powerpc.c 13.7 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 25
/*
 * 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/module.h>
#include <linux/vmalloc.h>
A
Alexander Graf 已提交
26
#include <linux/hrtimer.h>
27
#include <linux/fs.h>
28
#include <linux/slab.h>
29 30 31
#include <asm/cputable.h>
#include <asm/uaccess.h>
#include <asm/kvm_ppc.h>
32
#include <asm/tlbflush.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 42
	return !(v->arch.shared->msr & MSR_WE) ||
	       !!(v->arch.pending_exceptions);
43 44
}

45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
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) {
64 65 66 67 68
	case HC_VENDOR_KVM | KVM_HC_PPC_MAP_MAGIC_PAGE:
	{
		vcpu->arch.magic_page_pa = param1;
		vcpu->arch.magic_page_ea = param2;

69
		r2 = KVM_MAGIC_FEAT_SR;
70

71 72 73
		r = HC_EV_SUCCESS;
		break;
	}
74 75
	case HC_VENDOR_KVM | KVM_HC_FEATURES:
		r = HC_EV_SUCCESS;
76 77 78
#if defined(CONFIG_PPC_BOOK3S) /* XXX Missing magic page on BookE */
		r2 |= (1 << KVM_FEATURE_MAGIC_PAGE);
#endif
79 80 81 82 83 84 85 86

		/* Second return value is in r4 */
		break;
	default:
		r = HC_EV_UNIMPLEMENTED;
		break;
	}

87 88
	kvmppc_set_gpr(vcpu, 4, r2);

89 90
	return r;
}
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114

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__,
115
		       kvmppc_get_last_inst(vcpu));
116 117 118 119 120 121 122 123 124
		r = RESUME_HOST;
		break;
	default:
		BUG();
	}

	return r;
}

125
int kvm_arch_hardware_enable(void *garbage)
126
{
127
	return 0;
128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
}

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)
{
145
	*(int *)rtn = kvmppc_core_check_processor_compat();
146 147
}

148
int kvm_arch_init_vm(struct kvm *kvm)
149
{
150
	return 0;
151 152
}

153
void kvm_arch_destroy_vm(struct kvm *kvm)
154 155
{
	unsigned int i;
156
	struct kvm_vcpu *vcpu;
157

158 159 160 161 162 163 164 165 166
	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);
	mutex_unlock(&kvm->lock);
167 168
}

169 170 171 172
void kvm_arch_sync_events(struct kvm *kvm)
{
}

173 174 175 176 177
int kvm_dev_ioctl_check_extension(long ext)
{
	int r;

	switch (ext) {
178
	case KVM_CAP_PPC_SEGSTATE:
179
	case KVM_CAP_PPC_PAIRED_SINGLES:
180
	case KVM_CAP_PPC_UNSET_IRQ:
181
	case KVM_CAP_PPC_IRQ_LEVEL:
182
	case KVM_CAP_ENABLE_CAP:
183
	case KVM_CAP_PPC_OSI:
184
	case KVM_CAP_PPC_GET_PVINFO:
185 186
		r = 1;
		break;
187 188 189
	case KVM_CAP_COALESCED_MMIO:
		r = KVM_COALESCED_MMIO_PAGE_OFFSET;
		break;
190 191 192 193 194 195 196 197 198 199 200 201 202 203
	default:
		r = 0;
		break;
	}
	return r;

}

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

204 205 206 207 208
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)
209 210 211 212
{
	return 0;
}

213 214 215 216 217 218 219 220 221
void kvm_arch_commit_memory_region(struct kvm *kvm,
               struct kvm_userspace_memory_region *mem,
               struct kvm_memory_slot old,
               int user_alloc)
{
       return;
}


222 223 224 225
void kvm_arch_flush_shadow(struct kvm *kvm)
{
}

226 227
struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
{
228 229
	struct kvm_vcpu *vcpu;
	vcpu = kvmppc_core_vcpu_create(kvm, id);
230 231
	if (!IS_ERR(vcpu))
		kvmppc_create_vcpu_debugfs(vcpu, id);
232
	return vcpu;
233 234 235 236
}

void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
{
237 238 239 240
	/* Make sure we're not using the vcpu anymore */
	hrtimer_cancel(&vcpu->arch.dec_timer);
	tasklet_kill(&vcpu->arch.tasklet);

241
	kvmppc_remove_vcpu_debugfs(vcpu);
242
	kvmppc_core_vcpu_free(vcpu);
243 244 245 246 247 248 249 250 251
}

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

int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
{
252
	return kvmppc_core_pending_dec(vcpu);
253 254 255 256 257 258
}

static void kvmppc_decrementer_func(unsigned long data)
{
	struct kvm_vcpu *vcpu = (struct kvm_vcpu *)data;

259
	kvmppc_core_queue_dec(vcpu);
260 261 262 263 264

	if (waitqueue_active(&vcpu->wq)) {
		wake_up_interruptible(&vcpu->wq);
		vcpu->stat.halt_wakeup++;
	}
265 266
}

A
Alexander Graf 已提交
267 268 269 270 271 272 273 274 275 276 277 278 279 280
/*
 * 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;
}

281 282
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
A
Alexander Graf 已提交
283 284 285
	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;
286 287 288 289 290 291

	return 0;
}

void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
{
292
	kvmppc_mmu_destroy(vcpu);
293 294 295 296
}

void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
297
	kvmppc_core_vcpu_load(vcpu, cpu);
298 299 300 301
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
302
	kvmppc_core_vcpu_put(vcpu);
303 304
}

J
Jan Kiszka 已提交
305
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
306
                                        struct kvm_guest_debug *dbg)
307
{
308
	return -EINVAL;
309 310 311 312 313
}

static void kvmppc_complete_dcr_load(struct kvm_vcpu *vcpu,
                                     struct kvm_run *run)
{
314
	kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, run->dcr.data);
315 316 317 318 319
}

static void kvmppc_complete_mmio_load(struct kvm_vcpu *vcpu,
                                      struct kvm_run *run)
{
320
	u64 uninitialized_var(gpr);
321

322
	if (run->mmio.len > sizeof(gpr)) {
323 324 325 326 327 328
		printk(KERN_ERR "bad MMIO length: %d\n", run->mmio.len);
		return;
	}

	if (vcpu->arch.mmio_is_bigendian) {
		switch (run->mmio.len) {
329
		case 8: gpr = *(u64 *)run->mmio.data; break;
330 331 332
		case 4: gpr = *(u32 *)run->mmio.data; break;
		case 2: gpr = *(u16 *)run->mmio.data; break;
		case 1: gpr = *(u8 *)run->mmio.data; break;
333 334 335 336
		}
	} else {
		/* Convert BE data from userland back to LE. */
		switch (run->mmio.len) {
337 338 339
		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;
340 341
		}
	}
342

A
Alexander Graf 已提交
343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358
	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;
		}
	}

359
	kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
360 361 362 363 364 365 366 367

	switch (vcpu->arch.io_gpr & KVM_REG_EXT_MASK) {
	case KVM_REG_GPR:
		kvmppc_set_gpr(vcpu, vcpu->arch.io_gpr, gpr);
		break;
	case KVM_REG_FPR:
		vcpu->arch.fpr[vcpu->arch.io_gpr & KVM_REG_MASK] = gpr;
		break;
368
#ifdef CONFIG_PPC_BOOK3S
369 370 371 372 373 374 375
	case KVM_REG_QPR:
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_REG_MASK] = gpr;
		break;
	case KVM_REG_FQPR:
		vcpu->arch.fpr[vcpu->arch.io_gpr & KVM_REG_MASK] = gpr;
		vcpu->arch.qpr[vcpu->arch.io_gpr & KVM_REG_MASK] = gpr;
		break;
376
#endif
377 378 379
	default:
		BUG();
	}
380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397
}

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
398
	vcpu->arch.mmio_sign_extend = 0;
399 400 401 402

	return EMULATE_DO_MMIO;
}

A
Alexander Graf 已提交
403 404 405 406 407 408 409 410 411 412 413 414
/* 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;
}

415
int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
416
                        u64 val, unsigned int bytes, int is_bigendian)
417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433
{
	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) {
434
		case 8: *(u64 *)data = val; break;
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466
		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;
467 468 469 470 471 472 473
	} 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;
474 475
	}

476
	kvmppc_core_deliver_interrupts(vcpu);
477 478 479 480 481 482 483 484 485 486 487 488 489 490 491

	local_irq_disable();
	kvm_guest_enter();
	r = __kvmppc_vcpu_run(run, vcpu);
	kvm_guest_exit();
	local_irq_enable();

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

	return r;
}

int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq)
{
492 493 494 495
	if (irq->irq == KVM_INTERRUPT_UNSET)
		kvmppc_core_dequeue_external(vcpu, irq);
	else
		kvmppc_core_queue_external(vcpu, irq);
496 497 498 499 500 501

	if (waitqueue_active(&vcpu->wq)) {
		wake_up_interruptible(&vcpu->wq);
		vcpu->stat.halt_wakeup++;
	}

502 503 504
	return 0;
}

505 506 507 508 509 510 511 512 513
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) {
514 515 516 517
	case KVM_CAP_PPC_OSI:
		r = 0;
		vcpu->arch.osi_enabled = true;
		break;
518 519 520 521 522 523 524 525
	default:
		r = -EINVAL;
		break;
	}

	return r;
}

526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544
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;

545 546
	switch (ioctl) {
	case KVM_INTERRUPT: {
547 548 549
		struct kvm_interrupt irq;
		r = -EFAULT;
		if (copy_from_user(&irq, argp, sizeof(irq)))
550
			goto out;
551
		r = kvm_vcpu_ioctl_interrupt(vcpu, &irq);
552
		goto out;
553
	}
554

555 556 557 558 559 560 561 562 563
	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;
	}
564 565 566 567 568 569 570 571
	default:
		r = -EINVAL;
	}

out:
	return r;
}

572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596
static int kvm_vm_ioctl_get_pvinfo(struct kvm_ppc_pvinfo *pvinfo)
{
	u32 inst_lis = 0x3c000000;
	u32 inst_ori = 0x60000000;
	u32 inst_nop = 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;

	return 0;
}

597 598 599
long kvm_arch_vm_ioctl(struct file *filp,
                       unsigned int ioctl, unsigned long arg)
{
600
	void __user *argp = (void __user *)arg;
601 602 603
	long r;

	switch (ioctl) {
604 605
	case KVM_PPC_GET_PVINFO: {
		struct kvm_ppc_pvinfo pvinfo;
606
		memset(&pvinfo, 0, sizeof(pvinfo));
607 608 609 610 611 612 613 614
		r = kvm_vm_ioctl_get_pvinfo(&pvinfo);
		if (copy_to_user(argp, &pvinfo, sizeof(pvinfo))) {
			r = -EFAULT;
			goto out;
		}

		break;
	}
615
	default:
616
		r = -ENOTTY;
617 618
	}

619
out:
620 621 622 623 624 625 626 627 628 629 630
	return r;
}

int kvm_arch_init(void *opaque)
{
	return 0;
}

void kvm_arch_exit(void)
{
}