* derived from drivers/kvm/kvm_main.c
*
* Copyright (C) 2006 Qumranet, Inc.
+ * Copyright (C) 2008 Qumranet, Inc.
+ * Copyright IBM Corporation, 2008
*
* Authors:
* Avi Kivity <avi@qumranet.com>
* Yaniv Kamay <yaniv@qumranet.com>
+ * Amit Shah <amit.shah@qumranet.com>
+ * Ben-Ami Yassour <benami@il.ibm.com>
*
* This work is licensed under the terms of the GNU GPL, version 2. See
* the COPYING file in the top-level directory.
#include "mmu.h"
#include "i8254.h"
#include "tss.h"
+#include "kvm_cache_regs.h"
+#include "x86.h"
#include <linux/clocksource.h>
+#include <linux/interrupt.h>
#include <linux/kvm.h>
#include <linux/fs.h>
+#include <linux/pci.h>
#include <linux/vmalloc.h>
#include <linux/module.h>
#include <linux/mman.h>
struct kvm_cpuid_entry2 __user *entries);
struct kvm_x86_ops *kvm_x86_ops;
+EXPORT_SYMBOL_GPL(kvm_x86_ops);
struct kvm_stats_debugfs_item debugfs_entries[] = {
{ "pf_fixed", VCPU_STAT(pf_fixed) },
{ NULL }
};
+static struct kvm_assigned_dev_kernel *kvm_find_assigned_dev(struct list_head *head,
+ int assigned_dev_id)
+{
+ struct list_head *ptr;
+ struct kvm_assigned_dev_kernel *match;
+
+ list_for_each(ptr, head) {
+ match = list_entry(ptr, struct kvm_assigned_dev_kernel, list);
+ if (match->assigned_dev_id == assigned_dev_id)
+ return match;
+ }
+ return NULL;
+}
+
+static void kvm_assigned_dev_interrupt_work_handler(struct work_struct *work)
+{
+ struct kvm_assigned_dev_kernel *assigned_dev;
+
+ assigned_dev = container_of(work, struct kvm_assigned_dev_kernel,
+ interrupt_work);
+
+ /* This is taken to safely inject irq inside the guest. When
+ * the interrupt injection (or the ioapic code) uses a
+ * finer-grained lock, update this
+ */
+ mutex_lock(&assigned_dev->kvm->lock);
+ kvm_set_irq(assigned_dev->kvm,
+ assigned_dev->guest_irq, 1);
+ mutex_unlock(&assigned_dev->kvm->lock);
+ kvm_put_kvm(assigned_dev->kvm);
+}
+
+/* FIXME: Implement the OR logic needed to make shared interrupts on
+ * this line behave properly
+ */
+static irqreturn_t kvm_assigned_dev_intr(int irq, void *dev_id)
+{
+ struct kvm_assigned_dev_kernel *assigned_dev =
+ (struct kvm_assigned_dev_kernel *) dev_id;
+
+ kvm_get_kvm(assigned_dev->kvm);
+ schedule_work(&assigned_dev->interrupt_work);
+ disable_irq_nosync(irq);
+ return IRQ_HANDLED;
+}
+
+/* Ack the irq line for an assigned device */
+static void kvm_assigned_dev_ack_irq(struct kvm_irq_ack_notifier *kian)
+{
+ struct kvm_assigned_dev_kernel *dev;
+
+ if (kian->gsi == -1)
+ return;
+
+ dev = container_of(kian, struct kvm_assigned_dev_kernel,
+ ack_notifier);
+ kvm_set_irq(dev->kvm, dev->guest_irq, 0);
+ enable_irq(dev->host_irq);
+}
+
+static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
+ struct kvm_assigned_irq
+ *assigned_irq)
+{
+ int r = 0;
+ struct kvm_assigned_dev_kernel *match;
+
+ mutex_lock(&kvm->lock);
+
+ match = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head,
+ assigned_irq->assigned_dev_id);
+ if (!match) {
+ mutex_unlock(&kvm->lock);
+ return -EINVAL;
+ }
+
+ if (match->irq_requested) {
+ match->guest_irq = assigned_irq->guest_irq;
+ match->ack_notifier.gsi = assigned_irq->guest_irq;
+ mutex_unlock(&kvm->lock);
+ return 0;
+ }
+
+ INIT_WORK(&match->interrupt_work,
+ kvm_assigned_dev_interrupt_work_handler);
+
+ if (irqchip_in_kernel(kvm)) {
+ if (!capable(CAP_SYS_RAWIO)) {
+ return -EPERM;
+ goto out;
+ }
+
+ if (assigned_irq->host_irq)
+ match->host_irq = assigned_irq->host_irq;
+ else
+ match->host_irq = match->dev->irq;
+ match->guest_irq = assigned_irq->guest_irq;
+ match->ack_notifier.gsi = assigned_irq->guest_irq;
+ match->ack_notifier.irq_acked = kvm_assigned_dev_ack_irq;
+ kvm_register_irq_ack_notifier(kvm, &match->ack_notifier);
+
+ /* Even though this is PCI, we don't want to use shared
+ * interrupts. Sharing host devices with guest-assigned devices
+ * on the same interrupt line is not a happy situation: there
+ * are going to be long delays in accepting, acking, etc.
+ */
+ if (request_irq(match->host_irq, kvm_assigned_dev_intr, 0,
+ "kvm_assigned_device", (void *)match)) {
+ printk(KERN_INFO "%s: couldn't allocate irq for pv "
+ "device\n", __func__);
+ r = -EIO;
+ goto out;
+ }
+ }
+
+ match->irq_requested = true;
+out:
+ mutex_unlock(&kvm->lock);
+ return r;
+}
+
+static int kvm_vm_ioctl_assign_device(struct kvm *kvm,
+ struct kvm_assigned_pci_dev *assigned_dev)
+{
+ int r = 0;
+ struct kvm_assigned_dev_kernel *match;
+ struct pci_dev *dev;
+
+ mutex_lock(&kvm->lock);
+
+ match = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head,
+ assigned_dev->assigned_dev_id);
+ if (match) {
+ /* device already assigned */
+ r = -EINVAL;
+ goto out;
+ }
+
+ match = kzalloc(sizeof(struct kvm_assigned_dev_kernel), GFP_KERNEL);
+ if (match == NULL) {
+ printk(KERN_INFO "%s: Couldn't allocate memory\n",
+ __func__);
+ r = -ENOMEM;
+ goto out;
+ }
+ dev = pci_get_bus_and_slot(assigned_dev->busnr,
+ assigned_dev->devfn);
+ if (!dev) {
+ printk(KERN_INFO "%s: host device not found\n", __func__);
+ r = -EINVAL;
+ goto out_free;
+ }
+ if (pci_enable_device(dev)) {
+ printk(KERN_INFO "%s: Could not enable PCI device\n", __func__);
+ r = -EBUSY;
+ goto out_put;
+ }
+ r = pci_request_regions(dev, "kvm_assigned_device");
+ if (r) {
+ printk(KERN_INFO "%s: Could not get access to device regions\n",
+ __func__);
+ goto out_disable;
+ }
+ match->assigned_dev_id = assigned_dev->assigned_dev_id;
+ match->host_busnr = assigned_dev->busnr;
+ match->host_devfn = assigned_dev->devfn;
+ match->dev = dev;
+
+ match->kvm = kvm;
+
+ list_add(&match->list, &kvm->arch.assigned_dev_head);
+
+out:
+ mutex_unlock(&kvm->lock);
+ return r;
+out_disable:
+ pci_disable_device(dev);
+out_put:
+ pci_dev_put(dev);
+out_free:
+ kfree(match);
+ mutex_unlock(&kvm->lock);
+ return r;
+}
+
+static void kvm_free_assigned_devices(struct kvm *kvm)
+{
+ struct list_head *ptr, *ptr2;
+ struct kvm_assigned_dev_kernel *assigned_dev;
+
+ list_for_each_safe(ptr, ptr2, &kvm->arch.assigned_dev_head) {
+ assigned_dev = list_entry(ptr,
+ struct kvm_assigned_dev_kernel,
+ list);
+
+ if (irqchip_in_kernel(kvm) && assigned_dev->irq_requested) {
+ free_irq(assigned_dev->host_irq,
+ (void *)assigned_dev);
+
+ kvm_unregister_irq_ack_notifier(kvm,
+ &assigned_dev->
+ ack_notifier);
+ }
+
+ if (cancel_work_sync(&assigned_dev->interrupt_work))
+ /* We had pending work. That means we will have to take
+ * care of kvm_put_kvm.
+ */
+ kvm_put_kvm(kvm);
+
+ pci_release_regions(assigned_dev->dev);
+ pci_disable_device(assigned_dev->dev);
+ pci_dev_put(assigned_dev->dev);
+
+ list_del(&assigned_dev->list);
+ kfree(assigned_dev);
+ }
+}
unsigned long segment_base(u16 selector)
{
pr_unimpl(vcpu, "%s: MSR_IA32_MCG_CTL 0x%llx, nop\n",
__func__, data);
break;
+ case MSR_IA32_DEBUGCTLMSR:
+ if (!data) {
+ /* We support the non-activated case already */
+ break;
+ } else if (data & ~(DEBUGCTLMSR_LBR | DEBUGCTLMSR_BTF)) {
+ /* Values other than LBR and BTF are vendor-specific,
+ thus reserved and should throw a #GP */
+ return 1;
+ }
+ pr_unimpl(vcpu, "%s: MSR_IA32_DEBUGCTLMSR 0x%llx, nop\n",
+ __func__, data);
+ break;
case MSR_IA32_UCODE_REV:
case MSR_IA32_UCODE_WRITE:
break;
case MSR_IA32_MC0_MISC+16:
case MSR_IA32_UCODE_REV:
case MSR_IA32_EBL_CR_POWERON:
+ case MSR_IA32_DEBUGCTLMSR:
+ case MSR_IA32_LASTBRANCHFROMIP:
+ case MSR_IA32_LASTBRANCHTOIP:
+ case MSR_IA32_LASTINTFROMIP:
+ case MSR_IA32_LASTINTTOIP:
data = 0;
break;
case MSR_MTRRcap:
case KVM_CAP_PIT:
case KVM_CAP_NOP_IO_DELAY:
case KVM_CAP_MP_STATE:
+ case KVM_CAP_SYNC_MMU:
r = 1;
break;
case KVM_CAP_COALESCED_MMIO:
struct kvm_vcpu *vcpu = filp->private_data;
void __user *argp = (void __user *)arg;
int r;
+ struct kvm_lapic_state *lapic = NULL;
switch (ioctl) {
case KVM_GET_LAPIC: {
- struct kvm_lapic_state lapic;
+ lapic = kzalloc(sizeof(struct kvm_lapic_state), GFP_KERNEL);
- memset(&lapic, 0, sizeof lapic);
- r = kvm_vcpu_ioctl_get_lapic(vcpu, &lapic);
+ r = -ENOMEM;
+ if (!lapic)
+ goto out;
+ r = kvm_vcpu_ioctl_get_lapic(vcpu, lapic);
if (r)
goto out;
r = -EFAULT;
- if (copy_to_user(argp, &lapic, sizeof lapic))
+ if (copy_to_user(argp, lapic, sizeof(struct kvm_lapic_state)))
goto out;
r = 0;
break;
}
case KVM_SET_LAPIC: {
- struct kvm_lapic_state lapic;
-
+ lapic = kmalloc(sizeof(struct kvm_lapic_state), GFP_KERNEL);
+ r = -ENOMEM;
+ if (!lapic)
+ goto out;
r = -EFAULT;
- if (copy_from_user(&lapic, argp, sizeof lapic))
+ if (copy_from_user(lapic, argp, sizeof(struct kvm_lapic_state)))
goto out;
- r = kvm_vcpu_ioctl_set_lapic(vcpu, &lapic);;
+ r = kvm_vcpu_ioctl_set_lapic(vcpu, lapic);
if (r)
goto out;
r = 0;
r = -EINVAL;
}
out:
+ if (lapic)
+ kfree(lapic);
return r;
}
goto out;
down_write(&kvm->slots_lock);
+ spin_lock(&kvm->mmu_lock);
p = &kvm->arch.aliases[alias->slot];
p->base_gfn = alias->guest_phys_addr >> PAGE_SHIFT;
break;
kvm->arch.naliases = n;
+ spin_unlock(&kvm->mmu_lock);
kvm_mmu_zap_all(kvm);
up_write(&kvm->slots_lock);
struct kvm *kvm = filp->private_data;
void __user *argp = (void __user *)arg;
int r = -EINVAL;
+ /*
+ * This union makes it completely explicit to gcc-3.x
+ * that these two variables' stack usage should be
+ * combined, not added together.
+ */
+ union {
+ struct kvm_pit_state ps;
+ struct kvm_memory_alias alias;
+ } u;
switch (ioctl) {
case KVM_SET_TSS_ADDR:
case KVM_GET_NR_MMU_PAGES:
r = kvm_vm_ioctl_get_nr_mmu_pages(kvm);
break;
- case KVM_SET_MEMORY_ALIAS: {
- struct kvm_memory_alias alias;
-
+ case KVM_SET_MEMORY_ALIAS:
r = -EFAULT;
- if (copy_from_user(&alias, argp, sizeof alias))
+ if (copy_from_user(&u.alias, argp, sizeof(struct kvm_memory_alias)))
goto out;
- r = kvm_vm_ioctl_set_memory_alias(kvm, &alias);
+ r = kvm_vm_ioctl_set_memory_alias(kvm, &u.alias);
if (r)
goto out;
break;
- }
case KVM_CREATE_IRQCHIP:
r = -ENOMEM;
kvm->arch.vpic = kvm_create_pic(kvm);
goto out;
if (irqchip_in_kernel(kvm)) {
mutex_lock(&kvm->lock);
- if (irq_event.irq < 16)
- kvm_pic_set_irq(pic_irqchip(kvm),
- irq_event.irq,
- irq_event.level);
- kvm_ioapic_set_irq(kvm->arch.vioapic,
- irq_event.irq,
- irq_event.level);
+ kvm_set_irq(kvm, irq_event.irq, irq_event.level);
mutex_unlock(&kvm->lock);
r = 0;
}
}
case KVM_GET_IRQCHIP: {
/* 0: PIC master, 1: PIC slave, 2: IOAPIC */
- struct kvm_irqchip chip;
+ struct kvm_irqchip *chip = kmalloc(sizeof(*chip), GFP_KERNEL);
- r = -EFAULT;
- if (copy_from_user(&chip, argp, sizeof chip))
+ r = -ENOMEM;
+ if (!chip)
goto out;
+ r = -EFAULT;
+ if (copy_from_user(chip, argp, sizeof *chip))
+ goto get_irqchip_out;
r = -ENXIO;
if (!irqchip_in_kernel(kvm))
- goto out;
- r = kvm_vm_ioctl_get_irqchip(kvm, &chip);
+ goto get_irqchip_out;
+ r = kvm_vm_ioctl_get_irqchip(kvm, chip);
if (r)
- goto out;
+ goto get_irqchip_out;
r = -EFAULT;
- if (copy_to_user(argp, &chip, sizeof chip))
- goto out;
+ if (copy_to_user(argp, chip, sizeof *chip))
+ goto get_irqchip_out;
r = 0;
+ get_irqchip_out:
+ kfree(chip);
+ if (r)
+ goto out;
break;
}
case KVM_SET_IRQCHIP: {
/* 0: PIC master, 1: PIC slave, 2: IOAPIC */
- struct kvm_irqchip chip;
+ struct kvm_irqchip *chip = kmalloc(sizeof(*chip), GFP_KERNEL);
- r = -EFAULT;
- if (copy_from_user(&chip, argp, sizeof chip))
+ r = -ENOMEM;
+ if (!chip)
goto out;
+ r = -EFAULT;
+ if (copy_from_user(chip, argp, sizeof *chip))
+ goto set_irqchip_out;
r = -ENXIO;
if (!irqchip_in_kernel(kvm))
+ goto set_irqchip_out;
+ r = kvm_vm_ioctl_set_irqchip(kvm, chip);
+ if (r)
+ goto set_irqchip_out;
+ r = 0;
+ set_irqchip_out:
+ kfree(chip);
+ if (r)
goto out;
- r = kvm_vm_ioctl_set_irqchip(kvm, &chip);
+ break;
+ }
+ case KVM_ASSIGN_PCI_DEVICE: {
+ struct kvm_assigned_pci_dev assigned_dev;
+
+ r = -EFAULT;
+ if (copy_from_user(&assigned_dev, argp, sizeof assigned_dev))
+ goto out;
+ r = kvm_vm_ioctl_assign_device(kvm, &assigned_dev);
+ if (r)
+ goto out;
+ break;
+ }
+ case KVM_ASSIGN_IRQ: {
+ struct kvm_assigned_irq assigned_irq;
+
+ r = -EFAULT;
+ if (copy_from_user(&assigned_irq, argp, sizeof assigned_irq))
+ goto out;
+ r = kvm_vm_ioctl_assign_irq(kvm, &assigned_irq);
if (r)
goto out;
- r = 0;
break;
}
case KVM_GET_PIT: {
- struct kvm_pit_state ps;
r = -EFAULT;
- if (copy_from_user(&ps, argp, sizeof ps))
+ if (copy_from_user(&u.ps, argp, sizeof(struct kvm_pit_state)))
goto out;
r = -ENXIO;
if (!kvm->arch.vpit)
goto out;
- r = kvm_vm_ioctl_get_pit(kvm, &ps);
+ r = kvm_vm_ioctl_get_pit(kvm, &u.ps);
if (r)
goto out;
r = -EFAULT;
- if (copy_to_user(argp, &ps, sizeof ps))
+ if (copy_to_user(argp, &u.ps, sizeof(struct kvm_pit_state)))
goto out;
r = 0;
break;
}
case KVM_SET_PIT: {
- struct kvm_pit_state ps;
r = -EFAULT;
- if (copy_from_user(&ps, argp, sizeof ps))
+ if (copy_from_user(&u.ps, argp, sizeof u.ps))
goto out;
r = -ENXIO;
if (!kvm->arch.vpit)
goto out;
- r = kvm_vm_ioctl_set_pit(kvm, &ps);
+ r = kvm_vm_ioctl_set_pit(kvm, &u.ps);
if (r)
goto out;
r = 0;
void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context)
{
u8 opcodes[4];
- unsigned long rip = vcpu->arch.rip;
+ unsigned long rip = kvm_rip_read(vcpu);
unsigned long rip_linear;
if (!printk_ratelimit())
.cmpxchg_emulated = emulator_cmpxchg_emulated,
};
+static void cache_all_regs(struct kvm_vcpu *vcpu)
+{
+ kvm_register_read(vcpu, VCPU_REGS_RAX);
+ kvm_register_read(vcpu, VCPU_REGS_RSP);
+ kvm_register_read(vcpu, VCPU_REGS_RIP);
+ vcpu->arch.regs_dirty = ~0;
+}
+
int emulate_instruction(struct kvm_vcpu *vcpu,
struct kvm_run *run,
unsigned long cr2,
int r;
struct decode_cache *c;
+ kvm_clear_exception_queue(vcpu);
vcpu->arch.mmio_fault_cr2 = cr2;
- kvm_x86_ops->cache_regs(vcpu);
+ /*
+ * TODO: fix x86_emulate.c to use guest_read/write_register
+ * instead of direct ->regs accesses, can save hundred cycles
+ * on Intel for instructions that don't read/change RSP, for
+ * for example.
+ */
+ cache_all_regs(vcpu);
vcpu->mmio_is_write = 0;
vcpu->arch.pio.string = 0;
return EMULATE_DO_MMIO;
}
- kvm_x86_ops->decache_regs(vcpu);
kvm_x86_ops->set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags);
if (vcpu->mmio_is_write) {
struct kvm_pio_request *io = &vcpu->arch.pio;
long delta;
int r;
-
- kvm_x86_ops->cache_regs(vcpu);
+ unsigned long val;
if (!io->string) {
- if (io->in)
- memcpy(&vcpu->arch.regs[VCPU_REGS_RAX], vcpu->arch.pio_data,
- io->size);
+ if (io->in) {
+ val = kvm_register_read(vcpu, VCPU_REGS_RAX);
+ memcpy(&val, vcpu->arch.pio_data, io->size);
+ kvm_register_write(vcpu, VCPU_REGS_RAX, val);
+ }
} else {
if (io->in) {
r = pio_copy_data(vcpu);
- if (r) {
- kvm_x86_ops->cache_regs(vcpu);
+ if (r)
return r;
- }
}
delta = 1;
* The size of the register should really depend on
* current address size.
*/
- vcpu->arch.regs[VCPU_REGS_RCX] -= delta;
+ val = kvm_register_read(vcpu, VCPU_REGS_RCX);
+ val -= delta;
+ kvm_register_write(vcpu, VCPU_REGS_RCX, val);
}
if (io->down)
delta = -delta;
delta *= io->size;
- if (io->in)
- vcpu->arch.regs[VCPU_REGS_RDI] += delta;
- else
- vcpu->arch.regs[VCPU_REGS_RSI] += delta;
+ if (io->in) {
+ val = kvm_register_read(vcpu, VCPU_REGS_RDI);
+ val += delta;
+ kvm_register_write(vcpu, VCPU_REGS_RDI, val);
+ } else {
+ val = kvm_register_read(vcpu, VCPU_REGS_RSI);
+ val += delta;
+ kvm_register_write(vcpu, VCPU_REGS_RSI, val);
+ }
}
- kvm_x86_ops->decache_regs(vcpu);
-
io->count -= io->cur_count;
io->cur_count = 0;
int size, unsigned port)
{
struct kvm_io_device *pio_dev;
+ unsigned long val;
vcpu->run->exit_reason = KVM_EXIT_IO;
vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT;
KVMTRACE_2D(IO_WRITE, vcpu, vcpu->run->io.port, (u32)size,
handler);
- kvm_x86_ops->cache_regs(vcpu);
- memcpy(vcpu->arch.pio_data, &vcpu->arch.regs[VCPU_REGS_RAX], 4);
+ val = kvm_register_read(vcpu, VCPU_REGS_RAX);
+ memcpy(vcpu->arch.pio_data, &val, 4);
kvm_x86_ops->skip_emulated_instruction(vcpu);
unsigned long nr, a0, a1, a2, a3, ret;
int r = 1;
- kvm_x86_ops->cache_regs(vcpu);
-
- nr = vcpu->arch.regs[VCPU_REGS_RAX];
- a0 = vcpu->arch.regs[VCPU_REGS_RBX];
- a1 = vcpu->arch.regs[VCPU_REGS_RCX];
- a2 = vcpu->arch.regs[VCPU_REGS_RDX];
- a3 = vcpu->arch.regs[VCPU_REGS_RSI];
+ nr = kvm_register_read(vcpu, VCPU_REGS_RAX);
+ a0 = kvm_register_read(vcpu, VCPU_REGS_RBX);
+ a1 = kvm_register_read(vcpu, VCPU_REGS_RCX);
+ a2 = kvm_register_read(vcpu, VCPU_REGS_RDX);
+ a3 = kvm_register_read(vcpu, VCPU_REGS_RSI);
KVMTRACE_1D(VMMCALL, vcpu, (u32)nr, handler);
ret = -KVM_ENOSYS;
break;
}
- vcpu->arch.regs[VCPU_REGS_RAX] = ret;
- kvm_x86_ops->decache_regs(vcpu);
+ kvm_register_write(vcpu, VCPU_REGS_RAX, ret);
++vcpu->stat.hypercalls;
return r;
}
{
char instruction[3];
int ret = 0;
+ unsigned long rip = kvm_rip_read(vcpu);
/*
*/
kvm_mmu_zap_all(vcpu->kvm);
- kvm_x86_ops->cache_regs(vcpu);
kvm_x86_ops->patch_hypercall(vcpu, instruction);
- if (emulator_write_emulated(vcpu->arch.rip, instruction, 3, vcpu)
+ if (emulator_write_emulated(rip, instruction, 3, vcpu)
!= X86EMUL_CONTINUE)
ret = -EFAULT;
u32 function, index;
struct kvm_cpuid_entry2 *e, *best;
- kvm_x86_ops->cache_regs(vcpu);
- function = vcpu->arch.regs[VCPU_REGS_RAX];
- index = vcpu->arch.regs[VCPU_REGS_RCX];
- vcpu->arch.regs[VCPU_REGS_RAX] = 0;
- vcpu->arch.regs[VCPU_REGS_RBX] = 0;
- vcpu->arch.regs[VCPU_REGS_RCX] = 0;
- vcpu->arch.regs[VCPU_REGS_RDX] = 0;
+ function = kvm_register_read(vcpu, VCPU_REGS_RAX);
+ index = kvm_register_read(vcpu, VCPU_REGS_RCX);
+ kvm_register_write(vcpu, VCPU_REGS_RAX, 0);
+ kvm_register_write(vcpu, VCPU_REGS_RBX, 0);
+ kvm_register_write(vcpu, VCPU_REGS_RCX, 0);
+ kvm_register_write(vcpu, VCPU_REGS_RDX, 0);
best = NULL;
for (i = 0; i < vcpu->arch.cpuid_nent; ++i) {
e = &vcpu->arch.cpuid_entries[i];
best = e;
}
if (best) {
- vcpu->arch.regs[VCPU_REGS_RAX] = best->eax;
- vcpu->arch.regs[VCPU_REGS_RBX] = best->ebx;
- vcpu->arch.regs[VCPU_REGS_RCX] = best->ecx;
- vcpu->arch.regs[VCPU_REGS_RDX] = best->edx;
+ kvm_register_write(vcpu, VCPU_REGS_RAX, best->eax);
+ kvm_register_write(vcpu, VCPU_REGS_RBX, best->ebx);
+ kvm_register_write(vcpu, VCPU_REGS_RCX, best->ecx);
+ kvm_register_write(vcpu, VCPU_REGS_RDX, best->edx);
}
- kvm_x86_ops->decache_regs(vcpu);
kvm_x86_ops->skip_emulated_instruction(vcpu);
KVMTRACE_5D(CPUID, vcpu, function,
- (u32)vcpu->arch.regs[VCPU_REGS_RAX],
- (u32)vcpu->arch.regs[VCPU_REGS_RBX],
- (u32)vcpu->arch.regs[VCPU_REGS_RCX],
- (u32)vcpu->arch.regs[VCPU_REGS_RDX], handler);
+ (u32)kvm_register_read(vcpu, VCPU_REGS_RAX),
+ (u32)kvm_register_read(vcpu, VCPU_REGS_RBX),
+ (u32)kvm_register_read(vcpu, VCPU_REGS_RCX),
+ (u32)kvm_register_read(vcpu, VCPU_REGS_RDX), handler);
}
EXPORT_SYMBOL_GPL(kvm_emulate_cpuid);
down_read(&vcpu->kvm->slots_lock);
vapic_enter(vcpu);
-preempted:
- if (vcpu->guest_debug.enabled)
- kvm_x86_ops->guest_debug_pre(vcpu);
-
again:
if (vcpu->requests)
if (test_and_clear_bit(KVM_REQ_MMU_RELOAD, &vcpu->requests))
goto out;
}
+ if (vcpu->guest_debug.enabled)
+ kvm_x86_ops->guest_debug_pre(vcpu);
+
vcpu->guest_mode = 1;
/*
* Make sure that guest_mode assignment won't happen after
* Profile KVM exit RIPs:
*/
if (unlikely(prof_on == KVM_PROFILING)) {
- kvm_x86_ops->cache_regs(vcpu);
- profile_hit(KVM_PROFILING, (void *)vcpu->arch.rip);
+ unsigned long rip = kvm_rip_read(vcpu);
+ profile_hit(KVM_PROFILING, (void *)rip);
}
if (vcpu->arch.exception.pending && kvm_x86_ops->exception_injected(vcpu))
if (r > 0) {
kvm_resched(vcpu);
down_read(&vcpu->kvm->slots_lock);
- goto preempted;
+ goto again;
}
post_kvm_run_save(vcpu, kvm_run);
}
}
#endif
- if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) {
- kvm_x86_ops->cache_regs(vcpu);
- vcpu->arch.regs[VCPU_REGS_RAX] = kvm_run->hypercall.ret;
- kvm_x86_ops->decache_regs(vcpu);
- }
+ if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL)
+ kvm_register_write(vcpu, VCPU_REGS_RAX,
+ kvm_run->hypercall.ret);
r = __vcpu_run(vcpu, kvm_run);
{
vcpu_load(vcpu);
- kvm_x86_ops->cache_regs(vcpu);
-
- regs->rax = vcpu->arch.regs[VCPU_REGS_RAX];
- regs->rbx = vcpu->arch.regs[VCPU_REGS_RBX];
- regs->rcx = vcpu->arch.regs[VCPU_REGS_RCX];
- regs->rdx = vcpu->arch.regs[VCPU_REGS_RDX];
- regs->rsi = vcpu->arch.regs[VCPU_REGS_RSI];
- regs->rdi = vcpu->arch.regs[VCPU_REGS_RDI];
- regs->rsp = vcpu->arch.regs[VCPU_REGS_RSP];
- regs->rbp = vcpu->arch.regs[VCPU_REGS_RBP];
+ regs->rax = kvm_register_read(vcpu, VCPU_REGS_RAX);
+ regs->rbx = kvm_register_read(vcpu, VCPU_REGS_RBX);
+ regs->rcx = kvm_register_read(vcpu, VCPU_REGS_RCX);
+ regs->rdx = kvm_register_read(vcpu, VCPU_REGS_RDX);
+ regs->rsi = kvm_register_read(vcpu, VCPU_REGS_RSI);
+ regs->rdi = kvm_register_read(vcpu, VCPU_REGS_RDI);
+ regs->rsp = kvm_register_read(vcpu, VCPU_REGS_RSP);
+ regs->rbp = kvm_register_read(vcpu, VCPU_REGS_RBP);
#ifdef CONFIG_X86_64
- regs->r8 = vcpu->arch.regs[VCPU_REGS_R8];
- regs->r9 = vcpu->arch.regs[VCPU_REGS_R9];
- regs->r10 = vcpu->arch.regs[VCPU_REGS_R10];
- regs->r11 = vcpu->arch.regs[VCPU_REGS_R11];
- regs->r12 = vcpu->arch.regs[VCPU_REGS_R12];
- regs->r13 = vcpu->arch.regs[VCPU_REGS_R13];
- regs->r14 = vcpu->arch.regs[VCPU_REGS_R14];
- regs->r15 = vcpu->arch.regs[VCPU_REGS_R15];
+ regs->r8 = kvm_register_read(vcpu, VCPU_REGS_R8);
+ regs->r9 = kvm_register_read(vcpu, VCPU_REGS_R9);
+ regs->r10 = kvm_register_read(vcpu, VCPU_REGS_R10);
+ regs->r11 = kvm_register_read(vcpu, VCPU_REGS_R11);
+ regs->r12 = kvm_register_read(vcpu, VCPU_REGS_R12);
+ regs->r13 = kvm_register_read(vcpu, VCPU_REGS_R13);
+ regs->r14 = kvm_register_read(vcpu, VCPU_REGS_R14);
+ regs->r15 = kvm_register_read(vcpu, VCPU_REGS_R15);
#endif
- regs->rip = vcpu->arch.rip;
+ regs->rip = kvm_rip_read(vcpu);
regs->rflags = kvm_x86_ops->get_rflags(vcpu);
/*
{
vcpu_load(vcpu);
- vcpu->arch.regs[VCPU_REGS_RAX] = regs->rax;
- vcpu->arch.regs[VCPU_REGS_RBX] = regs->rbx;
- vcpu->arch.regs[VCPU_REGS_RCX] = regs->rcx;
- vcpu->arch.regs[VCPU_REGS_RDX] = regs->rdx;
- vcpu->arch.regs[VCPU_REGS_RSI] = regs->rsi;
- vcpu->arch.regs[VCPU_REGS_RDI] = regs->rdi;
- vcpu->arch.regs[VCPU_REGS_RSP] = regs->rsp;
- vcpu->arch.regs[VCPU_REGS_RBP] = regs->rbp;
+ kvm_register_write(vcpu, VCPU_REGS_RAX, regs->rax);
+ kvm_register_write(vcpu, VCPU_REGS_RBX, regs->rbx);
+ kvm_register_write(vcpu, VCPU_REGS_RCX, regs->rcx);
+ kvm_register_write(vcpu, VCPU_REGS_RDX, regs->rdx);
+ kvm_register_write(vcpu, VCPU_REGS_RSI, regs->rsi);
+ kvm_register_write(vcpu, VCPU_REGS_RDI, regs->rdi);
+ kvm_register_write(vcpu, VCPU_REGS_RSP, regs->rsp);
+ kvm_register_write(vcpu, VCPU_REGS_RBP, regs->rbp);
#ifdef CONFIG_X86_64
- vcpu->arch.regs[VCPU_REGS_R8] = regs->r8;
- vcpu->arch.regs[VCPU_REGS_R9] = regs->r9;
- vcpu->arch.regs[VCPU_REGS_R10] = regs->r10;
- vcpu->arch.regs[VCPU_REGS_R11] = regs->r11;
- vcpu->arch.regs[VCPU_REGS_R12] = regs->r12;
- vcpu->arch.regs[VCPU_REGS_R13] = regs->r13;
- vcpu->arch.regs[VCPU_REGS_R14] = regs->r14;
- vcpu->arch.regs[VCPU_REGS_R15] = regs->r15;
+ kvm_register_write(vcpu, VCPU_REGS_R8, regs->r8);
+ kvm_register_write(vcpu, VCPU_REGS_R9, regs->r9);
+ kvm_register_write(vcpu, VCPU_REGS_R10, regs->r10);
+ kvm_register_write(vcpu, VCPU_REGS_R11, regs->r11);
+ kvm_register_write(vcpu, VCPU_REGS_R12, regs->r12);
+ kvm_register_write(vcpu, VCPU_REGS_R13, regs->r13);
+ kvm_register_write(vcpu, VCPU_REGS_R14, regs->r14);
+ kvm_register_write(vcpu, VCPU_REGS_R15, regs->r15);
+
#endif
- vcpu->arch.rip = regs->rip;
+ kvm_rip_write(vcpu, regs->rip);
kvm_x86_ops->set_rflags(vcpu, regs->rflags);
- kvm_x86_ops->decache_regs(vcpu);
vcpu->arch.exception.pending = false;
struct tss_segment_32 *tss)
{
tss->cr3 = vcpu->arch.cr3;
- tss->eip = vcpu->arch.rip;
+ tss->eip = kvm_rip_read(vcpu);
tss->eflags = kvm_x86_ops->get_rflags(vcpu);
- tss->eax = vcpu->arch.regs[VCPU_REGS_RAX];
- tss->ecx = vcpu->arch.regs[VCPU_REGS_RCX];
- tss->edx = vcpu->arch.regs[VCPU_REGS_RDX];
- tss->ebx = vcpu->arch.regs[VCPU_REGS_RBX];
- tss->esp = vcpu->arch.regs[VCPU_REGS_RSP];
- tss->ebp = vcpu->arch.regs[VCPU_REGS_RBP];
- tss->esi = vcpu->arch.regs[VCPU_REGS_RSI];
- tss->edi = vcpu->arch.regs[VCPU_REGS_RDI];
-
+ tss->eax = kvm_register_read(vcpu, VCPU_REGS_RAX);
+ tss->ecx = kvm_register_read(vcpu, VCPU_REGS_RCX);
+ tss->edx = kvm_register_read(vcpu, VCPU_REGS_RDX);
+ tss->ebx = kvm_register_read(vcpu, VCPU_REGS_RBX);
+ tss->esp = kvm_register_read(vcpu, VCPU_REGS_RSP);
+ tss->ebp = kvm_register_read(vcpu, VCPU_REGS_RBP);
+ tss->esi = kvm_register_read(vcpu, VCPU_REGS_RSI);
+ tss->edi = kvm_register_read(vcpu, VCPU_REGS_RDI);
tss->es = get_segment_selector(vcpu, VCPU_SREG_ES);
tss->cs = get_segment_selector(vcpu, VCPU_SREG_CS);
tss->ss = get_segment_selector(vcpu, VCPU_SREG_SS);
{
kvm_set_cr3(vcpu, tss->cr3);
- vcpu->arch.rip = tss->eip;
+ kvm_rip_write(vcpu, tss->eip);
kvm_x86_ops->set_rflags(vcpu, tss->eflags | 2);
- vcpu->arch.regs[VCPU_REGS_RAX] = tss->eax;
- vcpu->arch.regs[VCPU_REGS_RCX] = tss->ecx;
- vcpu->arch.regs[VCPU_REGS_RDX] = tss->edx;
- vcpu->arch.regs[VCPU_REGS_RBX] = tss->ebx;
- vcpu->arch.regs[VCPU_REGS_RSP] = tss->esp;
- vcpu->arch.regs[VCPU_REGS_RBP] = tss->ebp;
- vcpu->arch.regs[VCPU_REGS_RSI] = tss->esi;
- vcpu->arch.regs[VCPU_REGS_RDI] = tss->edi;
+ kvm_register_write(vcpu, VCPU_REGS_RAX, tss->eax);
+ kvm_register_write(vcpu, VCPU_REGS_RCX, tss->ecx);
+ kvm_register_write(vcpu, VCPU_REGS_RDX, tss->edx);
+ kvm_register_write(vcpu, VCPU_REGS_RBX, tss->ebx);
+ kvm_register_write(vcpu, VCPU_REGS_RSP, tss->esp);
+ kvm_register_write(vcpu, VCPU_REGS_RBP, tss->ebp);
+ kvm_register_write(vcpu, VCPU_REGS_RSI, tss->esi);
+ kvm_register_write(vcpu, VCPU_REGS_RDI, tss->edi);
if (kvm_load_segment_descriptor(vcpu, tss->ldt_selector, 0, VCPU_SREG_LDTR))
return 1;
static void save_state_to_tss16(struct kvm_vcpu *vcpu,
struct tss_segment_16 *tss)
{
- tss->ip = vcpu->arch.rip;
+ tss->ip = kvm_rip_read(vcpu);
tss->flag = kvm_x86_ops->get_rflags(vcpu);
- tss->ax = vcpu->arch.regs[VCPU_REGS_RAX];
- tss->cx = vcpu->arch.regs[VCPU_REGS_RCX];
- tss->dx = vcpu->arch.regs[VCPU_REGS_RDX];
- tss->bx = vcpu->arch.regs[VCPU_REGS_RBX];
- tss->sp = vcpu->arch.regs[VCPU_REGS_RSP];
- tss->bp = vcpu->arch.regs[VCPU_REGS_RBP];
- tss->si = vcpu->arch.regs[VCPU_REGS_RSI];
- tss->di = vcpu->arch.regs[VCPU_REGS_RDI];
+ tss->ax = kvm_register_read(vcpu, VCPU_REGS_RAX);
+ tss->cx = kvm_register_read(vcpu, VCPU_REGS_RCX);
+ tss->dx = kvm_register_read(vcpu, VCPU_REGS_RDX);
+ tss->bx = kvm_register_read(vcpu, VCPU_REGS_RBX);
+ tss->sp = kvm_register_read(vcpu, VCPU_REGS_RSP);
+ tss->bp = kvm_register_read(vcpu, VCPU_REGS_RBP);
+ tss->si = kvm_register_read(vcpu, VCPU_REGS_RSI);
+ tss->di = kvm_register_read(vcpu, VCPU_REGS_RDI);
tss->es = get_segment_selector(vcpu, VCPU_SREG_ES);
tss->cs = get_segment_selector(vcpu, VCPU_SREG_CS);
static int load_state_from_tss16(struct kvm_vcpu *vcpu,
struct tss_segment_16 *tss)
{
- vcpu->arch.rip = tss->ip;
+ kvm_rip_write(vcpu, tss->ip);
kvm_x86_ops->set_rflags(vcpu, tss->flag | 2);
- vcpu->arch.regs[VCPU_REGS_RAX] = tss->ax;
- vcpu->arch.regs[VCPU_REGS_RCX] = tss->cx;
- vcpu->arch.regs[VCPU_REGS_RDX] = tss->dx;
- vcpu->arch.regs[VCPU_REGS_RBX] = tss->bx;
- vcpu->arch.regs[VCPU_REGS_RSP] = tss->sp;
- vcpu->arch.regs[VCPU_REGS_RBP] = tss->bp;
- vcpu->arch.regs[VCPU_REGS_RSI] = tss->si;
- vcpu->arch.regs[VCPU_REGS_RDI] = tss->di;
+ kvm_register_write(vcpu, VCPU_REGS_RAX, tss->ax);
+ kvm_register_write(vcpu, VCPU_REGS_RCX, tss->cx);
+ kvm_register_write(vcpu, VCPU_REGS_RDX, tss->dx);
+ kvm_register_write(vcpu, VCPU_REGS_RBX, tss->bx);
+ kvm_register_write(vcpu, VCPU_REGS_RSP, tss->sp);
+ kvm_register_write(vcpu, VCPU_REGS_RBP, tss->bp);
+ kvm_register_write(vcpu, VCPU_REGS_RSI, tss->si);
+ kvm_register_write(vcpu, VCPU_REGS_RDI, tss->di);
if (kvm_load_segment_descriptor(vcpu, tss->ldt, 0, VCPU_SREG_LDTR))
return 1;
}
kvm_x86_ops->skip_emulated_instruction(vcpu);
- kvm_x86_ops->cache_regs(vcpu);
if (nseg_desc.type & 8)
ret = kvm_task_switch_32(vcpu, tss_selector, old_tss_base,
tr_seg.type = 11;
kvm_set_segment(vcpu, &tr_seg, VCPU_SREG_TR);
out:
- kvm_x86_ops->decache_regs(vcpu);
return ret;
}
EXPORT_SYMBOL_GPL(kvm_task_switch);
return ERR_PTR(-ENOMEM);
INIT_LIST_HEAD(&kvm->arch.active_mmu_pages);
+ INIT_LIST_HEAD(&kvm->arch.assigned_dev_head);
return kvm;
}
void kvm_arch_destroy_vm(struct kvm *kvm)
{
+ kvm_free_assigned_devices(kvm);
kvm_free_pit(kvm);
kfree(kvm->arch.vpic);
kfree(kvm->arch.vioapic);
*/
if (!user_alloc) {
if (npages && !old.rmap) {
+ unsigned long userspace_addr;
+
down_write(¤t->mm->mmap_sem);
- memslot->userspace_addr = do_mmap(NULL, 0,
- npages * PAGE_SIZE,
- PROT_READ | PROT_WRITE,
- MAP_SHARED | MAP_ANONYMOUS,
- 0);
+ userspace_addr = do_mmap(NULL, 0,
+ npages * PAGE_SIZE,
+ PROT_READ | PROT_WRITE,
+ MAP_SHARED | MAP_ANONYMOUS,
+ 0);
up_write(¤t->mm->mmap_sem);
- if (IS_ERR((void *)memslot->userspace_addr))
- return PTR_ERR((void *)memslot->userspace_addr);
+ if (IS_ERR((void *)userspace_addr))
+ return PTR_ERR((void *)userspace_addr);
+
+ /* set userspace_addr atomically for kvm_hva_to_rmapp */
+ spin_lock(&kvm->mmu_lock);
+ memslot->userspace_addr = userspace_addr;
+ spin_unlock(&kvm->mmu_lock);
} else {
if (!old.user_alloc && old.rmap) {
int ret;
projects / cascardo / linux.git / blobdiff