guest.c 24.8 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0-only
2 3 4 5 6 7 8 9 10
/*
 * Copyright (C) 2012,2013 - ARM Ltd
 * Author: Marc Zyngier <marc.zyngier@arm.com>
 *
 * Derived from arch/arm/kvm/guest.c:
 * Copyright (C) 2012 - Virtual Open Systems and Columbia University
 * Author: Christoffer Dall <c.dall@virtualopensystems.com>
 */

11
#include <linux/bits.h>
12 13
#include <linux/errno.h>
#include <linux/err.h>
14
#include <linux/nospec.h>
15 16
#include <linux/kvm_host.h>
#include <linux/module.h>
17
#include <linux/stddef.h>
18
#include <linux/string.h>
19 20
#include <linux/vmalloc.h>
#include <linux/fs.h>
21
#include <kvm/arm_psci.h>
22
#include <asm/cputype.h>
23
#include <linux/uaccess.h>
24
#include <asm/fpsimd.h>
25 26 27
#include <asm/kvm.h>
#include <asm/kvm_emulate.h>
#include <asm/kvm_coproc.h>
28
#include <asm/sigcontext.h>
29

30 31
#include "trace.h"

32
struct kvm_stats_debugfs_item debugfs_entries[] = {
33 34 35 36 37 38 39 40 41 42
	VCPU_STAT("halt_successful_poll", halt_successful_poll),
	VCPU_STAT("halt_attempted_poll", halt_attempted_poll),
	VCPU_STAT("halt_poll_invalid", halt_poll_invalid),
	VCPU_STAT("halt_wakeup", halt_wakeup),
	VCPU_STAT("hvc_exit_stat", hvc_exit_stat),
	VCPU_STAT("wfe_exit_stat", wfe_exit_stat),
	VCPU_STAT("wfi_exit_stat", wfi_exit_stat),
	VCPU_STAT("mmio_exit_user", mmio_exit_user),
	VCPU_STAT("mmio_exit_kernel", mmio_exit_kernel),
	VCPU_STAT("exits", exits),
43 44
	VCPU_STAT("halt_poll_success_ns", halt_poll_success_ns),
	VCPU_STAT("halt_poll_fail_ns", halt_poll_fail_ns),
45 46 47 48 49 50 51 52 53 54 55 56 57
	{ "vcpu_stat", 0, KVM_STAT_DFX },
	{ NULL }
};

/* debugfs entries of Detail For vcpu stat EXtension */
struct dfx_kvm_stats_debugfs_item dfx_debugfs_entries[] = {
	DFX_STAT("pid", pid),
	DFX_STAT("hvc_exit_stat", hvc_exit_stat),
	DFX_STAT("wfe_exit_stat", wfe_exit_stat),
	DFX_STAT("wfi_exit_stat", wfi_exit_stat),
	DFX_STAT("mmio_exit_user", mmio_exit_user),
	DFX_STAT("mmio_exit_kernel", mmio_exit_kernel),
	DFX_STAT("exits", exits),
58 59 60 61 62 63 64 65 66 67 68 69 70 71 72
	DFX_STAT("fp_asimd_exit_stat", fp_asimd_exit_stat),
	DFX_STAT("irq_exit_stat", irq_exit_stat),
	DFX_STAT("sys64_exit_stat", sys64_exit_stat),
	DFX_STAT("mabt_exit_stat", mabt_exit_stat),
	DFX_STAT("fail_entry_exit_stat", fail_entry_exit_stat),
	DFX_STAT("internal_error_exit_stat", internal_error_exit_stat),
	DFX_STAT("unknown_ec_exit_stat", unknown_ec_exit_stat),
	DFX_STAT("cp15_32_exit_stat", cp15_32_exit_stat),
	DFX_STAT("cp15_64_exit_stat", cp15_64_exit_stat),
	DFX_STAT("cp14_mr_exit_stat", cp14_mr_exit_stat),
	DFX_STAT("cp14_ls_exit_stat", cp14_ls_exit_stat),
	DFX_STAT("cp14_64_exit_stat", cp14_64_exit_stat),
	DFX_STAT("smc_exit_stat", smc_exit_stat),
	DFX_STAT("sve_exit_stat", sve_exit_stat),
	DFX_STAT("debug_exit_stat", debug_exit_stat),
73 74 75
	{ NULL }
};

76 77 78 79 80 81
static bool core_reg_offset_is_vreg(u64 off)
{
	return off >= KVM_REG_ARM_CORE_REG(fp_regs.vregs) &&
		off < KVM_REG_ARM_CORE_REG(fp_regs.fpsr);
}

82 83 84 85 86
static u64 core_reg_offset_from_id(u64 id)
{
	return id & ~(KVM_REG_ARCH_MASK | KVM_REG_SIZE_MASK | KVM_REG_ARM_CORE);
}

87
static int core_reg_size_from_offset(const struct kvm_vcpu *vcpu, u64 off)
88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117
{
	int size;

	switch (off) {
	case KVM_REG_ARM_CORE_REG(regs.regs[0]) ...
	     KVM_REG_ARM_CORE_REG(regs.regs[30]):
	case KVM_REG_ARM_CORE_REG(regs.sp):
	case KVM_REG_ARM_CORE_REG(regs.pc):
	case KVM_REG_ARM_CORE_REG(regs.pstate):
	case KVM_REG_ARM_CORE_REG(sp_el1):
	case KVM_REG_ARM_CORE_REG(elr_el1):
	case KVM_REG_ARM_CORE_REG(spsr[0]) ...
	     KVM_REG_ARM_CORE_REG(spsr[KVM_NR_SPSR - 1]):
		size = sizeof(__u64);
		break;

	case KVM_REG_ARM_CORE_REG(fp_regs.vregs[0]) ...
	     KVM_REG_ARM_CORE_REG(fp_regs.vregs[31]):
		size = sizeof(__uint128_t);
		break;

	case KVM_REG_ARM_CORE_REG(fp_regs.fpsr):
	case KVM_REG_ARM_CORE_REG(fp_regs.fpcr):
		size = sizeof(__u32);
		break;

	default:
		return -EINVAL;
	}

118
	if (!IS_ALIGNED(off, size / sizeof(__u32)))
119
		return -EINVAL;
120

121 122 123 124 125 126 127 128
	/*
	 * The KVM_REG_ARM64_SVE regs must be used instead of
	 * KVM_REG_ARM_CORE for accessing the FPSIMD V-registers on
	 * SVE-enabled vcpus:
	 */
	if (vcpu_has_sve(vcpu) && core_reg_offset_is_vreg(off))
		return -EINVAL;

129 130 131
	return size;
}

132
static void *core_reg_addr(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
133 134 135 136 137
{
	u64 off = core_reg_offset_from_id(reg->id);
	int size = core_reg_size_from_offset(vcpu, off);

	if (size < 0)
138
		return NULL;
139 140

	if (KVM_REG_SIZE(reg->id) != size)
141
		return NULL;
142

143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159
	switch (off) {
	case KVM_REG_ARM_CORE_REG(regs.regs[0]) ...
	     KVM_REG_ARM_CORE_REG(regs.regs[30]):
		off -= KVM_REG_ARM_CORE_REG(regs.regs[0]);
		off /= 2;
		return &vcpu->arch.ctxt.regs.regs[off];

	case KVM_REG_ARM_CORE_REG(regs.sp):
		return &vcpu->arch.ctxt.regs.sp;

	case KVM_REG_ARM_CORE_REG(regs.pc):
		return &vcpu->arch.ctxt.regs.pc;

	case KVM_REG_ARM_CORE_REG(regs.pstate):
		return &vcpu->arch.ctxt.regs.pstate;

	case KVM_REG_ARM_CORE_REG(sp_el1):
160
		return __ctxt_sys_reg(&vcpu->arch.ctxt, SP_EL1);
161 162

	case KVM_REG_ARM_CORE_REG(elr_el1):
163
		return __ctxt_sys_reg(&vcpu->arch.ctxt, ELR_EL1);
164

165
	case KVM_REG_ARM_CORE_REG(spsr[KVM_SPSR_EL1]):
166
		return __ctxt_sys_reg(&vcpu->arch.ctxt, SPSR_EL1);
167 168 169 170 171 172 173 174 175 176 177 178

	case KVM_REG_ARM_CORE_REG(spsr[KVM_SPSR_ABT]):
		return &vcpu->arch.ctxt.spsr_abt;

	case KVM_REG_ARM_CORE_REG(spsr[KVM_SPSR_UND]):
		return &vcpu->arch.ctxt.spsr_und;

	case KVM_REG_ARM_CORE_REG(spsr[KVM_SPSR_IRQ]):
		return &vcpu->arch.ctxt.spsr_irq;

	case KVM_REG_ARM_CORE_REG(spsr[KVM_SPSR_FIQ]):
		return &vcpu->arch.ctxt.spsr_fiq;
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194

	case KVM_REG_ARM_CORE_REG(fp_regs.vregs[0]) ...
	     KVM_REG_ARM_CORE_REG(fp_regs.vregs[31]):
		off -= KVM_REG_ARM_CORE_REG(fp_regs.vregs[0]);
		off /= 4;
		return &vcpu->arch.ctxt.fp_regs.vregs[off];

	case KVM_REG_ARM_CORE_REG(fp_regs.fpsr):
		return &vcpu->arch.ctxt.fp_regs.fpsr;

	case KVM_REG_ARM_CORE_REG(fp_regs.fpcr):
		return &vcpu->arch.ctxt.fp_regs.fpcr;

	default:
		return NULL;
	}
195 196
}

197 198 199 200 201 202 203 204 205
static int get_core_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
{
	/*
	 * Because the kvm_regs structure is a mix of 32, 64 and
	 * 128bit fields, we index it as if it was a 32bit
	 * array. Hence below, nr_regs is the number of entries, and
	 * off the index in the "array".
	 */
	__u32 __user *uaddr = (__u32 __user *)(unsigned long)reg->addr;
206 207
	int nr_regs = sizeof(struct kvm_regs) / sizeof(__u32);
	void *addr;
208 209 210 211 212 213 214 215
	u32 off;

	/* Our ID is an index into the kvm_regs struct. */
	off = core_reg_offset_from_id(reg->id);
	if (off >= nr_regs ||
	    (off + (KVM_REG_SIZE(reg->id) / sizeof(__u32))) >= nr_regs)
		return -ENOENT;

216 217
	addr = core_reg_addr(vcpu, reg);
	if (!addr)
218 219
		return -EINVAL;

220
	if (copy_to_user(uaddr, addr, KVM_REG_SIZE(reg->id)))
221 222 223 224 225 226 227 228
		return -EFAULT;

	return 0;
}

static int set_core_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
{
	__u32 __user *uaddr = (__u32 __user *)(unsigned long)reg->addr;
229
	int nr_regs = sizeof(struct kvm_regs) / sizeof(__u32);
230
	__uint128_t tmp;
231
	void *valp = &tmp, *addr;
232 233 234 235 236 237 238 239 240
	u64 off;
	int err = 0;

	/* Our ID is an index into the kvm_regs struct. */
	off = core_reg_offset_from_id(reg->id);
	if (off >= nr_regs ||
	    (off + (KVM_REG_SIZE(reg->id) / sizeof(__u32))) >= nr_regs)
		return -ENOENT;

241 242
	addr = core_reg_addr(vcpu, reg);
	if (!addr)
243 244
		return -EINVAL;

245 246 247 248 249 250 251 252 253
	if (KVM_REG_SIZE(reg->id) > sizeof(tmp))
		return -EINVAL;

	if (copy_from_user(valp, uaddr, KVM_REG_SIZE(reg->id))) {
		err = -EFAULT;
		goto out;
	}

	if (off == KVM_REG_ARM_CORE_REG(regs.pstate)) {
254
		u64 mode = (*(u64 *)valp) & PSR_AA32_MODE_MASK;
255
		switch (mode) {
M
Mark Rutland 已提交
256
		case PSR_AA32_MODE_USR:
257 258 259
			if (!system_supports_32bit_el0())
				return -EINVAL;
			break;
M
Mark Rutland 已提交
260 261 262 263 264
		case PSR_AA32_MODE_FIQ:
		case PSR_AA32_MODE_IRQ:
		case PSR_AA32_MODE_SVC:
		case PSR_AA32_MODE_ABT:
		case PSR_AA32_MODE_UND:
265 266 267
			if (!vcpu_el1_is_32bit(vcpu))
				return -EINVAL;
			break;
268 269 270
		case PSR_MODE_EL0t:
		case PSR_MODE_EL1t:
		case PSR_MODE_EL1h:
271 272
			if (vcpu_el1_is_32bit(vcpu))
				return -EINVAL;
273 274 275 276 277 278 279
			break;
		default:
			err = -EINVAL;
			goto out;
		}
	}

280
	memcpy(addr, valp, KVM_REG_SIZE(reg->id));
281 282 283 284 285 286 287

	if (*vcpu_cpsr(vcpu) & PSR_MODE32_BIT) {
		int i;

		for (i = 0; i < 16; i++)
			*vcpu_reg32(vcpu, i) = (u32)*vcpu_reg32(vcpu, i);
	}
288 289 290 291
out:
	return err;
}

292 293
#define vq_word(vq) (((vq) - SVE_VQ_MIN) / 64)
#define vq_mask(vq) ((u64)1 << ((vq) - SVE_VQ_MIN) % 64)
294
#define vq_present(vqs, vq) (!!((vqs)[vq_word(vq)] & vq_mask(vq)))
295 296 297 298

static int get_sve_vls(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
{
	unsigned int max_vq, vq;
299
	u64 vqs[KVM_ARM64_SVE_VLS_WORDS];
300

301 302 303
	if (!vcpu_has_sve(vcpu))
		return -ENOENT;

304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
	if (WARN_ON(!sve_vl_valid(vcpu->arch.sve_max_vl)))
		return -EINVAL;

	memset(vqs, 0, sizeof(vqs));

	max_vq = sve_vq_from_vl(vcpu->arch.sve_max_vl);
	for (vq = SVE_VQ_MIN; vq <= max_vq; ++vq)
		if (sve_vq_available(vq))
			vqs[vq_word(vq)] |= vq_mask(vq);

	if (copy_to_user((void __user *)reg->addr, vqs, sizeof(vqs)))
		return -EFAULT;

	return 0;
}

static int set_sve_vls(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
{
	unsigned int max_vq, vq;
323
	u64 vqs[KVM_ARM64_SVE_VLS_WORDS];
324

325 326 327
	if (!vcpu_has_sve(vcpu))
		return -ENOENT;

328 329 330 331 332 333 334 335 336 337 338
	if (kvm_arm_vcpu_sve_finalized(vcpu))
		return -EPERM; /* too late! */

	if (WARN_ON(vcpu->arch.sve_state))
		return -EINVAL;

	if (copy_from_user(vqs, (const void __user *)reg->addr, sizeof(vqs)))
		return -EFAULT;

	max_vq = 0;
	for (vq = SVE_VQ_MIN; vq <= SVE_VQ_MAX; ++vq)
339
		if (vq_present(vqs, vq))
340 341 342 343 344
			max_vq = vq;

	if (max_vq > sve_vq_from_vl(kvm_sve_max_vl))
		return -EINVAL;

345 346 347
	/*
	 * Vector lengths supported by the host can't currently be
	 * hidden from the guest individually: instead we can only set a
F
Fuad Tabba 已提交
348
	 * maximum via ZCR_EL2.LEN.  So, make sure the available vector
349 350 351
	 * lengths match the set requested exactly up to the requested
	 * maximum:
	 */
352
	for (vq = SVE_VQ_MIN; vq <= max_vq; ++vq)
353
		if (vq_present(vqs, vq) != sve_vq_available(vq))
354 355 356 357 358 359 360 361 362 363 364 365
			return -EINVAL;

	/* Can't run with no vector lengths at all: */
	if (max_vq < SVE_VQ_MIN)
		return -EINVAL;

	/* vcpu->arch.sve_state will be alloc'd by kvm_vcpu_finalize_sve() */
	vcpu->arch.sve_max_vl = sve_vl_from_vq(max_vq);

	return 0;
}

366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381
#define SVE_REG_SLICE_SHIFT	0
#define SVE_REG_SLICE_BITS	5
#define SVE_REG_ID_SHIFT	(SVE_REG_SLICE_SHIFT + SVE_REG_SLICE_BITS)
#define SVE_REG_ID_BITS		5

#define SVE_REG_SLICE_MASK					\
	GENMASK(SVE_REG_SLICE_SHIFT + SVE_REG_SLICE_BITS - 1,	\
		SVE_REG_SLICE_SHIFT)
#define SVE_REG_ID_MASK							\
	GENMASK(SVE_REG_ID_SHIFT + SVE_REG_ID_BITS - 1, SVE_REG_ID_SHIFT)

#define SVE_NUM_SLICES (1 << SVE_REG_SLICE_BITS)

#define KVM_SVE_ZREG_SIZE KVM_REG_SIZE(KVM_REG_ARM64_SVE_ZREG(0, 0))
#define KVM_SVE_PREG_SIZE KVM_REG_SIZE(KVM_REG_ARM64_SVE_PREG(0, 0))

382
/*
383 384 385 386
 * Number of register slices required to cover each whole SVE register.
 * NOTE: Only the first slice every exists, for now.
 * If you are tempted to modify this, you must also rework sve_reg_to_region()
 * to match:
387 388 389
 */
#define vcpu_sve_slices(vcpu) 1

390 391 392 393 394 395 396
/* Bounds of a single SVE register slice within vcpu->arch.sve_state */
struct sve_state_reg_region {
	unsigned int koffset;	/* offset into sve_state in kernel memory */
	unsigned int klen;	/* length in kernel memory */
	unsigned int upad;	/* extra trailing padding in user memory */
};

397 398 399 400
/*
 * Validate SVE register ID and get sanitised bounds for user/kernel SVE
 * register copy
 */
401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417
static int sve_reg_to_region(struct sve_state_reg_region *region,
			     struct kvm_vcpu *vcpu,
			     const struct kvm_one_reg *reg)
{
	/* reg ID ranges for Z- registers */
	const u64 zreg_id_min = KVM_REG_ARM64_SVE_ZREG(0, 0);
	const u64 zreg_id_max = KVM_REG_ARM64_SVE_ZREG(SVE_NUM_ZREGS - 1,
						       SVE_NUM_SLICES - 1);

	/* reg ID ranges for P- registers and FFR (which are contiguous) */
	const u64 preg_id_min = KVM_REG_ARM64_SVE_PREG(0, 0);
	const u64 preg_id_max = KVM_REG_ARM64_SVE_FFR(SVE_NUM_SLICES - 1);

	unsigned int vq;
	unsigned int reg_num;

	unsigned int reqoffset, reqlen; /* User-requested offset and length */
F
Fuad Tabba 已提交
418
	unsigned int maxlen; /* Maximum permitted length */
419 420 421

	size_t sve_state_size;

422 423 424 425 426 427 428 429 430
	const u64 last_preg_id = KVM_REG_ARM64_SVE_PREG(SVE_NUM_PREGS - 1,
							SVE_NUM_SLICES - 1);

	/* Verify that the P-regs and FFR really do have contiguous IDs: */
	BUILD_BUG_ON(KVM_REG_ARM64_SVE_FFR(0) != last_preg_id + 1);

	/* Verify that we match the UAPI header: */
	BUILD_BUG_ON(SVE_NUM_SLICES != KVM_ARM64_SVE_MAX_SLICES);

431 432 433
	reg_num = (reg->id & SVE_REG_ID_MASK) >> SVE_REG_ID_SHIFT;

	if (reg->id >= zreg_id_min && reg->id <= zreg_id_max) {
434 435 436 437 438
		if (!vcpu_has_sve(vcpu) || (reg->id & SVE_REG_SLICE_MASK) > 0)
			return -ENOENT;

		vq = sve_vq_from_vl(vcpu->arch.sve_max_vl);

439 440 441 442 443
		reqoffset = SVE_SIG_ZREG_OFFSET(vq, reg_num) -
				SVE_SIG_REGS_OFFSET;
		reqlen = KVM_SVE_ZREG_SIZE;
		maxlen = SVE_SIG_ZREG_SIZE(vq);
	} else if (reg->id >= preg_id_min && reg->id <= preg_id_max) {
444 445 446 447 448
		if (!vcpu_has_sve(vcpu) || (reg->id & SVE_REG_SLICE_MASK) > 0)
			return -ENOENT;

		vq = sve_vq_from_vl(vcpu->arch.sve_max_vl);

449 450 451 452 453
		reqoffset = SVE_SIG_PREG_OFFSET(vq, reg_num) -
				SVE_SIG_REGS_OFFSET;
		reqlen = KVM_SVE_PREG_SIZE;
		maxlen = SVE_SIG_PREG_SIZE(vq);
	} else {
454
		return -EINVAL;
455 456 457
	}

	sve_state_size = vcpu_sve_state_size(vcpu);
458
	if (WARN_ON(!sve_state_size))
459 460 461 462 463 464 465 466 467 468 469
		return -EINVAL;

	region->koffset = array_index_nospec(reqoffset, sve_state_size);
	region->klen = min(maxlen, reqlen);
	region->upad = reqlen - region->klen;

	return 0;
}

static int get_sve_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
{
470
	int ret;
471 472 473
	struct sve_state_reg_region region;
	char __user *uptr = (char __user *)reg->addr;

474 475 476 477
	/* Handle the KVM_REG_ARM64_SVE_VLS pseudo-reg as a special case: */
	if (reg->id == KVM_REG_ARM64_SVE_VLS)
		return get_sve_vls(vcpu, reg);

478 479 480 481
	/* Try to interpret reg ID as an architectural SVE register... */
	ret = sve_reg_to_region(&region, vcpu, reg);
	if (ret)
		return ret;
482 483 484 485

	if (!kvm_arm_vcpu_sve_finalized(vcpu))
		return -EPERM;

486 487 488 489 490 491 492 493 494 495
	if (copy_to_user(uptr, vcpu->arch.sve_state + region.koffset,
			 region.klen) ||
	    clear_user(uptr + region.klen, region.upad))
		return -EFAULT;

	return 0;
}

static int set_sve_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
{
496
	int ret;
497 498 499
	struct sve_state_reg_region region;
	const char __user *uptr = (const char __user *)reg->addr;

500 501 502 503
	/* Handle the KVM_REG_ARM64_SVE_VLS pseudo-reg as a special case: */
	if (reg->id == KVM_REG_ARM64_SVE_VLS)
		return set_sve_vls(vcpu, reg);

504 505 506 507
	/* Try to interpret reg ID as an architectural SVE register... */
	ret = sve_reg_to_region(&region, vcpu, reg);
	if (ret)
		return ret;
508 509 510 511

	if (!kvm_arm_vcpu_sve_finalized(vcpu))
		return -EPERM;

512 513 514 515 516 517 518
	if (copy_from_user(vcpu->arch.sve_state + region.koffset, uptr,
			   region.klen))
		return -EFAULT;

	return 0;
}

519 520 521 522 523 524 525 526 527 528
int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
	return -EINVAL;
}

int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
	return -EINVAL;
}

529 530
static int copy_core_reg_indices(const struct kvm_vcpu *vcpu,
				 u64 __user *uindices)
531 532 533 534 535
{
	unsigned int i;
	int n = 0;

	for (i = 0; i < sizeof(struct kvm_regs) / sizeof(__u32); i++) {
536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556
		u64 reg = KVM_REG_ARM64 | KVM_REG_ARM_CORE | i;
		int size = core_reg_size_from_offset(vcpu, i);

		if (size < 0)
			continue;

		switch (size) {
		case sizeof(__u32):
			reg |= KVM_REG_SIZE_U32;
			break;

		case sizeof(__u64):
			reg |= KVM_REG_SIZE_U64;
			break;

		case sizeof(__uint128_t):
			reg |= KVM_REG_SIZE_U128;
			break;

		default:
			WARN_ON(1);
557
			continue;
558
		}
559

560
		if (uindices) {
561
			if (put_user(reg, uindices))
562 563 564 565 566 567 568 569 570 571
				return -EFAULT;
			uindices++;
		}

		n++;
	}

	return n;
}

572
static unsigned long num_core_regs(const struct kvm_vcpu *vcpu)
573
{
574
	return copy_core_reg_indices(vcpu, NULL);
575 576
}

577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615
/**
 * ARM64 versions of the TIMER registers, always available on arm64
 */

#define NUM_TIMER_REGS 3

static bool is_timer_reg(u64 index)
{
	switch (index) {
	case KVM_REG_ARM_TIMER_CTL:
	case KVM_REG_ARM_TIMER_CNT:
	case KVM_REG_ARM_TIMER_CVAL:
		return true;
	}
	return false;
}

static int copy_timer_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
{
	if (put_user(KVM_REG_ARM_TIMER_CTL, uindices))
		return -EFAULT;
	uindices++;
	if (put_user(KVM_REG_ARM_TIMER_CNT, uindices))
		return -EFAULT;
	uindices++;
	if (put_user(KVM_REG_ARM_TIMER_CVAL, uindices))
		return -EFAULT;

	return 0;
}

static int set_timer_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
{
	void __user *uaddr = (void __user *)(long)reg->addr;
	u64 val;
	int ret;

	ret = copy_from_user(&val, uaddr, KVM_REG_SIZE(reg->id));
	if (ret != 0)
616
		return -EFAULT;
617 618 619 620 621 622 623 624 625 626

	return kvm_arm_timer_set_reg(vcpu, reg->id, val);
}

static int get_timer_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
{
	void __user *uaddr = (void __user *)(long)reg->addr;
	u64 val;

	val = kvm_arm_timer_get_reg(vcpu, reg->id);
627
	return copy_to_user(uaddr, &val, KVM_REG_SIZE(reg->id)) ? -EFAULT : 0;
628 629
}

630 631 632 633 634 635 636
static unsigned long num_sve_regs(const struct kvm_vcpu *vcpu)
{
	const unsigned int slices = vcpu_sve_slices(vcpu);

	if (!vcpu_has_sve(vcpu))
		return 0;

637 638 639 640 641
	/* Policed by KVM_GET_REG_LIST: */
	WARN_ON(!kvm_arm_vcpu_sve_finalized(vcpu));

	return slices * (SVE_NUM_PREGS + SVE_NUM_ZREGS + 1 /* FFR */)
		+ 1; /* KVM_REG_ARM64_SVE_VLS */
642 643 644 645 646 647 648 649 650 651 652 653 654
}

static int copy_sve_reg_indices(const struct kvm_vcpu *vcpu,
				u64 __user *uindices)
{
	const unsigned int slices = vcpu_sve_slices(vcpu);
	u64 reg;
	unsigned int i, n;
	int num_regs = 0;

	if (!vcpu_has_sve(vcpu))
		return 0;

655 656 657 658 659 660 661 662 663 664 665 666
	/* Policed by KVM_GET_REG_LIST: */
	WARN_ON(!kvm_arm_vcpu_sve_finalized(vcpu));

	/*
	 * Enumerate this first, so that userspace can save/restore in
	 * the order reported by KVM_GET_REG_LIST:
	 */
	reg = KVM_REG_ARM64_SVE_VLS;
	if (put_user(reg, uindices++))
		return -EFAULT;
	++num_regs;

667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690
	for (i = 0; i < slices; i++) {
		for (n = 0; n < SVE_NUM_ZREGS; n++) {
			reg = KVM_REG_ARM64_SVE_ZREG(n, i);
			if (put_user(reg, uindices++))
				return -EFAULT;
			num_regs++;
		}

		for (n = 0; n < SVE_NUM_PREGS; n++) {
			reg = KVM_REG_ARM64_SVE_PREG(n, i);
			if (put_user(reg, uindices++))
				return -EFAULT;
			num_regs++;
		}

		reg = KVM_REG_ARM64_SVE_FFR(i);
		if (put_user(reg, uindices++))
			return -EFAULT;
		num_regs++;
	}

	return num_regs;
}

691 692 693 694 695 696 697
/**
 * kvm_arm_num_regs - how many registers do we present via KVM_GET_ONE_REG
 *
 * This is for all registers.
 */
unsigned long kvm_arm_num_regs(struct kvm_vcpu *vcpu)
{
698 699
	unsigned long res = 0;

700
	res += num_core_regs(vcpu);
701
	res += num_sve_regs(vcpu);
702 703 704 705 706
	res += kvm_arm_num_sys_reg_descs(vcpu);
	res += kvm_arm_get_fw_num_regs(vcpu);
	res += NUM_TIMER_REGS;

	return res;
707 708 709 710 711
}

/**
 * kvm_arm_copy_reg_indices - get indices of all registers.
 *
A
Andrea Gelmini 已提交
712
 * We do core registers right here, then we append system regs.
713 714 715
 */
int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
{
716
	int ret;
717

718
	ret = copy_core_reg_indices(vcpu, uindices);
719
	if (ret < 0)
720 721
		return ret;
	uindices += ret;
722

723
	ret = copy_sve_reg_indices(vcpu, uindices);
724
	if (ret < 0)
725 726 727
		return ret;
	uindices += ret;

728
	ret = kvm_arm_copy_fw_reg_indices(vcpu, uindices);
729
	if (ret < 0)
730 731 732
		return ret;
	uindices += kvm_arm_get_fw_num_regs(vcpu);

733
	ret = copy_timer_indices(vcpu, uindices);
734
	if (ret < 0)
735 736 737
		return ret;
	uindices += NUM_TIMER_REGS;

738 739 740 741 742 743 744 745 746
	return kvm_arm_copy_sys_reg_indices(vcpu, uindices);
}

int kvm_arm_get_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
{
	/* We currently use nothing arch-specific in upper 32 bits */
	if ((reg->id & ~KVM_REG_SIZE_MASK) >> 32 != KVM_REG_ARM64 >> 32)
		return -EINVAL;

747 748 749 750 751
	switch (reg->id & KVM_REG_ARM_COPROC_MASK) {
	case KVM_REG_ARM_CORE:	return get_core_reg(vcpu, reg);
	case KVM_REG_ARM_FW:	return kvm_arm_get_fw_reg(vcpu, reg);
	case KVM_REG_ARM64_SVE:	return get_sve_reg(vcpu, reg);
	}
752

753 754 755
	if (is_timer_reg(reg->id))
		return get_timer_reg(vcpu, reg);

756 757 758 759 760 761 762 763 764
	return kvm_arm_sys_reg_get_reg(vcpu, reg);
}

int kvm_arm_set_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
{
	/* We currently use nothing arch-specific in upper 32 bits */
	if ((reg->id & ~KVM_REG_SIZE_MASK) >> 32 != KVM_REG_ARM64 >> 32)
		return -EINVAL;

765 766 767 768 769
	switch (reg->id & KVM_REG_ARM_COPROC_MASK) {
	case KVM_REG_ARM_CORE:	return set_core_reg(vcpu, reg);
	case KVM_REG_ARM_FW:	return kvm_arm_set_fw_reg(vcpu, reg);
	case KVM_REG_ARM64_SVE:	return set_sve_reg(vcpu, reg);
	}
770

771 772 773
	if (is_timer_reg(reg->id))
		return set_timer_reg(vcpu, reg);

774 775 776 777 778 779 780 781 782 783 784 785 786 787 788
	return kvm_arm_sys_reg_set_reg(vcpu, reg);
}

int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
	return -EINVAL;
}

int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
	return -EINVAL;
}

789 790
int __kvm_arm_vcpu_get_events(struct kvm_vcpu *vcpu,
			      struct kvm_vcpu_events *events)
791 792 793 794 795 796 797
{
	events->exception.serror_pending = !!(vcpu->arch.hcr_el2 & HCR_VSE);
	events->exception.serror_has_esr = cpus_have_const_cap(ARM64_HAS_RAS_EXTN);

	if (events->exception.serror_pending && events->exception.serror_has_esr)
		events->exception.serror_esr = vcpu_get_vsesr(vcpu);

798 799 800 801 802 803
	/*
	 * We never return a pending ext_dabt here because we deliver it to
	 * the virtual CPU directly when setting the event and it's no longer
	 * 'pending' at this point.
	 */

804 805 806
	return 0;
}

807 808
int __kvm_arm_vcpu_set_events(struct kvm_vcpu *vcpu,
			      struct kvm_vcpu_events *events)
809 810 811
{
	bool serror_pending = events->exception.serror_pending;
	bool has_esr = events->exception.serror_has_esr;
812
	bool ext_dabt_pending = events->exception.ext_dabt_pending;
813 814 815 816 817 818 819 820 821 822 823 824 825

	if (serror_pending && has_esr) {
		if (!cpus_have_const_cap(ARM64_HAS_RAS_EXTN))
			return -EINVAL;

		if (!((events->exception.serror_esr) & ~ESR_ELx_ISS_MASK))
			kvm_set_sei_esr(vcpu, events->exception.serror_esr);
		else
			return -EINVAL;
	} else if (serror_pending) {
		kvm_inject_vabt(vcpu);
	}

826 827 828
	if (ext_dabt_pending)
		kvm_inject_dabt(vcpu, kvm_vcpu_get_hfar(vcpu));

829 830 831
	return 0;
}

832 833 834 835 836
int __attribute_const__ kvm_target_cpu(void)
{
	unsigned long implementor = read_cpuid_implementor();
	unsigned long part_number = read_cpuid_part_number();

837 838 839 840 841 842 843
	switch (implementor) {
	case ARM_CPU_IMP_ARM:
		switch (part_number) {
		case ARM_CPU_PART_AEM_V8:
			return KVM_ARM_TARGET_AEM_V8;
		case ARM_CPU_PART_FOUNDATION:
			return KVM_ARM_TARGET_FOUNDATION_V8;
844 845
		case ARM_CPU_PART_CORTEX_A53:
			return KVM_ARM_TARGET_CORTEX_A53;
846 847
		case ARM_CPU_PART_CORTEX_A57:
			return KVM_ARM_TARGET_CORTEX_A57;
848
		}
849 850 851 852 853
		break;
	case ARM_CPU_IMP_APM:
		switch (part_number) {
		case APM_CPU_PART_POTENZA:
			return KVM_ARM_TARGET_XGENE_POTENZA;
854
		}
855
		break;
856
	}
857

858 859
	/* Return a default generic target */
	return KVM_ARM_TARGET_GENERIC_V8;
860 861
}

862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881
int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init)
{
	int target = kvm_target_cpu();

	if (target < 0)
		return -ENODEV;

	memset(init, 0, sizeof(*init));

	/*
	 * For now, we don't return any features.
	 * In future, we might use features to return target
	 * specific features available for the preferred
	 * target type.
	 */
	init->target = (__u32)target;

	return 0;
}

882 883 884 885 886 887 888 889 890 891 892 893 894 895 896
int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
	return -EINVAL;
}

int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
	return -EINVAL;
}

int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
				  struct kvm_translation *tr)
{
	return -EINVAL;
}
897

898 899
#define KVM_GUESTDBG_VALID_MASK (KVM_GUESTDBG_ENABLE |    \
			    KVM_GUESTDBG_USE_SW_BP | \
900
			    KVM_GUESTDBG_USE_HW | \
901
			    KVM_GUESTDBG_SINGLESTEP)
902 903 904 905 906 907 908 909 910 911 912 913 914 915

/**
 * kvm_arch_vcpu_ioctl_set_guest_debug - set up guest debugging
 * @kvm:	pointer to the KVM struct
 * @kvm_guest_debug: the ioctl data buffer
 *
 * This sets up and enables the VM for guest debugging. Userspace
 * passes in a control flag to enable different debug types and
 * potentially other architecture specific information in the rest of
 * the structure.
 */
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
					struct kvm_guest_debug *dbg)
{
916 917
	int ret = 0;

918 919
	trace_kvm_set_guest_debug(vcpu, dbg->control);

920 921 922 923
	if (dbg->control & ~KVM_GUESTDBG_VALID_MASK) {
		ret = -EINVAL;
		goto out;
	}
924 925 926

	if (dbg->control & KVM_GUESTDBG_ENABLE) {
		vcpu->guest_debug = dbg->control;
927 928 929 930 931 932

		/* Hardware assisted Break and Watch points */
		if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW) {
			vcpu->arch.external_debug_state = dbg->arch;
		}

933 934 935 936
	} else {
		/* If not enabled clear all flags */
		vcpu->guest_debug = 0;
	}
937 938 939

out:
	return ret;
940
}
941 942 943 944 945 946 947 948 949 950

int kvm_arm_vcpu_arch_set_attr(struct kvm_vcpu *vcpu,
			       struct kvm_device_attr *attr)
{
	int ret;

	switch (attr->group) {
	case KVM_ARM_VCPU_PMU_V3_CTRL:
		ret = kvm_arm_pmu_v3_set_attr(vcpu, attr);
		break;
951 952 953
	case KVM_ARM_VCPU_TIMER_CTRL:
		ret = kvm_arm_timer_set_attr(vcpu, attr);
		break;
954 955 956
	case KVM_ARM_VCPU_PVTIME_CTRL:
		ret = kvm_arm_pvtime_set_attr(vcpu, attr);
		break;
957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

int kvm_arm_vcpu_arch_get_attr(struct kvm_vcpu *vcpu,
			       struct kvm_device_attr *attr)
{
	int ret;

	switch (attr->group) {
	case KVM_ARM_VCPU_PMU_V3_CTRL:
		ret = kvm_arm_pmu_v3_get_attr(vcpu, attr);
		break;
974 975 976
	case KVM_ARM_VCPU_TIMER_CTRL:
		ret = kvm_arm_timer_get_attr(vcpu, attr);
		break;
977 978 979
	case KVM_ARM_VCPU_PVTIME_CTRL:
		ret = kvm_arm_pvtime_get_attr(vcpu, attr);
		break;
980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

int kvm_arm_vcpu_arch_has_attr(struct kvm_vcpu *vcpu,
			       struct kvm_device_attr *attr)
{
	int ret;

	switch (attr->group) {
	case KVM_ARM_VCPU_PMU_V3_CTRL:
		ret = kvm_arm_pmu_v3_has_attr(vcpu, attr);
		break;
997 998 999
	case KVM_ARM_VCPU_TIMER_CTRL:
		ret = kvm_arm_timer_has_attr(vcpu, attr);
		break;
1000 1001 1002
	case KVM_ARM_VCPU_PVTIME_CTRL:
		ret = kvm_arm_pvtime_has_attr(vcpu, attr);
		break;
1003 1004 1005 1006 1007 1008 1009
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}