powerpc.c 42.8 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 42
#include <asm/xive.h>

43
#include "timing.h"
44
#include "irq.h"
P
Paul Mackerras 已提交
45
#include "../mm/mmu_decl.h"
46

47 48 49
#define CREATE_TRACE_POINTS
#include "trace.h"

50 51 52 53 54
struct kvmppc_ops *kvmppc_hv_ops;
EXPORT_SYMBOL_GPL(kvmppc_hv_ops);
struct kvmppc_ops *kvmppc_pr_ops;
EXPORT_SYMBOL_GPL(kvmppc_pr_ops);

55

56 57
int kvm_arch_vcpu_runnable(struct kvm_vcpu *v)
{
R
Radim Krčmář 已提交
58
	return !!(v->arch.pending_exceptions) || kvm_request_pending(v);
59 60
}

61 62 63 64 65
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
	return 1;
}

66 67 68 69
/*
 * Common checks before entering the guest world.  Call with interrupts
 * disabled.
 *
70 71 72 73
 * returns:
 *
 * == 1 if we're ready to go into guest state
 * <= 0 if we need to go back to the host with return value
74 75 76
 */
int kvmppc_prepare_to_enter(struct kvm_vcpu *vcpu)
{
S
Scott Wood 已提交
77 78 79 80
	int r;

	WARN_ON(irqs_disabled());
	hard_irq_disable();
81 82 83 84 85

	while (true) {
		if (need_resched()) {
			local_irq_enable();
			cond_resched();
S
Scott Wood 已提交
86
			hard_irq_disable();
87 88 89 90
			continue;
		}

		if (signal_pending(current)) {
91 92 93
			kvmppc_account_exit(vcpu, SIGNAL_EXITS);
			vcpu->run->exit_reason = KVM_EXIT_INTR;
			r = -EINTR;
94 95 96
			break;
		}

97 98 99 100 101 102 103
		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).
104 105 106
		 * 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.
107
		 */
108
		smp_mb();
109

R
Radim Krčmář 已提交
110
		if (kvm_request_pending(vcpu)) {
111 112 113
			/* Make sure we process requests preemptable */
			local_irq_enable();
			trace_kvm_check_requests(vcpu);
114
			r = kvmppc_core_check_requests(vcpu);
S
Scott Wood 已提交
115
			hard_irq_disable();
116 117 118
			if (r > 0)
				continue;
			break;
119 120 121 122 123 124 125 126
		}

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

127
		guest_enter_irqoff();
S
Scott Wood 已提交
128
		return 1;
129 130
	}

S
Scott Wood 已提交
131 132
	/* return to host */
	local_irq_enable();
133 134
	return r;
}
135
EXPORT_SYMBOL_GPL(kvmppc_prepare_to_enter);
136

137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157
#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

158 159 160 161 162 163 164 165 166 167
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;

168
	if (!(kvmppc_get_msr(vcpu) & MSR_SF)) {
169 170 171 172 173 174 175 176
		/* 32 bit mode */
		param1 &= 0xffffffff;
		param2 &= 0xffffffff;
		param3 &= 0xffffffff;
		param4 &= 0xffffffff;
	}

	switch (nr) {
177
	case KVM_HCALL_TOKEN(KVM_HC_PPC_MAP_MAGIC_PAGE):
178
	{
179 180 181 182 183 184 185 186 187 188
#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

189 190 191 192 193 194 195 196 197 198 199 200
		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;
201

202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220
#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

221
		r2 = KVM_MAGIC_FEAT_SR | KVM_MAGIC_FEAT_MAS0_TO_SPRG7;
222

223
		r = EV_SUCCESS;
224 225
		break;
	}
226 227
	case KVM_HCALL_TOKEN(KVM_HC_FEATURES):
		r = EV_SUCCESS;
228
#if defined(CONFIG_PPC_BOOK3S) || defined(CONFIG_KVM_E500V2)
229 230
		r2 |= (1 << KVM_FEATURE_MAGIC_PAGE);
#endif
231 232 233

		/* Second return value is in r4 */
		break;
234 235 236
	case EV_HCALL_TOKEN(EV_IDLE):
		r = EV_SUCCESS;
		kvm_vcpu_block(vcpu);
237
		kvm_clear_request(KVM_REQ_UNHALT, vcpu);
238
		break;
239
	default:
240
		r = EV_UNIMPLEMENTED;
241 242 243
		break;
	}

244 245
	kvmppc_set_gpr(vcpu, 4, r2);

246 247
	return r;
}
248
EXPORT_SYMBOL_GPL(kvmppc_kvm_pv);
249

250 251 252 253 254 255 256 257 258 259 260 261 262
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 */
263
	if (!vcpu->arch.papr_enabled && is_kvmppc_hv_enabled(vcpu->kvm))
264 265
		goto out;

266 267 268 269 270
#ifdef CONFIG_KVM_BOOKE_HV
	if (!cpu_has_feature(CPU_FTR_EMB_HV))
		goto out;
#endif

271 272 273 274 275 276
	r = true;

out:
	vcpu->arch.sane = r;
	return r ? 0 : -EINVAL;
}
277
EXPORT_SYMBOL_GPL(kvmppc_sanity_check);
278

279 280 281 282 283
int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu)
{
	enum emulation_result er;
	int r;

284
	er = kvmppc_emulate_loadstore(vcpu);
285 286 287 288 289 290
	switch (er) {
	case EMULATE_DONE:
		/* Future optimization: only reload non-volatiles if they were
		 * actually modified. */
		r = RESUME_GUEST_NV;
		break;
291 292 293
	case EMULATE_AGAIN:
		r = RESUME_GUEST;
		break;
294 295 296 297 298 299 300 301 302
	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:
303 304 305
	{
		u32 last_inst;

306
		kvmppc_get_last_inst(vcpu, INST_GENERIC, &last_inst);
307
		/* XXX Deliver Program interrupt to guest. */
308
		pr_emerg("%s: emulation failed (%08x)\n", __func__, last_inst);
309 310
		r = RESUME_HOST;
		break;
311
	}
312
	default:
313 314
		WARN_ON(1);
		r = RESUME_GUEST;
315 316 317 318
	}

	return r;
}
319
EXPORT_SYMBOL_GPL(kvmppc_emulate_mmio);
320

321 322 323
int kvmppc_st(struct kvm_vcpu *vcpu, ulong *eaddr, int size, void *ptr,
	      bool data)
{
324
	ulong mp_pa = vcpu->arch.magic_page_pa & KVM_PAM & PAGE_MASK;
325 326 327 328 329 330 331 332 333 334 335 336 337 338 339
	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;

340 341 342 343 344 345 346 347 348 349
	/* 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;
	}

350 351 352 353 354 355 356 357 358 359
	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)
{
360
	ulong mp_pa = vcpu->arch.magic_page_pa & KVM_PAM & PAGE_MASK;
361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378
	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;

379 380 381 382 383 384 385 386 387 388
	/* 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;
	}

389 390
	if (kvm_read_guest(vcpu->kvm, pte.raddr, ptr, size))
		return EMULATE_DO_MMIO;
391 392 393 394 395

	return EMULATE_DONE;
}
EXPORT_SYMBOL_GPL(kvmppc_ld);

396
int kvm_arch_hardware_enable(void)
397
{
398
	return 0;
399 400 401 402 403 404 405 406 407
}

int kvm_arch_hardware_setup(void)
{
	return 0;
}

void kvm_arch_check_processor_compat(void *rtn)
{
408
	*(int *)rtn = kvmppc_core_check_processor_compat();
409 410
}

411
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
412
{
413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438
	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;
439
	return kvmppc_core_init_vm(kvm);
440 441
err_out:
	return -EINVAL;
442 443
}

444 445 446 447 448 449 450 451 452 453
bool kvm_arch_has_vcpu_debugfs(void)
{
	return false;
}

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

454
void kvm_arch_destroy_vm(struct kvm *kvm)
455 456
{
	unsigned int i;
457
	struct kvm_vcpu *vcpu;
458

459 460 461 462 463 464 465 466 467 468
#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

469 470 471 472 473 474 475 476
	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);
477 478 479

	kvmppc_core_destroy_vm(kvm);

480
	mutex_unlock(&kvm->lock);
481 482 483

	/* drop the module reference */
	module_put(kvm->arch.kvm_ops->owner);
484 485
}

486
int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
487 488
{
	int r;
489
	/* Assume we're using HV mode when the HV module is loaded */
490
	int hv_enabled = kvmppc_hv_ops ? 1 : 0;
491

492 493 494 495 496 497 498 499
	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);
	}

500
	switch (ext) {
S
Scott Wood 已提交
501 502
#ifdef CONFIG_BOOKE
	case KVM_CAP_PPC_BOOKE_SREGS:
503
	case KVM_CAP_PPC_BOOKE_WATCHDOG:
504
	case KVM_CAP_PPC_EPR:
S
Scott Wood 已提交
505
#else
506
	case KVM_CAP_PPC_SEGSTATE:
507
	case KVM_CAP_PPC_HIOR:
508
	case KVM_CAP_PPC_PAPR:
S
Scott Wood 已提交
509
#endif
510
	case KVM_CAP_PPC_UNSET_IRQ:
511
	case KVM_CAP_PPC_IRQ_LEVEL:
512
	case KVM_CAP_ENABLE_CAP:
513
	case KVM_CAP_ENABLE_CAP_VM:
514
	case KVM_CAP_ONE_REG:
A
Alexander Graf 已提交
515
	case KVM_CAP_IOEVENTFD:
516
	case KVM_CAP_DEVICE_CTRL:
517
	case KVM_CAP_IMMEDIATE_EXIT:
518 519 520
		r = 1;
		break;
	case KVM_CAP_PPC_PAIRED_SINGLES:
521
	case KVM_CAP_PPC_OSI:
522
	case KVM_CAP_PPC_GET_PVINFO:
523
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
524
	case KVM_CAP_SW_TLB:
S
Scott Wood 已提交
525
#endif
526
		/* We support this only for PR */
527
		r = !hv_enabled;
528
		break;
529 530 531 532 533 534
#ifdef CONFIG_KVM_MPIC
	case KVM_CAP_IRQ_MPIC:
		r = 1;
		break;
#endif

535
#ifdef CONFIG_PPC_BOOK3S_64
536
	case KVM_CAP_SPAPR_TCE:
537
	case KVM_CAP_SPAPR_TCE_64:
538 539
		/* fallthrough */
	case KVM_CAP_SPAPR_TCE_VFIO:
540
	case KVM_CAP_PPC_RTAS:
541
	case KVM_CAP_PPC_FIXUP_HCALL:
542
	case KVM_CAP_PPC_ENABLE_HCALL:
543 544 545
#ifdef CONFIG_KVM_XICS
	case KVM_CAP_IRQ_XICS:
#endif
546 547
		r = 1;
		break;
548 549 550 551

	case KVM_CAP_PPC_ALLOC_HTAB:
		r = hv_enabled;
		break;
552
#endif /* CONFIG_PPC_BOOK3S_64 */
553
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
554
	case KVM_CAP_PPC_SMT:
555
		r = 0;
556 557 558 559 560 561
		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) {
562 563 564 565 566
			if (cpu_has_feature(CPU_FTR_ARCH_300))
				r = 1;
			else
				r = threads_per_subcore;
		}
567
		break;
568 569 570 571 572 573 574 575 576 577
	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;
578
	case KVM_CAP_PPC_RMA:
579
		r = 0;
580
		break;
581 582 583
	case KVM_CAP_PPC_HWRNG:
		r = kvmppc_hwrng_present();
		break;
584
	case KVM_CAP_PPC_MMU_RADIX:
585
		r = !!(hv_enabled && radix_enabled());
586 587
		break;
	case KVM_CAP_PPC_MMU_HASH_V3:
588
		r = !!(hv_enabled && !radix_enabled() &&
589 590
		       cpu_has_feature(CPU_FTR_ARCH_300));
		break;
591
#endif
592
	case KVM_CAP_SYNC_MMU:
593
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
594
		r = hv_enabled;
595 596 597 598
#elif defined(KVM_ARCH_WANT_MMU_NOTIFIER)
		r = 1;
#else
		r = 0;
599
#endif
600 601
		break;
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
602
	case KVM_CAP_PPC_HTAB_FD:
603
		r = hv_enabled;
604
		break;
605
#endif
606 607 608 609 610 611 612
	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.
		 */
613
		if (hv_enabled)
614 615 616
			r = num_present_cpus();
		else
			r = num_online_cpus();
617
		break;
618 619 620
	case KVM_CAP_NR_MEMSLOTS:
		r = KVM_USER_MEM_SLOTS;
		break;
621 622 623
	case KVM_CAP_MAX_VCPUS:
		r = KVM_MAX_VCPUS;
		break;
624 625 626 627
#ifdef CONFIG_PPC_BOOK3S_64
	case KVM_CAP_PPC_GET_SMMU_INFO:
		r = 1;
		break;
628 629 630
	case KVM_CAP_SPAPR_MULTITCE:
		r = 1;
		break;
631
	case KVM_CAP_SPAPR_RESIZE_HPT:
632 633
		/* Disable this on POWER9 until code handles new HPTE format */
		r = !!hv_enabled && !cpu_has_feature(CPU_FTR_ARCH_300);
634
		break;
635 636 637 638 639
#endif
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
	case KVM_CAP_PPC_FWNMI:
		r = hv_enabled;
		break;
640
#endif
641 642 643 644
	case KVM_CAP_PPC_HTM:
		r = cpu_has_feature(CPU_FTR_TM_COMP) &&
		    is_kvmppc_hv_enabled(kvm);
		break;
645 646 647 648 649 650 651 652 653 654 655 656 657 658
	default:
		r = 0;
		break;
	}
	return r;

}

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

659
void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
660 661
			   struct kvm_memory_slot *dont)
{
662
	kvmppc_core_free_memslot(kvm, free, dont);
663 664
}

665 666
int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
			    unsigned long npages)
667
{
668
	return kvmppc_core_create_memslot(kvm, slot, npages);
669 670
}

671
int kvm_arch_prepare_memory_region(struct kvm *kvm,
672
				   struct kvm_memory_slot *memslot,
673
				   const struct kvm_userspace_memory_region *mem,
674
				   enum kvm_mr_change change)
675
{
676
	return kvmppc_core_prepare_memory_region(kvm, memslot, mem);
677 678
}

679
void kvm_arch_commit_memory_region(struct kvm *kvm,
680
				   const struct kvm_userspace_memory_region *mem,
681
				   const struct kvm_memory_slot *old,
682
				   const struct kvm_memory_slot *new,
683
				   enum kvm_mr_change change)
684
{
685
	kvmppc_core_commit_memory_region(kvm, mem, old, new);
686 687
}

688 689
void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
				   struct kvm_memory_slot *slot)
690
{
691
	kvmppc_core_flush_memslot(kvm, slot);
692 693
}

694 695
struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
{
696 697
	struct kvm_vcpu *vcpu;
	vcpu = kvmppc_core_vcpu_create(kvm, id);
698 699
	if (!IS_ERR(vcpu)) {
		vcpu->arch.wqp = &vcpu->wq;
700
		kvmppc_create_vcpu_debugfs(vcpu, id);
701
	}
702
	return vcpu;
703 704
}

705
void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
706 707 708
{
}

709 710
void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
{
711 712 713
	/* Make sure we're not using the vcpu anymore */
	hrtimer_cancel(&vcpu->arch.dec_timer);

714
	kvmppc_remove_vcpu_debugfs(vcpu);
S
Scott Wood 已提交
715 716 717 718 719

	switch (vcpu->arch.irq_type) {
	case KVMPPC_IRQ_MPIC:
		kvmppc_mpic_disconnect_vcpu(vcpu->arch.mpic, vcpu);
		break;
720
	case KVMPPC_IRQ_XICS:
721 722 723 724
		if (xive_enabled())
			kvmppc_xive_cleanup_vcpu(vcpu);
		else
			kvmppc_xics_free_icp(vcpu);
725
		break;
S
Scott Wood 已提交
726 727
	}

728
	kvmppc_core_vcpu_free(vcpu);
729 730 731 732 733 734 735 736 737
}

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

int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
{
738
	return kvmppc_core_pending_dec(vcpu);
739 740
}

T
Thomas Huth 已提交
741
static enum hrtimer_restart kvmppc_decrementer_wakeup(struct hrtimer *timer)
A
Alexander Graf 已提交
742 743 744 745
{
	struct kvm_vcpu *vcpu;

	vcpu = container_of(timer, struct kvm_vcpu, arch.dec_timer);
746
	kvmppc_decrementer_func(vcpu);
A
Alexander Graf 已提交
747 748 749 750

	return HRTIMER_NORESTART;
}

751 752
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
753 754
	int ret;

A
Alexander Graf 已提交
755 756
	hrtimer_init(&vcpu->arch.dec_timer, CLOCK_REALTIME, HRTIMER_MODE_ABS);
	vcpu->arch.dec_timer.function = kvmppc_decrementer_wakeup;
757
	vcpu->arch.dec_expires = ~(u64)0;
758

759 760 761
#ifdef CONFIG_KVM_EXIT_TIMING
	mutex_init(&vcpu->arch.exit_timing_lock);
#endif
762 763
	ret = kvmppc_subarch_vcpu_init(vcpu);
	return ret;
764 765 766 767
}

void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
{
768
	kvmppc_mmu_destroy(vcpu);
769
	kvmppc_subarch_vcpu_uninit(vcpu);
770 771 772 773
}

void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
774 775 776 777 778 779 780 781 782 783
#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
784
	kvmppc_core_vcpu_load(vcpu, cpu);
785 786 787 788
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
789
	kvmppc_core_vcpu_put(vcpu);
790 791 792
#ifdef CONFIG_BOOKE
	vcpu->arch.vrsave = mfspr(SPRN_VRSAVE);
#endif
793 794
}

795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830
/*
 * 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);
}

831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953
#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;
	}
}

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 */

#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 */

954 955 956
static void kvmppc_complete_mmio_load(struct kvm_vcpu *vcpu,
                                      struct kvm_run *run)
{
957
	u64 uninitialized_var(gpr);
958

959
	if (run->mmio.len > sizeof(gpr)) {
960 961 962 963
		printk(KERN_ERR "bad MMIO length: %d\n", run->mmio.len);
		return;
	}

964
	if (!vcpu->arch.mmio_host_swabbed) {
965
		switch (run->mmio.len) {
966
		case 8: gpr = *(u64 *)run->mmio.data; break;
967 968 969
		case 4: gpr = *(u32 *)run->mmio.data; break;
		case 2: gpr = *(u16 *)run->mmio.data; break;
		case 1: gpr = *(u8 *)run->mmio.data; break;
970 971 972
		}
	} else {
		switch (run->mmio.len) {
973 974 975
		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;
976
		case 1: gpr = *(u8 *)run->mmio.data; break;
977 978
		}
	}
979

980 981 982 983
	/* 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
984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999
	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;
		}
	}

1000 1001
	switch (vcpu->arch.io_gpr & KVM_MMIO_REG_EXT_MASK) {
	case KVM_MMIO_REG_GPR:
1002 1003
		kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
		break;
1004
	case KVM_MMIO_REG_FPR:
1005
		VCPU_FPR(vcpu, vcpu->arch.io_gpr & KVM_MMIO_REG_MASK) = gpr;
1006
		break;
1007
#ifdef CONFIG_PPC_BOOK3S
1008 1009
	case KVM_MMIO_REG_QPR:
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
1010
		break;
1011
	case KVM_MMIO_REG_FQPR:
1012
		VCPU_FPR(vcpu, vcpu->arch.io_gpr & KVM_MMIO_REG_MASK) = gpr;
1013
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
1014
		break;
1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025
#endif
#ifdef CONFIG_VSX
	case KVM_MMIO_REG_VSX:
		if (vcpu->arch.mmio_vsx_copy_type == KVMPPC_VSX_COPY_DWORD)
			kvmppc_set_vsr_dword(vcpu, gpr);
		else if (vcpu->arch.mmio_vsx_copy_type == KVMPPC_VSX_COPY_WORD)
			kvmppc_set_vsr_word(vcpu, gpr);
		else if (vcpu->arch.mmio_vsx_copy_type ==
				KVMPPC_VSX_COPY_DWORD_LOAD_DUMP)
			kvmppc_set_vsr_dword_dump(vcpu, gpr);
		break;
1026
#endif
1027 1028 1029
	default:
		BUG();
	}
1030 1031
}

1032 1033 1034
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)
1035
{
1036
	int idx, ret;
1037
	bool host_swabbed;
1038

1039
	/* Pity C doesn't have a logical XOR operator */
1040
	if (kvmppc_need_byteswap(vcpu)) {
1041
		host_swabbed = is_default_endian;
1042
	} else {
1043
		host_swabbed = !is_default_endian;
1044
	}
1045

1046 1047 1048 1049 1050 1051 1052 1053 1054 1055
	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;
1056
	vcpu->arch.mmio_host_swabbed = host_swabbed;
1057 1058
	vcpu->mmio_needed = 1;
	vcpu->mmio_is_write = 0;
1059
	vcpu->arch.mmio_sign_extend = sign_extend;
1060

1061 1062
	idx = srcu_read_lock(&vcpu->kvm->srcu);

1063
	ret = kvm_io_bus_read(vcpu, KVM_MMIO_BUS, run->mmio.phys_addr,
1064 1065 1066 1067 1068
			      bytes, &run->mmio.data);

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

	if (!ret) {
A
Alexander Graf 已提交
1069 1070 1071 1072 1073
		kvmppc_complete_mmio_load(vcpu, run);
		vcpu->mmio_needed = 0;
		return EMULATE_DONE;
	}

1074 1075
	return EMULATE_DO_MMIO;
}
1076 1077 1078 1079 1080 1081 1082

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);
}
1083
EXPORT_SYMBOL_GPL(kvmppc_handle_load);
1084

A
Alexander Graf 已提交
1085 1086
/* Same as above, but sign extends */
int kvmppc_handle_loads(struct kvm_run *run, struct kvm_vcpu *vcpu,
1087 1088
			unsigned int rt, unsigned int bytes,
			int is_default_endian)
A
Alexander Graf 已提交
1089
{
1090
	return __kvmppc_handle_load(run, vcpu, rt, bytes, is_default_endian, 1);
A
Alexander Graf 已提交
1091 1092
}

1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121
#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;

	/* Currently, mmio_vsx_copy_nums only allowed to be less than 4 */
	if ( (vcpu->arch.mmio_vsx_copy_nums > 4) ||
		(vcpu->arch.mmio_vsx_copy_nums < 0) ) {
		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 */

1122
int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
1123
			u64 val, unsigned int bytes, int is_default_endian)
1124 1125
{
	void *data = run->mmio.data;
1126
	int idx, ret;
1127
	bool host_swabbed;
1128

1129
	/* Pity C doesn't have a logical XOR operator */
1130
	if (kvmppc_need_byteswap(vcpu)) {
1131
		host_swabbed = is_default_endian;
1132
	} else {
1133
		host_swabbed = !is_default_endian;
1134
	}
1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146

	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;

1147 1148 1149
	if ((vcpu->arch.mmio_sp64_extend) && (bytes == 4))
		val = dp_to_sp(val);

1150
	/* Store the value at the lowest bytes in 'data'. */
1151
	if (!host_swabbed) {
1152
		switch (bytes) {
1153
		case 8: *(u64 *)data = val; break;
1154 1155 1156 1157 1158 1159
		case 4: *(u32 *)data = val; break;
		case 2: *(u16 *)data = val; break;
		case 1: *(u8  *)data = val; break;
		}
	} else {
		switch (bytes) {
1160 1161 1162 1163
		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;
1164 1165 1166
		}
	}

1167 1168
	idx = srcu_read_lock(&vcpu->kvm->srcu);

1169
	ret = kvm_io_bus_write(vcpu, KVM_MMIO_BUS, run->mmio.phys_addr,
1170 1171 1172 1173 1174
			       bytes, &run->mmio.data);

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

	if (!ret) {
A
Alexander Graf 已提交
1175 1176 1177 1178
		vcpu->mmio_needed = 0;
		return EMULATE_DONE;
	}

1179 1180
	return EMULATE_DO_MMIO;
}
1181
EXPORT_SYMBOL_GPL(kvmppc_handle_store);
1182

1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305
#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;
	int copy_type = vcpu->arch.mmio_vsx_copy_type;
	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;

	/* Currently, mmio_vsx_copy_nums only allowed to be less than 4 */
	if ( (vcpu->arch.mmio_vsx_copy_nums > 4) ||
		(vcpu->arch.mmio_vsx_copy_nums < 0) ) {
		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 */

1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319
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) {
1320 1321 1322 1323 1324 1325
#ifdef CONFIG_ALTIVEC
		case KVM_REG_PPC_VR0 ... KVM_REG_PPC_VR31:
			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
				r = -ENXIO;
				break;
			}
1326
			val.vval = vcpu->arch.vr.vr[reg->id - KVM_REG_PPC_VR0];
1327 1328 1329 1330 1331 1332
			break;
		case KVM_REG_PPC_VSCR:
			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
				r = -ENXIO;
				break;
			}
1333
			val = get_reg_val(reg->id, vcpu->arch.vr.vscr.u[3]);
1334 1335
			break;
		case KVM_REG_PPC_VRSAVE:
1336
			val = get_reg_val(reg->id, vcpu->arch.vrsave);
1337 1338
			break;
#endif /* CONFIG_ALTIVEC */
1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370
		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) {
1371 1372 1373 1374 1375 1376
#ifdef CONFIG_ALTIVEC
		case KVM_REG_PPC_VR0 ... KVM_REG_PPC_VR31:
			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
				r = -ENXIO;
				break;
			}
1377
			vcpu->arch.vr.vr[reg->id - KVM_REG_PPC_VR0] = val.vval;
1378 1379 1380 1381 1382 1383
			break;
		case KVM_REG_PPC_VSCR:
			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
				r = -ENXIO;
				break;
			}
1384
			vcpu->arch.vr.vscr.u[3] = set_reg_val(reg->id, val);
1385 1386
			break;
		case KVM_REG_PPC_VRSAVE:
1387 1388 1389 1390 1391
			if (!cpu_has_feature(CPU_FTR_ALTIVEC)) {
				r = -ENXIO;
				break;
			}
			vcpu->arch.vrsave = set_reg_val(reg->id, val);
1392 1393
			break;
#endif /* CONFIG_ALTIVEC */
1394 1395 1396 1397 1398 1399 1400 1401 1402
		default:
			r = -EINVAL;
			break;
		}
	}

	return r;
}

1403 1404 1405 1406 1407 1408
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
{
	int r;
	sigset_t sigsaved;

	if (vcpu->mmio_needed) {
1409
		vcpu->mmio_needed = 0;
1410 1411
		if (!vcpu->mmio_is_write)
			kvmppc_complete_mmio_load(vcpu, run);
1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425
#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;
				return r;
			}
		}
#endif
1426 1427 1428 1429 1430 1431 1432
	} 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;
1433 1434 1435 1436 1437 1438 1439
	} 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;
1440 1441 1442 1443 1444
#ifdef CONFIG_BOOKE
	} else if (vcpu->arch.epr_needed) {
		kvmppc_set_epr(vcpu, run->epr.epr);
		vcpu->arch.epr_needed = 0;
#endif
1445 1446
	}

1447 1448 1449
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);

1450 1451 1452 1453
	if (run->immediate_exit)
		r = -EINTR;
	else
		r = kvmppc_vcpu_run(run, vcpu);
1454 1455 1456 1457 1458 1459 1460 1461 1462

	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &sigsaved, NULL);

	return r;
}

int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq)
{
1463
	if (irq->irq == KVM_INTERRUPT_UNSET) {
1464
		kvmppc_core_dequeue_external(vcpu);
1465 1466 1467 1468
		return 0;
	}

	kvmppc_core_queue_external(vcpu, irq);
1469

1470
	kvm_vcpu_kick(vcpu);
1471

1472 1473 1474
	return 0;
}

1475 1476 1477 1478 1479 1480 1481 1482 1483
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) {
1484 1485 1486 1487
	case KVM_CAP_PPC_OSI:
		r = 0;
		vcpu->arch.osi_enabled = true;
		break;
1488 1489 1490 1491
	case KVM_CAP_PPC_PAPR:
		r = 0;
		vcpu->arch.papr_enabled = true;
		break;
1492 1493
	case KVM_CAP_PPC_EPR:
		r = 0;
1494 1495 1496 1497
		if (cap->args[0])
			vcpu->arch.epr_flags |= KVMPPC_EPR_USER;
		else
			vcpu->arch.epr_flags &= ~KVMPPC_EPR_USER;
1498
		break;
1499 1500 1501 1502 1503 1504
#ifdef CONFIG_BOOKE
	case KVM_CAP_PPC_BOOKE_WATCHDOG:
		r = 0;
		vcpu->arch.watchdog_enabled = true;
		break;
#endif
1505
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
1506 1507 1508 1509 1510 1511 1512 1513 1514 1515
	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
1516 1517 1518 1519
	}
#endif
#ifdef CONFIG_KVM_MPIC
	case KVM_CAP_IRQ_MPIC: {
A
Al Viro 已提交
1520
		struct fd f;
S
Scott Wood 已提交
1521 1522 1523
		struct kvm_device *dev;

		r = -EBADF;
A
Al Viro 已提交
1524 1525
		f = fdget(cap->args[0]);
		if (!f.file)
S
Scott Wood 已提交
1526 1527 1528
			break;

		r = -EPERM;
A
Al Viro 已提交
1529
		dev = kvm_device_from_filp(f.file);
S
Scott Wood 已提交
1530 1531 1532
		if (dev)
			r = kvmppc_mpic_connect_vcpu(dev, vcpu, cap->args[1]);

A
Al Viro 已提交
1533
		fdput(f);
S
Scott Wood 已提交
1534
		break;
S
Scott Wood 已提交
1535 1536
	}
#endif
1537 1538
#ifdef CONFIG_KVM_XICS
	case KVM_CAP_IRQ_XICS: {
A
Al Viro 已提交
1539
		struct fd f;
1540 1541 1542
		struct kvm_device *dev;

		r = -EBADF;
A
Al Viro 已提交
1543 1544
		f = fdget(cap->args[0]);
		if (!f.file)
1545 1546 1547
			break;

		r = -EPERM;
A
Al Viro 已提交
1548
		dev = kvm_device_from_filp(f.file);
1549 1550 1551 1552 1553 1554
		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]);
		}
1555

A
Al Viro 已提交
1556
		fdput(f);
1557 1558 1559
		break;
	}
#endif /* CONFIG_KVM_XICS */
1560 1561 1562 1563 1564 1565 1566 1567 1568
#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 */
1569 1570 1571 1572 1573
	default:
		r = -EINVAL;
		break;
	}

1574 1575 1576
	if (!r)
		r = kvmppc_sanity_check(vcpu);

1577 1578 1579
	return r;
}

1580 1581 1582 1583 1584 1585 1586
bool kvm_arch_intc_initialized(struct kvm *kvm)
{
#ifdef CONFIG_KVM_MPIC
	if (kvm->arch.mpic)
		return true;
#endif
#ifdef CONFIG_KVM_XICS
1587
	if (kvm->arch.xics || kvm->arch.xive)
1588 1589 1590 1591 1592
		return true;
#endif
	return false;
}

1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611
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;
}

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;

1612 1613
	switch (ioctl) {
	case KVM_INTERRUPT: {
1614 1615 1616
		struct kvm_interrupt irq;
		r = -EFAULT;
		if (copy_from_user(&irq, argp, sizeof(irq)))
1617
			goto out;
1618
		r = kvm_vcpu_ioctl_interrupt(vcpu, &irq);
1619
		goto out;
1620
	}
1621

1622 1623 1624 1625 1626 1627 1628 1629 1630
	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
1631

1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645
	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;
	}

1646
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
1647 1648 1649 1650 1651 1652 1653 1654 1655
	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
1656 1657 1658 1659 1660 1661 1662 1663
	default:
		r = -EINVAL;
	}

out:
	return r;
}

1664 1665 1666 1667 1668
int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
{
	return VM_FAULT_SIGBUS;
}

1669 1670
static int kvm_vm_ioctl_get_pvinfo(struct kvm_ppc_pvinfo *pvinfo)
{
1671 1672 1673
	u32 inst_nop = 0x60000000;
#ifdef CONFIG_KVM_BOOKE_HV
	u32 inst_sc1 = 0x44000022;
1674 1675 1676 1677
	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);
1678
#else
1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692
	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
	 */
1693 1694 1695 1696
	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);
1697
#endif
1698

1699 1700
	pvinfo->flags = KVM_PPC_PVINFO_FLAGS_EV_IDLE;

1701 1702 1703
	return 0;
}

1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715
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;
}

1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733

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;
1734 1735
		if (!kvmppc_book3s_hcall_implemented(kvm, hcall))
			break;
1736 1737 1738 1739 1740 1741 1742
		if (cap->args[1])
			set_bit(hcall / 4, kvm->arch.enabled_hcalls);
		else
			clear_bit(hcall / 4, kvm->arch.enabled_hcalls);
		r = 0;
		break;
	}
1743 1744 1745 1746 1747 1748 1749 1750 1751
	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;
	}
1752 1753 1754 1755 1756 1757 1758 1759 1760
#endif
	default:
		r = -EINVAL;
		break;
	}

	return r;
}

1761 1762 1763
long kvm_arch_vm_ioctl(struct file *filp,
                       unsigned int ioctl, unsigned long arg)
{
1764
	struct kvm *kvm __maybe_unused = filp->private_data;
1765
	void __user *argp = (void __user *)arg;
1766 1767 1768
	long r;

	switch (ioctl) {
1769 1770
	case KVM_PPC_GET_PVINFO: {
		struct kvm_ppc_pvinfo pvinfo;
1771
		memset(&pvinfo, 0, sizeof(pvinfo));
1772 1773 1774 1775 1776 1777 1778 1779
		r = kvm_vm_ioctl_get_pvinfo(&pvinfo);
		if (copy_to_user(argp, &pvinfo, sizeof(pvinfo))) {
			r = -EFAULT;
			goto out;
		}

		break;
	}
1780 1781 1782 1783 1784 1785 1786 1787 1788
	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;
	}
1789
#ifdef CONFIG_SPAPR_TCE_IOMMU
1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802
	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;
	}
1803 1804
	case KVM_CREATE_SPAPR_TCE: {
		struct kvm_create_spapr_tce create_tce;
1805
		struct kvm_create_spapr_tce_64 create_tce_64;
1806 1807 1808 1809

		r = -EFAULT;
		if (copy_from_user(&create_tce, argp, sizeof(create_tce)))
			goto out;
1810 1811 1812 1813 1814 1815 1816 1817

		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);
1818 1819
		goto out;
	}
1820 1821
#endif
#ifdef CONFIG_PPC_BOOK3S_64
1822 1823
	case KVM_PPC_GET_SMMU_INFO: {
		struct kvm_ppc_smmu_info info;
1824
		struct kvm *kvm = filp->private_data;
1825 1826

		memset(&info, 0, sizeof(info));
1827
		r = kvm->arch.kvm_ops->get_smmu_info(kvm, &info);
1828 1829 1830 1831
		if (r >= 0 && copy_to_user(argp, &info, sizeof(info)))
			r = -EFAULT;
		break;
	}
1832 1833 1834 1835 1836 1837
	case KVM_PPC_RTAS_DEFINE_TOKEN: {
		struct kvm *kvm = filp->private_data;

		r = kvm_vm_ioctl_rtas_define_token(kvm, argp);
		break;
	}
1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862
	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;
	}
1863 1864 1865 1866
	default: {
		struct kvm *kvm = filp->private_data;
		r = kvm->arch.kvm_ops->arch_vm_ioctl(filp, ioctl, arg);
	}
1867
#else /* CONFIG_PPC_BOOK3S_64 */
1868
	default:
1869
		r = -ENOTTY;
1870
#endif
1871
	}
1872
out:
1873 1874 1875
	return r;
}

1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892
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;
}
1893
EXPORT_SYMBOL_GPL(kvmppc_alloc_lpid);
1894 1895 1896 1897 1898

void kvmppc_claim_lpid(long lpid)
{
	set_bit(lpid, lpid_inuse);
}
1899
EXPORT_SYMBOL_GPL(kvmppc_claim_lpid);
1900 1901 1902 1903 1904

void kvmppc_free_lpid(long lpid)
{
	clear_bit(lpid, lpid_inuse);
}
1905
EXPORT_SYMBOL_GPL(kvmppc_free_lpid);
1906 1907 1908 1909 1910 1911

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));
}
1912
EXPORT_SYMBOL_GPL(kvmppc_init_lpid);
1913

1914 1915 1916 1917 1918
int kvm_arch_init(void *opaque)
{
	return 0;
}

1919
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_ppc_instr);