kvm_main.c 69.3 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 18 19 20 21 22
/*
 * 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.
 *
 */

#include "kvm.h"

#include <linux/kvm.h>
#include <linux/module.h>
#include <linux/errno.h>
23
#include <linux/magic.h>
A
Avi Kivity 已提交
24 25 26 27 28 29 30 31 32 33 34 35 36 37
#include <asm/processor.h>
#include <linux/percpu.h>
#include <linux/gfp.h>
#include <asm/msr.h>
#include <linux/mm.h>
#include <linux/miscdevice.h>
#include <linux/vmalloc.h>
#include <asm/uaccess.h>
#include <linux/reboot.h>
#include <asm/io.h>
#include <linux/debugfs.h>
#include <linux/highmem.h>
#include <linux/file.h>
#include <asm/desc.h>
38
#include <linux/sysdev.h>
A
Avi Kivity 已提交
39
#include <linux/cpu.h>
40
#include <linux/file.h>
41 42
#include <linux/fs.h>
#include <linux/mount.h>
A
Alexey Dobriyan 已提交
43
#include <linux/sched.h>
A
Avi Kivity 已提交
44 45 46 47 48 49 50

#include "x86_emulate.h"
#include "segment_descriptor.h"

MODULE_AUTHOR("Qumranet");
MODULE_LICENSE("GPL");

51 52 53
static DEFINE_SPINLOCK(kvm_lock);
static LIST_HEAD(vm_list);

A
Avi Kivity 已提交
54
struct kvm_arch_ops *kvm_arch_ops;
A
Avi Kivity 已提交
55 56

#define STAT_OFFSET(x) offsetof(struct kvm_vcpu, stat.x)
A
Avi Kivity 已提交
57 58 59

static struct kvm_stats_debugfs_item {
	const char *name;
A
Avi Kivity 已提交
60
	int offset;
A
Avi Kivity 已提交
61 62
	struct dentry *dentry;
} debugfs_entries[] = {
A
Avi Kivity 已提交
63 64 65 66 67 68 69 70 71 72 73 74
	{ "pf_fixed", STAT_OFFSET(pf_fixed) },
	{ "pf_guest", STAT_OFFSET(pf_guest) },
	{ "tlb_flush", STAT_OFFSET(tlb_flush) },
	{ "invlpg", STAT_OFFSET(invlpg) },
	{ "exits", STAT_OFFSET(exits) },
	{ "io_exits", STAT_OFFSET(io_exits) },
	{ "mmio_exits", STAT_OFFSET(mmio_exits) },
	{ "signal_exits", STAT_OFFSET(signal_exits) },
	{ "irq_window", STAT_OFFSET(irq_window_exits) },
	{ "halt_exits", STAT_OFFSET(halt_exits) },
	{ "request_irq", STAT_OFFSET(request_irq_exits) },
	{ "irq_exits", STAT_OFFSET(irq_exits) },
75
	{ "light_exits", STAT_OFFSET(light_exits) },
76
	{ "efer_reload", STAT_OFFSET(efer_reload) },
A
Avi Kivity 已提交
77
	{ NULL }
A
Avi Kivity 已提交
78 79 80 81
};

static struct dentry *debugfs_dir;

82 83
struct vfsmount *kvmfs_mnt;

A
Avi Kivity 已提交
84 85 86 87 88 89 90 91
#define MAX_IO_MSRS 256

#define CR0_RESEVED_BITS 0xffffffff1ffaffc0ULL
#define LMSW_GUEST_MASK 0x0eULL
#define CR4_RESEVED_BITS (~((1ULL << 11) - 1))
#define CR8_RESEVED_BITS (~0x0fULL)
#define EFER_RESERVED_BITS 0xfffffffffffff2fe

92
#ifdef CONFIG_X86_64
A
Avi Kivity 已提交
93 94 95 96 97 98 99 100 101
// LDT or TSS descriptor in the GDT. 16 bytes.
struct segment_descriptor_64 {
	struct segment_descriptor s;
	u32 base_higher;
	u32 pad_zero;
};

#endif

A
Avi Kivity 已提交
102 103 104
static long kvm_vcpu_ioctl(struct file *file, unsigned int ioctl,
			   unsigned long arg);

105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153
static struct inode *kvmfs_inode(struct file_operations *fops)
{
	int error = -ENOMEM;
	struct inode *inode = new_inode(kvmfs_mnt->mnt_sb);

	if (!inode)
		goto eexit_1;

	inode->i_fop = fops;

	/*
	 * Mark the inode dirty from the very beginning,
	 * that way it will never be moved to the dirty
	 * list because mark_inode_dirty() will think
	 * that it already _is_ on the dirty list.
	 */
	inode->i_state = I_DIRTY;
	inode->i_mode = S_IRUSR | S_IWUSR;
	inode->i_uid = current->fsuid;
	inode->i_gid = current->fsgid;
	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	return inode;

eexit_1:
	return ERR_PTR(error);
}

static struct file *kvmfs_file(struct inode *inode, void *private_data)
{
	struct file *file = get_empty_filp();

	if (!file)
		return ERR_PTR(-ENFILE);

	file->f_path.mnt = mntget(kvmfs_mnt);
	file->f_path.dentry = d_alloc_anon(inode);
	if (!file->f_path.dentry)
		return ERR_PTR(-ENOMEM);
	file->f_mapping = inode->i_mapping;

	file->f_pos = 0;
	file->f_flags = O_RDWR;
	file->f_op = inode->i_fop;
	file->f_mode = FMODE_READ | FMODE_WRITE;
	file->f_version = 0;
	file->private_data = private_data;
	return file;
}

A
Avi Kivity 已提交
154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175
unsigned long segment_base(u16 selector)
{
	struct descriptor_table gdt;
	struct segment_descriptor *d;
	unsigned long table_base;
	typedef unsigned long ul;
	unsigned long v;

	if (selector == 0)
		return 0;

	asm ("sgdt %0" : "=m"(gdt));
	table_base = gdt.base;

	if (selector & 4) {           /* from ldt */
		u16 ldt_selector;

		asm ("sldt %0" : "=g"(ldt_selector));
		table_base = segment_base(ldt_selector);
	}
	d = (struct segment_descriptor *)(table_base + (selector & ~7));
	v = d->base_low | ((ul)d->base_mid << 16) | ((ul)d->base_high << 24);
176
#ifdef CONFIG_X86_64
A
Avi Kivity 已提交
177 178 179 180 181 182 183 184
	if (d->system == 0
	    && (d->type == 2 || d->type == 9 || d->type == 11))
		v |= ((ul)((struct segment_descriptor_64 *)d)->base_higher) << 32;
#endif
	return v;
}
EXPORT_SYMBOL_GPL(segment_base);

185 186 187 188 189
static inline int valid_vcpu(int n)
{
	return likely(n >= 0 && n < KVM_MAX_VCPUS);
}

A
Avi Kivity 已提交
190 191
int kvm_read_guest(struct kvm_vcpu *vcpu, gva_t addr, unsigned long size,
		   void *dest)
A
Avi Kivity 已提交
192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222
{
	unsigned char *host_buf = dest;
	unsigned long req_size = size;

	while (size) {
		hpa_t paddr;
		unsigned now;
		unsigned offset;
		hva_t guest_buf;

		paddr = gva_to_hpa(vcpu, addr);

		if (is_error_hpa(paddr))
			break;

		guest_buf = (hva_t)kmap_atomic(
					pfn_to_page(paddr >> PAGE_SHIFT),
					KM_USER0);
		offset = addr & ~PAGE_MASK;
		guest_buf |= offset;
		now = min(size, PAGE_SIZE - offset);
		memcpy(host_buf, (void*)guest_buf, now);
		host_buf += now;
		addr += now;
		size -= now;
		kunmap_atomic((void *)(guest_buf & PAGE_MASK), KM_USER0);
	}
	return req_size - size;
}
EXPORT_SYMBOL_GPL(kvm_read_guest);

A
Avi Kivity 已提交
223 224
int kvm_write_guest(struct kvm_vcpu *vcpu, gva_t addr, unsigned long size,
		    void *data)
A
Avi Kivity 已提交
225 226 227 228 229 230 231 232 233
{
	unsigned char *host_buf = data;
	unsigned long req_size = size;

	while (size) {
		hpa_t paddr;
		unsigned now;
		unsigned offset;
		hva_t guest_buf;
234
		gfn_t gfn;
A
Avi Kivity 已提交
235 236 237 238 239 240

		paddr = gva_to_hpa(vcpu, addr);

		if (is_error_hpa(paddr))
			break;

241 242
		gfn = vcpu->mmu.gva_to_gpa(vcpu, addr) >> PAGE_SHIFT;
		mark_page_dirty(vcpu->kvm, gfn);
A
Avi Kivity 已提交
243 244 245 246 247 248 249 250 251 252 253 254 255 256 257
		guest_buf = (hva_t)kmap_atomic(
				pfn_to_page(paddr >> PAGE_SHIFT), KM_USER0);
		offset = addr & ~PAGE_MASK;
		guest_buf |= offset;
		now = min(size, PAGE_SIZE - offset);
		memcpy((void*)guest_buf, host_buf, now);
		host_buf += now;
		addr += now;
		size -= now;
		kunmap_atomic((void *)(guest_buf & PAGE_MASK), KM_USER0);
	}
	return req_size - size;
}
EXPORT_SYMBOL_GPL(kvm_write_guest);

258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279
void kvm_load_guest_fpu(struct kvm_vcpu *vcpu)
{
	if (!vcpu->fpu_active || vcpu->guest_fpu_loaded)
		return;

	vcpu->guest_fpu_loaded = 1;
	fx_save(vcpu->host_fx_image);
	fx_restore(vcpu->guest_fx_image);
}
EXPORT_SYMBOL_GPL(kvm_load_guest_fpu);

void kvm_put_guest_fpu(struct kvm_vcpu *vcpu)
{
	if (!vcpu->guest_fpu_loaded)
		return;

	vcpu->guest_fpu_loaded = 0;
	fx_save(vcpu->guest_fx_image);
	fx_restore(vcpu->host_fx_image);
}
EXPORT_SYMBOL_GPL(kvm_put_guest_fpu);

A
Avi Kivity 已提交
280 281 282 283
/*
 * Switches to specified vcpu, until a matching vcpu_put()
 */
static void vcpu_load(struct kvm_vcpu *vcpu)
A
Avi Kivity 已提交
284
{
A
Avi Kivity 已提交
285 286
	mutex_lock(&vcpu->mutex);
	kvm_arch_ops->vcpu_load(vcpu);
A
Avi Kivity 已提交
287 288 289
}

/*
A
Avi Kivity 已提交
290 291
 * Switches to specified vcpu, until a matching vcpu_put(). Will return NULL
 * if the slot is not populated.
A
Avi Kivity 已提交
292
 */
A
Avi Kivity 已提交
293
static struct kvm_vcpu *vcpu_load_slot(struct kvm *kvm, int slot)
A
Avi Kivity 已提交
294
{
A
Avi Kivity 已提交
295
	struct kvm_vcpu *vcpu = &kvm->vcpus[slot];
A
Avi Kivity 已提交
296 297

	mutex_lock(&vcpu->mutex);
A
Avi Kivity 已提交
298
	if (!vcpu->vmcs) {
A
Avi Kivity 已提交
299
		mutex_unlock(&vcpu->mutex);
A
Al Viro 已提交
300
		return NULL;
A
Avi Kivity 已提交
301
	}
A
Avi Kivity 已提交
302 303
	kvm_arch_ops->vcpu_load(vcpu);
	return vcpu;
A
Avi Kivity 已提交
304 305 306 307 308 309 310 311
}

static void vcpu_put(struct kvm_vcpu *vcpu)
{
	kvm_arch_ops->vcpu_put(vcpu);
	mutex_unlock(&vcpu->mutex);
}

312
static struct kvm *kvm_create_vm(void)
A
Avi Kivity 已提交
313 314 315 316 317
{
	struct kvm *kvm = kzalloc(sizeof(struct kvm), GFP_KERNEL);
	int i;

	if (!kvm)
318
		return ERR_PTR(-ENOMEM);
A
Avi Kivity 已提交
319 320 321

	spin_lock_init(&kvm->lock);
	INIT_LIST_HEAD(&kvm->active_mmu_pages);
322 323 324
	spin_lock(&kvm_lock);
	list_add(&kvm->vm_list, &vm_list);
	spin_unlock(&kvm_lock);
A
Avi Kivity 已提交
325 326 327 328
	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
		struct kvm_vcpu *vcpu = &kvm->vcpus[i];

		mutex_init(&vcpu->mutex);
329
		vcpu->cpu = -1;
330
		vcpu->kvm = kvm;
A
Avi Kivity 已提交
331 332
		vcpu->mmu.root_hpa = INVALID_PAGE;
	}
333 334 335 336 337
	return kvm;
}

static int kvm_dev_open(struct inode *inode, struct file *filp)
{
A
Avi Kivity 已提交
338 339 340 341 342 343 344 345 346 347 348 349 350 351
	return 0;
}

/*
 * 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)
{
	int i;

	if (!dont || free->phys_mem != dont->phys_mem)
		if (free->phys_mem) {
			for (i = 0; i < free->npages; ++i)
A
Avi Kivity 已提交
352 353
				if (free->phys_mem[i])
					__free_page(free->phys_mem[i]);
A
Avi Kivity 已提交
354 355 356 357 358 359
			vfree(free->phys_mem);
		}

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

A
Al Viro 已提交
360
	free->phys_mem = NULL;
A
Avi Kivity 已提交
361
	free->npages = 0;
A
Al Viro 已提交
362
	free->dirty_bitmap = NULL;
A
Avi Kivity 已提交
363 364 365 366 367 368 369
}

static void kvm_free_physmem(struct kvm *kvm)
{
	int i;

	for (i = 0; i < kvm->nmemslots; ++i)
A
Al Viro 已提交
370
		kvm_free_physmem_slot(&kvm->memslots[i], NULL);
A
Avi Kivity 已提交
371 372
}

373 374 375 376 377 378 379 380 381 382 383
static void free_pio_guest_pages(struct kvm_vcpu *vcpu)
{
	int i;

	for (i = 0; i < 2; ++i)
		if (vcpu->pio.guest_pages[i]) {
			__free_page(vcpu->pio.guest_pages[i]);
			vcpu->pio.guest_pages[i] = NULL;
		}
}

A
Avi Kivity 已提交
384 385 386 387 388 389 390 391 392 393
static void kvm_unload_vcpu_mmu(struct kvm_vcpu *vcpu)
{
	if (!vcpu->vmcs)
		return;

	vcpu_load(vcpu);
	kvm_mmu_unload(vcpu);
	vcpu_put(vcpu);
}

A
Avi Kivity 已提交
394 395
static void kvm_free_vcpu(struct kvm_vcpu *vcpu)
{
A
Avi Kivity 已提交
396
	if (!vcpu->vmcs)
I
Ingo Molnar 已提交
397 398
		return;

A
Avi Kivity 已提交
399
	vcpu_load(vcpu);
A
Avi Kivity 已提交
400
	kvm_mmu_destroy(vcpu);
401
	vcpu_put(vcpu);
402
	kvm_arch_ops->vcpu_free(vcpu);
403 404
	free_page((unsigned long)vcpu->run);
	vcpu->run = NULL;
405 406 407
	free_page((unsigned long)vcpu->pio_data);
	vcpu->pio_data = NULL;
	free_pio_guest_pages(vcpu);
A
Avi Kivity 已提交
408 409 410 411 412 413
}

static void kvm_free_vcpus(struct kvm *kvm)
{
	unsigned int i;

A
Avi Kivity 已提交
414 415 416 417 418
	/*
	 * Unpin any mmu pages first.
	 */
	for (i = 0; i < KVM_MAX_VCPUS; ++i)
		kvm_unload_vcpu_mmu(&kvm->vcpus[i]);
A
Avi Kivity 已提交
419 420 421 422 423 424
	for (i = 0; i < KVM_MAX_VCPUS; ++i)
		kvm_free_vcpu(&kvm->vcpus[i]);
}

static int kvm_dev_release(struct inode *inode, struct file *filp)
{
425 426
	return 0;
}
A
Avi Kivity 已提交
427

428 429
static void kvm_destroy_vm(struct kvm *kvm)
{
430 431 432
	spin_lock(&kvm_lock);
	list_del(&kvm->vm_list);
	spin_unlock(&kvm_lock);
A
Avi Kivity 已提交
433 434 435
	kvm_free_vcpus(kvm);
	kvm_free_physmem(kvm);
	kfree(kvm);
436 437 438 439 440 441 442
}

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

	kvm_destroy_vm(kvm);
A
Avi Kivity 已提交
443 444 445 446 447 448 449 450
	return 0;
}

static void inject_gp(struct kvm_vcpu *vcpu)
{
	kvm_arch_ops->inject_gp(vcpu, 0);
}

451 452 453 454
/*
 * Load the pae pdptrs.  Return true is they are all valid.
 */
static int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3)
A
Avi Kivity 已提交
455 456
{
	gfn_t pdpt_gfn = cr3 >> PAGE_SHIFT;
457
	unsigned offset = ((cr3 & (PAGE_SIZE-1)) >> 5) << 2;
A
Avi Kivity 已提交
458 459 460
	int i;
	u64 pdpte;
	u64 *pdpt;
461
	int ret;
A
Avi Kivity 已提交
462
	struct page *page;
A
Avi Kivity 已提交
463 464

	spin_lock(&vcpu->kvm->lock);
A
Avi Kivity 已提交
465 466 467
	page = gfn_to_page(vcpu->kvm, pdpt_gfn);
	/* FIXME: !page - emulate? 0xff? */
	pdpt = kmap_atomic(page, KM_USER0);
A
Avi Kivity 已提交
468

469
	ret = 1;
A
Avi Kivity 已提交
470 471
	for (i = 0; i < 4; ++i) {
		pdpte = pdpt[offset + i];
472 473 474 475
		if ((pdpte & 1) && (pdpte & 0xfffffff0000001e6ull)) {
			ret = 0;
			goto out;
		}
A
Avi Kivity 已提交
476 477
	}

478 479 480 481
	for (i = 0; i < 4; ++i)
		vcpu->pdptrs[i] = pdpt[offset + i];

out:
A
Avi Kivity 已提交
482 483 484
	kunmap_atomic(pdpt, KM_USER0);
	spin_unlock(&vcpu->kvm->lock);

485
	return ret;
A
Avi Kivity 已提交
486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510
}

void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
{
	if (cr0 & CR0_RESEVED_BITS) {
		printk(KERN_DEBUG "set_cr0: 0x%lx #GP, reserved bits 0x%lx\n",
		       cr0, vcpu->cr0);
		inject_gp(vcpu);
		return;
	}

	if ((cr0 & CR0_NW_MASK) && !(cr0 & CR0_CD_MASK)) {
		printk(KERN_DEBUG "set_cr0: #GP, CD == 0 && NW == 1\n");
		inject_gp(vcpu);
		return;
	}

	if ((cr0 & CR0_PG_MASK) && !(cr0 & CR0_PE_MASK)) {
		printk(KERN_DEBUG "set_cr0: #GP, set PG flag "
		       "and a clear PE flag\n");
		inject_gp(vcpu);
		return;
	}

	if (!is_paging(vcpu) && (cr0 & CR0_PG_MASK)) {
511
#ifdef CONFIG_X86_64
A
Avi Kivity 已提交
512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530
		if ((vcpu->shadow_efer & EFER_LME)) {
			int cs_db, cs_l;

			if (!is_pae(vcpu)) {
				printk(KERN_DEBUG "set_cr0: #GP, start paging "
				       "in long mode while PAE is disabled\n");
				inject_gp(vcpu);
				return;
			}
			kvm_arch_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
			if (cs_l) {
				printk(KERN_DEBUG "set_cr0: #GP, start paging "
				       "in long mode while CS.L == 1\n");
				inject_gp(vcpu);
				return;

			}
		} else
#endif
531
		if (is_pae(vcpu) && !load_pdptrs(vcpu, vcpu->cr3)) {
A
Avi Kivity 已提交
532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563
			printk(KERN_DEBUG "set_cr0: #GP, pdptrs "
			       "reserved bits\n");
			inject_gp(vcpu);
			return;
		}

	}

	kvm_arch_ops->set_cr0(vcpu, cr0);
	vcpu->cr0 = cr0;

	spin_lock(&vcpu->kvm->lock);
	kvm_mmu_reset_context(vcpu);
	spin_unlock(&vcpu->kvm->lock);
	return;
}
EXPORT_SYMBOL_GPL(set_cr0);

void lmsw(struct kvm_vcpu *vcpu, unsigned long msw)
{
	set_cr0(vcpu, (vcpu->cr0 & ~0x0ful) | (msw & 0x0f));
}
EXPORT_SYMBOL_GPL(lmsw);

void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
{
	if (cr4 & CR4_RESEVED_BITS) {
		printk(KERN_DEBUG "set_cr4: #GP, reserved bits\n");
		inject_gp(vcpu);
		return;
	}

A
Avi Kivity 已提交
564
	if (is_long_mode(vcpu)) {
A
Avi Kivity 已提交
565 566 567 568 569 570 571
		if (!(cr4 & CR4_PAE_MASK)) {
			printk(KERN_DEBUG "set_cr4: #GP, clearing PAE while "
			       "in long mode\n");
			inject_gp(vcpu);
			return;
		}
	} else if (is_paging(vcpu) && !is_pae(vcpu) && (cr4 & CR4_PAE_MASK)
572
		   && !load_pdptrs(vcpu, vcpu->cr3)) {
A
Avi Kivity 已提交
573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590
		printk(KERN_DEBUG "set_cr4: #GP, pdptrs reserved bits\n");
		inject_gp(vcpu);
	}

	if (cr4 & CR4_VMXE_MASK) {
		printk(KERN_DEBUG "set_cr4: #GP, setting VMXE\n");
		inject_gp(vcpu);
		return;
	}
	kvm_arch_ops->set_cr4(vcpu, cr4);
	spin_lock(&vcpu->kvm->lock);
	kvm_mmu_reset_context(vcpu);
	spin_unlock(&vcpu->kvm->lock);
}
EXPORT_SYMBOL_GPL(set_cr4);

void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
{
A
Avi Kivity 已提交
591
	if (is_long_mode(vcpu)) {
A
Avi Kivity 已提交
592
		if (cr3 & CR3_L_MODE_RESEVED_BITS) {
A
Avi Kivity 已提交
593 594 595 596 597 598 599 600 601 602 603
			printk(KERN_DEBUG "set_cr3: #GP, reserved bits\n");
			inject_gp(vcpu);
			return;
		}
	} else {
		if (cr3 & CR3_RESEVED_BITS) {
			printk(KERN_DEBUG "set_cr3: #GP, reserved bits\n");
			inject_gp(vcpu);
			return;
		}
		if (is_paging(vcpu) && is_pae(vcpu) &&
604
		    !load_pdptrs(vcpu, cr3)) {
A
Avi Kivity 已提交
605 606 607 608 609 610 611 612 613
			printk(KERN_DEBUG "set_cr3: #GP, pdptrs "
			       "reserved bits\n");
			inject_gp(vcpu);
			return;
		}
	}

	vcpu->cr3 = cr3;
	spin_lock(&vcpu->kvm->lock);
614 615 616 617 618 619 620 621 622 623 624 625 626
	/*
	 * Does the new cr3 value map to physical memory? (Note, we
	 * catch an invalid cr3 even in real-mode, because it would
	 * cause trouble later on when we turn on paging anyway.)
	 *
	 * A real CPU would silently accept an invalid cr3 and would
	 * attempt to use it - with largely undefined (and often hard
	 * to debug) behavior on the guest side.
	 */
	if (unlikely(!gfn_to_memslot(vcpu->kvm, cr3 >> PAGE_SHIFT)))
		inject_gp(vcpu);
	else
		vcpu->mmu.new_cr3(vcpu);
A
Avi Kivity 已提交
627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667
	spin_unlock(&vcpu->kvm->lock);
}
EXPORT_SYMBOL_GPL(set_cr3);

void set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8)
{
	if ( cr8 & CR8_RESEVED_BITS) {
		printk(KERN_DEBUG "set_cr8: #GP, reserved bits 0x%lx\n", cr8);
		inject_gp(vcpu);
		return;
	}
	vcpu->cr8 = cr8;
}
EXPORT_SYMBOL_GPL(set_cr8);

void fx_init(struct kvm_vcpu *vcpu)
{
	struct __attribute__ ((__packed__)) fx_image_s {
		u16 control; //fcw
		u16 status; //fsw
		u16 tag; // ftw
		u16 opcode; //fop
		u64 ip; // fpu ip
		u64 operand;// fpu dp
		u32 mxcsr;
		u32 mxcsr_mask;

	} *fx_image;

	fx_save(vcpu->host_fx_image);
	fpu_init();
	fx_save(vcpu->guest_fx_image);
	fx_restore(vcpu->host_fx_image);

	fx_image = (struct fx_image_s *)vcpu->guest_fx_image;
	fx_image->mxcsr = 0x1f80;
	memset(vcpu->guest_fx_image + sizeof(struct fx_image_s),
	       0, FX_IMAGE_SIZE - sizeof(struct fx_image_s));
}
EXPORT_SYMBOL_GPL(fx_init);

668 669 670 671 672 673 674
static void do_remove_write_access(struct kvm_vcpu *vcpu, int slot)
{
	spin_lock(&vcpu->kvm->lock);
	kvm_mmu_slot_remove_write_access(vcpu, slot);
	spin_unlock(&vcpu->kvm->lock);
}

A
Avi Kivity 已提交
675 676 677 678 679 680
/*
 * Allocate some memory and give it an address in the guest physical address
 * space.
 *
 * Discontiguous memory is allowed, mostly for framebuffers.
 */
681 682
static int kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
					  struct kvm_memory_region *mem)
A
Avi Kivity 已提交
683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743
{
	int r;
	gfn_t base_gfn;
	unsigned long npages;
	unsigned long i;
	struct kvm_memory_slot *memslot;
	struct kvm_memory_slot old, new;
	int memory_config_version;

	r = -EINVAL;
	/* General sanity checks */
	if (mem->memory_size & (PAGE_SIZE - 1))
		goto out;
	if (mem->guest_phys_addr & (PAGE_SIZE - 1))
		goto out;
	if (mem->slot >= KVM_MEMORY_SLOTS)
		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;

raced:
	spin_lock(&kvm->lock);

	memory_config_version = kvm->memory_config_version;
	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)
		goto out_unlock;

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

		if (s == memslot)
			continue;
		if (!((base_gfn + npages <= s->base_gfn) ||
		      (base_gfn >= s->base_gfn + s->npages)))
			goto out_unlock;
	}
	/*
	 * Do memory allocations outside lock.  memory_config_version will
	 * detect any races.
	 */
	spin_unlock(&kvm->lock);

	/* Deallocate if slot is being removed */
	if (!npages)
A
Al Viro 已提交
744
		new.phys_mem = NULL;
A
Avi Kivity 已提交
745 746 747

	/* Free page dirty bitmap if unneeded */
	if (!(new.flags & KVM_MEM_LOG_DIRTY_PAGES))
A
Al Viro 已提交
748
		new.dirty_bitmap = NULL;
A
Avi Kivity 已提交
749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764

	r = -ENOMEM;

	/* Allocate if a slot is being created */
	if (npages && !new.phys_mem) {
		new.phys_mem = vmalloc(npages * sizeof(struct page *));

		if (!new.phys_mem)
			goto out_free;

		memset(new.phys_mem, 0, npages * sizeof(struct page *));
		for (i = 0; i < npages; ++i) {
			new.phys_mem[i] = alloc_page(GFP_HIGHUSER
						     | __GFP_ZERO);
			if (!new.phys_mem[i])
				goto out_free;
765
			set_page_private(new.phys_mem[i],0);
A
Avi Kivity 已提交
766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801
		}
	}

	/* 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)
			goto out_free;
		memset(new.dirty_bitmap, 0, dirty_bytes);
	}

	spin_lock(&kvm->lock);

	if (memory_config_version != kvm->memory_config_version) {
		spin_unlock(&kvm->lock);
		kvm_free_physmem_slot(&new, &old);
		goto raced;
	}

	r = -EAGAIN;
	if (kvm->busy)
		goto out_unlock;

	if (mem->slot >= kvm->nmemslots)
		kvm->nmemslots = mem->slot + 1;

	*memslot = new;
	++kvm->memory_config_version;

	spin_unlock(&kvm->lock);

	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
		struct kvm_vcpu *vcpu;

A
Avi Kivity 已提交
802
		vcpu = vcpu_load_slot(kvm, i);
A
Avi Kivity 已提交
803 804
		if (!vcpu)
			continue;
805 806
		if (new.flags & KVM_MEM_LOG_DIRTY_PAGES)
			do_remove_write_access(vcpu, mem->slot);
A
Avi Kivity 已提交
807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824
		kvm_mmu_reset_context(vcpu);
		vcpu_put(vcpu);
	}

	kvm_free_physmem_slot(&old, &new);
	return 0;

out_unlock:
	spin_unlock(&kvm->lock);
out_free:
	kvm_free_physmem_slot(&new, &old);
out:
	return r;
}

/*
 * Get (and clear) the dirty memory log for a memory slot.
 */
825 826
static int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
				      struct kvm_dirty_log *log)
A
Avi Kivity 已提交
827 828 829 830
{
	struct kvm_memory_slot *memslot;
	int r, i;
	int n;
831
	int cleared;
A
Avi Kivity 已提交
832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850
	unsigned long any = 0;

	spin_lock(&kvm->lock);

	/*
	 * Prevent changes to guest memory configuration even while the lock
	 * is not taken.
	 */
	++kvm->busy;
	spin_unlock(&kvm->lock);
	r = -EINVAL;
	if (log->slot >= KVM_MEMORY_SLOTS)
		goto out;

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

851
	n = ALIGN(memslot->npages, BITS_PER_LONG) / 8;
A
Avi Kivity 已提交
852

853
	for (i = 0; !any && i < n/sizeof(long); ++i)
A
Avi Kivity 已提交
854 855 856 857 858 859 860
		any = memslot->dirty_bitmap[i];

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

	if (any) {
861
		cleared = 0;
A
Avi Kivity 已提交
862
		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
A
Avi Kivity 已提交
863
			struct kvm_vcpu *vcpu;
A
Avi Kivity 已提交
864

A
Avi Kivity 已提交
865
			vcpu = vcpu_load_slot(kvm, i);
A
Avi Kivity 已提交
866 867
			if (!vcpu)
				continue;
868 869 870 871 872
			if (!cleared) {
				do_remove_write_access(vcpu, log->slot);
				memset(memslot->dirty_bitmap, 0, n);
				cleared = 1;
			}
A
Avi Kivity 已提交
873 874 875 876 877 878 879 880 881 882 883 884 885 886
			kvm_arch_ops->tlb_flush(vcpu);
			vcpu_put(vcpu);
		}
	}

	r = 0;

out:
	spin_lock(&kvm->lock);
	--kvm->busy;
	spin_unlock(&kvm->lock);
	return r;
}

887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953
/*
 * Set a new alias region.  Aliases map a portion of physical memory into
 * another portion.  This is useful for memory windows, for example the PC
 * VGA region.
 */
static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm,
					 struct kvm_memory_alias *alias)
{
	int r, n;
	struct kvm_mem_alias *p;

	r = -EINVAL;
	/* General sanity checks */
	if (alias->memory_size & (PAGE_SIZE - 1))
		goto out;
	if (alias->guest_phys_addr & (PAGE_SIZE - 1))
		goto out;
	if (alias->slot >= KVM_ALIAS_SLOTS)
		goto out;
	if (alias->guest_phys_addr + alias->memory_size
	    < alias->guest_phys_addr)
		goto out;
	if (alias->target_phys_addr + alias->memory_size
	    < alias->target_phys_addr)
		goto out;

	spin_lock(&kvm->lock);

	p = &kvm->aliases[alias->slot];
	p->base_gfn = alias->guest_phys_addr >> PAGE_SHIFT;
	p->npages = alias->memory_size >> PAGE_SHIFT;
	p->target_gfn = alias->target_phys_addr >> PAGE_SHIFT;

	for (n = KVM_ALIAS_SLOTS; n > 0; --n)
		if (kvm->aliases[n - 1].npages)
			break;
	kvm->naliases = n;

	spin_unlock(&kvm->lock);

	vcpu_load(&kvm->vcpus[0]);
	spin_lock(&kvm->lock);
	kvm_mmu_zap_all(&kvm->vcpus[0]);
	spin_unlock(&kvm->lock);
	vcpu_put(&kvm->vcpus[0]);

	return 0;

out:
	return r;
}

static gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn)
{
	int i;
	struct kvm_mem_alias *alias;

	for (i = 0; i < kvm->naliases; ++i) {
		alias = &kvm->aliases[i];
		if (gfn >= alias->base_gfn
		    && gfn < alias->base_gfn + alias->npages)
			return alias->target_gfn + gfn - alias->base_gfn;
	}
	return gfn;
}

static struct kvm_memory_slot *__gfn_to_memslot(struct kvm *kvm, gfn_t gfn)
A
Avi Kivity 已提交
954 955 956 957 958 959 960 961 962 963
{
	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
964
	return NULL;
A
Avi Kivity 已提交
965
}
966 967 968 969 970 971

struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn)
{
	gfn = unalias_gfn(kvm, gfn);
	return __gfn_to_memslot(kvm, gfn);
}
A
Avi Kivity 已提交
972

A
Avi Kivity 已提交
973 974 975 976
struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn)
{
	struct kvm_memory_slot *slot;

977 978
	gfn = unalias_gfn(kvm, gfn);
	slot = __gfn_to_memslot(kvm, gfn);
A
Avi Kivity 已提交
979 980 981 982 983 984
	if (!slot)
		return NULL;
	return slot->phys_mem[gfn - slot->base_gfn];
}
EXPORT_SYMBOL_GPL(gfn_to_page);

A
Avi Kivity 已提交
985 986 987
void mark_page_dirty(struct kvm *kvm, gfn_t gfn)
{
	int i;
988
	struct kvm_memory_slot *memslot;
A
Avi Kivity 已提交
989 990 991 992 993 994 995 996
	unsigned long rel_gfn;

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

		if (gfn >= memslot->base_gfn
		    && gfn < memslot->base_gfn + memslot->npages) {

997
			if (!memslot->dirty_bitmap)
A
Avi Kivity 已提交
998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010
				return;

			rel_gfn = gfn - memslot->base_gfn;

			/* avoid RMW */
			if (!test_bit(rel_gfn, memslot->dirty_bitmap))
				set_bit(rel_gfn, memslot->dirty_bitmap);
			return;
		}
	}
}

static int emulator_read_std(unsigned long addr,
1011
			     void *val,
A
Avi Kivity 已提交
1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022
			     unsigned int bytes,
			     struct x86_emulate_ctxt *ctxt)
{
	struct kvm_vcpu *vcpu = ctxt->vcpu;
	void *data = val;

	while (bytes) {
		gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);
		unsigned offset = addr & (PAGE_SIZE-1);
		unsigned tocopy = min(bytes, (unsigned)PAGE_SIZE - offset);
		unsigned long pfn;
A
Avi Kivity 已提交
1023 1024
		struct page *page;
		void *page_virt;
A
Avi Kivity 已提交
1025 1026 1027 1028

		if (gpa == UNMAPPED_GVA)
			return X86EMUL_PROPAGATE_FAULT;
		pfn = gpa >> PAGE_SHIFT;
A
Avi Kivity 已提交
1029 1030
		page = gfn_to_page(vcpu->kvm, pfn);
		if (!page)
A
Avi Kivity 已提交
1031
			return X86EMUL_UNHANDLEABLE;
A
Avi Kivity 已提交
1032
		page_virt = kmap_atomic(page, KM_USER0);
A
Avi Kivity 已提交
1033

A
Avi Kivity 已提交
1034
		memcpy(data, page_virt + offset, tocopy);
A
Avi Kivity 已提交
1035

A
Avi Kivity 已提交
1036
		kunmap_atomic(page_virt, KM_USER0);
A
Avi Kivity 已提交
1037 1038 1039 1040 1041 1042 1043 1044 1045 1046

		bytes -= tocopy;
		data += tocopy;
		addr += tocopy;
	}

	return X86EMUL_CONTINUE;
}

static int emulator_write_std(unsigned long addr,
1047
			      const void *val,
A
Avi Kivity 已提交
1048 1049 1050 1051 1052 1053 1054 1055 1056
			      unsigned int bytes,
			      struct x86_emulate_ctxt *ctxt)
{
	printk(KERN_ERR "emulator_write_std: addr %lx n %d\n",
	       addr, bytes);
	return X86EMUL_UNHANDLEABLE;
}

static int emulator_read_emulated(unsigned long addr,
1057
				  void *val,
A
Avi Kivity 已提交
1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071
				  unsigned int bytes,
				  struct x86_emulate_ctxt *ctxt)
{
	struct kvm_vcpu *vcpu = ctxt->vcpu;

	if (vcpu->mmio_read_completed) {
		memcpy(val, vcpu->mmio_data, bytes);
		vcpu->mmio_read_completed = 0;
		return X86EMUL_CONTINUE;
	} else if (emulator_read_std(addr, val, bytes, ctxt)
		   == X86EMUL_CONTINUE)
		return X86EMUL_CONTINUE;
	else {
		gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);
A
Avi Kivity 已提交
1072

A
Avi Kivity 已提交
1073
		if (gpa == UNMAPPED_GVA)
A
Avi Kivity 已提交
1074
			return X86EMUL_PROPAGATE_FAULT;
A
Avi Kivity 已提交
1075 1076 1077 1078 1079 1080 1081 1082 1083
		vcpu->mmio_needed = 1;
		vcpu->mmio_phys_addr = gpa;
		vcpu->mmio_size = bytes;
		vcpu->mmio_is_write = 0;

		return X86EMUL_UNHANDLEABLE;
	}
}

1084
static int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa,
1085
			       const void *val, int bytes)
1086 1087 1088
{
	struct page *page;
	void *virt;
1089
	unsigned offset = offset_in_page(gpa);
1090 1091 1092

	if (((gpa + bytes - 1) >> PAGE_SHIFT) != (gpa >> PAGE_SHIFT))
		return 0;
A
Avi Kivity 已提交
1093 1094
	page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
	if (!page)
1095
		return 0;
1096
	mark_page_dirty(vcpu->kvm, gpa >> PAGE_SHIFT);
1097
	virt = kmap_atomic(page, KM_USER0);
1098
	kvm_mmu_pte_write(vcpu, gpa, virt + offset, val, bytes);
1099
	memcpy(virt + offset_in_page(gpa), val, bytes);
1100 1101 1102 1103
	kunmap_atomic(virt, KM_USER0);
	return 1;
}

A
Avi Kivity 已提交
1104
static int emulator_write_emulated(unsigned long addr,
1105
				   const void *val,
A
Avi Kivity 已提交
1106 1107 1108 1109 1110 1111
				   unsigned int bytes,
				   struct x86_emulate_ctxt *ctxt)
{
	struct kvm_vcpu *vcpu = ctxt->vcpu;
	gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);

1112 1113
	if (gpa == UNMAPPED_GVA) {
		kvm_arch_ops->inject_page_fault(vcpu, addr, 2);
A
Avi Kivity 已提交
1114
		return X86EMUL_PROPAGATE_FAULT;
1115
	}
A
Avi Kivity 已提交
1116

1117 1118 1119
	if (emulator_write_phys(vcpu, gpa, val, bytes))
		return X86EMUL_CONTINUE;

A
Avi Kivity 已提交
1120 1121 1122 1123
	vcpu->mmio_needed = 1;
	vcpu->mmio_phys_addr = gpa;
	vcpu->mmio_size = bytes;
	vcpu->mmio_is_write = 1;
1124
	memcpy(vcpu->mmio_data, val, bytes);
A
Avi Kivity 已提交
1125 1126 1127 1128 1129

	return X86EMUL_CONTINUE;
}

static int emulator_cmpxchg_emulated(unsigned long addr,
1130 1131
				     const void *old,
				     const void *new,
A
Avi Kivity 已提交
1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155
				     unsigned int bytes,
				     struct x86_emulate_ctxt *ctxt)
{
	static int reported;

	if (!reported) {
		reported = 1;
		printk(KERN_WARNING "kvm: emulating exchange as write\n");
	}
	return emulator_write_emulated(addr, new, bytes, ctxt);
}

static unsigned long get_segment_base(struct kvm_vcpu *vcpu, int seg)
{
	return kvm_arch_ops->get_segment_base(vcpu, seg);
}

int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address)
{
	return X86EMUL_CONTINUE;
}

int emulate_clts(struct kvm_vcpu *vcpu)
{
1156
	unsigned long cr0;
A
Avi Kivity 已提交
1157

1158
	cr0 = vcpu->cr0 & ~CR0_TS_MASK;
A
Avi Kivity 已提交
1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227
	kvm_arch_ops->set_cr0(vcpu, cr0);
	return X86EMUL_CONTINUE;
}

int emulator_get_dr(struct x86_emulate_ctxt* ctxt, int dr, unsigned long *dest)
{
	struct kvm_vcpu *vcpu = ctxt->vcpu;

	switch (dr) {
	case 0 ... 3:
		*dest = kvm_arch_ops->get_dr(vcpu, dr);
		return X86EMUL_CONTINUE;
	default:
		printk(KERN_DEBUG "%s: unexpected dr %u\n",
		       __FUNCTION__, dr);
		return X86EMUL_UNHANDLEABLE;
	}
}

int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long value)
{
	unsigned long mask = (ctxt->mode == X86EMUL_MODE_PROT64) ? ~0ULL : ~0U;
	int exception;

	kvm_arch_ops->set_dr(ctxt->vcpu, dr, value & mask, &exception);
	if (exception) {
		/* FIXME: better handling */
		return X86EMUL_UNHANDLEABLE;
	}
	return X86EMUL_CONTINUE;
}

static void report_emulation_failure(struct x86_emulate_ctxt *ctxt)
{
	static int reported;
	u8 opcodes[4];
	unsigned long rip = ctxt->vcpu->rip;
	unsigned long rip_linear;

	rip_linear = rip + get_segment_base(ctxt->vcpu, VCPU_SREG_CS);

	if (reported)
		return;

	emulator_read_std(rip_linear, (void *)opcodes, 4, ctxt);

	printk(KERN_ERR "emulation failed but !mmio_needed?"
	       " rip %lx %02x %02x %02x %02x\n",
	       rip, opcodes[0], opcodes[1], opcodes[2], opcodes[3]);
	reported = 1;
}

struct x86_emulate_ops emulate_ops = {
	.read_std            = emulator_read_std,
	.write_std           = emulator_write_std,
	.read_emulated       = emulator_read_emulated,
	.write_emulated      = emulator_write_emulated,
	.cmpxchg_emulated    = emulator_cmpxchg_emulated,
};

int emulate_instruction(struct kvm_vcpu *vcpu,
			struct kvm_run *run,
			unsigned long cr2,
			u16 error_code)
{
	struct x86_emulate_ctxt emulate_ctxt;
	int r;
	int cs_db, cs_l;

1228
	vcpu->mmio_fault_cr2 = cr2;
A
Avi Kivity 已提交
1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266
	kvm_arch_ops->cache_regs(vcpu);

	kvm_arch_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);

	emulate_ctxt.vcpu = vcpu;
	emulate_ctxt.eflags = kvm_arch_ops->get_rflags(vcpu);
	emulate_ctxt.cr2 = cr2;
	emulate_ctxt.mode = (emulate_ctxt.eflags & X86_EFLAGS_VM)
		? X86EMUL_MODE_REAL : cs_l
		? X86EMUL_MODE_PROT64 :	cs_db
		? X86EMUL_MODE_PROT32 : X86EMUL_MODE_PROT16;

	if (emulate_ctxt.mode == X86EMUL_MODE_PROT64) {
		emulate_ctxt.cs_base = 0;
		emulate_ctxt.ds_base = 0;
		emulate_ctxt.es_base = 0;
		emulate_ctxt.ss_base = 0;
	} else {
		emulate_ctxt.cs_base = get_segment_base(vcpu, VCPU_SREG_CS);
		emulate_ctxt.ds_base = get_segment_base(vcpu, VCPU_SREG_DS);
		emulate_ctxt.es_base = get_segment_base(vcpu, VCPU_SREG_ES);
		emulate_ctxt.ss_base = get_segment_base(vcpu, VCPU_SREG_SS);
	}

	emulate_ctxt.gs_base = get_segment_base(vcpu, VCPU_SREG_GS);
	emulate_ctxt.fs_base = get_segment_base(vcpu, VCPU_SREG_FS);

	vcpu->mmio_is_write = 0;
	r = x86_emulate_memop(&emulate_ctxt, &emulate_ops);

	if ((r || vcpu->mmio_is_write) && run) {
		run->mmio.phys_addr = vcpu->mmio_phys_addr;
		memcpy(run->mmio.data, vcpu->mmio_data, 8);
		run->mmio.len = vcpu->mmio_size;
		run->mmio.is_write = vcpu->mmio_is_write;
	}

	if (r) {
1267 1268
		if (kvm_mmu_unprotect_page_virt(vcpu, cr2))
			return EMULATE_DONE;
A
Avi Kivity 已提交
1269 1270 1271 1272 1273 1274 1275 1276 1277 1278
		if (!vcpu->mmio_needed) {
			report_emulation_failure(&emulate_ctxt);
			return EMULATE_FAIL;
		}
		return EMULATE_DO_MMIO;
	}

	kvm_arch_ops->decache_regs(vcpu);
	kvm_arch_ops->set_rflags(vcpu, emulate_ctxt.eflags);

1279 1280
	if (vcpu->mmio_is_write) {
		vcpu->mmio_needed = 0;
A
Avi Kivity 已提交
1281
		return EMULATE_DO_MMIO;
1282
	}
A
Avi Kivity 已提交
1283 1284 1285 1286 1287

	return EMULATE_DONE;
}
EXPORT_SYMBOL_GPL(emulate_instruction);

1288 1289 1290 1291
int kvm_hypercall(struct kvm_vcpu *vcpu, struct kvm_run *run)
{
	unsigned long nr, a0, a1, a2, a3, a4, a5, ret;

1292
	kvm_arch_ops->cache_regs(vcpu);
1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315
	ret = -KVM_EINVAL;
#ifdef CONFIG_X86_64
	if (is_long_mode(vcpu)) {
		nr = vcpu->regs[VCPU_REGS_RAX];
		a0 = vcpu->regs[VCPU_REGS_RDI];
		a1 = vcpu->regs[VCPU_REGS_RSI];
		a2 = vcpu->regs[VCPU_REGS_RDX];
		a3 = vcpu->regs[VCPU_REGS_RCX];
		a4 = vcpu->regs[VCPU_REGS_R8];
		a5 = vcpu->regs[VCPU_REGS_R9];
	} else
#endif
	{
		nr = vcpu->regs[VCPU_REGS_RBX] & -1u;
		a0 = vcpu->regs[VCPU_REGS_RAX] & -1u;
		a1 = vcpu->regs[VCPU_REGS_RCX] & -1u;
		a2 = vcpu->regs[VCPU_REGS_RDX] & -1u;
		a3 = vcpu->regs[VCPU_REGS_RSI] & -1u;
		a4 = vcpu->regs[VCPU_REGS_RDI] & -1u;
		a5 = vcpu->regs[VCPU_REGS_RBP] & -1u;
	}
	switch (nr) {
	default:
1316 1317 1318 1319 1320 1321 1322 1323 1324 1325
		run->hypercall.args[0] = a0;
		run->hypercall.args[1] = a1;
		run->hypercall.args[2] = a2;
		run->hypercall.args[3] = a3;
		run->hypercall.args[4] = a4;
		run->hypercall.args[5] = a5;
		run->hypercall.ret = ret;
		run->hypercall.longmode = is_long_mode(vcpu);
		kvm_arch_ops->decache_regs(vcpu);
		return 0;
1326 1327
	}
	vcpu->regs[VCPU_REGS_RAX] = ret;
1328
	kvm_arch_ops->decache_regs(vcpu);
1329 1330 1331 1332
	return 1;
}
EXPORT_SYMBOL_GPL(kvm_hypercall);

A
Avi Kivity 已提交
1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360
static u64 mk_cr_64(u64 curr_cr, u32 new_val)
{
	return (curr_cr & ~((1ULL << 32) - 1)) | new_val;
}

void realmode_lgdt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base)
{
	struct descriptor_table dt = { limit, base };

	kvm_arch_ops->set_gdt(vcpu, &dt);
}

void realmode_lidt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base)
{
	struct descriptor_table dt = { limit, base };

	kvm_arch_ops->set_idt(vcpu, &dt);
}

void realmode_lmsw(struct kvm_vcpu *vcpu, unsigned long msw,
		   unsigned long *rflags)
{
	lmsw(vcpu, msw);
	*rflags = kvm_arch_ops->get_rflags(vcpu);
}

unsigned long realmode_get_cr(struct kvm_vcpu *vcpu, int cr)
{
1361
	kvm_arch_ops->decache_cr4_guest_bits(vcpu);
A
Avi Kivity 已提交
1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398
	switch (cr) {
	case 0:
		return vcpu->cr0;
	case 2:
		return vcpu->cr2;
	case 3:
		return vcpu->cr3;
	case 4:
		return vcpu->cr4;
	default:
		vcpu_printf(vcpu, "%s: unexpected cr %u\n", __FUNCTION__, cr);
		return 0;
	}
}

void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long val,
		     unsigned long *rflags)
{
	switch (cr) {
	case 0:
		set_cr0(vcpu, mk_cr_64(vcpu->cr0, val));
		*rflags = kvm_arch_ops->get_rflags(vcpu);
		break;
	case 2:
		vcpu->cr2 = val;
		break;
	case 3:
		set_cr3(vcpu, val);
		break;
	case 4:
		set_cr4(vcpu, mk_cr_64(vcpu->cr4, val));
		break;
	default:
		vcpu_printf(vcpu, "%s: unexpected cr %u\n", __FUNCTION__, cr);
	}
}

I
Ingo Molnar 已提交
1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423
/*
 * Register the para guest with the host:
 */
static int vcpu_register_para(struct kvm_vcpu *vcpu, gpa_t para_state_gpa)
{
	struct kvm_vcpu_para_state *para_state;
	hpa_t para_state_hpa, hypercall_hpa;
	struct page *para_state_page;
	unsigned char *hypercall;
	gpa_t hypercall_gpa;

	printk(KERN_DEBUG "kvm: guest trying to enter paravirtual mode\n");
	printk(KERN_DEBUG ".... para_state_gpa: %08Lx\n", para_state_gpa);

	/*
	 * Needs to be page aligned:
	 */
	if (para_state_gpa != PAGE_ALIGN(para_state_gpa))
		goto err_gp;

	para_state_hpa = gpa_to_hpa(vcpu, para_state_gpa);
	printk(KERN_DEBUG ".... para_state_hpa: %08Lx\n", para_state_hpa);
	if (is_error_hpa(para_state_hpa))
		goto err_gp;

1424
	mark_page_dirty(vcpu->kvm, para_state_gpa >> PAGE_SHIFT);
I
Ingo Molnar 已提交
1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453
	para_state_page = pfn_to_page(para_state_hpa >> PAGE_SHIFT);
	para_state = kmap_atomic(para_state_page, KM_USER0);

	printk(KERN_DEBUG "....  guest version: %d\n", para_state->guest_version);
	printk(KERN_DEBUG "....           size: %d\n", para_state->size);

	para_state->host_version = KVM_PARA_API_VERSION;
	/*
	 * We cannot support guests that try to register themselves
	 * with a newer API version than the host supports:
	 */
	if (para_state->guest_version > KVM_PARA_API_VERSION) {
		para_state->ret = -KVM_EINVAL;
		goto err_kunmap_skip;
	}

	hypercall_gpa = para_state->hypercall_gpa;
	hypercall_hpa = gpa_to_hpa(vcpu, hypercall_gpa);
	printk(KERN_DEBUG ".... hypercall_hpa: %08Lx\n", hypercall_hpa);
	if (is_error_hpa(hypercall_hpa)) {
		para_state->ret = -KVM_EINVAL;
		goto err_kunmap_skip;
	}

	printk(KERN_DEBUG "kvm: para guest successfully registered.\n");
	vcpu->para_state_page = para_state_page;
	vcpu->para_state_gpa = para_state_gpa;
	vcpu->hypercall_gpa = hypercall_gpa;

1454
	mark_page_dirty(vcpu->kvm, hypercall_gpa >> PAGE_SHIFT);
I
Ingo Molnar 已提交
1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467
	hypercall = kmap_atomic(pfn_to_page(hypercall_hpa >> PAGE_SHIFT),
				KM_USER1) + (hypercall_hpa & ~PAGE_MASK);
	kvm_arch_ops->patch_hypercall(vcpu, hypercall);
	kunmap_atomic(hypercall, KM_USER1);

	para_state->ret = 0;
err_kunmap_skip:
	kunmap_atomic(para_state, KM_USER0);
	return 0;
err_gp:
	return 1;
}

1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486
int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
{
	u64 data;

	switch (msr) {
	case 0xc0010010: /* SYSCFG */
	case 0xc0010015: /* HWCR */
	case MSR_IA32_PLATFORM_ID:
	case MSR_IA32_P5_MC_ADDR:
	case MSR_IA32_P5_MC_TYPE:
	case MSR_IA32_MC0_CTL:
	case MSR_IA32_MCG_STATUS:
	case MSR_IA32_MCG_CAP:
	case MSR_IA32_MC0_MISC:
	case MSR_IA32_MC0_MISC+4:
	case MSR_IA32_MC0_MISC+8:
	case MSR_IA32_MC0_MISC+12:
	case MSR_IA32_MC0_MISC+16:
	case MSR_IA32_UCODE_REV:
A
Avi Kivity 已提交
1487
	case MSR_IA32_PERF_STATUS:
1488
	case MSR_IA32_EBL_CR_POWERON:
1489 1490 1491 1492 1493
		/* MTRR registers */
	case 0xfe:
	case 0x200 ... 0x2ff:
		data = 0;
		break;
A
Avi Kivity 已提交
1494 1495 1496
	case 0xcd: /* fsb frequency */
		data = 3;
		break;
1497 1498 1499
	case MSR_IA32_APICBASE:
		data = vcpu->apic_base;
		break;
1500 1501 1502
	case MSR_IA32_MISC_ENABLE:
		data = vcpu->ia32_misc_enable_msr;
		break;
1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516
#ifdef CONFIG_X86_64
	case MSR_EFER:
		data = vcpu->shadow_efer;
		break;
#endif
	default:
		printk(KERN_ERR "kvm: unhandled rdmsr: 0x%x\n", msr);
		return 1;
	}
	*pdata = data;
	return 0;
}
EXPORT_SYMBOL_GPL(kvm_get_msr_common);

A
Avi Kivity 已提交
1517 1518 1519 1520 1521 1522 1523 1524 1525 1526
/*
 * Reads an msr value (of 'msr_index') into 'pdata'.
 * Returns 0 on success, non-0 otherwise.
 * Assumes vcpu_load() was already called.
 */
static int get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata)
{
	return kvm_arch_ops->get_msr(vcpu, msr_index, pdata);
}

1527
#ifdef CONFIG_X86_64
A
Avi Kivity 已提交
1528

1529
static void set_efer(struct kvm_vcpu *vcpu, u64 efer)
A
Avi Kivity 已提交
1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544
{
	if (efer & EFER_RESERVED_BITS) {
		printk(KERN_DEBUG "set_efer: 0x%llx #GP, reserved bits\n",
		       efer);
		inject_gp(vcpu);
		return;
	}

	if (is_paging(vcpu)
	    && (vcpu->shadow_efer & EFER_LME) != (efer & EFER_LME)) {
		printk(KERN_DEBUG "set_efer: #GP, change LME while paging\n");
		inject_gp(vcpu);
		return;
	}

1545 1546
	kvm_arch_ops->set_efer(vcpu, efer);

A
Avi Kivity 已提交
1547 1548 1549 1550 1551 1552 1553 1554
	efer &= ~EFER_LMA;
	efer |= vcpu->shadow_efer & EFER_LMA;

	vcpu->shadow_efer = efer;
}

#endif

1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566
int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
{
	switch (msr) {
#ifdef CONFIG_X86_64
	case MSR_EFER:
		set_efer(vcpu, data);
		break;
#endif
	case MSR_IA32_MC0_STATUS:
		printk(KERN_WARNING "%s: MSR_IA32_MC0_STATUS 0x%llx, nop\n",
		       __FUNCTION__, data);
		break;
1567 1568 1569 1570
	case MSR_IA32_MCG_STATUS:
		printk(KERN_WARNING "%s: MSR_IA32_MCG_STATUS 0x%llx, nop\n",
			__FUNCTION__, data);
		break;
1571 1572 1573 1574 1575 1576 1577
	case MSR_IA32_UCODE_REV:
	case MSR_IA32_UCODE_WRITE:
	case 0x200 ... 0x2ff: /* MTRRs */
		break;
	case MSR_IA32_APICBASE:
		vcpu->apic_base = data;
		break;
1578 1579 1580
	case MSR_IA32_MISC_ENABLE:
		vcpu->ia32_misc_enable_msr = data;
		break;
I
Ingo Molnar 已提交
1581 1582 1583 1584 1585 1586
	/*
	 * This is the 'probe whether the host is KVM' logic:
	 */
	case MSR_KVM_API_MAGIC:
		return vcpu_register_para(vcpu, data);

1587 1588 1589 1590 1591 1592 1593 1594
	default:
		printk(KERN_ERR "kvm: unhandled wrmsr: 0x%x\n", msr);
		return 1;
	}
	return 0;
}
EXPORT_SYMBOL_GPL(kvm_set_msr_common);

A
Avi Kivity 已提交
1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606
/*
 * Writes msr value into into the appropriate "register".
 * Returns 0 on success, non-0 otherwise.
 * Assumes vcpu_load() was already called.
 */
static int set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data)
{
	return kvm_arch_ops->set_msr(vcpu, msr_index, data);
}

void kvm_resched(struct kvm_vcpu *vcpu)
{
1607 1608
	if (!need_resched())
		return;
A
Avi Kivity 已提交
1609 1610
	vcpu_put(vcpu);
	cond_resched();
A
Avi Kivity 已提交
1611
	vcpu_load(vcpu);
A
Avi Kivity 已提交
1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632
}
EXPORT_SYMBOL_GPL(kvm_resched);

void load_msrs(struct vmx_msr_entry *e, int n)
{
	int i;

	for (i = 0; i < n; ++i)
		wrmsrl(e[i].index, e[i].data);
}
EXPORT_SYMBOL_GPL(load_msrs);

void save_msrs(struct vmx_msr_entry *e, int n)
{
	int i;

	for (i = 0; i < n; ++i)
		rdmsrl(e[i].index, e[i].data);
}
EXPORT_SYMBOL_GPL(save_msrs);

1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669
void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
{
	int i;
	u32 function;
	struct kvm_cpuid_entry *e, *best;

	kvm_arch_ops->cache_regs(vcpu);
	function = vcpu->regs[VCPU_REGS_RAX];
	vcpu->regs[VCPU_REGS_RAX] = 0;
	vcpu->regs[VCPU_REGS_RBX] = 0;
	vcpu->regs[VCPU_REGS_RCX] = 0;
	vcpu->regs[VCPU_REGS_RDX] = 0;
	best = NULL;
	for (i = 0; i < vcpu->cpuid_nent; ++i) {
		e = &vcpu->cpuid_entries[i];
		if (e->function == function) {
			best = e;
			break;
		}
		/*
		 * Both basic or both extended?
		 */
		if (((e->function ^ function) & 0x80000000) == 0)
			if (!best || e->function > best->function)
				best = e;
	}
	if (best) {
		vcpu->regs[VCPU_REGS_RAX] = best->eax;
		vcpu->regs[VCPU_REGS_RBX] = best->ebx;
		vcpu->regs[VCPU_REGS_RCX] = best->ecx;
		vcpu->regs[VCPU_REGS_RDX] = best->edx;
	}
	kvm_arch_ops->decache_regs(vcpu);
	kvm_arch_ops->skip_emulated_instruction(vcpu);
}
EXPORT_SYMBOL_GPL(kvm_emulate_cpuid);

1670
static int pio_copy_data(struct kvm_vcpu *vcpu)
1671
{
1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700
	void *p = vcpu->pio_data;
	void *q;
	unsigned bytes;
	int nr_pages = vcpu->pio.guest_pages[1] ? 2 : 1;

	kvm_arch_ops->vcpu_put(vcpu);
	q = vmap(vcpu->pio.guest_pages, nr_pages, VM_READ|VM_WRITE,
		 PAGE_KERNEL);
	if (!q) {
		kvm_arch_ops->vcpu_load(vcpu);
		free_pio_guest_pages(vcpu);
		return -ENOMEM;
	}
	q += vcpu->pio.guest_page_offset;
	bytes = vcpu->pio.size * vcpu->pio.cur_count;
	if (vcpu->pio.in)
		memcpy(q, p, bytes);
	else
		memcpy(p, q, bytes);
	q -= vcpu->pio.guest_page_offset;
	vunmap(q);
	kvm_arch_ops->vcpu_load(vcpu);
	free_pio_guest_pages(vcpu);
	return 0;
}

static int complete_pio(struct kvm_vcpu *vcpu)
{
	struct kvm_pio_request *io = &vcpu->pio;
1701
	long delta;
1702
	int r;
1703 1704 1705 1706

	kvm_arch_ops->cache_regs(vcpu);

	if (!io->string) {
1707 1708
		if (io->in)
			memcpy(&vcpu->regs[VCPU_REGS_RAX], vcpu->pio_data,
1709 1710
			       io->size);
	} else {
1711 1712 1713 1714 1715 1716 1717 1718
		if (io->in) {
			r = pio_copy_data(vcpu);
			if (r) {
				kvm_arch_ops->cache_regs(vcpu);
				return r;
			}
		}

1719 1720
		delta = 1;
		if (io->rep) {
1721
			delta *= io->cur_count;
1722 1723 1724 1725 1726 1727
			/*
			 * The size of the register should really depend on
			 * current address size.
			 */
			vcpu->regs[VCPU_REGS_RCX] -= delta;
		}
1728
		if (io->down)
1729 1730
			delta = -delta;
		delta *= io->size;
1731
		if (io->in)
1732 1733 1734 1735 1736 1737 1738
			vcpu->regs[VCPU_REGS_RDI] += delta;
		else
			vcpu->regs[VCPU_REGS_RSI] += delta;
	}

	kvm_arch_ops->decache_regs(vcpu);

1739 1740 1741 1742 1743 1744
	io->count -= io->cur_count;
	io->cur_count = 0;

	if (!io->count)
		kvm_arch_ops->skip_emulated_instruction(vcpu);
	return 0;
1745 1746
}

1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 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
int kvm_setup_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
		  int size, unsigned long count, int string, int down,
		  gva_t address, int rep, unsigned port)
{
	unsigned now, in_page;
	int i;
	int nr_pages = 1;
	struct page *page;

	vcpu->run->exit_reason = KVM_EXIT_IO;
	vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT;
	vcpu->run->io.size = size;
	vcpu->run->io.data_offset = KVM_PIO_PAGE_OFFSET * PAGE_SIZE;
	vcpu->run->io.count = count;
	vcpu->run->io.port = port;
	vcpu->pio.count = count;
	vcpu->pio.cur_count = count;
	vcpu->pio.size = size;
	vcpu->pio.in = in;
	vcpu->pio.string = string;
	vcpu->pio.down = down;
	vcpu->pio.guest_page_offset = offset_in_page(address);
	vcpu->pio.rep = rep;

	if (!string) {
		kvm_arch_ops->cache_regs(vcpu);
		memcpy(vcpu->pio_data, &vcpu->regs[VCPU_REGS_RAX], 4);
		kvm_arch_ops->decache_regs(vcpu);
		return 0;
	}

	if (!count) {
		kvm_arch_ops->skip_emulated_instruction(vcpu);
		return 1;
	}

	now = min(count, PAGE_SIZE / size);

	if (!down)
		in_page = PAGE_SIZE - offset_in_page(address);
	else
		in_page = offset_in_page(address) + size;
	now = min(count, (unsigned long)in_page / size);
	if (!now) {
		/*
		 * String I/O straddles page boundary.  Pin two guest pages
		 * so that we satisfy atomicity constraints.  Do just one
		 * transaction to avoid complexity.
		 */
		nr_pages = 2;
		now = 1;
	}
	if (down) {
		/*
		 * String I/O in reverse.  Yuck.  Kill the guest, fix later.
		 */
		printk(KERN_ERR "kvm: guest string pio down\n");
		inject_gp(vcpu);
		return 1;
	}
	vcpu->run->io.count = now;
	vcpu->pio.cur_count = now;

	for (i = 0; i < nr_pages; ++i) {
		spin_lock(&vcpu->kvm->lock);
		page = gva_to_page(vcpu, address + i * PAGE_SIZE);
		if (page)
			get_page(page);
		vcpu->pio.guest_pages[i] = page;
		spin_unlock(&vcpu->kvm->lock);
		if (!page) {
			inject_gp(vcpu);
			free_pio_guest_pages(vcpu);
			return 1;
		}
	}

	if (!vcpu->pio.in)
		return pio_copy_data(vcpu);
	return 0;
}
EXPORT_SYMBOL_GPL(kvm_setup_pio);

A
Avi Kivity 已提交
1830
static int kvm_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
A
Avi Kivity 已提交
1831 1832
{
	int r;
A
Avi Kivity 已提交
1833
	sigset_t sigsaved;
A
Avi Kivity 已提交
1834

A
Avi Kivity 已提交
1835
	vcpu_load(vcpu);
A
Avi Kivity 已提交
1836

A
Avi Kivity 已提交
1837 1838 1839
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);

1840 1841 1842
	/* re-sync apic's tpr */
	vcpu->cr8 = kvm_run->cr8;

1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861
	if (vcpu->pio.cur_count) {
		r = complete_pio(vcpu);
		if (r)
			goto out;
	}

	if (vcpu->mmio_needed) {
		memcpy(vcpu->mmio_data, kvm_run->mmio.data, 8);
		vcpu->mmio_read_completed = 1;
		vcpu->mmio_needed = 0;
		r = emulate_instruction(vcpu, kvm_run,
					vcpu->mmio_fault_cr2, 0);
		if (r == EMULATE_DO_MMIO) {
			/*
			 * Read-modify-write.  Back to userspace.
			 */
			kvm_run->exit_reason = KVM_EXIT_MMIO;
			r = 0;
			goto out;
1862
		}
A
Avi Kivity 已提交
1863 1864
	}

1865
	if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) {
1866 1867 1868 1869 1870
		kvm_arch_ops->cache_regs(vcpu);
		vcpu->regs[VCPU_REGS_RAX] = kvm_run->hypercall.ret;
		kvm_arch_ops->decache_regs(vcpu);
	}

A
Avi Kivity 已提交
1871 1872
	r = kvm_arch_ops->run(vcpu, kvm_run);

1873
out:
A
Avi Kivity 已提交
1874 1875 1876
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &sigsaved, NULL);

A
Avi Kivity 已提交
1877 1878 1879 1880
	vcpu_put(vcpu);
	return r;
}

A
Avi Kivity 已提交
1881 1882
static int kvm_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu,
				   struct kvm_regs *regs)
A
Avi Kivity 已提交
1883
{
A
Avi Kivity 已提交
1884
	vcpu_load(vcpu);
A
Avi Kivity 已提交
1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895

	kvm_arch_ops->cache_regs(vcpu);

	regs->rax = vcpu->regs[VCPU_REGS_RAX];
	regs->rbx = vcpu->regs[VCPU_REGS_RBX];
	regs->rcx = vcpu->regs[VCPU_REGS_RCX];
	regs->rdx = vcpu->regs[VCPU_REGS_RDX];
	regs->rsi = vcpu->regs[VCPU_REGS_RSI];
	regs->rdi = vcpu->regs[VCPU_REGS_RDI];
	regs->rsp = vcpu->regs[VCPU_REGS_RSP];
	regs->rbp = vcpu->regs[VCPU_REGS_RBP];
1896
#ifdef CONFIG_X86_64
A
Avi Kivity 已提交
1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920
	regs->r8 = vcpu->regs[VCPU_REGS_R8];
	regs->r9 = vcpu->regs[VCPU_REGS_R9];
	regs->r10 = vcpu->regs[VCPU_REGS_R10];
	regs->r11 = vcpu->regs[VCPU_REGS_R11];
	regs->r12 = vcpu->regs[VCPU_REGS_R12];
	regs->r13 = vcpu->regs[VCPU_REGS_R13];
	regs->r14 = vcpu->regs[VCPU_REGS_R14];
	regs->r15 = vcpu->regs[VCPU_REGS_R15];
#endif

	regs->rip = vcpu->rip;
	regs->rflags = kvm_arch_ops->get_rflags(vcpu);

	/*
	 * Don't leak debug flags in case they were set for guest debugging
	 */
	if (vcpu->guest_debug.enabled && vcpu->guest_debug.singlestep)
		regs->rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);

	vcpu_put(vcpu);

	return 0;
}

A
Avi Kivity 已提交
1921 1922
static int kvm_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu,
				   struct kvm_regs *regs)
A
Avi Kivity 已提交
1923
{
A
Avi Kivity 已提交
1924
	vcpu_load(vcpu);
A
Avi Kivity 已提交
1925 1926 1927 1928 1929 1930 1931 1932 1933

	vcpu->regs[VCPU_REGS_RAX] = regs->rax;
	vcpu->regs[VCPU_REGS_RBX] = regs->rbx;
	vcpu->regs[VCPU_REGS_RCX] = regs->rcx;
	vcpu->regs[VCPU_REGS_RDX] = regs->rdx;
	vcpu->regs[VCPU_REGS_RSI] = regs->rsi;
	vcpu->regs[VCPU_REGS_RDI] = regs->rdi;
	vcpu->regs[VCPU_REGS_RSP] = regs->rsp;
	vcpu->regs[VCPU_REGS_RBP] = regs->rbp;
1934
#ifdef CONFIG_X86_64
A
Avi Kivity 已提交
1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960
	vcpu->regs[VCPU_REGS_R8] = regs->r8;
	vcpu->regs[VCPU_REGS_R9] = regs->r9;
	vcpu->regs[VCPU_REGS_R10] = regs->r10;
	vcpu->regs[VCPU_REGS_R11] = regs->r11;
	vcpu->regs[VCPU_REGS_R12] = regs->r12;
	vcpu->regs[VCPU_REGS_R13] = regs->r13;
	vcpu->regs[VCPU_REGS_R14] = regs->r14;
	vcpu->regs[VCPU_REGS_R15] = regs->r15;
#endif

	vcpu->rip = regs->rip;
	kvm_arch_ops->set_rflags(vcpu, regs->rflags);

	kvm_arch_ops->decache_regs(vcpu);

	vcpu_put(vcpu);

	return 0;
}

static void get_segment(struct kvm_vcpu *vcpu,
			struct kvm_segment *var, int seg)
{
	return kvm_arch_ops->get_segment(vcpu, var, seg);
}

A
Avi Kivity 已提交
1961 1962
static int kvm_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
				    struct kvm_sregs *sregs)
A
Avi Kivity 已提交
1963 1964 1965
{
	struct descriptor_table dt;

A
Avi Kivity 已提交
1966
	vcpu_load(vcpu);
A
Avi Kivity 已提交
1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984

	get_segment(vcpu, &sregs->cs, VCPU_SREG_CS);
	get_segment(vcpu, &sregs->ds, VCPU_SREG_DS);
	get_segment(vcpu, &sregs->es, VCPU_SREG_ES);
	get_segment(vcpu, &sregs->fs, VCPU_SREG_FS);
	get_segment(vcpu, &sregs->gs, VCPU_SREG_GS);
	get_segment(vcpu, &sregs->ss, VCPU_SREG_SS);

	get_segment(vcpu, &sregs->tr, VCPU_SREG_TR);
	get_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR);

	kvm_arch_ops->get_idt(vcpu, &dt);
	sregs->idt.limit = dt.limit;
	sregs->idt.base = dt.base;
	kvm_arch_ops->get_gdt(vcpu, &dt);
	sregs->gdt.limit = dt.limit;
	sregs->gdt.base = dt.base;

1985
	kvm_arch_ops->decache_cr4_guest_bits(vcpu);
A
Avi Kivity 已提交
1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
	sregs->cr0 = vcpu->cr0;
	sregs->cr2 = vcpu->cr2;
	sregs->cr3 = vcpu->cr3;
	sregs->cr4 = vcpu->cr4;
	sregs->cr8 = vcpu->cr8;
	sregs->efer = vcpu->shadow_efer;
	sregs->apic_base = vcpu->apic_base;

	memcpy(sregs->interrupt_bitmap, vcpu->irq_pending,
	       sizeof sregs->interrupt_bitmap);

	vcpu_put(vcpu);

	return 0;
}

static void set_segment(struct kvm_vcpu *vcpu,
			struct kvm_segment *var, int seg)
{
	return kvm_arch_ops->set_segment(vcpu, var, seg);
}

A
Avi Kivity 已提交
2008 2009
static int kvm_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
				    struct kvm_sregs *sregs)
A
Avi Kivity 已提交
2010 2011 2012 2013 2014
{
	int mmu_reset_needed = 0;
	int i;
	struct descriptor_table dt;

A
Avi Kivity 已提交
2015
	vcpu_load(vcpu);
A
Avi Kivity 已提交
2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030

	dt.limit = sregs->idt.limit;
	dt.base = sregs->idt.base;
	kvm_arch_ops->set_idt(vcpu, &dt);
	dt.limit = sregs->gdt.limit;
	dt.base = sregs->gdt.base;
	kvm_arch_ops->set_gdt(vcpu, &dt);

	vcpu->cr2 = sregs->cr2;
	mmu_reset_needed |= vcpu->cr3 != sregs->cr3;
	vcpu->cr3 = sregs->cr3;

	vcpu->cr8 = sregs->cr8;

	mmu_reset_needed |= vcpu->shadow_efer != sregs->efer;
2031
#ifdef CONFIG_X86_64
A
Avi Kivity 已提交
2032 2033 2034 2035
	kvm_arch_ops->set_efer(vcpu, sregs->efer);
#endif
	vcpu->apic_base = sregs->apic_base;

2036
	kvm_arch_ops->decache_cr4_guest_bits(vcpu);
2037

A
Avi Kivity 已提交
2038
	mmu_reset_needed |= vcpu->cr0 != sregs->cr0;
2039
	kvm_arch_ops->set_cr0(vcpu, sregs->cr0);
A
Avi Kivity 已提交
2040 2041 2042

	mmu_reset_needed |= vcpu->cr4 != sregs->cr4;
	kvm_arch_ops->set_cr4(vcpu, sregs->cr4);
2043 2044
	if (!is_long_mode(vcpu) && is_pae(vcpu))
		load_pdptrs(vcpu, vcpu->cr3);
A
Avi Kivity 已提交
2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055

	if (mmu_reset_needed)
		kvm_mmu_reset_context(vcpu);

	memcpy(vcpu->irq_pending, sregs->interrupt_bitmap,
	       sizeof vcpu->irq_pending);
	vcpu->irq_summary = 0;
	for (i = 0; i < NR_IRQ_WORDS; ++i)
		if (vcpu->irq_pending[i])
			__set_bit(i, &vcpu->irq_summary);

2056 2057 2058 2059 2060 2061 2062 2063 2064 2065
	set_segment(vcpu, &sregs->cs, VCPU_SREG_CS);
	set_segment(vcpu, &sregs->ds, VCPU_SREG_DS);
	set_segment(vcpu, &sregs->es, VCPU_SREG_ES);
	set_segment(vcpu, &sregs->fs, VCPU_SREG_FS);
	set_segment(vcpu, &sregs->gs, VCPU_SREG_GS);
	set_segment(vcpu, &sregs->ss, VCPU_SREG_SS);

	set_segment(vcpu, &sregs->tr, VCPU_SREG_TR);
	set_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR);

A
Avi Kivity 已提交
2066 2067 2068 2069 2070 2071 2072 2073
	vcpu_put(vcpu);

	return 0;
}

/*
 * List of msr numbers which we expose to userspace through KVM_GET_MSRS
 * and KVM_SET_MSRS, and KVM_GET_MSR_INDEX_LIST.
2074 2075 2076
 *
 * This list is modified at module load time to reflect the
 * capabilities of the host cpu.
A
Avi Kivity 已提交
2077 2078 2079 2080
 */
static u32 msrs_to_save[] = {
	MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP,
	MSR_K6_STAR,
2081
#ifdef CONFIG_X86_64
A
Avi Kivity 已提交
2082 2083 2084 2085 2086
	MSR_CSTAR, MSR_KERNEL_GS_BASE, MSR_SYSCALL_MASK, MSR_LSTAR,
#endif
	MSR_IA32_TIME_STAMP_COUNTER,
};

2087 2088
static unsigned num_msrs_to_save;

2089 2090 2091 2092
static u32 emulated_msrs[] = {
	MSR_IA32_MISC_ENABLE,
};

2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106
static __init void kvm_init_msr_list(void)
{
	u32 dummy[2];
	unsigned i, j;

	for (i = j = 0; i < ARRAY_SIZE(msrs_to_save); i++) {
		if (rdmsr_safe(msrs_to_save[i], &dummy[0], &dummy[1]) < 0)
			continue;
		if (j < i)
			msrs_to_save[j] = msrs_to_save[i];
		j++;
	}
	num_msrs_to_save = j;
}
A
Avi Kivity 已提交
2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120

/*
 * Adapt set_msr() to msr_io()'s calling convention
 */
static int do_set_msr(struct kvm_vcpu *vcpu, unsigned index, u64 *data)
{
	return set_msr(vcpu, index, *data);
}

/*
 * Read or write a bunch of msrs. All parameters are kernel addresses.
 *
 * @return number of msrs set successfully.
 */
A
Avi Kivity 已提交
2121
static int __msr_io(struct kvm_vcpu *vcpu, struct kvm_msrs *msrs,
A
Avi Kivity 已提交
2122 2123 2124 2125 2126 2127
		    struct kvm_msr_entry *entries,
		    int (*do_msr)(struct kvm_vcpu *vcpu,
				  unsigned index, u64 *data))
{
	int i;

A
Avi Kivity 已提交
2128
	vcpu_load(vcpu);
A
Avi Kivity 已提交
2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143

	for (i = 0; i < msrs->nmsrs; ++i)
		if (do_msr(vcpu, entries[i].index, &entries[i].data))
			break;

	vcpu_put(vcpu);

	return i;
}

/*
 * Read or write a bunch of msrs. Parameters are user addresses.
 *
 * @return number of msrs set successfully.
 */
A
Avi Kivity 已提交
2144
static int msr_io(struct kvm_vcpu *vcpu, struct kvm_msrs __user *user_msrs,
A
Avi Kivity 已提交
2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171
		  int (*do_msr)(struct kvm_vcpu *vcpu,
				unsigned index, u64 *data),
		  int writeback)
{
	struct kvm_msrs msrs;
	struct kvm_msr_entry *entries;
	int r, n;
	unsigned size;

	r = -EFAULT;
	if (copy_from_user(&msrs, user_msrs, sizeof msrs))
		goto out;

	r = -E2BIG;
	if (msrs.nmsrs >= MAX_IO_MSRS)
		goto out;

	r = -ENOMEM;
	size = sizeof(struct kvm_msr_entry) * msrs.nmsrs;
	entries = vmalloc(size);
	if (!entries)
		goto out;

	r = -EFAULT;
	if (copy_from_user(entries, user_msrs->entries, size))
		goto out_free;

A
Avi Kivity 已提交
2172
	r = n = __msr_io(vcpu, &msrs, entries, do_msr);
A
Avi Kivity 已提交
2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190
	if (r < 0)
		goto out_free;

	r = -EFAULT;
	if (writeback && copy_to_user(user_msrs->entries, entries, size))
		goto out_free;

	r = n;

out_free:
	vfree(entries);
out:
	return r;
}

/*
 * Translate a guest virtual address to a guest physical address.
 */
A
Avi Kivity 已提交
2191 2192
static int kvm_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
				    struct kvm_translation *tr)
A
Avi Kivity 已提交
2193 2194 2195 2196
{
	unsigned long vaddr = tr->linear_address;
	gpa_t gpa;

A
Avi Kivity 已提交
2197 2198
	vcpu_load(vcpu);
	spin_lock(&vcpu->kvm->lock);
A
Avi Kivity 已提交
2199 2200 2201 2202 2203
	gpa = vcpu->mmu.gva_to_gpa(vcpu, vaddr);
	tr->physical_address = gpa;
	tr->valid = gpa != UNMAPPED_GVA;
	tr->writeable = 1;
	tr->usermode = 0;
A
Avi Kivity 已提交
2204
	spin_unlock(&vcpu->kvm->lock);
A
Avi Kivity 已提交
2205 2206 2207 2208 2209
	vcpu_put(vcpu);

	return 0;
}

A
Avi Kivity 已提交
2210 2211
static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu,
				    struct kvm_interrupt *irq)
A
Avi Kivity 已提交
2212 2213 2214
{
	if (irq->irq < 0 || irq->irq >= 256)
		return -EINVAL;
A
Avi Kivity 已提交
2215
	vcpu_load(vcpu);
A
Avi Kivity 已提交
2216 2217 2218 2219 2220 2221 2222 2223 2224

	set_bit(irq->irq, vcpu->irq_pending);
	set_bit(irq->irq / BITS_PER_LONG, &vcpu->irq_summary);

	vcpu_put(vcpu);

	return 0;
}

A
Avi Kivity 已提交
2225 2226
static int kvm_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
				      struct kvm_debug_guest *dbg)
A
Avi Kivity 已提交
2227 2228 2229
{
	int r;

A
Avi Kivity 已提交
2230
	vcpu_load(vcpu);
A
Avi Kivity 已提交
2231 2232 2233 2234 2235 2236 2237 2238

	r = kvm_arch_ops->set_guest_debug(vcpu, dbg);

	vcpu_put(vcpu);

	return r;
}

2239 2240 2241 2242 2243 2244 2245 2246 2247 2248
static struct page *kvm_vcpu_nopage(struct vm_area_struct *vma,
				    unsigned long address,
				    int *type)
{
	struct kvm_vcpu *vcpu = vma->vm_file->private_data;
	unsigned long pgoff;
	struct page *page;

	*type = VM_FAULT_MINOR;
	pgoff = ((address - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff;
2249 2250 2251 2252 2253
	if (pgoff == 0)
		page = virt_to_page(vcpu->run);
	else if (pgoff == KVM_PIO_PAGE_OFFSET)
		page = virt_to_page(vcpu->pio_data);
	else
2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268
		return NOPAGE_SIGBUS;
	get_page(page);
	return page;
}

static struct vm_operations_struct kvm_vcpu_vm_ops = {
	.nopage = kvm_vcpu_nopage,
};

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
2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280
static int kvm_vcpu_release(struct inode *inode, struct file *filp)
{
	struct kvm_vcpu *vcpu = filp->private_data;

	fput(vcpu->kvm->filp);
	return 0;
}

static struct file_operations kvm_vcpu_fops = {
	.release        = kvm_vcpu_release,
	.unlocked_ioctl = kvm_vcpu_ioctl,
	.compat_ioctl   = kvm_vcpu_ioctl,
2281
	.mmap           = kvm_vcpu_mmap,
A
Avi Kivity 已提交
2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322
};

/*
 * Allocates an inode for the vcpu.
 */
static int create_vcpu_fd(struct kvm_vcpu *vcpu)
{
	int fd, r;
	struct inode *inode;
	struct file *file;

	atomic_inc(&vcpu->kvm->filp->f_count);
	inode = kvmfs_inode(&kvm_vcpu_fops);
	if (IS_ERR(inode)) {
		r = PTR_ERR(inode);
		goto out1;
	}

	file = kvmfs_file(inode, vcpu);
	if (IS_ERR(file)) {
		r = PTR_ERR(file);
		goto out2;
	}

	r = get_unused_fd();
	if (r < 0)
		goto out3;
	fd = r;
	fd_install(fd, file);

	return fd;

out3:
	fput(file);
out2:
	iput(inode);
out1:
	fput(vcpu->kvm->filp);
	return r;
}

2323 2324 2325 2326 2327 2328 2329
/*
 * 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;
2330
	struct page *page;
2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344

	r = -EINVAL;
	if (!valid_vcpu(n))
		goto out;

	vcpu = &kvm->vcpus[n];

	mutex_lock(&vcpu->mutex);

	if (vcpu->vmcs) {
		mutex_unlock(&vcpu->mutex);
		return -EEXIST;
	}

2345 2346 2347 2348 2349 2350
	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
	r = -ENOMEM;
	if (!page)
		goto out_unlock;
	vcpu->run = page_address(page);

2351 2352 2353 2354 2355 2356
	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
	r = -ENOMEM;
	if (!page)
		goto out_free_run;
	vcpu->pio_data = page_address(page);

2357 2358 2359
	vcpu->host_fx_image = (char*)ALIGN((hva_t)vcpu->fx_buf,
					   FX_IMAGE_ALIGN);
	vcpu->guest_fx_image = vcpu->host_fx_image + FX_IMAGE_SIZE;
2360
	vcpu->cr0 = 0x10;
2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378

	r = kvm_arch_ops->vcpu_create(vcpu);
	if (r < 0)
		goto out_free_vcpus;

	r = kvm_mmu_create(vcpu);
	if (r < 0)
		goto out_free_vcpus;

	kvm_arch_ops->vcpu_load(vcpu);
	r = kvm_mmu_setup(vcpu);
	if (r >= 0)
		r = kvm_arch_ops->vcpu_setup(vcpu);
	vcpu_put(vcpu);

	if (r < 0)
		goto out_free_vcpus;

A
Avi Kivity 已提交
2379 2380 2381 2382 2383
	r = create_vcpu_fd(vcpu);
	if (r < 0)
		goto out_free_vcpus;

	return r;
2384 2385 2386

out_free_vcpus:
	kvm_free_vcpu(vcpu);
2387 2388 2389
out_free_run:
	free_page((unsigned long)vcpu->run);
	vcpu->run = NULL;
2390
out_unlock:
2391 2392 2393 2394 2395
	mutex_unlock(&vcpu->mutex);
out:
	return r;
}

2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416
static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu)
{
	u64 efer;
	int i;
	struct kvm_cpuid_entry *e, *entry;

	rdmsrl(MSR_EFER, efer);
	entry = NULL;
	for (i = 0; i < vcpu->cpuid_nent; ++i) {
		e = &vcpu->cpuid_entries[i];
		if (e->function == 0x80000001) {
			entry = e;
			break;
		}
	}
	if (entry && (entry->edx & EFER_NX) && !(efer & EFER_NX)) {
		entry->edx &= ~(1 << 20);
		printk(KERN_INFO ": guest NX capability removed\n");
	}
}

2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430
static int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu,
				    struct kvm_cpuid *cpuid,
				    struct kvm_cpuid_entry __user *entries)
{
	int r;

	r = -E2BIG;
	if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
		goto out;
	r = -EFAULT;
	if (copy_from_user(&vcpu->cpuid_entries, entries,
			   cpuid->nent * sizeof(struct kvm_cpuid_entry)))
		goto out;
	vcpu->cpuid_nent = cpuid->nent;
2431
	cpuid_fix_nx_cap(vcpu);
2432 2433 2434 2435 2436 2437
	return 0;

out:
	return r;
}

A
Avi Kivity 已提交
2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448
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;
}

A
Avi Kivity 已提交
2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509
/*
 * fxsave fpu state.  Taken from x86_64/processor.h.  To be killed when
 * we have asm/x86/processor.h
 */
struct fxsave {
	u16	cwd;
	u16	swd;
	u16	twd;
	u16	fop;
	u64	rip;
	u64	rdp;
	u32	mxcsr;
	u32	mxcsr_mask;
	u32	st_space[32];	/* 8*16 bytes for each FP-reg = 128 bytes */
#ifdef CONFIG_X86_64
	u32	xmm_space[64];	/* 16*16 bytes for each XMM-reg = 256 bytes */
#else
	u32	xmm_space[32];	/* 8*16 bytes for each XMM-reg = 128 bytes */
#endif
};

static int kvm_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
	struct fxsave *fxsave = (struct fxsave *)vcpu->guest_fx_image;

	vcpu_load(vcpu);

	memcpy(fpu->fpr, fxsave->st_space, 128);
	fpu->fcw = fxsave->cwd;
	fpu->fsw = fxsave->swd;
	fpu->ftwx = fxsave->twd;
	fpu->last_opcode = fxsave->fop;
	fpu->last_ip = fxsave->rip;
	fpu->last_dp = fxsave->rdp;
	memcpy(fpu->xmm, fxsave->xmm_space, sizeof fxsave->xmm_space);

	vcpu_put(vcpu);

	return 0;
}

static int kvm_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
	struct fxsave *fxsave = (struct fxsave *)vcpu->guest_fx_image;

	vcpu_load(vcpu);

	memcpy(fxsave->st_space, fpu->fpr, 128);
	fxsave->cwd = fpu->fcw;
	fxsave->swd = fpu->fsw;
	fxsave->twd = fpu->ftwx;
	fxsave->fop = fpu->last_opcode;
	fxsave->rip = fpu->last_ip;
	fxsave->rdp = fpu->last_dp;
	memcpy(fxsave->xmm_space, fpu->xmm, sizeof fxsave->xmm_space);

	vcpu_put(vcpu);

	return 0;
}

A
Avi Kivity 已提交
2510 2511
static long kvm_vcpu_ioctl(struct file *filp,
			   unsigned int ioctl, unsigned long arg)
A
Avi Kivity 已提交
2512
{
A
Avi Kivity 已提交
2513
	struct kvm_vcpu *vcpu = filp->private_data;
A
Al Viro 已提交
2514
	void __user *argp = (void __user *)arg;
A
Avi Kivity 已提交
2515 2516 2517
	int r = -EINVAL;

	switch (ioctl) {
2518
	case KVM_RUN:
2519 2520 2521
		r = -EINVAL;
		if (arg)
			goto out;
2522
		r = kvm_vcpu_ioctl_run(vcpu, vcpu->run);
A
Avi Kivity 已提交
2523 2524 2525 2526
		break;
	case KVM_GET_REGS: {
		struct kvm_regs kvm_regs;

A
Avi Kivity 已提交
2527 2528
		memset(&kvm_regs, 0, sizeof kvm_regs);
		r = kvm_vcpu_ioctl_get_regs(vcpu, &kvm_regs);
A
Avi Kivity 已提交
2529 2530 2531
		if (r)
			goto out;
		r = -EFAULT;
A
Al Viro 已提交
2532
		if (copy_to_user(argp, &kvm_regs, sizeof kvm_regs))
A
Avi Kivity 已提交
2533 2534 2535 2536 2537 2538 2539 2540
			goto out;
		r = 0;
		break;
	}
	case KVM_SET_REGS: {
		struct kvm_regs kvm_regs;

		r = -EFAULT;
A
Al Viro 已提交
2541
		if (copy_from_user(&kvm_regs, argp, sizeof kvm_regs))
A
Avi Kivity 已提交
2542
			goto out;
A
Avi Kivity 已提交
2543
		r = kvm_vcpu_ioctl_set_regs(vcpu, &kvm_regs);
A
Avi Kivity 已提交
2544 2545 2546 2547 2548 2549 2550 2551
		if (r)
			goto out;
		r = 0;
		break;
	}
	case KVM_GET_SREGS: {
		struct kvm_sregs kvm_sregs;

A
Avi Kivity 已提交
2552 2553
		memset(&kvm_sregs, 0, sizeof kvm_sregs);
		r = kvm_vcpu_ioctl_get_sregs(vcpu, &kvm_sregs);
A
Avi Kivity 已提交
2554 2555 2556
		if (r)
			goto out;
		r = -EFAULT;
A
Al Viro 已提交
2557
		if (copy_to_user(argp, &kvm_sregs, sizeof kvm_sregs))
A
Avi Kivity 已提交
2558 2559 2560 2561 2562 2563 2564 2565
			goto out;
		r = 0;
		break;
	}
	case KVM_SET_SREGS: {
		struct kvm_sregs kvm_sregs;

		r = -EFAULT;
A
Al Viro 已提交
2566
		if (copy_from_user(&kvm_sregs, argp, sizeof kvm_sregs))
A
Avi Kivity 已提交
2567
			goto out;
A
Avi Kivity 已提交
2568
		r = kvm_vcpu_ioctl_set_sregs(vcpu, &kvm_sregs);
A
Avi Kivity 已提交
2569 2570 2571 2572 2573 2574 2575 2576 2577
		if (r)
			goto out;
		r = 0;
		break;
	}
	case KVM_TRANSLATE: {
		struct kvm_translation tr;

		r = -EFAULT;
A
Al Viro 已提交
2578
		if (copy_from_user(&tr, argp, sizeof tr))
A
Avi Kivity 已提交
2579
			goto out;
A
Avi Kivity 已提交
2580
		r = kvm_vcpu_ioctl_translate(vcpu, &tr);
A
Avi Kivity 已提交
2581 2582 2583
		if (r)
			goto out;
		r = -EFAULT;
A
Al Viro 已提交
2584
		if (copy_to_user(argp, &tr, sizeof tr))
A
Avi Kivity 已提交
2585 2586 2587 2588 2589 2590 2591 2592
			goto out;
		r = 0;
		break;
	}
	case KVM_INTERRUPT: {
		struct kvm_interrupt irq;

		r = -EFAULT;
A
Al Viro 已提交
2593
		if (copy_from_user(&irq, argp, sizeof irq))
A
Avi Kivity 已提交
2594
			goto out;
A
Avi Kivity 已提交
2595
		r = kvm_vcpu_ioctl_interrupt(vcpu, &irq);
A
Avi Kivity 已提交
2596 2597 2598 2599 2600 2601 2602 2603 2604
		if (r)
			goto out;
		r = 0;
		break;
	}
	case KVM_DEBUG_GUEST: {
		struct kvm_debug_guest dbg;

		r = -EFAULT;
A
Al Viro 已提交
2605
		if (copy_from_user(&dbg, argp, sizeof dbg))
A
Avi Kivity 已提交
2606
			goto out;
A
Avi Kivity 已提交
2607
		r = kvm_vcpu_ioctl_debug_guest(vcpu, &dbg);
A
Avi Kivity 已提交
2608 2609 2610 2611 2612
		if (r)
			goto out;
		r = 0;
		break;
	}
A
Avi Kivity 已提交
2613 2614 2615 2616 2617 2618
	case KVM_GET_MSRS:
		r = msr_io(vcpu, argp, get_msr, 1);
		break;
	case KVM_SET_MSRS:
		r = msr_io(vcpu, argp, do_set_msr, 0);
		break;
2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630
	case KVM_SET_CPUID: {
		struct kvm_cpuid __user *cpuid_arg = argp;
		struct kvm_cpuid cpuid;

		r = -EFAULT;
		if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid))
			goto out;
		r = kvm_vcpu_ioctl_set_cpuid(vcpu, &cpuid, cpuid_arg->entries);
		if (r)
			goto out;
		break;
	}
A
Avi Kivity 已提交
2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653
	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
2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678
	case KVM_GET_FPU: {
		struct kvm_fpu fpu;

		memset(&fpu, 0, sizeof fpu);
		r = kvm_vcpu_ioctl_get_fpu(vcpu, &fpu);
		if (r)
			goto out;
		r = -EFAULT;
		if (copy_to_user(argp, &fpu, sizeof fpu))
			goto out;
		r = 0;
		break;
	}
	case KVM_SET_FPU: {
		struct kvm_fpu fpu;

		r = -EFAULT;
		if (copy_from_user(&fpu, argp, sizeof fpu))
			goto out;
		r = kvm_vcpu_ioctl_set_fpu(vcpu, &fpu);
		if (r)
			goto out;
		r = 0;
		break;
	}
A
Avi Kivity 已提交
2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698
	default:
		;
	}
out:
	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;
	int r = -EINVAL;

	switch (ioctl) {
	case KVM_CREATE_VCPU:
		r = kvm_vm_ioctl_create_vcpu(kvm, arg);
		if (r < 0)
			goto out;
		break;
A
Avi Kivity 已提交
2699 2700 2701 2702
	case KVM_SET_MEMORY_REGION: {
		struct kvm_memory_region kvm_mem;

		r = -EFAULT;
A
Al Viro 已提交
2703
		if (copy_from_user(&kvm_mem, argp, sizeof kvm_mem))
A
Avi Kivity 已提交
2704
			goto out;
2705
		r = kvm_vm_ioctl_set_memory_region(kvm, &kvm_mem);
A
Avi Kivity 已提交
2706 2707 2708 2709 2710 2711 2712 2713
		if (r)
			goto out;
		break;
	}
	case KVM_GET_DIRTY_LOG: {
		struct kvm_dirty_log log;

		r = -EFAULT;
A
Al Viro 已提交
2714
		if (copy_from_user(&log, argp, sizeof log))
A
Avi Kivity 已提交
2715
			goto out;
2716
		r = kvm_vm_ioctl_get_dirty_log(kvm, &log);
A
Avi Kivity 已提交
2717 2718 2719 2720
		if (r)
			goto out;
		break;
	}
2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731
	case KVM_SET_MEMORY_ALIAS: {
		struct kvm_memory_alias alias;

		r = -EFAULT;
		if (copy_from_user(&alias, argp, sizeof alias))
			goto out;
		r = kvm_vm_ioctl_set_memory_alias(kvm, &alias);
		if (r)
			goto out;
		break;
	}
2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748
	default:
		;
	}
out:
	return r;
}

static struct page *kvm_vm_nopage(struct vm_area_struct *vma,
				  unsigned long address,
				  int *type)
{
	struct kvm *kvm = vma->vm_file->private_data;
	unsigned long pgoff;
	struct page *page;

	*type = VM_FAULT_MINOR;
	pgoff = ((address - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff;
A
Avi Kivity 已提交
2749
	page = gfn_to_page(kvm, pgoff);
2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796
	if (!page)
		return NOPAGE_SIGBUS;
	get_page(page);
	return page;
}

static struct vm_operations_struct kvm_vm_vm_ops = {
	.nopage = kvm_vm_nopage,
};

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

static struct file_operations kvm_vm_fops = {
	.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)
{
	int fd, r;
	struct inode *inode;
	struct file *file;
	struct kvm *kvm;

	inode = kvmfs_inode(&kvm_vm_fops);
	if (IS_ERR(inode)) {
		r = PTR_ERR(inode);
		goto out1;
	}

	kvm = kvm_create_vm();
	if (IS_ERR(kvm)) {
		r = PTR_ERR(kvm);
		goto out2;
	}

	file = kvmfs_file(inode, kvm);
	if (IS_ERR(file)) {
		r = PTR_ERR(file);
		goto out3;
	}
A
Avi Kivity 已提交
2797
	kvm->filp = file;
2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820

	r = get_unused_fd();
	if (r < 0)
		goto out4;
	fd = r;
	fd_install(fd, file);

	return fd;

out4:
	fput(file);
out3:
	kvm_destroy_vm(kvm);
out2:
	iput(inode);
out1:
	return r;
}

static long kvm_dev_ioctl(struct file *filp,
			  unsigned int ioctl, unsigned long arg)
{
	void __user *argp = (void __user *)arg;
2821
	long r = -EINVAL;
2822 2823 2824

	switch (ioctl) {
	case KVM_GET_API_VERSION:
2825 2826 2827
		r = -EINVAL;
		if (arg)
			goto out;
2828 2829 2830
		r = KVM_API_VERSION;
		break;
	case KVM_CREATE_VM:
2831 2832 2833
		r = -EINVAL;
		if (arg)
			goto out;
2834 2835
		r = kvm_dev_ioctl_create_vm();
		break;
A
Avi Kivity 已提交
2836
	case KVM_GET_MSR_INDEX_LIST: {
A
Al Viro 已提交
2837
		struct kvm_msr_list __user *user_msr_list = argp;
A
Avi Kivity 已提交
2838 2839 2840 2841 2842 2843 2844
		struct kvm_msr_list msr_list;
		unsigned n;

		r = -EFAULT;
		if (copy_from_user(&msr_list, user_msr_list, sizeof msr_list))
			goto out;
		n = msr_list.nmsrs;
2845
		msr_list.nmsrs = num_msrs_to_save + ARRAY_SIZE(emulated_msrs);
A
Avi Kivity 已提交
2846 2847 2848
		if (copy_to_user(user_msr_list, &msr_list, sizeof msr_list))
			goto out;
		r = -E2BIG;
2849
		if (n < num_msrs_to_save)
A
Avi Kivity 已提交
2850 2851 2852
			goto out;
		r = -EFAULT;
		if (copy_to_user(user_msr_list->indices, &msrs_to_save,
2853
				 num_msrs_to_save * sizeof(u32)))
A
Avi Kivity 已提交
2854
			goto out;
2855 2856 2857 2858 2859
		if (copy_to_user(user_msr_list->indices
				 + num_msrs_to_save * sizeof(u32),
				 &emulated_msrs,
				 ARRAY_SIZE(emulated_msrs) * sizeof(u32)))
			goto out;
A
Avi Kivity 已提交
2860
		r = 0;
A
Avi Kivity 已提交
2861
		break;
A
Avi Kivity 已提交
2862
	}
2863 2864 2865 2866 2867 2868
	case KVM_CHECK_EXTENSION:
		/*
		 * No extensions defined at present.
		 */
		r = 0;
		break;
2869 2870 2871 2872
	case KVM_GET_VCPU_MMAP_SIZE:
		r = -EINVAL;
		if (arg)
			goto out;
2873
		r = 2 * PAGE_SIZE;
2874
		break;
A
Avi Kivity 已提交
2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889
	default:
		;
	}
out:
	return r;
}

static struct file_operations kvm_chardev_ops = {
	.open		= kvm_dev_open,
	.release        = kvm_dev_release,
	.unlocked_ioctl = kvm_dev_ioctl,
	.compat_ioctl   = kvm_dev_ioctl,
};

static struct miscdevice kvm_dev = {
A
Avi Kivity 已提交
2890
	KVM_MINOR,
A
Avi Kivity 已提交
2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903
	"kvm",
	&kvm_chardev_ops,
};

static int kvm_reboot(struct notifier_block *notifier, unsigned long val,
                       void *v)
{
	if (val == SYS_RESTART) {
		/*
		 * Some (well, at least mine) BIOSes hang on reboot if
		 * in vmx root mode.
		 */
		printk(KERN_INFO "kvm: exiting hardware virtualization\n");
A
Al Viro 已提交
2904
		on_each_cpu(kvm_arch_ops->hardware_disable, NULL, 0, 1);
A
Avi Kivity 已提交
2905 2906 2907 2908 2909 2910 2911 2912 2913
	}
	return NOTIFY_OK;
}

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

A
Avi Kivity 已提交
2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952
/*
 * Make sure that a cpu that is being hot-unplugged does not have any vcpus
 * cached on it.
 */
static void decache_vcpus_on_cpu(int cpu)
{
	struct kvm *vm;
	struct kvm_vcpu *vcpu;
	int i;

	spin_lock(&kvm_lock);
	list_for_each_entry(vm, &vm_list, vm_list)
		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
			vcpu = &vm->vcpus[i];
			/*
			 * If the vcpu is locked, then it is running on some
			 * other cpu and therefore it is not cached on the
			 * cpu in question.
			 *
			 * If it's not locked, check the last cpu it executed
			 * on.
			 */
			if (mutex_trylock(&vcpu->mutex)) {
				if (vcpu->cpu == cpu) {
					kvm_arch_ops->vcpu_decache(vcpu);
					vcpu->cpu = -1;
				}
				mutex_unlock(&vcpu->mutex);
			}
		}
	spin_unlock(&kvm_lock);
}

static int kvm_cpu_hotplug(struct notifier_block *notifier, unsigned long val,
			   void *v)
{
	int cpu = (long)v;

	switch (val) {
2953
	case CPU_DOWN_PREPARE:
2954
	case CPU_DOWN_PREPARE_FROZEN:
A
Avi Kivity 已提交
2955
	case CPU_UP_CANCELED:
2956
	case CPU_UP_CANCELED_FROZEN:
2957 2958
		printk(KERN_INFO "kvm: disabling virtualization on CPU%d\n",
		       cpu);
A
Avi Kivity 已提交
2959 2960 2961 2962
		decache_vcpus_on_cpu(cpu);
		smp_call_function_single(cpu, kvm_arch_ops->hardware_disable,
					 NULL, 0, 1);
		break;
2963
	case CPU_ONLINE:
2964
	case CPU_ONLINE_FROZEN:
2965 2966
		printk(KERN_INFO "kvm: enabling virtualization on CPU%d\n",
		       cpu);
A
Avi Kivity 已提交
2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978
		smp_call_function_single(cpu, kvm_arch_ops->hardware_enable,
					 NULL, 0, 1);
		break;
	}
	return NOTIFY_OK;
}

static struct notifier_block kvm_cpu_notifier = {
	.notifier_call = kvm_cpu_hotplug,
	.priority = 20, /* must be > scheduler priority */
};

A
Avi Kivity 已提交
2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002
static u64 stat_get(void *_offset)
{
	unsigned offset = (long)_offset;
	u64 total = 0;
	struct kvm *kvm;
	struct kvm_vcpu *vcpu;
	int i;

	spin_lock(&kvm_lock);
	list_for_each_entry(kvm, &vm_list, vm_list)
		for (i = 0; i < KVM_MAX_VCPUS; ++i) {
			vcpu = &kvm->vcpus[i];
			total += *(u32 *)((void *)vcpu + offset);
		}
	spin_unlock(&kvm_lock);
	return total;
}

static void stat_set(void *offset, u64 val)
{
}

DEFINE_SIMPLE_ATTRIBUTE(stat_fops, stat_get, stat_set, "%llu\n");

A
Avi Kivity 已提交
3003 3004 3005 3006
static __init void kvm_init_debug(void)
{
	struct kvm_stats_debugfs_item *p;

A
Al Viro 已提交
3007
	debugfs_dir = debugfs_create_dir("kvm", NULL);
A
Avi Kivity 已提交
3008
	for (p = debugfs_entries; p->name; ++p)
A
Avi Kivity 已提交
3009 3010 3011
		p->dentry = debugfs_create_file(p->name, 0444, debugfs_dir,
						(void *)(long)p->offset,
						&stat_fops);
A
Avi Kivity 已提交
3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022
}

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

	for (p = debugfs_entries; p->name; ++p)
		debugfs_remove(p->dentry);
	debugfs_remove(debugfs_dir);
}

3023 3024 3025
static int kvm_suspend(struct sys_device *dev, pm_message_t state)
{
	decache_vcpus_on_cpu(raw_smp_processor_id());
A
Avi Kivity 已提交
3026
	on_each_cpu(kvm_arch_ops->hardware_disable, NULL, 0, 1);
3027 3028 3029 3030 3031
	return 0;
}

static int kvm_resume(struct sys_device *dev)
{
A
Avi Kivity 已提交
3032
	on_each_cpu(kvm_arch_ops->hardware_enable, NULL, 0, 1);
3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046
	return 0;
}

static struct sysdev_class kvm_sysdev_class = {
	set_kset_name("kvm"),
	.suspend = kvm_suspend,
	.resume = kvm_resume,
};

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

A
Avi Kivity 已提交
3047 3048
hpa_t bad_page_address;

3049 3050 3051
static int kvmfs_get_sb(struct file_system_type *fs_type, int flags,
			const char *dev_name, void *data, struct vfsmount *mnt)
{
3052
	return get_sb_pseudo(fs_type, "kvm:", NULL, KVMFS_SUPER_MAGIC, mnt);
3053 3054 3055 3056 3057 3058 3059 3060
}

static struct file_system_type kvm_fs_type = {
	.name		= "kvmfs",
	.get_sb		= kvmfs_get_sb,
	.kill_sb	= kill_anon_super,
};

A
Avi Kivity 已提交
3061 3062 3063 3064
int kvm_init_arch(struct kvm_arch_ops *ops, struct module *module)
{
	int r;

3065 3066 3067 3068 3069
	if (kvm_arch_ops) {
		printk(KERN_ERR "kvm: already loaded the other module\n");
		return -EEXIST;
	}

3070
	if (!ops->cpu_has_kvm_support()) {
A
Avi Kivity 已提交
3071 3072 3073
		printk(KERN_ERR "kvm: no hardware support\n");
		return -EOPNOTSUPP;
	}
3074
	if (ops->disabled_by_bios()) {
A
Avi Kivity 已提交
3075 3076 3077 3078
		printk(KERN_ERR "kvm: disabled by bios\n");
		return -EOPNOTSUPP;
	}

3079 3080
	kvm_arch_ops = ops;

A
Avi Kivity 已提交
3081 3082
	r = kvm_arch_ops->hardware_setup();
	if (r < 0)
3083
		goto out;
A
Avi Kivity 已提交
3084

A
Al Viro 已提交
3085
	on_each_cpu(kvm_arch_ops->hardware_enable, NULL, 0, 1);
A
Avi Kivity 已提交
3086 3087 3088
	r = register_cpu_notifier(&kvm_cpu_notifier);
	if (r)
		goto out_free_1;
A
Avi Kivity 已提交
3089 3090
	register_reboot_notifier(&kvm_reboot_notifier);

3091 3092 3093 3094 3095 3096 3097 3098
	r = sysdev_class_register(&kvm_sysdev_class);
	if (r)
		goto out_free_2;

	r = sysdev_register(&kvm_sysdev);
	if (r)
		goto out_free_3;

A
Avi Kivity 已提交
3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109
	kvm_chardev_ops.owner = module;

	r = misc_register(&kvm_dev);
	if (r) {
		printk (KERN_ERR "kvm: misc device register failed\n");
		goto out_free;
	}

	return r;

out_free:
3110 3111 3112 3113
	sysdev_unregister(&kvm_sysdev);
out_free_3:
	sysdev_class_unregister(&kvm_sysdev_class);
out_free_2:
A
Avi Kivity 已提交
3114
	unregister_reboot_notifier(&kvm_reboot_notifier);
A
Avi Kivity 已提交
3115 3116
	unregister_cpu_notifier(&kvm_cpu_notifier);
out_free_1:
A
Al Viro 已提交
3117
	on_each_cpu(kvm_arch_ops->hardware_disable, NULL, 0, 1);
A
Avi Kivity 已提交
3118
	kvm_arch_ops->hardware_unsetup();
3119 3120
out:
	kvm_arch_ops = NULL;
A
Avi Kivity 已提交
3121 3122 3123 3124 3125 3126
	return r;
}

void kvm_exit_arch(void)
{
	misc_deregister(&kvm_dev);
3127 3128
	sysdev_unregister(&kvm_sysdev);
	sysdev_class_unregister(&kvm_sysdev_class);
A
Avi Kivity 已提交
3129
	unregister_reboot_notifier(&kvm_reboot_notifier);
3130
	unregister_cpu_notifier(&kvm_cpu_notifier);
A
Al Viro 已提交
3131
	on_each_cpu(kvm_arch_ops->hardware_disable, NULL, 0, 1);
A
Avi Kivity 已提交
3132
	kvm_arch_ops->hardware_unsetup();
3133
	kvm_arch_ops = NULL;
A
Avi Kivity 已提交
3134 3135 3136 3137 3138
}

static __init int kvm_init(void)
{
	static struct page *bad_page;
3139 3140
	int r;

3141 3142 3143 3144
	r = kvm_mmu_module_init();
	if (r)
		goto out4;

3145 3146 3147
	r = register_filesystem(&kvm_fs_type);
	if (r)
		goto out3;
A
Avi Kivity 已提交
3148

3149 3150 3151 3152
	kvmfs_mnt = kern_mount(&kvm_fs_type);
	r = PTR_ERR(kvmfs_mnt);
	if (IS_ERR(kvmfs_mnt))
		goto out2;
A
Avi Kivity 已提交
3153 3154
	kvm_init_debug();

3155 3156
	kvm_init_msr_list();

A
Avi Kivity 已提交
3157 3158 3159 3160 3161 3162 3163 3164
	if ((bad_page = alloc_page(GFP_KERNEL)) == NULL) {
		r = -ENOMEM;
		goto out;
	}

	bad_page_address = page_to_pfn(bad_page) << PAGE_SHIFT;
	memset(__va(bad_page_address), 0, PAGE_SIZE);

3165
	return 0;
A
Avi Kivity 已提交
3166 3167 3168

out:
	kvm_exit_debug();
3169 3170 3171 3172
	mntput(kvmfs_mnt);
out2:
	unregister_filesystem(&kvm_fs_type);
out3:
3173 3174
	kvm_mmu_module_exit();
out4:
A
Avi Kivity 已提交
3175 3176 3177 3178 3179 3180 3181
	return r;
}

static __exit void kvm_exit(void)
{
	kvm_exit_debug();
	__free_page(pfn_to_page(bad_page_address >> PAGE_SHIFT));
3182 3183
	mntput(kvmfs_mnt);
	unregister_filesystem(&kvm_fs_type);
3184
	kvm_mmu_module_exit();
A
Avi Kivity 已提交
3185 3186 3187 3188 3189 3190 3191
}

module_init(kvm_init)
module_exit(kvm_exit)

EXPORT_SYMBOL_GPL(kvm_init_arch);
EXPORT_SYMBOL_GPL(kvm_exit_arch);