kvm-ia64.c 44.3 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 26
/*
 * kvm_ia64.c: Basic KVM suppport On Itanium series processors
 *
 *
 * 	Copyright (C) 2007, Intel Corporation.
 *  	Xiantao Zhang  (xiantao.zhang@intel.com)
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms and conditions of the GNU General Public License,
 * version 2, as published by the Free Software Foundation.
 *
 * This program is distributed in the hope 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, Inc., 59 Temple
 * Place - Suite 330, Boston, MA 02111-1307 USA.
 *
 */

#include <linux/module.h>
#include <linux/errno.h>
#include <linux/percpu.h>
#include <linux/fs.h>
27
#include <linux/slab.h>
28 29 30 31 32 33
#include <linux/smp.h>
#include <linux/kvm_host.h>
#include <linux/kvm.h>
#include <linux/bitops.h>
#include <linux/hrtimer.h>
#include <linux/uaccess.h>
J
Joerg Roedel 已提交
34
#include <linux/iommu.h>
35
#include <linux/intel-iommu.h>
36 37 38 39 40 41 42

#include <asm/pgtable.h>
#include <asm/gcc_intrin.h>
#include <asm/pal.h>
#include <asm/cacheflush.h>
#include <asm/div64.h>
#include <asm/tlb.h>
43
#include <asm/elf.h>
44 45 46
#include <asm/sn/addrs.h>
#include <asm/sn/clksupport.h>
#include <asm/sn/shub_mmr.h>
47 48 49 50 51 52

#include "misc.h"
#include "vti.h"
#include "iodev.h"
#include "ioapic.h"
#include "lapic.h"
53
#include "irq.h"
54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70

static unsigned long kvm_vmm_base;
static unsigned long kvm_vsa_base;
static unsigned long kvm_vm_buffer;
static unsigned long kvm_vm_buffer_size;
unsigned long kvm_vmm_gp;

static long vp_env_info;

static struct kvm_vmm_info *kvm_vmm_info;

static DEFINE_PER_CPU(struct kvm_vcpu *, last_vcpu);

struct kvm_stats_debugfs_item debugfs_entries[] = {
	{ NULL }
};

71 72 73 74 75 76 77 78 79 80
static unsigned long kvm_get_itc(struct kvm_vcpu *vcpu)
{
#if defined(CONFIG_IA64_SGI_SN2) || defined(CONFIG_IA64_GENERIC)
	if (vcpu->kvm->arch.is_sn2)
		return rtc_time();
	else
#endif
		return ia64_getreg(_IA64_REG_AR_ITC);
}

81 82 83 84 85
static void kvm_flush_icache(unsigned long start, unsigned long len)
{
	int l;

	for (l = 0; l < (len + 32); l += 32)
I
Isaku Yamahata 已提交
86
		ia64_fc((void *)(start + l));
87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126

	ia64_sync_i();
	ia64_srlz_i();
}

static void kvm_flush_tlb_all(void)
{
	unsigned long i, j, count0, count1, stride0, stride1, addr;
	long flags;

	addr    = local_cpu_data->ptce_base;
	count0  = local_cpu_data->ptce_count[0];
	count1  = local_cpu_data->ptce_count[1];
	stride0 = local_cpu_data->ptce_stride[0];
	stride1 = local_cpu_data->ptce_stride[1];

	local_irq_save(flags);
	for (i = 0; i < count0; ++i) {
		for (j = 0; j < count1; ++j) {
			ia64_ptce(addr);
			addr += stride1;
		}
		addr += stride0;
	}
	local_irq_restore(flags);
	ia64_srlz_i();			/* srlz.i implies srlz.d */
}

long ia64_pal_vp_create(u64 *vpd, u64 *host_iva, u64 *opt_handler)
{
	struct ia64_pal_retval iprv;

	PAL_CALL_STK(iprv, PAL_VP_CREATE, (u64)vpd, (u64)host_iva,
			(u64)opt_handler);

	return iprv.status;
}

static  DEFINE_SPINLOCK(vp_lock);

127
int kvm_arch_hardware_enable(void *garbage)
128 129 130 131 132 133 134
{
	long  status;
	long  tmp_base;
	unsigned long pte;
	unsigned long saved_psr;
	int slot;

135
	pte = pte_val(mk_pte_phys(__pa(kvm_vmm_base), PAGE_KERNEL));
136 137
	local_irq_save(saved_psr);
	slot = ia64_itr_entry(0x3, KVM_VMM_BASE, pte, KVM_VMM_SHIFT);
138
	local_irq_restore(saved_psr);
139
	if (slot < 0)
140
		return -EINVAL;
141 142 143 144 145 146

	spin_lock(&vp_lock);
	status = ia64_pal_vp_init_env(kvm_vsa_base ?
				VP_INIT_ENV : VP_INIT_ENV_INITALIZE,
			__pa(kvm_vm_buffer), KVM_VM_BUFFER_BASE, &tmp_base);
	if (status != 0) {
147
		spin_unlock(&vp_lock);
148
		printk(KERN_WARNING"kvm: Failed to Enable VT Support!!!!\n");
149
		return -EINVAL;
150 151 152 153 154 155 156 157
	}

	if (!kvm_vsa_base) {
		kvm_vsa_base = tmp_base;
		printk(KERN_INFO"kvm: kvm_vsa_base:0x%lx\n", kvm_vsa_base);
	}
	spin_unlock(&vp_lock);
	ia64_ptr_entry(0x3, slot);
158 159

	return 0;
160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175
}

void kvm_arch_hardware_disable(void *garbage)
{

	long status;
	int slot;
	unsigned long pte;
	unsigned long saved_psr;
	unsigned long host_iva = ia64_getreg(_IA64_REG_CR_IVA);

	pte = pte_val(mk_pte_phys(__pa(kvm_vmm_base),
				PAGE_KERNEL));

	local_irq_save(saved_psr);
	slot = ia64_itr_entry(0x3, KVM_VMM_BASE, pte, KVM_VMM_SHIFT);
176
	local_irq_restore(saved_psr);
177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198
	if (slot < 0)
		return;

	status = ia64_pal_vp_exit_env(host_iva);
	if (status)
		printk(KERN_DEBUG"kvm: Failed to disable VT support! :%ld\n",
				status);
	ia64_ptr_entry(0x3, slot);
}

void kvm_arch_check_processor_compat(void *rtn)
{
	*(int *)rtn = 0;
}

int kvm_dev_ioctl_check_extension(long ext)
{

	int r;

	switch (ext) {
	case KVM_CAP_IRQCHIP:
199
	case KVM_CAP_MP_STATE:
200
	case KVM_CAP_IRQ_INJECT_STATUS:
201 202
		r = 1;
		break;
203 204 205
	case KVM_CAP_COALESCED_MMIO:
		r = KVM_COALESCED_MMIO_PAGE_OFFSET;
		break;
206
	case KVM_CAP_IOMMU:
J
Joerg Roedel 已提交
207
		r = iommu_found();
208
		break;
209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226
	default:
		r = 0;
	}
	return r;

}

static int handle_vm_error(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
	kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
	kvm_run->hw.hardware_exit_reason = 1;
	return 0;
}

static int handle_mmio(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
	struct kvm_mmio_req *p;
	struct kvm_io_device *mmio_dev;
227
	int r;
228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243

	p = kvm_get_vcpu_ioreq(vcpu);

	if ((p->addr & PAGE_MASK) == IOAPIC_DEFAULT_BASE_ADDRESS)
		goto mmio;
	vcpu->mmio_needed = 1;
	vcpu->mmio_phys_addr = kvm_run->mmio.phys_addr = p->addr;
	vcpu->mmio_size = kvm_run->mmio.len = p->size;
	vcpu->mmio_is_write = kvm_run->mmio.is_write = !p->dir;

	if (vcpu->mmio_is_write)
		memcpy(vcpu->mmio_data, &p->data, p->size);
	memcpy(kvm_run->mmio.data, &p->data, p->size);
	kvm_run->exit_reason = KVM_EXIT_MMIO;
	return 0;
mmio:
244
	if (p->dir)
M
Marcelo Tosatti 已提交
245
		r = kvm_io_bus_read(vcpu->kvm, KVM_MMIO_BUS, p->addr,
246 247
				    p->size, &p->data);
	else
M
Marcelo Tosatti 已提交
248
		r = kvm_io_bus_write(vcpu->kvm, KVM_MMIO_BUS, p->addr,
249 250
				     p->size, &p->data);
	if (r)
251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288
		printk(KERN_ERR"kvm: No iodevice found! addr:%lx\n", p->addr);
	p->state = STATE_IORESP_READY;

	return 1;
}

static int handle_pal_call(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
	struct exit_ctl_data *p;

	p = kvm_get_exit_data(vcpu);

	if (p->exit_reason == EXIT_REASON_PAL_CALL)
		return kvm_pal_emul(vcpu, kvm_run);
	else {
		kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
		kvm_run->hw.hardware_exit_reason = 2;
		return 0;
	}
}

static int handle_sal_call(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
	struct exit_ctl_data *p;

	p = kvm_get_exit_data(vcpu);

	if (p->exit_reason == EXIT_REASON_SAL_CALL) {
		kvm_sal_emul(vcpu);
		return 1;
	} else {
		kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
		kvm_run->hw.hardware_exit_reason = 3;
		return 0;
	}

}

289 290 291 292 293 294 295 296 297 298 299 300
static int __apic_accept_irq(struct kvm_vcpu *vcpu, uint64_t vector)
{
	struct vpd *vpd = to_host(vcpu->kvm, vcpu->arch.vpd);

	if (!test_and_set_bit(vector, &vpd->irr[0])) {
		vcpu->arch.irq_new_pending = 1;
		kvm_vcpu_kick(vcpu);
		return 1;
	}
	return 0;
}

301 302 303 304 305 306 307 308 309 310 311
/*
 *  offset: address offset to IPI space.
 *  value:  deliver value.
 */
static void vcpu_deliver_ipi(struct kvm_vcpu *vcpu, uint64_t dm,
				uint64_t vector)
{
	switch (dm) {
	case SAPIC_FIXED:
		break;
	case SAPIC_NMI:
312
		vector = 2;
313 314
		break;
	case SAPIC_EXTINT:
315
		vector = 0;
316 317 318 319 320
		break;
	case SAPIC_INIT:
	case SAPIC_PMI:
	default:
		printk(KERN_ERR"kvm: Unimplemented Deliver reserved IPI!\n");
321
		return;
322
	}
323
	__apic_accept_irq(vcpu, vector);
324 325 326 327 328 329 330
}

static struct kvm_vcpu *lid_to_vcpu(struct kvm *kvm, unsigned long id,
			unsigned long eid)
{
	union ia64_lid lid;
	int i;
331
	struct kvm_vcpu *vcpu;
332

333 334 335 336
	kvm_for_each_vcpu(i, vcpu, kvm) {
		lid.val = VCPU_LID(vcpu);
		if (lid.id == id && lid.eid == eid)
			return vcpu;
337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359
	}

	return NULL;
}

static int handle_ipi(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
	struct exit_ctl_data *p = kvm_get_exit_data(vcpu);
	struct kvm_vcpu *target_vcpu;
	struct kvm_pt_regs *regs;
	union ia64_ipi_a addr = p->u.ipi_data.addr;
	union ia64_ipi_d data = p->u.ipi_data.data;

	target_vcpu = lid_to_vcpu(vcpu->kvm, addr.id, addr.eid);
	if (!target_vcpu)
		return handle_vm_error(vcpu, kvm_run);

	if (!target_vcpu->arch.launched) {
		regs = vcpu_regs(target_vcpu);

		regs->cr_iip = vcpu->kvm->arch.rdv_sal_data.boot_ip;
		regs->r1 = vcpu->kvm->arch.rdv_sal_data.boot_gp;

360
		target_vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE;
361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401
		if (waitqueue_active(&target_vcpu->wq))
			wake_up_interruptible(&target_vcpu->wq);
	} else {
		vcpu_deliver_ipi(target_vcpu, data.dm, data.vector);
		if (target_vcpu != vcpu)
			kvm_vcpu_kick(target_vcpu);
	}

	return 1;
}

struct call_data {
	struct kvm_ptc_g ptc_g_data;
	struct kvm_vcpu *vcpu;
};

static void vcpu_global_purge(void *info)
{
	struct call_data *p = (struct call_data *)info;
	struct kvm_vcpu *vcpu = p->vcpu;

	if (test_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests))
		return;

	set_bit(KVM_REQ_PTC_G, &vcpu->requests);
	if (vcpu->arch.ptc_g_count < MAX_PTC_G_NUM) {
		vcpu->arch.ptc_g_data[vcpu->arch.ptc_g_count++] =
							p->ptc_g_data;
	} else {
		clear_bit(KVM_REQ_PTC_G, &vcpu->requests);
		vcpu->arch.ptc_g_count = 0;
		set_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests);
	}
}

static int handle_global_purge(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
	struct exit_ctl_data *p = kvm_get_exit_data(vcpu);
	struct kvm *kvm = vcpu->kvm;
	struct call_data call_data;
	int i;
402
	struct kvm_vcpu *vcpui;
403

404 405
	call_data.ptc_g_data = p->u.ptc_g_data;

406 407 408
	kvm_for_each_vcpu(i, vcpui, kvm) {
		if (vcpui->arch.mp_state == KVM_MP_STATE_UNINITIALIZED ||
				vcpu == vcpui)
409 410
			continue;

411 412
		if (waitqueue_active(&vcpui->wq))
			wake_up_interruptible(&vcpui->wq);
413

414 415 416
		if (vcpui->cpu != -1) {
			call_data.vcpu = vcpui;
			smp_call_function_single(vcpui->cpu,
417
					vcpu_global_purge, &call_data, 1);
418 419 420 421 422 423 424 425 426 427 428 429
		} else
			printk(KERN_WARNING"kvm: Uninit vcpu received ipi!\n");

	}
	return 1;
}

static int handle_switch_rr6(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
	return 1;
}

430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446
static int kvm_sn2_setup_mappings(struct kvm_vcpu *vcpu)
{
	unsigned long pte, rtc_phys_addr, map_addr;
	int slot;

	map_addr = KVM_VMM_BASE + (1UL << KVM_VMM_SHIFT);
	rtc_phys_addr = LOCAL_MMR_OFFSET | SH_RTC;
	pte = pte_val(mk_pte_phys(rtc_phys_addr, PAGE_KERNEL_UC));
	slot = ia64_itr_entry(0x3, map_addr, pte, PAGE_SHIFT);
	vcpu->arch.sn_rtc_tr_slot = slot;
	if (slot < 0) {
		printk(KERN_ERR "Mayday mayday! RTC mapping failed!\n");
		slot = 0;
	}
	return slot;
}

447 448 449 450 451 452 453 454 455 456 457
int kvm_emulate_halt(struct kvm_vcpu *vcpu)
{

	ktime_t kt;
	long itc_diff;
	unsigned long vcpu_now_itc;
	unsigned long expires;
	struct hrtimer *p_ht = &vcpu->arch.hlt_timer;
	unsigned long cyc_per_usec = local_cpu_data->cyc_per_usec;
	struct vpd *vpd = to_host(vcpu->kvm, vcpu->arch.vpd);

458
	if (irqchip_in_kernel(vcpu->kvm)) {
459

460
		vcpu_now_itc = kvm_get_itc(vcpu) + vcpu->arch.itc_offset;
461

462 463 464 465 466 467 468 469 470 471 472 473 474
		if (time_after(vcpu_now_itc, vpd->itm)) {
			vcpu->arch.timer_check = 1;
			return 1;
		}
		itc_diff = vpd->itm - vcpu_now_itc;
		if (itc_diff < 0)
			itc_diff = -itc_diff;

		expires = div64_u64(itc_diff, cyc_per_usec);
		kt = ktime_set(0, 1000 * expires);

		vcpu->arch.ht_active = 1;
		hrtimer_start(p_ht, kt, HRTIMER_MODE_ABS);
475

476
		vcpu->arch.mp_state = KVM_MP_STATE_HALTED;
477 478 479 480
		kvm_vcpu_block(vcpu);
		hrtimer_cancel(p_ht);
		vcpu->arch.ht_active = 0;

481 482
		if (test_and_clear_bit(KVM_REQ_UNHALT, &vcpu->requests) ||
				kvm_cpu_has_pending_timer(vcpu))
483
			if (vcpu->arch.mp_state == KVM_MP_STATE_HALTED)
484
				vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE;
485

486
		if (vcpu->arch.mp_state != KVM_MP_STATE_RUNNABLE)
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507
			return -EINTR;
		return 1;
	} else {
		printk(KERN_ERR"kvm: Unsupported userspace halt!");
		return 0;
	}
}

static int handle_vm_shutdown(struct kvm_vcpu *vcpu,
		struct kvm_run *kvm_run)
{
	kvm_run->exit_reason = KVM_EXIT_SHUTDOWN;
	return 0;
}

static int handle_external_interrupt(struct kvm_vcpu *vcpu,
		struct kvm_run *kvm_run)
{
	return 1;
}

508 509 510 511 512 513 514
static int handle_vcpu_debug(struct kvm_vcpu *vcpu,
				struct kvm_run *kvm_run)
{
	printk("VMM: %s", vcpu->arch.log_buf);
	return 1;
}

515 516 517 518 519 520 521 522 523 524 525
static int (*kvm_vti_exit_handlers[])(struct kvm_vcpu *vcpu,
		struct kvm_run *kvm_run) = {
	[EXIT_REASON_VM_PANIC]              = handle_vm_error,
	[EXIT_REASON_MMIO_INSTRUCTION]      = handle_mmio,
	[EXIT_REASON_PAL_CALL]              = handle_pal_call,
	[EXIT_REASON_SAL_CALL]              = handle_sal_call,
	[EXIT_REASON_SWITCH_RR6]            = handle_switch_rr6,
	[EXIT_REASON_VM_DESTROY]            = handle_vm_shutdown,
	[EXIT_REASON_EXTERNAL_INTERRUPT]    = handle_external_interrupt,
	[EXIT_REASON_IPI]		    = handle_ipi,
	[EXIT_REASON_PTC_G]		    = handle_global_purge,
526
	[EXIT_REASON_DEBUG]		    = handle_vcpu_debug,
527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584

};

static const int kvm_vti_max_exit_handlers =
		sizeof(kvm_vti_exit_handlers)/sizeof(*kvm_vti_exit_handlers);

static uint32_t kvm_get_exit_reason(struct kvm_vcpu *vcpu)
{
	struct exit_ctl_data *p_exit_data;

	p_exit_data = kvm_get_exit_data(vcpu);
	return p_exit_data->exit_reason;
}

/*
 * The guest has exited.  See if we can fix it or if we need userspace
 * assistance.
 */
static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
{
	u32 exit_reason = kvm_get_exit_reason(vcpu);
	vcpu->arch.last_exit = exit_reason;

	if (exit_reason < kvm_vti_max_exit_handlers
			&& kvm_vti_exit_handlers[exit_reason])
		return kvm_vti_exit_handlers[exit_reason](vcpu, kvm_run);
	else {
		kvm_run->exit_reason = KVM_EXIT_UNKNOWN;
		kvm_run->hw.hardware_exit_reason = exit_reason;
	}
	return 0;
}

static inline void vti_set_rr6(unsigned long rr6)
{
	ia64_set_rr(RR6, rr6);
	ia64_srlz_i();
}

static int kvm_insert_vmm_mapping(struct kvm_vcpu *vcpu)
{
	unsigned long pte;
	struct kvm *kvm = vcpu->kvm;
	int r;

	/*Insert a pair of tr to map vmm*/
	pte = pte_val(mk_pte_phys(__pa(kvm_vmm_base), PAGE_KERNEL));
	r = ia64_itr_entry(0x3, KVM_VMM_BASE, pte, KVM_VMM_SHIFT);
	if (r < 0)
		goto out;
	vcpu->arch.vmm_tr_slot = r;
	/*Insert a pairt of tr to map data of vm*/
	pte = pte_val(mk_pte_phys(__pa(kvm->arch.vm_base), PAGE_KERNEL));
	r = ia64_itr_entry(0x3, KVM_VM_DATA_BASE,
					pte, KVM_VM_DATA_SHIFT);
	if (r < 0)
		goto out;
	vcpu->arch.vm_tr_slot = r;
585 586 587 588 589 590 591 592 593

#if defined(CONFIG_IA64_SGI_SN2) || defined(CONFIG_IA64_GENERIC)
	if (kvm->arch.is_sn2) {
		r = kvm_sn2_setup_mappings(vcpu);
		if (r < 0)
			goto out;
	}
#endif

594 595 596 597 598 599 600
	r = 0;
out:
	return r;
}

static void kvm_purge_vmm_mapping(struct kvm_vcpu *vcpu)
{
601
	struct kvm *kvm = vcpu->kvm;
602 603
	ia64_ptr_entry(0x3, vcpu->arch.vmm_tr_slot);
	ia64_ptr_entry(0x3, vcpu->arch.vm_tr_slot);
604 605 606 607
#if defined(CONFIG_IA64_SGI_SN2) || defined(CONFIG_IA64_GENERIC)
	if (kvm->arch.is_sn2)
		ia64_ptr_entry(0x3, vcpu->arch.sn_rtc_tr_slot);
#endif
608 609 610 611
}

static int kvm_vcpu_pre_transition(struct kvm_vcpu *vcpu)
{
612 613
	unsigned long psr;
	int r;
614 615 616 617 618 619 620 621 622 623 624
	int cpu = smp_processor_id();

	if (vcpu->arch.last_run_cpu != cpu ||
			per_cpu(last_vcpu, cpu) != vcpu) {
		per_cpu(last_vcpu, cpu) = vcpu;
		vcpu->arch.last_run_cpu = cpu;
		kvm_flush_tlb_all();
	}

	vcpu->arch.host_rr6 = ia64_get_rr(RR6);
	vti_set_rr6(vcpu->arch.vmm_rr);
625 626 627 628
	local_irq_save(psr);
	r = kvm_insert_vmm_mapping(vcpu);
	local_irq_restore(psr);
	return r;
629
}
630

631 632 633 634 635 636
static void kvm_vcpu_post_transition(struct kvm_vcpu *vcpu)
{
	kvm_purge_vmm_mapping(vcpu);
	vti_set_rr6(vcpu->arch.host_rr6);
}

637
static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
638 639
{
	union context *host_ctx, *guest_ctx;
640
	int r, idx;
641

642
	idx = srcu_read_lock(&vcpu->kvm->srcu);
643 644 645 646 647 648 649 650

again:
	if (signal_pending(current)) {
		r = -EINTR;
		kvm_run->exit_reason = KVM_EXIT_INTR;
		goto out;
	}

651 652 653
	preempt_disable();
	local_irq_disable();

654 655 656 657
	/*Get host and guest context with guest address space.*/
	host_ctx = kvm_get_host_context(vcpu);
	guest_ctx = kvm_get_guest_context(vcpu);

658
	clear_bit(KVM_REQ_KICK, &vcpu->requests);
659 660 661 662 663

	r = kvm_vcpu_pre_transition(vcpu);
	if (r < 0)
		goto vcpu_run_fail;

664
	srcu_read_unlock(&vcpu->kvm->srcu, idx);
665
	kvm_guest_enter();
666 667 668 669 670 671 672

	/*
	 * Transition to the guest
	 */
	kvm_vmm_info->tramp_entry(host_ctx, guest_ctx);

	kvm_vcpu_post_transition(vcpu);
673 674

	vcpu->arch.launched = 1;
675
	set_bit(KVM_REQ_KICK, &vcpu->requests);
676 677 678 679 680 681 682 683 684 685 686 687
	local_irq_enable();

	/*
	 * We must have an instruction between local_irq_enable() and
	 * kvm_guest_exit(), so the timer interrupt isn't delayed by
	 * the interrupt shadow.  The stat.exits increment will do nicely.
	 * But we need to prevent reordering, hence this barrier():
	 */
	barrier();
	kvm_guest_exit();
	preempt_enable();

688
	idx = srcu_read_lock(&vcpu->kvm->srcu);
689

690 691 692 693 694 695 696 697
	r = kvm_handle_exit(kvm_run, vcpu);

	if (r > 0) {
		if (!need_resched())
			goto again;
	}

out:
698
	srcu_read_unlock(&vcpu->kvm->srcu, idx);
699 700
	if (r > 0) {
		kvm_resched(vcpu);
701
		idx = srcu_read_lock(&vcpu->kvm->srcu);
702 703 704 705
		goto again;
	}

	return r;
706 707 708 709 710 711

vcpu_run_fail:
	local_irq_enable();
	preempt_enable();
	kvm_run->exit_reason = KVM_EXIT_FAIL_ENTRY;
	goto out;
712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727
}

static void kvm_set_mmio_data(struct kvm_vcpu *vcpu)
{
	struct kvm_mmio_req *p = kvm_get_vcpu_ioreq(vcpu);

	if (!vcpu->mmio_is_write)
		memcpy(&p->data, vcpu->mmio_data, 8);
	p->state = STATE_IORESP_READY;
}

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

728 729 730
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);

731
	if (unlikely(vcpu->arch.mp_state == KVM_MP_STATE_UNINITIALIZED)) {
732
		kvm_vcpu_block(vcpu);
733
		clear_bit(KVM_REQ_UNHALT, &vcpu->requests);
734 735
		r = -EAGAIN;
		goto out;
736 737 738 739 740 741 742 743 744
	}

	if (vcpu->mmio_needed) {
		memcpy(vcpu->mmio_data, kvm_run->mmio.data, 8);
		kvm_set_mmio_data(vcpu);
		vcpu->mmio_read_completed = 1;
		vcpu->mmio_needed = 0;
	}
	r = __vcpu_run(vcpu, kvm_run);
745
out:
746 747 748 749 750 751 752 753 754 755 756 757
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &sigsaved, NULL);

	return r;
}

static struct kvm *kvm_alloc_kvm(void)
{

	struct kvm *kvm;
	uint64_t  vm_base;

758 759
	BUG_ON(sizeof(struct kvm) > KVM_VM_STRUCT_SIZE);

760 761 762 763 764 765
	vm_base = __get_free_pages(GFP_KERNEL, get_order(KVM_VM_DATA_SIZE));

	if (!vm_base)
		return ERR_PTR(-ENOMEM);

	memset((void *)vm_base, 0, KVM_VM_DATA_SIZE);
766 767
	kvm = (struct kvm *)(vm_base +
			offsetof(struct kvm_vm_data, kvm_vm_struct));
768
	kvm->arch.vm_base = vm_base;
769
	printk(KERN_DEBUG"kvm: vm's data area:0x%lx\n", vm_base);
770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821

	return kvm;
}

struct kvm_io_range {
	unsigned long start;
	unsigned long size;
	unsigned long type;
};

static const struct kvm_io_range io_ranges[] = {
	{VGA_IO_START, VGA_IO_SIZE, GPFN_FRAME_BUFFER},
	{MMIO_START, MMIO_SIZE, GPFN_LOW_MMIO},
	{LEGACY_IO_START, LEGACY_IO_SIZE, GPFN_LEGACY_IO},
	{IO_SAPIC_START, IO_SAPIC_SIZE, GPFN_IOSAPIC},
	{PIB_START, PIB_SIZE, GPFN_PIB},
};

static void kvm_build_io_pmt(struct kvm *kvm)
{
	unsigned long i, j;

	/* Mark I/O ranges */
	for (i = 0; i < (sizeof(io_ranges) / sizeof(struct kvm_io_range));
							i++) {
		for (j = io_ranges[i].start;
				j < io_ranges[i].start + io_ranges[i].size;
				j += PAGE_SIZE)
			kvm_set_pmt_entry(kvm, j >> PAGE_SHIFT,
					io_ranges[i].type, 0);
	}

}

/*Use unused rids to virtualize guest rid.*/
#define GUEST_PHYSICAL_RR0	0x1739
#define GUEST_PHYSICAL_RR4	0x2739
#define VMM_INIT_RR		0x1660

static void kvm_init_vm(struct kvm *kvm)
{
	BUG_ON(!kvm);

	kvm->arch.metaphysical_rr0 = GUEST_PHYSICAL_RR0;
	kvm->arch.metaphysical_rr4 = GUEST_PHYSICAL_RR4;
	kvm->arch.vmm_init_rr = VMM_INIT_RR;

	/*
	 *Fill P2M entries for MMIO/IO ranges
	 */
	kvm_build_io_pmt(kvm);

822
	INIT_LIST_HEAD(&kvm->arch.assigned_dev_head);
823 824 825

	/* Reserve bit 0 of irq_sources_bitmap for userspace irq source */
	set_bit(KVM_USERSPACE_IRQ_SOURCE_ID, &kvm->arch.irq_sources_bitmap);
826 827 828 829 830 831 832 833
}

struct  kvm *kvm_arch_create_vm(void)
{
	struct kvm *kvm = kvm_alloc_kvm();

	if (IS_ERR(kvm))
		return ERR_PTR(-ENOMEM);
834 835 836

	kvm->arch.is_sn2 = ia64_platform_is("sn2");

837 838 839 840 841 842 843 844 845 846 847 848 849 850
	kvm_init_vm(kvm);

	return kvm;

}

static int kvm_vm_ioctl_get_irqchip(struct kvm *kvm,
					struct kvm_irqchip *chip)
{
	int r;

	r = 0;
	switch (chip->chip_id) {
	case KVM_IRQCHIP_IOAPIC:
G
Gleb Natapov 已提交
851
		r = kvm_get_ioapic(kvm, &chip->chip.ioapic);
852 853 854 855 856 857 858 859 860 861 862 863 864 865 866
		break;
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

static int kvm_vm_ioctl_set_irqchip(struct kvm *kvm, struct kvm_irqchip *chip)
{
	int r;

	r = 0;
	switch (chip->chip_id) {
	case KVM_IRQCHIP_IOAPIC:
G
Gleb Natapov 已提交
867
		r = kvm_set_ioapic(kvm, &chip->chip.ioapic);
868 869 870 871 872 873 874 875 876 877 878 879 880
		break;
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

#define RESTORE_REGS(_x) vcpu->arch._x = regs->_x

int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
	struct vpd *vpd = to_host(vcpu->kvm, vcpu->arch.vpd);
881
	int i;
882 883 884 885 886 887 888 889 890 891 892 893 894 895

	for (i = 0; i < 16; i++) {
		vpd->vgr[i] = regs->vpd.vgr[i];
		vpd->vbgr[i] = regs->vpd.vbgr[i];
	}
	for (i = 0; i < 128; i++)
		vpd->vcr[i] = regs->vpd.vcr[i];
	vpd->vhpi = regs->vpd.vhpi;
	vpd->vnat = regs->vpd.vnat;
	vpd->vbnat = regs->vpd.vbnat;
	vpd->vpsr = regs->vpd.vpsr;

	vpd->vpr = regs->vpd.vpr;

896
	memcpy(&vcpu->arch.guest, &regs->saved_guest, sizeof(union context));
897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925

	RESTORE_REGS(mp_state);
	RESTORE_REGS(vmm_rr);
	memcpy(vcpu->arch.itrs, regs->itrs, sizeof(struct thash_data) * NITRS);
	memcpy(vcpu->arch.dtrs, regs->dtrs, sizeof(struct thash_data) * NDTRS);
	RESTORE_REGS(itr_regions);
	RESTORE_REGS(dtr_regions);
	RESTORE_REGS(tc_regions);
	RESTORE_REGS(irq_check);
	RESTORE_REGS(itc_check);
	RESTORE_REGS(timer_check);
	RESTORE_REGS(timer_pending);
	RESTORE_REGS(last_itc);
	for (i = 0; i < 8; i++) {
		vcpu->arch.vrr[i] = regs->vrr[i];
		vcpu->arch.ibr[i] = regs->ibr[i];
		vcpu->arch.dbr[i] = regs->dbr[i];
	}
	for (i = 0; i < 4; i++)
		vcpu->arch.insvc[i] = regs->insvc[i];
	RESTORE_REGS(xtp);
	RESTORE_REGS(metaphysical_rr0);
	RESTORE_REGS(metaphysical_rr4);
	RESTORE_REGS(metaphysical_saved_rr0);
	RESTORE_REGS(metaphysical_saved_rr4);
	RESTORE_REGS(fp_psr);
	RESTORE_REGS(saved_gp);

	vcpu->arch.irq_new_pending = 1;
926
	vcpu->arch.itc_offset = regs->saved_itc - kvm_get_itc(vcpu);
927 928
	set_bit(KVM_REQ_RESUME, &vcpu->requests);

929
	return 0;
930 931 932 933 934 935 936
}

long kvm_arch_vm_ioctl(struct file *filp,
		unsigned int ioctl, unsigned long arg)
{
	struct kvm *kvm = filp->private_data;
	void __user *argp = (void __user *)arg;
937
	int r = -ENOTTY;
938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962

	switch (ioctl) {
	case KVM_SET_MEMORY_REGION: {
		struct kvm_memory_region kvm_mem;
		struct kvm_userspace_memory_region kvm_userspace_mem;

		r = -EFAULT;
		if (copy_from_user(&kvm_mem, argp, sizeof kvm_mem))
			goto out;
		kvm_userspace_mem.slot = kvm_mem.slot;
		kvm_userspace_mem.flags = kvm_mem.flags;
		kvm_userspace_mem.guest_phys_addr =
					kvm_mem.guest_phys_addr;
		kvm_userspace_mem.memory_size = kvm_mem.memory_size;
		r = kvm_vm_ioctl_set_memory_region(kvm,
					&kvm_userspace_mem, 0);
		if (r)
			goto out;
		break;
		}
	case KVM_CREATE_IRQCHIP:
		r = -EFAULT;
		r = kvm_ioapic_init(kvm);
		if (r)
			goto out;
963 964
		r = kvm_setup_default_irq_routing(kvm);
		if (r) {
965
			kvm_ioapic_destroy(kvm);
966 967
			goto out;
		}
968
		break;
969
	case KVM_IRQ_LINE_STATUS:
970 971 972 973 974 975
	case KVM_IRQ_LINE: {
		struct kvm_irq_level irq_event;

		r = -EFAULT;
		if (copy_from_user(&irq_event, argp, sizeof irq_event))
			goto out;
976
		r = -ENXIO;
977
		if (irqchip_in_kernel(kvm)) {
978 979
			__s32 status;
			status = kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID,
980
				    irq_event.irq, irq_event.level);
981
			if (ioctl == KVM_IRQ_LINE_STATUS) {
982
				r = -EFAULT;
983 984 985 986 987
				irq_event.status = status;
				if (copy_to_user(argp, &irq_event,
							sizeof irq_event))
					goto out;
			}
988 989 990 991 992 993 994 995 996 997 998 999 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 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 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152
			r = 0;
		}
		break;
		}
	case KVM_GET_IRQCHIP: {
		/* 0: PIC master, 1: PIC slave, 2: IOAPIC */
		struct kvm_irqchip chip;

		r = -EFAULT;
		if (copy_from_user(&chip, argp, sizeof chip))
				goto out;
		r = -ENXIO;
		if (!irqchip_in_kernel(kvm))
			goto out;
		r = kvm_vm_ioctl_get_irqchip(kvm, &chip);
		if (r)
			goto out;
		r = -EFAULT;
		if (copy_to_user(argp, &chip, sizeof chip))
				goto out;
		r = 0;
		break;
		}
	case KVM_SET_IRQCHIP: {
		/* 0: PIC master, 1: PIC slave, 2: IOAPIC */
		struct kvm_irqchip chip;

		r = -EFAULT;
		if (copy_from_user(&chip, argp, sizeof chip))
				goto out;
		r = -ENXIO;
		if (!irqchip_in_kernel(kvm))
			goto out;
		r = kvm_vm_ioctl_set_irqchip(kvm, &chip);
		if (r)
			goto out;
		r = 0;
		break;
		}
	default:
		;
	}
out:
	return r;
}

int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
		struct kvm_sregs *sregs)
{
	return -EINVAL;
}

int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
		struct kvm_sregs *sregs)
{
	return -EINVAL;

}
int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
		struct kvm_translation *tr)
{

	return -EINVAL;
}

static int kvm_alloc_vmm_area(void)
{
	if (!kvm_vmm_base && (kvm_vm_buffer_size < KVM_VM_BUFFER_SIZE)) {
		kvm_vmm_base = __get_free_pages(GFP_KERNEL,
				get_order(KVM_VMM_SIZE));
		if (!kvm_vmm_base)
			return -ENOMEM;

		memset((void *)kvm_vmm_base, 0, KVM_VMM_SIZE);
		kvm_vm_buffer = kvm_vmm_base + VMM_SIZE;

		printk(KERN_DEBUG"kvm:VMM's Base Addr:0x%lx, vm_buffer:0x%lx\n",
				kvm_vmm_base, kvm_vm_buffer);
	}

	return 0;
}

static void kvm_free_vmm_area(void)
{
	if (kvm_vmm_base) {
		/*Zero this area before free to avoid bits leak!!*/
		memset((void *)kvm_vmm_base, 0, KVM_VMM_SIZE);
		free_pages(kvm_vmm_base, get_order(KVM_VMM_SIZE));
		kvm_vmm_base  = 0;
		kvm_vm_buffer = 0;
		kvm_vsa_base = 0;
	}
}

static int vti_init_vpd(struct kvm_vcpu *vcpu)
{
	int i;
	union cpuid3_t cpuid3;
	struct vpd *vpd = to_host(vcpu->kvm, vcpu->arch.vpd);

	if (IS_ERR(vpd))
		return PTR_ERR(vpd);

	/* CPUID init */
	for (i = 0; i < 5; i++)
		vpd->vcpuid[i] = ia64_get_cpuid(i);

	/* Limit the CPUID number to 5 */
	cpuid3.value = vpd->vcpuid[3];
	cpuid3.number = 4;	/* 5 - 1 */
	vpd->vcpuid[3] = cpuid3.value;

	/*Set vac and vdc fields*/
	vpd->vac.a_from_int_cr = 1;
	vpd->vac.a_to_int_cr = 1;
	vpd->vac.a_from_psr = 1;
	vpd->vac.a_from_cpuid = 1;
	vpd->vac.a_cover = 1;
	vpd->vac.a_bsw = 1;
	vpd->vac.a_int = 1;
	vpd->vdc.d_vmsw = 1;

	/*Set virtual buffer*/
	vpd->virt_env_vaddr = KVM_VM_BUFFER_BASE;

	return 0;
}

static int vti_create_vp(struct kvm_vcpu *vcpu)
{
	long ret;
	struct vpd *vpd = vcpu->arch.vpd;
	unsigned long  vmm_ivt;

	vmm_ivt = kvm_vmm_info->vmm_ivt;

	printk(KERN_DEBUG "kvm: vcpu:%p,ivt: 0x%lx\n", vcpu, vmm_ivt);

	ret = ia64_pal_vp_create((u64 *)vpd, (u64 *)vmm_ivt, 0);

	if (ret) {
		printk(KERN_ERR"kvm: ia64_pal_vp_create failed!\n");
		return -EINVAL;
	}
	return 0;
}

static void init_ptce_info(struct kvm_vcpu *vcpu)
{
	ia64_ptce_info_t ptce = {0};

	ia64_get_ptce(&ptce);
	vcpu->arch.ptce_base = ptce.base;
	vcpu->arch.ptce_count[0] = ptce.count[0];
	vcpu->arch.ptce_count[1] = ptce.count[1];
	vcpu->arch.ptce_stride[0] = ptce.stride[0];
	vcpu->arch.ptce_stride[1] = ptce.stride[1];
}

static void kvm_migrate_hlt_timer(struct kvm_vcpu *vcpu)
{
	struct hrtimer *p_ht = &vcpu->arch.hlt_timer;

	if (hrtimer_cancel(p_ht))
1153
		hrtimer_start_expires(p_ht, HRTIMER_MODE_ABS);
1154 1155 1156 1157 1158 1159 1160 1161
}

static enum hrtimer_restart hlt_timer_fn(struct hrtimer *data)
{
	struct kvm_vcpu *vcpu;
	wait_queue_head_t *q;

	vcpu  = container_of(data, struct kvm_vcpu, arch.hlt_timer);
1162 1163
	q = &vcpu->wq;

1164
	if (vcpu->arch.mp_state != KVM_MP_STATE_HALTED)
1165 1166
		goto out;

1167
	if (waitqueue_active(q))
1168
		wake_up_interruptible(q);
1169

1170
out:
1171
	vcpu->arch.timer_fired = 1;
1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193
	vcpu->arch.timer_check = 1;
	return HRTIMER_NORESTART;
}

#define PALE_RESET_ENTRY    0x80000000ffffffb0UL

int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
	struct kvm_vcpu *v;
	int r;
	int i;
	long itc_offset;
	struct kvm *kvm = vcpu->kvm;
	struct kvm_pt_regs *regs = vcpu_regs(vcpu);

	union context *p_ctx = &vcpu->arch.guest;
	struct kvm_vcpu *vmm_vcpu = to_guest(vcpu->kvm, vcpu);

	/*Init vcpu context for first run.*/
	if (IS_ERR(vmm_vcpu))
		return PTR_ERR(vmm_vcpu);

1194
	if (kvm_vcpu_is_bsp(vcpu)) {
1195
		vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE;
1196 1197 1198 1199

		/*Set entry address for first run.*/
		regs->cr_iip = PALE_RESET_ENTRY;

1200
		/*Initialize itc offset for vcpus*/
1201
		itc_offset = 0UL - kvm_get_itc(vcpu);
1202
		for (i = 0; i < KVM_MAX_VCPUS; i++) {
1203 1204
			v = (struct kvm_vcpu *)((char *)vcpu +
					sizeof(struct kvm_vcpu_data) * i);
1205 1206 1207 1208
			v->arch.itc_offset = itc_offset;
			v->arch.last_itc = 0;
		}
	} else
1209
		vcpu->arch.mp_state = KVM_MP_STATE_UNINITIALIZED;
1210 1211 1212 1213 1214 1215 1216 1217

	r = -ENOMEM;
	vcpu->arch.apic = kzalloc(sizeof(struct kvm_lapic), GFP_KERNEL);
	if (!vcpu->arch.apic)
		goto out;
	vcpu->arch.apic->vcpu = vcpu;

	p_ctx->gr[1] = 0;
1218
	p_ctx->gr[12] = (unsigned long)((char *)vmm_vcpu + KVM_STK_OFFSET);
1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232
	p_ctx->gr[13] = (unsigned long)vmm_vcpu;
	p_ctx->psr = 0x1008522000UL;
	p_ctx->ar[40] = FPSR_DEFAULT; /*fpsr*/
	p_ctx->caller_unat = 0;
	p_ctx->pr = 0x0;
	p_ctx->ar[36] = 0x0; /*unat*/
	p_ctx->ar[19] = 0x0; /*rnat*/
	p_ctx->ar[18] = (unsigned long)vmm_vcpu +
				((sizeof(struct kvm_vcpu)+15) & ~15);
	p_ctx->ar[64] = 0x0; /*pfs*/
	p_ctx->cr[0] = 0x7e04UL;
	p_ctx->cr[2] = (unsigned long)kvm_vmm_info->vmm_ivt;
	p_ctx->cr[8] = 0x3c;

1233
	/*Initialize region register*/
1234 1235 1236 1237 1238 1239 1240 1241
	p_ctx->rr[0] = 0x30;
	p_ctx->rr[1] = 0x30;
	p_ctx->rr[2] = 0x30;
	p_ctx->rr[3] = 0x30;
	p_ctx->rr[4] = 0x30;
	p_ctx->rr[5] = 0x30;
	p_ctx->rr[7] = 0x30;

1242
	/*Initialize branch register 0*/
1243 1244 1245 1246 1247 1248 1249 1250 1251 1252
	p_ctx->br[0] = *(unsigned long *)kvm_vmm_info->vmm_entry;

	vcpu->arch.vmm_rr = kvm->arch.vmm_init_rr;
	vcpu->arch.metaphysical_rr0 = kvm->arch.metaphysical_rr0;
	vcpu->arch.metaphysical_rr4 = kvm->arch.metaphysical_rr4;

	hrtimer_init(&vcpu->arch.hlt_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
	vcpu->arch.hlt_timer.function = hlt_timer_fn;

	vcpu->arch.last_run_cpu = -1;
1253
	vcpu->arch.vpd = (struct vpd *)VPD_BASE(vcpu->vcpu_id);
1254 1255 1256
	vcpu->arch.vsa_base = kvm_vsa_base;
	vcpu->arch.__gp = kvm_vmm_gp;
	vcpu->arch.dirty_log_lock_pa = __pa(&kvm->arch.dirty_log_lock);
1257 1258
	vcpu->arch.vhpt.hash = (struct thash_data *)VHPT_BASE(vcpu->vcpu_id);
	vcpu->arch.vtlb.hash = (struct thash_data *)VTLB_BASE(vcpu->vcpu_id);
1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272
	init_ptce_info(vcpu);

	r = 0;
out:
	return r;
}

static int vti_vcpu_setup(struct kvm_vcpu *vcpu, int id)
{
	unsigned long psr;
	int r;

	local_irq_save(psr);
	r = kvm_insert_vmm_mapping(vcpu);
1273
	local_irq_restore(psr);
1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306
	if (r)
		goto fail;
	r = kvm_vcpu_init(vcpu, vcpu->kvm, id);
	if (r)
		goto fail;

	r = vti_init_vpd(vcpu);
	if (r) {
		printk(KERN_DEBUG"kvm: vpd init error!!\n");
		goto uninit;
	}

	r = vti_create_vp(vcpu);
	if (r)
		goto uninit;

	kvm_purge_vmm_mapping(vcpu);

	return 0;
uninit:
	kvm_vcpu_uninit(vcpu);
fail:
	return r;
}

struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
		unsigned int id)
{
	struct kvm_vcpu *vcpu;
	unsigned long vm_base = kvm->arch.vm_base;
	int r;
	int cpu;

1307 1308 1309 1310 1311 1312 1313 1314 1315
	BUG_ON(sizeof(struct kvm_vcpu) > VCPU_STRUCT_SIZE/2);

	r = -EINVAL;
	if (id >= KVM_MAX_VCPUS) {
		printk(KERN_ERR"kvm: Can't configure vcpus > %ld",
				KVM_MAX_VCPUS);
		goto fail;
	}

1316 1317 1318 1319 1320
	r = -ENOMEM;
	if (!vm_base) {
		printk(KERN_ERR"kvm: Create vcpu[%d] error!\n", id);
		goto fail;
	}
1321 1322
	vcpu = (struct kvm_vcpu *)(vm_base + offsetof(struct kvm_vm_data,
					vcpu_data[id].vcpu_struct));
1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353
	vcpu->kvm = kvm;

	cpu = get_cpu();
	r = vti_vcpu_setup(vcpu, id);
	put_cpu();

	if (r) {
		printk(KERN_DEBUG"kvm: vcpu_setup error!!\n");
		goto fail;
	}

	return vcpu;
fail:
	return ERR_PTR(r);
}

int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
{
	return 0;
}

int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
	return -EINVAL;
}

int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
	return -EINVAL;
}

J
Jan Kiszka 已提交
1354 1355
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
					struct kvm_guest_debug *dbg)
1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
{
	return -EINVAL;
}

static void free_kvm(struct kvm *kvm)
{
	unsigned long vm_base = kvm->arch.vm_base;

	if (vm_base) {
		memset((void *)vm_base, 0, KVM_VM_DATA_SIZE);
		free_pages(vm_base, get_order(KVM_VM_DATA_SIZE));
	}

}

static void kvm_release_vm_pages(struct kvm *kvm)
{
1373
	struct kvm_memslots *slots;
1374 1375 1376 1377
	struct kvm_memory_slot *memslot;
	int i, j;
	unsigned long base_gfn;

1378
	slots = kvm_memslots(kvm);
1379 1380
	for (i = 0; i < slots->nmemslots; i++) {
		memslot = &slots->memslots[i];
1381 1382 1383 1384 1385 1386 1387 1388 1389
		base_gfn = memslot->base_gfn;

		for (j = 0; j < memslot->npages; j++) {
			if (memslot->rmap[j])
				put_page((struct page *)memslot->rmap[j]);
		}
	}
}

1390 1391 1392 1393
void kvm_arch_sync_events(struct kvm *kvm)
{
}

1394 1395
void kvm_arch_destroy_vm(struct kvm *kvm)
{
1396 1397 1398 1399
	kvm_iommu_unmap_guest(kvm);
#ifdef  KVM_CAP_DEVICE_ASSIGNMENT
	kvm_free_all_assigned_devices(kvm);
#endif
1400 1401 1402
	kfree(kvm->arch.vioapic);
	kvm_release_vm_pages(kvm);
	kvm_free_physmem(kvm);
1403
	cleanup_srcu_struct(&kvm->srcu);
1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424
	free_kvm(kvm);
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
}

void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
	if (cpu != vcpu->cpu) {
		vcpu->cpu = cpu;
		if (vcpu->arch.ht_active)
			kvm_migrate_hlt_timer(vcpu);
	}
}

#define SAVE_REGS(_x) 	regs->_x = vcpu->arch._x

int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
	struct vpd *vpd = to_host(vcpu->kvm, vcpu->arch.vpd);
1425 1426
	int i;

1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440
	vcpu_load(vcpu);

	for (i = 0; i < 16; i++) {
		regs->vpd.vgr[i] = vpd->vgr[i];
		regs->vpd.vbgr[i] = vpd->vbgr[i];
	}
	for (i = 0; i < 128; i++)
		regs->vpd.vcr[i] = vpd->vcr[i];
	regs->vpd.vhpi = vpd->vhpi;
	regs->vpd.vnat = vpd->vnat;
	regs->vpd.vbnat = vpd->vbnat;
	regs->vpd.vpsr = vpd->vpsr;
	regs->vpd.vpr = vpd->vpr;

1441 1442
	memcpy(&regs->saved_guest, &vcpu->arch.guest, sizeof(union context));

1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461
	SAVE_REGS(mp_state);
	SAVE_REGS(vmm_rr);
	memcpy(regs->itrs, vcpu->arch.itrs, sizeof(struct thash_data) * NITRS);
	memcpy(regs->dtrs, vcpu->arch.dtrs, sizeof(struct thash_data) * NDTRS);
	SAVE_REGS(itr_regions);
	SAVE_REGS(dtr_regions);
	SAVE_REGS(tc_regions);
	SAVE_REGS(irq_check);
	SAVE_REGS(itc_check);
	SAVE_REGS(timer_check);
	SAVE_REGS(timer_pending);
	SAVE_REGS(last_itc);
	for (i = 0; i < 8; i++) {
		regs->vrr[i] = vcpu->arch.vrr[i];
		regs->ibr[i] = vcpu->arch.ibr[i];
		regs->dbr[i] = vcpu->arch.dbr[i];
	}
	for (i = 0; i < 4; i++)
		regs->insvc[i] = vcpu->arch.insvc[i];
1462
	regs->saved_itc = vcpu->arch.itc_offset + kvm_get_itc(vcpu);
1463 1464 1465 1466 1467 1468 1469
	SAVE_REGS(xtp);
	SAVE_REGS(metaphysical_rr0);
	SAVE_REGS(metaphysical_rr4);
	SAVE_REGS(metaphysical_saved_rr0);
	SAVE_REGS(metaphysical_saved_rr4);
	SAVE_REGS(fp_psr);
	SAVE_REGS(saved_gp);
1470

1471
	vcpu_put(vcpu);
1472
	return 0;
1473 1474
}

1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491
int kvm_arch_vcpu_ioctl_get_stack(struct kvm_vcpu *vcpu,
				  struct kvm_ia64_vcpu_stack *stack)
{
	memcpy(stack, vcpu, sizeof(struct kvm_ia64_vcpu_stack));
	return 0;
}

int kvm_arch_vcpu_ioctl_set_stack(struct kvm_vcpu *vcpu,
				  struct kvm_ia64_vcpu_stack *stack)
{
	memcpy(vcpu + 1, &stack->stack[0] + sizeof(struct kvm_vcpu),
	       sizeof(struct kvm_ia64_vcpu_stack) - sizeof(struct kvm_vcpu));

	vcpu->arch.exit_data = ((struct kvm_vcpu *)stack)->arch.exit_data;
	return 0;
}

1492 1493 1494 1495 1496 1497 1498 1499 1500
void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
{

	hrtimer_cancel(&vcpu->arch.hlt_timer);
	kfree(vcpu->arch.apic);
}


long kvm_arch_vcpu_ioctl(struct file *filp,
1501
			 unsigned int ioctl, unsigned long arg)
1502
{
1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533
	struct kvm_vcpu *vcpu = filp->private_data;
	void __user *argp = (void __user *)arg;
	struct kvm_ia64_vcpu_stack *stack = NULL;
	long r;

	switch (ioctl) {
	case KVM_IA64_VCPU_GET_STACK: {
		struct kvm_ia64_vcpu_stack __user *user_stack;
	        void __user *first_p = argp;

		r = -EFAULT;
		if (copy_from_user(&user_stack, first_p, sizeof(void *)))
			goto out;

		if (!access_ok(VERIFY_WRITE, user_stack,
			       sizeof(struct kvm_ia64_vcpu_stack))) {
			printk(KERN_INFO "KVM_IA64_VCPU_GET_STACK: "
			       "Illegal user destination address for stack\n");
			goto out;
		}
		stack = kzalloc(sizeof(struct kvm_ia64_vcpu_stack), GFP_KERNEL);
		if (!stack) {
			r = -ENOMEM;
			goto out;
		}

		r = kvm_arch_vcpu_ioctl_get_stack(vcpu, stack);
		if (r)
			goto out;

		if (copy_to_user(user_stack, stack,
1534 1535
				 sizeof(struct kvm_ia64_vcpu_stack))) {
			r = -EFAULT;
1536
			goto out;
1537
		}
1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574

		break;
	}
	case KVM_IA64_VCPU_SET_STACK: {
		struct kvm_ia64_vcpu_stack __user *user_stack;
	        void __user *first_p = argp;

		r = -EFAULT;
		if (copy_from_user(&user_stack, first_p, sizeof(void *)))
			goto out;

		if (!access_ok(VERIFY_READ, user_stack,
			    sizeof(struct kvm_ia64_vcpu_stack))) {
			printk(KERN_INFO "KVM_IA64_VCPU_SET_STACK: "
			       "Illegal user address for stack\n");
			goto out;
		}
		stack = kmalloc(sizeof(struct kvm_ia64_vcpu_stack), GFP_KERNEL);
		if (!stack) {
			r = -ENOMEM;
			goto out;
		}
		if (copy_from_user(stack, user_stack,
				   sizeof(struct kvm_ia64_vcpu_stack)))
			goto out;

		r = kvm_arch_vcpu_ioctl_set_stack(vcpu, stack);
		break;
	}

	default:
		r = -EINVAL;
	}

out:
	kfree(stack);
	return r;
1575 1576
}

1577 1578
int kvm_arch_prepare_memory_region(struct kvm *kvm,
		struct kvm_memory_slot *memslot,
1579
		struct kvm_memory_slot old,
1580
		struct kvm_userspace_memory_region *mem,
1581 1582 1583
		int user_alloc)
{
	unsigned long i;
1584
	unsigned long pfn;
1585
	int npages = memslot->npages;
1586 1587
	unsigned long base_gfn = memslot->base_gfn;

1588 1589 1590
	if (base_gfn + npages > (KVM_MAX_MEM_SIZE >> PAGE_SHIFT))
		return -ENOMEM;

1591
	for (i = 0; i < npages; i++) {
1592 1593 1594 1595
		pfn = gfn_to_pfn(kvm, base_gfn + i);
		if (!kvm_is_mmio_pfn(pfn)) {
			kvm_set_pmt_entry(kvm, base_gfn + i,
					pfn << PAGE_SHIFT,
1596
				_PAGE_AR_RWX | _PAGE_MA_WB);
1597 1598 1599
			memslot->rmap[i] = (unsigned long)pfn_to_page(pfn);
		} else {
			kvm_set_pmt_entry(kvm, base_gfn + i,
1600
					GPFN_PHYS_MMIO | (pfn << PAGE_SHIFT),
1601 1602 1603
					_PAGE_MA_UC);
			memslot->rmap[i] = 0;
			}
1604 1605 1606 1607 1608
	}

	return 0;
}

1609 1610 1611 1612 1613 1614 1615 1616
void kvm_arch_commit_memory_region(struct kvm *kvm,
		struct kvm_userspace_memory_region *mem,
		struct kvm_memory_slot old,
		int user_alloc)
{
	return;
}

1617 1618
void kvm_arch_flush_shadow(struct kvm *kvm)
{
1619
	kvm_flush_remote_tlbs(kvm);
1620
}
1621 1622

long kvm_arch_dev_ioctl(struct file *filp,
1623
			unsigned int ioctl, unsigned long arg)
1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661
{
	return -EINVAL;
}

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

static int vti_cpu_has_kvm_support(void)
{
	long  avail = 1, status = 1, control = 1;
	long ret;

	ret = ia64_pal_proc_get_features(&avail, &status, &control, 0);
	if (ret)
		goto out;

	if (!(avail & PAL_PROC_VM_BIT))
		goto out;

	printk(KERN_DEBUG"kvm: Hardware Supports VT\n");

	ret = ia64_pal_vp_env_info(&kvm_vm_buffer_size, &vp_env_info);
	if (ret)
		goto out;
	printk(KERN_DEBUG"kvm: VM Buffer Size:0x%lx\n", kvm_vm_buffer_size);

	if (!(vp_env_info & VP_OPCODE)) {
		printk(KERN_WARNING"kvm: No opcode ability on hardware, "
				"vm_env_info:0x%lx\n", vp_env_info);
	}

	return 1;
out:
	return 0;
}

1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690

/*
 * On SN2, the ITC isn't stable, so copy in fast path code to use the
 * SN2 RTC, replacing the ITC based default verion.
 */
static void kvm_patch_vmm(struct kvm_vmm_info *vmm_info,
			  struct module *module)
{
	unsigned long new_ar, new_ar_sn2;
	unsigned long module_base;

	if (!ia64_platform_is("sn2"))
		return;

	module_base = (unsigned long)module->module_core;

	new_ar = kvm_vmm_base + vmm_info->patch_mov_ar - module_base;
	new_ar_sn2 = kvm_vmm_base + vmm_info->patch_mov_ar_sn2 - module_base;

	printk(KERN_INFO "kvm: Patching ITC emulation to use SGI SN2 RTC "
	       "as source\n");

	/*
	 * Copy the SN2 version of mov_ar into place. They are both
	 * the same size, so 6 bundles is sufficient (6 * 0x10).
	 */
	memcpy((void *)new_ar, (void *)new_ar_sn2, 0x60);
}

1691
static int kvm_relocate_vmm(struct kvm_vmm_info *vmm_info,
1692
			    struct module *module)
1693 1694 1695 1696 1697 1698 1699 1700 1701 1702
{
	unsigned long module_base;
	unsigned long vmm_size;

	unsigned long vmm_offset, func_offset, fdesc_offset;
	struct fdesc *p_fdesc;

	BUG_ON(!module);

	if (!kvm_vmm_base) {
1703
		printk("kvm: kvm area hasn't been initialized yet!!\n");
1704 1705 1706 1707 1708 1709 1710 1711 1712 1713
		return -EFAULT;
	}

	/*Calculate new position of relocated vmm module.*/
	module_base = (unsigned long)module->module_core;
	vmm_size = module->core_size;
	if (unlikely(vmm_size > KVM_VMM_SIZE))
		return -EFAULT;

	memcpy((void *)kvm_vmm_base, (void *)module_base, vmm_size);
1714
	kvm_patch_vmm(vmm_info, module);
1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797
	kvm_flush_icache(kvm_vmm_base, vmm_size);

	/*Recalculate kvm_vmm_info based on new VMM*/
	vmm_offset = vmm_info->vmm_ivt - module_base;
	kvm_vmm_info->vmm_ivt = KVM_VMM_BASE + vmm_offset;
	printk(KERN_DEBUG"kvm: Relocated VMM's IVT Base Addr:%lx\n",
			kvm_vmm_info->vmm_ivt);

	fdesc_offset = (unsigned long)vmm_info->vmm_entry - module_base;
	kvm_vmm_info->vmm_entry = (kvm_vmm_entry *)(KVM_VMM_BASE +
							fdesc_offset);
	func_offset = *(unsigned long *)vmm_info->vmm_entry - module_base;
	p_fdesc = (struct fdesc *)(kvm_vmm_base + fdesc_offset);
	p_fdesc->ip = KVM_VMM_BASE + func_offset;
	p_fdesc->gp = KVM_VMM_BASE+(p_fdesc->gp - module_base);

	printk(KERN_DEBUG"kvm: Relocated VMM's Init Entry Addr:%lx\n",
			KVM_VMM_BASE+func_offset);

	fdesc_offset = (unsigned long)vmm_info->tramp_entry - module_base;
	kvm_vmm_info->tramp_entry = (kvm_tramp_entry *)(KVM_VMM_BASE +
			fdesc_offset);
	func_offset = *(unsigned long *)vmm_info->tramp_entry - module_base;
	p_fdesc = (struct fdesc *)(kvm_vmm_base + fdesc_offset);
	p_fdesc->ip = KVM_VMM_BASE + func_offset;
	p_fdesc->gp = KVM_VMM_BASE + (p_fdesc->gp - module_base);

	kvm_vmm_gp = p_fdesc->gp;

	printk(KERN_DEBUG"kvm: Relocated VMM's Entry IP:%p\n",
						kvm_vmm_info->vmm_entry);
	printk(KERN_DEBUG"kvm: Relocated VMM's Trampoline Entry IP:0x%lx\n",
						KVM_VMM_BASE + func_offset);

	return 0;
}

int kvm_arch_init(void *opaque)
{
	int r;
	struct kvm_vmm_info *vmm_info = (struct kvm_vmm_info *)opaque;

	if (!vti_cpu_has_kvm_support()) {
		printk(KERN_ERR "kvm: No Hardware Virtualization Support!\n");
		r = -EOPNOTSUPP;
		goto out;
	}

	if (kvm_vmm_info) {
		printk(KERN_ERR "kvm: Already loaded VMM module!\n");
		r = -EEXIST;
		goto out;
	}

	r = -ENOMEM;
	kvm_vmm_info = kzalloc(sizeof(struct kvm_vmm_info), GFP_KERNEL);
	if (!kvm_vmm_info)
		goto out;

	if (kvm_alloc_vmm_area())
		goto out_free0;

	r = kvm_relocate_vmm(vmm_info, vmm_info->module);
	if (r)
		goto out_free1;

	return 0;

out_free1:
	kvm_free_vmm_area();
out_free0:
	kfree(kvm_vmm_info);
out:
	return r;
}

void kvm_arch_exit(void)
{
	kvm_free_vmm_area();
	kfree(kvm_vmm_info);
	kvm_vmm_info = NULL;
}

1798 1799
static void kvm_ia64_sync_dirty_log(struct kvm *kvm,
				    struct kvm_memory_slot *memslot)
1800
{
1801
	int i;
1802 1803
	long base;
	unsigned long n;
1804 1805
	unsigned long *dirty_bitmap = (unsigned long *)(kvm->arch.vm_base +
			offsetof(struct kvm_vm_data, kvm_mem_dirty_log));
1806

1807
	n = kvm_dirty_bitmap_bytes(memslot);
1808 1809
	base = memslot->base_gfn / BITS_PER_LONG;

1810
	spin_lock(&kvm->arch.dirty_log_lock);
1811 1812 1813 1814
	for (i = 0; i < n/sizeof(long); ++i) {
		memslot->dirty_bitmap[i] = dirty_bitmap[base + i];
		dirty_bitmap[base + i] = 0;
	}
1815
	spin_unlock(&kvm->arch.dirty_log_lock);
1816 1817 1818 1819 1820 1821
}

int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
		struct kvm_dirty_log *log)
{
	int r;
1822
	unsigned long n;
1823 1824 1825
	struct kvm_memory_slot *memslot;
	int is_dirty = 0;

1826
	mutex_lock(&kvm->slots_lock);
1827

1828 1829 1830 1831 1832 1833 1834
	r = -EINVAL;
	if (log->slot >= KVM_MEMORY_SLOTS)
		goto out;

	memslot = &kvm->memslots->memslots[log->slot];
	r = -ENOENT;
	if (!memslot->dirty_bitmap)
1835 1836
		goto out;

1837
	kvm_ia64_sync_dirty_log(kvm, memslot);
1838 1839 1840 1841 1842 1843 1844
	r = kvm_get_dirty_log(kvm, log, &is_dirty);
	if (r)
		goto out;

	/* If nothing is dirty, don't bother messing with page tables. */
	if (is_dirty) {
		kvm_flush_remote_tlbs(kvm);
1845
		n = kvm_dirty_bitmap_bytes(memslot);
1846 1847 1848 1849
		memset(memslot->dirty_bitmap, 0, n);
	}
	r = 0;
out:
1850
	mutex_unlock(&kvm->slots_lock);
1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864
	return r;
}

int kvm_arch_hardware_setup(void)
{
	return 0;
}

void kvm_arch_hardware_unsetup(void)
{
}

void kvm_vcpu_kick(struct kvm_vcpu *vcpu)
{
1865 1866
	int me;
	int cpu = vcpu->cpu;
1867 1868 1869 1870

	if (waitqueue_active(&vcpu->wq))
		wake_up_interruptible(&vcpu->wq);

1871 1872 1873 1874
	me = get_cpu();
	if (cpu != me && (unsigned) cpu < nr_cpu_ids && cpu_online(cpu))
		if (!test_and_set_bit(KVM_REQ_KICK, &vcpu->requests))
			smp_send_reschedule(cpu);
1875
	put_cpu();
1876 1877
}

1878
int kvm_apic_set_irq(struct kvm_vcpu *vcpu, struct kvm_lapic_irq *irq)
1879
{
1880
	return __apic_accept_irq(vcpu, irq->vector);
1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892
}

int kvm_apic_match_physical_addr(struct kvm_lapic *apic, u16 dest)
{
	return apic->vcpu->vcpu_id == dest;
}

int kvm_apic_match_logical_addr(struct kvm_lapic *apic, u8 mda)
{
	return 0;
}

1893
int kvm_apic_compare_prio(struct kvm_vcpu *vcpu1, struct kvm_vcpu *vcpu2)
1894
{
1895
	return vcpu1->arch.xtp - vcpu2->arch.xtp;
1896 1897
}

1898 1899 1900
int kvm_apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic *source,
		int short_hand, int dest, int dest_mode)
{
1901
	struct kvm_lapic *target = vcpu->arch.apic;
1902 1903 1904 1905 1906
	return (dest_mode == 0) ?
		kvm_apic_match_physical_addr(target, dest) :
		kvm_apic_match_logical_addr(target, dest);
}

1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935
static int find_highest_bits(int *dat)
{
	u32  bits, bitnum;
	int i;

	/* loop for all 256 bits */
	for (i = 7; i >= 0 ; i--) {
		bits = dat[i];
		if (bits) {
			bitnum = fls(bits);
			return i * 32 + bitnum - 1;
		}
	}

	return -1;
}

int kvm_highest_pending_irq(struct kvm_vcpu *vcpu)
{
    struct vpd *vpd = to_host(vcpu->kvm, vcpu->arch.vpd);

    if (vpd->irr[0] & (1UL << NMI_VECTOR))
		return NMI_VECTOR;
    if (vpd->irr[0] & (1UL << ExtINT_VECTOR))
		return ExtINT_VECTOR;

    return find_highest_bits((int *)&vpd->irr[0]);
}

1936 1937
int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
{
1938
	return vcpu->arch.timer_fired;
1939 1940
}

1941 1942
int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
{
1943 1944
	return (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE) ||
		(kvm_highest_pending_irq(vcpu) != -1);
1945
}
1946 1947 1948 1949

int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
1950 1951 1952 1953 1954 1955 1956 1957 1958 1959
	mp_state->mp_state = vcpu->arch.mp_state;
	return 0;
}

static int vcpu_reset(struct kvm_vcpu *vcpu)
{
	int r;
	long psr;
	local_irq_save(psr);
	r = kvm_insert_vmm_mapping(vcpu);
1960
	local_irq_restore(psr);
1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973
	if (r)
		goto fail;

	vcpu->arch.launched = 0;
	kvm_arch_vcpu_uninit(vcpu);
	r = kvm_arch_vcpu_init(vcpu);
	if (r)
		goto fail;

	kvm_purge_vmm_mapping(vcpu);
	r = 0;
fail:
	return r;
1974 1975 1976 1977 1978
}

int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
1979 1980 1981 1982 1983 1984
	int r = 0;

	vcpu->arch.mp_state = mp_state->mp_state;
	if (vcpu->arch.mp_state == KVM_MP_STATE_UNINITIALIZED)
		r = vcpu_reset(vcpu);
	return r;
1985
}