powerpc.c 20.0 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/fs.h>
27
#include <linux/slab.h>
28 29 30
#include <asm/cputable.h>
#include <asm/uaccess.h>
#include <asm/kvm_ppc.h>
31
#include <asm/tlbflush.h>
32
#include <asm/cputhreads.h>
33
#include "timing.h"
P
Paul Mackerras 已提交
34
#include "../mm/mmu_decl.h"
35

36 37 38
#define CREATE_TRACE_POINTS
#include "trace.h"

39 40
int kvm_arch_vcpu_runnable(struct kvm_vcpu *v)
{
41
	return !!(v->arch.pending_exceptions) ||
42
	       v->requests;
43 44
}

45 46 47 48 49
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
	return 1;
}

50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68
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;

	if (!(vcpu->arch.shared->msr & MSR_SF)) {
		/* 32 bit mode */
		param1 &= 0xffffffff;
		param2 &= 0xffffffff;
		param3 &= 0xffffffff;
		param4 &= 0xffffffff;
	}

	switch (nr) {
69
	case KVM_HCALL_TOKEN(KVM_HC_PPC_MAP_MAGIC_PAGE):
70 71 72 73
	{
		vcpu->arch.magic_page_pa = param1;
		vcpu->arch.magic_page_ea = param2;

74
		r2 = KVM_MAGIC_FEAT_SR | KVM_MAGIC_FEAT_MAS0_TO_SPRG7;
75

76
		r = EV_SUCCESS;
77 78
		break;
	}
79 80
	case KVM_HCALL_TOKEN(KVM_HC_FEATURES):
		r = EV_SUCCESS;
81
#if defined(CONFIG_PPC_BOOK3S) || defined(CONFIG_KVM_E500V2)
S
Scott Wood 已提交
82
		/* XXX Missing magic page on 44x */
83 84
		r2 |= (1 << KVM_FEATURE_MAGIC_PAGE);
#endif
85 86 87

		/* Second return value is in r4 */
		break;
88 89 90 91 92
	case EV_HCALL_TOKEN(EV_IDLE):
		r = EV_SUCCESS;
		kvm_vcpu_block(vcpu);
		clear_bit(KVM_REQ_UNHALT, &vcpu->requests);
		break;
93
	default:
94
		r = EV_UNIMPLEMENTED;
95 96 97
		break;
	}

98 99
	kvmppc_set_gpr(vcpu, 4, r2);

100 101
	return r;
}
102

103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120
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;

#ifdef CONFIG_KVM_BOOK3S_64_HV
	/* HV KVM can only do PAPR mode for now */
	if (!vcpu->arch.papr_enabled)
		goto out;
#endif

121 122 123 124 125
#ifdef CONFIG_KVM_BOOKE_HV
	if (!cpu_has_feature(CPU_FTR_EMB_HV))
		goto out;
#endif

126 127 128 129 130 131 132
	r = true;

out:
	vcpu->arch.sane = r;
	return r ? 0 : -EINVAL;
}

133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155
int kvmppc_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu)
{
	enum emulation_result er;
	int r;

	er = kvmppc_emulate_instruction(run, vcpu);
	switch (er) {
	case EMULATE_DONE:
		/* Future optimization: only reload non-volatiles if they were
		 * actually modified. */
		r = RESUME_GUEST_NV;
		break;
	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:
		/* XXX Deliver Program interrupt to guest. */
		printk(KERN_EMERG "%s: emulation failed (%08x)\n", __func__,
156
		       kvmppc_get_last_inst(vcpu));
157 158 159 160 161 162 163 164 165
		r = RESUME_HOST;
		break;
	default:
		BUG();
	}

	return r;
}

166
int kvm_arch_hardware_enable(void *garbage)
167
{
168
	return 0;
169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185
}

void kvm_arch_hardware_disable(void *garbage)
{
}

int kvm_arch_hardware_setup(void)
{
	return 0;
}

void kvm_arch_hardware_unsetup(void)
{
}

void kvm_arch_check_processor_compat(void *rtn)
{
186
	*(int *)rtn = kvmppc_core_check_processor_compat();
187 188
}

189
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
190
{
191 192 193
	if (type)
		return -EINVAL;

194
	return kvmppc_core_init_vm(kvm);
195 196
}

197
void kvm_arch_destroy_vm(struct kvm *kvm)
198 199
{
	unsigned int i;
200
	struct kvm_vcpu *vcpu;
201

202 203 204 205 206 207 208 209
	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);
210 211 212

	kvmppc_core_destroy_vm(kvm);

213
	mutex_unlock(&kvm->lock);
214 215
}

216 217 218 219
void kvm_arch_sync_events(struct kvm *kvm)
{
}

220 221 222 223 224
int kvm_dev_ioctl_check_extension(long ext)
{
	int r;

	switch (ext) {
S
Scott Wood 已提交
225 226 227
#ifdef CONFIG_BOOKE
	case KVM_CAP_PPC_BOOKE_SREGS:
#else
228
	case KVM_CAP_PPC_SEGSTATE:
229
	case KVM_CAP_PPC_HIOR:
230
	case KVM_CAP_PPC_PAPR:
S
Scott Wood 已提交
231
#endif
232
	case KVM_CAP_PPC_UNSET_IRQ:
233
	case KVM_CAP_PPC_IRQ_LEVEL:
234
	case KVM_CAP_ENABLE_CAP:
235
	case KVM_CAP_ONE_REG:
236 237 238 239
		r = 1;
		break;
#ifndef CONFIG_KVM_BOOK3S_64_HV
	case KVM_CAP_PPC_PAIRED_SINGLES:
240
	case KVM_CAP_PPC_OSI:
241
	case KVM_CAP_PPC_GET_PVINFO:
242
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
243 244
	case KVM_CAP_SW_TLB:
#endif
245 246
		r = 1;
		break;
247 248 249
	case KVM_CAP_COALESCED_MMIO:
		r = KVM_COALESCED_MMIO_PAGE_OFFSET;
		break;
250
#endif
251
#ifdef CONFIG_PPC_BOOK3S_64
252
	case KVM_CAP_SPAPR_TCE:
253
	case KVM_CAP_PPC_ALLOC_HTAB:
254 255
		r = 1;
		break;
256 257
#endif /* CONFIG_PPC_BOOK3S_64 */
#ifdef CONFIG_KVM_BOOK3S_64_HV
258 259 260
	case KVM_CAP_PPC_SMT:
		r = threads_per_core;
		break;
261 262
	case KVM_CAP_PPC_RMA:
		r = 1;
263 264 265
		/* PPC970 requires an RMA */
		if (cpu_has_feature(CPU_FTR_ARCH_201))
			r = 2;
266
		break;
267
#endif
268
	case KVM_CAP_SYNC_MMU:
269
#ifdef CONFIG_KVM_BOOK3S_64_HV
270
		r = cpu_has_feature(CPU_FTR_ARCH_206) ? 1 : 0;
271 272 273 274
#elif defined(KVM_ARCH_WANT_MMU_NOTIFIER)
		r = 1;
#else
		r = 0;
275
#endif
276
		break;
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292
	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.
		 */
#ifdef CONFIG_KVM_BOOK3S_64_HV
		r = num_present_cpus();
#else
		r = num_online_cpus();
#endif
		break;
	case KVM_CAP_MAX_VCPUS:
		r = KVM_MAX_VCPUS;
		break;
293 294 295 296 297
#ifdef CONFIG_PPC_BOOK3S_64
	case KVM_CAP_PPC_GET_SMMU_INFO:
		r = 1;
		break;
#endif
298 299 300 301 302 303 304 305 306 307 308 309 310 311
	default:
		r = 0;
		break;
	}
	return r;

}

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

312 313 314
void kvm_arch_free_memslot(struct kvm_memory_slot *free,
			   struct kvm_memory_slot *dont)
{
315 316 317 318
	if (!dont || free->arch.rmap != dont->arch.rmap) {
		vfree(free->arch.rmap);
		free->arch.rmap = NULL;
	}
319 320 321 322
}

int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
{
323 324 325 326
	slot->arch.rmap = vzalloc(npages * sizeof(*slot->arch.rmap));
	if (!slot->arch.rmap)
		return -ENOMEM;

327 328 329
	return 0;
}

330 331 332 333 334
int kvm_arch_prepare_memory_region(struct kvm *kvm,
                                   struct kvm_memory_slot *memslot,
                                   struct kvm_memory_slot old,
                                   struct kvm_userspace_memory_region *mem,
                                   int user_alloc)
335
{
336
	return kvmppc_core_prepare_memory_region(kvm, mem);
337 338
}

339 340 341 342 343
void kvm_arch_commit_memory_region(struct kvm *kvm,
               struct kvm_userspace_memory_region *mem,
               struct kvm_memory_slot old,
               int user_alloc)
{
344
	kvmppc_core_commit_memory_region(kvm, mem);
345 346
}

347 348 349
void kvm_arch_flush_shadow_all(struct kvm *kvm)
{
}
350

351 352
void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
				   struct kvm_memory_slot *slot)
353 354 355
{
}

356 357
struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
{
358 359
	struct kvm_vcpu *vcpu;
	vcpu = kvmppc_core_vcpu_create(kvm, id);
360 361
	if (!IS_ERR(vcpu)) {
		vcpu->arch.wqp = &vcpu->wq;
362
		kvmppc_create_vcpu_debugfs(vcpu, id);
363
	}
364
	return vcpu;
365 366 367 368
}

void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
{
369 370 371 372
	/* Make sure we're not using the vcpu anymore */
	hrtimer_cancel(&vcpu->arch.dec_timer);
	tasklet_kill(&vcpu->arch.tasklet);

373
	kvmppc_remove_vcpu_debugfs(vcpu);
374
	kvmppc_core_vcpu_free(vcpu);
375 376 377 378 379 380 381 382 383
}

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

int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
{
384
	return kvmppc_core_pending_dec(vcpu);
385 386
}

A
Alexander Graf 已提交
387 388 389 390 391 392 393 394 395 396 397 398 399 400
/*
 * low level hrtimer wake routine. Because this runs in hardirq context
 * we schedule a tasklet to do the real work.
 */
enum hrtimer_restart kvmppc_decrementer_wakeup(struct hrtimer *timer)
{
	struct kvm_vcpu *vcpu;

	vcpu = container_of(timer, struct kvm_vcpu, arch.dec_timer);
	tasklet_schedule(&vcpu->arch.tasklet);

	return HRTIMER_NORESTART;
}

401 402
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
A
Alexander Graf 已提交
403 404 405
	hrtimer_init(&vcpu->arch.dec_timer, CLOCK_REALTIME, HRTIMER_MODE_ABS);
	tasklet_init(&vcpu->arch.tasklet, kvmppc_decrementer_func, (ulong)vcpu);
	vcpu->arch.dec_timer.function = kvmppc_decrementer_wakeup;
406
	vcpu->arch.dec_expires = ~(u64)0;
407

408 409 410 411
#ifdef CONFIG_KVM_EXIT_TIMING
	mutex_init(&vcpu->arch.exit_timing_lock);
#endif

412 413 414 415 416
	return 0;
}

void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
{
417
	kvmppc_mmu_destroy(vcpu);
418 419 420 421
}

void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
422 423 424 425 426 427 428 429 430 431
#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
432
	kvmppc_core_vcpu_load(vcpu, cpu);
433
	vcpu->cpu = smp_processor_id();
434 435 436 437
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
438
	kvmppc_core_vcpu_put(vcpu);
439 440 441
#ifdef CONFIG_BOOKE
	vcpu->arch.vrsave = mfspr(SPRN_VRSAVE);
#endif
442
	vcpu->cpu = -1;
443 444
}

J
Jan Kiszka 已提交
445
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
446
                                        struct kvm_guest_debug *dbg)
447
{
448
	return -EINVAL;
449 450 451 452 453
}

static void kvmppc_complete_dcr_load(struct kvm_vcpu *vcpu,
                                     struct kvm_run *run)
{
454
	kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, run->dcr.data);
455 456 457 458 459
}

static void kvmppc_complete_mmio_load(struct kvm_vcpu *vcpu,
                                      struct kvm_run *run)
{
460
	u64 uninitialized_var(gpr);
461

462
	if (run->mmio.len > sizeof(gpr)) {
463 464 465 466 467 468
		printk(KERN_ERR "bad MMIO length: %d\n", run->mmio.len);
		return;
	}

	if (vcpu->arch.mmio_is_bigendian) {
		switch (run->mmio.len) {
469
		case 8: gpr = *(u64 *)run->mmio.data; break;
470 471 472
		case 4: gpr = *(u32 *)run->mmio.data; break;
		case 2: gpr = *(u16 *)run->mmio.data; break;
		case 1: gpr = *(u8 *)run->mmio.data; break;
473 474 475 476
		}
	} else {
		/* Convert BE data from userland back to LE. */
		switch (run->mmio.len) {
477 478 479
		case 4: gpr = ld_le32((u32 *)run->mmio.data); break;
		case 2: gpr = ld_le16((u16 *)run->mmio.data); break;
		case 1: gpr = *(u8 *)run->mmio.data; break;
480 481
		}
	}
482

A
Alexander Graf 已提交
483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498
	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;
		}
	}

499
	kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
500

501 502
	switch (vcpu->arch.io_gpr & KVM_MMIO_REG_EXT_MASK) {
	case KVM_MMIO_REG_GPR:
503 504
		kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
		break;
505 506
	case KVM_MMIO_REG_FPR:
		vcpu->arch.fpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
507
		break;
508
#ifdef CONFIG_PPC_BOOK3S
509 510
	case KVM_MMIO_REG_QPR:
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
511
		break;
512 513 514
	case KVM_MMIO_REG_FQPR:
		vcpu->arch.fpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
515
		break;
516
#endif
517 518 519
	default:
		BUG();
	}
520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537
}

int kvmppc_handle_load(struct kvm_run *run, struct kvm_vcpu *vcpu,
                       unsigned int rt, unsigned int bytes, int is_bigendian)
{
	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;
	vcpu->arch.mmio_is_bigendian = is_bigendian;
	vcpu->mmio_needed = 1;
	vcpu->mmio_is_write = 0;
A
Alexander Graf 已提交
538
	vcpu->arch.mmio_sign_extend = 0;
539 540 541 542

	return EMULATE_DO_MMIO;
}

A
Alexander Graf 已提交
543 544 545 546 547 548 549 550 551 552 553 554
/* Same as above, but sign extends */
int kvmppc_handle_loads(struct kvm_run *run, struct kvm_vcpu *vcpu,
                        unsigned int rt, unsigned int bytes, int is_bigendian)
{
	int r;

	r = kvmppc_handle_load(run, vcpu, rt, bytes, is_bigendian);
	vcpu->arch.mmio_sign_extend = 1;

	return r;
}

555
int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
556
                        u64 val, unsigned int bytes, int is_bigendian)
557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573
{
	void *data = run->mmio.data;

	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;

	/* Store the value at the lowest bytes in 'data'. */
	if (is_bigendian) {
		switch (bytes) {
574
		case 8: *(u64 *)data = val; break;
575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606
		case 4: *(u32 *)data = val; break;
		case 2: *(u16 *)data = val; break;
		case 1: *(u8  *)data = val; break;
		}
	} else {
		/* Store LE value into 'data'. */
		switch (bytes) {
		case 4: st_le32(data, val); break;
		case 2: st_le16(data, val); break;
		case 1: *(u8 *)data = val; break;
		}
	}

	return EMULATE_DO_MMIO;
}

int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
{
	int r;
	sigset_t sigsaved;

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

	if (vcpu->mmio_needed) {
		if (!vcpu->mmio_is_write)
			kvmppc_complete_mmio_load(vcpu, run);
		vcpu->mmio_needed = 0;
	} else if (vcpu->arch.dcr_needed) {
		if (!vcpu->arch.dcr_is_write)
			kvmppc_complete_dcr_load(vcpu, run);
		vcpu->arch.dcr_needed = 0;
607 608 609 610 611 612 613
	} 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;
614 615 616 617 618 619 620
	} 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;
621 622
	}

623
	r = kvmppc_vcpu_run(run, vcpu);
624 625 626 627 628 629 630 631 632

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

	return r;
}

int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq)
{
633
	if (irq->irq == KVM_INTERRUPT_UNSET) {
634
		kvmppc_core_dequeue_external(vcpu, irq);
635 636 637 638
		return 0;
	}

	kvmppc_core_queue_external(vcpu, irq);
639

640
	kvm_vcpu_kick(vcpu);
641

642 643 644
	return 0;
}

645 646 647 648 649 650 651 652 653
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) {
654 655 656 657
	case KVM_CAP_PPC_OSI:
		r = 0;
		vcpu->arch.osi_enabled = true;
		break;
658 659 660 661
	case KVM_CAP_PPC_PAPR:
		r = 0;
		vcpu->arch.papr_enabled = true;
		break;
662
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
663 664 665 666 667 668 669 670 671 672 673 674
	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;
	}
#endif
675 676 677 678 679
	default:
		r = -EINVAL;
		break;
	}

680 681 682
	if (!r)
		r = kvmppc_sanity_check(vcpu);

683 684 685
	return r;
}

686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704
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;

705 706
	switch (ioctl) {
	case KVM_INTERRUPT: {
707 708 709
		struct kvm_interrupt irq;
		r = -EFAULT;
		if (copy_from_user(&irq, argp, sizeof(irq)))
710
			goto out;
711
		r = kvm_vcpu_ioctl_interrupt(vcpu, &irq);
712
		goto out;
713
	}
714

715 716 717 718 719 720 721 722 723
	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
724

725 726 727 728 729 730 731 732 733 734 735 736 737 738
	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;
	}

739
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
740 741 742 743 744 745 746 747 748
	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
749 750 751 752 753 754 755 756
	default:
		r = -EINVAL;
	}

out:
	return r;
}

757 758 759 760 761
int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
{
	return VM_FAULT_SIGBUS;
}

762 763
static int kvm_vm_ioctl_get_pvinfo(struct kvm_ppc_pvinfo *pvinfo)
{
764 765 766 767 768 769 770 771
	u32 inst_nop = 0x60000000;
#ifdef CONFIG_KVM_BOOKE_HV
	u32 inst_sc1 = 0x44000022;
	pvinfo->hcall[0] = inst_sc1;
	pvinfo->hcall[1] = inst_nop;
	pvinfo->hcall[2] = inst_nop;
	pvinfo->hcall[3] = inst_nop;
#else
772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789
	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
	 */
	pvinfo->hcall[0] = inst_lis | ((KVM_SC_MAGIC_R0 >> 16) & inst_imm_mask);
	pvinfo->hcall[1] = inst_ori | (KVM_SC_MAGIC_R0 & inst_imm_mask);
	pvinfo->hcall[2] = inst_sc;
	pvinfo->hcall[3] = inst_nop;
790
#endif
791

792 793
	pvinfo->flags = KVM_PPC_PVINFO_FLAGS_EV_IDLE;

794 795 796
	return 0;
}

797 798 799
long kvm_arch_vm_ioctl(struct file *filp,
                       unsigned int ioctl, unsigned long arg)
{
800
	void __user *argp = (void __user *)arg;
801 802 803
	long r;

	switch (ioctl) {
804 805
	case KVM_PPC_GET_PVINFO: {
		struct kvm_ppc_pvinfo pvinfo;
806
		memset(&pvinfo, 0, sizeof(pvinfo));
807 808 809 810 811 812 813 814
		r = kvm_vm_ioctl_get_pvinfo(&pvinfo);
		if (copy_to_user(argp, &pvinfo, sizeof(pvinfo))) {
			r = -EFAULT;
			goto out;
		}

		break;
	}
815
#ifdef CONFIG_PPC_BOOK3S_64
816 817 818 819 820 821 822 823 824 825
	case KVM_CREATE_SPAPR_TCE: {
		struct kvm_create_spapr_tce create_tce;
		struct kvm *kvm = filp->private_data;

		r = -EFAULT;
		if (copy_from_user(&create_tce, argp, sizeof(create_tce)))
			goto out;
		r = kvm_vm_ioctl_create_spapr_tce(kvm, &create_tce);
		goto out;
	}
826
#endif /* CONFIG_PPC_BOOK3S_64 */
827

828
#ifdef CONFIG_KVM_BOOK3S_64_HV
829 830 831 832 833 834 835 836 837
	case KVM_ALLOCATE_RMA: {
		struct kvm *kvm = filp->private_data;
		struct kvm_allocate_rma rma;

		r = kvm_vm_ioctl_allocate_rma(kvm, &rma);
		if (r >= 0 && copy_to_user(argp, &rma, sizeof(rma)))
			r = -EFAULT;
		break;
	}
838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854

	case KVM_PPC_ALLOCATE_HTAB: {
		struct kvm *kvm = filp->private_data;
		u32 htab_order;

		r = -EFAULT;
		if (get_user(htab_order, (u32 __user *)argp))
			break;
		r = kvmppc_alloc_reset_hpt(kvm, &htab_order);
		if (r)
			break;
		r = -EFAULT;
		if (put_user(htab_order, (u32 __user *)argp))
			break;
		r = 0;
		break;
	}
855 856
#endif /* CONFIG_KVM_BOOK3S_64_HV */

857 858 859 860 861 862 863 864 865 866 867 868
#ifdef CONFIG_PPC_BOOK3S_64
	case KVM_PPC_GET_SMMU_INFO: {
		struct kvm *kvm = filp->private_data;
		struct kvm_ppc_smmu_info info;

		memset(&info, 0, sizeof(info));
		r = kvm_vm_ioctl_get_smmu_info(kvm, &info);
		if (r >= 0 && copy_to_user(argp, &info, sizeof(info)))
			r = -EFAULT;
		break;
	}
#endif /* CONFIG_PPC_BOOK3S_64 */
869
	default:
870
		r = -ENOTTY;
871 872
	}

873
out:
874 875 876
	return r;
}

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
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;
}

void kvmppc_claim_lpid(long lpid)
{
	set_bit(lpid, lpid_inuse);
}

void kvmppc_free_lpid(long lpid)
{
	clear_bit(lpid, lpid_inuse);
}

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));
}

911 912 913 914 915 916 917 918
int kvm_arch_init(void *opaque)
{
	return 0;
}

void kvm_arch_exit(void)
{
}