kvm_main.c 60.3 KB
Newer Older
A
Avi Kivity 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 * Kernel-based Virtual Machine driver for Linux
 *
 * This module enables machines with Intel VT-x extensions to run virtual
 * machines without emulation or binary translation.
 *
 * Copyright (C) 2006 Qumranet, Inc.
 *
 * Authors:
 *   Avi Kivity   <avi@qumranet.com>
 *   Yaniv Kamay  <yaniv@qumranet.com>
 *
 * This work is licensed under the terms of the GNU GPL, version 2.  See
 * the COPYING file in the top-level directory.
 *
 */

18
#include "iodev.h"
A
Avi Kivity 已提交
19

20
#include <linux/kvm_host.h>
A
Avi Kivity 已提交
21 22 23 24 25 26 27 28 29 30 31 32
#include <linux/kvm.h>
#include <linux/module.h>
#include <linux/errno.h>
#include <linux/percpu.h>
#include <linux/gfp.h>
#include <linux/mm.h>
#include <linux/miscdevice.h>
#include <linux/vmalloc.h>
#include <linux/reboot.h>
#include <linux/debugfs.h>
#include <linux/highmem.h>
#include <linux/file.h>
33
#include <linux/sysdev.h>
A
Avi Kivity 已提交
34
#include <linux/cpu.h>
A
Alexey Dobriyan 已提交
35
#include <linux/sched.h>
36 37
#include <linux/cpumask.h>
#include <linux/smp.h>
38
#include <linux/anon_inodes.h>
39
#include <linux/profile.h>
40
#include <linux/kvm_para.h>
41
#include <linux/pagemap.h>
42
#include <linux/mman.h>
43
#include <linux/swap.h>
44
#include <linux/bitops.h>
45
#include <linux/spinlock.h>
A
Avi Kivity 已提交
46

A
Avi Kivity 已提交
47 48 49
#include <asm/processor.h>
#include <asm/io.h>
#include <asm/uaccess.h>
50
#include <asm/pgtable.h>
A
Avi Kivity 已提交
51

52 53 54 55
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
#include "coalesced_mmio.h"
#endif

56 57 58 59 60 61
#ifdef KVM_CAP_DEVICE_ASSIGNMENT
#include <linux/pci.h>
#include <linux/interrupt.h>
#include "irq.h"
#endif

A
Avi Kivity 已提交
62 63 64
MODULE_AUTHOR("Qumranet");
MODULE_LICENSE("GPL");

65 66
DEFINE_SPINLOCK(kvm_lock);
LIST_HEAD(vm_list);
67

68
static cpumask_var_t cpus_hardware_enabled;
69

70 71
struct kmem_cache *kvm_vcpu_cache;
EXPORT_SYMBOL_GPL(kvm_vcpu_cache);
A
Avi Kivity 已提交
72

73 74
static __read_mostly struct preempt_ops kvm_preempt_ops;

75
struct dentry *kvm_debugfs_dir;
A
Avi Kivity 已提交
76

A
Avi Kivity 已提交
77 78 79
static long kvm_vcpu_ioctl(struct file *file, unsigned int ioctl,
			   unsigned long arg);

H
Hannes Eder 已提交
80
static bool kvm_rebooting;
81

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
#ifdef KVM_CAP_DEVICE_ASSIGNMENT
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;
}

97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118
static int find_index_from_host_irq(struct kvm_assigned_dev_kernel
				    *assigned_dev, int irq)
{
	int i, index;
	struct msix_entry *host_msix_entries;

	host_msix_entries = assigned_dev->host_msix_entries;

	index = -1;
	for (i = 0; i < assigned_dev->entries_nr; i++)
		if (irq == host_msix_entries[i].vector) {
			index = i;
			break;
		}
	if (index < 0) {
		printk(KERN_WARNING "Fail to find correlated MSI-X entry!\n");
		return 0;
	}

	return index;
}

119 120 121
static void kvm_assigned_dev_interrupt_work_handler(struct work_struct *work)
{
	struct kvm_assigned_dev_kernel *assigned_dev;
122 123
	struct kvm *kvm;
	int irq, i;
124 125 126

	assigned_dev = container_of(work, struct kvm_assigned_dev_kernel,
				    interrupt_work);
127
	kvm = assigned_dev->kvm;
128 129 130 131 132

	/* 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
	 */
133
	mutex_lock(&kvm->lock);
134
	spin_lock_irq(&assigned_dev->assigned_dev_lock);
135
	if (assigned_dev->irq_requested_type & KVM_DEV_IRQ_HOST_MSIX) {
136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154
		struct kvm_guest_msix_entry *guest_entries =
			assigned_dev->guest_msix_entries;
		for (i = 0; i < assigned_dev->entries_nr; i++) {
			if (!(guest_entries[i].flags &
					KVM_ASSIGNED_MSIX_PENDING))
				continue;
			guest_entries[i].flags &= ~KVM_ASSIGNED_MSIX_PENDING;
			kvm_set_irq(assigned_dev->kvm,
				    assigned_dev->irq_source_id,
				    guest_entries[i].vector, 1);
			irq = assigned_dev->host_msix_entries[i].vector;
			if (irq != 0)
				enable_irq(irq);
			assigned_dev->host_irq_disabled = false;
		}
	} else {
		kvm_set_irq(assigned_dev->kvm, assigned_dev->irq_source_id,
			    assigned_dev->guest_irq, 1);
		if (assigned_dev->irq_requested_type &
155
				KVM_DEV_IRQ_GUEST_MSI) {
156 157 158
			enable_irq(assigned_dev->host_irq);
			assigned_dev->host_irq_disabled = false;
		}
S
Sheng Yang 已提交
159
	}
160

161
	spin_unlock_irq(&assigned_dev->assigned_dev_lock);
162 163 164 165 166
	mutex_unlock(&assigned_dev->kvm->lock);
}

static irqreturn_t kvm_assigned_dev_intr(int irq, void *dev_id)
{
167
	unsigned long flags;
168 169 170
	struct kvm_assigned_dev_kernel *assigned_dev =
		(struct kvm_assigned_dev_kernel *) dev_id;

171
	spin_lock_irqsave(&assigned_dev->assigned_dev_lock, flags);
172
	if (assigned_dev->irq_requested_type & KVM_DEV_IRQ_HOST_MSIX) {
173 174
		int index = find_index_from_host_irq(assigned_dev, irq);
		if (index < 0)
175
			goto out;
176 177 178 179
		assigned_dev->guest_msix_entries[index].flags |=
			KVM_ASSIGNED_MSIX_PENDING;
	}

180
	schedule_work(&assigned_dev->interrupt_work);
181

182
	disable_irq_nosync(irq);
183 184
	assigned_dev->host_irq_disabled = true;

185 186
out:
	spin_unlock_irqrestore(&assigned_dev->assigned_dev_lock, flags);
187 188 189 190 191 192 193
	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;
194
	unsigned long flags;
195 196 197 198 199 200

	if (kian->gsi == -1)
		return;

	dev = container_of(kian, struct kvm_assigned_dev_kernel,
			   ack_notifier);
201

202
	kvm_set_irq(dev->kvm, dev->irq_source_id, dev->guest_irq, 0);
203 204 205 206

	/* The guest irq may be shared so this ack may be
	 * from another device.
	 */
207
	spin_lock_irqsave(&dev->assigned_dev_lock, flags);
208 209 210 211
	if (dev->host_irq_disabled) {
		enable_irq(dev->host_irq);
		dev->host_irq_disabled = false;
	}
212
	spin_unlock_irqrestore(&dev->assigned_dev_lock, flags);
213 214
}

215 216
static void deassign_guest_irq(struct kvm *kvm,
			       struct kvm_assigned_dev_kernel *assigned_dev)
217
{
218
	kvm_unregister_irq_ack_notifier(&assigned_dev->ack_notifier);
219
	assigned_dev->ack_notifier.gsi = -1;
220 221 222 223

	if (assigned_dev->irq_source_id != -1)
		kvm_free_irq_source_id(kvm, assigned_dev->irq_source_id);
	assigned_dev->irq_source_id = -1;
224 225
	assigned_dev->irq_requested_type &= ~(KVM_DEV_IRQ_GUEST_MASK);
}
226

227 228 229 230
/* The function implicit hold kvm->lock mutex due to cancel_work_sync() */
static void deassign_host_irq(struct kvm *kvm,
			      struct kvm_assigned_dev_kernel *assigned_dev)
{
231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246
	/*
	 * In kvm_free_device_irq, cancel_work_sync return true if:
	 * 1. work is scheduled, and then cancelled.
	 * 2. work callback is executed.
	 *
	 * The first one ensured that the irq is disabled and no more events
	 * would happen. But for the second one, the irq may be enabled (e.g.
	 * for MSI). So we disable irq here to prevent further events.
	 *
	 * Notice this maybe result in nested disable if the interrupt type is
	 * INTx, but it's OK for we are going to free it.
	 *
	 * If this function is a part of VM destroy, please ensure that till
	 * now, the kvm state is still legal for probably we also have to wait
	 * interrupt_work done.
	 */
247
	if (assigned_dev->irq_requested_type & KVM_DEV_IRQ_HOST_MSIX) {
248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266
		int i;
		for (i = 0; i < assigned_dev->entries_nr; i++)
			disable_irq_nosync(assigned_dev->
					   host_msix_entries[i].vector);

		cancel_work_sync(&assigned_dev->interrupt_work);

		for (i = 0; i < assigned_dev->entries_nr; i++)
			free_irq(assigned_dev->host_msix_entries[i].vector,
				 (void *)assigned_dev);

		assigned_dev->entries_nr = 0;
		kfree(assigned_dev->host_msix_entries);
		kfree(assigned_dev->guest_msix_entries);
		pci_disable_msix(assigned_dev->dev);
	} else {
		/* Deal with MSI and INTx */
		disable_irq_nosync(assigned_dev->host_irq);
		cancel_work_sync(&assigned_dev->interrupt_work);
267

268
		free_irq(assigned_dev->host_irq, (void *)assigned_dev);
269

270
		if (assigned_dev->irq_requested_type & KVM_DEV_IRQ_HOST_MSI)
271 272
			pci_disable_msi(assigned_dev->dev);
	}
273

274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297
	assigned_dev->irq_requested_type &= ~(KVM_DEV_IRQ_HOST_MASK);
}

static int kvm_deassign_irq(struct kvm *kvm,
			    struct kvm_assigned_dev_kernel *assigned_dev,
			    unsigned long irq_requested_type)
{
	unsigned long guest_irq_type, host_irq_type;

	if (!irqchip_in_kernel(kvm))
		return -EINVAL;
	/* no irq assignment to deassign */
	if (!assigned_dev->irq_requested_type)
		return -ENXIO;

	host_irq_type = irq_requested_type & KVM_DEV_IRQ_HOST_MASK;
	guest_irq_type = irq_requested_type & KVM_DEV_IRQ_GUEST_MASK;

	if (host_irq_type)
		deassign_host_irq(kvm, assigned_dev);
	if (guest_irq_type)
		deassign_guest_irq(kvm, assigned_dev);

	return 0;
298 299
}

300 301 302 303 304
static void kvm_free_assigned_irq(struct kvm *kvm,
				  struct kvm_assigned_dev_kernel *assigned_dev)
{
	kvm_deassign_irq(kvm, assigned_dev, assigned_dev->irq_requested_type);
}
305 306 307 308 309 310 311

static void kvm_free_assigned_device(struct kvm *kvm,
				     struct kvm_assigned_dev_kernel
				     *assigned_dev)
{
	kvm_free_assigned_irq(kvm, assigned_dev);

312 313
	pci_reset_function(assigned_dev->dev);

314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335
	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);
}

void kvm_free_all_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);

		kvm_free_assigned_device(kvm, assigned_dev);
	}
}

336 337
static int assigned_device_enable_host_intx(struct kvm *kvm,
					    struct kvm_assigned_dev_kernel *dev)
338
{
339 340 341 342 343 344 345 346 347 348 349
	dev->host_irq = dev->dev->irq;
	/* 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(dev->host_irq, kvm_assigned_dev_intr,
			0, "kvm_assigned_intx_device", (void *)dev))
		return -EIO;
	return 0;
}
S
Sheng Yang 已提交
350

351 352 353 354 355
#ifdef __KVM_HAVE_MSI
static int assigned_device_enable_host_msi(struct kvm *kvm,
					   struct kvm_assigned_dev_kernel *dev)
{
	int r;
356

357 358 359 360 361
	if (!dev->dev->msi_enabled) {
		r = pci_enable_msi(dev->dev);
		if (r)
			return r;
	}
362

363 364 365 366 367
	dev->host_irq = dev->dev->irq;
	if (request_irq(dev->host_irq, kvm_assigned_dev_intr, 0,
			"kvm_assigned_msi_device", (void *)dev)) {
		pci_disable_msi(dev->dev);
		return -EIO;
368 369 370 371
	}

	return 0;
}
372
#endif
373

374 375 376
#ifdef __KVM_HAVE_MSIX
static int assigned_device_enable_host_msix(struct kvm *kvm,
					    struct kvm_assigned_dev_kernel *dev)
S
Sheng Yang 已提交
377
{
378
	int i, r = -EINVAL;
S
Sheng Yang 已提交
379

380 381 382 383
	/* host_msix_entries and guest_msix_entries should have been
	 * initialized */
	if (dev->entries_nr == 0)
		return r;
S
Sheng Yang 已提交
384

385 386 387
	r = pci_enable_msix(dev->dev, dev->host_msix_entries, dev->entries_nr);
	if (r)
		return r;
S
Sheng Yang 已提交
388

389 390 391 392 393 394 395 396 397
	for (i = 0; i < dev->entries_nr; i++) {
		r = request_irq(dev->host_msix_entries[i].vector,
				kvm_assigned_dev_intr, 0,
				"kvm_assigned_msix_device",
				(void *)dev);
		/* FIXME: free requested_irq's on failure */
		if (r)
			return r;
	}
S
Sheng Yang 已提交
398

399 400
	return 0;
}
S
Sheng Yang 已提交
401

402
#endif
S
Sheng Yang 已提交
403

404 405 406 407 408 409 410 411
static int assigned_device_enable_guest_intx(struct kvm *kvm,
				struct kvm_assigned_dev_kernel *dev,
				struct kvm_assigned_irq *irq)
{
	dev->guest_irq = irq->guest_irq;
	dev->ack_notifier.gsi = irq->guest_irq;
	return 0;
}
S
Sheng Yang 已提交
412

413 414 415 416 417 418 419
#ifdef __KVM_HAVE_MSI
static int assigned_device_enable_guest_msi(struct kvm *kvm,
			struct kvm_assigned_dev_kernel *dev,
			struct kvm_assigned_irq *irq)
{
	dev->guest_irq = irq->guest_irq;
	dev->ack_notifier.gsi = -1;
S
Sheng Yang 已提交
420 421 422
	return 0;
}
#endif
423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441
#ifdef __KVM_HAVE_MSIX
static int assigned_device_enable_guest_msix(struct kvm *kvm,
			struct kvm_assigned_dev_kernel *dev,
			struct kvm_assigned_irq *irq)
{
	dev->guest_irq = irq->guest_irq;
	dev->ack_notifier.gsi = -1;
	return 0;
}
#endif

static int assign_host_irq(struct kvm *kvm,
			   struct kvm_assigned_dev_kernel *dev,
			   __u32 host_irq_type)
{
	int r = -EEXIST;

	if (dev->irq_requested_type & KVM_DEV_IRQ_HOST_MASK)
		return r;
S
Sheng Yang 已提交
442

443 444 445 446 447 448 449 450 451
	switch (host_irq_type) {
	case KVM_DEV_IRQ_HOST_INTX:
		r = assigned_device_enable_host_intx(kvm, dev);
		break;
#ifdef __KVM_HAVE_MSI
	case KVM_DEV_IRQ_HOST_MSI:
		r = assigned_device_enable_host_msi(kvm, dev);
		break;
#endif
452
#ifdef __KVM_HAVE_MSIX
453 454 455 456 457 458 459
	case KVM_DEV_IRQ_HOST_MSIX:
		r = assigned_device_enable_host_msix(kvm, dev);
		break;
#endif
	default:
		r = -EINVAL;
	}
460

461 462
	if (!r)
		dev->irq_requested_type |= host_irq_type;
463

464 465
	return r;
}
466

467 468 469 470 471 472 473
static int assign_guest_irq(struct kvm *kvm,
			    struct kvm_assigned_dev_kernel *dev,
			    struct kvm_assigned_irq *irq,
			    unsigned long guest_irq_type)
{
	int id;
	int r = -EEXIST;
474

475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499
	if (dev->irq_requested_type & KVM_DEV_IRQ_GUEST_MASK)
		return r;

	id = kvm_request_irq_source_id(kvm);
	if (id < 0)
		return id;

	dev->irq_source_id = id;

	switch (guest_irq_type) {
	case KVM_DEV_IRQ_GUEST_INTX:
		r = assigned_device_enable_guest_intx(kvm, dev, irq);
		break;
#ifdef __KVM_HAVE_MSI
	case KVM_DEV_IRQ_GUEST_MSI:
		r = assigned_device_enable_guest_msi(kvm, dev, irq);
		break;
#endif
#ifdef __KVM_HAVE_MSIX
	case KVM_DEV_IRQ_GUEST_MSIX:
		r = assigned_device_enable_guest_msix(kvm, dev, irq);
		break;
#endif
	default:
		r = -EINVAL;
500 501
	}

502 503 504 505 506
	if (!r) {
		dev->irq_requested_type |= guest_irq_type;
		kvm_register_irq_ack_notifier(kvm, &dev->ack_notifier);
	} else
		kvm_free_irq_source_id(kvm, dev->irq_source_id);
507

508
	return r;
509 510
}

511
/* TODO Deal with KVM_DEV_IRQ_ASSIGNED_MASK_MSIX */
512
static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
513
				   struct kvm_assigned_irq *assigned_irq)
514
{
515
	int r = -EINVAL;
516
	struct kvm_assigned_dev_kernel *match;
517
	unsigned long host_irq_type, guest_irq_type;
518

519 520
	if (!capable(CAP_SYS_RAWIO))
		return -EPERM;
521

522 523 524 525 526
	if (!irqchip_in_kernel(kvm))
		return r;

	mutex_lock(&kvm->lock);
	r = -ENODEV;
527 528
	match = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head,
				      assigned_irq->assigned_dev_id);
529 530
	if (!match)
		goto out;
531

532 533
	host_irq_type = (assigned_irq->flags & KVM_DEV_IRQ_HOST_MASK);
	guest_irq_type = (assigned_irq->flags & KVM_DEV_IRQ_GUEST_MASK);
534

535 536 537 538 539 540 541 542
	r = -EINVAL;
	/* can only assign one type at a time */
	if (hweight_long(host_irq_type) > 1)
		goto out;
	if (hweight_long(guest_irq_type) > 1)
		goto out;
	if (host_irq_type == 0 && guest_irq_type == 0)
		goto out;
543

544 545 546 547 548
	r = 0;
	if (host_irq_type)
		r = assign_host_irq(kvm, match, host_irq_type);
	if (r)
		goto out;
549

550 551 552
	if (guest_irq_type)
		r = assign_guest_irq(kvm, match, assigned_irq, guest_irq_type);
out:
553 554
	mutex_unlock(&kvm->lock);
	return r;
555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572
}

static int kvm_vm_ioctl_deassign_dev_irq(struct kvm *kvm,
					 struct kvm_assigned_irq
					 *assigned_irq)
{
	int r = -ENODEV;
	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)
		goto out;

	r = kvm_deassign_irq(kvm, match, assigned_irq->flags);
out:
573 574 575 576 577 578 579 580 581 582 583
	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;

584
	down_read(&kvm->slots_lock);
585 586 587 588 589 590
	mutex_lock(&kvm->lock);

	match = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head,
				      assigned_dev->assigned_dev_id);
	if (match) {
		/* device already assigned */
591
		r = -EEXIST;
592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619
		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;
	}
620 621 622

	pci_reset_function(dev);

623 624 625
	match->assigned_dev_id = assigned_dev->assigned_dev_id;
	match->host_busnr = assigned_dev->busnr;
	match->host_devfn = assigned_dev->devfn;
626
	match->flags = assigned_dev->flags;
627
	match->dev = dev;
628
	spin_lock_init(&match->assigned_dev_lock);
629
	match->irq_source_id = -1;
630
	match->kvm = kvm;
631 632 633
	match->ack_notifier.irq_acked = kvm_assigned_dev_ack_irq;
	INIT_WORK(&match->interrupt_work,
		  kvm_assigned_dev_interrupt_work_handler);
634 635 636 637

	list_add(&match->list, &kvm->arch.assigned_dev_head);

	if (assigned_dev->flags & KVM_DEV_ASSIGN_ENABLE_IOMMU) {
J
Joerg Roedel 已提交
638
		if (!kvm->arch.iommu_domain) {
W
Weidong Han 已提交
639 640 641 642 643
			r = kvm_iommu_map_guest(kvm);
			if (r)
				goto out_list_del;
		}
		r = kvm_assign_device(kvm, match);
644 645 646 647 648 649
		if (r)
			goto out_list_del;
	}

out:
	mutex_unlock(&kvm->lock);
650
	up_read(&kvm->slots_lock);
651 652 653 654 655 656 657 658 659 660 661
	return r;
out_list_del:
	list_del(&match->list);
	pci_release_regions(dev);
out_disable:
	pci_disable_device(dev);
out_put:
	pci_dev_put(dev);
out_free:
	kfree(match);
	mutex_unlock(&kvm->lock);
662
	up_read(&kvm->slots_lock);
663 664 665 666
	return r;
}
#endif

W
Weidong Han 已提交
667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684
#ifdef KVM_CAP_DEVICE_DEASSIGNMENT
static int kvm_vm_ioctl_deassign_device(struct kvm *kvm,
		struct kvm_assigned_pci_dev *assigned_dev)
{
	int r = 0;
	struct kvm_assigned_dev_kernel *match;

	mutex_lock(&kvm->lock);

	match = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head,
				      assigned_dev->assigned_dev_id);
	if (!match) {
		printk(KERN_INFO "%s: device hasn't been assigned before, "
		  "so cannot be deassigned\n", __func__);
		r = -EINVAL;
		goto out;
	}

685
	if (match->flags & KVM_DEV_ASSIGN_ENABLE_IOMMU)
W
Weidong Han 已提交
686 687 688 689 690 691 692 693 694 695
		kvm_deassign_device(kvm, match);

	kvm_free_assigned_device(kvm, match);

out:
	mutex_unlock(&kvm->lock);
	return r;
}
#endif

696 697 698 699 700
static inline int valid_vcpu(int n)
{
	return likely(n >= 0 && n < KVM_MAX_VCPUS);
}

701
inline int kvm_is_mmio_pfn(pfn_t pfn)
B
Ben-Ami Yassour 已提交
702
{
703 704 705 706
	if (pfn_valid(pfn)) {
		struct page *page = compound_head(pfn_to_page(pfn));
		return PageReserved(page);
	}
B
Ben-Ami Yassour 已提交
707 708 709 710

	return true;
}

A
Avi Kivity 已提交
711 712 713
/*
 * Switches to specified vcpu, until a matching vcpu_put()
 */
714
void vcpu_load(struct kvm_vcpu *vcpu)
A
Avi Kivity 已提交
715
{
716 717
	int cpu;

A
Avi Kivity 已提交
718
	mutex_lock(&vcpu->mutex);
719 720
	cpu = get_cpu();
	preempt_notifier_register(&vcpu->preempt_notifier);
721
	kvm_arch_vcpu_load(vcpu, cpu);
722
	put_cpu();
A
Avi Kivity 已提交
723 724
}

725
void vcpu_put(struct kvm_vcpu *vcpu)
A
Avi Kivity 已提交
726
{
727
	preempt_disable();
728
	kvm_arch_vcpu_put(vcpu);
729 730
	preempt_notifier_unregister(&vcpu->preempt_notifier);
	preempt_enable();
A
Avi Kivity 已提交
731 732 733
	mutex_unlock(&vcpu->mutex);
}

734 735 736 737
static void ack_flush(void *_completed)
{
}

738
static bool make_all_cpus_request(struct kvm *kvm, unsigned int req)
739
{
740
	int i, cpu, me;
741 742
	cpumask_var_t cpus;
	bool called = true;
743 744
	struct kvm_vcpu *vcpu;

745 746 747
	if (alloc_cpumask_var(&cpus, GFP_ATOMIC))
		cpumask_clear(cpus);

748
	me = get_cpu();
R
Rusty Russell 已提交
749 750 751 752
	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
		vcpu = kvm->vcpus[i];
		if (!vcpu)
			continue;
753
		if (test_and_set_bit(req, &vcpu->requests))
754 755
			continue;
		cpu = vcpu->cpu;
756 757
		if (cpus != NULL && cpu != -1 && cpu != me)
			cpumask_set_cpu(cpu, cpus);
758
	}
759 760 761 762 763 764
	if (unlikely(cpus == NULL))
		smp_call_function_many(cpu_online_mask, ack_flush, NULL, 1);
	else if (!cpumask_empty(cpus))
		smp_call_function_many(cpus, ack_flush, NULL, 1);
	else
		called = false;
765
	put_cpu();
766
	free_cpumask_var(cpus);
767
	return called;
768 769
}

770
void kvm_flush_remote_tlbs(struct kvm *kvm)
771
{
772 773
	if (make_all_cpus_request(kvm, KVM_REQ_TLB_FLUSH))
		++kvm->stat.remote_tlb_flush;
774 775
}

776 777 778 779
void kvm_reload_remote_mmus(struct kvm *kvm)
{
	make_all_cpus_request(kvm, KVM_REQ_MMU_RELOAD);
}
780

R
Rusty Russell 已提交
781 782 783 784 785 786 787 788 789
int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id)
{
	struct page *page;
	int r;

	mutex_init(&vcpu->mutex);
	vcpu->cpu = -1;
	vcpu->kvm = kvm;
	vcpu->vcpu_id = id;
E
Eddie Dong 已提交
790
	init_waitqueue_head(&vcpu->wq);
R
Rusty Russell 已提交
791 792 793 794 795 796 797 798

	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
	if (!page) {
		r = -ENOMEM;
		goto fail;
	}
	vcpu->run = page_address(page);

799
	r = kvm_arch_vcpu_init(vcpu);
R
Rusty Russell 已提交
800
	if (r < 0)
801
		goto fail_free_run;
R
Rusty Russell 已提交
802 803 804 805 806
	return 0;

fail_free_run:
	free_page((unsigned long)vcpu->run);
fail:
807
	return r;
R
Rusty Russell 已提交
808 809 810 811 812
}
EXPORT_SYMBOL_GPL(kvm_vcpu_init);

void kvm_vcpu_uninit(struct kvm_vcpu *vcpu)
{
813
	kvm_arch_vcpu_uninit(vcpu);
R
Rusty Russell 已提交
814 815 816 817
	free_page((unsigned long)vcpu->run);
}
EXPORT_SYMBOL_GPL(kvm_vcpu_uninit);

818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926
#if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER)
static inline struct kvm *mmu_notifier_to_kvm(struct mmu_notifier *mn)
{
	return container_of(mn, struct kvm, mmu_notifier);
}

static void kvm_mmu_notifier_invalidate_page(struct mmu_notifier *mn,
					     struct mm_struct *mm,
					     unsigned long address)
{
	struct kvm *kvm = mmu_notifier_to_kvm(mn);
	int need_tlb_flush;

	/*
	 * When ->invalidate_page runs, the linux pte has been zapped
	 * already but the page is still allocated until
	 * ->invalidate_page returns. So if we increase the sequence
	 * here the kvm page fault will notice if the spte can't be
	 * established because the page is going to be freed. If
	 * instead the kvm page fault establishes the spte before
	 * ->invalidate_page runs, kvm_unmap_hva will release it
	 * before returning.
	 *
	 * The sequence increase only need to be seen at spin_unlock
	 * time, and not at spin_lock time.
	 *
	 * Increasing the sequence after the spin_unlock would be
	 * unsafe because the kvm page fault could then establish the
	 * pte after kvm_unmap_hva returned, without noticing the page
	 * is going to be freed.
	 */
	spin_lock(&kvm->mmu_lock);
	kvm->mmu_notifier_seq++;
	need_tlb_flush = kvm_unmap_hva(kvm, address);
	spin_unlock(&kvm->mmu_lock);

	/* we've to flush the tlb before the pages can be freed */
	if (need_tlb_flush)
		kvm_flush_remote_tlbs(kvm);

}

static void kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
						    struct mm_struct *mm,
						    unsigned long start,
						    unsigned long end)
{
	struct kvm *kvm = mmu_notifier_to_kvm(mn);
	int need_tlb_flush = 0;

	spin_lock(&kvm->mmu_lock);
	/*
	 * The count increase must become visible at unlock time as no
	 * spte can be established without taking the mmu_lock and
	 * count is also read inside the mmu_lock critical section.
	 */
	kvm->mmu_notifier_count++;
	for (; start < end; start += PAGE_SIZE)
		need_tlb_flush |= kvm_unmap_hva(kvm, start);
	spin_unlock(&kvm->mmu_lock);

	/* we've to flush the tlb before the pages can be freed */
	if (need_tlb_flush)
		kvm_flush_remote_tlbs(kvm);
}

static void kvm_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn,
						  struct mm_struct *mm,
						  unsigned long start,
						  unsigned long end)
{
	struct kvm *kvm = mmu_notifier_to_kvm(mn);

	spin_lock(&kvm->mmu_lock);
	/*
	 * This sequence increase will notify the kvm page fault that
	 * the page that is going to be mapped in the spte could have
	 * been freed.
	 */
	kvm->mmu_notifier_seq++;
	/*
	 * The above sequence increase must be visible before the
	 * below count decrease but both values are read by the kvm
	 * page fault under mmu_lock spinlock so we don't need to add
	 * a smb_wmb() here in between the two.
	 */
	kvm->mmu_notifier_count--;
	spin_unlock(&kvm->mmu_lock);

	BUG_ON(kvm->mmu_notifier_count < 0);
}

static int kvm_mmu_notifier_clear_flush_young(struct mmu_notifier *mn,
					      struct mm_struct *mm,
					      unsigned long address)
{
	struct kvm *kvm = mmu_notifier_to_kvm(mn);
	int young;

	spin_lock(&kvm->mmu_lock);
	young = kvm_age_hva(kvm, address);
	spin_unlock(&kvm->mmu_lock);

	if (young)
		kvm_flush_remote_tlbs(kvm);

	return young;
}

927 928 929 930 931 932 933
static void kvm_mmu_notifier_release(struct mmu_notifier *mn,
				     struct mm_struct *mm)
{
	struct kvm *kvm = mmu_notifier_to_kvm(mn);
	kvm_arch_flush_shadow(kvm);
}

934 935 936 937 938
static const struct mmu_notifier_ops kvm_mmu_notifier_ops = {
	.invalidate_page	= kvm_mmu_notifier_invalidate_page,
	.invalidate_range_start	= kvm_mmu_notifier_invalidate_range_start,
	.invalidate_range_end	= kvm_mmu_notifier_invalidate_range_end,
	.clear_flush_young	= kvm_mmu_notifier_clear_flush_young,
939
	.release		= kvm_mmu_notifier_release,
940 941 942
};
#endif /* CONFIG_MMU_NOTIFIER && KVM_ARCH_WANT_MMU_NOTIFIER */

943
static struct kvm *kvm_create_vm(void)
A
Avi Kivity 已提交
944
{
945
	struct kvm *kvm = kvm_arch_create_vm();
946 947 948
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
	struct page *page;
#endif
A
Avi Kivity 已提交
949

950 951
	if (IS_ERR(kvm))
		goto out;
952
#ifdef CONFIG_HAVE_KVM_IRQCHIP
953
	INIT_LIST_HEAD(&kvm->irq_routing);
954 955
	INIT_HLIST_HEAD(&kvm->mask_notifier_list);
#endif
A
Avi Kivity 已提交
956

957 958 959 960 961 962 963 964 965 966
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
	if (!page) {
		kfree(kvm);
		return ERR_PTR(-ENOMEM);
	}
	kvm->coalesced_mmio_ring =
			(struct kvm_coalesced_mmio_ring *)page_address(page);
#endif

967 968 969 970 971 972 973 974 975 976 977 978 979 980 981
#if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER)
	{
		int err;
		kvm->mmu_notifier.ops = &kvm_mmu_notifier_ops;
		err = mmu_notifier_register(&kvm->mmu_notifier, current->mm);
		if (err) {
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
			put_page(page);
#endif
			kfree(kvm);
			return ERR_PTR(err);
		}
	}
#endif

982 983
	kvm->mm = current->mm;
	atomic_inc(&kvm->mm->mm_count);
984
	spin_lock_init(&kvm->mmu_lock);
985
	kvm_io_bus_init(&kvm->pio_bus);
S
Shaohua Li 已提交
986
	mutex_init(&kvm->lock);
987
	kvm_io_bus_init(&kvm->mmio_bus);
988
	init_rwsem(&kvm->slots_lock);
I
Izik Eidus 已提交
989
	atomic_set(&kvm->users_count, 1);
990 991 992
	spin_lock(&kvm_lock);
	list_add(&kvm->vm_list, &vm_list);
	spin_unlock(&kvm_lock);
993 994 995
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
	kvm_coalesced_mmio_init(kvm);
#endif
996
out:
997 998 999
	return kvm;
}

A
Avi Kivity 已提交
1000 1001 1002 1003 1004 1005
/*
 * Free any memory in @free but not in @dont.
 */
static void kvm_free_physmem_slot(struct kvm_memory_slot *free,
				  struct kvm_memory_slot *dont)
{
1006 1007
	if (!dont || free->rmap != dont->rmap)
		vfree(free->rmap);
A
Avi Kivity 已提交
1008 1009 1010 1011

	if (!dont || free->dirty_bitmap != dont->dirty_bitmap)
		vfree(free->dirty_bitmap);

M
Marcelo Tosatti 已提交
1012 1013 1014
	if (!dont || free->lpage_info != dont->lpage_info)
		vfree(free->lpage_info);

A
Avi Kivity 已提交
1015
	free->npages = 0;
A
Al Viro 已提交
1016
	free->dirty_bitmap = NULL;
1017
	free->rmap = NULL;
M
Marcelo Tosatti 已提交
1018
	free->lpage_info = NULL;
A
Avi Kivity 已提交
1019 1020
}

1021
void kvm_free_physmem(struct kvm *kvm)
A
Avi Kivity 已提交
1022 1023 1024 1025
{
	int i;

	for (i = 0; i < kvm->nmemslots; ++i)
A
Al Viro 已提交
1026
		kvm_free_physmem_slot(&kvm->memslots[i], NULL);
A
Avi Kivity 已提交
1027 1028
}

1029 1030
static void kvm_destroy_vm(struct kvm *kvm)
{
1031 1032
	struct mm_struct *mm = kvm->mm;

1033
	kvm_arch_sync_events(kvm);
1034 1035 1036
	spin_lock(&kvm_lock);
	list_del(&kvm->vm_list);
	spin_unlock(&kvm_lock);
1037
	kvm_free_irq_routing(kvm);
1038
	kvm_io_bus_destroy(&kvm->pio_bus);
1039
	kvm_io_bus_destroy(&kvm->mmio_bus);
1040 1041 1042
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
	if (kvm->coalesced_mmio_ring != NULL)
		free_page((unsigned long)kvm->coalesced_mmio_ring);
1043 1044 1045
#endif
#if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER)
	mmu_notifier_unregister(&kvm->mmu_notifier, kvm->mm);
1046 1047
#else
	kvm_arch_flush_shadow(kvm);
1048
#endif
1049
	kvm_arch_destroy_vm(kvm);
1050
	mmdrop(mm);
1051 1052
}

I
Izik Eidus 已提交
1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066
void kvm_get_kvm(struct kvm *kvm)
{
	atomic_inc(&kvm->users_count);
}
EXPORT_SYMBOL_GPL(kvm_get_kvm);

void kvm_put_kvm(struct kvm *kvm)
{
	if (atomic_dec_and_test(&kvm->users_count))
		kvm_destroy_vm(kvm);
}
EXPORT_SYMBOL_GPL(kvm_put_kvm);


1067 1068 1069 1070
static int kvm_vm_release(struct inode *inode, struct file *filp)
{
	struct kvm *kvm = filp->private_data;

I
Izik Eidus 已提交
1071
	kvm_put_kvm(kvm);
A
Avi Kivity 已提交
1072 1073 1074 1075 1076 1077 1078 1079
	return 0;
}

/*
 * Allocate some memory and give it an address in the guest physical address
 * space.
 *
 * Discontiguous memory is allowed, mostly for framebuffers.
1080
 *
1081
 * Must be called holding mmap_sem for write.
A
Avi Kivity 已提交
1082
 */
1083 1084 1085
int __kvm_set_memory_region(struct kvm *kvm,
			    struct kvm_userspace_memory_region *mem,
			    int user_alloc)
A
Avi Kivity 已提交
1086 1087 1088
{
	int r;
	gfn_t base_gfn;
1089
	unsigned long npages, ugfn;
1090
	int largepages;
A
Avi Kivity 已提交
1091 1092 1093 1094 1095 1096 1097 1098 1099 1100
	unsigned long i;
	struct kvm_memory_slot *memslot;
	struct kvm_memory_slot old, new;

	r = -EINVAL;
	/* General sanity checks */
	if (mem->memory_size & (PAGE_SIZE - 1))
		goto out;
	if (mem->guest_phys_addr & (PAGE_SIZE - 1))
		goto out;
S
Sheng Yang 已提交
1101
	if (user_alloc && (mem->userspace_addr & (PAGE_SIZE - 1)))
1102
		goto out;
1103
	if (mem->slot >= KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS)
A
Avi Kivity 已提交
1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123
		goto out;
	if (mem->guest_phys_addr + mem->memory_size < mem->guest_phys_addr)
		goto out;

	memslot = &kvm->memslots[mem->slot];
	base_gfn = mem->guest_phys_addr >> PAGE_SHIFT;
	npages = mem->memory_size >> PAGE_SHIFT;

	if (!npages)
		mem->flags &= ~KVM_MEM_LOG_DIRTY_PAGES;

	new = old = *memslot;

	new.base_gfn = base_gfn;
	new.npages = npages;
	new.flags = mem->flags;

	/* Disallow changing a memory slot's size. */
	r = -EINVAL;
	if (npages && old.npages && npages != old.npages)
1124
		goto out_free;
A
Avi Kivity 已提交
1125 1126 1127 1128 1129 1130

	/* Check for overlaps */
	r = -EEXIST;
	for (i = 0; i < KVM_MEMORY_SLOTS; ++i) {
		struct kvm_memory_slot *s = &kvm->memslots[i];

1131
		if (s == memslot || !s->npages)
A
Avi Kivity 已提交
1132 1133 1134
			continue;
		if (!((base_gfn + npages <= s->base_gfn) ||
		      (base_gfn >= s->base_gfn + s->npages)))
1135
			goto out_free;
A
Avi Kivity 已提交
1136 1137 1138 1139
	}

	/* Free page dirty bitmap if unneeded */
	if (!(new.flags & KVM_MEM_LOG_DIRTY_PAGES))
A
Al Viro 已提交
1140
		new.dirty_bitmap = NULL;
A
Avi Kivity 已提交
1141 1142 1143 1144

	r = -ENOMEM;

	/* Allocate if a slot is being created */
1145
#ifndef CONFIG_S390
1146
	if (npages && !new.rmap) {
M
Mike Day 已提交
1147
		new.rmap = vmalloc(npages * sizeof(struct page *));
1148 1149

		if (!new.rmap)
1150
			goto out_free;
1151 1152

		memset(new.rmap, 0, npages * sizeof(*new.rmap));
1153

1154
		new.user_alloc = user_alloc;
1155 1156 1157 1158 1159 1160 1161 1162 1163
		/*
		 * hva_to_rmmap() serialzies with the mmu_lock and to be
		 * safe it has to ignore memslots with !user_alloc &&
		 * !userspace_addr.
		 */
		if (user_alloc)
			new.userspace_addr = mem->userspace_addr;
		else
			new.userspace_addr = 0;
A
Avi Kivity 已提交
1164
	}
M
Marcelo Tosatti 已提交
1165
	if (npages && !new.lpage_info) {
1166 1167
		largepages = 1 + (base_gfn + npages - 1) / KVM_PAGES_PER_HPAGE;
		largepages -= base_gfn / KVM_PAGES_PER_HPAGE;
M
Marcelo Tosatti 已提交
1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179

		new.lpage_info = vmalloc(largepages * sizeof(*new.lpage_info));

		if (!new.lpage_info)
			goto out_free;

		memset(new.lpage_info, 0, largepages * sizeof(*new.lpage_info));

		if (base_gfn % KVM_PAGES_PER_HPAGE)
			new.lpage_info[0].write_count = 1;
		if ((base_gfn+npages) % KVM_PAGES_PER_HPAGE)
			new.lpage_info[largepages-1].write_count = 1;
1180 1181 1182 1183 1184 1185 1186 1187
		ugfn = new.userspace_addr >> PAGE_SHIFT;
		/*
		 * If the gfn and userspace address are not aligned wrt each
		 * other, disable large page support for this slot
		 */
		if ((base_gfn ^ ugfn) & (KVM_PAGES_PER_HPAGE - 1))
			for (i = 0; i < largepages; ++i)
				new.lpage_info[i].write_count = 1;
M
Marcelo Tosatti 已提交
1188
	}
A
Avi Kivity 已提交
1189 1190 1191 1192 1193 1194 1195

	/* Allocate page dirty bitmap if needed */
	if ((new.flags & KVM_MEM_LOG_DIRTY_PAGES) && !new.dirty_bitmap) {
		unsigned dirty_bytes = ALIGN(npages, BITS_PER_LONG) / 8;

		new.dirty_bitmap = vmalloc(dirty_bytes);
		if (!new.dirty_bitmap)
1196
			goto out_free;
A
Avi Kivity 已提交
1197 1198
		memset(new.dirty_bitmap, 0, dirty_bytes);
	}
1199
#endif /* not defined CONFIG_S390 */
A
Avi Kivity 已提交
1200

1201 1202 1203
	if (!npages)
		kvm_arch_flush_shadow(kvm);

1204 1205 1206 1207
	spin_lock(&kvm->mmu_lock);
	if (mem->slot >= kvm->nmemslots)
		kvm->nmemslots = mem->slot + 1;

1208
	*memslot = new;
1209
	spin_unlock(&kvm->mmu_lock);
1210

1211 1212
	r = kvm_arch_set_memory_region(kvm, mem, old, user_alloc);
	if (r) {
1213
		spin_lock(&kvm->mmu_lock);
1214
		*memslot = old;
1215
		spin_unlock(&kvm->mmu_lock);
1216
		goto out_free;
1217 1218
	}

1219 1220
	kvm_free_physmem_slot(&old, npages ? &new : NULL);
	/* Slot deletion case: we have to update the current slot */
1221
	spin_lock(&kvm->mmu_lock);
1222 1223
	if (!npages)
		*memslot = old;
1224
	spin_unlock(&kvm->mmu_lock);
1225
#ifdef CONFIG_DMAR
B
Ben-Ami Yassour 已提交
1226 1227 1228 1229
	/* map the pages in iommu page table */
	r = kvm_iommu_map_pages(kvm, base_gfn, npages);
	if (r)
		goto out;
1230
#endif
A
Avi Kivity 已提交
1231 1232
	return 0;

1233
out_free:
A
Avi Kivity 已提交
1234 1235 1236
	kvm_free_physmem_slot(&new, &old);
out:
	return r;
1237 1238

}
1239 1240 1241 1242 1243 1244 1245 1246
EXPORT_SYMBOL_GPL(__kvm_set_memory_region);

int kvm_set_memory_region(struct kvm *kvm,
			  struct kvm_userspace_memory_region *mem,
			  int user_alloc)
{
	int r;

1247
	down_write(&kvm->slots_lock);
1248
	r = __kvm_set_memory_region(kvm, mem, user_alloc);
1249
	up_write(&kvm->slots_lock);
1250 1251
	return r;
}
1252 1253
EXPORT_SYMBOL_GPL(kvm_set_memory_region);

1254 1255 1256 1257
int kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
				   struct
				   kvm_userspace_memory_region *mem,
				   int user_alloc)
1258
{
1259 1260
	if (mem->slot >= KVM_MEMORY_SLOTS)
		return -EINVAL;
1261
	return kvm_set_memory_region(kvm, mem, user_alloc);
A
Avi Kivity 已提交
1262 1263
}

1264 1265
int kvm_get_dirty_log(struct kvm *kvm,
			struct kvm_dirty_log *log, int *is_dirty)
A
Avi Kivity 已提交
1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280
{
	struct kvm_memory_slot *memslot;
	int r, i;
	int n;
	unsigned long any = 0;

	r = -EINVAL;
	if (log->slot >= KVM_MEMORY_SLOTS)
		goto out;

	memslot = &kvm->memslots[log->slot];
	r = -ENOENT;
	if (!memslot->dirty_bitmap)
		goto out;

1281
	n = ALIGN(memslot->npages, BITS_PER_LONG) / 8;
A
Avi Kivity 已提交
1282

1283
	for (i = 0; !any && i < n/sizeof(long); ++i)
A
Avi Kivity 已提交
1284 1285 1286 1287 1288 1289
		any = memslot->dirty_bitmap[i];

	r = -EFAULT;
	if (copy_to_user(log->dirty_bitmap, memslot->dirty_bitmap, n))
		goto out;

1290 1291
	if (any)
		*is_dirty = 1;
A
Avi Kivity 已提交
1292 1293 1294 1295 1296 1297

	r = 0;
out:
	return r;
}

1298 1299 1300 1301 1302 1303
int is_error_page(struct page *page)
{
	return page == bad_page;
}
EXPORT_SYMBOL_GPL(is_error_page);

1304 1305 1306 1307 1308 1309
int is_error_pfn(pfn_t pfn)
{
	return pfn == bad_pfn;
}
EXPORT_SYMBOL_GPL(is_error_pfn);

I
Izik Eidus 已提交
1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320
static inline unsigned long bad_hva(void)
{
	return PAGE_OFFSET;
}

int kvm_is_error_hva(unsigned long addr)
{
	return addr == bad_hva();
}
EXPORT_SYMBOL_GPL(kvm_is_error_hva);

1321
struct kvm_memory_slot *gfn_to_memslot_unaliased(struct kvm *kvm, gfn_t gfn)
A
Avi Kivity 已提交
1322 1323 1324 1325 1326 1327 1328 1329 1330 1331
{
	int i;

	for (i = 0; i < kvm->nmemslots; ++i) {
		struct kvm_memory_slot *memslot = &kvm->memslots[i];

		if (gfn >= memslot->base_gfn
		    && gfn < memslot->base_gfn + memslot->npages)
			return memslot;
	}
A
Al Viro 已提交
1332
	return NULL;
A
Avi Kivity 已提交
1333
}
1334
EXPORT_SYMBOL_GPL(gfn_to_memslot_unaliased);
1335 1336 1337 1338

struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn)
{
	gfn = unalias_gfn(kvm, gfn);
1339
	return gfn_to_memslot_unaliased(kvm, gfn);
1340
}
A
Avi Kivity 已提交
1341

1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357
int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn)
{
	int i;

	gfn = unalias_gfn(kvm, gfn);
	for (i = 0; i < KVM_MEMORY_SLOTS; ++i) {
		struct kvm_memory_slot *memslot = &kvm->memslots[i];

		if (gfn >= memslot->base_gfn
		    && gfn < memslot->base_gfn + memslot->npages)
			return 1;
	}
	return 0;
}
EXPORT_SYMBOL_GPL(kvm_is_visible_gfn);

M
Marcelo Tosatti 已提交
1358
unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn)
I
Izik Eidus 已提交
1359 1360 1361 1362
{
	struct kvm_memory_slot *slot;

	gfn = unalias_gfn(kvm, gfn);
1363
	slot = gfn_to_memslot_unaliased(kvm, gfn);
I
Izik Eidus 已提交
1364 1365 1366 1367
	if (!slot)
		return bad_hva();
	return (slot->userspace_addr + (gfn - slot->base_gfn) * PAGE_SIZE);
}
1368
EXPORT_SYMBOL_GPL(gfn_to_hva);
I
Izik Eidus 已提交
1369

1370
pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn)
A
Avi Kivity 已提交
1371
{
1372
	struct page *page[1];
I
Izik Eidus 已提交
1373
	unsigned long addr;
1374
	int npages;
1375
	pfn_t pfn;
A
Avi Kivity 已提交
1376

1377 1378
	might_sleep();

I
Izik Eidus 已提交
1379 1380
	addr = gfn_to_hva(kvm, gfn);
	if (kvm_is_error_hva(addr)) {
1381
		get_page(bad_page);
1382
		return page_to_pfn(bad_page);
1383
	}
1384

1385
	npages = get_user_pages_fast(addr, 1, 1, page);
I
Izik Eidus 已提交
1386

1387 1388 1389
	if (unlikely(npages != 1)) {
		struct vm_area_struct *vma;

1390
		down_read(&current->mm->mmap_sem);
1391
		vma = find_vma(current->mm, addr);
1392

1393 1394
		if (vma == NULL || addr < vma->vm_start ||
		    !(vma->vm_flags & VM_PFNMAP)) {
1395
			up_read(&current->mm->mmap_sem);
1396 1397 1398 1399 1400
			get_page(bad_page);
			return page_to_pfn(bad_page);
		}

		pfn = ((addr - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff;
1401
		up_read(&current->mm->mmap_sem);
1402
		BUG_ON(!kvm_is_mmio_pfn(pfn));
1403 1404
	} else
		pfn = page_to_pfn(page[0]);
1405

1406
	return pfn;
1407 1408 1409 1410 1411 1412
}

EXPORT_SYMBOL_GPL(gfn_to_pfn);

struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn)
{
1413 1414 1415
	pfn_t pfn;

	pfn = gfn_to_pfn(kvm, gfn);
1416
	if (!kvm_is_mmio_pfn(pfn))
1417 1418
		return pfn_to_page(pfn);

1419
	WARN_ON(kvm_is_mmio_pfn(pfn));
1420 1421 1422

	get_page(bad_page);
	return bad_page;
A
Avi Kivity 已提交
1423
}
1424

A
Avi Kivity 已提交
1425 1426
EXPORT_SYMBOL_GPL(gfn_to_page);

1427 1428
void kvm_release_page_clean(struct page *page)
{
1429
	kvm_release_pfn_clean(page_to_pfn(page));
1430 1431 1432
}
EXPORT_SYMBOL_GPL(kvm_release_page_clean);

1433 1434
void kvm_release_pfn_clean(pfn_t pfn)
{
1435
	if (!kvm_is_mmio_pfn(pfn))
1436
		put_page(pfn_to_page(pfn));
1437 1438 1439
}
EXPORT_SYMBOL_GPL(kvm_release_pfn_clean);

1440
void kvm_release_page_dirty(struct page *page)
1441
{
1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460
	kvm_release_pfn_dirty(page_to_pfn(page));
}
EXPORT_SYMBOL_GPL(kvm_release_page_dirty);

void kvm_release_pfn_dirty(pfn_t pfn)
{
	kvm_set_pfn_dirty(pfn);
	kvm_release_pfn_clean(pfn);
}
EXPORT_SYMBOL_GPL(kvm_release_pfn_dirty);

void kvm_set_page_dirty(struct page *page)
{
	kvm_set_pfn_dirty(page_to_pfn(page));
}
EXPORT_SYMBOL_GPL(kvm_set_page_dirty);

void kvm_set_pfn_dirty(pfn_t pfn)
{
1461
	if (!kvm_is_mmio_pfn(pfn)) {
1462 1463 1464 1465
		struct page *page = pfn_to_page(pfn);
		if (!PageReserved(page))
			SetPageDirty(page);
	}
1466
}
1467 1468 1469 1470
EXPORT_SYMBOL_GPL(kvm_set_pfn_dirty);

void kvm_set_pfn_accessed(pfn_t pfn)
{
1471
	if (!kvm_is_mmio_pfn(pfn))
1472
		mark_page_accessed(pfn_to_page(pfn));
1473 1474 1475 1476 1477
}
EXPORT_SYMBOL_GPL(kvm_set_pfn_accessed);

void kvm_get_pfn(pfn_t pfn)
{
1478
	if (!kvm_is_mmio_pfn(pfn))
1479
		get_page(pfn_to_page(pfn));
1480 1481
}
EXPORT_SYMBOL_GPL(kvm_get_pfn);
1482

1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493
static int next_segment(unsigned long len, int offset)
{
	if (len > PAGE_SIZE - offset)
		return PAGE_SIZE - offset;
	else
		return len;
}

int kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset,
			int len)
{
1494 1495
	int r;
	unsigned long addr;
1496

1497 1498 1499 1500 1501
	addr = gfn_to_hva(kvm, gfn);
	if (kvm_is_error_hva(addr))
		return -EFAULT;
	r = copy_from_user(data, (void __user *)addr + offset, len);
	if (r)
1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526
		return -EFAULT;
	return 0;
}
EXPORT_SYMBOL_GPL(kvm_read_guest_page);

int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len)
{
	gfn_t gfn = gpa >> PAGE_SHIFT;
	int seg;
	int offset = offset_in_page(gpa);
	int ret;

	while ((seg = next_segment(len, offset)) != 0) {
		ret = kvm_read_guest_page(kvm, gfn, data, offset, seg);
		if (ret < 0)
			return ret;
		offset = 0;
		len -= seg;
		data += seg;
		++gfn;
	}
	return 0;
}
EXPORT_SYMBOL_GPL(kvm_read_guest);

1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537
int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
			  unsigned long len)
{
	int r;
	unsigned long addr;
	gfn_t gfn = gpa >> PAGE_SHIFT;
	int offset = offset_in_page(gpa);

	addr = gfn_to_hva(kvm, gfn);
	if (kvm_is_error_hva(addr))
		return -EFAULT;
1538
	pagefault_disable();
1539
	r = __copy_from_user_inatomic(data, (void __user *)addr + offset, len);
1540
	pagefault_enable();
1541 1542 1543 1544 1545 1546
	if (r)
		return -EFAULT;
	return 0;
}
EXPORT_SYMBOL(kvm_read_guest_atomic);

1547 1548 1549
int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn, const void *data,
			 int offset, int len)
{
1550 1551
	int r;
	unsigned long addr;
1552

1553 1554 1555 1556 1557
	addr = gfn_to_hva(kvm, gfn);
	if (kvm_is_error_hva(addr))
		return -EFAULT;
	r = copy_to_user((void __user *)addr + offset, data, len);
	if (r)
1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585
		return -EFAULT;
	mark_page_dirty(kvm, gfn);
	return 0;
}
EXPORT_SYMBOL_GPL(kvm_write_guest_page);

int kvm_write_guest(struct kvm *kvm, gpa_t gpa, const void *data,
		    unsigned long len)
{
	gfn_t gfn = gpa >> PAGE_SHIFT;
	int seg;
	int offset = offset_in_page(gpa);
	int ret;

	while ((seg = next_segment(len, offset)) != 0) {
		ret = kvm_write_guest_page(kvm, gfn, data, offset, seg);
		if (ret < 0)
			return ret;
		offset = 0;
		len -= seg;
		data += seg;
		++gfn;
	}
	return 0;
}

int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len)
{
1586
	return kvm_write_guest_page(kvm, gfn, empty_zero_page, offset, len);
1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608
}
EXPORT_SYMBOL_GPL(kvm_clear_guest_page);

int kvm_clear_guest(struct kvm *kvm, gpa_t gpa, unsigned long len)
{
	gfn_t gfn = gpa >> PAGE_SHIFT;
	int seg;
	int offset = offset_in_page(gpa);
	int ret;

        while ((seg = next_segment(len, offset)) != 0) {
		ret = kvm_clear_guest_page(kvm, gfn, offset, seg);
		if (ret < 0)
			return ret;
		offset = 0;
		len -= seg;
		++gfn;
	}
	return 0;
}
EXPORT_SYMBOL_GPL(kvm_clear_guest);

A
Avi Kivity 已提交
1609 1610
void mark_page_dirty(struct kvm *kvm, gfn_t gfn)
{
1611
	struct kvm_memory_slot *memslot;
A
Avi Kivity 已提交
1612

1613
	gfn = unalias_gfn(kvm, gfn);
1614
	memslot = gfn_to_memslot_unaliased(kvm, gfn);
R
Rusty Russell 已提交
1615 1616
	if (memslot && memslot->dirty_bitmap) {
		unsigned long rel_gfn = gfn - memslot->base_gfn;
A
Avi Kivity 已提交
1617

R
Rusty Russell 已提交
1618 1619 1620
		/* avoid RMW */
		if (!test_bit(rel_gfn, memslot->dirty_bitmap))
			set_bit(rel_gfn, memslot->dirty_bitmap);
A
Avi Kivity 已提交
1621 1622 1623
	}
}

E
Eddie Dong 已提交
1624 1625 1626
/*
 * The vCPU has executed a HLT instruction with in-kernel mode enabled.
 */
1627
void kvm_vcpu_block(struct kvm_vcpu *vcpu)
1628
{
1629 1630 1631 1632 1633
	DEFINE_WAIT(wait);

	for (;;) {
		prepare_to_wait(&vcpu->wq, &wait, TASK_INTERRUPTIBLE);

1634 1635
		if ((kvm_arch_interrupt_allowed(vcpu) &&
					kvm_cpu_has_interrupt(vcpu)) ||
1636
				kvm_arch_vcpu_runnable(vcpu)) {
1637
			set_bit(KVM_REQ_UNHALT, &vcpu->requests);
1638
			break;
1639
		}
1640 1641
		if (kvm_cpu_has_pending_timer(vcpu))
			break;
1642 1643 1644
		if (signal_pending(current))
			break;

E
Eddie Dong 已提交
1645 1646 1647 1648
		vcpu_put(vcpu);
		schedule();
		vcpu_load(vcpu);
	}
1649

1650
	finish_wait(&vcpu->wq, &wait);
E
Eddie Dong 已提交
1651 1652
}

A
Avi Kivity 已提交
1653 1654
void kvm_resched(struct kvm_vcpu *vcpu)
{
1655 1656
	if (!need_resched())
		return;
A
Avi Kivity 已提交
1657 1658 1659 1660
	cond_resched();
}
EXPORT_SYMBOL_GPL(kvm_resched);

1661
static int kvm_vcpu_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
1662 1663 1664 1665
{
	struct kvm_vcpu *vcpu = vma->vm_file->private_data;
	struct page *page;

1666
	if (vmf->pgoff == 0)
1667
		page = virt_to_page(vcpu->run);
A
Avi Kivity 已提交
1668
#ifdef CONFIG_X86
1669
	else if (vmf->pgoff == KVM_PIO_PAGE_OFFSET)
1670
		page = virt_to_page(vcpu->arch.pio_data);
1671 1672 1673 1674
#endif
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
	else if (vmf->pgoff == KVM_COALESCED_MMIO_PAGE_OFFSET)
		page = virt_to_page(vcpu->kvm->coalesced_mmio_ring);
A
Avi Kivity 已提交
1675
#endif
1676
	else
1677
		return VM_FAULT_SIGBUS;
1678
	get_page(page);
1679 1680
	vmf->page = page;
	return 0;
1681 1682 1683
}

static struct vm_operations_struct kvm_vcpu_vm_ops = {
1684
	.fault = kvm_vcpu_fault,
1685 1686 1687 1688 1689 1690 1691 1692
};

static int kvm_vcpu_mmap(struct file *file, struct vm_area_struct *vma)
{
	vma->vm_ops = &kvm_vcpu_vm_ops;
	return 0;
}

A
Avi Kivity 已提交
1693 1694 1695 1696
static int kvm_vcpu_release(struct inode *inode, struct file *filp)
{
	struct kvm_vcpu *vcpu = filp->private_data;

A
Al Viro 已提交
1697
	kvm_put_kvm(vcpu->kvm);
A
Avi Kivity 已提交
1698 1699 1700
	return 0;
}

1701
static struct file_operations kvm_vcpu_fops = {
A
Avi Kivity 已提交
1702 1703 1704
	.release        = kvm_vcpu_release,
	.unlocked_ioctl = kvm_vcpu_ioctl,
	.compat_ioctl   = kvm_vcpu_ioctl,
1705
	.mmap           = kvm_vcpu_mmap,
A
Avi Kivity 已提交
1706 1707 1708 1709 1710 1711 1712
};

/*
 * Allocates an inode for the vcpu.
 */
static int create_vcpu_fd(struct kvm_vcpu *vcpu)
{
1713
	int fd = anon_inode_getfd("kvm-vcpu", &kvm_vcpu_fops, vcpu, 0);
A
Al Viro 已提交
1714
	if (fd < 0)
A
Al Viro 已提交
1715
		kvm_put_kvm(vcpu->kvm);
A
Avi Kivity 已提交
1716 1717 1718
	return fd;
}

1719 1720 1721 1722 1723 1724 1725 1726 1727
/*
 * Creates some virtual cpus.  Good luck creating more than one.
 */
static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, int n)
{
	int r;
	struct kvm_vcpu *vcpu;

	if (!valid_vcpu(n))
R
Rusty Russell 已提交
1728
		return -EINVAL;
1729

1730
	vcpu = kvm_arch_vcpu_create(kvm, n);
R
Rusty Russell 已提交
1731 1732
	if (IS_ERR(vcpu))
		return PTR_ERR(vcpu);
1733

1734 1735
	preempt_notifier_init(&vcpu->preempt_notifier, &kvm_preempt_ops);

1736 1737
	r = kvm_arch_vcpu_setup(vcpu);
	if (r)
1738
		return r;
1739

S
Shaohua Li 已提交
1740
	mutex_lock(&kvm->lock);
R
Rusty Russell 已提交
1741 1742
	if (kvm->vcpus[n]) {
		r = -EEXIST;
1743
		goto vcpu_destroy;
R
Rusty Russell 已提交
1744 1745
	}
	kvm->vcpus[n] = vcpu;
S
Shaohua Li 已提交
1746
	mutex_unlock(&kvm->lock);
1747

R
Rusty Russell 已提交
1748
	/* Now it's all set up, let userspace reach it */
A
Al Viro 已提交
1749
	kvm_get_kvm(kvm);
A
Avi Kivity 已提交
1750 1751
	r = create_vcpu_fd(vcpu);
	if (r < 0)
R
Rusty Russell 已提交
1752 1753
		goto unlink;
	return r;
1754

R
Rusty Russell 已提交
1755
unlink:
S
Shaohua Li 已提交
1756
	mutex_lock(&kvm->lock);
R
Rusty Russell 已提交
1757
	kvm->vcpus[n] = NULL;
1758
vcpu_destroy:
1759
	mutex_unlock(&kvm->lock);
1760
	kvm_arch_vcpu_destroy(vcpu);
1761 1762 1763
	return r;
}

A
Avi Kivity 已提交
1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774
static int kvm_vcpu_ioctl_set_sigmask(struct kvm_vcpu *vcpu, sigset_t *sigset)
{
	if (sigset) {
		sigdelsetmask(sigset, sigmask(SIGKILL)|sigmask(SIGSTOP));
		vcpu->sigset_active = 1;
		vcpu->sigset = *sigset;
	} else
		vcpu->sigset_active = 0;
	return 0;
}

S
Sheng Yang 已提交
1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856
#ifdef __KVM_HAVE_MSIX
static int kvm_vm_ioctl_set_msix_nr(struct kvm *kvm,
				    struct kvm_assigned_msix_nr *entry_nr)
{
	int r = 0;
	struct kvm_assigned_dev_kernel *adev;

	mutex_lock(&kvm->lock);

	adev = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head,
				      entry_nr->assigned_dev_id);
	if (!adev) {
		r = -EINVAL;
		goto msix_nr_out;
	}

	if (adev->entries_nr == 0) {
		adev->entries_nr = entry_nr->entry_nr;
		if (adev->entries_nr == 0 ||
		    adev->entries_nr >= KVM_MAX_MSIX_PER_DEV) {
			r = -EINVAL;
			goto msix_nr_out;
		}

		adev->host_msix_entries = kzalloc(sizeof(struct msix_entry) *
						entry_nr->entry_nr,
						GFP_KERNEL);
		if (!adev->host_msix_entries) {
			r = -ENOMEM;
			goto msix_nr_out;
		}
		adev->guest_msix_entries = kzalloc(
				sizeof(struct kvm_guest_msix_entry) *
				entry_nr->entry_nr, GFP_KERNEL);
		if (!adev->guest_msix_entries) {
			kfree(adev->host_msix_entries);
			r = -ENOMEM;
			goto msix_nr_out;
		}
	} else /* Not allowed set MSI-X number twice */
		r = -EINVAL;
msix_nr_out:
	mutex_unlock(&kvm->lock);
	return r;
}

static int kvm_vm_ioctl_set_msix_entry(struct kvm *kvm,
				       struct kvm_assigned_msix_entry *entry)
{
	int r = 0, i;
	struct kvm_assigned_dev_kernel *adev;

	mutex_lock(&kvm->lock);

	adev = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head,
				      entry->assigned_dev_id);

	if (!adev) {
		r = -EINVAL;
		goto msix_entry_out;
	}

	for (i = 0; i < adev->entries_nr; i++)
		if (adev->guest_msix_entries[i].vector == 0 ||
		    adev->guest_msix_entries[i].entry == entry->entry) {
			adev->guest_msix_entries[i].entry = entry->entry;
			adev->guest_msix_entries[i].vector = entry->gsi;
			adev->host_msix_entries[i].entry = entry->entry;
			break;
		}
	if (i == adev->entries_nr) {
		r = -ENOSPC;
		goto msix_entry_out;
	}

msix_entry_out:
	mutex_unlock(&kvm->lock);

	return r;
}
#endif

A
Avi Kivity 已提交
1857 1858
static long kvm_vcpu_ioctl(struct file *filp,
			   unsigned int ioctl, unsigned long arg)
A
Avi Kivity 已提交
1859
{
A
Avi Kivity 已提交
1860
	struct kvm_vcpu *vcpu = filp->private_data;
A
Al Viro 已提交
1861
	void __user *argp = (void __user *)arg;
1862
	int r;
1863 1864
	struct kvm_fpu *fpu = NULL;
	struct kvm_sregs *kvm_sregs = NULL;
A
Avi Kivity 已提交
1865

1866 1867
	if (vcpu->kvm->mm != current->mm)
		return -EIO;
A
Avi Kivity 已提交
1868
	switch (ioctl) {
1869
	case KVM_RUN:
1870 1871 1872
		r = -EINVAL;
		if (arg)
			goto out;
1873
		r = kvm_arch_vcpu_ioctl_run(vcpu, vcpu->run);
A
Avi Kivity 已提交
1874 1875
		break;
	case KVM_GET_REGS: {
1876
		struct kvm_regs *kvm_regs;
A
Avi Kivity 已提交
1877

1878 1879 1880
		r = -ENOMEM;
		kvm_regs = kzalloc(sizeof(struct kvm_regs), GFP_KERNEL);
		if (!kvm_regs)
A
Avi Kivity 已提交
1881
			goto out;
1882 1883 1884
		r = kvm_arch_vcpu_ioctl_get_regs(vcpu, kvm_regs);
		if (r)
			goto out_free1;
A
Avi Kivity 已提交
1885
		r = -EFAULT;
1886 1887
		if (copy_to_user(argp, kvm_regs, sizeof(struct kvm_regs)))
			goto out_free1;
A
Avi Kivity 已提交
1888
		r = 0;
1889 1890
out_free1:
		kfree(kvm_regs);
A
Avi Kivity 已提交
1891 1892 1893
		break;
	}
	case KVM_SET_REGS: {
1894
		struct kvm_regs *kvm_regs;
A
Avi Kivity 已提交
1895

1896 1897 1898
		r = -ENOMEM;
		kvm_regs = kzalloc(sizeof(struct kvm_regs), GFP_KERNEL);
		if (!kvm_regs)
A
Avi Kivity 已提交
1899
			goto out;
1900 1901 1902 1903
		r = -EFAULT;
		if (copy_from_user(kvm_regs, argp, sizeof(struct kvm_regs)))
			goto out_free2;
		r = kvm_arch_vcpu_ioctl_set_regs(vcpu, kvm_regs);
A
Avi Kivity 已提交
1904
		if (r)
1905
			goto out_free2;
A
Avi Kivity 已提交
1906
		r = 0;
1907 1908
out_free2:
		kfree(kvm_regs);
A
Avi Kivity 已提交
1909 1910 1911
		break;
	}
	case KVM_GET_SREGS: {
1912 1913 1914 1915 1916
		kvm_sregs = kzalloc(sizeof(struct kvm_sregs), GFP_KERNEL);
		r = -ENOMEM;
		if (!kvm_sregs)
			goto out;
		r = kvm_arch_vcpu_ioctl_get_sregs(vcpu, kvm_sregs);
A
Avi Kivity 已提交
1917 1918 1919
		if (r)
			goto out;
		r = -EFAULT;
1920
		if (copy_to_user(argp, kvm_sregs, sizeof(struct kvm_sregs)))
A
Avi Kivity 已提交
1921 1922 1923 1924 1925
			goto out;
		r = 0;
		break;
	}
	case KVM_SET_SREGS: {
1926 1927 1928 1929
		kvm_sregs = kmalloc(sizeof(struct kvm_sregs), GFP_KERNEL);
		r = -ENOMEM;
		if (!kvm_sregs)
			goto out;
A
Avi Kivity 已提交
1930
		r = -EFAULT;
1931
		if (copy_from_user(kvm_sregs, argp, sizeof(struct kvm_sregs)))
A
Avi Kivity 已提交
1932
			goto out;
1933
		r = kvm_arch_vcpu_ioctl_set_sregs(vcpu, kvm_sregs);
A
Avi Kivity 已提交
1934 1935 1936 1937 1938
		if (r)
			goto out;
		r = 0;
		break;
	}
1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962
	case KVM_GET_MP_STATE: {
		struct kvm_mp_state mp_state;

		r = kvm_arch_vcpu_ioctl_get_mpstate(vcpu, &mp_state);
		if (r)
			goto out;
		r = -EFAULT;
		if (copy_to_user(argp, &mp_state, sizeof mp_state))
			goto out;
		r = 0;
		break;
	}
	case KVM_SET_MP_STATE: {
		struct kvm_mp_state mp_state;

		r = -EFAULT;
		if (copy_from_user(&mp_state, argp, sizeof mp_state))
			goto out;
		r = kvm_arch_vcpu_ioctl_set_mpstate(vcpu, &mp_state);
		if (r)
			goto out;
		r = 0;
		break;
	}
A
Avi Kivity 已提交
1963 1964 1965 1966
	case KVM_TRANSLATE: {
		struct kvm_translation tr;

		r = -EFAULT;
A
Al Viro 已提交
1967
		if (copy_from_user(&tr, argp, sizeof tr))
A
Avi Kivity 已提交
1968
			goto out;
1969
		r = kvm_arch_vcpu_ioctl_translate(vcpu, &tr);
A
Avi Kivity 已提交
1970 1971 1972
		if (r)
			goto out;
		r = -EFAULT;
A
Al Viro 已提交
1973
		if (copy_to_user(argp, &tr, sizeof tr))
A
Avi Kivity 已提交
1974 1975 1976 1977
			goto out;
		r = 0;
		break;
	}
J
Jan Kiszka 已提交
1978 1979
	case KVM_SET_GUEST_DEBUG: {
		struct kvm_guest_debug dbg;
A
Avi Kivity 已提交
1980 1981

		r = -EFAULT;
A
Al Viro 已提交
1982
		if (copy_from_user(&dbg, argp, sizeof dbg))
A
Avi Kivity 已提交
1983
			goto out;
J
Jan Kiszka 已提交
1984
		r = kvm_arch_vcpu_ioctl_set_guest_debug(vcpu, &dbg);
A
Avi Kivity 已提交
1985 1986 1987 1988 1989
		if (r)
			goto out;
		r = 0;
		break;
	}
A
Avi Kivity 已提交
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
	case KVM_SET_SIGNAL_MASK: {
		struct kvm_signal_mask __user *sigmask_arg = argp;
		struct kvm_signal_mask kvm_sigmask;
		sigset_t sigset, *p;

		p = NULL;
		if (argp) {
			r = -EFAULT;
			if (copy_from_user(&kvm_sigmask, argp,
					   sizeof kvm_sigmask))
				goto out;
			r = -EINVAL;
			if (kvm_sigmask.len != sizeof sigset)
				goto out;
			r = -EFAULT;
			if (copy_from_user(&sigset, sigmask_arg->sigset,
					   sizeof sigset))
				goto out;
			p = &sigset;
		}
		r = kvm_vcpu_ioctl_set_sigmask(vcpu, &sigset);
		break;
	}
A
Avi Kivity 已提交
2013
	case KVM_GET_FPU: {
2014 2015 2016 2017 2018
		fpu = kzalloc(sizeof(struct kvm_fpu), GFP_KERNEL);
		r = -ENOMEM;
		if (!fpu)
			goto out;
		r = kvm_arch_vcpu_ioctl_get_fpu(vcpu, fpu);
A
Avi Kivity 已提交
2019 2020 2021
		if (r)
			goto out;
		r = -EFAULT;
2022
		if (copy_to_user(argp, fpu, sizeof(struct kvm_fpu)))
A
Avi Kivity 已提交
2023 2024 2025 2026 2027
			goto out;
		r = 0;
		break;
	}
	case KVM_SET_FPU: {
2028 2029 2030 2031
		fpu = kmalloc(sizeof(struct kvm_fpu), GFP_KERNEL);
		r = -ENOMEM;
		if (!fpu)
			goto out;
A
Avi Kivity 已提交
2032
		r = -EFAULT;
2033
		if (copy_from_user(fpu, argp, sizeof(struct kvm_fpu)))
A
Avi Kivity 已提交
2034
			goto out;
2035
		r = kvm_arch_vcpu_ioctl_set_fpu(vcpu, fpu);
A
Avi Kivity 已提交
2036 2037 2038 2039 2040
		if (r)
			goto out;
		r = 0;
		break;
	}
A
Avi Kivity 已提交
2041
	default:
2042
		r = kvm_arch_vcpu_ioctl(filp, ioctl, arg);
A
Avi Kivity 已提交
2043 2044
	}
out:
2045 2046
	kfree(fpu);
	kfree(kvm_sregs);
A
Avi Kivity 已提交
2047 2048 2049 2050 2051 2052 2053 2054
	return r;
}

static long kvm_vm_ioctl(struct file *filp,
			   unsigned int ioctl, unsigned long arg)
{
	struct kvm *kvm = filp->private_data;
	void __user *argp = (void __user *)arg;
2055
	int r;
A
Avi Kivity 已提交
2056

2057 2058
	if (kvm->mm != current->mm)
		return -EIO;
A
Avi Kivity 已提交
2059 2060 2061 2062 2063 2064
	switch (ioctl) {
	case KVM_CREATE_VCPU:
		r = kvm_vm_ioctl_create_vcpu(kvm, arg);
		if (r < 0)
			goto out;
		break;
2065 2066 2067 2068 2069 2070 2071 2072 2073
	case KVM_SET_USER_MEMORY_REGION: {
		struct kvm_userspace_memory_region kvm_userspace_mem;

		r = -EFAULT;
		if (copy_from_user(&kvm_userspace_mem, argp,
						sizeof kvm_userspace_mem))
			goto out;

		r = kvm_vm_ioctl_set_memory_region(kvm, &kvm_userspace_mem, 1);
A
Avi Kivity 已提交
2074 2075 2076 2077 2078 2079 2080 2081
		if (r)
			goto out;
		break;
	}
	case KVM_GET_DIRTY_LOG: {
		struct kvm_dirty_log log;

		r = -EFAULT;
A
Al Viro 已提交
2082
		if (copy_from_user(&log, argp, sizeof log))
A
Avi Kivity 已提交
2083
			goto out;
2084
		r = kvm_vm_ioctl_get_dirty_log(kvm, &log);
A
Avi Kivity 已提交
2085 2086 2087 2088
		if (r)
			goto out;
		break;
	}
2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
	case KVM_REGISTER_COALESCED_MMIO: {
		struct kvm_coalesced_mmio_zone zone;
		r = -EFAULT;
		if (copy_from_user(&zone, argp, sizeof zone))
			goto out;
		r = -ENXIO;
		r = kvm_vm_ioctl_register_coalesced_mmio(kvm, &zone);
		if (r)
			goto out;
		r = 0;
		break;
	}
	case KVM_UNREGISTER_COALESCED_MMIO: {
		struct kvm_coalesced_mmio_zone zone;
		r = -EFAULT;
		if (copy_from_user(&zone, argp, sizeof zone))
			goto out;
		r = -ENXIO;
		r = kvm_vm_ioctl_unregister_coalesced_mmio(kvm, &zone);
		if (r)
			goto out;
		r = 0;
		break;
	}
2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127
#endif
#ifdef KVM_CAP_DEVICE_ASSIGNMENT
	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: {
2128 2129 2130 2131 2132
		r = -EOPNOTSUPP;
		break;
	}
#ifdef KVM_CAP_ASSIGN_DEV_IRQ
	case KVM_ASSIGN_DEV_IRQ: {
2133 2134 2135 2136 2137 2138 2139 2140 2141 2142
		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;
		break;
	}
2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154
	case KVM_DEASSIGN_DEV_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_deassign_dev_irq(kvm, &assigned_irq);
		if (r)
			goto out;
		break;
	}
#endif
W
Weidong Han 已提交
2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167
#endif
#ifdef KVM_CAP_DEVICE_DEASSIGNMENT
	case KVM_DEASSIGN_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_deassign_device(kvm, &assigned_dev);
		if (r)
			goto out;
		break;
	}
2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197
#endif
#ifdef KVM_CAP_IRQ_ROUTING
	case KVM_SET_GSI_ROUTING: {
		struct kvm_irq_routing routing;
		struct kvm_irq_routing __user *urouting;
		struct kvm_irq_routing_entry *entries;

		r = -EFAULT;
		if (copy_from_user(&routing, argp, sizeof(routing)))
			goto out;
		r = -EINVAL;
		if (routing.nr >= KVM_MAX_IRQ_ROUTES)
			goto out;
		if (routing.flags)
			goto out;
		r = -ENOMEM;
		entries = vmalloc(routing.nr * sizeof(*entries));
		if (!entries)
			goto out;
		r = -EFAULT;
		urouting = argp;
		if (copy_from_user(entries, urouting->entries,
				   routing.nr * sizeof(*entries)))
			goto out_free_irq_routing;
		r = kvm_set_irq_routing(kvm, entries, routing.nr,
					routing.flags);
	out_free_irq_routing:
		vfree(entries);
		break;
	}
S
Sheng Yang 已提交
2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218
#ifdef __KVM_HAVE_MSIX
	case KVM_ASSIGN_SET_MSIX_NR: {
		struct kvm_assigned_msix_nr entry_nr;
		r = -EFAULT;
		if (copy_from_user(&entry_nr, argp, sizeof entry_nr))
			goto out;
		r = kvm_vm_ioctl_set_msix_nr(kvm, &entry_nr);
		if (r)
			goto out;
		break;
	}
	case KVM_ASSIGN_SET_MSIX_ENTRY: {
		struct kvm_assigned_msix_entry entry;
		r = -EFAULT;
		if (copy_from_user(&entry, argp, sizeof entry))
			goto out;
		r = kvm_vm_ioctl_set_msix_entry(kvm, &entry);
		if (r)
			goto out;
		break;
	}
2219
#endif
S
Sheng Yang 已提交
2220
#endif /* KVM_CAP_IRQ_ROUTING */
2221
	default:
2222
		r = kvm_arch_vm_ioctl(filp, ioctl, arg);
2223 2224 2225 2226 2227
	}
out:
	return r;
}

2228
static int kvm_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
2229
{
2230 2231 2232 2233
	struct page *page[1];
	unsigned long addr;
	int npages;
	gfn_t gfn = vmf->pgoff;
2234 2235
	struct kvm *kvm = vma->vm_file->private_data;

2236 2237
	addr = gfn_to_hva(kvm, gfn);
	if (kvm_is_error_hva(addr))
2238
		return VM_FAULT_SIGBUS;
2239 2240 2241 2242

	npages = get_user_pages(current, current->mm, addr, 1, 1, 0, page,
				NULL);
	if (unlikely(npages != 1))
2243
		return VM_FAULT_SIGBUS;
2244 2245

	vmf->page = page[0];
2246
	return 0;
2247 2248 2249
}

static struct vm_operations_struct kvm_vm_vm_ops = {
2250
	.fault = kvm_vm_fault,
2251 2252 2253 2254 2255 2256 2257 2258
};

static int kvm_vm_mmap(struct file *file, struct vm_area_struct *vma)
{
	vma->vm_ops = &kvm_vm_vm_ops;
	return 0;
}

2259
static struct file_operations kvm_vm_fops = {
2260 2261 2262 2263 2264 2265 2266 2267
	.release        = kvm_vm_release,
	.unlocked_ioctl = kvm_vm_ioctl,
	.compat_ioctl   = kvm_vm_ioctl,
	.mmap           = kvm_vm_mmap,
};

static int kvm_dev_ioctl_create_vm(void)
{
A
Al Viro 已提交
2268
	int fd;
2269 2270 2271
	struct kvm *kvm;

	kvm = kvm_create_vm();
2272 2273
	if (IS_ERR(kvm))
		return PTR_ERR(kvm);
2274
	fd = anon_inode_getfd("kvm-vm", &kvm_vm_fops, kvm, 0);
A
Al Viro 已提交
2275
	if (fd < 0)
A
Al Viro 已提交
2276
		kvm_put_kvm(kvm);
2277 2278 2279 2280

	return fd;
}

2281 2282 2283
static long kvm_dev_ioctl_check_extension_generic(long arg)
{
	switch (arg) {
2284
	case KVM_CAP_USER_MEMORY:
2285
	case KVM_CAP_DESTROY_MEMORY_REGION_WORKS:
2286
	case KVM_CAP_JOIN_MEMORY_REGIONS_WORKS:
2287
		return 1;
2288 2289
#ifdef CONFIG_HAVE_KVM_IRQCHIP
	case KVM_CAP_IRQ_ROUTING:
2290
		return KVM_MAX_IRQ_ROUTES;
2291
#endif
2292 2293 2294 2295 2296 2297
	default:
		break;
	}
	return kvm_dev_ioctl_check_extension(arg);
}

2298 2299 2300
static long kvm_dev_ioctl(struct file *filp,
			  unsigned int ioctl, unsigned long arg)
{
2301
	long r = -EINVAL;
2302 2303 2304

	switch (ioctl) {
	case KVM_GET_API_VERSION:
2305 2306 2307
		r = -EINVAL;
		if (arg)
			goto out;
2308 2309 2310
		r = KVM_API_VERSION;
		break;
	case KVM_CREATE_VM:
2311 2312 2313
		r = -EINVAL;
		if (arg)
			goto out;
2314 2315
		r = kvm_dev_ioctl_create_vm();
		break;
2316
	case KVM_CHECK_EXTENSION:
2317
		r = kvm_dev_ioctl_check_extension_generic(arg);
2318
		break;
2319 2320 2321 2322
	case KVM_GET_VCPU_MMAP_SIZE:
		r = -EINVAL;
		if (arg)
			goto out;
2323 2324 2325
		r = PAGE_SIZE;     /* struct kvm_run */
#ifdef CONFIG_X86
		r += PAGE_SIZE;    /* pio data page */
2326 2327 2328
#endif
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
		r += PAGE_SIZE;    /* coalesced mmio ring page */
2329
#endif
2330
		break;
2331 2332 2333 2334 2335
	case KVM_TRACE_ENABLE:
	case KVM_TRACE_PAUSE:
	case KVM_TRACE_DISABLE:
		r = kvm_trace_ioctl(ioctl, arg);
		break;
A
Avi Kivity 已提交
2336
	default:
2337
		return kvm_arch_dev_ioctl(filp, ioctl, arg);
A
Avi Kivity 已提交
2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348
	}
out:
	return r;
}

static struct file_operations kvm_chardev_ops = {
	.unlocked_ioctl = kvm_dev_ioctl,
	.compat_ioctl   = kvm_dev_ioctl,
};

static struct miscdevice kvm_dev = {
A
Avi Kivity 已提交
2349
	KVM_MINOR,
A
Avi Kivity 已提交
2350 2351 2352 2353
	"kvm",
	&kvm_chardev_ops,
};

2354 2355 2356 2357
static void hardware_enable(void *junk)
{
	int cpu = raw_smp_processor_id();

2358
	if (cpumask_test_cpu(cpu, cpus_hardware_enabled))
2359
		return;
2360
	cpumask_set_cpu(cpu, cpus_hardware_enabled);
2361
	kvm_arch_hardware_enable(NULL);
2362 2363 2364 2365 2366 2367
}

static void hardware_disable(void *junk)
{
	int cpu = raw_smp_processor_id();

2368
	if (!cpumask_test_cpu(cpu, cpus_hardware_enabled))
2369
		return;
2370
	cpumask_clear_cpu(cpu, cpus_hardware_enabled);
2371
	kvm_arch_hardware_disable(NULL);
2372 2373
}

A
Avi Kivity 已提交
2374 2375 2376 2377 2378
static int kvm_cpu_hotplug(struct notifier_block *notifier, unsigned long val,
			   void *v)
{
	int cpu = (long)v;

2379
	val &= ~CPU_TASKS_FROZEN;
A
Avi Kivity 已提交
2380
	switch (val) {
2381
	case CPU_DYING:
2382 2383 2384 2385
		printk(KERN_INFO "kvm: disabling virtualization on CPU%d\n",
		       cpu);
		hardware_disable(NULL);
		break;
A
Avi Kivity 已提交
2386
	case CPU_UP_CANCELED:
2387 2388
		printk(KERN_INFO "kvm: disabling virtualization on CPU%d\n",
		       cpu);
2389
		smp_call_function_single(cpu, hardware_disable, NULL, 1);
A
Avi Kivity 已提交
2390
		break;
2391 2392 2393
	case CPU_ONLINE:
		printk(KERN_INFO "kvm: enabling virtualization on CPU%d\n",
		       cpu);
2394
		smp_call_function_single(cpu, hardware_enable, NULL, 1);
A
Avi Kivity 已提交
2395 2396 2397 2398 2399
		break;
	}
	return NOTIFY_OK;
}

2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411

asmlinkage void kvm_handle_fault_on_reboot(void)
{
	if (kvm_rebooting)
		/* spin while reset goes on */
		while (true)
			;
	/* Fault while not rebooting.  We want the trace. */
	BUG();
}
EXPORT_SYMBOL_GPL(kvm_handle_fault_on_reboot);

2412
static int kvm_reboot(struct notifier_block *notifier, unsigned long val,
M
Mike Day 已提交
2413
		      void *v)
2414
{
2415 2416 2417 2418 2419 2420 2421 2422 2423
	/*
	 * Some (well, at least mine) BIOSes hang on reboot if
	 * in vmx root mode.
	 *
	 * And Intel TXT required VMX off for all cpu when system shutdown.
	 */
	printk(KERN_INFO "kvm: exiting hardware virtualization\n");
	kvm_rebooting = true;
	on_each_cpu(hardware_disable, NULL, 1);
2424 2425 2426 2427 2428 2429 2430 2431
	return NOTIFY_OK;
}

static struct notifier_block kvm_reboot_notifier = {
	.notifier_call = kvm_reboot,
	.priority = 0,
};

2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447
void kvm_io_bus_init(struct kvm_io_bus *bus)
{
	memset(bus, 0, sizeof(*bus));
}

void kvm_io_bus_destroy(struct kvm_io_bus *bus)
{
	int i;

	for (i = 0; i < bus->dev_count; i++) {
		struct kvm_io_device *pos = bus->devs[i];

		kvm_iodevice_destructor(pos);
	}
}

2448 2449
struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus,
					  gpa_t addr, int len, int is_write)
2450 2451 2452 2453 2454 2455
{
	int i;

	for (i = 0; i < bus->dev_count; i++) {
		struct kvm_io_device *pos = bus->devs[i];

2456
		if (pos->in_range(pos, addr, len, is_write))
2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469
			return pos;
	}

	return NULL;
}

void kvm_io_bus_register_dev(struct kvm_io_bus *bus, struct kvm_io_device *dev)
{
	BUG_ON(bus->dev_count > (NR_IOBUS_DEVS-1));

	bus->devs[bus->dev_count++] = dev;
}

A
Avi Kivity 已提交
2470 2471 2472 2473 2474
static struct notifier_block kvm_cpu_notifier = {
	.notifier_call = kvm_cpu_hotplug,
	.priority = 20, /* must be > scheduler priority */
};

2475
static int vm_stat_get(void *_offset, u64 *val)
2476 2477 2478 2479
{
	unsigned offset = (long)_offset;
	struct kvm *kvm;

2480
	*val = 0;
2481 2482
	spin_lock(&kvm_lock);
	list_for_each_entry(kvm, &vm_list, vm_list)
2483
		*val += *(u32 *)((void *)kvm + offset);
2484
	spin_unlock(&kvm_lock);
2485
	return 0;
2486 2487 2488 2489
}

DEFINE_SIMPLE_ATTRIBUTE(vm_stat_fops, vm_stat_get, NULL, "%llu\n");

2490
static int vcpu_stat_get(void *_offset, u64 *val)
A
Avi Kivity 已提交
2491 2492 2493 2494 2495 2496
{
	unsigned offset = (long)_offset;
	struct kvm *kvm;
	struct kvm_vcpu *vcpu;
	int i;

2497
	*val = 0;
A
Avi Kivity 已提交
2498 2499 2500
	spin_lock(&kvm_lock);
	list_for_each_entry(kvm, &vm_list, vm_list)
		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
R
Rusty Russell 已提交
2501 2502
			vcpu = kvm->vcpus[i];
			if (vcpu)
2503
				*val += *(u32 *)((void *)vcpu + offset);
A
Avi Kivity 已提交
2504 2505
		}
	spin_unlock(&kvm_lock);
2506
	return 0;
A
Avi Kivity 已提交
2507 2508
}

2509 2510 2511 2512 2513 2514
DEFINE_SIMPLE_ATTRIBUTE(vcpu_stat_fops, vcpu_stat_get, NULL, "%llu\n");

static struct file_operations *stat_fops[] = {
	[KVM_STAT_VCPU] = &vcpu_stat_fops,
	[KVM_STAT_VM]   = &vm_stat_fops,
};
A
Avi Kivity 已提交
2515

2516
static void kvm_init_debug(void)
A
Avi Kivity 已提交
2517 2518 2519
{
	struct kvm_stats_debugfs_item *p;

2520
	kvm_debugfs_dir = debugfs_create_dir("kvm", NULL);
A
Avi Kivity 已提交
2521
	for (p = debugfs_entries; p->name; ++p)
2522
		p->dentry = debugfs_create_file(p->name, 0444, kvm_debugfs_dir,
A
Avi Kivity 已提交
2523
						(void *)(long)p->offset,
2524
						stat_fops[p->kind]);
A
Avi Kivity 已提交
2525 2526 2527 2528 2529 2530 2531 2532
}

static void kvm_exit_debug(void)
{
	struct kvm_stats_debugfs_item *p;

	for (p = debugfs_entries; p->name; ++p)
		debugfs_remove(p->dentry);
2533
	debugfs_remove(kvm_debugfs_dir);
A
Avi Kivity 已提交
2534 2535
}

2536 2537
static int kvm_suspend(struct sys_device *dev, pm_message_t state)
{
A
Avi Kivity 已提交
2538
	hardware_disable(NULL);
2539 2540 2541 2542 2543
	return 0;
}

static int kvm_resume(struct sys_device *dev)
{
A
Avi Kivity 已提交
2544
	hardware_enable(NULL);
2545 2546 2547 2548
	return 0;
}

static struct sysdev_class kvm_sysdev_class = {
2549
	.name = "kvm",
2550 2551 2552 2553 2554 2555 2556 2557 2558
	.suspend = kvm_suspend,
	.resume = kvm_resume,
};

static struct sys_device kvm_sysdev = {
	.id = 0,
	.cls = &kvm_sysdev_class,
};

2559
struct page *bad_page;
2560
pfn_t bad_pfn;
A
Avi Kivity 已提交
2561

2562 2563 2564 2565 2566 2567 2568 2569 2570 2571
static inline
struct kvm_vcpu *preempt_notifier_to_vcpu(struct preempt_notifier *pn)
{
	return container_of(pn, struct kvm_vcpu, preempt_notifier);
}

static void kvm_sched_in(struct preempt_notifier *pn, int cpu)
{
	struct kvm_vcpu *vcpu = preempt_notifier_to_vcpu(pn);

2572
	kvm_arch_vcpu_load(vcpu, cpu);
2573 2574 2575 2576 2577 2578 2579
}

static void kvm_sched_out(struct preempt_notifier *pn,
			  struct task_struct *next)
{
	struct kvm_vcpu *vcpu = preempt_notifier_to_vcpu(pn);

2580
	kvm_arch_vcpu_put(vcpu);
2581 2582
}

2583
int kvm_init(void *opaque, unsigned int vcpu_size,
2584
		  struct module *module)
A
Avi Kivity 已提交
2585 2586
{
	int r;
Y
Yang, Sheng 已提交
2587
	int cpu;
A
Avi Kivity 已提交
2588

2589 2590
	kvm_init_debug();

2591 2592
	r = kvm_arch_init(opaque);
	if (r)
2593
		goto out_fail;
2594 2595 2596 2597 2598 2599 2600 2601

	bad_page = alloc_page(GFP_KERNEL | __GFP_ZERO);

	if (bad_page == NULL) {
		r = -ENOMEM;
		goto out;
	}

2602 2603
	bad_pfn = page_to_pfn(bad_page);

2604
	if (!zalloc_cpumask_var(&cpus_hardware_enabled, GFP_KERNEL)) {
2605 2606 2607
		r = -ENOMEM;
		goto out_free_0;
	}
2608
	cpumask_clear(cpus_hardware_enabled);
2609

2610
	r = kvm_arch_hardware_setup();
A
Avi Kivity 已提交
2611
	if (r < 0)
2612
		goto out_free_0a;
A
Avi Kivity 已提交
2613

Y
Yang, Sheng 已提交
2614 2615
	for_each_online_cpu(cpu) {
		smp_call_function_single(cpu,
2616
				kvm_arch_check_processor_compat,
2617
				&r, 1);
Y
Yang, Sheng 已提交
2618
		if (r < 0)
2619
			goto out_free_1;
Y
Yang, Sheng 已提交
2620 2621
	}

2622
	on_each_cpu(hardware_enable, NULL, 1);
A
Avi Kivity 已提交
2623 2624
	r = register_cpu_notifier(&kvm_cpu_notifier);
	if (r)
2625
		goto out_free_2;
A
Avi Kivity 已提交
2626 2627
	register_reboot_notifier(&kvm_reboot_notifier);

2628 2629
	r = sysdev_class_register(&kvm_sysdev_class);
	if (r)
2630
		goto out_free_3;
2631 2632 2633

	r = sysdev_register(&kvm_sysdev);
	if (r)
2634
		goto out_free_4;
2635

2636 2637
	/* A kmem cache lets us meet the alignment requirements of fx_save. */
	kvm_vcpu_cache = kmem_cache_create("kvm_vcpu", vcpu_size,
J
Joe Perches 已提交
2638 2639
					   __alignof__(struct kvm_vcpu),
					   0, NULL);
2640 2641
	if (!kvm_vcpu_cache) {
		r = -ENOMEM;
2642
		goto out_free_5;
2643 2644
	}

A
Avi Kivity 已提交
2645
	kvm_chardev_ops.owner = module;
2646 2647
	kvm_vm_fops.owner = module;
	kvm_vcpu_fops.owner = module;
A
Avi Kivity 已提交
2648 2649 2650

	r = misc_register(&kvm_dev);
	if (r) {
M
Mike Day 已提交
2651
		printk(KERN_ERR "kvm: misc device register failed\n");
A
Avi Kivity 已提交
2652 2653 2654
		goto out_free;
	}

2655 2656 2657
	kvm_preempt_ops.sched_in = kvm_sched_in;
	kvm_preempt_ops.sched_out = kvm_sched_out;

2658
	return 0;
A
Avi Kivity 已提交
2659 2660

out_free:
2661
	kmem_cache_destroy(kvm_vcpu_cache);
2662
out_free_5:
2663
	sysdev_unregister(&kvm_sysdev);
2664
out_free_4:
2665
	sysdev_class_unregister(&kvm_sysdev_class);
2666
out_free_3:
A
Avi Kivity 已提交
2667
	unregister_reboot_notifier(&kvm_reboot_notifier);
A
Avi Kivity 已提交
2668
	unregister_cpu_notifier(&kvm_cpu_notifier);
2669
out_free_2:
2670
	on_each_cpu(hardware_disable, NULL, 1);
2671
out_free_1:
2672
	kvm_arch_hardware_unsetup();
2673 2674
out_free_0a:
	free_cpumask_var(cpus_hardware_enabled);
2675 2676
out_free_0:
	__free_page(bad_page);
2677
out:
2678
	kvm_arch_exit();
2679
	kvm_exit_debug();
2680
out_fail:
A
Avi Kivity 已提交
2681 2682
	return r;
}
2683
EXPORT_SYMBOL_GPL(kvm_init);
A
Avi Kivity 已提交
2684

2685
void kvm_exit(void)
A
Avi Kivity 已提交
2686
{
2687
	kvm_trace_cleanup();
A
Avi Kivity 已提交
2688
	misc_deregister(&kvm_dev);
2689
	kmem_cache_destroy(kvm_vcpu_cache);
2690 2691
	sysdev_unregister(&kvm_sysdev);
	sysdev_class_unregister(&kvm_sysdev_class);
A
Avi Kivity 已提交
2692
	unregister_reboot_notifier(&kvm_reboot_notifier);
2693
	unregister_cpu_notifier(&kvm_cpu_notifier);
2694
	on_each_cpu(hardware_disable, NULL, 1);
2695
	kvm_arch_hardware_unsetup();
2696
	kvm_arch_exit();
A
Avi Kivity 已提交
2697
	kvm_exit_debug();
2698
	free_cpumask_var(cpus_hardware_enabled);
2699
	__free_page(bad_page);
A
Avi Kivity 已提交
2700
}
2701
EXPORT_SYMBOL_GPL(kvm_exit);