powerpc.c 21.4 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 <asm/irqflags.h>
34
#include "timing.h"
P
Paul Mackerras 已提交
35
#include "../mm/mmu_decl.h"
36

37 38 39
#define CREATE_TRACE_POINTS
#include "trace.h"

40 41
int kvm_arch_vcpu_runnable(struct kvm_vcpu *v)
{
42
	return !!(v->arch.pending_exceptions) ||
43
	       v->requests;
44 45
}

46 47 48 49 50
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
	return 1;
}

51 52 53 54 55
#ifndef CONFIG_KVM_BOOK3S_64_HV
/*
 * Common checks before entering the guest world.  Call with interrupts
 * disabled.
 *
56 57 58 59
 * returns:
 *
 * == 1 if we're ready to go into guest state
 * <= 0 if we need to go back to the host with return value
60 61 62
 */
int kvmppc_prepare_to_enter(struct kvm_vcpu *vcpu)
{
63
	int r = 1;
64 65 66 67 68 69 70 71 72 73 74

	WARN_ON_ONCE(!irqs_disabled());
	while (true) {
		if (need_resched()) {
			local_irq_enable();
			cond_resched();
			local_irq_disable();
			continue;
		}

		if (signal_pending(current)) {
75 76 77
			kvmppc_account_exit(vcpu, SIGNAL_EXITS);
			vcpu->run->exit_reason = KVM_EXIT_INTR;
			r = -EINTR;
78 79 80 81 82 83 84 85
			break;
		}

		smp_mb();
		if (vcpu->requests) {
			/* Make sure we process requests preemptable */
			local_irq_enable();
			trace_kvm_check_requests(vcpu);
86
			r = kvmppc_core_check_requests(vcpu);
87
			local_irq_disable();
88 89 90
			if (r > 0)
				continue;
			break;
91 92 93 94 95 96 97 98
		}

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

99 100 101 102 103 104 105
#ifdef CONFIG_PPC64
		/* lazy EE magic */
		hard_irq_disable();
		if (lazy_irq_pending()) {
			/* Got an interrupt in between, try again */
			local_irq_enable();
			local_irq_disable();
106
			kvm_guest_exit();
107 108 109 110 111 112
			continue;
		}

		trace_hardirqs_on();
#endif

113 114
		kvm_guest_enter();

115 116 117 118 119 120 121 122 123 124 125
		/* Going into guest context! Yay! */
		vcpu->mode = IN_GUEST_MODE;
		smp_wmb();

		break;
	}

	return r;
}
#endif /* CONFIG_KVM_BOOK3S_64_HV */

126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
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) {
145
	case KVM_HCALL_TOKEN(KVM_HC_PPC_MAP_MAGIC_PAGE):
146 147 148 149
	{
		vcpu->arch.magic_page_pa = param1;
		vcpu->arch.magic_page_ea = param2;

150
		r2 = KVM_MAGIC_FEAT_SR | KVM_MAGIC_FEAT_MAS0_TO_SPRG7;
151

152
		r = EV_SUCCESS;
153 154
		break;
	}
155 156
	case KVM_HCALL_TOKEN(KVM_HC_FEATURES):
		r = EV_SUCCESS;
157
#if defined(CONFIG_PPC_BOOK3S) || defined(CONFIG_KVM_E500V2)
S
Scott Wood 已提交
158
		/* XXX Missing magic page on 44x */
159 160
		r2 |= (1 << KVM_FEATURE_MAGIC_PAGE);
#endif
161 162 163

		/* Second return value is in r4 */
		break;
164 165 166 167 168
	case EV_HCALL_TOKEN(EV_IDLE):
		r = EV_SUCCESS;
		kvm_vcpu_block(vcpu);
		clear_bit(KVM_REQ_UNHALT, &vcpu->requests);
		break;
169
	default:
170
		r = EV_UNIMPLEMENTED;
171 172 173
		break;
	}

174 175
	kvmppc_set_gpr(vcpu, 4, r2);

176 177
	return r;
}
178

179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196
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

197 198 199 200 201
#ifdef CONFIG_KVM_BOOKE_HV
	if (!cpu_has_feature(CPU_FTR_EMB_HV))
		goto out;
#endif

202 203 204 205 206 207 208
	r = true;

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

209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
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__,
232
		       kvmppc_get_last_inst(vcpu));
233 234 235 236 237 238 239 240 241
		r = RESUME_HOST;
		break;
	default:
		BUG();
	}

	return r;
}

242
int kvm_arch_hardware_enable(void *garbage)
243
{
244
	return 0;
245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261
}

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)
{
262
	*(int *)rtn = kvmppc_core_check_processor_compat();
263 264
}

265
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
266
{
267 268 269
	if (type)
		return -EINVAL;

270
	return kvmppc_core_init_vm(kvm);
271 272
}

273
void kvm_arch_destroy_vm(struct kvm *kvm)
274 275
{
	unsigned int i;
276
	struct kvm_vcpu *vcpu;
277

278 279 280 281 282 283 284 285
	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);
286 287 288

	kvmppc_core_destroy_vm(kvm);

289
	mutex_unlock(&kvm->lock);
290 291
}

292 293 294 295
void kvm_arch_sync_events(struct kvm *kvm)
{
}

296 297 298 299 300
int kvm_dev_ioctl_check_extension(long ext)
{
	int r;

	switch (ext) {
S
Scott Wood 已提交
301 302 303
#ifdef CONFIG_BOOKE
	case KVM_CAP_PPC_BOOKE_SREGS:
#else
304
	case KVM_CAP_PPC_SEGSTATE:
305
	case KVM_CAP_PPC_HIOR:
306
	case KVM_CAP_PPC_PAPR:
S
Scott Wood 已提交
307
#endif
308
	case KVM_CAP_PPC_UNSET_IRQ:
309
	case KVM_CAP_PPC_IRQ_LEVEL:
310
	case KVM_CAP_ENABLE_CAP:
311
	case KVM_CAP_ONE_REG:
312 313 314 315
		r = 1;
		break;
#ifndef CONFIG_KVM_BOOK3S_64_HV
	case KVM_CAP_PPC_PAIRED_SINGLES:
316
	case KVM_CAP_PPC_OSI:
317
	case KVM_CAP_PPC_GET_PVINFO:
318
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
319 320
	case KVM_CAP_SW_TLB:
#endif
321 322
		r = 1;
		break;
323 324 325
	case KVM_CAP_COALESCED_MMIO:
		r = KVM_COALESCED_MMIO_PAGE_OFFSET;
		break;
326
#endif
327
#ifdef CONFIG_PPC_BOOK3S_64
328
	case KVM_CAP_SPAPR_TCE:
329
	case KVM_CAP_PPC_ALLOC_HTAB:
330 331
		r = 1;
		break;
332 333
#endif /* CONFIG_PPC_BOOK3S_64 */
#ifdef CONFIG_KVM_BOOK3S_64_HV
334 335 336
	case KVM_CAP_PPC_SMT:
		r = threads_per_core;
		break;
337 338
	case KVM_CAP_PPC_RMA:
		r = 1;
339 340 341
		/* PPC970 requires an RMA */
		if (cpu_has_feature(CPU_FTR_ARCH_201))
			r = 2;
342
		break;
343
#endif
344
	case KVM_CAP_SYNC_MMU:
345
#ifdef CONFIG_KVM_BOOK3S_64_HV
346
		r = cpu_has_feature(CPU_FTR_ARCH_206) ? 1 : 0;
347 348 349 350
#elif defined(KVM_ARCH_WANT_MMU_NOTIFIER)
		r = 1;
#else
		r = 0;
351
#endif
352
		break;
353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368
	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;
369 370 371 372 373
#ifdef CONFIG_PPC_BOOK3S_64
	case KVM_CAP_PPC_GET_SMMU_INFO:
		r = 1;
		break;
#endif
374 375 376 377 378 379 380 381 382 383 384 385 386 387
	default:
		r = 0;
		break;
	}
	return r;

}

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

388 389 390
void kvm_arch_free_memslot(struct kvm_memory_slot *free,
			   struct kvm_memory_slot *dont)
{
391 392 393 394
	if (!dont || free->arch.rmap != dont->arch.rmap) {
		vfree(free->arch.rmap);
		free->arch.rmap = NULL;
	}
395 396 397 398
}

int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
{
399 400 401 402
	slot->arch.rmap = vzalloc(npages * sizeof(*slot->arch.rmap));
	if (!slot->arch.rmap)
		return -ENOMEM;

403 404 405
	return 0;
}

406 407 408 409 410
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)
411
{
412
	return kvmppc_core_prepare_memory_region(kvm, mem);
413 414
}

415 416 417 418 419
void kvm_arch_commit_memory_region(struct kvm *kvm,
               struct kvm_userspace_memory_region *mem,
               struct kvm_memory_slot old,
               int user_alloc)
{
420
	kvmppc_core_commit_memory_region(kvm, mem);
421 422
}

423 424 425
void kvm_arch_flush_shadow_all(struct kvm *kvm)
{
}
426

427 428
void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
				   struct kvm_memory_slot *slot)
429 430 431
{
}

432 433
struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
{
434 435
	struct kvm_vcpu *vcpu;
	vcpu = kvmppc_core_vcpu_create(kvm, id);
436 437
	if (!IS_ERR(vcpu)) {
		vcpu->arch.wqp = &vcpu->wq;
438
		kvmppc_create_vcpu_debugfs(vcpu, id);
439
	}
440
	return vcpu;
441 442 443 444
}

void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
{
445 446 447 448
	/* Make sure we're not using the vcpu anymore */
	hrtimer_cancel(&vcpu->arch.dec_timer);
	tasklet_kill(&vcpu->arch.tasklet);

449
	kvmppc_remove_vcpu_debugfs(vcpu);
450
	kvmppc_core_vcpu_free(vcpu);
451 452 453 454 455 456 457 458 459
}

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

int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
{
460
	return kvmppc_core_pending_dec(vcpu);
461 462
}

A
Alexander Graf 已提交
463 464 465 466 467 468 469 470 471 472 473 474 475 476
/*
 * 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;
}

477 478
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
A
Alexander Graf 已提交
479 480 481
	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;
482
	vcpu->arch.dec_expires = ~(u64)0;
483

484 485 486 487
#ifdef CONFIG_KVM_EXIT_TIMING
	mutex_init(&vcpu->arch.exit_timing_lock);
#endif

488 489 490 491 492
	return 0;
}

void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
{
493
	kvmppc_mmu_destroy(vcpu);
494 495 496 497
}

void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
498 499 500 501 502 503 504 505 506 507
#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
508
	kvmppc_core_vcpu_load(vcpu, cpu);
509
	vcpu->cpu = smp_processor_id();
510 511 512 513
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
514
	kvmppc_core_vcpu_put(vcpu);
515 516 517
#ifdef CONFIG_BOOKE
	vcpu->arch.vrsave = mfspr(SPRN_VRSAVE);
#endif
518
	vcpu->cpu = -1;
519 520
}

J
Jan Kiszka 已提交
521
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
522
                                        struct kvm_guest_debug *dbg)
523
{
524
	return -EINVAL;
525 526 527 528 529
}

static void kvmppc_complete_dcr_load(struct kvm_vcpu *vcpu,
                                     struct kvm_run *run)
{
530
	kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, run->dcr.data);
531 532 533 534 535
}

static void kvmppc_complete_mmio_load(struct kvm_vcpu *vcpu,
                                      struct kvm_run *run)
{
536
	u64 uninitialized_var(gpr);
537

538
	if (run->mmio.len > sizeof(gpr)) {
539 540 541 542 543 544
		printk(KERN_ERR "bad MMIO length: %d\n", run->mmio.len);
		return;
	}

	if (vcpu->arch.mmio_is_bigendian) {
		switch (run->mmio.len) {
545
		case 8: gpr = *(u64 *)run->mmio.data; break;
546 547 548
		case 4: gpr = *(u32 *)run->mmio.data; break;
		case 2: gpr = *(u16 *)run->mmio.data; break;
		case 1: gpr = *(u8 *)run->mmio.data; break;
549 550 551 552
		}
	} else {
		/* Convert BE data from userland back to LE. */
		switch (run->mmio.len) {
553 554 555
		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;
556 557
		}
	}
558

A
Alexander Graf 已提交
559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574
	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;
		}
	}

575
	kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
576

577 578
	switch (vcpu->arch.io_gpr & KVM_MMIO_REG_EXT_MASK) {
	case KVM_MMIO_REG_GPR:
579 580
		kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
		break;
581 582
	case KVM_MMIO_REG_FPR:
		vcpu->arch.fpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
583
		break;
584
#ifdef CONFIG_PPC_BOOK3S
585 586
	case KVM_MMIO_REG_QPR:
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
587
		break;
588 589 590
	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;
591
		break;
592
#endif
593 594 595
	default:
		BUG();
	}
596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613
}

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
614
	vcpu->arch.mmio_sign_extend = 0;
615 616 617 618

	return EMULATE_DO_MMIO;
}

A
Alexander Graf 已提交
619 620 621 622 623 624 625 626 627 628 629 630
/* 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;
}

631
int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
632
                        u64 val, unsigned int bytes, int is_bigendian)
633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649
{
	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) {
650
		case 8: *(u64 *)data = val; break;
651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682
		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;
683 684 685 686 687 688 689
	} 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;
690 691 692 693 694 695 696
	} 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;
697 698
	}

699
	r = kvmppc_vcpu_run(run, vcpu);
700 701 702 703 704 705 706 707 708

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

	return r;
}

int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq)
{
709
	if (irq->irq == KVM_INTERRUPT_UNSET) {
710
		kvmppc_core_dequeue_external(vcpu, irq);
711 712 713 714
		return 0;
	}

	kvmppc_core_queue_external(vcpu, irq);
715

716
	kvm_vcpu_kick(vcpu);
717

718 719 720
	return 0;
}

721 722 723 724 725 726 727 728 729
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) {
730 731 732 733
	case KVM_CAP_PPC_OSI:
		r = 0;
		vcpu->arch.osi_enabled = true;
		break;
734 735 736 737
	case KVM_CAP_PPC_PAPR:
		r = 0;
		vcpu->arch.papr_enabled = true;
		break;
738
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
739 740 741 742 743 744 745 746 747 748 749 750
	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
751 752 753 754 755
	default:
		r = -EINVAL;
		break;
	}

756 757 758
	if (!r)
		r = kvmppc_sanity_check(vcpu);

759 760 761
	return r;
}

762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780
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;

781 782
	switch (ioctl) {
	case KVM_INTERRUPT: {
783 784 785
		struct kvm_interrupt irq;
		r = -EFAULT;
		if (copy_from_user(&irq, argp, sizeof(irq)))
786
			goto out;
787
		r = kvm_vcpu_ioctl_interrupt(vcpu, &irq);
788
		goto out;
789
	}
790

791 792 793 794 795 796 797 798 799
	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
800

801 802 803 804 805 806 807 808 809 810 811 812 813 814
	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;
	}

815
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
816 817 818 819 820 821 822 823 824
	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
825 826 827 828 829 830 831 832
	default:
		r = -EINVAL;
	}

out:
	return r;
}

833 834 835 836 837
int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
{
	return VM_FAULT_SIGBUS;
}

838 839
static int kvm_vm_ioctl_get_pvinfo(struct kvm_ppc_pvinfo *pvinfo)
{
840 841 842 843 844 845 846 847
	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
848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865
	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;
866
#endif
867

868 869
	pvinfo->flags = KVM_PPC_PVINFO_FLAGS_EV_IDLE;

870 871 872
	return 0;
}

873 874 875
long kvm_arch_vm_ioctl(struct file *filp,
                       unsigned int ioctl, unsigned long arg)
{
876
	void __user *argp = (void __user *)arg;
877 878 879
	long r;

	switch (ioctl) {
880 881
	case KVM_PPC_GET_PVINFO: {
		struct kvm_ppc_pvinfo pvinfo;
882
		memset(&pvinfo, 0, sizeof(pvinfo));
883 884 885 886 887 888 889 890
		r = kvm_vm_ioctl_get_pvinfo(&pvinfo);
		if (copy_to_user(argp, &pvinfo, sizeof(pvinfo))) {
			r = -EFAULT;
			goto out;
		}

		break;
	}
891
#ifdef CONFIG_PPC_BOOK3S_64
892 893 894 895 896 897 898 899 900 901
	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;
	}
902
#endif /* CONFIG_PPC_BOOK3S_64 */
903

904
#ifdef CONFIG_KVM_BOOK3S_64_HV
905 906 907 908 909 910 911 912 913
	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;
	}
914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930

	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;
	}
931 932
#endif /* CONFIG_KVM_BOOK3S_64_HV */

933 934 935 936 937 938 939 940 941 942 943 944
#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 */
945
	default:
946
		r = -ENOTTY;
947 948
	}

949
out:
950 951 952
	return r;
}

953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986
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));
}

987 988 989 990 991 992 993 994
int kvm_arch_init(void *opaque)
{
	return 0;
}

void kvm_arch_exit(void)
{
}