powerpc.c 22.4 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
			break;
		}

81 82 83 84 85 86 87 88
		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).
		 */
89
		smp_mb();
90

91 92 93 94
		if (vcpu->requests) {
			/* Make sure we process requests preemptable */
			local_irq_enable();
			trace_kvm_check_requests(vcpu);
95
			r = kvmppc_core_check_requests(vcpu);
96
			local_irq_disable();
97 98 99
			if (r > 0)
				continue;
			break;
100 101 102 103 104 105 106 107
		}

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

108 109 110 111 112 113 114
#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();
115
			kvm_guest_exit();
116 117 118 119 120 121
			continue;
		}

		trace_hardirqs_on();
#endif

122
		kvm_guest_enter();
123 124 125 126 127 128 129
		break;
	}

	return r;
}
#endif /* CONFIG_KVM_BOOK3S_64_HV */

130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148
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) {
149
	case KVM_HCALL_TOKEN(KVM_HC_PPC_MAP_MAGIC_PAGE):
150 151 152 153
	{
		vcpu->arch.magic_page_pa = param1;
		vcpu->arch.magic_page_ea = param2;

154
		r2 = KVM_MAGIC_FEAT_SR | KVM_MAGIC_FEAT_MAS0_TO_SPRG7;
155

156
		r = EV_SUCCESS;
157 158
		break;
	}
159 160
	case KVM_HCALL_TOKEN(KVM_HC_FEATURES):
		r = EV_SUCCESS;
161
#if defined(CONFIG_PPC_BOOK3S) || defined(CONFIG_KVM_E500V2)
S
Scott Wood 已提交
162
		/* XXX Missing magic page on 44x */
163 164
		r2 |= (1 << KVM_FEATURE_MAGIC_PAGE);
#endif
165 166 167

		/* Second return value is in r4 */
		break;
168 169 170 171 172
	case EV_HCALL_TOKEN(EV_IDLE):
		r = EV_SUCCESS;
		kvm_vcpu_block(vcpu);
		clear_bit(KVM_REQ_UNHALT, &vcpu->requests);
		break;
173
	default:
174
		r = EV_UNIMPLEMENTED;
175 176 177
		break;
	}

178 179
	kvmppc_set_gpr(vcpu, 4, r2);

180 181
	return r;
}
182

183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
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

201 202 203 204 205
#ifdef CONFIG_KVM_BOOKE_HV
	if (!cpu_has_feature(CPU_FTR_EMB_HV))
		goto out;
#endif

206 207 208 209 210 211 212
	r = true;

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

213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
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__,
236
		       kvmppc_get_last_inst(vcpu));
237 238 239
		r = RESUME_HOST;
		break;
	default:
240 241
		WARN_ON(1);
		r = RESUME_GUEST;
242 243 244 245 246
	}

	return r;
}

247
int kvm_arch_hardware_enable(void *garbage)
248
{
249
	return 0;
250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266
}

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)
{
267
	*(int *)rtn = kvmppc_core_check_processor_compat();
268 269
}

270
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
271
{
272 273 274
	if (type)
		return -EINVAL;

275
	return kvmppc_core_init_vm(kvm);
276 277
}

278
void kvm_arch_destroy_vm(struct kvm *kvm)
279 280
{
	unsigned int i;
281
	struct kvm_vcpu *vcpu;
282

283 284 285 286 287 288 289 290
	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);
291 292 293

	kvmppc_core_destroy_vm(kvm);

294
	mutex_unlock(&kvm->lock);
295 296
}

297 298 299 300
void kvm_arch_sync_events(struct kvm *kvm)
{
}

301 302 303 304 305
int kvm_dev_ioctl_check_extension(long ext)
{
	int r;

	switch (ext) {
S
Scott Wood 已提交
306 307
#ifdef CONFIG_BOOKE
	case KVM_CAP_PPC_BOOKE_SREGS:
308
	case KVM_CAP_PPC_BOOKE_WATCHDOG:
309
	case KVM_CAP_PPC_EPR:
S
Scott Wood 已提交
310
#else
311
	case KVM_CAP_PPC_SEGSTATE:
312
	case KVM_CAP_PPC_HIOR:
313
	case KVM_CAP_PPC_PAPR:
S
Scott Wood 已提交
314
#endif
315
	case KVM_CAP_PPC_UNSET_IRQ:
316
	case KVM_CAP_PPC_IRQ_LEVEL:
317
	case KVM_CAP_ENABLE_CAP:
318
	case KVM_CAP_ONE_REG:
A
Alexander Graf 已提交
319
	case KVM_CAP_IOEVENTFD:
320
	case KVM_CAP_DEVICE_CTRL:
321 322 323 324
		r = 1;
		break;
#ifndef CONFIG_KVM_BOOK3S_64_HV
	case KVM_CAP_PPC_PAIRED_SINGLES:
325
	case KVM_CAP_PPC_OSI:
326
	case KVM_CAP_PPC_GET_PVINFO:
327
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
328 329
	case KVM_CAP_SW_TLB:
#endif
330 331
		r = 1;
		break;
332 333 334
	case KVM_CAP_COALESCED_MMIO:
		r = KVM_COALESCED_MMIO_PAGE_OFFSET;
		break;
335
#endif
336
#ifdef CONFIG_PPC_BOOK3S_64
337
	case KVM_CAP_SPAPR_TCE:
338
	case KVM_CAP_PPC_ALLOC_HTAB:
339 340
		r = 1;
		break;
341 342
#endif /* CONFIG_PPC_BOOK3S_64 */
#ifdef CONFIG_KVM_BOOK3S_64_HV
343 344 345
	case KVM_CAP_PPC_SMT:
		r = threads_per_core;
		break;
346 347
	case KVM_CAP_PPC_RMA:
		r = 1;
348 349 350
		/* PPC970 requires an RMA */
		if (cpu_has_feature(CPU_FTR_ARCH_201))
			r = 2;
351
		break;
352
#endif
353
	case KVM_CAP_SYNC_MMU:
354
#ifdef CONFIG_KVM_BOOK3S_64_HV
355
		r = cpu_has_feature(CPU_FTR_ARCH_206) ? 1 : 0;
356 357 358 359
#elif defined(KVM_ARCH_WANT_MMU_NOTIFIER)
		r = 1;
#else
		r = 0;
360 361 362 363 364 365
		break;
#endif
#ifdef CONFIG_KVM_BOOK3S_64_HV
	case KVM_CAP_PPC_HTAB_FD:
		r = 1;
		break;
366
#endif
367
		break;
368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383
	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;
384 385 386 387 388
#ifdef CONFIG_PPC_BOOK3S_64
	case KVM_CAP_PPC_GET_SMMU_INFO:
		r = 1;
		break;
#endif
389 390 391 392 393 394 395 396 397 398 399 400 401 402
	default:
		r = 0;
		break;
	}
	return r;

}

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

403 404 405
void kvm_arch_free_memslot(struct kvm_memory_slot *free,
			   struct kvm_memory_slot *dont)
{
406
	kvmppc_core_free_memslot(free, dont);
407 408 409 410
}

int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
{
411
	return kvmppc_core_create_memslot(slot, npages);
412 413
}

414
int kvm_arch_prepare_memory_region(struct kvm *kvm,
415
				   struct kvm_memory_slot *memslot,
416 417
				   struct kvm_userspace_memory_region *mem,
				   enum kvm_mr_change change)
418
{
419
	return kvmppc_core_prepare_memory_region(kvm, memslot, mem);
420 421
}

422
void kvm_arch_commit_memory_region(struct kvm *kvm,
423
				   struct kvm_userspace_memory_region *mem,
424 425
				   const struct kvm_memory_slot *old,
				   enum kvm_mr_change change)
426
{
427
	kvmppc_core_commit_memory_region(kvm, mem, old);
428 429
}

430 431 432
void kvm_arch_flush_shadow_all(struct kvm *kvm)
{
}
433

434 435
void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
				   struct kvm_memory_slot *slot)
436
{
437
	kvmppc_core_flush_memslot(kvm, slot);
438 439
}

440 441
struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
{
442 443
	struct kvm_vcpu *vcpu;
	vcpu = kvmppc_core_vcpu_create(kvm, id);
444 445
	if (!IS_ERR(vcpu)) {
		vcpu->arch.wqp = &vcpu->wq;
446
		kvmppc_create_vcpu_debugfs(vcpu, id);
447
	}
448
	return vcpu;
449 450
}

451 452 453 454 455
int kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
{
	return 0;
}

456 457
void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
{
458 459 460 461
	/* Make sure we're not using the vcpu anymore */
	hrtimer_cancel(&vcpu->arch.dec_timer);
	tasklet_kill(&vcpu->arch.tasklet);

462
	kvmppc_remove_vcpu_debugfs(vcpu);
463
	kvmppc_core_vcpu_free(vcpu);
464 465 466 467 468 469 470 471 472
}

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

int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
{
473
	return kvmppc_core_pending_dec(vcpu);
474 475
}

A
Alexander Graf 已提交
476 477 478 479 480 481 482 483 484 485 486 487 488 489
/*
 * 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;
}

490 491
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
492 493
	int ret;

A
Alexander Graf 已提交
494 495 496
	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;
497
	vcpu->arch.dec_expires = ~(u64)0;
498

499 500 501
#ifdef CONFIG_KVM_EXIT_TIMING
	mutex_init(&vcpu->arch.exit_timing_lock);
#endif
502 503
	ret = kvmppc_subarch_vcpu_init(vcpu);
	return ret;
504 505 506 507
}

void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
{
508
	kvmppc_mmu_destroy(vcpu);
509
	kvmppc_subarch_vcpu_uninit(vcpu);
510 511 512 513
}

void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
514 515 516 517 518 519 520 521 522 523
#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
524
	kvmppc_core_vcpu_load(vcpu, cpu);
525 526 527 528
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
529
	kvmppc_core_vcpu_put(vcpu);
530 531 532
#ifdef CONFIG_BOOKE
	vcpu->arch.vrsave = mfspr(SPRN_VRSAVE);
#endif
533 534 535 536 537
}

static void kvmppc_complete_dcr_load(struct kvm_vcpu *vcpu,
                                     struct kvm_run *run)
{
538
	kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, run->dcr.data);
539 540 541 542 543
}

static void kvmppc_complete_mmio_load(struct kvm_vcpu *vcpu,
                                      struct kvm_run *run)
{
544
	u64 uninitialized_var(gpr);
545

546
	if (run->mmio.len > sizeof(gpr)) {
547 548 549 550 551 552
		printk(KERN_ERR "bad MMIO length: %d\n", run->mmio.len);
		return;
	}

	if (vcpu->arch.mmio_is_bigendian) {
		switch (run->mmio.len) {
553
		case 8: gpr = *(u64 *)run->mmio.data; break;
554 555 556
		case 4: gpr = *(u32 *)run->mmio.data; break;
		case 2: gpr = *(u16 *)run->mmio.data; break;
		case 1: gpr = *(u8 *)run->mmio.data; break;
557 558 559 560
		}
	} else {
		/* Convert BE data from userland back to LE. */
		switch (run->mmio.len) {
561 562 563
		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;
564 565
		}
	}
566

A
Alexander Graf 已提交
567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582
	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;
		}
	}

583
	kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
584

585 586
	switch (vcpu->arch.io_gpr & KVM_MMIO_REG_EXT_MASK) {
	case KVM_MMIO_REG_GPR:
587 588
		kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
		break;
589 590
	case KVM_MMIO_REG_FPR:
		vcpu->arch.fpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
591
		break;
592
#ifdef CONFIG_PPC_BOOK3S
593 594
	case KVM_MMIO_REG_QPR:
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_MMIO_REG_MASK] = gpr;
595
		break;
596 597 598
	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;
599
		break;
600
#endif
601 602 603
	default:
		BUG();
	}
604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621
}

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
622
	vcpu->arch.mmio_sign_extend = 0;
623

A
Alexander Graf 已提交
624 625 626 627 628 629 630
	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;
	}

631 632 633
	return EMULATE_DO_MMIO;
}

A
Alexander Graf 已提交
634 635 636 637 638 639 640
/* 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
641
	r = kvmppc_handle_load(run, vcpu, rt, bytes, is_bigendian);
A
Alexander Graf 已提交
642 643 644 645

	return r;
}

646
int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
647
                        u64 val, unsigned int bytes, int is_bigendian)
648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664
{
	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) {
665
		case 8: *(u64 *)data = val; break;
666 667 668 669 670 671 672 673 674 675 676 677 678
		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
679 680 681 682 683 684
	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;
	}

685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703
	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;
704 705 706 707 708 709 710
	} 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;
711 712 713 714 715 716 717
	} 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;
718 719 720 721 722
#ifdef CONFIG_BOOKE
	} else if (vcpu->arch.epr_needed) {
		kvmppc_set_epr(vcpu, run->epr.epr);
		vcpu->arch.epr_needed = 0;
#endif
723 724
	}

725
	r = kvmppc_vcpu_run(run, vcpu);
726 727 728 729 730 731 732 733 734

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

	return r;
}

int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq)
{
735
	if (irq->irq == KVM_INTERRUPT_UNSET) {
736
		kvmppc_core_dequeue_external(vcpu);
737 738 739 740
		return 0;
	}

	kvmppc_core_queue_external(vcpu, irq);
741

742
	kvm_vcpu_kick(vcpu);
743

744 745 746
	return 0;
}

747 748 749 750 751 752 753 754 755
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) {
756 757 758 759
	case KVM_CAP_PPC_OSI:
		r = 0;
		vcpu->arch.osi_enabled = true;
		break;
760 761 762 763
	case KVM_CAP_PPC_PAPR:
		r = 0;
		vcpu->arch.papr_enabled = true;
		break;
764 765
	case KVM_CAP_PPC_EPR:
		r = 0;
766 767 768 769
		if (cap->args[0])
			vcpu->arch.epr_flags |= KVMPPC_EPR_USER;
		else
			vcpu->arch.epr_flags &= ~KVMPPC_EPR_USER;
770
		break;
771 772 773 774 775 776
#ifdef CONFIG_BOOKE
	case KVM_CAP_PPC_BOOKE_WATCHDOG:
		r = 0;
		vcpu->arch.watchdog_enabled = true;
		break;
#endif
777
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
778 779 780 781 782 783 784 785 786 787 788 789
	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
790 791 792 793 794
	default:
		r = -EINVAL;
		break;
	}

795 796 797
	if (!r)
		r = kvmppc_sanity_check(vcpu);

798 799 800
	return r;
}

801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819
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;

820 821
	switch (ioctl) {
	case KVM_INTERRUPT: {
822 823 824
		struct kvm_interrupt irq;
		r = -EFAULT;
		if (copy_from_user(&irq, argp, sizeof(irq)))
825
			goto out;
826
		r = kvm_vcpu_ioctl_interrupt(vcpu, &irq);
827
		goto out;
828
	}
829

830 831 832 833 834 835 836 837 838
	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
839

840 841 842 843 844 845 846 847 848 849 850 851 852 853
	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;
	}

854
#if defined(CONFIG_KVM_E500V2) || defined(CONFIG_KVM_E500MC)
S
Scott Wood 已提交
855 856 857 858 859 860 861 862 863
	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
864 865 866 867 868 869 870 871
	default:
		r = -EINVAL;
	}

out:
	return r;
}

872 873 874 875 876
int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
{
	return VM_FAULT_SIGBUS;
}

877 878
static int kvm_vm_ioctl_get_pvinfo(struct kvm_ppc_pvinfo *pvinfo)
{
879 880 881 882 883 884 885 886
	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
887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904
	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;
905
#endif
906

907 908
	pvinfo->flags = KVM_PPC_PVINFO_FLAGS_EV_IDLE;

909 910 911
	return 0;
}

912 913 914
long kvm_arch_vm_ioctl(struct file *filp,
                       unsigned int ioctl, unsigned long arg)
{
915
	struct kvm *kvm __maybe_unused = filp->private_data;
916
	void __user *argp = (void __user *)arg;
917 918 919
	long r;

	switch (ioctl) {
920 921
	case KVM_PPC_GET_PVINFO: {
		struct kvm_ppc_pvinfo pvinfo;
922
		memset(&pvinfo, 0, sizeof(pvinfo));
923 924 925 926 927 928 929 930
		r = kvm_vm_ioctl_get_pvinfo(&pvinfo);
		if (copy_to_user(argp, &pvinfo, sizeof(pvinfo))) {
			r = -EFAULT;
			goto out;
		}

		break;
	}
931
#ifdef CONFIG_PPC_BOOK3S_64
932 933 934 935 936 937 938 939 940
	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;
	}
941
#endif /* CONFIG_PPC_BOOK3S_64 */
942

943
#ifdef CONFIG_KVM_BOOK3S_64_HV
944 945 946 947 948 949 950 951
	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;
	}
952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967

	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;
	}
968 969 970 971 972 973 974 975 976 977

	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;
	}
978 979
#endif /* CONFIG_KVM_BOOK3S_64_HV */

980 981 982 983 984 985 986 987 988 989 990
#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 */
991
	default:
992
		r = -ENOTTY;
993 994
	}

995
out:
996 997 998
	return r;
}

999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032
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));
}

1033 1034 1035 1036 1037 1038 1039 1040
int kvm_arch_init(void *opaque)
{
	return 0;
}

void kvm_arch_exit(void)
{
}