kvm_host.h 25.6 KB
Newer Older
1 2
#ifndef __KVM_HOST_H
#define __KVM_HOST_H
A
Avi Kivity 已提交
3 4 5 6 7 8 9

/*
 * This work is licensed under the terms of the GNU GPL, version 2.  See
 * the COPYING file in the top-level directory.
 */

#include <linux/types.h>
10
#include <linux/hardirq.h>
A
Avi Kivity 已提交
11 12 13
#include <linux/list.h>
#include <linux/mutex.h>
#include <linux/spinlock.h>
M
Markus Rechberger 已提交
14 15
#include <linux/signal.h>
#include <linux/sched.h>
16
#include <linux/bug.h>
A
Avi Kivity 已提交
17
#include <linux/mm.h>
18
#include <linux/mmu_notifier.h>
19
#include <linux/preempt.h>
20
#include <linux/msi.h>
21
#include <linux/slab.h>
22
#include <linux/rcupdate.h>
23
#include <linux/ratelimit.h>
X
Xiao Guangrong 已提交
24
#include <linux/err.h>
A
Alexey Dobriyan 已提交
25
#include <asm/signal.h>
A
Avi Kivity 已提交
26 27

#include <linux/kvm.h>
I
Ingo Molnar 已提交
28
#include <linux/kvm_para.h>
A
Avi Kivity 已提交
29

30
#include <linux/kvm_types.h>
31

32
#include <asm/kvm_host.h>
33

A
Avi Kivity 已提交
34 35 36 37
#ifndef KVM_MMIO_SIZE
#define KVM_MMIO_SIZE 8
#endif

A
Avi Kivity 已提交
38 39 40 41 42 43 44 45 46 47 48 49 50 51
/*
 * If we support unaligned MMIO, at most one fragment will be split into two:
 */
#ifdef KVM_UNALIGNED_MMIO
#  define KVM_EXTRA_MMIO_FRAGMENTS 1
#else
#  define KVM_EXTRA_MMIO_FRAGMENTS 0
#endif

#define KVM_USER_MMIO_SIZE 8

#define KVM_MAX_MMIO_FRAGMENTS \
	(KVM_MMIO_SIZE / KVM_USER_MMIO_SIZE + KVM_EXTRA_MMIO_FRAGMENTS)

52
#define KVM_PFN_ERR_FAULT	(-EFAULT)
53
#define KVM_PFN_ERR_HWPOISON	(-EHWPOISON)
X
Xiao Guangrong 已提交
54
#define KVM_PFN_ERR_BAD		(-ENOENT)
55

X
Xiao Guangrong 已提交
56 57 58 59 60 61 62 63 64 65 66 67 68 69 70
static inline int is_error_pfn(pfn_t pfn)
{
	return IS_ERR_VALUE(pfn);
}

static inline int is_noslot_pfn(pfn_t pfn)
{
	return pfn == -ENOENT;
}

static inline int is_invalid_pfn(pfn_t pfn)
{
	return !is_noslot_pfn(pfn) && is_error_pfn(pfn);
}

71 72 73
/*
 * vcpu->requests bit members
 */
74
#define KVM_REQ_TLB_FLUSH          0
75
#define KVM_REQ_MIGRATE_TIMER      1
76
#define KVM_REQ_REPORT_TPR_ACCESS  2
77
#define KVM_REQ_MMU_RELOAD         3
J
Joerg Roedel 已提交
78
#define KVM_REQ_TRIPLE_FAULT       4
79
#define KVM_REQ_PENDING_TIMER      5
80
#define KVM_REQ_UNHALT             6
81
#define KVM_REQ_MMU_SYNC           7
Z
Zachary Amsden 已提交
82
#define KVM_REQ_CLOCK_UPDATE       8
83
#define KVM_REQ_KICK               9
84
#define KVM_REQ_DEACTIVATE_FPU    10
85
#define KVM_REQ_EVENT             11
86
#define KVM_REQ_APF_HALT          12
G
Glauber Costa 已提交
87
#define KVM_REQ_STEAL_UPDATE      13
A
Avi Kivity 已提交
88
#define KVM_REQ_NMI               14
89
#define KVM_REQ_IMMEDIATE_EXIT    15
90 91
#define KVM_REQ_PMU               16
#define KVM_REQ_PMI               17
A
Avi Kivity 已提交
92

93 94
#define KVM_USERSPACE_IRQ_SOURCE_ID	0

95
struct kvm;
A
Avi Kivity 已提交
96
struct kvm_vcpu;
97
extern struct kmem_cache *kvm_vcpu_cache;
A
Avi Kivity 已提交
98

99 100 101 102 103 104
struct kvm_io_range {
	gpa_t addr;
	int len;
	struct kvm_io_device *dev;
};

105
#define NR_IOBUS_DEVS 1000
106

107 108
struct kvm_io_bus {
	int                   dev_count;
109
	struct kvm_io_range range[];
110 111
};

M
Marcelo Tosatti 已提交
112 113 114 115 116 117 118 119 120
enum kvm_bus {
	KVM_MMIO_BUS,
	KVM_PIO_BUS,
	KVM_NR_BUSES
};

int kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
		     int len, const void *val);
int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, int len,
121
		    void *val);
122 123
int kvm_io_bus_register_dev(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
			    int len, struct kvm_io_device *dev);
M
Marcelo Tosatti 已提交
124 125
int kvm_io_bus_unregister_dev(struct kvm *kvm, enum kvm_bus bus_idx,
			      struct kvm_io_device *dev);
126

127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
#ifdef CONFIG_KVM_ASYNC_PF
struct kvm_async_pf {
	struct work_struct work;
	struct list_head link;
	struct list_head queue;
	struct kvm_vcpu *vcpu;
	struct mm_struct *mm;
	gva_t gva;
	unsigned long addr;
	struct kvm_arch_async_pf arch;
	struct page *page;
	bool done;
};

void kvm_clear_async_pf_completion_queue(struct kvm_vcpu *vcpu);
void kvm_check_async_pf_completion(struct kvm_vcpu *vcpu);
int kvm_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn,
		       struct kvm_arch_async_pf *arch);
145
int kvm_async_pf_wakeup_all(struct kvm_vcpu *vcpu);
146 147
#endif

148 149 150
enum {
	OUTSIDE_GUEST_MODE,
	IN_GUEST_MODE,
151 152
	EXITING_GUEST_MODE,
	READING_SHADOW_PAGE_TABLES,
153 154
};

A
Avi Kivity 已提交
155 156 157 158 159 160 161 162 163 164
/*
 * Sometimes a large or cross-page mmio needs to be broken up into separate
 * exits for userspace servicing.
 */
struct kvm_mmio_fragment {
	gpa_t gpa;
	void *data;
	unsigned len;
};

165 166
struct kvm_vcpu {
	struct kvm *kvm;
167
#ifdef CONFIG_PREEMPT_NOTIFIERS
168
	struct preempt_notifier preempt_notifier;
169
#endif
170
	int cpu;
171
	int vcpu_id;
172 173
	int srcu_idx;
	int mode;
174
	unsigned long requests;
J
Jan Kiszka 已提交
175
	unsigned long guest_debug;
176 177 178

	struct mutex mutex;
	struct kvm_run *run;
179

180
	int fpu_active;
181
	int guest_fpu_loaded, guest_xcr0_loaded;
182
	wait_queue_head_t wq;
183
	struct pid *pid;
184 185 186 187
	int sigset_active;
	sigset_t sigset;
	struct kvm_vcpu_stat stat;

188
#ifdef CONFIG_HAS_IOMEM
189 190 191
	int mmio_needed;
	int mmio_read_completed;
	int mmio_is_write;
A
Avi Kivity 已提交
192 193 194
	int mmio_cur_fragment;
	int mmio_nr_fragments;
	struct kvm_mmio_fragment mmio_fragments[KVM_MAX_MMIO_FRAGMENTS];
195
#endif
A
Avi Kivity 已提交
196

197 198 199 200 201 202 203 204 205
#ifdef CONFIG_KVM_ASYNC_PF
	struct {
		u32 queued;
		struct list_head queue;
		struct list_head done;
		spinlock_t lock;
	} async_pf;
#endif

206 207 208 209 210 211 212 213 214 215 216 217
#ifdef CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT
	/*
	 * Cpu relax intercept or pause loop exit optimization
	 * in_spin_loop: set when a vcpu does a pause loop exit
	 *  or cpu relax intercepted.
	 * dy_eligible: indicates whether vcpu is eligible for directed yield.
	 */
	struct {
		bool in_spin_loop;
		bool dy_eligible;
	} spin_loop;
#endif
218 219 220
	struct kvm_vcpu_arch arch;
};

221 222 223 224 225
static inline int kvm_vcpu_exiting_guest_mode(struct kvm_vcpu *vcpu)
{
	return cmpxchg(&vcpu->mode, IN_GUEST_MODE, EXITING_GUEST_MODE);
}

226 227 228 229 230 231
/*
 * Some of the bitops functions do not support too long bitmaps.
 * This number must be determined not to exceed such limits.
 */
#define KVM_MEM_MAX_NR_PAGES ((1UL << 31) - 1)

A
Avi Kivity 已提交
232 233 234 235 236
struct kvm_memory_slot {
	gfn_t base_gfn;
	unsigned long npages;
	unsigned long flags;
	unsigned long *dirty_bitmap;
237
	struct kvm_arch_memory_slot arch;
238
	unsigned long userspace_addr;
239
	int user_alloc;
240
	int id;
A
Avi Kivity 已提交
241 242
};

243 244 245 246 247
static inline unsigned long kvm_dirty_bitmap_bytes(struct kvm_memory_slot *memslot)
{
	return ALIGN(memslot->npages, BITS_PER_LONG) / 8;
}

248 249
struct kvm_kernel_irq_routing_entry {
	u32 gsi;
250
	u32 type;
251
	int (*set)(struct kvm_kernel_irq_routing_entry *e,
252
		   struct kvm *kvm, int irq_source_id, int level);
253 254 255 256 257
	union {
		struct {
			unsigned irqchip;
			unsigned pin;
		} irqchip;
S
Sheng Yang 已提交
258
		struct msi_msg msi;
259
	};
260 261 262
	struct hlist_node link;
};

263 264
#ifdef __KVM_HAVE_IOAPIC

265
struct kvm_irq_routing_table {
266
	int chip[KVM_NR_IRQCHIPS][KVM_IOAPIC_NUM_PINS];
267 268 269 270 271 272 273
	struct kvm_kernel_irq_routing_entry *rt_entries;
	u32 nr_rt_entries;
	/*
	 * Array indexed by gsi. Each entry contains list of irq chips
	 * the gsi is connected to.
	 */
	struct hlist_head map[0];
274 275
};

276 277 278 279 280 281
#else

struct kvm_irq_routing_table {};

#endif

282 283 284 285
#ifndef KVM_MEM_SLOTS_NUM
#define KVM_MEM_SLOTS_NUM (KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS)
#endif

286 287 288 289 290
/*
 * Note:
 * memslots are not sorted by id anymore, please use id_to_memslot()
 * to get the memslot by its id.
 */
291
struct kvm_memslots {
292
	u64 generation;
293
	struct kvm_memory_slot memslots[KVM_MEM_SLOTS_NUM];
294 295
	/* The mapping table from slot id to the index in memslots[]. */
	int id_to_index[KVM_MEM_SLOTS_NUM];
296 297
};

A
Avi Kivity 已提交
298
struct kvm {
299
	spinlock_t mmu_lock;
300
	struct mutex slots_lock;
301
	struct mm_struct *mm; /* userspace tied to this vm */
302
	struct kvm_memslots *memslots;
303
	struct srcu_struct srcu;
304 305 306
#ifdef CONFIG_KVM_APIC_ARCHITECTURE
	u32 bsp_vcpu_id;
#endif
R
Rusty Russell 已提交
307
	struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];
308
	atomic_t online_vcpus;
309
	int last_boosted_vcpu;
310
	struct list_head vm_list;
311
	struct mutex lock;
M
Marcelo Tosatti 已提交
312
	struct kvm_io_bus *buses[KVM_NR_BUSES];
G
Gregory Haskins 已提交
313 314 315 316 317
#ifdef CONFIG_HAVE_KVM_EVENTFD
	struct {
		spinlock_t        lock;
		struct list_head  items;
	} irqfds;
G
Gregory Haskins 已提交
318
	struct list_head ioeventfds;
G
Gregory Haskins 已提交
319
#endif
320
	struct kvm_vm_stat stat;
321
	struct kvm_arch arch;
I
Izik Eidus 已提交
322
	atomic_t users_count;
323 324
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
	struct kvm_coalesced_mmio_ring *coalesced_mmio_ring;
325 326
	spinlock_t ring_lock;
	struct list_head coalesced_zones;
327
#endif
328

329
	struct mutex irq_lock;
330
#ifdef CONFIG_HAVE_KVM_IRQCHIP
331 332 333 334
	/*
	 * Update side is protected by irq_lock and,
	 * if configured, irqfds.lock.
	 */
A
Arnd Bergmann 已提交
335
	struct kvm_irq_routing_table __rcu *irq_routing;
336
	struct hlist_head mask_notifier_list;
337
	struct hlist_head irq_ack_notifier_list;
338 339
#endif

340
#if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER)
341 342 343 344
	struct mmu_notifier mmu_notifier;
	unsigned long mmu_notifier_seq;
	long mmu_notifier_count;
#endif
345
	long tlbs_dirty;
A
Avi Kivity 已提交
346 347
};

348 349 350 351 352 353 354 355 356
#define kvm_err(fmt, ...) \
	pr_err("kvm [%i]: " fmt, task_pid_nr(current), ## __VA_ARGS__)
#define kvm_info(fmt, ...) \
	pr_info("kvm [%i]: " fmt, task_pid_nr(current), ## __VA_ARGS__)
#define kvm_debug(fmt, ...) \
	pr_debug("kvm [%i]: " fmt, task_pid_nr(current), ## __VA_ARGS__)
#define kvm_pr_unimpl(fmt, ...) \
	pr_err_ratelimited("kvm [%i]: " fmt, \
			   task_tgid_nr(current), ## __VA_ARGS__)
357

358 359 360
/* The guest did something we don't support. */
#define vcpu_unimpl(vcpu, fmt, ...)					\
	kvm_pr_unimpl("vcpu%i " fmt, (vcpu)->vcpu_id, ## __VA_ARGS__)
A
Avi Kivity 已提交
361

362 363 364 365 366 367 368
static inline struct kvm_vcpu *kvm_get_vcpu(struct kvm *kvm, int i)
{
	smp_rmb();
	return kvm->vcpus[i];
}

#define kvm_for_each_vcpu(idx, vcpup, kvm) \
369 370 371 372
	for (idx = 0; \
	     idx < atomic_read(&kvm->online_vcpus) && \
	     (vcpup = kvm_get_vcpu(kvm, idx)) != NULL; \
	     idx++)
373

374 375
#define kvm_for_each_memslot(memslot, slots)	\
	for (memslot = &slots->memslots[0];	\
376 377
	      memslot < slots->memslots + KVM_MEM_SLOTS_NUM && memslot->npages;\
		memslot++)
378

R
Rusty Russell 已提交
379 380 381
int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id);
void kvm_vcpu_uninit(struct kvm_vcpu *vcpu);

382 383 384
void vcpu_load(struct kvm_vcpu *vcpu);
void vcpu_put(struct kvm_vcpu *vcpu);

385
int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align,
386
		  struct module *module);
387
void kvm_exit(void);
A
Avi Kivity 已提交
388

I
Izik Eidus 已提交
389 390
void kvm_get_kvm(struct kvm *kvm);
void kvm_put_kvm(struct kvm *kvm);
391
void update_memslots(struct kvm_memslots *slots, struct kvm_memory_slot *new);
I
Izik Eidus 已提交
392

393 394 395 396 397 398 399
static inline struct kvm_memslots *kvm_memslots(struct kvm *kvm)
{
	return rcu_dereference_check(kvm->memslots,
			srcu_read_lock_held(&kvm->srcu)
			|| lockdep_is_held(&kvm->slots_lock));
}

400 401 402
static inline struct kvm_memory_slot *
id_to_memslot(struct kvm_memslots *slots, int id)
{
403 404
	int index = slots->id_to_index[id];
	struct kvm_memory_slot *slot;
405

406
	slot = &slots->memslots[index];
407

408 409
	WARN_ON(slot->id != id);
	return slot;
410 411
}

412
int is_error_page(struct page *page);
I
Izik Eidus 已提交
413
int kvm_is_error_hva(unsigned long addr);
414 415 416
int kvm_set_memory_region(struct kvm *kvm,
			  struct kvm_userspace_memory_region *mem,
			  int user_alloc);
417 418 419
int __kvm_set_memory_region(struct kvm *kvm,
			    struct kvm_userspace_memory_region *mem,
			    int user_alloc);
420 421 422
void kvm_arch_free_memslot(struct kvm_memory_slot *free,
			   struct kvm_memory_slot *dont);
int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages);
423 424 425 426 427 428
int kvm_arch_prepare_memory_region(struct kvm *kvm,
				struct kvm_memory_slot *memslot,
				struct kvm_memory_slot old,
				struct kvm_userspace_memory_region *mem,
				int user_alloc);
void kvm_arch_commit_memory_region(struct kvm *kvm,
429 430 431
				struct kvm_userspace_memory_region *mem,
				struct kvm_memory_slot old,
				int user_alloc);
432
bool kvm_largepages_enabled(void);
433
void kvm_disable_largepages(void);
434
void kvm_arch_flush_shadow(struct kvm *kvm);
435

436 437 438
int gfn_to_page_many_atomic(struct kvm *kvm, gfn_t gfn, struct page **pages,
			    int nr_pages);

X
Xiao Guangrong 已提交
439
struct page *get_bad_page(void);
A
Avi Kivity 已提交
440
struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn);
M
Marcelo Tosatti 已提交
441
unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn);
442 443
void kvm_release_page_clean(struct page *page);
void kvm_release_page_dirty(struct page *page);
444 445 446
void kvm_set_page_dirty(struct page *page);
void kvm_set_page_accessed(struct page *page);

447
pfn_t hva_to_pfn_atomic(unsigned long addr);
448
pfn_t gfn_to_pfn_atomic(struct kvm *kvm, gfn_t gfn);
449 450
pfn_t gfn_to_pfn_async(struct kvm *kvm, gfn_t gfn, bool *async,
		       bool write_fault, bool *writable);
451
pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn);
452 453
pfn_t gfn_to_pfn_prot(struct kvm *kvm, gfn_t gfn, bool write_fault,
		      bool *writable);
454
pfn_t gfn_to_pfn_memslot(struct kvm_memory_slot *slot, gfn_t gfn);
455 456 457 458 459 460
void kvm_release_pfn_dirty(pfn_t);
void kvm_release_pfn_clean(pfn_t pfn);
void kvm_set_pfn_dirty(pfn_t pfn);
void kvm_set_pfn_accessed(pfn_t pfn);
void kvm_get_pfn(pfn_t pfn);

461 462
int kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset,
			int len);
463 464
int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
			  unsigned long len);
465
int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len);
466 467
int kvm_read_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
			   void *data, unsigned long len);
468 469 470 471
int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn, const void *data,
			 int offset, int len);
int kvm_write_guest(struct kvm *kvm, gpa_t gpa, const void *data,
		    unsigned long len);
472 473 474 475
int kvm_write_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
			   void *data, unsigned long len);
int kvm_gfn_to_hva_cache_init(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
			      gpa_t gpa);
476 477
int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len);
int kvm_clear_guest(struct kvm *kvm, gpa_t gpa, unsigned long len);
A
Avi Kivity 已提交
478
struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn);
479
int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn);
J
Joerg Roedel 已提交
480
unsigned long kvm_host_page_size(struct kvm *kvm, gfn_t gfn);
A
Avi Kivity 已提交
481
void mark_page_dirty(struct kvm *kvm, gfn_t gfn);
482 483
void mark_page_dirty_in_slot(struct kvm *kvm, struct kvm_memory_slot *memslot,
			     gfn_t gfn);
A
Avi Kivity 已提交
484

485
void kvm_vcpu_block(struct kvm_vcpu *vcpu);
486
void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
487
bool kvm_vcpu_yield_to(struct kvm_vcpu *target);
Z
Zhai, Edwin 已提交
488
void kvm_vcpu_on_spin(struct kvm_vcpu *vcpu);
A
Avi Kivity 已提交
489
void kvm_resched(struct kvm_vcpu *vcpu);
490 491
void kvm_load_guest_fpu(struct kvm_vcpu *vcpu);
void kvm_put_guest_fpu(struct kvm_vcpu *vcpu);
492

493
void kvm_flush_remote_tlbs(struct kvm *kvm);
494
void kvm_reload_remote_mmus(struct kvm *kvm);
A
Avi Kivity 已提交
495

496 497
long kvm_arch_dev_ioctl(struct file *filp,
			unsigned int ioctl, unsigned long arg);
498 499
long kvm_arch_vcpu_ioctl(struct file *filp,
			 unsigned int ioctl, unsigned long arg);
500
int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf);
501 502 503

int kvm_dev_ioctl_check_extension(long ext);

504 505 506 507 508
int kvm_get_dirty_log(struct kvm *kvm,
			struct kvm_dirty_log *log, int *is_dirty);
int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
				struct kvm_dirty_log *log);

509 510 511 512
int kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
				   struct
				   kvm_userspace_memory_region *mem,
				   int user_alloc);
513
int kvm_vm_ioctl_irq_line(struct kvm *kvm, struct kvm_irq_level *irq_level);
514 515
long kvm_arch_vm_ioctl(struct file *filp,
		       unsigned int ioctl, unsigned long arg);
516

517 518 519
int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu);
int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu);

520 521 522
int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
				    struct kvm_translation *tr);

523 524 525 526 527 528
int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs);
int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs);
int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs);
int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs);
529 530 531 532
int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state);
int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state);
J
Jan Kiszka 已提交
533 534
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
					struct kvm_guest_debug *dbg);
535 536
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run);

537 538
int kvm_arch_init(void *opaque);
void kvm_arch_exit(void);
539

540 541 542 543 544 545 546
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu);
void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu);

void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu);
void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu);
void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu);
struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id);
547
int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu);
548
void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu);
549 550

int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu);
551
int kvm_arch_hardware_enable(void *garbage);
552 553 554 555
void kvm_arch_hardware_disable(void *garbage);
int kvm_arch_hardware_setup(void);
void kvm_arch_hardware_unsetup(void);
void kvm_arch_check_processor_compat(void *rtn);
556
int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu);
557
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu);
558

559 560
void kvm_free_physmem(struct kvm *kvm);

561 562 563
void *kvm_kvzalloc(unsigned long size);
void kvm_kvfree(const void *addr);

564 565 566 567 568 569 570 571 572 573 574 575
#ifndef __KVM_HAVE_ARCH_VM_ALLOC
static inline struct kvm *kvm_arch_alloc_vm(void)
{
	return kzalloc(sizeof(struct kvm), GFP_KERNEL);
}

static inline void kvm_arch_free_vm(struct kvm *kvm)
{
	kfree(kvm);
}
#endif

576 577
static inline wait_queue_head_t *kvm_arch_vcpu_wq(struct kvm_vcpu *vcpu)
{
578 579 580
#ifdef __KVM_HAVE_ARCH_WQP
	return vcpu->arch.wqp;
#else
581 582
	return &vcpu->wq;
#endif
583
}
584

585
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type);
586
void kvm_arch_destroy_vm(struct kvm *kvm);
587
void kvm_free_all_assigned_devices(struct kvm *kvm);
588
void kvm_arch_sync_events(struct kvm *kvm);
589

590
int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu);
591
void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
592

X
Xiao Guangrong 已提交
593
bool kvm_is_mmio_pfn(pfn_t pfn);
594

B
Ben-Ami Yassour 已提交
595 596 597 598 599 600 601 602 603 604
struct kvm_irq_ack_notifier {
	struct hlist_node link;
	unsigned gsi;
	void (*irq_acked)(struct kvm_irq_ack_notifier *kian);
};

struct kvm_assigned_dev_kernel {
	struct kvm_irq_ack_notifier ack_notifier;
	struct list_head list;
	int assigned_dev_id;
605
	int host_segnr;
B
Ben-Ami Yassour 已提交
606 607
	int host_busnr;
	int host_devfn;
S
Sheng Yang 已提交
608
	unsigned int entries_nr;
B
Ben-Ami Yassour 已提交
609
	int host_irq;
610
	bool host_irq_disabled;
611
	bool pci_2_3;
S
Sheng Yang 已提交
612
	struct msix_entry *host_msix_entries;
B
Ben-Ami Yassour 已提交
613
	int guest_irq;
614
	struct msix_entry *guest_msix_entries;
615
	unsigned long irq_requested_type;
616
	int irq_source_id;
617
	int flags;
B
Ben-Ami Yassour 已提交
618 619
	struct pci_dev *dev;
	struct kvm *kvm;
620
	spinlock_t intx_lock;
621
	spinlock_t intx_mask_lock;
622
	char irq_name[32];
623
	struct pci_saved_state *pci_saved_state;
B
Ben-Ami Yassour 已提交
624
};
625 626 627 628 629 630 631 632 633 634 635

struct kvm_irq_mask_notifier {
	void (*func)(struct kvm_irq_mask_notifier *kimn, bool masked);
	int irq;
	struct hlist_node link;
};

void kvm_register_irq_mask_notifier(struct kvm *kvm, int irq,
				    struct kvm_irq_mask_notifier *kimn);
void kvm_unregister_irq_mask_notifier(struct kvm *kvm, int irq,
				      struct kvm_irq_mask_notifier *kimn);
636 637
void kvm_fire_mask_notifiers(struct kvm *kvm, unsigned irqchip, unsigned pin,
			     bool mask);
638

639 640 641 642 643 644
#ifdef __KVM_HAVE_IOAPIC
void kvm_get_intr_delivery_bitmask(struct kvm_ioapic *ioapic,
				   union kvm_ioapic_redirect_entry *entry,
				   unsigned long *deliver_bitmask);
#endif
int kvm_set_irq(struct kvm *kvm, int irq_source_id, u32 irq, int level);
645 646
int kvm_set_msi(struct kvm_kernel_irq_routing_entry *irq_entry, struct kvm *kvm,
		int irq_source_id, int level);
647
void kvm_notify_acked_irq(struct kvm *kvm, unsigned irqchip, unsigned pin);
648 649
void kvm_register_irq_ack_notifier(struct kvm *kvm,
				   struct kvm_irq_ack_notifier *kian);
650 651
void kvm_unregister_irq_ack_notifier(struct kvm *kvm,
				   struct kvm_irq_ack_notifier *kian);
652 653
int kvm_request_irq_source_id(struct kvm *kvm);
void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id);
B
Ben-Ami Yassour 已提交
654

655 656 657
/* For vcpu->arch.iommu_flags */
#define KVM_IOMMU_CACHE_COHERENCY	0x1

J
Joerg Roedel 已提交
658
#ifdef CONFIG_IOMMU_API
659
int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot);
660
void kvm_iommu_unmap_pages(struct kvm *kvm, struct kvm_memory_slot *slot);
W
Weidong Han 已提交
661
int kvm_iommu_map_guest(struct kvm *kvm);
B
Ben-Ami Yassour 已提交
662
int kvm_iommu_unmap_guest(struct kvm *kvm);
W
Weidong Han 已提交
663 664
int kvm_assign_device(struct kvm *kvm,
		      struct kvm_assigned_dev_kernel *assigned_dev);
W
Weidong Han 已提交
665 666
int kvm_deassign_device(struct kvm *kvm,
			struct kvm_assigned_dev_kernel *assigned_dev);
J
Joerg Roedel 已提交
667
#else /* CONFIG_IOMMU_API */
B
Ben-Ami Yassour 已提交
668
static inline int kvm_iommu_map_pages(struct kvm *kvm,
669
				      struct kvm_memory_slot *slot)
B
Ben-Ami Yassour 已提交
670 671 672 673
{
	return 0;
}

674 675 676 677 678
static inline void kvm_iommu_unmap_pages(struct kvm *kvm,
					 struct kvm_memory_slot *slot)
{
}

W
Weidong Han 已提交
679
static inline int kvm_iommu_map_guest(struct kvm *kvm)
B
Ben-Ami Yassour 已提交
680 681 682 683 684 685 686 687
{
	return -ENODEV;
}

static inline int kvm_iommu_unmap_guest(struct kvm *kvm)
{
	return 0;
}
W
Weidong Han 已提交
688 689 690 691 692 693

static inline int kvm_assign_device(struct kvm *kvm,
		struct kvm_assigned_dev_kernel *assigned_dev)
{
	return 0;
}
W
Weidong Han 已提交
694 695 696 697 698 699

static inline int kvm_deassign_device(struct kvm *kvm,
		struct kvm_assigned_dev_kernel *assigned_dev)
{
	return 0;
}
J
Joerg Roedel 已提交
700
#endif /* CONFIG_IOMMU_API */
B
Ben-Ami Yassour 已提交
701

702 703
static inline void kvm_guest_enter(void)
{
704
	BUG_ON(preemptible());
705
	account_system_vtime(current);
706
	current->flags |= PF_VCPU;
707 708 709 710 711 712 713 714
	/* KVM does not hold any references to rcu protected data when it
	 * switches CPU into a guest mode. In fact switching to a guest mode
	 * is very similar to exiting to userspase from rcu point of view. In
	 * addition CPU may stay in a guest mode for quite a long time (up to
	 * one time slice). Lets treat guest mode as quiescent state, just like
	 * we do with user-mode execution.
	 */
	rcu_virt_note_context_switch(smp_processor_id());
715 716 717 718
}

static inline void kvm_guest_exit(void)
{
719
	account_system_vtime(current);
720 721 722
	current->flags &= ~PF_VCPU;
}

723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747
/*
 * search_memslots() and __gfn_to_memslot() are here because they are
 * used in non-modular code in arch/powerpc/kvm/book3s_hv_rm_mmu.c.
 * gfn_to_memslot() itself isn't here as an inline because that would
 * bloat other code too much.
 */
static inline struct kvm_memory_slot *
search_memslots(struct kvm_memslots *slots, gfn_t gfn)
{
	struct kvm_memory_slot *memslot;

	kvm_for_each_memslot(memslot, slots)
		if (gfn >= memslot->base_gfn &&
		      gfn < memslot->base_gfn + memslot->npages)
			return memslot;

	return NULL;
}

static inline struct kvm_memory_slot *
__gfn_to_memslot(struct kvm_memslots *slots, gfn_t gfn)
{
	return search_memslots(slots, gfn);
}

748 749 750 751 752
static inline int memslot_id(struct kvm *kvm, gfn_t gfn)
{
	return gfn_to_memslot(kvm, gfn)->id;
}

753 754 755 756 757 758 759
static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level)
{
	/* KVM_HPAGE_GFN_SHIFT(PT_PAGE_TABLE_LEVEL) must be 0. */
	return (gfn >> KVM_HPAGE_GFN_SHIFT(level)) -
		(base_gfn >> KVM_HPAGE_GFN_SHIFT(level));
}

760 761 762 763 764 765 766 767
static inline gfn_t
hva_to_gfn_memslot(unsigned long hva, struct kvm_memory_slot *slot)
{
	gfn_t gfn_offset = (hva - slot->userspace_addr) >> PAGE_SHIFT;

	return slot->base_gfn + gfn_offset;
}

768 769 770 771 772 773
static inline unsigned long gfn_to_hva_memslot(struct kvm_memory_slot *slot,
					       gfn_t gfn)
{
	return slot->userspace_addr + (gfn - slot->base_gfn) * PAGE_SIZE;
}

A
Avi Kivity 已提交
774 775 776 777
static inline gpa_t gfn_to_gpa(gfn_t gfn)
{
	return (gpa_t)gfn << PAGE_SHIFT;
}
A
Avi Kivity 已提交
778

779 780 781 782 783
static inline gfn_t gpa_to_gfn(gpa_t gpa)
{
	return (gfn_t)(gpa >> PAGE_SHIFT);
}

B
Ben-Ami Yassour 已提交
784 785 786 787 788
static inline hpa_t pfn_to_hpa(pfn_t pfn)
{
	return (hpa_t)pfn << PAGE_SHIFT;
}

M
Marcelo Tosatti 已提交
789
static inline void kvm_migrate_timers(struct kvm_vcpu *vcpu)
790 791 792 793
{
	set_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests);
}

794 795 796 797 798
enum kvm_stat_kind {
	KVM_STAT_VM,
	KVM_STAT_VCPU,
};

799 800 801
struct kvm_stats_debugfs_item {
	const char *name;
	int offset;
802
	enum kvm_stat_kind kind;
803 804 805
	struct dentry *dentry;
};
extern struct kvm_stats_debugfs_item debugfs_entries[];
806
extern struct dentry *kvm_debugfs_dir;
807

808
#if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER)
809 810 811 812 813
static inline int mmu_notifier_retry(struct kvm_vcpu *vcpu, unsigned long mmu_seq)
{
	if (unlikely(vcpu->kvm->mmu_notifier_count))
		return 1;
	/*
814 815 816 817 818 819 820 821
	 * Ensure the read of mmu_notifier_count happens before the read
	 * of mmu_notifier_seq.  This interacts with the smp_wmb() in
	 * mmu_notifier_invalidate_range_end to make sure that the caller
	 * either sees the old (non-zero) value of mmu_notifier_count or
	 * the new (incremented) value of mmu_notifier_seq.
	 * PowerPC Book3s HV KVM calls this under a per-page lock
	 * rather than under kvm->mmu_lock, for scalability, so
	 * can't rely on kvm->mmu_lock to keep things ordered.
822
	 */
823
	smp_rmb();
824 825 826 827 828 829
	if (vcpu->kvm->mmu_notifier_seq != mmu_seq)
		return 1;
	return 0;
}
#endif

830
#ifdef KVM_CAP_IRQ_ROUTING
831 832 833 834 835 836 837 838 839 840

#define KVM_MAX_IRQ_ROUTES 1024

int kvm_setup_default_irq_routing(struct kvm *kvm);
int kvm_set_irq_routing(struct kvm *kvm,
			const struct kvm_irq_routing_entry *entries,
			unsigned nr,
			unsigned flags);
void kvm_free_irq_routing(struct kvm *kvm);

841 842
int kvm_send_userspace_msi(struct kvm *kvm, struct kvm_msi *msi);

843 844 845 846 847 848
#else

static inline void kvm_free_irq_routing(struct kvm *kvm) {}

#endif

G
Gregory Haskins 已提交
849 850
#ifdef CONFIG_HAVE_KVM_EVENTFD

G
Gregory Haskins 已提交
851
void kvm_eventfd_init(struct kvm *kvm);
852
int kvm_irqfd(struct kvm *kvm, struct kvm_irqfd *args);
G
Gregory Haskins 已提交
853
void kvm_irqfd_release(struct kvm *kvm);
854
void kvm_irq_routing_update(struct kvm *, struct kvm_irq_routing_table *);
G
Gregory Haskins 已提交
855
int kvm_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args);
G
Gregory Haskins 已提交
856 857 858

#else

G
Gregory Haskins 已提交
859
static inline void kvm_eventfd_init(struct kvm *kvm) {}
860

861
static inline int kvm_irqfd(struct kvm *kvm, struct kvm_irqfd *args)
G
Gregory Haskins 已提交
862 863 864 865 866
{
	return -EINVAL;
}

static inline void kvm_irqfd_release(struct kvm *kvm) {}
867

A
Alexander Graf 已提交
868
#ifdef CONFIG_HAVE_KVM_IRQCHIP
869 870 871 872 873
static inline void kvm_irq_routing_update(struct kvm *kvm,
					  struct kvm_irq_routing_table *irq_rt)
{
	rcu_assign_pointer(kvm->irq_routing, irq_rt);
}
A
Alexander Graf 已提交
874
#endif
875

G
Gregory Haskins 已提交
876 877 878 879
static inline int kvm_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args)
{
	return -ENOSYS;
}
G
Gregory Haskins 已提交
880 881 882

#endif /* CONFIG_HAVE_KVM_EVENTFD */

883
#ifdef CONFIG_KVM_APIC_ARCHITECTURE
884 885
static inline bool kvm_vcpu_is_bsp(struct kvm_vcpu *vcpu)
{
886
	return vcpu->kvm->bsp_vcpu_id == vcpu->vcpu_id;
887
}
888 889 890 891 892 893 894

bool kvm_vcpu_compatible(struct kvm_vcpu *vcpu);

#else

static inline bool kvm_vcpu_compatible(struct kvm_vcpu *vcpu) { return true; }

A
Avi Kivity 已提交
895
#endif
896 897 898 899 900 901 902 903 904 905 906 907 908 909

#ifdef __KVM_HAVE_DEVICE_ASSIGNMENT

long kvm_vm_ioctl_assigned_device(struct kvm *kvm, unsigned ioctl,
				  unsigned long arg);

#else

static inline long kvm_vm_ioctl_assigned_device(struct kvm *kvm, unsigned ioctl,
						unsigned long arg)
{
	return -ENOTTY;
}

910
#endif
911

912 913 914 915 916 917 918
static inline void kvm_make_request(int req, struct kvm_vcpu *vcpu)
{
	set_bit(req, &vcpu->requests);
}

static inline bool kvm_check_request(int req, struct kvm_vcpu *vcpu)
{
919 920 921 922 923 924
	if (test_bit(req, &vcpu->requests)) {
		clear_bit(req, &vcpu->requests);
		return true;
	} else {
		return false;
	}
925 926
}

927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947
#ifdef CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT

static inline void kvm_vcpu_set_in_spin_loop(struct kvm_vcpu *vcpu, bool val)
{
	vcpu->spin_loop.in_spin_loop = val;
}
static inline void kvm_vcpu_set_dy_eligible(struct kvm_vcpu *vcpu, bool val)
{
	vcpu->spin_loop.dy_eligible = val;
}

#else /* !CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT */

static inline void kvm_vcpu_set_in_spin_loop(struct kvm_vcpu *vcpu, bool val)
{
}

static inline void kvm_vcpu_set_dy_eligible(struct kvm_vcpu *vcpu, bool val)
{
}

948 949 950 951 952
static inline bool kvm_vcpu_eligible_for_directed_yield(struct kvm_vcpu *vcpu)
{
	return true;
}

953
#endif /* CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT */
954 955
#endif