM_MAX_MCE_BANKS;
+ if (!zalloc_cpumask_var(&vcpu->arch.wbinvd_dirty_mask, GFP_KERNEL))
+ goto fail_free_mce_banks;
+
return 0;
+fail_free_mce_banks:
+ kfree(vcpu->arch.mce_banks);
fail_free_lapic:
kvm_free_lapic(vcpu);
fail_mmu_destroy:
if (!kvm)
return ERR_PTR(-ENOMEM);
- kvm->arch.aliases = kzalloc(sizeof(struct kvm_mem_aliases), GFP_KERNEL);
- if (!kvm->arch.aliases) {
- kfree(kvm);
- return ERR_PTR(-ENOMEM);
- }
-
INIT_LIST_HEAD(&kvm->arch.active_mmu_pages);
INIT_LIST_HEAD(&kvm->arch.assigned_dev_head);
void kvm_arch_sync_events(struct kvm *kvm)
{
kvm_free_all_assigned_devices(kvm);
+ kvm_free_pit(kvm);
}
void kvm_arch_destroy_vm(struct kvm *kvm)
{
kvm_iommu_unmap_guest(kvm);
- kvm_free_pit(kvm);
kfree(kvm->arch.vpic);
kfree(kvm->arch.vioapic);
kvm_free_vcpus(kvm);
if (kvm->arch.ept_identity_pagetable)
put_page(kvm->arch.ept_identity_pagetable);
cleanup_srcu_struct(&kvm->srcu);
- kfree(kvm->arch.aliases);
kfree(kvm);
}
int user_alloc)
{
int npages = memslot->npages;
+ int map_flags = MAP_PRIVATE | MAP_ANONYMOUS;
+
+ /* Prevent internal slot pages from being moved by fork()/COW. */
+ if (memslot->id >= KVM_MEMORY_SLOTS)
+ map_flags = MAP_SHARED | MAP_ANONYMOUS;
/*To keep backward compatibility with older userspace,
*x86 needs to hanlde !user_alloc case.
userspace_addr = do_mmap(NULL, 0,
npages * PAGE_SIZE,
PROT_READ | PROT_WRITE,
- MAP_PRIVATE | MAP_ANONYMOUS,
+ map_flags,
0);
up_write(¤t->mm->mmap_sem);
projects / mv-sheeva.git / blobdiff