powerpc.c 22.9 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>
S
Scott Wood 已提交
28
#include <linux/file.h>
29 30 31
#include <asm/cputable.h>
#include <asm/uaccess.h>
#include <asm/kvm_ppc.h>
32
#include <asm/tlbflush.h>
33
#include <asm/cputhreads.h>
34
#include <asm/irqflags.h>
35
#include "timing.h"
P
Paul Mackerras 已提交
36
#include "../mm/mmu_decl.h"
37

38 39 40
#define CREATE_TRACE_POINTS
#include "trace.h"

41 42
int kvm_arch_vcpu_runnable(struct kvm_vcpu *v)
{
43
	return !!(v->arch.pending_exceptions) ||
44
	       v->requests;
45 46
}

47 48 49 50 51
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
	return 1;
}

52 53 54 55 56
#ifndef CONFIG_KVM_BOOK3S_64_HV
/*
 * Common checks before entering the guest world.  Call with interrupts
 * disabled.
 *
57 58 59 60
 * returns:
 *
 * == 1 if we're ready to go into guest state
 * <= 0 if we need to go back to the host with return value
61 62 63
 */
int kvmppc_prepare_to_enter(struct kvm_vcpu *vcpu)
{
64
	int r = 1;
65 66 67 68 69 70 71 72 73 74 75

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

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

82 83 84 85 86 87 88 89
		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).
		 */
90
		smp_mb();
91

92 93 94 95
		if (vcpu->requests) {
			/* Make sure we process requests preemptable */
			local_irq_enable();
			trace_kvm_check_requests(vcpu);
96
			r = kvmppc_core_check_requests(vcpu);
97
			local_irq_disable();
98 99 100
			if (r > 0)
				continue;
			break;
101 102 103 104 105 106 107 108
		}

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

109 110 111 112 113 114 115
#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();
116
			kvm_guest_exit();
117 118 119 120 121 122
			continue;
		}

		trace_hardirqs_on();
#endif

123
		kvm_guest_enter();
124 125 126 127 128 129 130
		break;
	}

	return r;
}
#endif /* CONFIG_KVM_BOOK3S_64_HV */

131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149
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) {
150
	case KVM_HCALL_TOKEN(KVM_HC_PPC_MAP_MAGIC_PAGE):
151 152 153 154
	{
		vcpu->arch.magic_page_pa = param1;
		vcpu->arch.magic_page_ea = param2;

155
		r2 = KVM_MAGIC_FEAT_SR | KVM_MAGIC_FEAT_MAS0_TO_SPRG7;
156

157
		r = EV_SUCCESS;
158 159
		break;
	}
160 161
	case KVM_HCALL_TOKEN(KVM_HC_FEATURES):
		r = EV_SUCCESS;
162
#if defined(CONFIG_PPC_BOOK3S) || defined(CONFIG_KVM_E500V2)
S
Scott Wood 已提交
163
		/* XXX Missing magic page on 44x */
164 165
		r2 |= (1 << KVM_FEATURE_MAGIC_PAGE);
#endif
166 167 168

		/* Second return value is in r4 */
		break;
169 170 171 172 173
	case EV_HCALL_TOKEN(EV_IDLE):
		r = EV_SUCCESS;
		kvm_vcpu_block(vcpu);
		clear_bit(KVM_REQ_UNHALT, &vcpu->requests);
		break;
174
	default:
175
		r = EV_UNIMPLEMENTED;
176 177 178
		break;
	}

179 180
	kvmppc_set_gpr(vcpu, 4, r2);

181 182
	return r;
}
183

184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201
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

202 203 204 205 206
#ifdef CONFIG_KVM_BOOKE_HV
	if (!cpu_has_feature(CPU_FTR_EMB_HV))
		goto out;
#endif

207 208 209 210 211 212 213
	r = true;

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

214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236
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__,
237
		       kvmppc_get_last_inst(vcpu));
238 239 240
		r = RESUME_HOST;
		break;
	default:
241 242
		WARN_ON(1);
		r = RESUME_GUEST;
243 244 245 246 247
	}

	return r;
}

248
int kvm_arch_hardware_enable(void *garbage)
249
{
250
	return 0;
251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267
}

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)
{
268
	*(int *)rtn = kvmppc_core_check_processor_compat();
269 270
}

271
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
272
{
273 274 275
	if (type)
		return -EINVAL;

276
	return kvmppc_core_init_vm(kvm);
277 278
}

279
void kvm_arch_destroy_vm(struct kvm *kvm)
280 281
{
	unsigned int i;
282
	struct kvm_vcpu *vcpu;
283

284 285 286 287 288 289 290 291
	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);
292 293 294

	kvmppc_core_destroy_vm(kvm);

295
	mutex_unlock(&kvm->lock);
296 297
}

298 299 300 301
void kvm_arch_sync_events(struct kvm *kvm)
{
}

302 303 304 305 306
int kvm_dev_ioctl_check_extension(long ext)
{
	int r;

	switch (ext) {
S
Scott Wood 已提交
307 308
#ifdef CONFIG_BOOKE
	case KVM_CAP_PPC_BOOKE_SREGS:
309
	case KVM_CAP_PPC_BOOKE_WATCHDOG:
310
	case KVM_CAP_PPC_EPR:
S
Scott Wood 已提交
311
#else
312
	case KVM_CAP_PPC_SEGSTATE:
313
	case KVM_CAP_PPC_HIOR:
314
	case KVM_CAP_PPC_PAPR:
S
Scott Wood 已提交
315
#endif
316
	case KVM_CAP_PPC_UNSET_IRQ:
317
	case KVM_CAP_PPC_IRQ_LEVEL:
318
	case KVM_CAP_ENABLE_CAP:
319
	case KVM_CAP_ONE_REG:
A
Alexander Graf 已提交
320
	case KVM_CAP_IOEVENTFD:
321
	case KVM_CAP_DEVICE_CTRL:
322 323 324 325
		r = 1;
		break;
#ifndef CONFIG_KVM_BOOK3S_64_HV
	case KVM_CAP_PPC_PAIRED_SINGLES:
326
	case KVM_CAP_PPC_OSI:
327
	case KVM_CAP_PPC_GET_PVINFO:
328
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
329
	case KVM_CAP_SW_TLB:
S
Scott Wood 已提交
330 331 332
#endif
#ifdef CONFIG_KVM_MPIC
	case KVM_CAP_IRQ_MPIC:
S
Scott Wood 已提交
333
#endif
334 335
		r = 1;
		break;
336 337 338
	case KVM_CAP_COALESCED_MMIO:
		r = KVM_COALESCED_MMIO_PAGE_OFFSET;
		break;
339
#endif
340
#ifdef CONFIG_PPC_BOOK3S_64
341
	case KVM_CAP_SPAPR_TCE:
342
	case KVM_CAP_PPC_ALLOC_HTAB:
343 344
		r = 1;
		break;
345 346
#endif /* CONFIG_PPC_BOOK3S_64 */
#ifdef CONFIG_KVM_BOOK3S_64_HV
347 348 349
	case KVM_CAP_PPC_SMT:
		r = threads_per_core;
		break;
350 351
	case KVM_CAP_PPC_RMA:
		r = 1;
352 353 354
		/* PPC970 requires an RMA */
		if (cpu_has_feature(CPU_FTR_ARCH_201))
			r = 2;
355
		break;
356
#endif
357
	case KVM_CAP_SYNC_MMU:
358
#ifdef CONFIG_KVM_BOOK3S_64_HV
359
		r = cpu_has_feature(CPU_FTR_ARCH_206) ? 1 : 0;
360 361 362 363
#elif defined(KVM_ARCH_WANT_MMU_NOTIFIER)
		r = 1;
#else
		r = 0;
364 365 366 367 368 369
		break;
#endif
#ifdef CONFIG_KVM_BOOK3S_64_HV
	case KVM_CAP_PPC_HTAB_FD:
		r = 1;
		break;
370
#endif
371
		break;
372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387
	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;
388 389 390 391 392
#ifdef CONFIG_PPC_BOOK3S_64
	case KVM_CAP_PPC_GET_SMMU_INFO:
		r = 1;
		break;
#endif
393 394 395 396 397 398 399 400 401 402 403 404 405 406
	default:
		r = 0;
		break;
	}
	return r;

}

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

407 408 409
void kvm_arch_free_memslot(struct kvm_memory_slot *free,
			   struct kvm_memory_slot *dont)
{
410
	kvmppc_core_free_memslot(free, dont);
411 412 413 414
}

int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
{
415
	return kvmppc_core_create_memslot(slot, npages);
416 417
}

418
int kvm_arch_prepare_memory_region(struct kvm *kvm,
419
				   struct kvm_memory_slot *memslot,
420 421
				   struct kvm_userspace_memory_region *mem,
				   enum kvm_mr_change change)
422
{
423
	return kvmppc_core_prepare_memory_region(kvm, memslot, mem);
424 425
}

426
void kvm_arch_commit_memory_region(struct kvm *kvm,
427
				   struct kvm_userspace_memory_region *mem,
428 429
				   const struct kvm_memory_slot *old,
				   enum kvm_mr_change change)
430
{
431
	kvmppc_core_commit_memory_region(kvm, mem, old);
432 433
}

434 435 436
void kvm_arch_flush_shadow_all(struct kvm *kvm)
{
}
437

438 439
void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
				   struct kvm_memory_slot *slot)
440
{
441
	kvmppc_core_flush_memslot(kvm, slot);
442 443
}

444 445
struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
{
446 447
	struct kvm_vcpu *vcpu;
	vcpu = kvmppc_core_vcpu_create(kvm, id);
448 449
	if (!IS_ERR(vcpu)) {
		vcpu->arch.wqp = &vcpu->wq;
450
		kvmppc_create_vcpu_debugfs(vcpu, id);
451
	}
452
	return vcpu;
453 454
}

455 456 457 458 459
int kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
{
	return 0;
}

460 461
void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
{
462 463 464 465
	/* Make sure we're not using the vcpu anymore */
	hrtimer_cancel(&vcpu->arch.dec_timer);
	tasklet_kill(&vcpu->arch.tasklet);

466
	kvmppc_remove_vcpu_debugfs(vcpu);
S
Scott Wood 已提交
467 468 469 470 471 472 473

	switch (vcpu->arch.irq_type) {
	case KVMPPC_IRQ_MPIC:
		kvmppc_mpic_disconnect_vcpu(vcpu->arch.mpic, vcpu);
		break;
	}

474
	kvmppc_core_vcpu_free(vcpu);
475 476 477 478 479 480 481 482 483
}

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

int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
{
484
	return kvmppc_core_pending_dec(vcpu);
485 486
}

A
Alexander Graf 已提交
487 488 489 490 491 492 493 494 495 496 497 498 499 500
/*
 * 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;
}

501 502
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
503 504
	int ret;

A
Alexander Graf 已提交
505 506 507
	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;
508
	vcpu->arch.dec_expires = ~(u64)0;
509

510 511 512
#ifdef CONFIG_KVM_EXIT_TIMING
	mutex_init(&vcpu->arch.exit_timing_lock);
#endif
513 514
	ret = kvmppc_subarch_vcpu_init(vcpu);
	return ret;
515 516 517 518
}

void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
{
519
	kvmppc_mmu_destroy(vcpu);
520
	kvmppc_subarch_vcpu_uninit(vcpu);
521 522 523 524
}

void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
525 526 527 528 529 530 531 532 533 534
#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
535
	kvmppc_core_vcpu_load(vcpu, cpu);
536 537 538 539
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
540
	kvmppc_core_vcpu_put(vcpu);
541 542 543
#ifdef CONFIG_BOOKE
	vcpu->arch.vrsave = mfspr(SPRN_VRSAVE);
#endif
544 545 546 547 548
}

static void kvmppc_complete_dcr_load(struct kvm_vcpu *vcpu,
                                     struct kvm_run *run)
{
549
	kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, run->dcr.data);
550 551 552 553 554
}

static void kvmppc_complete_mmio_load(struct kvm_vcpu *vcpu,
                                      struct kvm_run *run)
{
555
	u64 uninitialized_var(gpr);
556

557
	if (run->mmio.len > sizeof(gpr)) {
558 559 560 561 562 563
		printk(KERN_ERR "bad MMIO length: %d\n", run->mmio.len);
		return;
	}

	if (vcpu->arch.mmio_is_bigendian) {
		switch (run->mmio.len) {
564
		case 8: gpr = *(u64 *)run->mmio.data; break;
565 566 567
		case 4: gpr = *(u32 *)run->mmio.data; break;
		case 2: gpr = *(u16 *)run->mmio.data; break;
		case 1: gpr = *(u8 *)run->mmio.data; break;
568 569 570 571
		}
	} else {
		/* Convert BE data from userland back to LE. */
		switch (run->mmio.len) {
572 573 574
		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;
575 576
		}
	}
577

A
Alexander Graf 已提交
578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593
	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;
		}
	}

594
	kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
595

596 597
	switch (vcpu->arch.io_gpr & KVM_MMIO_REG_EXT_MASK) {
	case KVM_MMIO_REG_GPR:
598 599
		kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
		break;
600 601
	case KVM_MMIO_REG_FPR:
		vcpu->arch.fpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
602
		break;
603
#ifdef CONFIG_PPC_BOOK3S
604 605
	case KVM_MMIO_REG_QPR:
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
606
		break;
607 608 609
	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;
610
		break;
611
#endif
612 613 614
	default:
		BUG();
	}
615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632
}

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
633
	vcpu->arch.mmio_sign_extend = 0;
634

A
Alexander Graf 已提交
635 636 637 638 639 640 641
	if (!kvm_io_bus_read(vcpu->kvm, KVM_MMIO_BUS, run->mmio.phys_addr,
			     bytes, &run->mmio.data)) {
		kvmppc_complete_mmio_load(vcpu, run);
		vcpu->mmio_needed = 0;
		return EMULATE_DONE;
	}

642 643 644
	return EMULATE_DO_MMIO;
}

A
Alexander Graf 已提交
645 646 647 648 649 650 651
/* 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;

	vcpu->arch.mmio_sign_extend = 1;
A
Alexander Graf 已提交
652
	r = kvmppc_handle_load(run, vcpu, rt, bytes, is_bigendian);
A
Alexander Graf 已提交
653 654 655 656

	return r;
}

657
int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
658
                        u64 val, unsigned int bytes, int is_bigendian)
659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675
{
	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) {
676
		case 8: *(u64 *)data = val; break;
677 678 679 680 681 682 683 684 685 686 687 688 689
		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;
		}
	}

A
Alexander Graf 已提交
690 691 692 693 694 695
	if (!kvm_io_bus_write(vcpu->kvm, KVM_MMIO_BUS, run->mmio.phys_addr,
			      bytes, &run->mmio.data)) {
		vcpu->mmio_needed = 0;
		return EMULATE_DONE;
	}

696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714
	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;
715 716 717 718 719 720 721
	} 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;
722 723 724 725 726 727 728
	} 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;
729 730 731 732 733
#ifdef CONFIG_BOOKE
	} else if (vcpu->arch.epr_needed) {
		kvmppc_set_epr(vcpu, run->epr.epr);
		vcpu->arch.epr_needed = 0;
#endif
734 735
	}

736
	r = kvmppc_vcpu_run(run, vcpu);
737 738 739 740 741 742 743 744 745

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

	return r;
}

int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq)
{
746
	if (irq->irq == KVM_INTERRUPT_UNSET) {
747
		kvmppc_core_dequeue_external(vcpu);
748 749 750 751
		return 0;
	}

	kvmppc_core_queue_external(vcpu, irq);
752

753
	kvm_vcpu_kick(vcpu);
754

755 756 757
	return 0;
}

758 759 760 761 762 763 764 765 766
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) {
767 768 769 770
	case KVM_CAP_PPC_OSI:
		r = 0;
		vcpu->arch.osi_enabled = true;
		break;
771 772 773 774
	case KVM_CAP_PPC_PAPR:
		r = 0;
		vcpu->arch.papr_enabled = true;
		break;
775 776
	case KVM_CAP_PPC_EPR:
		r = 0;
777 778 779 780
		if (cap->args[0])
			vcpu->arch.epr_flags |= KVMPPC_EPR_USER;
		else
			vcpu->arch.epr_flags &= ~KVMPPC_EPR_USER;
781
		break;
782 783 784 785 786 787
#ifdef CONFIG_BOOKE
	case KVM_CAP_PPC_BOOKE_WATCHDOG:
		r = 0;
		vcpu->arch.watchdog_enabled = true;
		break;
#endif
788
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
789 790 791 792 793 794 795 796 797 798
	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
799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817
	}
#endif
#ifdef CONFIG_KVM_MPIC
	case KVM_CAP_IRQ_MPIC: {
		struct file *filp;
		struct kvm_device *dev;

		r = -EBADF;
		filp = fget(cap->args[0]);
		if (!filp)
			break;

		r = -EPERM;
		dev = kvm_device_from_filp(filp);
		if (dev)
			r = kvmppc_mpic_connect_vcpu(dev, vcpu, cap->args[1]);

		fput(filp);
		break;
S
Scott Wood 已提交
818 819
	}
#endif
820 821 822 823 824
	default:
		r = -EINVAL;
		break;
	}

825 826 827
	if (!r)
		r = kvmppc_sanity_check(vcpu);

828 829 830
	return r;
}

831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849
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;

850 851
	switch (ioctl) {
	case KVM_INTERRUPT: {
852 853 854
		struct kvm_interrupt irq;
		r = -EFAULT;
		if (copy_from_user(&irq, argp, sizeof(irq)))
855
			goto out;
856
		r = kvm_vcpu_ioctl_interrupt(vcpu, &irq);
857
		goto out;
858
	}
859

860 861 862 863 864 865 866 867 868
	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
869

870 871 872 873 874 875 876 877 878 879 880 881 882 883
	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;
	}

884
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
885 886 887 888 889 890 891 892 893
	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
894 895 896 897 898 899 900 901
	default:
		r = -EINVAL;
	}

out:
	return r;
}

902 903 904 905 906
int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
{
	return VM_FAULT_SIGBUS;
}

907 908
static int kvm_vm_ioctl_get_pvinfo(struct kvm_ppc_pvinfo *pvinfo)
{
909 910 911 912 913 914 915 916
	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
917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934
	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;
935
#endif
936

937 938
	pvinfo->flags = KVM_PPC_PVINFO_FLAGS_EV_IDLE;

939 940 941
	return 0;
}

942 943 944
long kvm_arch_vm_ioctl(struct file *filp,
                       unsigned int ioctl, unsigned long arg)
{
945
	struct kvm *kvm __maybe_unused = filp->private_data;
946
	void __user *argp = (void __user *)arg;
947 948 949
	long r;

	switch (ioctl) {
950 951
	case KVM_PPC_GET_PVINFO: {
		struct kvm_ppc_pvinfo pvinfo;
952
		memset(&pvinfo, 0, sizeof(pvinfo));
953 954 955 956 957 958 959 960
		r = kvm_vm_ioctl_get_pvinfo(&pvinfo);
		if (copy_to_user(argp, &pvinfo, sizeof(pvinfo))) {
			r = -EFAULT;
			goto out;
		}

		break;
	}
961
#ifdef CONFIG_PPC_BOOK3S_64
962 963 964 965 966 967 968 969 970
	case KVM_CREATE_SPAPR_TCE: {
		struct kvm_create_spapr_tce create_tce;

		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;
	}
971
#endif /* CONFIG_PPC_BOOK3S_64 */
972

973
#ifdef CONFIG_KVM_BOOK3S_64_HV
974 975 976 977 978 979 980 981
	case KVM_ALLOCATE_RMA: {
		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;
	}
982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997

	case KVM_PPC_ALLOCATE_HTAB: {
		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;
	}
998 999 1000 1001 1002 1003 1004 1005 1006 1007

	case KVM_PPC_GET_HTAB_FD: {
		struct kvm_get_htab_fd ghf;

		r = -EFAULT;
		if (copy_from_user(&ghf, argp, sizeof(ghf)))
			break;
		r = kvm_vm_ioctl_get_htab_fd(kvm, &ghf);
		break;
	}
1008 1009
#endif /* CONFIG_KVM_BOOK3S_64_HV */

1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020
#ifdef CONFIG_PPC_BOOK3S_64
	case KVM_PPC_GET_SMMU_INFO: {
		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 */
1021
	default:
1022
		r = -ENOTTY;
1023 1024
	}

1025
out:
1026 1027 1028
	return r;
}

1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062
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));
}

1063 1064 1065 1066 1067 1068 1069 1070
int kvm_arch_init(void *opaque)
{
	return 0;
}

void kvm_arch_exit(void)
{
}