kvm_host.h 24.8 KB
Newer Older
1
/*
2 3 4 5 6 7 8 9 10
 * Kernel-based Virtual Machine driver for Linux
 *
 * This header defines architecture specific interfaces, x86 version
 *
 * This work is licensed under the terms of the GNU GPL, version 2.  See
 * the COPYING file in the top-level directory.
 *
 */

H
H. Peter Anvin 已提交
11 12
#ifndef _ASM_X86_KVM_HOST_H
#define _ASM_X86_KVM_HOST_H
13

14 15
#include <linux/types.h>
#include <linux/mm.h>
16
#include <linux/mmu_notifier.h>
17
#include <linux/tracepoint.h>
18
#include <linux/cpumask.h>
19 20 21

#include <linux/kvm.h>
#include <linux/kvm_para.h>
22
#include <linux/kvm_types.h>
23

24
#include <asm/pvclock-abi.h>
25
#include <asm/desc.h>
S
Sheng Yang 已提交
26
#include <asm/mtrr.h>
27
#include <asm/msr-index.h>
28

A
Avi Kivity 已提交
29
#define KVM_MAX_VCPUS 64
30 31 32
#define KVM_MEMORY_SLOTS 32
/* memory slots that does not exposed to userspace */
#define KVM_PRIVATE_MEM_SLOTS 4
A
Avi Kivity 已提交
33
#define KVM_MMIO_SIZE 16
34 35

#define KVM_PIO_PAGE_OFFSET 1
36
#define KVM_COALESCED_MMIO_PAGE_OFFSET 2
37

38 39 40 41 42
#define CR0_RESERVED_BITS                                               \
	(~(unsigned long)(X86_CR0_PE | X86_CR0_MP | X86_CR0_EM | X86_CR0_TS \
			  | X86_CR0_ET | X86_CR0_NE | X86_CR0_WP | X86_CR0_AM \
			  | X86_CR0_NW | X86_CR0_CD | X86_CR0_PG))

43 44
#define CR3_PAE_RESERVED_BITS ((X86_CR3_PWT | X86_CR3_PCD) - 1)
#define CR3_NONPAE_RESERVED_BITS ((PAGE_SIZE-1) & ~(X86_CR3_PWT | X86_CR3_PCD))
45 46
#define CR3_L_MODE_RESERVED_BITS (CR3_NONPAE_RESERVED_BITS |	\
				  0xFFFFFF0000000000ULL)
47 48 49 50
#define CR4_RESERVED_BITS                                               \
	(~(unsigned long)(X86_CR4_VME | X86_CR4_PVI | X86_CR4_TSD | X86_CR4_DE\
			  | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_MCE     \
			  | X86_CR4_PGE | X86_CR4_PCE | X86_CR4_OSFXSR  \
51
			  | X86_CR4_OSXSAVE | X86_CR4_SMEP | X86_CR4_RDWRGSFS \
52 53 54 55 56
			  | X86_CR4_OSXMMEXCPT | X86_CR4_VMXE))

#define CR8_RESERVED_BITS (~(unsigned long)X86_CR8_TPR)


57 58

#define INVALID_PAGE (~(hpa_t)0)
59 60
#define VALID_PAGE(x) ((x) != INVALID_PAGE)

61 62
#define UNMAPPED_GVA (~(gpa_t)0)

63
/* KVM Hugepage definitions for x86 */
64
#define KVM_NR_PAGE_SIZES	3
65 66
#define KVM_HPAGE_GFN_SHIFT(x)	(((x) - 1) * 9)
#define KVM_HPAGE_SHIFT(x)	(PAGE_SHIFT + KVM_HPAGE_GFN_SHIFT(x))
67 68 69
#define KVM_HPAGE_SIZE(x)	(1UL << KVM_HPAGE_SHIFT(x))
#define KVM_HPAGE_MASK(x)	(~(KVM_HPAGE_SIZE(x) - 1))
#define KVM_PAGES_PER_HPAGE(x)	(KVM_HPAGE_SIZE(x) / PAGE_SIZE)
M
Marcelo Tosatti 已提交
70

71
#define DE_VECTOR 0
J
Jan Kiszka 已提交
72
#define DB_VECTOR 1
73 74 75
#define BP_VECTOR 3
#define OF_VECTOR 4
#define BR_VECTOR 5
76 77 78 79 80 81 82 83
#define UD_VECTOR 6
#define NM_VECTOR 7
#define DF_VECTOR 8
#define TS_VECTOR 10
#define NP_VECTOR 11
#define SS_VECTOR 12
#define GP_VECTOR 13
#define PF_VECTOR 14
84
#define MF_VECTOR 16
85
#define MC_VECTOR 18
86 87 88 89 90 91

#define SELECTOR_TI_MASK (1 << 2)
#define SELECTOR_RPL_MASK 0x03

#define IOPL_SHIFT 12

92 93
#define KVM_PERMILLE_MMU_PAGES 20
#define KVM_MIN_ALLOC_MMU_PAGES 64
94 95
#define KVM_MMU_HASH_SHIFT 10
#define KVM_NUM_MMU_PAGES (1 << KVM_MMU_HASH_SHIFT)
96 97
#define KVM_MIN_FREE_MMU_PAGES 5
#define KVM_REFILL_PAGES 25
98
#define KVM_MAX_CPUID_ENTRIES 80
S
Sheng Yang 已提交
99
#define KVM_NR_FIXED_MTRR_REGION 88
A
Avi Kivity 已提交
100
#define KVM_NR_VAR_MTRR 8
101

102 103
#define ASYNC_PF_PER_VCPU 64

104
extern raw_spinlock_t kvm_lock;
105 106
extern struct list_head vm_list;

107 108
struct kvm_vcpu;
struct kvm;
109
struct kvm_async_pf;
110

111
enum kvm_reg {
112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129
	VCPU_REGS_RAX = 0,
	VCPU_REGS_RCX = 1,
	VCPU_REGS_RDX = 2,
	VCPU_REGS_RBX = 3,
	VCPU_REGS_RSP = 4,
	VCPU_REGS_RBP = 5,
	VCPU_REGS_RSI = 6,
	VCPU_REGS_RDI = 7,
#ifdef CONFIG_X86_64
	VCPU_REGS_R8 = 8,
	VCPU_REGS_R9 = 9,
	VCPU_REGS_R10 = 10,
	VCPU_REGS_R11 = 11,
	VCPU_REGS_R12 = 12,
	VCPU_REGS_R13 = 13,
	VCPU_REGS_R14 = 14,
	VCPU_REGS_R15 = 15,
#endif
130
	VCPU_REGS_RIP,
131 132 133
	NR_VCPU_REGS
};

A
Avi Kivity 已提交
134 135
enum kvm_reg_ex {
	VCPU_EXREG_PDPTR = NR_VCPU_REGS,
136
	VCPU_EXREG_CR3,
A
Avi Kivity 已提交
137
	VCPU_EXREG_RFLAGS,
A
Avi Kivity 已提交
138
	VCPU_EXREG_CPL,
A
Avi Kivity 已提交
139
	VCPU_EXREG_SEGMENTS,
A
Avi Kivity 已提交
140 141
};

142
enum {
143
	VCPU_SREG_ES,
144
	VCPU_SREG_CS,
145
	VCPU_SREG_SS,
146 147 148 149 150 151 152
	VCPU_SREG_DS,
	VCPU_SREG_FS,
	VCPU_SREG_GS,
	VCPU_SREG_TR,
	VCPU_SREG_LDTR,
};

153
#include <asm/kvm_emulate.h>
154

155 156
#define KVM_NR_MEM_OBJS 40

157 158 159 160 161 162 163 164 165 166 167 168 169
#define KVM_NR_DB_REGS	4

#define DR6_BD		(1 << 13)
#define DR6_BS		(1 << 14)
#define DR6_FIXED_1	0xffff0ff0
#define DR6_VOLATILE	0x0000e00f

#define DR7_BP_EN_MASK	0x000000ff
#define DR7_GE		(1 << 9)
#define DR7_GD		(1 << 13)
#define DR7_FIXED_1	0x00000400
#define DR7_VOLATILE	0xffff23ff

170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191
/*
 * We don't want allocation failures within the mmu code, so we preallocate
 * enough memory for a single page fault in a cache.
 */
struct kvm_mmu_memory_cache {
	int nobjs;
	void *objects[KVM_NR_MEM_OBJS];
};

#define NR_PTE_CHAIN_ENTRIES 5

struct kvm_pte_chain {
	u64 *parent_ptes[NR_PTE_CHAIN_ENTRIES];
	struct hlist_node link;
};

/*
 * kvm_mmu_page_role, below, is defined as:
 *
 *   bits 0:3 - total guest paging levels (2-4, or zero for real mode)
 *   bits 4:7 - page table level for this shadow (1-4)
 *   bits 8:9 - page table quadrant for 2-level guests
192 193
 *   bit   16 - direct mapping of virtual to physical mapping at gfn
 *              used for real mode and two-dimensional paging
194 195 196 197 198
 *   bits 17:19 - common access permissions for all ptes in this shadow page
 */
union kvm_mmu_page_role {
	unsigned word;
	struct {
199
		unsigned level:4;
200
		unsigned cr4_pae:1;
201 202
		unsigned quadrant:2;
		unsigned pad_for_nice_hex_output:6;
203
		unsigned direct:1;
204
		unsigned access:3;
205
		unsigned invalid:1;
206
		unsigned nxe:1;
207
		unsigned cr0_wp:1;
208
		unsigned smep_andnot_wp:1;
209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225
	};
};

struct kvm_mmu_page {
	struct list_head link;
	struct hlist_node hash_link;

	/*
	 * The following two entries are used to key the shadow page in the
	 * hash table.
	 */
	gfn_t gfn;
	union kvm_mmu_page_role role;

	u64 *spt;
	/* hold the gfn of each spte inside spt */
	gfn_t *gfns;
226 227 228 229 230
	/*
	 * One bit set per slot which has memory
	 * in this shadow page.
	 */
	DECLARE_BITMAP(slot_bitmap, KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS);
231
	bool unsync;
232
	int root_count;          /* Currently serving as active root */
233
	unsigned int unsync_children;
234
	unsigned long parent_ptes;	/* Reverse mapping for parent_pte */
235
	DECLARE_BITMAP(unsync_child_bitmap, 512);
236 237
};

238 239 240 241 242 243 244
struct kvm_pv_mmu_op_buffer {
	void *ptr;
	unsigned len;
	unsigned processed;
	char buf[512] __aligned(sizeof(long));
};

245 246 247 248 249 250 251
struct kvm_pio_request {
	unsigned long count;
	int in;
	int port;
	int size;
};

252 253 254 255 256 257 258
/*
 * x86 supports 3 paging modes (4-level 64-bit, 3-level 64-bit, and 2-level
 * 32-bit).  The kvm_mmu structure abstracts the details of the current mmu
 * mode.
 */
struct kvm_mmu {
	void (*new_cr3)(struct kvm_vcpu *vcpu);
259
	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long root);
260
	unsigned long (*get_cr3)(struct kvm_vcpu *vcpu);
261 262
	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err,
			  bool prefault);
263 264
	void (*inject_page_fault)(struct kvm_vcpu *vcpu,
				  struct x86_exception *fault);
265
	void (*free)(struct kvm_vcpu *vcpu);
266
	gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva, u32 access,
267
			    struct x86_exception *exception);
268
	gpa_t (*translate_gpa)(struct kvm_vcpu *vcpu, gpa_t gpa, u32 access);
269 270
	void (*prefetch_page)(struct kvm_vcpu *vcpu,
			      struct kvm_mmu_page *page);
271
	int (*sync_page)(struct kvm_vcpu *vcpu,
272
			 struct kvm_mmu_page *sp);
M
Marcelo Tosatti 已提交
273
	void (*invlpg)(struct kvm_vcpu *vcpu, gva_t gva);
274
	void (*update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
275
			   u64 *spte, const void *pte);
276 277 278
	hpa_t root_hpa;
	int root_level;
	int shadow_root_level;
279
	union kvm_mmu_page_role base_role;
280
	bool direct_map;
281 282

	u64 *pae_root;
283
	u64 *lm_root;
284
	u64 rsvd_bits_mask[2][4];
285

286 287
	bool nx;

288
	u64 pdptrs[4]; /* pae */
289 290
};

291
struct kvm_vcpu_arch {
292 293 294 295 296 297 298
	/*
	 * rip and regs accesses must go through
	 * kvm_{register,rip}_{read,write} functions.
	 */
	unsigned long regs[NR_VCPU_REGS];
	u32 regs_avail;
	u32 regs_dirty;
299 300

	unsigned long cr0;
301
	unsigned long cr0_guest_owned_bits;
302 303 304
	unsigned long cr2;
	unsigned long cr3;
	unsigned long cr4;
305
	unsigned long cr4_guest_owned_bits;
306
	unsigned long cr8;
307
	u32 hflags;
308
	u64 efer;
309 310
	u64 apic_base;
	struct kvm_lapic *apic;    /* kernel irqchip context */
311
	int32_t apic_arb_prio;
312 313 314
	int mp_state;
	int sipi_vector;
	u64 ia32_misc_enable_msr;
315
	bool tpr_access_reporting;
316

317 318 319 320 321 322 323
	/*
	 * Paging state of the vcpu
	 *
	 * If the vcpu runs in guest mode with two level paging this still saves
	 * the paging mode of the l1 guest. This context is always used to
	 * handle faults.
	 */
324
	struct kvm_mmu mmu;
325

326 327 328 329 330 331 332 333 334 335
	/*
	 * Paging state of an L2 guest (used for nested npt)
	 *
	 * This context will save all necessary information to walk page tables
	 * of the an L2 guest. This context is only initialized for page table
	 * walking and not for faulting since we never handle l2 page faults on
	 * the host.
	 */
	struct kvm_mmu nested_mmu;

336 337 338 339 340 341
	/*
	 * Pointer to the mmu context currently used for
	 * gva_to_gpa translations.
	 */
	struct kvm_mmu *walk_mmu;

342 343 344
	/* only needed in kvm_pv_mmu_op() path, but it's hot so
	 * put it here to avoid allocation */
	struct kvm_pv_mmu_op_buffer mmu_op_buffer;
345

346
	struct kvm_mmu_memory_cache mmu_pte_list_desc_cache;
347 348 349 350 351 352
	struct kvm_mmu_memory_cache mmu_page_cache;
	struct kvm_mmu_memory_cache mmu_page_header_cache;

	gfn_t last_pt_write_gfn;
	int   last_pt_write_count;
	u64  *last_pte_updated;
353
	gfn_t last_pte_gfn;
354

S
Sheng Yang 已提交
355
	struct fpu guest_fpu;
356
	u64 xcr0;
357 358 359 360

	struct kvm_pio_request pio;
	void *pio_data;

361 362
	u8 event_exit_inst_len;

363 364 365
	struct kvm_queued_exception {
		bool pending;
		bool has_error_code;
366
		bool reinject;
367 368 369 370
		u8 nr;
		u32 error_code;
	} exception;

A
Avi Kivity 已提交
371 372
	struct kvm_queued_interrupt {
		bool pending;
373
		bool soft;
A
Avi Kivity 已提交
374 375 376
		u8 nr;
	} interrupt;

377 378 379
	int halt_request; /* real mode on Intel only */

	int cpuid_nent;
380
	struct kvm_cpuid_entry2 cpuid_entries[KVM_MAX_CPUID_ENTRIES];
381 382 383
	/* emulate context */

	struct x86_emulate_ctxt emulate_ctxt;
384 385
	bool emulate_regs_need_sync_to_vcpu;
	bool emulate_regs_need_sync_from_vcpu;
386 387

	gpa_t time;
388
	struct pvclock_vcpu_time_info hv_clock;
Z
Zachary Amsden 已提交
389
	unsigned int hw_tsc_khz;
390 391
	unsigned int time_offset;
	struct page *time_page;
G
Glauber Costa 已提交
392 393 394 395 396 397 398 399 400

	struct {
		u64 msr_val;
		u64 last_steal;
		u64 accum_steal;
		struct gfn_to_hva_cache stime;
		struct kvm_steal_time steal;
	} st;

401 402
	u64 last_guest_tsc;
	u64 last_kernel_ns;
Z
Zachary Amsden 已提交
403 404
	u64 last_tsc_nsec;
	u64 last_tsc_write;
405
	u32 virtual_tsc_khz;
Z
Zachary Amsden 已提交
406
	bool tsc_catchup;
407 408
	u32  tsc_catchup_mult;
	s8   tsc_catchup_shift;
409 410

	bool nmi_pending;
411
	bool nmi_injected;
A
Avi Kivity 已提交
412

S
Sheng Yang 已提交
413 414
	struct mtrr_state_type mtrr_state;
	u32 pat;
415 416 417 418 419 420

	int switch_db_regs;
	unsigned long db[KVM_NR_DB_REGS];
	unsigned long dr6;
	unsigned long dr7;
	unsigned long eff_db[KVM_NR_DB_REGS];
H
Huang Ying 已提交
421 422 423 424 425

	u64 mcg_cap;
	u64 mcg_status;
	u64 mcg_ctl;
	u64 *mce_banks;
426 427 428

	/* used for guest single stepping over the given code position */
	unsigned long singlestep_rip;
J
Jan Kiszka 已提交
429

G
Gleb Natapov 已提交
430 431
	/* fields used by HYPER-V emulation */
	u64 hv_vapic;
432 433

	cpumask_var_t wbinvd_dirty_mask;
434 435 436 437

	struct {
		bool halted;
		gfn_t gfns[roundup_pow_of_two(ASYNC_PF_PER_VCPU)];
438 439
		struct gfn_to_hva_cache data;
		u64 msr_val;
440
		u32 id;
441
		bool send_user_only;
442
	} apf;
443 444
};

445
struct kvm_arch {
446
	unsigned int n_used_mmu_pages;
447
	unsigned int n_requested_mmu_pages;
448
	unsigned int n_max_mmu_pages;
449
	unsigned int indirect_shadow_pages;
450
	atomic_t invlpg_counter;
451 452 453 454 455
	struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES];
	/*
	 * Hash table of struct kvm_mmu_page.
	 */
	struct list_head active_mmu_pages;
B
Ben-Ami Yassour 已提交
456
	struct list_head assigned_dev_head;
J
Joerg Roedel 已提交
457
	struct iommu_domain *iommu_domain;
458
	int iommu_flags;
459 460
	struct kvm_pic *vpic;
	struct kvm_ioapic *vioapic;
S
Sheng Yang 已提交
461
	struct kvm_pit *vpit;
462
	int vapics_in_nmi_mode;
463 464 465

	unsigned int tss_addr;
	struct page *apic_access_page;
466 467

	gpa_t wall_clock;
468 469 470

	struct page *ept_identity_pagetable;
	bool ept_identity_pagetable_done;
471
	gpa_t ept_identity_map_addr;
472 473

	unsigned long irq_sources_bitmap;
474
	s64 kvmclock_offset;
475
	raw_spinlock_t tsc_write_lock;
Z
Zachary Amsden 已提交
476 477 478
	u64 last_tsc_nsec;
	u64 last_tsc_offset;
	u64 last_tsc_write;
E
Ed Swierk 已提交
479 480

	struct kvm_xen_hvm_config xen_hvm_config;
481 482 483 484

	/* fields used by HYPER-V emulation */
	u64 hv_guest_os_id;
	u64 hv_hypercall;
485 486 487 488

	#ifdef CONFIG_KVM_MMU_AUDIT
	int audit_point;
	#endif
489 490
};

491 492 493 494 495 496 497
struct kvm_vm_stat {
	u32 mmu_shadow_zapped;
	u32 mmu_pte_write;
	u32 mmu_pte_updated;
	u32 mmu_pde_zapped;
	u32 mmu_flooded;
	u32 mmu_recycled;
A
Avi Kivity 已提交
498
	u32 mmu_cache_miss;
499
	u32 mmu_unsync;
500
	u32 remote_tlb_flush;
M
Marcelo Tosatti 已提交
501
	u32 lpages;
502 503
};

504 505 506 507 508 509 510 511 512 513 514
struct kvm_vcpu_stat {
	u32 pf_fixed;
	u32 pf_guest;
	u32 tlb_flush;
	u32 invlpg;

	u32 exits;
	u32 io_exits;
	u32 mmio_exits;
	u32 signal_exits;
	u32 irq_window_exits;
515
	u32 nmi_window_exits;
516 517 518 519 520 521 522 523 524
	u32 halt_exits;
	u32 halt_wakeup;
	u32 request_irq_exits;
	u32 irq_exits;
	u32 host_state_reload;
	u32 efer_reload;
	u32 fpu_reload;
	u32 insn_emulation;
	u32 insn_emulation_fail;
A
Amit Shah 已提交
525
	u32 hypercalls;
526
	u32 irq_injections;
527
	u32 nmi_injections;
528
};
529

530 531
struct x86_instruction_info;

532 533 534
struct kvm_x86_ops {
	int (*cpu_has_kvm_support)(void);          /* __init */
	int (*disabled_by_bios)(void);             /* __init */
535
	int (*hardware_enable)(void *dummy);
536 537 538 539
	void (*hardware_disable)(void *dummy);
	void (*check_processor_compatibility)(void *rtn);
	int (*hardware_setup)(void);               /* __init */
	void (*hardware_unsetup)(void);            /* __exit */
540
	bool (*cpu_has_accelerated_tpr)(void);
541
	void (*cpuid_update)(struct kvm_vcpu *vcpu);
542 543 544 545 546 547 548 549 550 551

	/* Create, but do not attach this VCPU */
	struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned id);
	void (*vcpu_free)(struct kvm_vcpu *vcpu);
	int (*vcpu_reset)(struct kvm_vcpu *vcpu);

	void (*prepare_guest_switch)(struct kvm_vcpu *vcpu);
	void (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu);
	void (*vcpu_put)(struct kvm_vcpu *vcpu);

552 553
	void (*set_guest_debug)(struct kvm_vcpu *vcpu,
				struct kvm_guest_debug *dbg);
554 555 556 557 558
	int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata);
	int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data);
	u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg);
	void (*get_segment)(struct kvm_vcpu *vcpu,
			    struct kvm_segment *var, int seg);
559
	int (*get_cpl)(struct kvm_vcpu *vcpu);
560 561 562
	void (*set_segment)(struct kvm_vcpu *vcpu,
			    struct kvm_segment *var, int seg);
	void (*get_cs_db_l_bits)(struct kvm_vcpu *vcpu, int *db, int *l);
563
	void (*decache_cr0_guest_bits)(struct kvm_vcpu *vcpu);
564
	void (*decache_cr3)(struct kvm_vcpu *vcpu);
565 566 567
	void (*decache_cr4_guest_bits)(struct kvm_vcpu *vcpu);
	void (*set_cr0)(struct kvm_vcpu *vcpu, unsigned long cr0);
	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long cr3);
568
	int (*set_cr4)(struct kvm_vcpu *vcpu, unsigned long cr4);
569
	void (*set_efer)(struct kvm_vcpu *vcpu, u64 efer);
570 571 572 573
	void (*get_idt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
	void (*set_idt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
	void (*get_gdt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
	void (*set_gdt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
574
	void (*set_dr7)(struct kvm_vcpu *vcpu, unsigned long value);
575
	void (*cache_reg)(struct kvm_vcpu *vcpu, enum kvm_reg reg);
576 577
	unsigned long (*get_rflags)(struct kvm_vcpu *vcpu);
	void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags);
A
Avi Kivity 已提交
578
	void (*fpu_activate)(struct kvm_vcpu *vcpu);
579
	void (*fpu_deactivate)(struct kvm_vcpu *vcpu);
580 581 582

	void (*tlb_flush)(struct kvm_vcpu *vcpu);

A
Avi Kivity 已提交
583 584
	void (*run)(struct kvm_vcpu *vcpu);
	int (*handle_exit)(struct kvm_vcpu *vcpu);
585
	void (*skip_emulated_instruction)(struct kvm_vcpu *vcpu);
586 587
	void (*set_interrupt_shadow)(struct kvm_vcpu *vcpu, int mask);
	u32 (*get_interrupt_shadow)(struct kvm_vcpu *vcpu, int mask);
588 589
	void (*patch_hypercall)(struct kvm_vcpu *vcpu,
				unsigned char *hypercall_addr);
590
	void (*set_irq)(struct kvm_vcpu *vcpu);
591
	void (*set_nmi)(struct kvm_vcpu *vcpu);
592
	void (*queue_exception)(struct kvm_vcpu *vcpu, unsigned nr,
593 594
				bool has_error_code, u32 error_code,
				bool reinject);
A
Avi Kivity 已提交
595
	void (*cancel_injection)(struct kvm_vcpu *vcpu);
596
	int (*interrupt_allowed)(struct kvm_vcpu *vcpu);
597
	int (*nmi_allowed)(struct kvm_vcpu *vcpu);
J
Jan Kiszka 已提交
598 599
	bool (*get_nmi_mask)(struct kvm_vcpu *vcpu);
	void (*set_nmi_mask)(struct kvm_vcpu *vcpu, bool masked);
600 601 602
	void (*enable_nmi_window)(struct kvm_vcpu *vcpu);
	void (*enable_irq_window)(struct kvm_vcpu *vcpu);
	void (*update_cr8_intercept)(struct kvm_vcpu *vcpu, int tpr, int irr);
603
	int (*set_tss_addr)(struct kvm *kvm, unsigned int addr);
604
	int (*get_tdp_level)(void);
605
	u64 (*get_mt_mask)(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio);
606
	int (*get_lpage_level)(void);
607
	bool (*rdtscp_supported)(void);
Z
Zachary Amsden 已提交
608
	void (*adjust_tsc_offset)(struct kvm_vcpu *vcpu, s64 adjustment);
609

610 611
	void (*set_tdp_cr3)(struct kvm_vcpu *vcpu, unsigned long cr3);

612 613
	void (*set_supported_cpuid)(u32 func, struct kvm_cpuid_entry2 *entry);

614 615
	bool (*has_wbinvd_exit)(void);

616
	void (*set_tsc_khz)(struct kvm_vcpu *vcpu, u32 user_tsc_khz);
617 618
	void (*write_tsc_offset)(struct kvm_vcpu *vcpu, u64 offset);

619 620
	u64 (*compute_tsc_offset)(struct kvm_vcpu *vcpu, u64 target_tsc);

621
	void (*get_exit_info)(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2);
622 623 624 625 626

	int (*check_intercept)(struct kvm_vcpu *vcpu,
			       struct x86_instruction_info *info,
			       enum x86_intercept_stage stage);

627
	const struct trace_print_flags *exit_reasons_str;
628 629
};

630
struct kvm_arch_async_pf {
631
	u32 token;
632
	gfn_t gfn;
X
Xiao Guangrong 已提交
633
	unsigned long cr3;
634
	bool direct_map;
635 636
};

637 638
extern struct kvm_x86_ops *kvm_x86_ops;

639 640 641 642 643 644 645
int kvm_mmu_module_init(void);
void kvm_mmu_module_exit(void);

void kvm_mmu_destroy(struct kvm_vcpu *vcpu);
int kvm_mmu_create(struct kvm_vcpu *vcpu);
int kvm_mmu_setup(struct kvm_vcpu *vcpu);
void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte);
S
Sheng Yang 已提交
646
void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask,
647
		u64 dirty_mask, u64 nx_mask, u64 x_mask);
648 649 650 651

int kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot);
void kvm_mmu_zap_all(struct kvm *kvm);
652
unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm);
653 654
void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages);

655
int load_pdptrs(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, unsigned long cr3);
656

657
int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa,
658
			  const void *val, int bytes);
659 660
int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes,
		  gpa_t addr, unsigned long *ret);
661
u8 kvm_get_guest_memory_type(struct kvm_vcpu *vcpu, gfn_t gfn);
662 663

extern bool tdp_enabled;
664

665 666 667 668 669 670 671
/* control of guest tsc rate supported? */
extern bool kvm_has_tsc_control;
/* minimum supported tsc_khz for guests */
extern u32  kvm_min_guest_tsc_khz;
/* maximum supported tsc_khz for guests */
extern u32  kvm_max_guest_tsc_khz;

672 673 674 675 676 677
enum emulation_result {
	EMULATE_DONE,       /* no further processing */
	EMULATE_DO_MMIO,      /* kvm_run filled with mmio request */
	EMULATE_FAIL,         /* can't emulate this instruction */
};

678 679
#define EMULTYPE_NO_DECODE	    (1 << 0)
#define EMULTYPE_TRAP_UD	    (1 << 1)
680
#define EMULTYPE_SKIP		    (1 << 2)
681 682
int x86_emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2,
			    int emulation_type, void *insn, int insn_len);
683 684 685 686

static inline int emulate_instruction(struct kvm_vcpu *vcpu,
			int emulation_type)
{
687
	return x86_emulate_instruction(vcpu, 0, emulation_type, NULL, 0);
688 689
}

690
void kvm_enable_efer_bits(u64);
691 692 693 694 695
int kvm_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *data);
int kvm_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data);

struct x86_emulate_ctxt;

696
int kvm_fast_pio_out(struct kvm_vcpu *vcpu, int size, unsigned short port);
697 698
void kvm_emulate_cpuid(struct kvm_vcpu *vcpu);
int kvm_emulate_halt(struct kvm_vcpu *vcpu);
699
int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu);
700

701
void kvm_get_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg);
702
int kvm_load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, int seg);
703

704 705
int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason,
		    bool has_error_code, u32 error_code);
706

707
int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0);
708
int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3);
709
int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4);
A
Andre Przywara 已提交
710
int kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8);
711 712
int kvm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long val);
int kvm_get_dr(struct kvm_vcpu *vcpu, int dr, unsigned long *val);
713 714
unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu);
void kvm_lmsw(struct kvm_vcpu *vcpu, unsigned long msw);
715
void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l);
716
int kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr);
717 718 719 720

int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata);
int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data);

721 722 723
unsigned long kvm_get_rflags(struct kvm_vcpu *vcpu);
void kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags);

724 725
void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr);
void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
726 727
void kvm_requeue_exception(struct kvm_vcpu *vcpu, unsigned nr);
void kvm_requeue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
728
void kvm_inject_page_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault);
729 730 731
int kvm_read_guest_page_mmu(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
			    gfn_t gfn, void *data, int offset, int len,
			    u32 access);
732
void kvm_propagate_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault);
733
bool kvm_require_cpl(struct kvm_vcpu *vcpu, int required_cpl);
734

735
int kvm_pic_set_irq(void *opaque, int irq, int level);
736

737 738
void kvm_inject_nmi(struct kvm_vcpu *vcpu);

739
int fx_init(struct kvm_vcpu *vcpu);
740

741
void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu);
742
void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
743 744
		       const u8 *new, int bytes,
		       bool guest_initiated);
745 746 747 748
int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva);
void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu);
int kvm_mmu_load(struct kvm_vcpu *vcpu);
void kvm_mmu_unload(struct kvm_vcpu *vcpu);
749
void kvm_mmu_sync_roots(struct kvm_vcpu *vcpu);
750 751 752 753 754 755 756 757
gpa_t kvm_mmu_gva_to_gpa_read(struct kvm_vcpu *vcpu, gva_t gva,
			      struct x86_exception *exception);
gpa_t kvm_mmu_gva_to_gpa_fetch(struct kvm_vcpu *vcpu, gva_t gva,
			       struct x86_exception *exception);
gpa_t kvm_mmu_gva_to_gpa_write(struct kvm_vcpu *vcpu, gva_t gva,
			       struct x86_exception *exception);
gpa_t kvm_mmu_gva_to_gpa_system(struct kvm_vcpu *vcpu, gva_t gva,
				struct x86_exception *exception);
758 759 760

int kvm_emulate_hypercall(struct kvm_vcpu *vcpu);

761 762
int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t gva, u32 error_code,
		       void *insn, int insn_len);
M
Marcelo Tosatti 已提交
763
void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva);
764

765
void kvm_enable_tdp(void);
766
void kvm_disable_tdp(void);
767

768
int complete_pio(struct kvm_vcpu *vcpu);
769
bool kvm_check_iopl(struct kvm_vcpu *vcpu);
770 771 772 773 774 775 776 777

static inline struct kvm_mmu_page *page_header(hpa_t shadow_page)
{
	struct page *page = pfn_to_page(shadow_page >> PAGE_SHIFT);

	return (struct kvm_mmu_page *)page_private(page);
}

778
static inline u16 kvm_read_ldt(void)
779 780 781 782 783 784
{
	u16 ldt;
	asm("sldt %0" : "=g"(ldt));
	return ldt;
}

785
static inline void kvm_load_ldt(u16 sel)
786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804
{
	asm("lldt %0" : : "rm"(sel));
}

#ifdef CONFIG_X86_64
static inline unsigned long read_msr(unsigned long msr)
{
	u64 value;

	rdmsrl(msr, value);
	return value;
}
#endif

static inline u32 get_rdx_init_val(void)
{
	return 0x600; /* P6 family */
}

805 806 807 808 809
static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code)
{
	kvm_queue_exception_e(vcpu, GP_VECTOR, error_code);
}

810 811 812 813
#define TSS_IOPB_BASE_OFFSET 0x66
#define TSS_BASE_SIZE 0x68
#define TSS_IOPB_SIZE (65536 / 8)
#define TSS_REDIRECTION_SIZE (256 / 8)
814 815
#define RMODE_TSS_SIZE							\
	(TSS_BASE_SIZE + TSS_REDIRECTION_SIZE + TSS_IOPB_SIZE + 1)
816

817 818 819 820 821 822 823
enum {
	TASK_SWITCH_CALL = 0,
	TASK_SWITCH_IRET = 1,
	TASK_SWITCH_JMP = 2,
	TASK_SWITCH_GATE = 3,
};

824
#define HF_GIF_MASK		(1 << 0)
A
Alexander Graf 已提交
825 826
#define HF_HIF_MASK		(1 << 1)
#define HF_VINTR_MASK		(1 << 2)
827
#define HF_NMI_MASK		(1 << 3)
828
#define HF_IRET_MASK		(1 << 4)
829
#define HF_GUEST_MASK		(1 << 5) /* VCPU is in guest-mode */
830

831 832 833 834 835
/*
 * Hardware virtualization extension instructions may fault if a
 * reboot turns off virtualization while processes are running.
 * Trap the fault and ignore the instruction if that happens.
 */
836 837
asmlinkage void kvm_spurious_fault(void);
extern bool kvm_rebooting;
838

839
#define ____kvm_handle_fault_on_reboot(insn, cleanup_insn)	\
840
	"666: " insn "\n\t" \
841
	"668: \n\t"                           \
842
	".pushsection .fixup, \"ax\" \n" \
843
	"667: \n\t" \
844
	cleanup_insn "\n\t"		      \
845 846
	"cmpb $0, kvm_rebooting \n\t"	      \
	"jne 668b \n\t"      		      \
847
	__ASM_SIZE(push) " $666b \n\t"	      \
848
	"call kvm_spurious_fault \n\t"	      \
849 850
	".popsection \n\t" \
	".pushsection __ex_table, \"a\" \n\t" \
851
	_ASM_PTR " 666b, 667b \n\t" \
852 853
	".popsection"

854 855 856
#define __kvm_handle_fault_on_reboot(insn)		\
	____kvm_handle_fault_on_reboot(insn, "")

857 858 859
#define KVM_ARCH_WANT_MMU_NOTIFIER
int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
int kvm_age_hva(struct kvm *kvm, unsigned long hva);
A
Andrea Arcangeli 已提交
860
int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
861
void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
862
int cpuid_maxphyaddr(struct kvm_vcpu *vcpu);
863 864
int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu);
865
int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
866

A
Avi Kivity 已提交
867
void kvm_define_shared_msr(unsigned index, u32 msr);
868
void kvm_set_shared_msr(unsigned index, u64 val, u64 mask);
A
Avi Kivity 已提交
869

J
Jan Kiszka 已提交
870 871
bool kvm_is_linear_rip(struct kvm_vcpu *vcpu, unsigned long linear_rip);

872 873 874 875
void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu,
				     struct kvm_async_pf *work);
void kvm_arch_async_page_present(struct kvm_vcpu *vcpu,
				 struct kvm_async_pf *work);
G
Gleb Natapov 已提交
876 877
void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu,
			       struct kvm_async_pf *work);
878
bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu);
879 880
extern bool kvm_find_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn);

881 882
void kvm_complete_insn_gp(struct kvm_vcpu *vcpu, int err);

H
H. Peter Anvin 已提交
883
#endif /* _ASM_X86_KVM_HOST_H */