powerpc.c 54.6 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
/*
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License, version 2, as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 *
 * Copyright IBM Corp. 2007
 *
 * Authors: Hollis Blanchard <hollisb@us.ibm.com>
 *          Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
 */

#include <linux/errno.h>
#include <linux/err.h>
#include <linux/kvm_host.h>
#include <linux/vmalloc.h>
A
Alexander Graf 已提交
25
#include <linux/hrtimer.h>
26
#include <linux/sched/signal.h>
27
#include <linux/fs.h>
28
#include <linux/slab.h>
S
Scott Wood 已提交
29
#include <linux/file.h>
30
#include <linux/module.h>
31 32
#include <linux/irqbypass.h>
#include <linux/kvm_irqfd.h>
33
#include <asm/cputable.h>
34
#include <linux/uaccess.h>
35
#include <asm/kvm_ppc.h>
36
#include <asm/tlbflush.h>
37
#include <asm/cputhreads.h>
38
#include <asm/irqflags.h>
39
#include <asm/iommu.h>
40
#include <asm/switch_to.h>
41
#include <asm/xive.h>
42 43 44 45
#ifdef CONFIG_PPC_PSERIES
#include <asm/hvcall.h>
#include <asm/plpar_wrappers.h>
#endif
46

47
#include "timing.h"
48
#include "irq.h"
P
Paul Mackerras 已提交
49
#include "../mm/mmu_decl.h"
50

51 52 53
#define CREATE_TRACE_POINTS
#include "trace.h"

54 55 56 57 58
struct kvmppc_ops *kvmppc_hv_ops;
EXPORT_SYMBOL_GPL(kvmppc_hv_ops);
struct kvmppc_ops *kvmppc_pr_ops;
EXPORT_SYMBOL_GPL(kvmppc_pr_ops);

59

60 61
int kvm_arch_vcpu_runnable(struct kvm_vcpu *v)
{
R
Radim Krčmář 已提交
62
	return !!(v->arch.pending_exceptions) || kvm_request_pending(v);
63 64
}

65 66 67 68 69
bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
{
	return false;
}

70 71 72 73 74
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
	return 1;
}

75 76 77 78
/*
 * Common checks before entering the guest world.  Call with interrupts
 * disabled.
 *
79 80 81 82
 * returns:
 *
 * == 1 if we're ready to go into guest state
 * <= 0 if we need to go back to the host with return value
83 84 85
 */
int kvmppc_prepare_to_enter(struct kvm_vcpu *vcpu)
{
S
Scott Wood 已提交
86 87 88 89
	int r;

	WARN_ON(irqs_disabled());
	hard_irq_disable();
90 91 92 93 94

	while (true) {
		if (need_resched()) {
			local_irq_enable();
			cond_resched();
S
Scott Wood 已提交
95
			hard_irq_disable();
96 97 98 99
			continue;
		}

		if (signal_pending(current)) {
100 101 102
			kvmppc_account_exit(vcpu, SIGNAL_EXITS);
			vcpu->run->exit_reason = KVM_EXIT_INTR;
			r = -EINTR;
103 104 105
			break;
		}

106 107 108 109 110 111 112
		vcpu->mode = IN_GUEST_MODE;

		/*
		 * Reading vcpu->requests must happen after setting vcpu->mode,
		 * so we don't miss a request because the requester sees
		 * OUTSIDE_GUEST_MODE and assumes we'll be checking requests
		 * before next entering the guest (and thus doesn't IPI).
113 114 115
		 * This also orders the write to mode from any reads
		 * to the page tables done while the VCPU is running.
		 * Please see the comment in kvm_flush_remote_tlbs.
116
		 */
117
		smp_mb();
118

R
Radim Krčmář 已提交
119
		if (kvm_request_pending(vcpu)) {
120 121 122
			/* Make sure we process requests preemptable */
			local_irq_enable();
			trace_kvm_check_requests(vcpu);
123
			r = kvmppc_core_check_requests(vcpu);
S
Scott Wood 已提交
124
			hard_irq_disable();
125 126 127
			if (r > 0)
				continue;
			break;
128 129 130 131 132 133 134 135
		}

		if (kvmppc_core_prepare_to_enter(vcpu)) {
			/* interrupts got enabled in between, so we
			   are back at square 1 */
			continue;
		}

136
		guest_enter_irqoff();
S
Scott Wood 已提交
137
		return 1;
138 139
	}

S
Scott Wood 已提交
140 141
	/* return to host */
	local_irq_enable();
142 143
	return r;
}
144
EXPORT_SYMBOL_GPL(kvmppc_prepare_to_enter);
145

146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166
#if defined(CONFIG_PPC_BOOK3S_64) && defined(CONFIG_KVM_BOOK3S_PR_POSSIBLE)
static void kvmppc_swab_shared(struct kvm_vcpu *vcpu)
{
	struct kvm_vcpu_arch_shared *shared = vcpu->arch.shared;
	int i;

	shared->sprg0 = swab64(shared->sprg0);
	shared->sprg1 = swab64(shared->sprg1);
	shared->sprg2 = swab64(shared->sprg2);
	shared->sprg3 = swab64(shared->sprg3);
	shared->srr0 = swab64(shared->srr0);
	shared->srr1 = swab64(shared->srr1);
	shared->dar = swab64(shared->dar);
	shared->msr = swab64(shared->msr);
	shared->dsisr = swab32(shared->dsisr);
	shared->int_pending = swab32(shared->int_pending);
	for (i = 0; i < ARRAY_SIZE(shared->sr); i++)
		shared->sr[i] = swab32(shared->sr[i]);
}
#endif

167 168 169 170 171 172 173 174 175 176
int kvmppc_kvm_pv(struct kvm_vcpu *vcpu)
{
	int nr = kvmppc_get_gpr(vcpu, 11);
	int r;
	unsigned long __maybe_unused param1 = kvmppc_get_gpr(vcpu, 3);
	unsigned long __maybe_unused param2 = kvmppc_get_gpr(vcpu, 4);
	unsigned long __maybe_unused param3 = kvmppc_get_gpr(vcpu, 5);
	unsigned long __maybe_unused param4 = kvmppc_get_gpr(vcpu, 6);
	unsigned long r2 = 0;

177
	if (!(kvmppc_get_msr(vcpu) & MSR_SF)) {
178 179 180 181 182 183 184 185
		/* 32 bit mode */
		param1 &= 0xffffffff;
		param2 &= 0xffffffff;
		param3 &= 0xffffffff;
		param4 &= 0xffffffff;
	}

	switch (nr) {
186
	case KVM_HCALL_TOKEN(KVM_HC_PPC_MAP_MAGIC_PAGE):
187
	{
188 189 190 191 192 193 194 195 196 197
#if defined(CONFIG_PPC_BOOK3S_64) && defined(CONFIG_KVM_BOOK3S_PR_POSSIBLE)
		/* Book3S can be little endian, find it out here */
		int shared_big_endian = true;
		if (vcpu->arch.intr_msr & MSR_LE)
			shared_big_endian = false;
		if (shared_big_endian != vcpu->arch.shared_big_endian)
			kvmppc_swab_shared(vcpu);
		vcpu->arch.shared_big_endian = shared_big_endian;
#endif

198 199 200 201 202 203 204 205 206 207 208 209
		if (!(param2 & MAGIC_PAGE_FLAG_NOT_MAPPED_NX)) {
			/*
			 * Older versions of the Linux magic page code had
			 * a bug where they would map their trampoline code
			 * NX. If that's the case, remove !PR NX capability.
			 */
			vcpu->arch.disable_kernel_nx = true;
			kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
		}

		vcpu->arch.magic_page_pa = param1 & ~0xfffULL;
		vcpu->arch.magic_page_ea = param2 & ~0xfffULL;
210

211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229
#ifdef CONFIG_PPC_64K_PAGES
		/*
		 * Make sure our 4k magic page is in the same window of a 64k
		 * page within the guest and within the host's page.
		 */
		if ((vcpu->arch.magic_page_pa & 0xf000) !=
		    ((ulong)vcpu->arch.shared & 0xf000)) {
			void *old_shared = vcpu->arch.shared;
			ulong shared = (ulong)vcpu->arch.shared;
			void *new_shared;

			shared &= PAGE_MASK;
			shared |= vcpu->arch.magic_page_pa & 0xf000;
			new_shared = (void*)shared;
			memcpy(new_shared, old_shared, 0x1000);
			vcpu->arch.shared = new_shared;
		}
#endif

230
		r2 = KVM_MAGIC_FEAT_SR | KVM_MAGIC_FEAT_MAS0_TO_SPRG7;
231

232
		r = EV_SUCCESS;
233 234
		break;
	}
235 236
	case KVM_HCALL_TOKEN(KVM_HC_FEATURES):
		r = EV_SUCCESS;
237
#if defined(CONFIG_PPC_BOOK3S) || defined(CONFIG_KVM_E500V2)
238 239
		r2 |= (1 << KVM_FEATURE_MAGIC_PAGE);
#endif
240 241 242

		/* Second return value is in r4 */
		break;
243 244 245
	case EV_HCALL_TOKEN(EV_IDLE):
		r = EV_SUCCESS;
		kvm_vcpu_block(vcpu);
246
		kvm_clear_request(KVM_REQ_UNHALT, vcpu);
247
		break;
248
	default:
249
		r = EV_UNIMPLEMENTED;
250 251 252
		break;
	}

253 254
	kvmppc_set_gpr(vcpu, 4, r2);

255 256
	return r;
}
257
EXPORT_SYMBOL_GPL(kvmppc_kvm_pv);
258

259 260 261 262 263 264 265 266 267 268 269 270 271
int kvmppc_sanity_check(struct kvm_vcpu *vcpu)
{
	int r = false;

	/* We have to know what CPU to virtualize */
	if (!vcpu->arch.pvr)
		goto out;

	/* PAPR only works with book3s_64 */
	if ((vcpu->arch.cpu_type != KVM_CPU_3S_64) && vcpu->arch.papr_enabled)
		goto out;

	/* HV KVM can only do PAPR mode for now */
272
	if (!vcpu->arch.papr_enabled && is_kvmppc_hv_enabled(vcpu->kvm))
273 274
		goto out;

275 276 277 278 279
#ifdef CONFIG_KVM_BOOKE_HV
	if (!cpu_has_feature(CPU_FTR_EMB_HV))
		goto out;
#endif

280 281 282 283 284 285
	r = true;

out:
	vcpu->arch.sane = r;
	return r ? 0 : -EINVAL;
}
286
EXPORT_SYMBOL_GPL(kvmppc_sanity_check);
287

288 289 290 291 292
int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu)
{
	enum emulation_result er;
	int r;

293
	er = kvmppc_emulate_loadstore(vcpu);
294 295 296 297 298 299
	switch (er) {
	case EMULATE_DONE:
		/* Future optimization: only reload non-volatiles if they were
		 * actually modified. */
		r = RESUME_GUEST_NV;
		break;
300 301 302
	case EMULATE_AGAIN:
		r = RESUME_GUEST;
		break;
303 304 305 306 307 308 309 310 311
	case EMULATE_DO_MMIO:
		run->exit_reason = KVM_EXIT_MMIO;
		/* We must reload nonvolatiles because "update" load/store
		 * instructions modify register state. */
		/* Future optimization: only reload non-volatiles if they were
		 * actually modified. */
		r = RESUME_HOST_NV;
		break;
	case EMULATE_FAIL:
312 313 314
	{
		u32 last_inst;

315
		kvmppc_get_last_inst(vcpu, INST_GENERIC, &last_inst);
316
		/* XXX Deliver Program interrupt to guest. */
317
		pr_emerg("%s: emulation failed (%08x)\n", __func__, last_inst);
318 319
		r = RESUME_HOST;
		break;
320
	}
321
	default:
322 323
		WARN_ON(1);
		r = RESUME_GUEST;
324 325 326 327
	}

	return r;
}
328
EXPORT_SYMBOL_GPL(kvmppc_emulate_mmio);
329

330 331 332
int kvmppc_st(struct kvm_vcpu *vcpu, ulong *eaddr, int size, void *ptr,
	      bool data)
{
333
	ulong mp_pa = vcpu->arch.magic_page_pa & KVM_PAM & PAGE_MASK;
334 335 336 337 338 339 340 341 342 343 344 345 346 347 348
	struct kvmppc_pte pte;
	int r;

	vcpu->stat.st++;

	r = kvmppc_xlate(vcpu, *eaddr, data ? XLATE_DATA : XLATE_INST,
			 XLATE_WRITE, &pte);
	if (r < 0)
		return r;

	*eaddr = pte.raddr;

	if (!pte.may_write)
		return -EPERM;

349 350 351 352 353 354 355 356 357 358
	/* Magic page override */
	if (kvmppc_supports_magic_page(vcpu) && mp_pa &&
	    ((pte.raddr & KVM_PAM & PAGE_MASK) == mp_pa) &&
	    !(kvmppc_get_msr(vcpu) & MSR_PR)) {
		void *magic = vcpu->arch.shared;
		magic += pte.eaddr & 0xfff;
		memcpy(magic, ptr, size);
		return EMULATE_DONE;
	}

359 360 361 362 363 364 365 366 367 368
	if (kvm_write_guest(vcpu->kvm, pte.raddr, ptr, size))
		return EMULATE_DO_MMIO;

	return EMULATE_DONE;
}
EXPORT_SYMBOL_GPL(kvmppc_st);

int kvmppc_ld(struct kvm_vcpu *vcpu, ulong *eaddr, int size, void *ptr,
		      bool data)
{
369
	ulong mp_pa = vcpu->arch.magic_page_pa & KVM_PAM & PAGE_MASK;
370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387
	struct kvmppc_pte pte;
	int rc;

	vcpu->stat.ld++;

	rc = kvmppc_xlate(vcpu, *eaddr, data ? XLATE_DATA : XLATE_INST,
			  XLATE_READ, &pte);
	if (rc)
		return rc;

	*eaddr = pte.raddr;

	if (!pte.may_read)
		return -EPERM;

	if (!data && !pte.may_execute)
		return -ENOEXEC;

388 389 390 391 392 393 394 395 396 397
	/* Magic page override */
	if (kvmppc_supports_magic_page(vcpu) && mp_pa &&
	    ((pte.raddr & KVM_PAM & PAGE_MASK) == mp_pa) &&
	    !(kvmppc_get_msr(vcpu) & MSR_PR)) {
		void *magic = vcpu->arch.shared;
		magic += pte.eaddr & 0xfff;
		memcpy(ptr, magic, size);
		return EMULATE_DONE;
	}

398 399
	if (kvm_read_guest(vcpu->kvm, pte.raddr, ptr, size))
		return EMULATE_DO_MMIO;
400 401 402 403 404

	return EMULATE_DONE;
}
EXPORT_SYMBOL_GPL(kvmppc_ld);

405
int kvm_arch_hardware_enable(void)
406
{
407
	return 0;
408 409 410 411 412 413 414 415 416
}

int kvm_arch_hardware_setup(void)
{
	return 0;
}

void kvm_arch_check_processor_compat(void *rtn)
{
417
	*(int *)rtn = kvmppc_core_check_processor_compat();
418 419
}

420
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
421
{
422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447
	struct kvmppc_ops *kvm_ops = NULL;
	/*
	 * if we have both HV and PR enabled, default is HV
	 */
	if (type == 0) {
		if (kvmppc_hv_ops)
			kvm_ops = kvmppc_hv_ops;
		else
			kvm_ops = kvmppc_pr_ops;
		if (!kvm_ops)
			goto err_out;
	} else	if (type == KVM_VM_PPC_HV) {
		if (!kvmppc_hv_ops)
			goto err_out;
		kvm_ops = kvmppc_hv_ops;
	} else if (type == KVM_VM_PPC_PR) {
		if (!kvmppc_pr_ops)
			goto err_out;
		kvm_ops = kvmppc_pr_ops;
	} else
		goto err_out;

	if (kvm_ops->owner && !try_module_get(kvm_ops->owner))
		return -ENOENT;

	kvm->arch.kvm_ops = kvm_ops;
448
	return kvmppc_core_init_vm(kvm);
449 450
err_out:
	return -EINVAL;
451 452
}

453 454 455 456 457 458 459 460 461 462
bool kvm_arch_has_vcpu_debugfs(void)
{
	return false;
}

int kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
{
	return 0;
}

463
void kvm_arch_destroy_vm(struct kvm *kvm)
464 465
{
	unsigned int i;
466
	struct kvm_vcpu *vcpu;
467

468 469 470 471 472 473 474 475 476 477
#ifdef CONFIG_KVM_XICS
	/*
	 * We call kick_all_cpus_sync() to ensure that all
	 * CPUs have executed any pending IPIs before we
	 * continue and free VCPUs structures below.
	 */
	if (is_kvmppc_hv_enabled(kvm))
		kick_all_cpus_sync();
#endif

478 479 480 481 482 483 484 485
	kvm_for_each_vcpu(i, vcpu, kvm)
		kvm_arch_vcpu_free(vcpu);

	mutex_lock(&kvm->lock);
	for (i = 0; i < atomic_read(&kvm->online_vcpus); i++)
		kvm->vcpus[i] = NULL;

	atomic_set(&kvm->online_vcpus, 0);
486 487 488

	kvmppc_core_destroy_vm(kvm);

489
	mutex_unlock(&kvm->lock);
490 491 492

	/* drop the module reference */
	module_put(kvm->arch.kvm_ops->owner);
493 494
}

495
int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
496 497
{
	int r;
498
	/* Assume we're using HV mode when the HV module is loaded */
499
	int hv_enabled = kvmppc_hv_ops ? 1 : 0;
500

501 502 503 504 505 506 507 508
	if (kvm) {
		/*
		 * Hooray - we know which VM type we're running on. Depend on
		 * that rather than the guess above.
		 */
		hv_enabled = is_kvmppc_hv_enabled(kvm);
	}

509
	switch (ext) {
S
Scott Wood 已提交
510 511
#ifdef CONFIG_BOOKE
	case KVM_CAP_PPC_BOOKE_SREGS:
512
	case KVM_CAP_PPC_BOOKE_WATCHDOG:
513
	case KVM_CAP_PPC_EPR:
S
Scott Wood 已提交
514
#else
515
	case KVM_CAP_PPC_SEGSTATE:
516
	case KVM_CAP_PPC_HIOR:
517
	case KVM_CAP_PPC_PAPR:
S
Scott Wood 已提交
518
#endif
519
	case KVM_CAP_PPC_UNSET_IRQ:
520
	case KVM_CAP_PPC_IRQ_LEVEL:
521
	case KVM_CAP_ENABLE_CAP:
522
	case KVM_CAP_ENABLE_CAP_VM:
523
	case KVM_CAP_ONE_REG:
A
Alexander Graf 已提交
524
	case KVM_CAP_IOEVENTFD:
525
	case KVM_CAP_DEVICE_CTRL:
526
	case KVM_CAP_IMMEDIATE_EXIT:
527 528 529
		r = 1;
		break;
	case KVM_CAP_PPC_PAIRED_SINGLES:
530
	case KVM_CAP_PPC_OSI:
531
	case KVM_CAP_PPC_GET_PVINFO:
532
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
533
	case KVM_CAP_SW_TLB:
S
Scott Wood 已提交
534
#endif
535
		/* We support this only for PR */
536
		r = !hv_enabled;
537
		break;
538 539 540 541 542 543
#ifdef CONFIG_KVM_MPIC
	case KVM_CAP_IRQ_MPIC:
		r = 1;
		break;
#endif

544
#ifdef CONFIG_PPC_BOOK3S_64
545
	case KVM_CAP_SPAPR_TCE:
546
	case KVM_CAP_SPAPR_TCE_64:
547 548
		/* fallthrough */
	case KVM_CAP_SPAPR_TCE_VFIO:
549
	case KVM_CAP_PPC_RTAS:
550
	case KVM_CAP_PPC_FIXUP_HCALL:
551
	case KVM_CAP_PPC_ENABLE_HCALL:
552 553 554
#ifdef CONFIG_KVM_XICS
	case KVM_CAP_IRQ_XICS:
#endif
555
	case KVM_CAP_PPC_GET_CPU_CHAR:
556 557
		r = 1;
		break;
558 559 560 561

	case KVM_CAP_PPC_ALLOC_HTAB:
		r = hv_enabled;
		break;
562
#endif /* CONFIG_PPC_BOOK3S_64 */
563
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
564
	case KVM_CAP_PPC_SMT:
565
		r = 0;
566 567 568 569 570 571
		if (kvm) {
			if (kvm->arch.emul_smt_mode > 1)
				r = kvm->arch.emul_smt_mode;
			else
				r = kvm->arch.smt_mode;
		} else if (hv_enabled) {
572 573 574 575 576
			if (cpu_has_feature(CPU_FTR_ARCH_300))
				r = 1;
			else
				r = threads_per_subcore;
		}
577
		break;
578 579 580 581 582 583 584 585 586 587
	case KVM_CAP_PPC_SMT_POSSIBLE:
		r = 1;
		if (hv_enabled) {
			if (!cpu_has_feature(CPU_FTR_ARCH_300))
				r = ((threads_per_subcore << 1) - 1);
			else
				/* P9 can emulate dbells, so allow any mode */
				r = 8 | 4 | 2 | 1;
		}
		break;
588
	case KVM_CAP_PPC_RMA:
589
		r = 0;
590
		break;
591 592 593
	case KVM_CAP_PPC_HWRNG:
		r = kvmppc_hwrng_present();
		break;
594
	case KVM_CAP_PPC_MMU_RADIX:
595
		r = !!(hv_enabled && radix_enabled());
596 597
		break;
	case KVM_CAP_PPC_MMU_HASH_V3:
598
		r = !!(hv_enabled && cpu_has_feature(CPU_FTR_ARCH_300));
599
		break;
600
#endif
601
	case KVM_CAP_SYNC_MMU:
602
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
603
		r = hv_enabled;
604 605 606 607
#elif defined(KVM_ARCH_WANT_MMU_NOTIFIER)
		r = 1;
#else
		r = 0;
608
#endif
609 610
		break;
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
611
	case KVM_CAP_PPC_HTAB_FD:
612
		r = hv_enabled;
613
		break;
614
#endif
615 616 617 618 619 620 621
	case KVM_CAP_NR_VCPUS:
		/*
		 * Recommending a number of CPUs is somewhat arbitrary; we
		 * return the number of present CPUs for -HV (since a host
		 * will have secondary threads "offline"), and for other KVM
		 * implementations just count online CPUs.
		 */
622
		if (hv_enabled)
623 624 625
			r = num_present_cpus();
		else
			r = num_online_cpus();
626
		break;
627 628 629
	case KVM_CAP_NR_MEMSLOTS:
		r = KVM_USER_MEM_SLOTS;
		break;
630 631 632
	case KVM_CAP_MAX_VCPUS:
		r = KVM_MAX_VCPUS;
		break;
633 634 635 636
#ifdef CONFIG_PPC_BOOK3S_64
	case KVM_CAP_PPC_GET_SMMU_INFO:
		r = 1;
		break;
637 638 639
	case KVM_CAP_SPAPR_MULTITCE:
		r = 1;
		break;
640
	case KVM_CAP_SPAPR_RESIZE_HPT:
641
		r = !!hv_enabled;
642
		break;
643 644 645 646 647
#endif
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
	case KVM_CAP_PPC_FWNMI:
		r = hv_enabled;
		break;
648
#endif
649
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
650
	case KVM_CAP_PPC_HTM:
651
		r = hv_enabled &&
652 653
		    (!!(cur_cpu_spec->cpu_user_features2 & PPC_FEATURE2_HTM) ||
		     cpu_has_feature(CPU_FTR_P9_TM_HV_ASSIST));
654
		break;
655
#endif
656 657 658 659 660 661 662 663 664 665 666 667 668 669
	default:
		r = 0;
		break;
	}
	return r;

}

long kvm_arch_dev_ioctl(struct file *filp,
                        unsigned int ioctl, unsigned long arg)
{
	return -EINVAL;
}

670
void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
671 672
			   struct kvm_memory_slot *dont)
{
673
	kvmppc_core_free_memslot(kvm, free, dont);
674 675
}

676 677
int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
			    unsigned long npages)
678
{
679
	return kvmppc_core_create_memslot(kvm, slot, npages);
680 681
}

682
int kvm_arch_prepare_memory_region(struct kvm *kvm,
683
				   struct kvm_memory_slot *memslot,
684
				   const struct kvm_userspace_memory_region *mem,
685
				   enum kvm_mr_change change)
686
{
687
	return kvmppc_core_prepare_memory_region(kvm, memslot, mem);
688 689
}

690
void kvm_arch_commit_memory_region(struct kvm *kvm,
691
				   const struct kvm_userspace_memory_region *mem,
692
				   const struct kvm_memory_slot *old,
693
				   const struct kvm_memory_slot *new,
694
				   enum kvm_mr_change change)
695
{
696
	kvmppc_core_commit_memory_region(kvm, mem, old, new);
697 698
}

699 700
void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
				   struct kvm_memory_slot *slot)
701
{
702
	kvmppc_core_flush_memslot(kvm, slot);
703 704
}

705 706
struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
{
707 708
	struct kvm_vcpu *vcpu;
	vcpu = kvmppc_core_vcpu_create(kvm, id);
709 710
	if (!IS_ERR(vcpu)) {
		vcpu->arch.wqp = &vcpu->wq;
711
		kvmppc_create_vcpu_debugfs(vcpu, id);
712
	}
713
	return vcpu;
714 715
}

716
void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
717 718 719
{
}

720 721
void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
{
722 723 724
	/* Make sure we're not using the vcpu anymore */
	hrtimer_cancel(&vcpu->arch.dec_timer);

725
	kvmppc_remove_vcpu_debugfs(vcpu);
S
Scott Wood 已提交
726 727 728 729 730

	switch (vcpu->arch.irq_type) {
	case KVMPPC_IRQ_MPIC:
		kvmppc_mpic_disconnect_vcpu(vcpu->arch.mpic, vcpu);
		break;
731
	case KVMPPC_IRQ_XICS:
732 733 734 735
		if (xive_enabled())
			kvmppc_xive_cleanup_vcpu(vcpu);
		else
			kvmppc_xics_free_icp(vcpu);
736
		break;
S
Scott Wood 已提交
737 738
	}

739
	kvmppc_core_vcpu_free(vcpu);
740 741 742 743 744 745 746 747 748
}

void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
{
	kvm_arch_vcpu_free(vcpu);
}

int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
{
749
	return kvmppc_core_pending_dec(vcpu);
750 751
}

T
Thomas Huth 已提交
752
static enum hrtimer_restart kvmppc_decrementer_wakeup(struct hrtimer *timer)
A
Alexander Graf 已提交
753 754 755 756
{
	struct kvm_vcpu *vcpu;

	vcpu = container_of(timer, struct kvm_vcpu, arch.dec_timer);
757
	kvmppc_decrementer_func(vcpu);
A
Alexander Graf 已提交
758 759 760 761

	return HRTIMER_NORESTART;
}

762 763
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
764 765
	int ret;

A
Alexander Graf 已提交
766 767
	hrtimer_init(&vcpu->arch.dec_timer, CLOCK_REALTIME, HRTIMER_MODE_ABS);
	vcpu->arch.dec_timer.function = kvmppc_decrementer_wakeup;
768
	vcpu->arch.dec_expires = get_tb();
769

770 771 772
#ifdef CONFIG_KVM_EXIT_TIMING
	mutex_init(&vcpu->arch.exit_timing_lock);
#endif
773 774
	ret = kvmppc_subarch_vcpu_init(vcpu);
	return ret;
775 776 777 778
}

void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
{
779
	kvmppc_mmu_destroy(vcpu);
780
	kvmppc_subarch_vcpu_uninit(vcpu);
781 782 783 784
}

void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
785 786 787 788 789 790 791 792 793 794
#ifdef CONFIG_BOOKE
	/*
	 * vrsave (formerly usprg0) isn't used by Linux, but may
	 * be used by the guest.
	 *
	 * On non-booke this is associated with Altivec and
	 * is handled by code in book3s.c.
	 */
	mtspr(SPRN_VRSAVE, vcpu->arch.vrsave);
#endif
795
	kvmppc_core_vcpu_load(vcpu, cpu);
796 797 798 799
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
800
	kvmppc_core_vcpu_put(vcpu);
801 802 803
#ifdef CONFIG_BOOKE
	vcpu->arch.vrsave = mfspr(SPRN_VRSAVE);
#endif
804 805
}

806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841
/*
 * irq_bypass_add_producer and irq_bypass_del_producer are only
 * useful if the architecture supports PCI passthrough.
 * irq_bypass_stop and irq_bypass_start are not needed and so
 * kvm_ops are not defined for them.
 */
bool kvm_arch_has_irq_bypass(void)
{
	return ((kvmppc_hv_ops && kvmppc_hv_ops->irq_bypass_add_producer) ||
		(kvmppc_pr_ops && kvmppc_pr_ops->irq_bypass_add_producer));
}

int kvm_arch_irq_bypass_add_producer(struct irq_bypass_consumer *cons,
				     struct irq_bypass_producer *prod)
{
	struct kvm_kernel_irqfd *irqfd =
		container_of(cons, struct kvm_kernel_irqfd, consumer);
	struct kvm *kvm = irqfd->kvm;

	if (kvm->arch.kvm_ops->irq_bypass_add_producer)
		return kvm->arch.kvm_ops->irq_bypass_add_producer(cons, prod);

	return 0;
}

void kvm_arch_irq_bypass_del_producer(struct irq_bypass_consumer *cons,
				      struct irq_bypass_producer *prod)
{
	struct kvm_kernel_irqfd *irqfd =
		container_of(cons, struct kvm_kernel_irqfd, consumer);
	struct kvm *kvm = irqfd->kvm;

	if (kvm->arch.kvm_ops->irq_bypass_del_producer)
		kvm->arch.kvm_ops->irq_bypass_del_producer(cons, prod);
}

842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909
#ifdef CONFIG_VSX
static inline int kvmppc_get_vsr_dword_offset(int index)
{
	int offset;

	if ((index != 0) && (index != 1))
		return -1;

#ifdef __BIG_ENDIAN
	offset =  index;
#else
	offset = 1 - index;
#endif

	return offset;
}

static inline int kvmppc_get_vsr_word_offset(int index)
{
	int offset;

	if ((index > 3) || (index < 0))
		return -1;

#ifdef __BIG_ENDIAN
	offset = index;
#else
	offset = 3 - index;
#endif
	return offset;
}

static inline void kvmppc_set_vsr_dword(struct kvm_vcpu *vcpu,
	u64 gpr)
{
	union kvmppc_one_reg val;
	int offset = kvmppc_get_vsr_dword_offset(vcpu->arch.mmio_vsx_offset);
	int index = vcpu->arch.io_gpr & KVM_MMIO_REG_MASK;

	if (offset == -1)
		return;

	if (vcpu->arch.mmio_vsx_tx_sx_enabled) {
		val.vval = VCPU_VSX_VR(vcpu, index);
		val.vsxval[offset] = gpr;
		VCPU_VSX_VR(vcpu, index) = val.vval;
	} else {
		VCPU_VSX_FPR(vcpu, index, offset) = gpr;
	}
}

static inline void kvmppc_set_vsr_dword_dump(struct kvm_vcpu *vcpu,
	u64 gpr)
{
	union kvmppc_one_reg val;
	int index = vcpu->arch.io_gpr & KVM_MMIO_REG_MASK;

	if (vcpu->arch.mmio_vsx_tx_sx_enabled) {
		val.vval = VCPU_VSX_VR(vcpu, index);
		val.vsxval[0] = gpr;
		val.vsxval[1] = gpr;
		VCPU_VSX_VR(vcpu, index) = val.vval;
	} else {
		VCPU_VSX_FPR(vcpu, index, 0) = gpr;
		VCPU_VSX_FPR(vcpu, index, 1) = gpr;
	}
}

910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929
static inline void kvmppc_set_vsr_word_dump(struct kvm_vcpu *vcpu,
	u32 gpr)
{
	union kvmppc_one_reg val;
	int index = vcpu->arch.io_gpr & KVM_MMIO_REG_MASK;

	if (vcpu->arch.mmio_vsx_tx_sx_enabled) {
		val.vsx32val[0] = gpr;
		val.vsx32val[1] = gpr;
		val.vsx32val[2] = gpr;
		val.vsx32val[3] = gpr;
		VCPU_VSX_VR(vcpu, index) = val.vval;
	} else {
		val.vsx32val[0] = gpr;
		val.vsx32val[1] = gpr;
		VCPU_VSX_FPR(vcpu, index, 0) = val.vsxval[0];
		VCPU_VSX_FPR(vcpu, index, 1) = val.vsxval[0];
	}
}

930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954
static inline void kvmppc_set_vsr_word(struct kvm_vcpu *vcpu,
	u32 gpr32)
{
	union kvmppc_one_reg val;
	int offset = kvmppc_get_vsr_word_offset(vcpu->arch.mmio_vsx_offset);
	int index = vcpu->arch.io_gpr & KVM_MMIO_REG_MASK;
	int dword_offset, word_offset;

	if (offset == -1)
		return;

	if (vcpu->arch.mmio_vsx_tx_sx_enabled) {
		val.vval = VCPU_VSX_VR(vcpu, index);
		val.vsx32val[offset] = gpr32;
		VCPU_VSX_VR(vcpu, index) = val.vval;
	} else {
		dword_offset = offset / 2;
		word_offset = offset % 2;
		val.vsxval[0] = VCPU_VSX_FPR(vcpu, index, dword_offset);
		val.vsx32val[word_offset] = gpr32;
		VCPU_VSX_FPR(vcpu, index, dword_offset) = val.vsxval[0];
	}
}
#endif /* CONFIG_VSX */

955
#ifdef CONFIG_ALTIVEC
956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997
static inline int kvmppc_get_vmx_offset_generic(struct kvm_vcpu *vcpu,
		int index, int element_size)
{
	int offset;
	int elts = sizeof(vector128)/element_size;

	if ((index < 0) || (index >= elts))
		return -1;

	if (kvmppc_need_byteswap(vcpu))
		offset = elts - index - 1;
	else
		offset = index;

	return offset;
}

static inline int kvmppc_get_vmx_dword_offset(struct kvm_vcpu *vcpu,
		int index)
{
	return kvmppc_get_vmx_offset_generic(vcpu, index, 8);
}

static inline int kvmppc_get_vmx_word_offset(struct kvm_vcpu *vcpu,
		int index)
{
	return kvmppc_get_vmx_offset_generic(vcpu, index, 4);
}

static inline int kvmppc_get_vmx_hword_offset(struct kvm_vcpu *vcpu,
		int index)
{
	return kvmppc_get_vmx_offset_generic(vcpu, index, 2);
}

static inline int kvmppc_get_vmx_byte_offset(struct kvm_vcpu *vcpu,
		int index)
{
	return kvmppc_get_vmx_offset_generic(vcpu, index, 1);
}


998
static inline void kvmppc_set_vmx_dword(struct kvm_vcpu *vcpu,
999
	u64 gpr)
1000
{
1001 1002 1003
	union kvmppc_one_reg val;
	int offset = kvmppc_get_vmx_dword_offset(vcpu,
			vcpu->arch.mmio_vmx_offset);
1004 1005
	int index = vcpu->arch.io_gpr & KVM_MMIO_REG_MASK;

1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028
	if (offset == -1)
		return;

	val.vval = VCPU_VSX_VR(vcpu, index);
	val.vsxval[offset] = gpr;
	VCPU_VSX_VR(vcpu, index) = val.vval;
}

static inline void kvmppc_set_vmx_word(struct kvm_vcpu *vcpu,
	u32 gpr32)
{
	union kvmppc_one_reg val;
	int offset = kvmppc_get_vmx_word_offset(vcpu,
			vcpu->arch.mmio_vmx_offset);
	int index = vcpu->arch.io_gpr & KVM_MMIO_REG_MASK;

	if (offset == -1)
		return;

	val.vval = VCPU_VSX_VR(vcpu, index);
	val.vsx32val[offset] = gpr32;
	VCPU_VSX_VR(vcpu, index) = val.vval;
}
1029

1030 1031 1032 1033 1034 1035 1036 1037 1038
static inline void kvmppc_set_vmx_hword(struct kvm_vcpu *vcpu,
	u16 gpr16)
{
	union kvmppc_one_reg val;
	int offset = kvmppc_get_vmx_hword_offset(vcpu,
			vcpu->arch.mmio_vmx_offset);
	int index = vcpu->arch.io_gpr & KVM_MMIO_REG_MASK;

	if (offset == -1)
1039 1040
		return;

1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052
	val.vval = VCPU_VSX_VR(vcpu, index);
	val.vsx16val[offset] = gpr16;
	VCPU_VSX_VR(vcpu, index) = val.vval;
}

static inline void kvmppc_set_vmx_byte(struct kvm_vcpu *vcpu,
	u8 gpr8)
{
	union kvmppc_one_reg val;
	int offset = kvmppc_get_vmx_byte_offset(vcpu,
			vcpu->arch.mmio_vmx_offset);
	int index = vcpu->arch.io_gpr & KVM_MMIO_REG_MASK;
1053

1054 1055 1056 1057 1058 1059
	if (offset == -1)
		return;

	val.vval = VCPU_VSX_VR(vcpu, index);
	val.vsx8val[offset] = gpr8;
	VCPU_VSX_VR(vcpu, index) = val.vval;
1060 1061 1062
}
#endif /* CONFIG_ALTIVEC */

1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092
#ifdef CONFIG_PPC_FPU
static inline u64 sp_to_dp(u32 fprs)
{
	u64 fprd;

	preempt_disable();
	enable_kernel_fp();
	asm ("lfs%U1%X1 0,%1; stfd%U0%X0 0,%0" : "=m" (fprd) : "m" (fprs)
	     : "fr0");
	preempt_enable();
	return fprd;
}

static inline u32 dp_to_sp(u64 fprd)
{
	u32 fprs;

	preempt_disable();
	enable_kernel_fp();
	asm ("lfd%U1%X1 0,%1; stfs%U0%X0 0,%0" : "=m" (fprs) : "m" (fprd)
	     : "fr0");
	preempt_enable();
	return fprs;
}

#else
#define sp_to_dp(x)	(x)
#define dp_to_sp(x)	(x)
#endif /* CONFIG_PPC_FPU */

1093 1094 1095
static void kvmppc_complete_mmio_load(struct kvm_vcpu *vcpu,
                                      struct kvm_run *run)
{
1096
	u64 uninitialized_var(gpr);
1097

1098
	if (run->mmio.len > sizeof(gpr)) {
1099 1100 1101 1102
		printk(KERN_ERR "bad MMIO length: %d\n", run->mmio.len);
		return;
	}

1103
	if (!vcpu->arch.mmio_host_swabbed) {
1104
		switch (run->mmio.len) {
1105
		case 8: gpr = *(u64 *)run->mmio.data; break;
1106 1107 1108
		case 4: gpr = *(u32 *)run->mmio.data; break;
		case 2: gpr = *(u16 *)run->mmio.data; break;
		case 1: gpr = *(u8 *)run->mmio.data; break;
1109 1110 1111
		}
	} else {
		switch (run->mmio.len) {
1112 1113 1114
		case 8: gpr = swab64(*(u64 *)run->mmio.data); break;
		case 4: gpr = swab32(*(u32 *)run->mmio.data); break;
		case 2: gpr = swab16(*(u16 *)run->mmio.data); break;
1115
		case 1: gpr = *(u8 *)run->mmio.data; break;
1116 1117
		}
	}
1118

1119 1120 1121 1122
	/* conversion between single and double precision */
	if ((vcpu->arch.mmio_sp64_extend) && (run->mmio.len == 4))
		gpr = sp_to_dp(gpr);

A
Alexander Graf 已提交
1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138
	if (vcpu->arch.mmio_sign_extend) {
		switch (run->mmio.len) {
#ifdef CONFIG_PPC64
		case 4:
			gpr = (s64)(s32)gpr;
			break;
#endif
		case 2:
			gpr = (s64)(s16)gpr;
			break;
		case 1:
			gpr = (s64)(s8)gpr;
			break;
		}
	}

1139 1140
	switch (vcpu->arch.io_gpr & KVM_MMIO_REG_EXT_MASK) {
	case KVM_MMIO_REG_GPR:
1141 1142
		kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
		break;
1143
	case KVM_MMIO_REG_FPR:
1144 1145 1146
		if (vcpu->kvm->arch.kvm_ops->giveup_ext)
			vcpu->kvm->arch.kvm_ops->giveup_ext(vcpu, MSR_FP);

1147
		VCPU_FPR(vcpu, vcpu->arch.io_gpr & KVM_MMIO_REG_MASK) = gpr;
1148
		break;
1149
#ifdef CONFIG_PPC_BOOK3S
1150 1151
	case KVM_MMIO_REG_QPR:
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
1152
		break;
1153
	case KVM_MMIO_REG_FQPR:
1154
		VCPU_FPR(vcpu, vcpu->arch.io_gpr & KVM_MMIO_REG_MASK) = gpr;
1155
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
1156
		break;
1157 1158 1159
#endif
#ifdef CONFIG_VSX
	case KVM_MMIO_REG_VSX:
1160 1161 1162
		if (vcpu->kvm->arch.kvm_ops->giveup_ext)
			vcpu->kvm->arch.kvm_ops->giveup_ext(vcpu, MSR_VSX);

1163
		if (vcpu->arch.mmio_copy_type == KVMPPC_VSX_COPY_DWORD)
1164
			kvmppc_set_vsr_dword(vcpu, gpr);
1165
		else if (vcpu->arch.mmio_copy_type == KVMPPC_VSX_COPY_WORD)
1166
			kvmppc_set_vsr_word(vcpu, gpr);
1167
		else if (vcpu->arch.mmio_copy_type ==
1168 1169
				KVMPPC_VSX_COPY_DWORD_LOAD_DUMP)
			kvmppc_set_vsr_dword_dump(vcpu, gpr);
1170
		else if (vcpu->arch.mmio_copy_type ==
1171 1172
				KVMPPC_VSX_COPY_WORD_LOAD_DUMP)
			kvmppc_set_vsr_word_dump(vcpu, gpr);
1173
		break;
1174 1175 1176
#endif
#ifdef CONFIG_ALTIVEC
	case KVM_MMIO_REG_VMX:
1177 1178 1179
		if (vcpu->kvm->arch.kvm_ops->giveup_ext)
			vcpu->kvm->arch.kvm_ops->giveup_ext(vcpu, MSR_VEC);

1180 1181 1182 1183 1184 1185 1186 1187 1188 1189
		if (vcpu->arch.mmio_copy_type == KVMPPC_VMX_COPY_DWORD)
			kvmppc_set_vmx_dword(vcpu, gpr);
		else if (vcpu->arch.mmio_copy_type == KVMPPC_VMX_COPY_WORD)
			kvmppc_set_vmx_word(vcpu, gpr);
		else if (vcpu->arch.mmio_copy_type ==
				KVMPPC_VMX_COPY_HWORD)
			kvmppc_set_vmx_hword(vcpu, gpr);
		else if (vcpu->arch.mmio_copy_type ==
				KVMPPC_VMX_COPY_BYTE)
			kvmppc_set_vmx_byte(vcpu, gpr);
1190
		break;
1191
#endif
1192 1193 1194
	default:
		BUG();
	}
1195 1196
}

1197 1198 1199
static int __kvmppc_handle_load(struct kvm_run *run, struct kvm_vcpu *vcpu,
				unsigned int rt, unsigned int bytes,
				int is_default_endian, int sign_extend)
1200
{
1201
	int idx, ret;
1202
	bool host_swabbed;
1203

1204
	/* Pity C doesn't have a logical XOR operator */
1205
	if (kvmppc_need_byteswap(vcpu)) {
1206
		host_swabbed = is_default_endian;
1207
	} else {
1208
		host_swabbed = !is_default_endian;
1209
	}
1210

1211 1212 1213 1214 1215 1216 1217 1218 1219 1220
	if (bytes > sizeof(run->mmio.data)) {
		printk(KERN_ERR "%s: bad MMIO length: %d\n", __func__,
		       run->mmio.len);
	}

	run->mmio.phys_addr = vcpu->arch.paddr_accessed;
	run->mmio.len = bytes;
	run->mmio.is_write = 0;

	vcpu->arch.io_gpr = rt;
1221
	vcpu->arch.mmio_host_swabbed = host_swabbed;
1222 1223
	vcpu->mmio_needed = 1;
	vcpu->mmio_is_write = 0;
1224
	vcpu->arch.mmio_sign_extend = sign_extend;
1225

1226 1227
	idx = srcu_read_lock(&vcpu->kvm->srcu);

1228
	ret = kvm_io_bus_read(vcpu, KVM_MMIO_BUS, run->mmio.phys_addr,
1229 1230 1231 1232 1233
			      bytes, &run->mmio.data);

	srcu_read_unlock(&vcpu->kvm->srcu, idx);

	if (!ret) {
A
Alexander Graf 已提交
1234 1235 1236 1237 1238
		kvmppc_complete_mmio_load(vcpu, run);
		vcpu->mmio_needed = 0;
		return EMULATE_DONE;
	}

1239 1240
	return EMULATE_DO_MMIO;
}
1241 1242 1243 1244 1245 1246 1247

int kvmppc_handle_load(struct kvm_run *run, struct kvm_vcpu *vcpu,
		       unsigned int rt, unsigned int bytes,
		       int is_default_endian)
{
	return __kvmppc_handle_load(run, vcpu, rt, bytes, is_default_endian, 0);
}
1248
EXPORT_SYMBOL_GPL(kvmppc_handle_load);
1249

A
Alexander Graf 已提交
1250 1251
/* Same as above, but sign extends */
int kvmppc_handle_loads(struct kvm_run *run, struct kvm_vcpu *vcpu,
1252 1253
			unsigned int rt, unsigned int bytes,
			int is_default_endian)
A
Alexander Graf 已提交
1254
{
1255
	return __kvmppc_handle_load(run, vcpu, rt, bytes, is_default_endian, 1);
A
Alexander Graf 已提交
1256 1257
}

1258 1259 1260 1261 1262 1263 1264
#ifdef CONFIG_VSX
int kvmppc_handle_vsx_load(struct kvm_run *run, struct kvm_vcpu *vcpu,
			unsigned int rt, unsigned int bytes,
			int is_default_endian, int mmio_sign_extend)
{
	enum emulation_result emulated = EMULATE_DONE;

1265 1266
	/* Currently, mmio_vsx_copy_nums only allowed to be 4 or less */
	if (vcpu->arch.mmio_vsx_copy_nums > 4)
1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284
		return EMULATE_FAIL;

	while (vcpu->arch.mmio_vsx_copy_nums) {
		emulated = __kvmppc_handle_load(run, vcpu, rt, bytes,
			is_default_endian, mmio_sign_extend);

		if (emulated != EMULATE_DONE)
			break;

		vcpu->arch.paddr_accessed += run->mmio.len;

		vcpu->arch.mmio_vsx_copy_nums--;
		vcpu->arch.mmio_vsx_offset++;
	}
	return emulated;
}
#endif /* CONFIG_VSX */

1285
int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
1286
			u64 val, unsigned int bytes, int is_default_endian)
1287 1288
{
	void *data = run->mmio.data;
1289
	int idx, ret;
1290
	bool host_swabbed;
1291

1292
	/* Pity C doesn't have a logical XOR operator */
1293
	if (kvmppc_need_byteswap(vcpu)) {
1294
		host_swabbed = is_default_endian;
1295
	} else {
1296
		host_swabbed = !is_default_endian;
1297
	}
1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309

	if (bytes > sizeof(run->mmio.data)) {
		printk(KERN_ERR "%s: bad MMIO length: %d\n", __func__,
		       run->mmio.len);
	}

	run->mmio.phys_addr = vcpu->arch.paddr_accessed;
	run->mmio.len = bytes;
	run->mmio.is_write = 1;
	vcpu->mmio_needed = 1;
	vcpu->mmio_is_write = 1;

1310 1311 1312
	if ((vcpu->arch.mmio_sp64_extend) && (bytes == 4))
		val = dp_to_sp(val);

1313
	/* Store the value at the lowest bytes in 'data'. */
1314
	if (!host_swabbed) {
1315
		switch (bytes) {
1316
		case 8: *(u64 *)data = val; break;
1317 1318 1319 1320 1321 1322
		case 4: *(u32 *)data = val; break;
		case 2: *(u16 *)data = val; break;
		case 1: *(u8  *)data = val; break;
		}
	} else {
		switch (bytes) {
1323 1324 1325 1326
		case 8: *(u64 *)data = swab64(val); break;
		case 4: *(u32 *)data = swab32(val); break;
		case 2: *(u16 *)data = swab16(val); break;
		case 1: *(u8  *)data = val; break;
1327 1328 1329
		}
	}

1330 1331
	idx = srcu_read_lock(&vcpu->kvm->srcu);

1332
	ret = kvm_io_bus_write(vcpu, KVM_MMIO_BUS, run->mmio.phys_addr,
1333 1334 1335 1336 1337
			       bytes, &run->mmio.data);

	srcu_read_unlock(&vcpu->kvm->srcu, idx);

	if (!ret) {
A
Alexander Graf 已提交
1338 1339 1340 1341
		vcpu->mmio_needed = 0;
		return EMULATE_DONE;
	}

1342 1343
	return EMULATE_DO_MMIO;
}
1344
EXPORT_SYMBOL_GPL(kvmppc_handle_store);
1345

1346 1347 1348 1349 1350 1351
#ifdef CONFIG_VSX
static inline int kvmppc_get_vsr_data(struct kvm_vcpu *vcpu, int rs, u64 *val)
{
	u32 dword_offset, word_offset;
	union kvmppc_one_reg reg;
	int vsx_offset = 0;
1352
	int copy_type = vcpu->arch.mmio_copy_type;
1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408
	int result = 0;

	switch (copy_type) {
	case KVMPPC_VSX_COPY_DWORD:
		vsx_offset =
			kvmppc_get_vsr_dword_offset(vcpu->arch.mmio_vsx_offset);

		if (vsx_offset == -1) {
			result = -1;
			break;
		}

		if (!vcpu->arch.mmio_vsx_tx_sx_enabled) {
			*val = VCPU_VSX_FPR(vcpu, rs, vsx_offset);
		} else {
			reg.vval = VCPU_VSX_VR(vcpu, rs);
			*val = reg.vsxval[vsx_offset];
		}
		break;

	case KVMPPC_VSX_COPY_WORD:
		vsx_offset =
			kvmppc_get_vsr_word_offset(vcpu->arch.mmio_vsx_offset);

		if (vsx_offset == -1) {
			result = -1;
			break;
		}

		if (!vcpu->arch.mmio_vsx_tx_sx_enabled) {
			dword_offset = vsx_offset / 2;
			word_offset = vsx_offset % 2;
			reg.vsxval[0] = VCPU_VSX_FPR(vcpu, rs, dword_offset);
			*val = reg.vsx32val[word_offset];
		} else {
			reg.vval = VCPU_VSX_VR(vcpu, rs);
			*val = reg.vsx32val[vsx_offset];
		}
		break;

	default:
		result = -1;
		break;
	}

	return result;
}

int kvmppc_handle_vsx_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
			int rs, unsigned int bytes, int is_default_endian)
{
	u64 val;
	enum emulation_result emulated = EMULATE_DONE;

	vcpu->arch.io_gpr = rs;

1409 1410
	/* Currently, mmio_vsx_copy_nums only allowed to be 4 or less */
	if (vcpu->arch.mmio_vsx_copy_nums > 4)
1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466
		return EMULATE_FAIL;

	while (vcpu->arch.mmio_vsx_copy_nums) {
		if (kvmppc_get_vsr_data(vcpu, rs, &val) == -1)
			return EMULATE_FAIL;

		emulated = kvmppc_handle_store(run, vcpu,
			 val, bytes, is_default_endian);

		if (emulated != EMULATE_DONE)
			break;

		vcpu->arch.paddr_accessed += run->mmio.len;

		vcpu->arch.mmio_vsx_copy_nums--;
		vcpu->arch.mmio_vsx_offset++;
	}

	return emulated;
}

static int kvmppc_emulate_mmio_vsx_loadstore(struct kvm_vcpu *vcpu,
			struct kvm_run *run)
{
	enum emulation_result emulated = EMULATE_FAIL;
	int r;

	vcpu->arch.paddr_accessed += run->mmio.len;

	if (!vcpu->mmio_is_write) {
		emulated = kvmppc_handle_vsx_load(run, vcpu, vcpu->arch.io_gpr,
			 run->mmio.len, 1, vcpu->arch.mmio_sign_extend);
	} else {
		emulated = kvmppc_handle_vsx_store(run, vcpu,
			 vcpu->arch.io_gpr, run->mmio.len, 1);
	}

	switch (emulated) {
	case EMULATE_DO_MMIO:
		run->exit_reason = KVM_EXIT_MMIO;
		r = RESUME_HOST;
		break;
	case EMULATE_FAIL:
		pr_info("KVM: MMIO emulation failed (VSX repeat)\n");
		run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
		run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION;
		r = RESUME_HOST;
		break;
	default:
		r = RESUME_GUEST;
		break;
	}
	return r;
}
#endif /* CONFIG_VSX */

1467
#ifdef CONFIG_ALTIVEC
1468 1469
int kvmppc_handle_vmx_load(struct kvm_run *run, struct kvm_vcpu *vcpu,
		unsigned int rt, unsigned int bytes, int is_default_endian)
1470
{
1471
	enum emulation_result emulated = EMULATE_DONE;
1472

1473 1474 1475
	if (vcpu->arch.mmio_vsx_copy_nums > 2)
		return EMULATE_FAIL;

1476
	while (vcpu->arch.mmio_vmx_copy_nums) {
1477
		emulated = __kvmppc_handle_load(run, vcpu, rt, bytes,
1478 1479 1480 1481 1482 1483 1484
				is_default_endian, 0);

		if (emulated != EMULATE_DONE)
			break;

		vcpu->arch.paddr_accessed += run->mmio.len;
		vcpu->arch.mmio_vmx_copy_nums--;
1485
		vcpu->arch.mmio_vmx_offset++;
1486 1487 1488 1489 1490
	}

	return emulated;
}

1491
int kvmppc_get_vmx_dword(struct kvm_vcpu *vcpu, int index, u64 *val)
1492
{
1493 1494 1495 1496 1497 1498
	union kvmppc_one_reg reg;
	int vmx_offset = 0;
	int result = 0;

	vmx_offset =
		kvmppc_get_vmx_dword_offset(vcpu, vcpu->arch.mmio_vmx_offset);
1499

1500
	if (vmx_offset == -1)
1501 1502
		return -1;

1503 1504
	reg.vval = VCPU_VSX_VR(vcpu, index);
	*val = reg.vsxval[vmx_offset];
1505

1506 1507
	return result;
}
1508

1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560
int kvmppc_get_vmx_word(struct kvm_vcpu *vcpu, int index, u64 *val)
{
	union kvmppc_one_reg reg;
	int vmx_offset = 0;
	int result = 0;

	vmx_offset =
		kvmppc_get_vmx_word_offset(vcpu, vcpu->arch.mmio_vmx_offset);

	if (vmx_offset == -1)
		return -1;

	reg.vval = VCPU_VSX_VR(vcpu, index);
	*val = reg.vsx32val[vmx_offset];

	return result;
}

int kvmppc_get_vmx_hword(struct kvm_vcpu *vcpu, int index, u64 *val)
{
	union kvmppc_one_reg reg;
	int vmx_offset = 0;
	int result = 0;

	vmx_offset =
		kvmppc_get_vmx_hword_offset(vcpu, vcpu->arch.mmio_vmx_offset);

	if (vmx_offset == -1)
		return -1;

	reg.vval = VCPU_VSX_VR(vcpu, index);
	*val = reg.vsx16val[vmx_offset];

	return result;
}

int kvmppc_get_vmx_byte(struct kvm_vcpu *vcpu, int index, u64 *val)
{
	union kvmppc_one_reg reg;
	int vmx_offset = 0;
	int result = 0;

	vmx_offset =
		kvmppc_get_vmx_byte_offset(vcpu, vcpu->arch.mmio_vmx_offset);

	if (vmx_offset == -1)
		return -1;

	reg.vval = VCPU_VSX_VR(vcpu, index);
	*val = reg.vsx8val[vmx_offset];

	return result;
1561 1562
}

1563 1564
int kvmppc_handle_vmx_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
		unsigned int rs, unsigned int bytes, int is_default_endian)
1565 1566
{
	u64 val = 0;
1567
	unsigned int index = rs & KVM_MMIO_REG_MASK;
1568 1569
	enum emulation_result emulated = EMULATE_DONE;

1570 1571 1572
	if (vcpu->arch.mmio_vsx_copy_nums > 2)
		return EMULATE_FAIL;

1573 1574 1575
	vcpu->arch.io_gpr = rs;

	while (vcpu->arch.mmio_vmx_copy_nums) {
1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594
		switch (vcpu->arch.mmio_copy_type) {
		case KVMPPC_VMX_COPY_DWORD:
			if (kvmppc_get_vmx_dword(vcpu, index, &val) == -1)
				return EMULATE_FAIL;

			break;
		case KVMPPC_VMX_COPY_WORD:
			if (kvmppc_get_vmx_word(vcpu, index, &val) == -1)
				return EMULATE_FAIL;
			break;
		case KVMPPC_VMX_COPY_HWORD:
			if (kvmppc_get_vmx_hword(vcpu, index, &val) == -1)
				return EMULATE_FAIL;
			break;
		case KVMPPC_VMX_COPY_BYTE:
			if (kvmppc_get_vmx_byte(vcpu, index, &val) == -1)
				return EMULATE_FAIL;
			break;
		default:
1595
			return EMULATE_FAIL;
1596
		}
1597

1598
		emulated = kvmppc_handle_store(run, vcpu, val, bytes,
1599 1600 1601 1602 1603 1604
				is_default_endian);
		if (emulated != EMULATE_DONE)
			break;

		vcpu->arch.paddr_accessed += run->mmio.len;
		vcpu->arch.mmio_vmx_copy_nums--;
1605
		vcpu->arch.mmio_vmx_offset++;
1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619
	}

	return emulated;
}

static int kvmppc_emulate_mmio_vmx_loadstore(struct kvm_vcpu *vcpu,
		struct kvm_run *run)
{
	enum emulation_result emulated = EMULATE_FAIL;
	int r;

	vcpu->arch.paddr_accessed += run->mmio.len;

	if (!vcpu->mmio_is_write) {
1620 1621
		emulated = kvmppc_handle_vmx_load(run, vcpu,
				vcpu->arch.io_gpr, run->mmio.len, 1);
1622
	} else {
1623 1624
		emulated = kvmppc_handle_vmx_store(run, vcpu,
				vcpu->arch.io_gpr, run->mmio.len, 1);
1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645
	}

	switch (emulated) {
	case EMULATE_DO_MMIO:
		run->exit_reason = KVM_EXIT_MMIO;
		r = RESUME_HOST;
		break;
	case EMULATE_FAIL:
		pr_info("KVM: MMIO emulation failed (VMX repeat)\n");
		run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
		run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION;
		r = RESUME_HOST;
		break;
	default:
		r = RESUME_GUEST;
		break;
	}
	return r;
}
#endif /* CONFIG_ALTIVEC */

1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659
int kvm_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
{
	int r = 0;
	union kvmppc_one_reg val;
	int size;

	size = one_reg_size(reg->id);
	if (size > sizeof(val))
		return -EINVAL;

	r = kvmppc_get_one_reg(vcpu, reg->id, &val);
	if (r == -EINVAL) {
		r = 0;
		switch (reg->id) {
1660 1661 1662 1663 1664 1665
#ifdef CONFIG_ALTIVEC
		case KVM_REG_PPC_VR0 ... KVM_REG_PPC_VR31:
			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
				r = -ENXIO;
				break;
			}
1666
			val.vval = vcpu->arch.vr.vr[reg->id - KVM_REG_PPC_VR0];
1667 1668 1669 1670 1671 1672
			break;
		case KVM_REG_PPC_VSCR:
			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
				r = -ENXIO;
				break;
			}
1673
			val = get_reg_val(reg->id, vcpu->arch.vr.vscr.u[3]);
1674 1675
			break;
		case KVM_REG_PPC_VRSAVE:
1676
			val = get_reg_val(reg->id, vcpu->arch.vrsave);
1677 1678
			break;
#endif /* CONFIG_ALTIVEC */
1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710
		default:
			r = -EINVAL;
			break;
		}
	}

	if (r)
		return r;

	if (copy_to_user((char __user *)(unsigned long)reg->addr, &val, size))
		r = -EFAULT;

	return r;
}

int kvm_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, struct kvm_one_reg *reg)
{
	int r;
	union kvmppc_one_reg val;
	int size;

	size = one_reg_size(reg->id);
	if (size > sizeof(val))
		return -EINVAL;

	if (copy_from_user(&val, (char __user *)(unsigned long)reg->addr, size))
		return -EFAULT;

	r = kvmppc_set_one_reg(vcpu, reg->id, &val);
	if (r == -EINVAL) {
		r = 0;
		switch (reg->id) {
1711 1712 1713 1714 1715 1716
#ifdef CONFIG_ALTIVEC
		case KVM_REG_PPC_VR0 ... KVM_REG_PPC_VR31:
			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
				r = -ENXIO;
				break;
			}
1717
			vcpu->arch.vr.vr[reg->id - KVM_REG_PPC_VR0] = val.vval;
1718 1719 1720 1721 1722 1723
			break;
		case KVM_REG_PPC_VSCR:
			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
				r = -ENXIO;
				break;
			}
1724
			vcpu->arch.vr.vscr.u[3] = set_reg_val(reg->id, val);
1725 1726
			break;
		case KVM_REG_PPC_VRSAVE:
1727 1728 1729 1730 1731
			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
				r = -ENXIO;
				break;
			}
			vcpu->arch.vrsave = set_reg_val(reg->id, val);
1732 1733
			break;
#endif /* CONFIG_ALTIVEC */
1734 1735 1736 1737 1738 1739 1740 1741 1742
		default:
			r = -EINVAL;
			break;
		}
	}

	return r;
}

1743 1744 1745 1746
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
{
	int r;

1747 1748
	vcpu_load(vcpu);

1749
	if (vcpu->mmio_needed) {
1750
		vcpu->mmio_needed = 0;
1751 1752
		if (!vcpu->mmio_is_write)
			kvmppc_complete_mmio_load(vcpu, run);
1753 1754 1755 1756 1757 1758 1759 1760 1761 1762
#ifdef CONFIG_VSX
		if (vcpu->arch.mmio_vsx_copy_nums > 0) {
			vcpu->arch.mmio_vsx_copy_nums--;
			vcpu->arch.mmio_vsx_offset++;
		}

		if (vcpu->arch.mmio_vsx_copy_nums > 0) {
			r = kvmppc_emulate_mmio_vsx_loadstore(vcpu, run);
			if (r == RESUME_HOST) {
				vcpu->mmio_needed = 1;
1763
				goto out;
1764 1765
			}
		}
1766 1767
#endif
#ifdef CONFIG_ALTIVEC
1768
		if (vcpu->arch.mmio_vmx_copy_nums > 0) {
1769
			vcpu->arch.mmio_vmx_copy_nums--;
1770 1771
			vcpu->arch.mmio_vmx_offset++;
		}
1772 1773 1774 1775 1776

		if (vcpu->arch.mmio_vmx_copy_nums > 0) {
			r = kvmppc_emulate_mmio_vmx_loadstore(vcpu, run);
			if (r == RESUME_HOST) {
				vcpu->mmio_needed = 1;
1777
				goto out;
1778 1779
			}
		}
1780
#endif
1781 1782 1783 1784 1785 1786 1787
	} else if (vcpu->arch.osi_needed) {
		u64 *gprs = run->osi.gprs;
		int i;

		for (i = 0; i < 32; i++)
			kvmppc_set_gpr(vcpu, i, gprs[i]);
		vcpu->arch.osi_needed = 0;
1788 1789 1790 1791 1792 1793 1794
	} else if (vcpu->arch.hcall_needed) {
		int i;

		kvmppc_set_gpr(vcpu, 3, run->papr_hcall.ret);
		for (i = 0; i < 9; ++i)
			kvmppc_set_gpr(vcpu, 4 + i, run->papr_hcall.args[i]);
		vcpu->arch.hcall_needed = 0;
1795 1796 1797 1798 1799
#ifdef CONFIG_BOOKE
	} else if (vcpu->arch.epr_needed) {
		kvmppc_set_epr(vcpu, run->epr.epr);
		vcpu->arch.epr_needed = 0;
#endif
1800 1801
	}

1802
	kvm_sigset_activate(vcpu);
1803

1804 1805 1806 1807
	if (run->immediate_exit)
		r = -EINTR;
	else
		r = kvmppc_vcpu_run(run, vcpu);
1808

1809
	kvm_sigset_deactivate(vcpu);
1810

1811
#ifdef CONFIG_ALTIVEC
1812
out:
1813
#endif
1814
	vcpu_put(vcpu);
1815 1816 1817 1818 1819
	return r;
}

int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq)
{
1820
	if (irq->irq == KVM_INTERRUPT_UNSET) {
1821
		kvmppc_core_dequeue_external(vcpu);
1822 1823 1824 1825
		return 0;
	}

	kvmppc_core_queue_external(vcpu, irq);
1826

1827
	kvm_vcpu_kick(vcpu);
1828

1829 1830 1831
	return 0;
}

1832 1833 1834 1835 1836 1837 1838 1839 1840
static int kvm_vcpu_ioctl_enable_cap(struct kvm_vcpu *vcpu,
				     struct kvm_enable_cap *cap)
{
	int r;

	if (cap->flags)
		return -EINVAL;

	switch (cap->cap) {
1841 1842 1843 1844
	case KVM_CAP_PPC_OSI:
		r = 0;
		vcpu->arch.osi_enabled = true;
		break;
1845 1846 1847 1848
	case KVM_CAP_PPC_PAPR:
		r = 0;
		vcpu->arch.papr_enabled = true;
		break;
1849 1850
	case KVM_CAP_PPC_EPR:
		r = 0;
1851 1852 1853 1854
		if (cap->args[0])
			vcpu->arch.epr_flags |= KVMPPC_EPR_USER;
		else
			vcpu->arch.epr_flags &= ~KVMPPC_EPR_USER;
1855
		break;
1856 1857 1858 1859 1860 1861
#ifdef CONFIG_BOOKE
	case KVM_CAP_PPC_BOOKE_WATCHDOG:
		r = 0;
		vcpu->arch.watchdog_enabled = true;
		break;
#endif
1862
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
1863 1864 1865 1866 1867 1868 1869 1870 1871 1872
	case KVM_CAP_SW_TLB: {
		struct kvm_config_tlb cfg;
		void __user *user_ptr = (void __user *)(uintptr_t)cap->args[0];

		r = -EFAULT;
		if (copy_from_user(&cfg, user_ptr, sizeof(cfg)))
			break;

		r = kvm_vcpu_ioctl_config_tlb(vcpu, &cfg);
		break;
S
Scott Wood 已提交
1873 1874 1875 1876
	}
#endif
#ifdef CONFIG_KVM_MPIC
	case KVM_CAP_IRQ_MPIC: {
A
Al Viro 已提交
1877
		struct fd f;
S
Scott Wood 已提交
1878 1879 1880
		struct kvm_device *dev;

		r = -EBADF;
A
Al Viro 已提交
1881 1882
		f = fdget(cap->args[0]);
		if (!f.file)
S
Scott Wood 已提交
1883 1884 1885
			break;

		r = -EPERM;
A
Al Viro 已提交
1886
		dev = kvm_device_from_filp(f.file);
S
Scott Wood 已提交
1887 1888 1889
		if (dev)
			r = kvmppc_mpic_connect_vcpu(dev, vcpu, cap->args[1]);

A
Al Viro 已提交
1890
		fdput(f);
S
Scott Wood 已提交
1891
		break;
S
Scott Wood 已提交
1892 1893
	}
#endif
1894 1895
#ifdef CONFIG_KVM_XICS
	case KVM_CAP_IRQ_XICS: {
A
Al Viro 已提交
1896
		struct fd f;
1897 1898 1899
		struct kvm_device *dev;

		r = -EBADF;
A
Al Viro 已提交
1900 1901
		f = fdget(cap->args[0]);
		if (!f.file)
1902 1903 1904
			break;

		r = -EPERM;
A
Al Viro 已提交
1905
		dev = kvm_device_from_filp(f.file);
1906 1907 1908 1909 1910 1911
		if (dev) {
			if (xive_enabled())
				r = kvmppc_xive_connect_vcpu(dev, vcpu, cap->args[1]);
			else
				r = kvmppc_xics_connect_vcpu(dev, vcpu, cap->args[1]);
		}
1912

A
Al Viro 已提交
1913
		fdput(f);
1914 1915 1916
		break;
	}
#endif /* CONFIG_KVM_XICS */
1917 1918 1919 1920 1921 1922 1923 1924 1925
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
	case KVM_CAP_PPC_FWNMI:
		r = -EINVAL;
		if (!is_kvmppc_hv_enabled(vcpu->kvm))
			break;
		r = 0;
		vcpu->kvm->arch.fwnmi_enabled = true;
		break;
#endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */
1926 1927 1928 1929 1930
	default:
		r = -EINVAL;
		break;
	}

1931 1932 1933
	if (!r)
		r = kvmppc_sanity_check(vcpu);

1934 1935 1936
	return r;
}

1937 1938 1939 1940 1941 1942 1943
bool kvm_arch_intc_initialized(struct kvm *kvm)
{
#ifdef CONFIG_KVM_MPIC
	if (kvm->arch.mpic)
		return true;
#endif
#ifdef CONFIG_KVM_XICS
1944
	if (kvm->arch.xics || kvm->arch.xive)
1945 1946 1947 1948 1949
		return true;
#endif
	return false;
}

1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961
int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
                                    struct kvm_mp_state *mp_state)
{
	return -EINVAL;
}

int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
                                    struct kvm_mp_state *mp_state)
{
	return -EINVAL;
}

1962 1963
long kvm_arch_vcpu_async_ioctl(struct file *filp,
			       unsigned int ioctl, unsigned long arg)
1964 1965 1966 1967
{
	struct kvm_vcpu *vcpu = filp->private_data;
	void __user *argp = (void __user *)arg;

1968
	if (ioctl == KVM_INTERRUPT) {
1969 1970
		struct kvm_interrupt irq;
		if (copy_from_user(&irq, argp, sizeof(irq)))
1971 1972
			return -EFAULT;
		return kvm_vcpu_ioctl_interrupt(vcpu, &irq);
1973
	}
1974 1975 1976 1977 1978 1979 1980 1981 1982
	return -ENOIOCTLCMD;
}

long kvm_arch_vcpu_ioctl(struct file *filp,
                         unsigned int ioctl, unsigned long arg)
{
	struct kvm_vcpu *vcpu = filp->private_data;
	void __user *argp = (void __user *)arg;
	long r;
1983

1984 1985 1986
	vcpu_load(vcpu);

	switch (ioctl) {
1987 1988 1989 1990 1991 1992 1993 1994 1995
	case KVM_ENABLE_CAP:
	{
		struct kvm_enable_cap cap;
		r = -EFAULT;
		if (copy_from_user(&cap, argp, sizeof(cap)))
			goto out;
		r = kvm_vcpu_ioctl_enable_cap(vcpu, &cap);
		break;
	}
S
Scott Wood 已提交
1996

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
	case KVM_SET_ONE_REG:
	case KVM_GET_ONE_REG:
	{
		struct kvm_one_reg reg;
		r = -EFAULT;
		if (copy_from_user(&reg, argp, sizeof(reg)))
			goto out;
		if (ioctl == KVM_SET_ONE_REG)
			r = kvm_vcpu_ioctl_set_one_reg(vcpu, &reg);
		else
			r = kvm_vcpu_ioctl_get_one_reg(vcpu, &reg);
		break;
	}

2011
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
2012 2013 2014 2015 2016 2017 2018 2019 2020
	case KVM_DIRTY_TLB: {
		struct kvm_dirty_tlb dirty;
		r = -EFAULT;
		if (copy_from_user(&dirty, argp, sizeof(dirty)))
			goto out;
		r = kvm_vcpu_ioctl_dirty_tlb(vcpu, &dirty);
		break;
	}
#endif
2021 2022 2023 2024 2025
	default:
		r = -EINVAL;
	}

out:
2026
	vcpu_put(vcpu);
2027 2028 2029
	return r;
}

2030 2031 2032 2033 2034
int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
{
	return VM_FAULT_SIGBUS;
}

2035 2036
static int kvm_vm_ioctl_get_pvinfo(struct kvm_ppc_pvinfo *pvinfo)
{
2037 2038 2039
	u32 inst_nop = 0x60000000;
#ifdef CONFIG_KVM_BOOKE_HV
	u32 inst_sc1 = 0x44000022;
2040 2041 2042 2043
	pvinfo->hcall[0] = cpu_to_be32(inst_sc1);
	pvinfo->hcall[1] = cpu_to_be32(inst_nop);
	pvinfo->hcall[2] = cpu_to_be32(inst_nop);
	pvinfo->hcall[3] = cpu_to_be32(inst_nop);
2044
#else
2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058
	u32 inst_lis = 0x3c000000;
	u32 inst_ori = 0x60000000;
	u32 inst_sc = 0x44000002;
	u32 inst_imm_mask = 0xffff;

	/*
	 * The hypercall to get into KVM from within guest context is as
	 * follows:
	 *
	 *    lis r0, r0, KVM_SC_MAGIC_R0@h
	 *    ori r0, KVM_SC_MAGIC_R0@l
	 *    sc
	 *    nop
	 */
2059 2060 2061 2062
	pvinfo->hcall[0] = cpu_to_be32(inst_lis | ((KVM_SC_MAGIC_R0 >> 16) & inst_imm_mask));
	pvinfo->hcall[1] = cpu_to_be32(inst_ori | (KVM_SC_MAGIC_R0 & inst_imm_mask));
	pvinfo->hcall[2] = cpu_to_be32(inst_sc);
	pvinfo->hcall[3] = cpu_to_be32(inst_nop);
2063
#endif
2064

2065 2066
	pvinfo->flags = KVM_PPC_PVINFO_FLAGS_EV_IDLE;

2067 2068 2069
	return 0;
}

2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081
int kvm_vm_ioctl_irq_line(struct kvm *kvm, struct kvm_irq_level *irq_event,
			  bool line_status)
{
	if (!irqchip_in_kernel(kvm))
		return -ENXIO;

	irq_event->status = kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID,
					irq_event->irq, irq_event->level,
					line_status);
	return 0;
}

2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099

static int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
				   struct kvm_enable_cap *cap)
{
	int r;

	if (cap->flags)
		return -EINVAL;

	switch (cap->cap) {
#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
	case KVM_CAP_PPC_ENABLE_HCALL: {
		unsigned long hcall = cap->args[0];

		r = -EINVAL;
		if (hcall > MAX_HCALL_OPCODE || (hcall & 3) ||
		    cap->args[1] > 1)
			break;
2100 2101
		if (!kvmppc_book3s_hcall_implemented(kvm, hcall))
			break;
2102 2103 2104 2105 2106 2107 2108
		if (cap->args[1])
			set_bit(hcall / 4, kvm->arch.enabled_hcalls);
		else
			clear_bit(hcall / 4, kvm->arch.enabled_hcalls);
		r = 0;
		break;
	}
2109 2110 2111 2112 2113 2114 2115 2116 2117
	case KVM_CAP_PPC_SMT: {
		unsigned long mode = cap->args[0];
		unsigned long flags = cap->args[1];

		r = -EINVAL;
		if (kvm->arch.kvm_ops->set_smt_mode)
			r = kvm->arch.kvm_ops->set_smt_mode(kvm, mode, flags);
		break;
	}
2118 2119 2120 2121 2122 2123 2124 2125 2126
#endif
	default:
		r = -EINVAL;
		break;
	}

	return r;
}

2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244
#ifdef CONFIG_PPC_BOOK3S_64
/*
 * These functions check whether the underlying hardware is safe
 * against attacks based on observing the effects of speculatively
 * executed instructions, and whether it supplies instructions for
 * use in workarounds.  The information comes from firmware, either
 * via the device tree on powernv platforms or from an hcall on
 * pseries platforms.
 */
#ifdef CONFIG_PPC_PSERIES
static int pseries_get_cpu_char(struct kvm_ppc_cpu_char *cp)
{
	struct h_cpu_char_result c;
	unsigned long rc;

	if (!machine_is(pseries))
		return -ENOTTY;

	rc = plpar_get_cpu_characteristics(&c);
	if (rc == H_SUCCESS) {
		cp->character = c.character;
		cp->behaviour = c.behaviour;
		cp->character_mask = KVM_PPC_CPU_CHAR_SPEC_BAR_ORI31 |
			KVM_PPC_CPU_CHAR_BCCTRL_SERIALISED |
			KVM_PPC_CPU_CHAR_L1D_FLUSH_ORI30 |
			KVM_PPC_CPU_CHAR_L1D_FLUSH_TRIG2 |
			KVM_PPC_CPU_CHAR_L1D_THREAD_PRIV |
			KVM_PPC_CPU_CHAR_BR_HINT_HONOURED |
			KVM_PPC_CPU_CHAR_MTTRIG_THR_RECONF |
			KVM_PPC_CPU_CHAR_COUNT_CACHE_DIS;
		cp->behaviour_mask = KVM_PPC_CPU_BEHAV_FAVOUR_SECURITY |
			KVM_PPC_CPU_BEHAV_L1D_FLUSH_PR |
			KVM_PPC_CPU_BEHAV_BNDS_CHK_SPEC_BAR;
	}
	return 0;
}
#else
static int pseries_get_cpu_char(struct kvm_ppc_cpu_char *cp)
{
	return -ENOTTY;
}
#endif

static inline bool have_fw_feat(struct device_node *fw_features,
				const char *state, const char *name)
{
	struct device_node *np;
	bool r = false;

	np = of_get_child_by_name(fw_features, name);
	if (np) {
		r = of_property_read_bool(np, state);
		of_node_put(np);
	}
	return r;
}

static int kvmppc_get_cpu_char(struct kvm_ppc_cpu_char *cp)
{
	struct device_node *np, *fw_features;
	int r;

	memset(cp, 0, sizeof(*cp));
	r = pseries_get_cpu_char(cp);
	if (r != -ENOTTY)
		return r;

	np = of_find_node_by_name(NULL, "ibm,opal");
	if (np) {
		fw_features = of_get_child_by_name(np, "fw-features");
		of_node_put(np);
		if (!fw_features)
			return 0;
		if (have_fw_feat(fw_features, "enabled",
				 "inst-spec-barrier-ori31,31,0"))
			cp->character |= KVM_PPC_CPU_CHAR_SPEC_BAR_ORI31;
		if (have_fw_feat(fw_features, "enabled",
				 "fw-bcctrl-serialized"))
			cp->character |= KVM_PPC_CPU_CHAR_BCCTRL_SERIALISED;
		if (have_fw_feat(fw_features, "enabled",
				 "inst-l1d-flush-ori30,30,0"))
			cp->character |= KVM_PPC_CPU_CHAR_L1D_FLUSH_ORI30;
		if (have_fw_feat(fw_features, "enabled",
				 "inst-l1d-flush-trig2"))
			cp->character |= KVM_PPC_CPU_CHAR_L1D_FLUSH_TRIG2;
		if (have_fw_feat(fw_features, "enabled",
				 "fw-l1d-thread-split"))
			cp->character |= KVM_PPC_CPU_CHAR_L1D_THREAD_PRIV;
		if (have_fw_feat(fw_features, "enabled",
				 "fw-count-cache-disabled"))
			cp->character |= KVM_PPC_CPU_CHAR_COUNT_CACHE_DIS;
		cp->character_mask = KVM_PPC_CPU_CHAR_SPEC_BAR_ORI31 |
			KVM_PPC_CPU_CHAR_BCCTRL_SERIALISED |
			KVM_PPC_CPU_CHAR_L1D_FLUSH_ORI30 |
			KVM_PPC_CPU_CHAR_L1D_FLUSH_TRIG2 |
			KVM_PPC_CPU_CHAR_L1D_THREAD_PRIV |
			KVM_PPC_CPU_CHAR_COUNT_CACHE_DIS;

		if (have_fw_feat(fw_features, "enabled",
				 "speculation-policy-favor-security"))
			cp->behaviour |= KVM_PPC_CPU_BEHAV_FAVOUR_SECURITY;
		if (!have_fw_feat(fw_features, "disabled",
				  "needs-l1d-flush-msr-pr-0-to-1"))
			cp->behaviour |= KVM_PPC_CPU_BEHAV_L1D_FLUSH_PR;
		if (!have_fw_feat(fw_features, "disabled",
				  "needs-spec-barrier-for-bound-checks"))
			cp->behaviour |= KVM_PPC_CPU_BEHAV_BNDS_CHK_SPEC_BAR;
		cp->behaviour_mask = KVM_PPC_CPU_BEHAV_FAVOUR_SECURITY |
			KVM_PPC_CPU_BEHAV_L1D_FLUSH_PR |
			KVM_PPC_CPU_BEHAV_BNDS_CHK_SPEC_BAR;

		of_node_put(fw_features);
	}

	return 0;
}
#endif

2245 2246 2247
long kvm_arch_vm_ioctl(struct file *filp,
                       unsigned int ioctl, unsigned long arg)
{
2248
	struct kvm *kvm __maybe_unused = filp->private_data;
2249
	void __user *argp = (void __user *)arg;
2250 2251 2252
	long r;

	switch (ioctl) {
2253 2254
	case KVM_PPC_GET_PVINFO: {
		struct kvm_ppc_pvinfo pvinfo;
2255
		memset(&pvinfo, 0, sizeof(pvinfo));
2256 2257 2258 2259 2260 2261 2262 2263
		r = kvm_vm_ioctl_get_pvinfo(&pvinfo);
		if (copy_to_user(argp, &pvinfo, sizeof(pvinfo))) {
			r = -EFAULT;
			goto out;
		}

		break;
	}
2264 2265 2266 2267 2268 2269 2270 2271 2272
	case KVM_ENABLE_CAP:
	{
		struct kvm_enable_cap cap;
		r = -EFAULT;
		if (copy_from_user(&cap, argp, sizeof(cap)))
			goto out;
		r = kvm_vm_ioctl_enable_cap(kvm, &cap);
		break;
	}
2273
#ifdef CONFIG_SPAPR_TCE_IOMMU
2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286
	case KVM_CREATE_SPAPR_TCE_64: {
		struct kvm_create_spapr_tce_64 create_tce_64;

		r = -EFAULT;
		if (copy_from_user(&create_tce_64, argp, sizeof(create_tce_64)))
			goto out;
		if (create_tce_64.flags) {
			r = -EINVAL;
			goto out;
		}
		r = kvm_vm_ioctl_create_spapr_tce(kvm, &create_tce_64);
		goto out;
	}
2287 2288
	case KVM_CREATE_SPAPR_TCE: {
		struct kvm_create_spapr_tce create_tce;
2289
		struct kvm_create_spapr_tce_64 create_tce_64;
2290 2291 2292 2293

		r = -EFAULT;
		if (copy_from_user(&create_tce, argp, sizeof(create_tce)))
			goto out;
2294 2295 2296 2297 2298 2299 2300 2301

		create_tce_64.liobn = create_tce.liobn;
		create_tce_64.page_shift = IOMMU_PAGE_SHIFT_4K;
		create_tce_64.offset = 0;
		create_tce_64.size = create_tce.window_size >>
				IOMMU_PAGE_SHIFT_4K;
		create_tce_64.flags = 0;
		r = kvm_vm_ioctl_create_spapr_tce(kvm, &create_tce_64);
2302 2303
		goto out;
	}
2304 2305
#endif
#ifdef CONFIG_PPC_BOOK3S_64
2306 2307
	case KVM_PPC_GET_SMMU_INFO: {
		struct kvm_ppc_smmu_info info;
2308
		struct kvm *kvm = filp->private_data;
2309 2310

		memset(&info, 0, sizeof(info));
2311
		r = kvm->arch.kvm_ops->get_smmu_info(kvm, &info);
2312 2313 2314 2315
		if (r >= 0 && copy_to_user(argp, &info, sizeof(info)))
			r = -EFAULT;
		break;
	}
2316 2317 2318 2319 2320 2321
	case KVM_PPC_RTAS_DEFINE_TOKEN: {
		struct kvm *kvm = filp->private_data;

		r = kvm_vm_ioctl_rtas_define_token(kvm, argp);
		break;
	}
2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346
	case KVM_PPC_CONFIGURE_V3_MMU: {
		struct kvm *kvm = filp->private_data;
		struct kvm_ppc_mmuv3_cfg cfg;

		r = -EINVAL;
		if (!kvm->arch.kvm_ops->configure_mmu)
			goto out;
		r = -EFAULT;
		if (copy_from_user(&cfg, argp, sizeof(cfg)))
			goto out;
		r = kvm->arch.kvm_ops->configure_mmu(kvm, &cfg);
		break;
	}
	case KVM_PPC_GET_RMMU_INFO: {
		struct kvm *kvm = filp->private_data;
		struct kvm_ppc_rmmu_info info;

		r = -EINVAL;
		if (!kvm->arch.kvm_ops->get_rmmu_info)
			goto out;
		r = kvm->arch.kvm_ops->get_rmmu_info(kvm, &info);
		if (r >= 0 && copy_to_user(argp, &info, sizeof(info)))
			r = -EFAULT;
		break;
	}
2347 2348 2349 2350 2351 2352 2353 2354
	case KVM_PPC_GET_CPU_CHAR: {
		struct kvm_ppc_cpu_char cpuchar;

		r = kvmppc_get_cpu_char(&cpuchar);
		if (r >= 0 && copy_to_user(argp, &cpuchar, sizeof(cpuchar)))
			r = -EFAULT;
		break;
	}
2355 2356 2357 2358
	default: {
		struct kvm *kvm = filp->private_data;
		r = kvm->arch.kvm_ops->arch_vm_ioctl(filp, ioctl, arg);
	}
2359
#else /* CONFIG_PPC_BOOK3S_64 */
2360
	default:
2361
		r = -ENOTTY;
2362
#endif
2363
	}
2364
out:
2365 2366 2367
	return r;
}

2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384
static unsigned long lpid_inuse[BITS_TO_LONGS(KVMPPC_NR_LPIDS)];
static unsigned long nr_lpids;

long kvmppc_alloc_lpid(void)
{
	long lpid;

	do {
		lpid = find_first_zero_bit(lpid_inuse, KVMPPC_NR_LPIDS);
		if (lpid >= nr_lpids) {
			pr_err("%s: No LPIDs free\n", __func__);
			return -ENOMEM;
		}
	} while (test_and_set_bit(lpid, lpid_inuse));

	return lpid;
}
2385
EXPORT_SYMBOL_GPL(kvmppc_alloc_lpid);
2386 2387 2388 2389 2390

void kvmppc_claim_lpid(long lpid)
{
	set_bit(lpid, lpid_inuse);
}
2391
EXPORT_SYMBOL_GPL(kvmppc_claim_lpid);
2392 2393 2394 2395 2396

void kvmppc_free_lpid(long lpid)
{
	clear_bit(lpid, lpid_inuse);
}
2397
EXPORT_SYMBOL_GPL(kvmppc_free_lpid);
2398 2399 2400 2401 2402 2403

void kvmppc_init_lpid(unsigned long nr_lpids_param)
{
	nr_lpids = min_t(unsigned long, KVMPPC_NR_LPIDS, nr_lpids_param);
	memset(lpid_inuse, 0, sizeof(lpid_inuse));
}
2404
EXPORT_SYMBOL_GPL(kvmppc_init_lpid);
2405

2406 2407 2408 2409 2410
int kvm_arch_init(void *opaque)
{
	return 0;
}

2411
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_ppc_instr);