kvm_host.h 25.2 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>
A
Alexey Dobriyan 已提交
24
#include <asm/signal.h>
A
Avi Kivity 已提交
25 26

#include <linux/kvm.h>
I
Ingo Molnar 已提交
27
#include <linux/kvm_para.h>
A
Avi Kivity 已提交
28

29
#include <linux/kvm_types.h>
30

31
#include <asm/kvm_host.h>
32

A
Avi Kivity 已提交
33 34 35 36
#ifndef KVM_MMIO_SIZE
#define KVM_MMIO_SIZE 8
#endif

A
Avi Kivity 已提交
37 38 39 40 41 42 43 44 45 46 47 48 49 50
/*
 * 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)

51 52 53
/*
 * vcpu->requests bit members
 */
54
#define KVM_REQ_TLB_FLUSH          0
55
#define KVM_REQ_MIGRATE_TIMER      1
56
#define KVM_REQ_REPORT_TPR_ACCESS  2
57
#define KVM_REQ_MMU_RELOAD         3
J
Joerg Roedel 已提交
58
#define KVM_REQ_TRIPLE_FAULT       4
59
#define KVM_REQ_PENDING_TIMER      5
60
#define KVM_REQ_UNHALT             6
61
#define KVM_REQ_MMU_SYNC           7
Z
Zachary Amsden 已提交
62
#define KVM_REQ_CLOCK_UPDATE       8
63
#define KVM_REQ_KICK               9
64
#define KVM_REQ_DEACTIVATE_FPU    10
65
#define KVM_REQ_EVENT             11
66
#define KVM_REQ_APF_HALT          12
G
Glauber Costa 已提交
67
#define KVM_REQ_STEAL_UPDATE      13
A
Avi Kivity 已提交
68
#define KVM_REQ_NMI               14
69
#define KVM_REQ_IMMEDIATE_EXIT    15
70 71
#define KVM_REQ_PMU               16
#define KVM_REQ_PMI               17
A
Avi Kivity 已提交
72

73 74
#define KVM_USERSPACE_IRQ_SOURCE_ID	0

75
struct kvm;
A
Avi Kivity 已提交
76
struct kvm_vcpu;
77
extern struct kmem_cache *kvm_vcpu_cache;
A
Avi Kivity 已提交
78

79 80 81 82 83 84
struct kvm_io_range {
	gpa_t addr;
	int len;
	struct kvm_io_device *dev;
};

85
#define NR_IOBUS_DEVS 1000
86

87 88
struct kvm_io_bus {
	int                   dev_count;
89
	struct kvm_io_range range[];
90 91
};

M
Marcelo Tosatti 已提交
92 93 94 95 96 97 98 99 100
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,
101
		    void *val);
102 103
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
104 105
int kvm_io_bus_unregister_dev(struct kvm *kvm, enum kvm_bus bus_idx,
			      struct kvm_io_device *dev);
106

107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
#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);
125
int kvm_async_pf_wakeup_all(struct kvm_vcpu *vcpu);
126 127
#endif

128 129 130
enum {
	OUTSIDE_GUEST_MODE,
	IN_GUEST_MODE,
131 132
	EXITING_GUEST_MODE,
	READING_SHADOW_PAGE_TABLES,
133 134
};

A
Avi Kivity 已提交
135 136 137 138 139 140 141 142 143 144
/*
 * 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;
};

145 146
struct kvm_vcpu {
	struct kvm *kvm;
147
#ifdef CONFIG_PREEMPT_NOTIFIERS
148
	struct preempt_notifier preempt_notifier;
149
#endif
150
	int cpu;
151
	int vcpu_id;
152 153
	int srcu_idx;
	int mode;
154
	unsigned long requests;
J
Jan Kiszka 已提交
155
	unsigned long guest_debug;
156 157 158

	struct mutex mutex;
	struct kvm_run *run;
159

160
	int fpu_active;
161
	int guest_fpu_loaded, guest_xcr0_loaded;
162
	wait_queue_head_t wq;
163
	struct pid *pid;
164 165 166 167
	int sigset_active;
	sigset_t sigset;
	struct kvm_vcpu_stat stat;

168
#ifdef CONFIG_HAS_IOMEM
169 170 171
	int mmio_needed;
	int mmio_read_completed;
	int mmio_is_write;
A
Avi Kivity 已提交
172 173 174
	int mmio_cur_fragment;
	int mmio_nr_fragments;
	struct kvm_mmio_fragment mmio_fragments[KVM_MAX_MMIO_FRAGMENTS];
175
#endif
A
Avi Kivity 已提交
176

177 178 179 180 181 182 183 184 185
#ifdef CONFIG_KVM_ASYNC_PF
	struct {
		u32 queued;
		struct list_head queue;
		struct list_head done;
		spinlock_t lock;
	} async_pf;
#endif

186 187 188 189 190 191 192 193 194 195 196 197
#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
198 199 200
	struct kvm_vcpu_arch arch;
};

201 202 203 204 205
static inline int kvm_vcpu_exiting_guest_mode(struct kvm_vcpu *vcpu)
{
	return cmpxchg(&vcpu->mode, IN_GUEST_MODE, EXITING_GUEST_MODE);
}

206 207 208 209 210 211
/*
 * 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
212 213 214 215
struct kvm_memory_slot {
	gfn_t base_gfn;
	unsigned long npages;
	unsigned long flags;
216
	unsigned long *rmap;
A
Avi Kivity 已提交
217
	unsigned long *dirty_bitmap;
218
	struct kvm_arch_memory_slot arch;
219
	unsigned long userspace_addr;
220
	int user_alloc;
221
	int id;
A
Avi Kivity 已提交
222 223
};

224 225 226 227 228
static inline unsigned long kvm_dirty_bitmap_bytes(struct kvm_memory_slot *memslot)
{
	return ALIGN(memslot->npages, BITS_PER_LONG) / 8;
}

229 230
struct kvm_kernel_irq_routing_entry {
	u32 gsi;
231
	u32 type;
232
	int (*set)(struct kvm_kernel_irq_routing_entry *e,
233
		   struct kvm *kvm, int irq_source_id, int level);
234 235 236 237 238
	union {
		struct {
			unsigned irqchip;
			unsigned pin;
		} irqchip;
S
Sheng Yang 已提交
239
		struct msi_msg msi;
240
	};
241 242 243
	struct hlist_node link;
};

244 245
#ifdef __KVM_HAVE_IOAPIC

246
struct kvm_irq_routing_table {
247
	int chip[KVM_NR_IRQCHIPS][KVM_IOAPIC_NUM_PINS];
248 249 250 251 252 253 254
	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];
255 256
};

257 258 259 260 261 262
#else

struct kvm_irq_routing_table {};

#endif

263 264 265 266
#ifndef KVM_MEM_SLOTS_NUM
#define KVM_MEM_SLOTS_NUM (KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS)
#endif

267 268 269 270 271
/*
 * Note:
 * memslots are not sorted by id anymore, please use id_to_memslot()
 * to get the memslot by its id.
 */
272
struct kvm_memslots {
273
	u64 generation;
274
	struct kvm_memory_slot memslots[KVM_MEM_SLOTS_NUM];
275 276
	/* The mapping table from slot id to the index in memslots[]. */
	int id_to_index[KVM_MEM_SLOTS_NUM];
277 278
};

A
Avi Kivity 已提交
279
struct kvm {
280
	spinlock_t mmu_lock;
281
	struct mutex slots_lock;
282
	struct mm_struct *mm; /* userspace tied to this vm */
283
	struct kvm_memslots *memslots;
284
	struct srcu_struct srcu;
285 286 287
#ifdef CONFIG_KVM_APIC_ARCHITECTURE
	u32 bsp_vcpu_id;
#endif
R
Rusty Russell 已提交
288
	struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];
289
	atomic_t online_vcpus;
290
	int last_boosted_vcpu;
291
	struct list_head vm_list;
292
	struct mutex lock;
M
Marcelo Tosatti 已提交
293
	struct kvm_io_bus *buses[KVM_NR_BUSES];
G
Gregory Haskins 已提交
294 295 296 297 298
#ifdef CONFIG_HAVE_KVM_EVENTFD
	struct {
		spinlock_t        lock;
		struct list_head  items;
	} irqfds;
G
Gregory Haskins 已提交
299
	struct list_head ioeventfds;
G
Gregory Haskins 已提交
300
#endif
301
	struct kvm_vm_stat stat;
302
	struct kvm_arch arch;
I
Izik Eidus 已提交
303
	atomic_t users_count;
304 305
#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
	struct kvm_coalesced_mmio_ring *coalesced_mmio_ring;
306 307
	spinlock_t ring_lock;
	struct list_head coalesced_zones;
308
#endif
309

310
	struct mutex irq_lock;
311
#ifdef CONFIG_HAVE_KVM_IRQCHIP
312 313 314 315
	/*
	 * Update side is protected by irq_lock and,
	 * if configured, irqfds.lock.
	 */
A
Arnd Bergmann 已提交
316
	struct kvm_irq_routing_table __rcu *irq_routing;
317
	struct hlist_head mask_notifier_list;
318
	struct hlist_head irq_ack_notifier_list;
319 320
#endif

321
#if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER)
322 323 324 325
	struct mmu_notifier mmu_notifier;
	unsigned long mmu_notifier_seq;
	long mmu_notifier_count;
#endif
326
	long tlbs_dirty;
A
Avi Kivity 已提交
327 328
};

329 330 331 332 333 334 335 336 337
#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__)
338

339 340 341
/* 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
342

343 344 345 346 347 348 349
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) \
350 351 352 353
	for (idx = 0; \
	     idx < atomic_read(&kvm->online_vcpus) && \
	     (vcpup = kvm_get_vcpu(kvm, idx)) != NULL; \
	     idx++)
354

355 356
#define kvm_for_each_memslot(memslot, slots)	\
	for (memslot = &slots->memslots[0];	\
357 358
	      memslot < slots->memslots + KVM_MEM_SLOTS_NUM && memslot->npages;\
		memslot++)
359

R
Rusty Russell 已提交
360 361 362
int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id);
void kvm_vcpu_uninit(struct kvm_vcpu *vcpu);

363 364 365
void vcpu_load(struct kvm_vcpu *vcpu);
void vcpu_put(struct kvm_vcpu *vcpu);

366
int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align,
367
		  struct module *module);
368
void kvm_exit(void);
A
Avi Kivity 已提交
369

I
Izik Eidus 已提交
370 371
void kvm_get_kvm(struct kvm *kvm);
void kvm_put_kvm(struct kvm *kvm);
372
void update_memslots(struct kvm_memslots *slots, struct kvm_memory_slot *new);
I
Izik Eidus 已提交
373

374 375 376 377 378 379 380
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));
}

381 382 383
static inline struct kvm_memory_slot *
id_to_memslot(struct kvm_memslots *slots, int id)
{
384 385
	int index = slots->id_to_index[id];
	struct kvm_memory_slot *slot;
386

387
	slot = &slots->memslots[index];
388

389 390
	WARN_ON(slot->id != id);
	return slot;
391 392
}

393
extern struct page *bad_page;
A
Avi Kivity 已提交
394

395
int is_error_page(struct page *page);
396
int is_error_pfn(pfn_t pfn);
397
int is_hwpoison_pfn(pfn_t pfn);
398 399
int is_noslot_pfn(pfn_t pfn);
int is_invalid_pfn(pfn_t pfn);
I
Izik Eidus 已提交
400
int kvm_is_error_hva(unsigned long addr);
401 402 403
int kvm_set_memory_region(struct kvm *kvm,
			  struct kvm_userspace_memory_region *mem,
			  int user_alloc);
404 405 406
int __kvm_set_memory_region(struct kvm *kvm,
			    struct kvm_userspace_memory_region *mem,
			    int user_alloc);
407 408 409
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);
410 411 412 413 414 415
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,
416 417 418
				struct kvm_userspace_memory_region *mem,
				struct kvm_memory_slot old,
				int user_alloc);
419
bool kvm_largepages_enabled(void);
420
void kvm_disable_largepages(void);
421
void kvm_arch_flush_shadow(struct kvm *kvm);
422

423 424 425
int gfn_to_page_many_atomic(struct kvm *kvm, gfn_t gfn, struct page **pages,
			    int nr_pages);

A
Avi Kivity 已提交
426
struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn);
M
Marcelo Tosatti 已提交
427
unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn);
428 429
void kvm_release_page_clean(struct page *page);
void kvm_release_page_dirty(struct page *page);
430 431 432
void kvm_set_page_dirty(struct page *page);
void kvm_set_page_accessed(struct page *page);

433
pfn_t hva_to_pfn_atomic(unsigned long addr);
434
pfn_t gfn_to_pfn_atomic(struct kvm *kvm, gfn_t gfn);
435 436
pfn_t gfn_to_pfn_async(struct kvm *kvm, gfn_t gfn, bool *async,
		       bool write_fault, bool *writable);
437
pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn);
438 439
pfn_t gfn_to_pfn_prot(struct kvm *kvm, gfn_t gfn, bool write_fault,
		      bool *writable);
440
pfn_t gfn_to_pfn_memslot(struct kvm_memory_slot *slot, gfn_t gfn);
441 442 443 444 445
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);
446
pfn_t get_fault_pfn(void);
447

448 449
int kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset,
			int len);
450 451
int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
			  unsigned long len);
452
int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len);
453 454
int kvm_read_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc,
			   void *data, unsigned long len);
455 456 457 458
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);
459 460 461 462
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);
463 464
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
465
struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn);
466
int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn);
J
Joerg Roedel 已提交
467
unsigned long kvm_host_page_size(struct kvm *kvm, gfn_t gfn);
A
Avi Kivity 已提交
468
void mark_page_dirty(struct kvm *kvm, gfn_t gfn);
469 470
void mark_page_dirty_in_slot(struct kvm *kvm, struct kvm_memory_slot *memslot,
			     gfn_t gfn);
A
Avi Kivity 已提交
471

472
void kvm_vcpu_block(struct kvm_vcpu *vcpu);
473
void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
474
bool kvm_vcpu_yield_to(struct kvm_vcpu *target);
Z
Zhai, Edwin 已提交
475
void kvm_vcpu_on_spin(struct kvm_vcpu *vcpu);
A
Avi Kivity 已提交
476
void kvm_resched(struct kvm_vcpu *vcpu);
477 478
void kvm_load_guest_fpu(struct kvm_vcpu *vcpu);
void kvm_put_guest_fpu(struct kvm_vcpu *vcpu);
479

480
void kvm_flush_remote_tlbs(struct kvm *kvm);
481
void kvm_reload_remote_mmus(struct kvm *kvm);
A
Avi Kivity 已提交
482

483 484
long kvm_arch_dev_ioctl(struct file *filp,
			unsigned int ioctl, unsigned long arg);
485 486
long kvm_arch_vcpu_ioctl(struct file *filp,
			 unsigned int ioctl, unsigned long arg);
487
int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf);
488 489 490

int kvm_dev_ioctl_check_extension(long ext);

491 492 493 494 495
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);

496 497 498 499 500 501
int kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
				   struct
				   kvm_userspace_memory_region *mem,
				   int user_alloc);
long kvm_arch_vm_ioctl(struct file *filp,
		       unsigned int ioctl, unsigned long arg);
502

503 504 505
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);

506 507 508
int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
				    struct kvm_translation *tr);

509 510 511 512 513 514
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);
515 516 517 518
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
519 520
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
					struct kvm_guest_debug *dbg);
521 522
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run);

523 524
int kvm_arch_init(void *opaque);
void kvm_arch_exit(void);
525

526 527 528 529 530 531 532
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);
533
int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu);
534
void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu);
535 536

int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu);
537
int kvm_arch_hardware_enable(void *garbage);
538 539 540 541
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);
542
int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu);
543
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu);
544

545 546
void kvm_free_physmem(struct kvm *kvm);

547 548 549
void *kvm_kvzalloc(unsigned long size);
void kvm_kvfree(const void *addr);

550 551 552 553 554 555 556 557 558 559 560 561
#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

562 563
static inline wait_queue_head_t *kvm_arch_vcpu_wq(struct kvm_vcpu *vcpu)
{
564 565 566
#ifdef __KVM_HAVE_ARCH_WQP
	return vcpu->arch.wqp;
#else
567 568
	return &vcpu->wq;
#endif
569
}
570

571
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type);
572
void kvm_arch_destroy_vm(struct kvm *kvm);
573
void kvm_free_all_assigned_devices(struct kvm *kvm);
574
void kvm_arch_sync_events(struct kvm *kvm);
575

576
int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu);
577
void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
578

579 580
int kvm_is_mmio_pfn(pfn_t pfn);

B
Ben-Ami Yassour 已提交
581 582 583 584 585 586 587 588 589 590
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;
591
	int host_segnr;
B
Ben-Ami Yassour 已提交
592 593
	int host_busnr;
	int host_devfn;
S
Sheng Yang 已提交
594
	unsigned int entries_nr;
B
Ben-Ami Yassour 已提交
595
	int host_irq;
596
	bool host_irq_disabled;
597
	bool pci_2_3;
S
Sheng Yang 已提交
598
	struct msix_entry *host_msix_entries;
B
Ben-Ami Yassour 已提交
599
	int guest_irq;
600
	struct msix_entry *guest_msix_entries;
601
	unsigned long irq_requested_type;
602
	int irq_source_id;
603
	int flags;
B
Ben-Ami Yassour 已提交
604 605
	struct pci_dev *dev;
	struct kvm *kvm;
606
	spinlock_t intx_lock;
607
	spinlock_t intx_mask_lock;
608
	char irq_name[32];
609
	struct pci_saved_state *pci_saved_state;
B
Ben-Ami Yassour 已提交
610
};
611 612 613 614 615 616 617 618 619 620 621

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);
622 623
void kvm_fire_mask_notifiers(struct kvm *kvm, unsigned irqchip, unsigned pin,
			     bool mask);
624

625 626 627 628 629 630
#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);
631 632
int kvm_set_msi(struct kvm_kernel_irq_routing_entry *irq_entry, struct kvm *kvm,
		int irq_source_id, int level);
633
void kvm_notify_acked_irq(struct kvm *kvm, unsigned irqchip, unsigned pin);
634 635
void kvm_register_irq_ack_notifier(struct kvm *kvm,
				   struct kvm_irq_ack_notifier *kian);
636 637
void kvm_unregister_irq_ack_notifier(struct kvm *kvm,
				   struct kvm_irq_ack_notifier *kian);
638 639
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
640

641 642 643
/* For vcpu->arch.iommu_flags */
#define KVM_IOMMU_CACHE_COHERENCY	0x1

J
Joerg Roedel 已提交
644
#ifdef CONFIG_IOMMU_API
645
int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot);
646
void kvm_iommu_unmap_pages(struct kvm *kvm, struct kvm_memory_slot *slot);
W
Weidong Han 已提交
647
int kvm_iommu_map_guest(struct kvm *kvm);
B
Ben-Ami Yassour 已提交
648
int kvm_iommu_unmap_guest(struct kvm *kvm);
W
Weidong Han 已提交
649 650
int kvm_assign_device(struct kvm *kvm,
		      struct kvm_assigned_dev_kernel *assigned_dev);
W
Weidong Han 已提交
651 652
int kvm_deassign_device(struct kvm *kvm,
			struct kvm_assigned_dev_kernel *assigned_dev);
J
Joerg Roedel 已提交
653
#else /* CONFIG_IOMMU_API */
B
Ben-Ami Yassour 已提交
654
static inline int kvm_iommu_map_pages(struct kvm *kvm,
655
				      struct kvm_memory_slot *slot)
B
Ben-Ami Yassour 已提交
656 657 658 659
{
	return 0;
}

660 661 662 663 664
static inline void kvm_iommu_unmap_pages(struct kvm *kvm,
					 struct kvm_memory_slot *slot)
{
}

W
Weidong Han 已提交
665
static inline int kvm_iommu_map_guest(struct kvm *kvm)
B
Ben-Ami Yassour 已提交
666 667 668 669 670 671 672 673
{
	return -ENODEV;
}

static inline int kvm_iommu_unmap_guest(struct kvm *kvm)
{
	return 0;
}
W
Weidong Han 已提交
674 675 676 677 678 679

static inline int kvm_assign_device(struct kvm *kvm,
		struct kvm_assigned_dev_kernel *assigned_dev)
{
	return 0;
}
W
Weidong Han 已提交
680 681 682 683 684 685

static inline int kvm_deassign_device(struct kvm *kvm,
		struct kvm_assigned_dev_kernel *assigned_dev)
{
	return 0;
}
J
Joerg Roedel 已提交
686
#endif /* CONFIG_IOMMU_API */
B
Ben-Ami Yassour 已提交
687

688 689
static inline void kvm_guest_enter(void)
{
690
	BUG_ON(preemptible());
691
	account_system_vtime(current);
692
	current->flags |= PF_VCPU;
693 694 695 696 697 698 699 700
	/* 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());
701 702 703 704
}

static inline void kvm_guest_exit(void)
{
705
	account_system_vtime(current);
706 707 708
	current->flags &= ~PF_VCPU;
}

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
/*
 * 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);
}

734 735 736 737 738
static inline int memslot_id(struct kvm *kvm, gfn_t gfn)
{
	return gfn_to_memslot(kvm, gfn)->id;
}

739 740 741 742 743 744 745
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));
}

746 747 748 749 750 751 752 753
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;
}

754 755 756 757 758 759
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
760 761 762 763
static inline gpa_t gfn_to_gpa(gfn_t gfn)
{
	return (gpa_t)gfn << PAGE_SHIFT;
}
A
Avi Kivity 已提交
764

765 766 767 768 769
static inline gfn_t gpa_to_gfn(gpa_t gpa)
{
	return (gfn_t)(gpa >> PAGE_SHIFT);
}

B
Ben-Ami Yassour 已提交
770 771 772 773 774
static inline hpa_t pfn_to_hpa(pfn_t pfn)
{
	return (hpa_t)pfn << PAGE_SHIFT;
}

M
Marcelo Tosatti 已提交
775
static inline void kvm_migrate_timers(struct kvm_vcpu *vcpu)
776 777 778 779
{
	set_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests);
}

780 781 782 783 784
enum kvm_stat_kind {
	KVM_STAT_VM,
	KVM_STAT_VCPU,
};

785 786 787
struct kvm_stats_debugfs_item {
	const char *name;
	int offset;
788
	enum kvm_stat_kind kind;
789 790 791
	struct dentry *dentry;
};
extern struct kvm_stats_debugfs_item debugfs_entries[];
792
extern struct dentry *kvm_debugfs_dir;
793

794
#if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER)
795 796 797 798 799
static inline int mmu_notifier_retry(struct kvm_vcpu *vcpu, unsigned long mmu_seq)
{
	if (unlikely(vcpu->kvm->mmu_notifier_count))
		return 1;
	/*
800 801 802 803 804 805 806 807
	 * 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.
808
	 */
809
	smp_rmb();
810 811 812 813 814 815
	if (vcpu->kvm->mmu_notifier_seq != mmu_seq)
		return 1;
	return 0;
}
#endif

816
#ifdef KVM_CAP_IRQ_ROUTING
817 818 819 820 821 822 823 824 825 826

#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);

827 828
int kvm_send_userspace_msi(struct kvm *kvm, struct kvm_msi *msi);

829 830 831 832 833 834
#else

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

#endif

G
Gregory Haskins 已提交
835 836
#ifdef CONFIG_HAVE_KVM_EVENTFD

G
Gregory Haskins 已提交
837
void kvm_eventfd_init(struct kvm *kvm);
G
Gregory Haskins 已提交
838 839
int kvm_irqfd(struct kvm *kvm, int fd, int gsi, int flags);
void kvm_irqfd_release(struct kvm *kvm);
840
void kvm_irq_routing_update(struct kvm *, struct kvm_irq_routing_table *);
G
Gregory Haskins 已提交
841
int kvm_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args);
G
Gregory Haskins 已提交
842 843 844

#else

G
Gregory Haskins 已提交
845
static inline void kvm_eventfd_init(struct kvm *kvm) {}
846

G
Gregory Haskins 已提交
847 848 849 850 851 852
static inline int kvm_irqfd(struct kvm *kvm, int fd, int gsi, int flags)
{
	return -EINVAL;
}

static inline void kvm_irqfd_release(struct kvm *kvm) {}
853

A
Alexander Graf 已提交
854
#ifdef CONFIG_HAVE_KVM_IRQCHIP
855 856 857 858 859
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
860
#endif
861

G
Gregory Haskins 已提交
862 863 864 865
static inline int kvm_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args)
{
	return -ENOSYS;
}
G
Gregory Haskins 已提交
866 867 868

#endif /* CONFIG_HAVE_KVM_EVENTFD */

869
#ifdef CONFIG_KVM_APIC_ARCHITECTURE
870 871
static inline bool kvm_vcpu_is_bsp(struct kvm_vcpu *vcpu)
{
872
	return vcpu->kvm->bsp_vcpu_id == vcpu->vcpu_id;
873
}
874 875 876 877 878 879 880

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
881
#endif
882 883 884 885 886 887 888 889 890 891 892 893 894 895

#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;
}

896
#endif
897

898 899 900 901 902 903 904
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)
{
905 906 907 908 909 910
	if (test_bit(req, &vcpu->requests)) {
		clear_bit(req, &vcpu->requests);
		return true;
	} else {
		return false;
	}
911 912
}

913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934
#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)
{
}

#endif /* CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT */
935 936
#endif