cpuid.c 21.8 KB
Newer Older
A
Avi Kivity 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
/*
 * Kernel-based Virtual Machine driver for Linux
 * cpuid support routines
 *
 * derived from arch/x86/kvm/x86.c
 *
 * Copyright 2011 Red Hat, Inc. and/or its affiliates.
 * Copyright IBM Corporation, 2008
 *
 * This work is licensed under the terms of the GNU GPL, version 2.  See
 * the COPYING file in the top-level directory.
 *
 */

#include <linux/kvm_host.h>
#include <linux/module.h>
17 18
#include <linux/vmalloc.h>
#include <linux/uaccess.h>
19
#include <asm/fpu/internal.h> /* For use_eager_fpu.  Ugh! */
A
Avi Kivity 已提交
20
#include <asm/user.h>
21
#include <asm/fpu/xstate.h>
A
Avi Kivity 已提交
22 23 24 25
#include "cpuid.h"
#include "lapic.h"
#include "mmu.h"
#include "trace.h"
26
#include "pmu.h"
A
Avi Kivity 已提交
27

28
static u32 xstate_required_size(u64 xstate_bv, bool compacted)
29 30 31 32
{
	int feature_bit = 0;
	u32 ret = XSAVE_HDR_SIZE + XSAVE_HDR_OFFSET;

33
	xstate_bv &= XSTATE_EXTEND_MASK;
34 35
	while (xstate_bv) {
		if (xstate_bv & 0x1) {
36
		        u32 eax, ebx, ecx, edx, offset;
37
		        cpuid_count(0xD, feature_bit, &eax, &ebx, &ecx, &edx);
38 39
			offset = compacted ? ret : ebx;
			ret = max(ret, offset + eax);
40 41 42 43 44 45 46 47 48
		}

		xstate_bv >>= 1;
		feature_bit++;
	}

	return ret;
}

49 50 51 52
u64 kvm_supported_xcr0(void)
{
	u64 xcr0 = KVM_SUPPORTED_XCR0 & host_xcr0;

53
	if (!kvm_x86_ops->mpx_supported())
54 55 56 57 58
		xcr0 &= ~(XSTATE_BNDREGS | XSTATE_BNDCSR);

	return xcr0;
}

59 60
#define F(x) bit(X86_FEATURE_##x)

61
int kvm_update_cpuid(struct kvm_vcpu *vcpu)
A
Avi Kivity 已提交
62 63 64 65 66 67
{
	struct kvm_cpuid_entry2 *best;
	struct kvm_lapic *apic = vcpu->arch.apic;

	best = kvm_find_cpuid_entry(vcpu, 1, 0);
	if (!best)
68
		return 0;
A
Avi Kivity 已提交
69 70 71

	/* Update OSXSAVE bit */
	if (cpu_has_xsave && best->function == 0x1) {
72
		best->ecx &= ~F(OSXSAVE);
A
Avi Kivity 已提交
73
		if (kvm_read_cr4_bits(vcpu, X86_CR4_OSXSAVE))
74
			best->ecx |= F(OSXSAVE);
A
Avi Kivity 已提交
75 76 77
	}

	if (apic) {
78
		if (best->ecx & F(TSC_DEADLINE_TIMER))
A
Avi Kivity 已提交
79 80 81 82
			apic->lapic_timer.timer_mode_mask = 3 << 17;
		else
			apic->lapic_timer.timer_mode_mask = 1 << 17;
	}
83

84
	best = kvm_find_cpuid_entry(vcpu, 0xD, 0);
85
	if (!best) {
86
		vcpu->arch.guest_supported_xcr0 = 0;
87 88
		vcpu->arch.guest_xstate_size = XSAVE_HDR_SIZE + XSAVE_HDR_OFFSET;
	} else {
89 90
		vcpu->arch.guest_supported_xcr0 =
			(best->eax | ((u64)best->edx << 32)) &
91
			kvm_supported_xcr0();
92
		vcpu->arch.guest_xstate_size = best->ebx =
93
			xstate_required_size(vcpu->arch.xcr0, false);
94
	}
95

96 97 98 99
	best = kvm_find_cpuid_entry(vcpu, 0xD, 1);
	if (best && (best->eax & (F(XSAVES) | F(XSAVEC))))
		best->ebx = xstate_required_size(vcpu->arch.xcr0, true);

100
	vcpu->arch.eager_fpu = use_eager_fpu() || guest_cpuid_has_mpx(vcpu);
101 102
	if (vcpu->arch.eager_fpu)
		kvm_x86_ops->fpu_activate(vcpu);
103

104 105 106 107 108 109 110 111 112
	/*
	 * The existing code assumes virtual address is 48-bit in the canonical
	 * address checks; exit if it is ever changed.
	 */
	best = kvm_find_cpuid_entry(vcpu, 0x80000008, 0);
	if (best && ((best->eax & 0xff00) >> 8) != 48 &&
		((best->eax & 0xff00) >> 8) != 0)
		return -EINVAL;

113 114 115
	/* Update physical-address width */
	vcpu->arch.maxphyaddr = cpuid_query_maxphyaddr(vcpu);

116
	kvm_pmu_refresh(vcpu);
117
	return 0;
A
Avi Kivity 已提交
118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140
}

static int is_efer_nx(void)
{
	unsigned long long efer = 0;

	rdmsrl_safe(MSR_EFER, &efer);
	return efer & EFER_NX;
}

static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu)
{
	int i;
	struct kvm_cpuid_entry2 *e, *entry;

	entry = NULL;
	for (i = 0; i < vcpu->arch.cpuid_nent; ++i) {
		e = &vcpu->arch.cpuid_entries[i];
		if (e->function == 0x80000001) {
			entry = e;
			break;
		}
	}
141 142
	if (entry && (entry->edx & F(NX)) && !is_efer_nx()) {
		entry->edx &= ~F(NX);
A
Avi Kivity 已提交
143 144 145 146
		printk(KERN_INFO "kvm: guest NX capability removed\n");
	}
}

147 148 149 150 151 152 153 154 155 156 157 158 159 160 161
int cpuid_query_maxphyaddr(struct kvm_vcpu *vcpu)
{
	struct kvm_cpuid_entry2 *best;

	best = kvm_find_cpuid_entry(vcpu, 0x80000000, 0);
	if (!best || best->eax < 0x80000008)
		goto not_found;
	best = kvm_find_cpuid_entry(vcpu, 0x80000008, 0);
	if (best)
		return best->eax & 0xff;
not_found:
	return 36;
}
EXPORT_SYMBOL_GPL(cpuid_query_maxphyaddr);

A
Avi Kivity 已提交
162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196
/* when an old userspace process fills a new kernel module */
int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu,
			     struct kvm_cpuid *cpuid,
			     struct kvm_cpuid_entry __user *entries)
{
	int r, i;
	struct kvm_cpuid_entry *cpuid_entries;

	r = -E2BIG;
	if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
		goto out;
	r = -ENOMEM;
	cpuid_entries = vmalloc(sizeof(struct kvm_cpuid_entry) * cpuid->nent);
	if (!cpuid_entries)
		goto out;
	r = -EFAULT;
	if (copy_from_user(cpuid_entries, entries,
			   cpuid->nent * sizeof(struct kvm_cpuid_entry)))
		goto out_free;
	for (i = 0; i < cpuid->nent; i++) {
		vcpu->arch.cpuid_entries[i].function = cpuid_entries[i].function;
		vcpu->arch.cpuid_entries[i].eax = cpuid_entries[i].eax;
		vcpu->arch.cpuid_entries[i].ebx = cpuid_entries[i].ebx;
		vcpu->arch.cpuid_entries[i].ecx = cpuid_entries[i].ecx;
		vcpu->arch.cpuid_entries[i].edx = cpuid_entries[i].edx;
		vcpu->arch.cpuid_entries[i].index = 0;
		vcpu->arch.cpuid_entries[i].flags = 0;
		vcpu->arch.cpuid_entries[i].padding[0] = 0;
		vcpu->arch.cpuid_entries[i].padding[1] = 0;
		vcpu->arch.cpuid_entries[i].padding[2] = 0;
	}
	vcpu->arch.cpuid_nent = cpuid->nent;
	cpuid_fix_nx_cap(vcpu);
	kvm_apic_set_version(vcpu);
	kvm_x86_ops->cpuid_update(vcpu);
197
	r = kvm_update_cpuid(vcpu);
A
Avi Kivity 已提交
198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220

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

int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu,
			      struct kvm_cpuid2 *cpuid,
			      struct kvm_cpuid_entry2 __user *entries)
{
	int r;

	r = -E2BIG;
	if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
		goto out;
	r = -EFAULT;
	if (copy_from_user(&vcpu->arch.cpuid_entries, entries,
			   cpuid->nent * sizeof(struct kvm_cpuid_entry2)))
		goto out;
	vcpu->arch.cpuid_nent = cpuid->nent;
	kvm_apic_set_version(vcpu);
	kvm_x86_ops->cpuid_update(vcpu);
221
	r = kvm_update_cpuid(vcpu);
A
Avi Kivity 已提交
222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260
out:
	return r;
}

int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu,
			      struct kvm_cpuid2 *cpuid,
			      struct kvm_cpuid_entry2 __user *entries)
{
	int r;

	r = -E2BIG;
	if (cpuid->nent < vcpu->arch.cpuid_nent)
		goto out;
	r = -EFAULT;
	if (copy_to_user(entries, &vcpu->arch.cpuid_entries,
			 vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2)))
		goto out;
	return 0;

out:
	cpuid->nent = vcpu->arch.cpuid_nent;
	return r;
}

static void cpuid_mask(u32 *word, int wordnum)
{
	*word &= boot_cpu_data.x86_capability[wordnum];
}

static void do_cpuid_1_ent(struct kvm_cpuid_entry2 *entry, u32 function,
			   u32 index)
{
	entry->function = function;
	entry->index = index;
	cpuid_count(entry->function, entry->index,
		    &entry->eax, &entry->ebx, &entry->ecx, &entry->edx);
	entry->flags = 0;
}

B
Borislav Petkov 已提交
261 262 263
static int __do_cpuid_ent_emulated(struct kvm_cpuid_entry2 *entry,
				   u32 func, u32 index, int *nent, int maxnent)
{
B
Borislav Petkov 已提交
264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279
	switch (func) {
	case 0:
		entry->eax = 1;		/* only one leaf currently */
		++*nent;
		break;
	case 1:
		entry->ecx = F(MOVBE);
		++*nent;
		break;
	default:
		break;
	}

	entry->function = func;
	entry->index = index;

B
Borislav Petkov 已提交
280 281 282 283 284
	return 0;
}

static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
				 u32 index, int *nent, int maxnent)
A
Avi Kivity 已提交
285
{
286
	int r;
A
Avi Kivity 已提交
287 288 289 290 291 292 293 294 295 296
	unsigned f_nx = is_efer_nx() ? F(NX) : 0;
#ifdef CONFIG_X86_64
	unsigned f_gbpages = (kvm_x86_ops->get_lpage_level() == PT_PDPE_LEVEL)
				? F(GBPAGES) : 0;
	unsigned f_lm = F(LM);
#else
	unsigned f_gbpages = 0;
	unsigned f_lm = 0;
#endif
	unsigned f_rdtscp = kvm_x86_ops->rdtscp_supported() ? F(RDTSCP) : 0;
297
	unsigned f_invpcid = kvm_x86_ops->invpcid_supported() ? F(INVPCID) : 0;
298
	unsigned f_mpx = kvm_x86_ops->mpx_supported() ? F(MPX) : 0;
299
	unsigned f_xsaves = kvm_x86_ops->xsaves_supported() ? F(XSAVES) : 0;
A
Avi Kivity 已提交
300 301 302 303 304 305 306

	/* cpuid 1.edx */
	const u32 kvm_supported_word0_x86_features =
		F(FPU) | F(VME) | F(DE) | F(PSE) |
		F(TSC) | F(MSR) | F(PAE) | F(MCE) |
		F(CX8) | F(APIC) | 0 /* Reserved */ | F(SEP) |
		F(MTRR) | F(PGE) | F(MCA) | F(CMOV) |
307
		F(PAT) | F(PSE36) | 0 /* PSN */ | F(CLFLUSH) |
A
Avi Kivity 已提交
308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
		0 /* Reserved, DS, ACPI */ | F(MMX) |
		F(FXSR) | F(XMM) | F(XMM2) | F(SELFSNOOP) |
		0 /* HTT, TM, Reserved, PBE */;
	/* cpuid 0x80000001.edx */
	const u32 kvm_supported_word1_x86_features =
		F(FPU) | F(VME) | F(DE) | F(PSE) |
		F(TSC) | F(MSR) | F(PAE) | F(MCE) |
		F(CX8) | F(APIC) | 0 /* Reserved */ | F(SYSCALL) |
		F(MTRR) | F(PGE) | F(MCA) | F(CMOV) |
		F(PAT) | F(PSE36) | 0 /* Reserved */ |
		f_nx | 0 /* Reserved */ | F(MMXEXT) | F(MMX) |
		F(FXSR) | F(FXSR_OPT) | f_gbpages | f_rdtscp |
		0 /* Reserved */ | f_lm | F(3DNOWEXT) | F(3DNOW);
	/* cpuid 1.ecx */
	const u32 kvm_supported_word4_x86_features =
323 324
		/* NOTE: MONITOR (and MWAIT) are emulated as NOP,
		 * but *not* advertised to guests via CPUID ! */
A
Avi Kivity 已提交
325 326 327
		F(XMM3) | F(PCLMULQDQ) | 0 /* DTES64, MONITOR */ |
		0 /* DS-CPL, VMX, SMX, EST */ |
		0 /* TM2 */ | F(SSSE3) | 0 /* CNXT-ID */ | 0 /* Reserved */ |
328
		F(FMA) | F(CX16) | 0 /* xTPR Update, PDCM */ |
329
		F(PCID) | 0 /* Reserved, DCA */ | F(XMM4_1) |
A
Avi Kivity 已提交
330 331 332 333 334 335 336
		F(XMM4_2) | F(X2APIC) | F(MOVBE) | F(POPCNT) |
		0 /* Reserved*/ | F(AES) | F(XSAVE) | 0 /* OSXSAVE */ | F(AVX) |
		F(F16C) | F(RDRAND);
	/* cpuid 0x80000001.ecx */
	const u32 kvm_supported_word6_x86_features =
		F(LAHF_LM) | F(CMP_LEGACY) | 0 /*SVM*/ | 0 /* ExtApicSpace */ |
		F(CR8_LEGACY) | F(ABM) | F(SSE4A) | F(MISALIGNSSE) |
337
		F(3DNOWPREFETCH) | F(OSVW) | 0 /* IBS */ | F(XOP) |
A
Avi Kivity 已提交
338 339 340 341 342 343 344 345 346 347
		0 /* SKINIT, WDT, LWP */ | F(FMA4) | F(TBM);

	/* cpuid 0xC0000001.edx */
	const u32 kvm_supported_word5_x86_features =
		F(XSTORE) | F(XSTORE_EN) | F(XCRYPT) | F(XCRYPT_EN) |
		F(ACE2) | F(ACE2_EN) | F(PHE) | F(PHE_EN) |
		F(PMM) | F(PMM_EN);

	/* cpuid 7.0.ebx */
	const u32 kvm_supported_word9_x86_features =
348
		F(FSGSBASE) | F(BMI1) | F(HLE) | F(AVX2) | F(SMEP) |
349
		F(BMI2) | F(ERMS) | f_invpcid | F(RTM) | f_mpx | F(RDSEED) |
350 351
		F(ADX) | F(SMAP) | F(AVX512F) | F(AVX512PF) | F(AVX512ER) |
		F(AVX512CD);
A
Avi Kivity 已提交
352

P
Paolo Bonzini 已提交
353 354
	/* cpuid 0xD.1.eax */
	const u32 kvm_supported_word10_x86_features =
355
		F(XSAVEOPT) | F(XSAVEC) | F(XGETBV1) | f_xsaves;
P
Paolo Bonzini 已提交
356

A
Avi Kivity 已提交
357 358
	/* all calls to cpuid_count() should be made on the same cpu */
	get_cpu();
359 360 361 362 363 364

	r = -E2BIG;

	if (*nent >= maxnent)
		goto out;

A
Avi Kivity 已提交
365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389
	do_cpuid_1_ent(entry, function, index);
	++*nent;

	switch (function) {
	case 0:
		entry->eax = min(entry->eax, (u32)0xd);
		break;
	case 1:
		entry->edx &= kvm_supported_word0_x86_features;
		cpuid_mask(&entry->edx, 0);
		entry->ecx &= kvm_supported_word4_x86_features;
		cpuid_mask(&entry->ecx, 4);
		/* we support x2apic emulation even if host does not support
		 * it since we emulate x2apic in software */
		entry->ecx |= F(X2APIC);
		break;
	/* function 2 entries are STATEFUL. That is, repeated cpuid commands
	 * may return different values. This forces us to get_cpu() before
	 * issuing the first command, and also to emulate this annoying behavior
	 * in kvm_emulate_cpuid() using KVM_CPUID_FLAG_STATE_READ_NEXT */
	case 2: {
		int t, times = entry->eax & 0xff;

		entry->flags |= KVM_CPUID_FLAG_STATEFUL_FUNC;
		entry->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT;
390 391 392 393
		for (t = 1; t < times; ++t) {
			if (*nent >= maxnent)
				goto out;

A
Avi Kivity 已提交
394 395 396 397 398 399 400 401 402 403 404 405
			do_cpuid_1_ent(&entry[t], function, 0);
			entry[t].flags |= KVM_CPUID_FLAG_STATEFUL_FUNC;
			++*nent;
		}
		break;
	}
	/* function 4 has additional index. */
	case 4: {
		int i, cache_type;

		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
		/* read more entries until cache_type is zero */
406 407 408 409
		for (i = 1; ; ++i) {
			if (*nent >= maxnent)
				goto out;

A
Avi Kivity 已提交
410 411 412 413 414 415 416 417 418 419
			cache_type = entry[i - 1].eax & 0x1f;
			if (!cache_type)
				break;
			do_cpuid_1_ent(&entry[i], function, i);
			entry[i].flags |=
			       KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
			++*nent;
		}
		break;
	}
J
Jan Kiszka 已提交
420 421 422 423 424 425
	case 6: /* Thermal management */
		entry->eax = 0x4; /* allow ARAT */
		entry->ebx = 0;
		entry->ecx = 0;
		entry->edx = 0;
		break;
A
Avi Kivity 已提交
426 427
	case 7: {
		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
G
Guo Chao 已提交
428
		/* Mask ebx against host capability word 9 */
A
Avi Kivity 已提交
429 430 431
		if (index == 0) {
			entry->ebx &= kvm_supported_word9_x86_features;
			cpuid_mask(&entry->ebx, 9);
W
Will Auld 已提交
432 433
			// TSC_ADJUST is emulated
			entry->ebx |= F(TSC_ADJUST);
A
Avi Kivity 已提交
434 435 436 437 438 439 440 441 442
		} else
			entry->ebx = 0;
		entry->eax = 0;
		entry->ecx = 0;
		entry->edx = 0;
		break;
	}
	case 9:
		break;
443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471
	case 0xa: { /* Architectural Performance Monitoring */
		struct x86_pmu_capability cap;
		union cpuid10_eax eax;
		union cpuid10_edx edx;

		perf_get_x86_pmu_capability(&cap);

		/*
		 * Only support guest architectural pmu on a host
		 * with architectural pmu.
		 */
		if (!cap.version)
			memset(&cap, 0, sizeof(cap));

		eax.split.version_id = min(cap.version, 2);
		eax.split.num_counters = cap.num_counters_gp;
		eax.split.bit_width = cap.bit_width_gp;
		eax.split.mask_length = cap.events_mask_len;

		edx.split.num_counters_fixed = cap.num_counters_fixed;
		edx.split.bit_width_fixed = cap.bit_width_fixed;
		edx.split.reserved = 0;

		entry->eax = eax.full;
		entry->ebx = cap.events_mask;
		entry->ecx = 0;
		entry->edx = edx.full;
		break;
	}
A
Avi Kivity 已提交
472 473 474 475 476 477
	/* function 0xb has additional index. */
	case 0xb: {
		int i, level_type;

		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
		/* read more entries until level_type is zero */
478 479 480 481
		for (i = 1; ; ++i) {
			if (*nent >= maxnent)
				goto out;

A
Avi Kivity 已提交
482 483 484 485 486 487 488 489 490 491 492 493
			level_type = entry[i - 1].ecx & 0xff00;
			if (!level_type)
				break;
			do_cpuid_1_ent(&entry[i], function, i);
			entry[i].flags |=
			       KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
			++*nent;
		}
		break;
	}
	case 0xd: {
		int idx, i;
494
		u64 supported = kvm_supported_xcr0();
A
Avi Kivity 已提交
495

496
		entry->eax &= supported;
497 498
		entry->ebx = xstate_required_size(supported, false);
		entry->ecx = entry->ebx;
499
		entry->edx &= supported >> 32;
A
Avi Kivity 已提交
500
		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
P
Paolo Bonzini 已提交
501 502 503
		if (!supported)
			break;

504
		for (idx = 1, i = 1; idx < 64; ++idx) {
505
			u64 mask = ((u64)1 << idx);
506 507 508
			if (*nent >= maxnent)
				goto out;

A
Avi Kivity 已提交
509
			do_cpuid_1_ent(&entry[i], function, idx);
510
			if (idx == 1) {
P
Paolo Bonzini 已提交
511
				entry[i].eax &= kvm_supported_word10_x86_features;
512 513 514 515 516
				entry[i].ebx = 0;
				if (entry[i].eax & (F(XSAVES)|F(XSAVEC)))
					entry[i].ebx =
						xstate_required_size(supported,
								     true);
517 518 519 520 521 522 523 524
			} else {
				if (entry[i].eax == 0 || !(supported & mask))
					continue;
				if (WARN_ON_ONCE(entry[i].ecx & 1))
					continue;
			}
			entry[i].ecx = 0;
			entry[i].edx = 0;
A
Avi Kivity 已提交
525 526 527 528 529 530 531 532
			entry[i].flags |=
			       KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
			++*nent;
			++i;
		}
		break;
	}
	case KVM_CPUID_SIGNATURE: {
533 534
		static const char signature[12] = "KVMKVMKVM\0\0";
		const u32 *sigptr = (const u32 *)signature;
535
		entry->eax = KVM_CPUID_FEATURES;
A
Avi Kivity 已提交
536 537 538 539 540 541 542 543 544 545
		entry->ebx = sigptr[0];
		entry->ecx = sigptr[1];
		entry->edx = sigptr[2];
		break;
	}
	case KVM_CPUID_FEATURES:
		entry->eax = (1 << KVM_FEATURE_CLOCKSOURCE) |
			     (1 << KVM_FEATURE_NOP_IO_DELAY) |
			     (1 << KVM_FEATURE_CLOCKSOURCE2) |
			     (1 << KVM_FEATURE_ASYNC_PF) |
546
			     (1 << KVM_FEATURE_PV_EOI) |
547 548
			     (1 << KVM_FEATURE_CLOCKSOURCE_STABLE_BIT) |
			     (1 << KVM_FEATURE_PV_UNHALT);
A
Avi Kivity 已提交
549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565

		if (sched_info_on())
			entry->eax |= (1 << KVM_FEATURE_STEAL_TIME);

		entry->ebx = 0;
		entry->ecx = 0;
		entry->edx = 0;
		break;
	case 0x80000000:
		entry->eax = min(entry->eax, 0x8000001a);
		break;
	case 0x80000001:
		entry->edx &= kvm_supported_word1_x86_features;
		cpuid_mask(&entry->edx, 1);
		entry->ecx &= kvm_supported_word6_x86_features;
		cpuid_mask(&entry->ecx, 6);
		break;
566 567 568 569 570 571 572
	case 0x80000007: /* Advanced power management */
		/* invariant TSC is CPUID.80000007H:EDX[8] */
		entry->edx &= (1 << 8);
		/* mask against host */
		entry->edx &= boot_cpu_data.x86_power;
		entry->eax = entry->ebx = entry->ecx = 0;
		break;
A
Avi Kivity 已提交
573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611
	case 0x80000008: {
		unsigned g_phys_as = (entry->eax >> 16) & 0xff;
		unsigned virt_as = max((entry->eax >> 8) & 0xff, 48U);
		unsigned phys_as = entry->eax & 0xff;

		if (!g_phys_as)
			g_phys_as = phys_as;
		entry->eax = g_phys_as | (virt_as << 8);
		entry->ebx = entry->edx = 0;
		break;
	}
	case 0x80000019:
		entry->ecx = entry->edx = 0;
		break;
	case 0x8000001a:
		break;
	case 0x8000001d:
		break;
	/*Add support for Centaur's CPUID instruction*/
	case 0xC0000000:
		/*Just support up to 0xC0000004 now*/
		entry->eax = min(entry->eax, 0xC0000004);
		break;
	case 0xC0000001:
		entry->edx &= kvm_supported_word5_x86_features;
		cpuid_mask(&entry->edx, 5);
		break;
	case 3: /* Processor serial number */
	case 5: /* MONITOR/MWAIT */
	case 0xC0000002:
	case 0xC0000003:
	case 0xC0000004:
	default:
		entry->eax = entry->ebx = entry->ecx = entry->edx = 0;
		break;
	}

	kvm_x86_ops->set_supported_cpuid(function, entry);

612 613 614
	r = 0;

out:
A
Avi Kivity 已提交
615
	put_cpu();
616 617

	return r;
A
Avi Kivity 已提交
618 619
}

B
Borislav Petkov 已提交
620 621 622 623 624 625 626 627 628
static int do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 func,
			u32 idx, int *nent, int maxnent, unsigned int type)
{
	if (type == KVM_GET_EMULATED_CPUID)
		return __do_cpuid_ent_emulated(entry, func, idx, nent, maxnent);

	return __do_cpuid_ent(entry, func, idx, nent, maxnent);
}

A
Avi Kivity 已提交
629 630
#undef F

631 632 633 634
struct kvm_cpuid_param {
	u32 func;
	u32 idx;
	bool has_leaf_count;
635
	bool (*qualifier)(const struct kvm_cpuid_param *param);
636 637
};

638
static bool is_centaur_cpu(const struct kvm_cpuid_param *param)
639 640 641 642
{
	return boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR;
}

B
Borislav Petkov 已提交
643 644 645 646
static bool sanity_check_entries(struct kvm_cpuid_entry2 __user *entries,
				 __u32 num_entries, unsigned int ioctl_type)
{
	int i;
B
Borislav Petkov 已提交
647
	__u32 pad[3];
B
Borislav Petkov 已提交
648 649 650 651 652 653 654 655 656 657 658 659 660

	if (ioctl_type != KVM_GET_EMULATED_CPUID)
		return false;

	/*
	 * We want to make sure that ->padding is being passed clean from
	 * userspace in case we want to use it for something in the future.
	 *
	 * Sadly, this wasn't enforced for KVM_GET_SUPPORTED_CPUID and so we
	 * have to give ourselves satisfied only with the emulated side. /me
	 * sheds a tear.
	 */
	for (i = 0; i < num_entries; i++) {
B
Borislav Petkov 已提交
661 662 663 664
		if (copy_from_user(pad, entries[i].padding, sizeof(pad)))
			return true;

		if (pad[0] || pad[1] || pad[2])
B
Borislav Petkov 已提交
665 666 667 668 669 670 671 672
			return true;
	}
	return false;
}

int kvm_dev_ioctl_get_cpuid(struct kvm_cpuid2 *cpuid,
			    struct kvm_cpuid_entry2 __user *entries,
			    unsigned int type)
A
Avi Kivity 已提交
673 674
{
	struct kvm_cpuid_entry2 *cpuid_entries;
675
	int limit, nent = 0, r = -E2BIG, i;
A
Avi Kivity 已提交
676
	u32 func;
677
	static const struct kvm_cpuid_param param[] = {
678 679 680 681 682 683
		{ .func = 0, .has_leaf_count = true },
		{ .func = 0x80000000, .has_leaf_count = true },
		{ .func = 0xC0000000, .qualifier = is_centaur_cpu, .has_leaf_count = true },
		{ .func = KVM_CPUID_SIGNATURE },
		{ .func = KVM_CPUID_FEATURES },
	};
A
Avi Kivity 已提交
684 685 686 687 688

	if (cpuid->nent < 1)
		goto out;
	if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
		cpuid->nent = KVM_MAX_CPUID_ENTRIES;
B
Borislav Petkov 已提交
689 690 691 692

	if (sanity_check_entries(entries, cpuid->nent, type))
		return -EINVAL;

A
Avi Kivity 已提交
693
	r = -ENOMEM;
B
Borislav Petkov 已提交
694
	cpuid_entries = vzalloc(sizeof(struct kvm_cpuid_entry2) * cpuid->nent);
A
Avi Kivity 已提交
695 696 697
	if (!cpuid_entries)
		goto out;

698 699
	r = 0;
	for (i = 0; i < ARRAY_SIZE(param); i++) {
700
		const struct kvm_cpuid_param *ent = &param[i];
A
Avi Kivity 已提交
701

702 703
		if (ent->qualifier && !ent->qualifier(ent))
			continue;
A
Avi Kivity 已提交
704

705
		r = do_cpuid_ent(&cpuid_entries[nent], ent->func, ent->idx,
B
Borislav Petkov 已提交
706
				&nent, cpuid->nent, type);
A
Avi Kivity 已提交
707

708
		if (r)
A
Avi Kivity 已提交
709 710
			goto out_free;

711 712 713
		if (!ent->has_leaf_count)
			continue;

A
Avi Kivity 已提交
714
		limit = cpuid_entries[nent - 1].eax;
715 716
		for (func = ent->func + 1; func <= limit && nent < cpuid->nent && r == 0; ++func)
			r = do_cpuid_ent(&cpuid_entries[nent], func, ent->idx,
B
Borislav Petkov 已提交
717
				     &nent, cpuid->nent, type);
A
Avi Kivity 已提交
718

719
		if (r)
A
Avi Kivity 已提交
720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809
			goto out_free;
	}

	r = -EFAULT;
	if (copy_to_user(entries, cpuid_entries,
			 nent * sizeof(struct kvm_cpuid_entry2)))
		goto out_free;
	cpuid->nent = nent;
	r = 0;

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

static int move_to_next_stateful_cpuid_entry(struct kvm_vcpu *vcpu, int i)
{
	struct kvm_cpuid_entry2 *e = &vcpu->arch.cpuid_entries[i];
	int j, nent = vcpu->arch.cpuid_nent;

	e->flags &= ~KVM_CPUID_FLAG_STATE_READ_NEXT;
	/* when no next entry is found, the current entry[i] is reselected */
	for (j = i + 1; ; j = (j + 1) % nent) {
		struct kvm_cpuid_entry2 *ej = &vcpu->arch.cpuid_entries[j];
		if (ej->function == e->function) {
			ej->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT;
			return j;
		}
	}
	return 0; /* silence gcc, even though control never reaches here */
}

/* find an entry with matching function, matching index (if needed), and that
 * should be read next (if it's stateful) */
static int is_matching_cpuid_entry(struct kvm_cpuid_entry2 *e,
	u32 function, u32 index)
{
	if (e->function != function)
		return 0;
	if ((e->flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX) && e->index != index)
		return 0;
	if ((e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) &&
	    !(e->flags & KVM_CPUID_FLAG_STATE_READ_NEXT))
		return 0;
	return 1;
}

struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu,
					      u32 function, u32 index)
{
	int i;
	struct kvm_cpuid_entry2 *best = NULL;

	for (i = 0; i < vcpu->arch.cpuid_nent; ++i) {
		struct kvm_cpuid_entry2 *e;

		e = &vcpu->arch.cpuid_entries[i];
		if (is_matching_cpuid_entry(e, function, index)) {
			if (e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC)
				move_to_next_stateful_cpuid_entry(vcpu, i);
			best = e;
			break;
		}
	}
	return best;
}
EXPORT_SYMBOL_GPL(kvm_find_cpuid_entry);

/*
 * If no match is found, check whether we exceed the vCPU's limit
 * and return the content of the highest valid _standard_ leaf instead.
 * This is to satisfy the CPUID specification.
 */
static struct kvm_cpuid_entry2* check_cpuid_limit(struct kvm_vcpu *vcpu,
                                                  u32 function, u32 index)
{
	struct kvm_cpuid_entry2 *maxlevel;

	maxlevel = kvm_find_cpuid_entry(vcpu, function & 0x80000000, 0);
	if (!maxlevel || maxlevel->eax >= function)
		return NULL;
	if (function & 0x80000000) {
		maxlevel = kvm_find_cpuid_entry(vcpu, 0, 0);
		if (!maxlevel)
			return NULL;
	}
	return kvm_find_cpuid_entry(vcpu, maxlevel->eax, index);
}

810
void kvm_cpuid(struct kvm_vcpu *vcpu, u32 *eax, u32 *ebx, u32 *ecx, u32 *edx)
A
Avi Kivity 已提交
811
{
812
	u32 function = *eax, index = *ecx;
A
Avi Kivity 已提交
813 814 815 816 817 818 819
	struct kvm_cpuid_entry2 *best;

	best = kvm_find_cpuid_entry(vcpu, function, index);

	if (!best)
		best = check_cpuid_limit(vcpu, function, index);

820 821 822 823 824 825
	/*
	 * Perfmon not yet supported for L2 guest.
	 */
	if (is_guest_mode(vcpu) && function == 0xa)
		best = NULL;

A
Avi Kivity 已提交
826
	if (best) {
827 828 829 830 831 832
		*eax = best->eax;
		*ebx = best->ebx;
		*ecx = best->ecx;
		*edx = best->edx;
	} else
		*eax = *ebx = *ecx = *edx = 0;
833
	trace_kvm_cpuid(function, *eax, *ebx, *ecx, *edx);
834
}
835
EXPORT_SYMBOL_GPL(kvm_cpuid);
836 837 838 839 840 841 842 843 844 845 846 847

void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
{
	u32 function, eax, ebx, ecx, edx;

	function = eax = kvm_register_read(vcpu, VCPU_REGS_RAX);
	ecx = kvm_register_read(vcpu, VCPU_REGS_RCX);
	kvm_cpuid(vcpu, &eax, &ebx, &ecx, &edx);
	kvm_register_write(vcpu, VCPU_REGS_RAX, eax);
	kvm_register_write(vcpu, VCPU_REGS_RBX, ebx);
	kvm_register_write(vcpu, VCPU_REGS_RCX, ecx);
	kvm_register_write(vcpu, VCPU_REGS_RDX, edx);
A
Avi Kivity 已提交
848 849 850
	kvm_x86_ops->skip_emulated_instruction(vcpu);
}
EXPORT_SYMBOL_GPL(kvm_emulate_cpuid);