kvm-s390.c 70.2 KB
Newer Older
1
/*
2
 * hosting zSeries kernel virtual machines
3
 *
4
 * Copyright IBM Corp. 2008, 2009
5 6 7 8 9 10 11 12
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License (version 2 only)
 * as published by the Free Software Foundation.
 *
 *    Author(s): Carsten Otte <cotte@de.ibm.com>
 *               Christian Borntraeger <borntraeger@de.ibm.com>
 *               Heiko Carstens <heiko.carstens@de.ibm.com>
13
 *               Christian Ehrhardt <ehrhardt@de.ibm.com>
14
 *               Jason J. Herne <jjherne@us.ibm.com>
15 16 17 18 19
 */

#include <linux/compiler.h>
#include <linux/err.h>
#include <linux/fs.h>
20
#include <linux/hrtimer.h>
21 22 23 24
#include <linux/init.h>
#include <linux/kvm.h>
#include <linux/kvm_host.h>
#include <linux/module.h>
25
#include <linux/random.h>
26
#include <linux/slab.h>
27
#include <linux/timer.h>
28
#include <linux/vmalloc.h>
29
#include <asm/asm-offsets.h>
30
#include <asm/lowcore.h>
31
#include <asm/etr.h>
32
#include <asm/pgtable.h>
33
#include <asm/nmi.h>
34
#include <asm/switch_to.h>
35
#include <asm/isc.h>
36
#include <asm/sclp.h>
37
#include "kvm-s390.h"
38 39
#include "gaccess.h"

40 41 42 43
#define KMSG_COMPONENT "kvm-s390"
#undef pr_fmt
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt

44 45
#define CREATE_TRACE_POINTS
#include "trace.h"
46
#include "trace-s390.h"
47

48
#define MEM_OP_MAX_SIZE 65536	/* Maximum transfer size for KVM_S390_MEM_OP */
49 50 51
#define LOCAL_IRQS 32
#define VCPU_IRQS_MAX_BUF (sizeof(struct kvm_s390_irq) * \
			   (KVM_MAX_VCPUS + LOCAL_IRQS))
52

53 54 55 56
#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU

struct kvm_stats_debugfs_item debugfs_entries[] = {
	{ "userspace_handled", VCPU_STAT(exit_userspace) },
57
	{ "exit_null", VCPU_STAT(exit_null) },
58 59 60 61
	{ "exit_validity", VCPU_STAT(exit_validity) },
	{ "exit_stop_request", VCPU_STAT(exit_stop_request) },
	{ "exit_external_request", VCPU_STAT(exit_external_request) },
	{ "exit_external_interrupt", VCPU_STAT(exit_external_interrupt) },
62 63 64
	{ "exit_instruction", VCPU_STAT(exit_instruction) },
	{ "exit_program_interruption", VCPU_STAT(exit_program_interruption) },
	{ "exit_instr_and_program_int", VCPU_STAT(exit_instr_and_program) },
65
	{ "halt_successful_poll", VCPU_STAT(halt_successful_poll) },
66
	{ "halt_attempted_poll", VCPU_STAT(halt_attempted_poll) },
67
	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
68
	{ "instruction_lctlg", VCPU_STAT(instruction_lctlg) },
69
	{ "instruction_lctl", VCPU_STAT(instruction_lctl) },
70 71
	{ "instruction_stctl", VCPU_STAT(instruction_stctl) },
	{ "instruction_stctg", VCPU_STAT(instruction_stctg) },
72
	{ "deliver_emergency_signal", VCPU_STAT(deliver_emergency_signal) },
73
	{ "deliver_external_call", VCPU_STAT(deliver_external_call) },
74 75 76 77 78 79 80
	{ "deliver_service_signal", VCPU_STAT(deliver_service_signal) },
	{ "deliver_virtio_interrupt", VCPU_STAT(deliver_virtio_interrupt) },
	{ "deliver_stop_signal", VCPU_STAT(deliver_stop_signal) },
	{ "deliver_prefix_signal", VCPU_STAT(deliver_prefix_signal) },
	{ "deliver_restart_signal", VCPU_STAT(deliver_restart_signal) },
	{ "deliver_program_interruption", VCPU_STAT(deliver_program_int) },
	{ "exit_wait_state", VCPU_STAT(exit_wait_state) },
81
	{ "instruction_pfmf", VCPU_STAT(instruction_pfmf) },
82 83 84 85 86
	{ "instruction_stidp", VCPU_STAT(instruction_stidp) },
	{ "instruction_spx", VCPU_STAT(instruction_spx) },
	{ "instruction_stpx", VCPU_STAT(instruction_stpx) },
	{ "instruction_stap", VCPU_STAT(instruction_stap) },
	{ "instruction_storage_key", VCPU_STAT(instruction_storage_key) },
87
	{ "instruction_ipte_interlock", VCPU_STAT(instruction_ipte_interlock) },
88 89
	{ "instruction_stsch", VCPU_STAT(instruction_stsch) },
	{ "instruction_chsc", VCPU_STAT(instruction_chsc) },
90
	{ "instruction_essa", VCPU_STAT(instruction_essa) },
91 92
	{ "instruction_stsi", VCPU_STAT(instruction_stsi) },
	{ "instruction_stfl", VCPU_STAT(instruction_stfl) },
93
	{ "instruction_tprot", VCPU_STAT(instruction_tprot) },
94
	{ "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) },
95
	{ "instruction_sigp_sense_running", VCPU_STAT(instruction_sigp_sense_running) },
96
	{ "instruction_sigp_external_call", VCPU_STAT(instruction_sigp_external_call) },
97
	{ "instruction_sigp_emergency", VCPU_STAT(instruction_sigp_emergency) },
98 99
	{ "instruction_sigp_cond_emergency", VCPU_STAT(instruction_sigp_cond_emergency) },
	{ "instruction_sigp_start", VCPU_STAT(instruction_sigp_start) },
100
	{ "instruction_sigp_stop", VCPU_STAT(instruction_sigp_stop) },
101 102
	{ "instruction_sigp_stop_store_status", VCPU_STAT(instruction_sigp_stop_store_status) },
	{ "instruction_sigp_store_status", VCPU_STAT(instruction_sigp_store_status) },
103
	{ "instruction_sigp_store_adtl_status", VCPU_STAT(instruction_sigp_store_adtl_status) },
104 105 106
	{ "instruction_sigp_set_arch", VCPU_STAT(instruction_sigp_arch) },
	{ "instruction_sigp_set_prefix", VCPU_STAT(instruction_sigp_prefix) },
	{ "instruction_sigp_restart", VCPU_STAT(instruction_sigp_restart) },
107 108 109
	{ "instruction_sigp_cpu_reset", VCPU_STAT(instruction_sigp_cpu_reset) },
	{ "instruction_sigp_init_cpu_reset", VCPU_STAT(instruction_sigp_init_cpu_reset) },
	{ "instruction_sigp_unknown", VCPU_STAT(instruction_sigp_unknown) },
110
	{ "diagnose_10", VCPU_STAT(diagnose_10) },
111
	{ "diagnose_44", VCPU_STAT(diagnose_44) },
112
	{ "diagnose_9c", VCPU_STAT(diagnose_9c) },
113 114 115
	{ "diagnose_258", VCPU_STAT(diagnose_258) },
	{ "diagnose_308", VCPU_STAT(diagnose_308) },
	{ "diagnose_500", VCPU_STAT(diagnose_500) },
116 117 118
	{ NULL }
};

119 120
/* upper facilities limit for kvm */
unsigned long kvm_s390_fac_list_mask[] = {
121
	0xffe6fffbfcfdfc40UL,
122
	0x005e800000000000UL,
123
};
124

125
unsigned long kvm_s390_fac_list_mask_size(void)
126
{
127 128
	BUILD_BUG_ON(ARRAY_SIZE(kvm_s390_fac_list_mask) > S390_ARCH_FAC_MASK_SIZE_U64);
	return ARRAY_SIZE(kvm_s390_fac_list_mask);
129 130
}

131
static struct gmap_notifier gmap_notifier;
132
debug_info_t *kvm_s390_dbf;
133

134
/* Section: not file related */
135
int kvm_arch_hardware_enable(void)
136 137
{
	/* every s390 is virtualization enabled ;-) */
138
	return 0;
139 140
}

141 142
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long address);

143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169
/*
 * This callback is executed during stop_machine(). All CPUs are therefore
 * temporarily stopped. In order not to change guest behavior, we have to
 * disable preemption whenever we touch the epoch of kvm and the VCPUs,
 * so a CPU won't be stopped while calculating with the epoch.
 */
static int kvm_clock_sync(struct notifier_block *notifier, unsigned long val,
			  void *v)
{
	struct kvm *kvm;
	struct kvm_vcpu *vcpu;
	int i;
	unsigned long long *delta = v;

	list_for_each_entry(kvm, &vm_list, vm_list) {
		kvm->arch.epoch -= *delta;
		kvm_for_each_vcpu(i, vcpu, kvm) {
			vcpu->arch.sie_block->epoch -= *delta;
		}
	}
	return NOTIFY_OK;
}

static struct notifier_block kvm_clock_notifier = {
	.notifier_call = kvm_clock_sync,
};

170 171
int kvm_arch_hardware_setup(void)
{
172 173
	gmap_notifier.notifier_call = kvm_gmap_notifier;
	gmap_register_ipte_notifier(&gmap_notifier);
174 175
	atomic_notifier_chain_register(&s390_epoch_delta_notifier,
				       &kvm_clock_notifier);
176 177 178 179 180
	return 0;
}

void kvm_arch_hardware_unsetup(void)
{
181
	gmap_unregister_ipte_notifier(&gmap_notifier);
182 183
	atomic_notifier_chain_unregister(&s390_epoch_delta_notifier,
					 &kvm_clock_notifier);
184 185 186 187
}

int kvm_arch_init(void *opaque)
{
188 189 190 191 192 193 194 195 196
	kvm_s390_dbf = debug_register("kvm-trace", 32, 1, 7 * sizeof(long));
	if (!kvm_s390_dbf)
		return -ENOMEM;

	if (debug_register_view(kvm_s390_dbf, &debug_sprintf_view)) {
		debug_unregister(kvm_s390_dbf);
		return -ENOMEM;
	}

197 198
	/* Register floating interrupt controller interface. */
	return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
199 200
}

201 202 203 204 205
void kvm_arch_exit(void)
{
	debug_unregister(kvm_s390_dbf);
}

206 207 208 209 210 211 212 213 214
/* Section: device related */
long kvm_arch_dev_ioctl(struct file *filp,
			unsigned int ioctl, unsigned long arg)
{
	if (ioctl == KVM_S390_ENABLE_SIE)
		return s390_enable_sie();
	return -EINVAL;
}

215
int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
216
{
217 218
	int r;

219
	switch (ext) {
220
	case KVM_CAP_S390_PSW:
221
	case KVM_CAP_S390_GMAP:
222
	case KVM_CAP_SYNC_MMU:
223 224 225
#ifdef CONFIG_KVM_S390_UCONTROL
	case KVM_CAP_S390_UCONTROL:
#endif
226
	case KVM_CAP_ASYNC_PF:
227
	case KVM_CAP_SYNC_REGS:
228
	case KVM_CAP_ONE_REG:
229
	case KVM_CAP_ENABLE_CAP:
230
	case KVM_CAP_S390_CSS_SUPPORT:
C
Cornelia Huck 已提交
231
	case KVM_CAP_IOEVENTFD:
232
	case KVM_CAP_DEVICE_CTRL:
233
	case KVM_CAP_ENABLE_CAP_VM:
234
	case KVM_CAP_S390_IRQCHIP:
235
	case KVM_CAP_VM_ATTRIBUTES:
236
	case KVM_CAP_MP_STATE:
237
	case KVM_CAP_S390_INJECT_IRQ:
238
	case KVM_CAP_S390_USER_SIGP:
239
	case KVM_CAP_S390_USER_STSI:
240
	case KVM_CAP_S390_SKEYS:
241
	case KVM_CAP_S390_IRQ_STATE:
242 243
		r = 1;
		break;
244 245 246
	case KVM_CAP_S390_MEM_OP:
		r = MEM_OP_MAX_SIZE;
		break;
247 248 249 250
	case KVM_CAP_NR_VCPUS:
	case KVM_CAP_MAX_VCPUS:
		r = KVM_MAX_VCPUS;
		break;
251 252 253
	case KVM_CAP_NR_MEMSLOTS:
		r = KVM_USER_MEM_SLOTS;
		break;
254
	case KVM_CAP_S390_COW:
255
		r = MACHINE_HAS_ESOP;
256
		break;
257 258 259
	case KVM_CAP_S390_VECTOR_REGISTERS:
		r = MACHINE_HAS_VX;
		break;
260
	default:
261
		r = 0;
262
	}
263
	return r;
264 265
}

266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284
static void kvm_s390_sync_dirty_log(struct kvm *kvm,
					struct kvm_memory_slot *memslot)
{
	gfn_t cur_gfn, last_gfn;
	unsigned long address;
	struct gmap *gmap = kvm->arch.gmap;

	down_read(&gmap->mm->mmap_sem);
	/* Loop over all guest pages */
	last_gfn = memslot->base_gfn + memslot->npages;
	for (cur_gfn = memslot->base_gfn; cur_gfn <= last_gfn; cur_gfn++) {
		address = gfn_to_hva_memslot(memslot, cur_gfn);

		if (gmap_test_and_clear_dirty(address, gmap))
			mark_page_dirty(kvm, cur_gfn);
	}
	up_read(&gmap->mm->mmap_sem);
}

285 286 287 288 289 290 291
/* Section: vm related */
/*
 * Get (and clear) the dirty memory log for a memory slot.
 */
int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
			       struct kvm_dirty_log *log)
{
292 293
	int r;
	unsigned long n;
294
	struct kvm_memslots *slots;
295 296 297 298 299 300 301 302 303
	struct kvm_memory_slot *memslot;
	int is_dirty = 0;

	mutex_lock(&kvm->slots_lock);

	r = -EINVAL;
	if (log->slot >= KVM_USER_MEM_SLOTS)
		goto out;

304 305
	slots = kvm_memslots(kvm);
	memslot = id_to_memslot(slots, log->slot);
306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323
	r = -ENOENT;
	if (!memslot->dirty_bitmap)
		goto out;

	kvm_s390_sync_dirty_log(kvm, memslot);
	r = kvm_get_dirty_log(kvm, log, &is_dirty);
	if (r)
		goto out;

	/* Clear the dirty log */
	if (is_dirty) {
		n = kvm_dirty_bitmap_bytes(memslot);
		memset(memslot->dirty_bitmap, 0, n);
	}
	r = 0;
out:
	mutex_unlock(&kvm->slots_lock);
	return r;
324 325
}

326 327 328 329 330 331 332 333
static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap)
{
	int r;

	if (cap->flags)
		return -EINVAL;

	switch (cap->cap) {
334
	case KVM_CAP_S390_IRQCHIP:
335
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_IRQCHIP");
336 337 338
		kvm->arch.use_irqchip = 1;
		r = 0;
		break;
339
	case KVM_CAP_S390_USER_SIGP:
340
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_SIGP");
341 342 343
		kvm->arch.user_sigp = 1;
		r = 0;
		break;
344
	case KVM_CAP_S390_VECTOR_REGISTERS:
345 346 347 348
		mutex_lock(&kvm->lock);
		if (atomic_read(&kvm->online_vcpus)) {
			r = -EBUSY;
		} else if (MACHINE_HAS_VX) {
349 350 351 352 353
			set_kvm_facility(kvm->arch.model.fac->mask, 129);
			set_kvm_facility(kvm->arch.model.fac->list, 129);
			r = 0;
		} else
			r = -EINVAL;
354
		mutex_unlock(&kvm->lock);
355 356
		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_VECTOR_REGISTERS %s",
			 r ? "(not available)" : "(success)");
357
		break;
358
	case KVM_CAP_S390_USER_STSI:
359
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_STSI");
360 361 362
		kvm->arch.user_stsi = 1;
		r = 0;
		break;
363 364 365 366 367 368 369
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

370 371 372 373 374 375 376
static int kvm_s390_get_mem_control(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret;

	switch (attr->attr) {
	case KVM_S390_VM_MEM_LIMIT_SIZE:
		ret = 0;
377 378
		VM_EVENT(kvm, 3, "QUERY: max guest memory: %lu bytes",
			 kvm->arch.gmap->asce_end);
379 380 381 382 383 384 385 386 387 388 389
		if (put_user(kvm->arch.gmap->asce_end, (u64 __user *)attr->addr))
			ret = -EFAULT;
		break;
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *attr)
390 391 392 393 394
{
	int ret;
	unsigned int idx;
	switch (attr->attr) {
	case KVM_S390_VM_MEM_ENABLE_CMMA:
395 396 397 398 399
		/* enable CMMA only for z10 and later (EDAT_1) */
		ret = -EINVAL;
		if (!MACHINE_IS_LPAR || !MACHINE_HAS_EDAT1)
			break;

400
		ret = -EBUSY;
401
		VM_EVENT(kvm, 3, "%s", "ENABLE: CMMA support");
402 403 404 405 406 407 408 409
		mutex_lock(&kvm->lock);
		if (atomic_read(&kvm->online_vcpus) == 0) {
			kvm->arch.use_cmma = 1;
			ret = 0;
		}
		mutex_unlock(&kvm->lock);
		break;
	case KVM_S390_VM_MEM_CLR_CMMA:
410 411 412 413
		ret = -EINVAL;
		if (!kvm->arch.use_cmma)
			break;

414
		VM_EVENT(kvm, 3, "%s", "RESET: CMMA states");
415 416
		mutex_lock(&kvm->lock);
		idx = srcu_read_lock(&kvm->srcu);
417
		s390_reset_cmma(kvm->arch.gmap->mm);
418 419 420 421
		srcu_read_unlock(&kvm->srcu, idx);
		mutex_unlock(&kvm->lock);
		ret = 0;
		break;
422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449
	case KVM_S390_VM_MEM_LIMIT_SIZE: {
		unsigned long new_limit;

		if (kvm_is_ucontrol(kvm))
			return -EINVAL;

		if (get_user(new_limit, (u64 __user *)attr->addr))
			return -EFAULT;

		if (new_limit > kvm->arch.gmap->asce_end)
			return -E2BIG;

		ret = -EBUSY;
		mutex_lock(&kvm->lock);
		if (atomic_read(&kvm->online_vcpus) == 0) {
			/* gmap_alloc will round the limit up */
			struct gmap *new = gmap_alloc(current->mm, new_limit);

			if (!new) {
				ret = -ENOMEM;
			} else {
				gmap_free(kvm->arch.gmap);
				new->private = kvm;
				kvm->arch.gmap = new;
				ret = 0;
			}
		}
		mutex_unlock(&kvm->lock);
450
		VM_EVENT(kvm, 3, "SET: max guest memory: %lu bytes", new_limit);
451 452
		break;
	}
453 454 455 456 457 458 459
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

460 461 462 463 464 465 466
static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu);

static int kvm_s390_vm_set_crypto(struct kvm *kvm, struct kvm_device_attr *attr)
{
	struct kvm_vcpu *vcpu;
	int i;

467
	if (!test_kvm_facility(kvm, 76))
468 469 470 471 472 473 474 475 476
		return -EINVAL;

	mutex_lock(&kvm->lock);
	switch (attr->attr) {
	case KVM_S390_VM_CRYPTO_ENABLE_AES_KW:
		get_random_bytes(
			kvm->arch.crypto.crycb->aes_wrapping_key_mask,
			sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask));
		kvm->arch.crypto.aes_kw = 1;
477
		VM_EVENT(kvm, 3, "%s", "ENABLE: AES keywrapping support");
478 479 480 481 482 483
		break;
	case KVM_S390_VM_CRYPTO_ENABLE_DEA_KW:
		get_random_bytes(
			kvm->arch.crypto.crycb->dea_wrapping_key_mask,
			sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask));
		kvm->arch.crypto.dea_kw = 1;
484
		VM_EVENT(kvm, 3, "%s", "ENABLE: DEA keywrapping support");
485 486 487 488 489
		break;
	case KVM_S390_VM_CRYPTO_DISABLE_AES_KW:
		kvm->arch.crypto.aes_kw = 0;
		memset(kvm->arch.crypto.crycb->aes_wrapping_key_mask, 0,
			sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask));
490
		VM_EVENT(kvm, 3, "%s", "DISABLE: AES keywrapping support");
491 492 493 494 495
		break;
	case KVM_S390_VM_CRYPTO_DISABLE_DEA_KW:
		kvm->arch.crypto.dea_kw = 0;
		memset(kvm->arch.crypto.crycb->dea_wrapping_key_mask, 0,
			sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask));
496
		VM_EVENT(kvm, 3, "%s", "DISABLE: DEA keywrapping support");
497 498 499 500 501 502 503 504 505 506 507 508 509 510
		break;
	default:
		mutex_unlock(&kvm->lock);
		return -ENXIO;
	}

	kvm_for_each_vcpu(i, vcpu, kvm) {
		kvm_s390_vcpu_crypto_setup(vcpu);
		exit_sie(vcpu);
	}
	mutex_unlock(&kvm->lock);
	return 0;
}

511 512 513 514 515 516 517 518 519 520
static int kvm_s390_set_tod_high(struct kvm *kvm, struct kvm_device_attr *attr)
{
	u8 gtod_high;

	if (copy_from_user(&gtod_high, (void __user *)attr->addr,
					   sizeof(gtod_high)))
		return -EFAULT;

	if (gtod_high != 0)
		return -EINVAL;
521
	VM_EVENT(kvm, 3, "SET: TOD extension: 0x%x", gtod_high);
522 523 524 525 526 527

	return 0;
}

static int kvm_s390_set_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
528
	u64 gtod;
529 530 531 532

	if (copy_from_user(&gtod, (void __user *)attr->addr, sizeof(gtod)))
		return -EFAULT;

533
	kvm_s390_set_tod_clock(kvm, gtod);
534
	VM_EVENT(kvm, 3, "SET: TOD base: 0x%llx", gtod);
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
	return 0;
}

static int kvm_s390_set_tod(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret;

	if (attr->flags)
		return -EINVAL;

	switch (attr->attr) {
	case KVM_S390_VM_TOD_HIGH:
		ret = kvm_s390_set_tod_high(kvm, attr);
		break;
	case KVM_S390_VM_TOD_LOW:
		ret = kvm_s390_set_tod_low(kvm, attr);
		break;
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

static int kvm_s390_get_tod_high(struct kvm *kvm, struct kvm_device_attr *attr)
{
	u8 gtod_high = 0;

	if (copy_to_user((void __user *)attr->addr, &gtod_high,
					 sizeof(gtod_high)))
		return -EFAULT;
566
	VM_EVENT(kvm, 3, "QUERY: TOD extension: 0x%x", gtod_high);
567 568 569 570 571 572

	return 0;
}

static int kvm_s390_get_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
573
	u64 gtod;
574

575
	gtod = kvm_s390_get_tod_clock_fast(kvm);
576 577
	if (copy_to_user((void __user *)attr->addr, &gtod, sizeof(gtod)))
		return -EFAULT;
578
	VM_EVENT(kvm, 3, "QUERY: TOD base: 0x%llx", gtod);
579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603

	return 0;
}

static int kvm_s390_get_tod(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret;

	if (attr->flags)
		return -EINVAL;

	switch (attr->attr) {
	case KVM_S390_VM_TOD_HIGH:
		ret = kvm_s390_get_tod_high(kvm, attr);
		break;
	case KVM_S390_VM_TOD_LOW:
		ret = kvm_s390_get_tod_low(kvm, attr);
		break;
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623
static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_processor *proc;
	int ret = 0;

	mutex_lock(&kvm->lock);
	if (atomic_read(&kvm->online_vcpus)) {
		ret = -EBUSY;
		goto out;
	}
	proc = kzalloc(sizeof(*proc), GFP_KERNEL);
	if (!proc) {
		ret = -ENOMEM;
		goto out;
	}
	if (!copy_from_user(proc, (void __user *)attr->addr,
			    sizeof(*proc))) {
		memcpy(&kvm->arch.model.cpu_id, &proc->cpuid,
		       sizeof(struct cpuid));
		kvm->arch.model.ibc = proc->ibc;
624
		memcpy(kvm->arch.model.fac->list, proc->fac_list,
625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657
		       S390_ARCH_FAC_LIST_SIZE_BYTE);
	} else
		ret = -EFAULT;
	kfree(proc);
out:
	mutex_unlock(&kvm->lock);
	return ret;
}

static int kvm_s390_set_cpu_model(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret = -ENXIO;

	switch (attr->attr) {
	case KVM_S390_VM_CPU_PROCESSOR:
		ret = kvm_s390_set_processor(kvm, attr);
		break;
	}
	return ret;
}

static int kvm_s390_get_processor(struct kvm *kvm, struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_processor *proc;
	int ret = 0;

	proc = kzalloc(sizeof(*proc), GFP_KERNEL);
	if (!proc) {
		ret = -ENOMEM;
		goto out;
	}
	memcpy(&proc->cpuid, &kvm->arch.model.cpu_id, sizeof(struct cpuid));
	proc->ibc = kvm->arch.model.ibc;
658
	memcpy(&proc->fac_list, kvm->arch.model.fac->list, S390_ARCH_FAC_LIST_SIZE_BYTE);
659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676
	if (copy_to_user((void __user *)attr->addr, proc, sizeof(*proc)))
		ret = -EFAULT;
	kfree(proc);
out:
	return ret;
}

static int kvm_s390_get_machine(struct kvm *kvm, struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_machine *mach;
	int ret = 0;

	mach = kzalloc(sizeof(*mach), GFP_KERNEL);
	if (!mach) {
		ret = -ENOMEM;
		goto out;
	}
	get_cpu_id((struct cpuid *) &mach->cpuid);
677
	mach->ibc = sclp.ibc;
678 679
	memcpy(&mach->fac_mask, kvm->arch.model.fac->mask,
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
680
	memcpy((unsigned long *)&mach->fac_list, S390_lowcore.stfle_fac_list,
681
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703
	if (copy_to_user((void __user *)attr->addr, mach, sizeof(*mach)))
		ret = -EFAULT;
	kfree(mach);
out:
	return ret;
}

static int kvm_s390_get_cpu_model(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret = -ENXIO;

	switch (attr->attr) {
	case KVM_S390_VM_CPU_PROCESSOR:
		ret = kvm_s390_get_processor(kvm, attr);
		break;
	case KVM_S390_VM_CPU_MACHINE:
		ret = kvm_s390_get_machine(kvm, attr);
		break;
	}
	return ret;
}

704 705 706 707 708
static int kvm_s390_vm_set_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret;

	switch (attr->group) {
709
	case KVM_S390_VM_MEM_CTRL:
710
		ret = kvm_s390_set_mem_control(kvm, attr);
711
		break;
712 713 714
	case KVM_S390_VM_TOD:
		ret = kvm_s390_set_tod(kvm, attr);
		break;
715 716 717
	case KVM_S390_VM_CPU_MODEL:
		ret = kvm_s390_set_cpu_model(kvm, attr);
		break;
718 719 720
	case KVM_S390_VM_CRYPTO:
		ret = kvm_s390_vm_set_crypto(kvm, attr);
		break;
721 722 723 724 725 726 727 728 729 730
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

static int kvm_s390_vm_get_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
731 732 733 734 735 736
	int ret;

	switch (attr->group) {
	case KVM_S390_VM_MEM_CTRL:
		ret = kvm_s390_get_mem_control(kvm, attr);
		break;
737 738 739
	case KVM_S390_VM_TOD:
		ret = kvm_s390_get_tod(kvm, attr);
		break;
740 741 742
	case KVM_S390_VM_CPU_MODEL:
		ret = kvm_s390_get_cpu_model(kvm, attr);
		break;
743 744 745 746 747 748
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
749 750 751 752 753 754 755
}

static int kvm_s390_vm_has_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret;

	switch (attr->group) {
756 757 758 759
	case KVM_S390_VM_MEM_CTRL:
		switch (attr->attr) {
		case KVM_S390_VM_MEM_ENABLE_CMMA:
		case KVM_S390_VM_MEM_CLR_CMMA:
760
		case KVM_S390_VM_MEM_LIMIT_SIZE:
761 762 763 764 765 766 767
			ret = 0;
			break;
		default:
			ret = -ENXIO;
			break;
		}
		break;
768 769 770 771 772 773 774 775 776 777 778
	case KVM_S390_VM_TOD:
		switch (attr->attr) {
		case KVM_S390_VM_TOD_LOW:
		case KVM_S390_VM_TOD_HIGH:
			ret = 0;
			break;
		default:
			ret = -ENXIO;
			break;
		}
		break;
779 780 781 782 783 784 785 786 787 788 789
	case KVM_S390_VM_CPU_MODEL:
		switch (attr->attr) {
		case KVM_S390_VM_CPU_PROCESSOR:
		case KVM_S390_VM_CPU_MACHINE:
			ret = 0;
			break;
		default:
			ret = -ENXIO;
			break;
		}
		break;
790 791 792 793 794 795 796 797 798 799 800 801 802
	case KVM_S390_VM_CRYPTO:
		switch (attr->attr) {
		case KVM_S390_VM_CRYPTO_ENABLE_AES_KW:
		case KVM_S390_VM_CRYPTO_ENABLE_DEA_KW:
		case KVM_S390_VM_CRYPTO_DISABLE_AES_KW:
		case KVM_S390_VM_CRYPTO_DISABLE_DEA_KW:
			ret = 0;
			break;
		default:
			ret = -ENXIO;
			break;
		}
		break;
803 804 805 806 807 808 809 810
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887
static long kvm_s390_get_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
{
	uint8_t *keys;
	uint64_t hva;
	unsigned long curkey;
	int i, r = 0;

	if (args->flags != 0)
		return -EINVAL;

	/* Is this guest using storage keys? */
	if (!mm_use_skey(current->mm))
		return KVM_S390_GET_SKEYS_NONE;

	/* Enforce sane limit on memory allocation */
	if (args->count < 1 || args->count > KVM_S390_SKEYS_MAX)
		return -EINVAL;

	keys = kmalloc_array(args->count, sizeof(uint8_t),
			     GFP_KERNEL | __GFP_NOWARN);
	if (!keys)
		keys = vmalloc(sizeof(uint8_t) * args->count);
	if (!keys)
		return -ENOMEM;

	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
			goto out;
		}

		curkey = get_guest_storage_key(current->mm, hva);
		if (IS_ERR_VALUE(curkey)) {
			r = curkey;
			goto out;
		}
		keys[i] = curkey;
	}

	r = copy_to_user((uint8_t __user *)args->skeydata_addr, keys,
			 sizeof(uint8_t) * args->count);
	if (r)
		r = -EFAULT;
out:
	kvfree(keys);
	return r;
}

static long kvm_s390_set_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
{
	uint8_t *keys;
	uint64_t hva;
	int i, r = 0;

	if (args->flags != 0)
		return -EINVAL;

	/* Enforce sane limit on memory allocation */
	if (args->count < 1 || args->count > KVM_S390_SKEYS_MAX)
		return -EINVAL;

	keys = kmalloc_array(args->count, sizeof(uint8_t),
			     GFP_KERNEL | __GFP_NOWARN);
	if (!keys)
		keys = vmalloc(sizeof(uint8_t) * args->count);
	if (!keys)
		return -ENOMEM;

	r = copy_from_user(keys, (uint8_t __user *)args->skeydata_addr,
			   sizeof(uint8_t) * args->count);
	if (r) {
		r = -EFAULT;
		goto out;
	}

	/* Enable storage key handling for the guest */
888 889 890
	r = s390_enable_skey();
	if (r)
		goto out;
891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914

	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
			goto out;
		}

		/* Lowest order bit is reserved */
		if (keys[i] & 0x01) {
			r = -EINVAL;
			goto out;
		}

		r = set_guest_storage_key(current->mm, hva,
					  (unsigned long)keys[i], 0);
		if (r)
			goto out;
	}
out:
	kvfree(keys);
	return r;
}

915 916 917 918 919
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;
920
	struct kvm_device_attr attr;
921 922 923
	int r;

	switch (ioctl) {
924 925 926 927 928 929 930 931 932
	case KVM_S390_INTERRUPT: {
		struct kvm_s390_interrupt s390int;

		r = -EFAULT;
		if (copy_from_user(&s390int, argp, sizeof(s390int)))
			break;
		r = kvm_s390_inject_vm(kvm, &s390int);
		break;
	}
933 934 935 936 937 938 939 940
	case KVM_ENABLE_CAP: {
		struct kvm_enable_cap cap;
		r = -EFAULT;
		if (copy_from_user(&cap, argp, sizeof(cap)))
			break;
		r = kvm_vm_ioctl_enable_cap(kvm, &cap);
		break;
	}
941 942 943 944 945 946 947
	case KVM_CREATE_IRQCHIP: {
		struct kvm_irq_routing_entry routing;

		r = -EINVAL;
		if (kvm->arch.use_irqchip) {
			/* Set up dummy routing. */
			memset(&routing, 0, sizeof(routing));
948
			r = kvm_set_irq_routing(kvm, &routing, 0, 0);
949 950 951
		}
		break;
	}
952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972
	case KVM_SET_DEVICE_ATTR: {
		r = -EFAULT;
		if (copy_from_user(&attr, (void __user *)arg, sizeof(attr)))
			break;
		r = kvm_s390_vm_set_attr(kvm, &attr);
		break;
	}
	case KVM_GET_DEVICE_ATTR: {
		r = -EFAULT;
		if (copy_from_user(&attr, (void __user *)arg, sizeof(attr)))
			break;
		r = kvm_s390_vm_get_attr(kvm, &attr);
		break;
	}
	case KVM_HAS_DEVICE_ATTR: {
		r = -EFAULT;
		if (copy_from_user(&attr, (void __user *)arg, sizeof(attr)))
			break;
		r = kvm_s390_vm_has_attr(kvm, &attr);
		break;
	}
973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
	case KVM_S390_GET_SKEYS: {
		struct kvm_s390_skeys args;

		r = -EFAULT;
		if (copy_from_user(&args, argp,
				   sizeof(struct kvm_s390_skeys)))
			break;
		r = kvm_s390_get_skeys(kvm, &args);
		break;
	}
	case KVM_S390_SET_SKEYS: {
		struct kvm_s390_skeys args;

		r = -EFAULT;
		if (copy_from_user(&args, argp,
				   sizeof(struct kvm_s390_skeys)))
			break;
		r = kvm_s390_set_skeys(kvm, &args);
		break;
	}
993
	default:
994
		r = -ENOTTY;
995 996 997 998 999
	}

	return r;
}

1000 1001 1002
static int kvm_s390_query_ap_config(u8 *config)
{
	u32 fcn_code = 0x04000000UL;
1003
	u32 cc = 0;
1004

1005
	memset(config, 0, 128);
1006 1007 1008 1009
	asm volatile(
		"lgr 0,%1\n"
		"lgr 2,%2\n"
		".long 0xb2af0000\n"		/* PQAP(QCI) */
1010
		"0: ipm %0\n"
1011
		"srl %0,28\n"
1012 1013 1014
		"1:\n"
		EX_TABLE(0b, 1b)
		: "+r" (cc)
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
		: "r" (fcn_code), "r" (config)
		: "cc", "0", "2", "memory"
	);

	return cc;
}

static int kvm_s390_apxa_installed(void)
{
	u8 config[128];
	int cc;

	if (test_facility(2) && test_facility(12)) {
		cc = kvm_s390_query_ap_config(config);

		if (cc)
			pr_err("PQAP(QCI) failed with cc=%d", cc);
		else
			return config[0] & 0x40;
	}

	return 0;
}

static void kvm_s390_set_crycb_format(struct kvm *kvm)
{
	kvm->arch.crypto.crycbd = (__u32)(unsigned long) kvm->arch.crypto.crycb;

	if (kvm_s390_apxa_installed())
		kvm->arch.crypto.crycbd |= CRYCB_FORMAT2;
	else
		kvm->arch.crypto.crycbd |= CRYCB_FORMAT1;
}

1049 1050 1051 1052 1053 1054
static void kvm_s390_get_cpu_id(struct cpuid *cpu_id)
{
	get_cpu_id(cpu_id);
	cpu_id->version = 0xff;
}

1055 1056
static int kvm_s390_crypto_init(struct kvm *kvm)
{
1057
	if (!test_kvm_facility(kvm, 76))
1058 1059 1060 1061 1062 1063 1064
		return 0;

	kvm->arch.crypto.crycb = kzalloc(sizeof(*kvm->arch.crypto.crycb),
					 GFP_KERNEL | GFP_DMA);
	if (!kvm->arch.crypto.crycb)
		return -ENOMEM;

1065
	kvm_s390_set_crycb_format(kvm);
1066

1067 1068 1069 1070 1071 1072 1073
	/* Enable AES/DEA protected key functions by default */
	kvm->arch.crypto.aes_kw = 1;
	kvm->arch.crypto.dea_kw = 1;
	get_random_bytes(kvm->arch.crypto.crycb->aes_wrapping_key_mask,
			 sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask));
	get_random_bytes(kvm->arch.crypto.crycb->dea_wrapping_key_mask,
			 sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask));
1074

1075 1076 1077
	return 0;
}

1078
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
1079
{
1080
	int i, rc;
1081
	char debug_name[16];
1082
	static unsigned long sca_offset;
1083

1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094
	rc = -EINVAL;
#ifdef CONFIG_KVM_S390_UCONTROL
	if (type & ~KVM_VM_S390_UCONTROL)
		goto out_err;
	if ((type & KVM_VM_S390_UCONTROL) && (!capable(CAP_SYS_ADMIN)))
		goto out_err;
#else
	if (type)
		goto out_err;
#endif

1095 1096
	rc = s390_enable_sie();
	if (rc)
1097
		goto out_err;
1098

1099 1100
	rc = -ENOMEM;

1101 1102
	kvm->arch.sca = (struct sca_block *) get_zeroed_page(GFP_KERNEL);
	if (!kvm->arch.sca)
1103
		goto out_err;
1104
	spin_lock(&kvm_lock);
1105 1106 1107
	sca_offset += 16;
	if (sca_offset + sizeof(struct sca_block) > PAGE_SIZE)
		sca_offset = 0;
1108 1109
	kvm->arch.sca = (struct sca_block *) ((char *) kvm->arch.sca + sca_offset);
	spin_unlock(&kvm_lock);
1110 1111 1112

	sprintf(debug_name, "kvm-%u", current->pid);

1113
	kvm->arch.dbf = debug_register(debug_name, 32, 1, 7 * sizeof(long));
1114
	if (!kvm->arch.dbf)
1115
		goto out_err;
1116

1117 1118 1119
	/*
	 * The architectural maximum amount of facilities is 16 kbit. To store
	 * this amount, 2 kbyte of memory is required. Thus we need a full
1120 1121
	 * page to hold the guest facility list (arch.model.fac->list) and the
	 * facility mask (arch.model.fac->mask). Its address size has to be
1122 1123 1124
	 * 31 bits and word aligned.
	 */
	kvm->arch.model.fac =
1125
		(struct kvm_s390_fac *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
1126
	if (!kvm->arch.model.fac)
1127
		goto out_err;
1128

1129
	/* Populate the facility mask initially. */
1130
	memcpy(kvm->arch.model.fac->mask, S390_lowcore.stfle_fac_list,
1131
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
1132 1133
	for (i = 0; i < S390_ARCH_FAC_LIST_SIZE_U64; i++) {
		if (i < kvm_s390_fac_list_mask_size())
1134
			kvm->arch.model.fac->mask[i] &= kvm_s390_fac_list_mask[i];
1135
		else
1136
			kvm->arch.model.fac->mask[i] = 0UL;
1137 1138
	}

1139 1140 1141 1142
	/* Populate the facility list initially. */
	memcpy(kvm->arch.model.fac->list, kvm->arch.model.fac->mask,
	       S390_ARCH_FAC_LIST_SIZE_BYTE);

1143
	kvm_s390_get_cpu_id(&kvm->arch.model.cpu_id);
1144
	kvm->arch.model.ibc = sclp.ibc & 0x0fff;
1145

1146
	if (kvm_s390_crypto_init(kvm) < 0)
1147
		goto out_err;
1148

1149
	spin_lock_init(&kvm->arch.float_int.lock);
1150 1151
	for (i = 0; i < FIRQ_LIST_COUNT; i++)
		INIT_LIST_HEAD(&kvm->arch.float_int.lists[i]);
1152
	init_waitqueue_head(&kvm->arch.ipte_wq);
1153
	mutex_init(&kvm->arch.ipte_mutex);
1154

1155
	debug_register_view(kvm->arch.dbf, &debug_sprintf_view);
1156
	VM_EVENT(kvm, 3, "vm created with type %lu", type);
1157

1158 1159 1160
	if (type & KVM_VM_S390_UCONTROL) {
		kvm->arch.gmap = NULL;
	} else {
1161
		kvm->arch.gmap = gmap_alloc(current->mm, (1UL << 44) - 1);
1162
		if (!kvm->arch.gmap)
1163
			goto out_err;
1164
		kvm->arch.gmap->private = kvm;
1165
		kvm->arch.gmap->pfault_enabled = 0;
1166
	}
1167 1168

	kvm->arch.css_support = 0;
1169
	kvm->arch.use_irqchip = 0;
1170
	kvm->arch.epoch = 0;
1171

1172
	spin_lock_init(&kvm->arch.start_stop_lock);
1173
	KVM_EVENT(3, "vm 0x%p created by pid %u", kvm, current->pid);
1174

1175
	return 0;
1176
out_err:
1177
	kfree(kvm->arch.crypto.crycb);
1178
	free_page((unsigned long)kvm->arch.model.fac);
1179
	debug_unregister(kvm->arch.dbf);
1180
	free_page((unsigned long)(kvm->arch.sca));
1181
	KVM_EVENT(3, "creation of vm failed: %d", rc);
1182
	return rc;
1183 1184
}

1185 1186 1187
void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
{
	VCPU_EVENT(vcpu, 3, "%s", "free cpu");
1188
	trace_kvm_s390_destroy_vcpu(vcpu->vcpu_id);
1189
	kvm_s390_clear_local_irqs(vcpu);
1190
	kvm_clear_async_pf_completion_queue(vcpu);
C
Carsten Otte 已提交
1191 1192 1193 1194 1195 1196 1197
	if (!kvm_is_ucontrol(vcpu->kvm)) {
		clear_bit(63 - vcpu->vcpu_id,
			  (unsigned long *) &vcpu->kvm->arch.sca->mcn);
		if (vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sda ==
		    (__u64) vcpu->arch.sie_block)
			vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sda = 0;
	}
1198
	smp_mb();
1199 1200 1201 1202

	if (kvm_is_ucontrol(vcpu->kvm))
		gmap_free(vcpu->arch.gmap);

1203
	if (vcpu->kvm->arch.use_cmma)
1204
		kvm_s390_vcpu_unsetup_cmma(vcpu);
1205
	free_page((unsigned long)(vcpu->arch.sie_block));
1206

1207
	kvm_vcpu_uninit(vcpu);
1208
	kmem_cache_free(kvm_vcpu_cache, vcpu);
1209 1210 1211 1212 1213
}

static void kvm_free_vcpus(struct kvm *kvm)
{
	unsigned int i;
1214
	struct kvm_vcpu *vcpu;
1215

1216 1217 1218 1219 1220 1221 1222 1223 1224
	kvm_for_each_vcpu(i, vcpu, kvm)
		kvm_arch_vcpu_destroy(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);
1225 1226
}

1227 1228
void kvm_arch_destroy_vm(struct kvm *kvm)
{
1229
	kvm_free_vcpus(kvm);
1230
	free_page((unsigned long)kvm->arch.model.fac);
1231
	free_page((unsigned long)(kvm->arch.sca));
1232
	debug_unregister(kvm->arch.dbf);
1233
	kfree(kvm->arch.crypto.crycb);
1234 1235
	if (!kvm_is_ucontrol(kvm))
		gmap_free(kvm->arch.gmap);
1236
	kvm_s390_destroy_adapters(kvm);
1237
	kvm_s390_clear_float_irqs(kvm);
1238
	KVM_EVENT(3, "vm 0x%p destroyed", kvm);
1239 1240 1241
}

/* Section: vcpu related */
1242 1243 1244 1245 1246 1247 1248 1249 1250 1251
static int __kvm_ucontrol_vcpu_init(struct kvm_vcpu *vcpu)
{
	vcpu->arch.gmap = gmap_alloc(current->mm, -1UL);
	if (!vcpu->arch.gmap)
		return -ENOMEM;
	vcpu->arch.gmap->private = vcpu->kvm;

	return 0;
}

1252 1253
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
1254 1255
	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
	kvm_clear_async_pf_completion_queue(vcpu);
1256 1257
	vcpu->run->kvm_valid_regs = KVM_SYNC_PREFIX |
				    KVM_SYNC_GPRS |
1258
				    KVM_SYNC_ACRS |
1259 1260 1261
				    KVM_SYNC_CRS |
				    KVM_SYNC_ARCH0 |
				    KVM_SYNC_PFAULT;
1262 1263
	if (test_kvm_facility(vcpu->kvm, 129))
		vcpu->run->kvm_valid_regs |= KVM_SYNC_VRS;
1264 1265 1266 1267

	if (kvm_is_ucontrol(vcpu->kvm))
		return __kvm_ucontrol_vcpu_init(vcpu);

1268 1269 1270
	return 0;
}

1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291
/*
 * Backs up the current FP/VX register save area on a particular
 * destination.  Used to switch between different register save
 * areas.
 */
static inline void save_fpu_to(struct fpu *dst)
{
	dst->fpc = current->thread.fpu.fpc;
	dst->regs = current->thread.fpu.regs;
}

/*
 * Switches the FP/VX register save area from which to lazy
 * restore register contents.
 */
static inline void load_fpu_from(struct fpu *from)
{
	current->thread.fpu.fpc = from->fpc;
	current->thread.fpu.regs = from->regs;
}

1292 1293
void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
1294
	/* Save host register state */
1295
	save_fpu_regs();
1296
	save_fpu_to(&vcpu->arch.host_fpregs);
1297

1298
	if (test_kvm_facility(vcpu->kvm, 129)) {
1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309
		current->thread.fpu.fpc = vcpu->run->s.regs.fpc;
		/*
		 * Use the register save area in the SIE-control block
		 * for register restore and save in kvm_arch_vcpu_put()
		 */
		current->thread.fpu.vxrs =
			(__vector128 *)&vcpu->run->s.regs.vrs;
	} else
		load_fpu_from(&vcpu->arch.guest_fpregs);

	if (test_fp_ctl(current->thread.fpu.fpc))
1310
		/* User space provided an invalid FPC, let's clear it */
1311 1312 1313
		current->thread.fpu.fpc = 0;

	save_access_regs(vcpu->arch.host_acrs);
1314
	restore_access_regs(vcpu->run->s.regs.acrs);
1315
	gmap_enable(vcpu->arch.gmap);
1316
	atomic_or(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
1317 1318 1319 1320
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
1321
	atomic_andnot(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
1322
	gmap_disable(vcpu->arch.gmap);
1323

1324
	save_fpu_regs();
1325

1326
	if (test_kvm_facility(vcpu->kvm, 129))
1327 1328 1329 1330 1331 1332 1333
		/*
		 * kvm_arch_vcpu_load() set up the register save area to
		 * the &vcpu->run->s.regs.vrs and, thus, the vector registers
		 * are already saved.  Only the floating-point control must be
		 * copied.
		 */
		vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
1334
	else
1335 1336 1337 1338
		save_fpu_to(&vcpu->arch.guest_fpregs);
	load_fpu_from(&vcpu->arch.host_fpregs);

	save_access_regs(vcpu->run->s.regs.acrs);
1339 1340 1341 1342 1343 1344 1345 1346
	restore_access_regs(vcpu->arch.host_acrs);
}

static void kvm_s390_vcpu_initial_reset(struct kvm_vcpu *vcpu)
{
	/* this equals initial cpu reset in pop, but we don't switch to ESA */
	vcpu->arch.sie_block->gpsw.mask = 0UL;
	vcpu->arch.sie_block->gpsw.addr = 0UL;
1347
	kvm_s390_set_prefix(vcpu, 0);
1348 1349 1350 1351 1352 1353 1354 1355 1356
	vcpu->arch.sie_block->cputm     = 0UL;
	vcpu->arch.sie_block->ckc       = 0UL;
	vcpu->arch.sie_block->todpr     = 0;
	memset(vcpu->arch.sie_block->gcr, 0, 16 * sizeof(__u64));
	vcpu->arch.sie_block->gcr[0]  = 0xE0UL;
	vcpu->arch.sie_block->gcr[14] = 0xC2000000UL;
	vcpu->arch.guest_fpregs.fpc = 0;
	asm volatile("lfpc %0" : : "Q" (vcpu->arch.guest_fpregs.fpc));
	vcpu->arch.sie_block->gbea = 1;
1357
	vcpu->arch.sie_block->pp = 0;
1358 1359
	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
	kvm_clear_async_pf_completion_queue(vcpu);
1360 1361
	if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm))
		kvm_s390_vcpu_stop(vcpu);
1362
	kvm_s390_clear_local_irqs(vcpu);
1363 1364
}

1365
void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
1366
{
1367
	mutex_lock(&vcpu->kvm->lock);
1368
	preempt_disable();
1369
	vcpu->arch.sie_block->epoch = vcpu->kvm->arch.epoch;
1370
	preempt_enable();
1371
	mutex_unlock(&vcpu->kvm->lock);
1372 1373
	if (!kvm_is_ucontrol(vcpu->kvm))
		vcpu->arch.gmap = vcpu->kvm->arch.gmap;
1374 1375
}

1376 1377
static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu)
{
1378
	if (!test_kvm_facility(vcpu->kvm, 76))
1379 1380
		return;

1381 1382 1383 1384 1385 1386 1387
	vcpu->arch.sie_block->ecb3 &= ~(ECB3_AES | ECB3_DEA);

	if (vcpu->kvm->arch.crypto.aes_kw)
		vcpu->arch.sie_block->ecb3 |= ECB3_AES;
	if (vcpu->kvm->arch.crypto.dea_kw)
		vcpu->arch.sie_block->ecb3 |= ECB3_DEA;

1388 1389 1390
	vcpu->arch.sie_block->crycbd = vcpu->kvm->arch.crypto.crycbd;
}

1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407
void kvm_s390_vcpu_unsetup_cmma(struct kvm_vcpu *vcpu)
{
	free_page(vcpu->arch.sie_block->cbrlo);
	vcpu->arch.sie_block->cbrlo = 0;
}

int kvm_s390_vcpu_setup_cmma(struct kvm_vcpu *vcpu)
{
	vcpu->arch.sie_block->cbrlo = get_zeroed_page(GFP_KERNEL);
	if (!vcpu->arch.sie_block->cbrlo)
		return -ENOMEM;

	vcpu->arch.sie_block->ecb2 |= 0x80;
	vcpu->arch.sie_block->ecb2 &= ~0x08;
	return 0;
}

1408 1409 1410 1411 1412 1413 1414 1415 1416
static void kvm_s390_vcpu_setup_model(struct kvm_vcpu *vcpu)
{
	struct kvm_s390_cpu_model *model = &vcpu->kvm->arch.model;

	vcpu->arch.cpu_id = model->cpu_id;
	vcpu->arch.sie_block->ibc = model->ibc;
	vcpu->arch.sie_block->fac = (int) (long) model->fac->list;
}

1417 1418
int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
{
1419
	int rc = 0;
1420

1421 1422
	atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH |
						    CPUSTAT_SM |
1423 1424
						    CPUSTAT_STOPPED);

1425
	if (test_kvm_facility(vcpu->kvm, 78))
1426
		atomic_or(CPUSTAT_GED2, &vcpu->arch.sie_block->cpuflags);
1427
	else if (test_kvm_facility(vcpu->kvm, 8))
1428
		atomic_or(CPUSTAT_GED, &vcpu->arch.sie_block->cpuflags);
1429

1430 1431
	kvm_s390_vcpu_setup_model(vcpu);

1432
	vcpu->arch.sie_block->ecb   = 6;
1433
	if (test_kvm_facility(vcpu->kvm, 50) && test_kvm_facility(vcpu->kvm, 73))
1434 1435
		vcpu->arch.sie_block->ecb |= 0x10;

1436
	vcpu->arch.sie_block->ecb2  = 8;
1437
	vcpu->arch.sie_block->eca   = 0xC1002000U;
1438
	if (sclp.has_siif)
1439
		vcpu->arch.sie_block->eca |= 1;
1440
	if (sclp.has_sigpif)
1441
		vcpu->arch.sie_block->eca |= 0x10000000U;
1442
	if (test_kvm_facility(vcpu->kvm, 129)) {
1443 1444 1445
		vcpu->arch.sie_block->eca |= 0x00020000;
		vcpu->arch.sie_block->ecd |= 0x20000000;
	}
1446
	vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE;
1447

1448
	if (vcpu->kvm->arch.use_cmma) {
1449 1450 1451
		rc = kvm_s390_vcpu_setup_cmma(vcpu);
		if (rc)
			return rc;
1452
	}
1453
	hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
1454
	vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup;
1455

1456 1457
	kvm_s390_vcpu_crypto_setup(vcpu);

1458
	return rc;
1459 1460 1461 1462 1463
}

struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
				      unsigned int id)
{
1464
	struct kvm_vcpu *vcpu;
1465
	struct sie_page *sie_page;
1466 1467 1468 1469 1470 1471
	int rc = -EINVAL;

	if (id >= KVM_MAX_VCPUS)
		goto out;

	rc = -ENOMEM;
1472

1473
	vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
1474
	if (!vcpu)
1475
		goto out;
1476

1477 1478
	sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
	if (!sie_page)
1479 1480
		goto out_free_cpu;

1481 1482 1483
	vcpu->arch.sie_block = &sie_page->sie_block;
	vcpu->arch.sie_block->itdba = (unsigned long) &sie_page->itdb;

1484
	vcpu->arch.sie_block->icpua = id;
C
Carsten Otte 已提交
1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497
	if (!kvm_is_ucontrol(kvm)) {
		if (!kvm->arch.sca) {
			WARN_ON_ONCE(1);
			goto out_free_cpu;
		}
		if (!kvm->arch.sca->cpu[id].sda)
			kvm->arch.sca->cpu[id].sda =
				(__u64) vcpu->arch.sie_block;
		vcpu->arch.sie_block->scaoh =
			(__u32)(((__u64)kvm->arch.sca) >> 32);
		vcpu->arch.sie_block->scaol = (__u32)(__u64)kvm->arch.sca;
		set_bit(63 - id, (unsigned long *) &kvm->arch.sca->mcn);
	}
1498

1499 1500
	spin_lock_init(&vcpu->arch.local_int.lock);
	vcpu->arch.local_int.float_int = &kvm->arch.float_int;
1501
	vcpu->arch.local_int.wq = &vcpu->wq;
1502
	vcpu->arch.local_int.cpuflags = &vcpu->arch.sie_block->cpuflags;
1503

1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516
	/*
	 * Allocate a save area for floating-point registers.  If the vector
	 * extension is available, register contents are saved in the SIE
	 * control block.  The allocated save area is still required in
	 * particular places, for example, in kvm_s390_vcpu_store_status().
	 */
	vcpu->arch.guest_fpregs.fprs = kzalloc(sizeof(freg_t) * __NUM_FPRS,
					       GFP_KERNEL);
	if (!vcpu->arch.guest_fpregs.fprs) {
		rc = -ENOMEM;
		goto out_free_sie_block;
	}

1517 1518
	rc = kvm_vcpu_init(vcpu, kvm, id);
	if (rc)
1519
		goto out_free_sie_block;
1520 1521
	VM_EVENT(kvm, 3, "create cpu %d at %p, sie block at %p", id, vcpu,
		 vcpu->arch.sie_block);
1522
	trace_kvm_s390_create_vcpu(id, vcpu, vcpu->arch.sie_block);
1523 1524

	return vcpu;
1525 1526
out_free_sie_block:
	free_page((unsigned long)(vcpu->arch.sie_block));
1527
out_free_cpu:
1528
	kmem_cache_free(kvm_vcpu_cache, vcpu);
1529
out:
1530 1531 1532 1533 1534
	return ERR_PTR(rc);
}

int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
{
1535
	return kvm_s390_vcpu_has_irq(vcpu, 0);
1536 1537
}

1538
void kvm_s390_vcpu_block(struct kvm_vcpu *vcpu)
1539
{
1540
	atomic_or(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
1541
	exit_sie(vcpu);
1542 1543
}

1544
void kvm_s390_vcpu_unblock(struct kvm_vcpu *vcpu)
1545
{
1546
	atomic_andnot(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
1547 1548
}

1549 1550
static void kvm_s390_vcpu_request(struct kvm_vcpu *vcpu)
{
1551
	atomic_or(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
1552
	exit_sie(vcpu);
1553 1554 1555 1556
}

static void kvm_s390_vcpu_request_handled(struct kvm_vcpu *vcpu)
{
1557
	atomic_andnot(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
1558 1559
}

1560 1561 1562 1563 1564 1565
/*
 * Kick a guest cpu out of SIE and wait until SIE is not running.
 * If the CPU is not running (e.g. waiting as idle) the function will
 * return immediately. */
void exit_sie(struct kvm_vcpu *vcpu)
{
1566
	atomic_or(CPUSTAT_STOP_INT, &vcpu->arch.sie_block->cpuflags);
1567 1568 1569 1570
	while (vcpu->arch.sie_block->prog0c & PROG_IN_SIE)
		cpu_relax();
}

1571 1572
/* Kick a guest cpu out of SIE to process a request synchronously */
void kvm_s390_sync_request(int req, struct kvm_vcpu *vcpu)
1573
{
1574 1575
	kvm_make_request(req, vcpu);
	kvm_s390_vcpu_request(vcpu);
1576 1577
}

1578 1579 1580 1581 1582 1583 1584 1585
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long address)
{
	int i;
	struct kvm *kvm = gmap->private;
	struct kvm_vcpu *vcpu;

	kvm_for_each_vcpu(i, vcpu, kvm) {
		/* match against both prefix pages */
1586
		if (kvm_s390_get_prefix(vcpu) == (address & ~0x1000UL)) {
1587
			VCPU_EVENT(vcpu, 2, "gmap notifier for %lx", address);
1588
			kvm_s390_sync_request(KVM_REQ_MMU_RELOAD, vcpu);
1589 1590 1591 1592
		}
	}
}

1593 1594 1595 1596 1597 1598 1599
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
	/* kvm common code refers to this, but never calls it */
	BUG();
	return 0;
}

1600 1601 1602 1603 1604 1605
static int kvm_arch_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu,
					   struct kvm_one_reg *reg)
{
	int r = -EINVAL;

	switch (reg->id) {
1606 1607 1608 1609 1610 1611 1612 1613
	case KVM_REG_S390_TODPR:
		r = put_user(vcpu->arch.sie_block->todpr,
			     (u32 __user *)reg->addr);
		break;
	case KVM_REG_S390_EPOCHDIFF:
		r = put_user(vcpu->arch.sie_block->epoch,
			     (u64 __user *)reg->addr);
		break;
1614 1615 1616 1617 1618 1619 1620 1621
	case KVM_REG_S390_CPU_TIMER:
		r = put_user(vcpu->arch.sie_block->cputm,
			     (u64 __user *)reg->addr);
		break;
	case KVM_REG_S390_CLOCK_COMP:
		r = put_user(vcpu->arch.sie_block->ckc,
			     (u64 __user *)reg->addr);
		break;
1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633
	case KVM_REG_S390_PFTOKEN:
		r = put_user(vcpu->arch.pfault_token,
			     (u64 __user *)reg->addr);
		break;
	case KVM_REG_S390_PFCOMPARE:
		r = put_user(vcpu->arch.pfault_compare,
			     (u64 __user *)reg->addr);
		break;
	case KVM_REG_S390_PFSELECT:
		r = put_user(vcpu->arch.pfault_select,
			     (u64 __user *)reg->addr);
		break;
1634 1635 1636 1637
	case KVM_REG_S390_PP:
		r = put_user(vcpu->arch.sie_block->pp,
			     (u64 __user *)reg->addr);
		break;
1638 1639 1640 1641
	case KVM_REG_S390_GBEA:
		r = put_user(vcpu->arch.sie_block->gbea,
			     (u64 __user *)reg->addr);
		break;
1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654
	default:
		break;
	}

	return r;
}

static int kvm_arch_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu,
					   struct kvm_one_reg *reg)
{
	int r = -EINVAL;

	switch (reg->id) {
1655 1656 1657 1658 1659 1660 1661 1662
	case KVM_REG_S390_TODPR:
		r = get_user(vcpu->arch.sie_block->todpr,
			     (u32 __user *)reg->addr);
		break;
	case KVM_REG_S390_EPOCHDIFF:
		r = get_user(vcpu->arch.sie_block->epoch,
			     (u64 __user *)reg->addr);
		break;
1663 1664 1665 1666 1667 1668 1669 1670
	case KVM_REG_S390_CPU_TIMER:
		r = get_user(vcpu->arch.sie_block->cputm,
			     (u64 __user *)reg->addr);
		break;
	case KVM_REG_S390_CLOCK_COMP:
		r = get_user(vcpu->arch.sie_block->ckc,
			     (u64 __user *)reg->addr);
		break;
1671 1672 1673
	case KVM_REG_S390_PFTOKEN:
		r = get_user(vcpu->arch.pfault_token,
			     (u64 __user *)reg->addr);
1674 1675
		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
			kvm_clear_async_pf_completion_queue(vcpu);
1676 1677 1678 1679 1680 1681 1682 1683 1684
		break;
	case KVM_REG_S390_PFCOMPARE:
		r = get_user(vcpu->arch.pfault_compare,
			     (u64 __user *)reg->addr);
		break;
	case KVM_REG_S390_PFSELECT:
		r = get_user(vcpu->arch.pfault_select,
			     (u64 __user *)reg->addr);
		break;
1685 1686 1687 1688
	case KVM_REG_S390_PP:
		r = get_user(vcpu->arch.sie_block->pp,
			     (u64 __user *)reg->addr);
		break;
1689 1690 1691 1692
	case KVM_REG_S390_GBEA:
		r = get_user(vcpu->arch.sie_block->gbea,
			     (u64 __user *)reg->addr);
		break;
1693 1694 1695 1696 1697 1698
	default:
		break;
	}

	return r;
}
1699

1700 1701 1702 1703 1704 1705 1706 1707
static int kvm_arch_vcpu_ioctl_initial_reset(struct kvm_vcpu *vcpu)
{
	kvm_s390_vcpu_initial_reset(vcpu);
	return 0;
}

int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
1708
	memcpy(&vcpu->run->s.regs.gprs, &regs->gprs, sizeof(regs->gprs));
1709 1710 1711 1712 1713
	return 0;
}

int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
1714
	memcpy(&regs->gprs, &vcpu->run->s.regs.gprs, sizeof(regs->gprs));
1715 1716 1717 1718 1719 1720
	return 0;
}

int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
1721
	memcpy(&vcpu->run->s.regs.acrs, &sregs->acrs, sizeof(sregs->acrs));
1722
	memcpy(&vcpu->arch.sie_block->gcr, &sregs->crs, sizeof(sregs->crs));
1723
	restore_access_regs(vcpu->run->s.regs.acrs);
1724 1725 1726 1727 1728 1729
	return 0;
}

int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
1730
	memcpy(&sregs->acrs, &vcpu->run->s.regs.acrs, sizeof(sregs->acrs));
1731 1732 1733 1734 1735 1736
	memcpy(&sregs->crs, &vcpu->arch.sie_block->gcr, sizeof(sregs->crs));
	return 0;
}

int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
1737 1738
	if (test_fp_ctl(fpu->fpc))
		return -EINVAL;
1739
	memcpy(vcpu->arch.guest_fpregs.fprs, &fpu->fprs, sizeof(fpu->fprs));
1740
	vcpu->arch.guest_fpregs.fpc = fpu->fpc;
1741
	save_fpu_regs();
1742
	load_fpu_from(&vcpu->arch.guest_fpregs);
1743 1744 1745 1746 1747
	return 0;
}

int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
1748
	memcpy(&fpu->fprs, vcpu->arch.guest_fpregs.fprs, sizeof(fpu->fprs));
1749 1750 1751 1752 1753 1754 1755 1756
	fpu->fpc = vcpu->arch.guest_fpregs.fpc;
	return 0;
}

static int kvm_arch_vcpu_ioctl_set_initial_psw(struct kvm_vcpu *vcpu, psw_t psw)
{
	int rc = 0;

1757
	if (!is_vcpu_stopped(vcpu))
1758
		rc = -EBUSY;
1759 1760 1761 1762
	else {
		vcpu->run->psw_mask = psw.mask;
		vcpu->run->psw_addr = psw.addr;
	}
1763 1764 1765 1766 1767 1768 1769 1770 1771
	return rc;
}

int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
				  struct kvm_translation *tr)
{
	return -EINVAL; /* not implemented yet */
}

1772 1773 1774 1775
#define VALID_GUESTDBG_FLAGS (KVM_GUESTDBG_SINGLESTEP | \
			      KVM_GUESTDBG_USE_HW_BP | \
			      KVM_GUESTDBG_ENABLE)

J
Jan Kiszka 已提交
1776 1777
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
					struct kvm_guest_debug *dbg)
1778
{
1779 1780 1781 1782 1783
	int rc = 0;

	vcpu->guest_debug = 0;
	kvm_s390_clear_bp_data(vcpu);

1784
	if (dbg->control & ~VALID_GUESTDBG_FLAGS)
1785 1786 1787 1788 1789
		return -EINVAL;

	if (dbg->control & KVM_GUESTDBG_ENABLE) {
		vcpu->guest_debug = dbg->control;
		/* enforce guest PER */
1790
		atomic_or(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
1791 1792 1793 1794

		if (dbg->control & KVM_GUESTDBG_USE_HW_BP)
			rc = kvm_s390_import_bp_data(vcpu, dbg);
	} else {
1795
		atomic_andnot(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
1796 1797 1798 1799 1800 1801
		vcpu->arch.guestdbg.last_bp = 0;
	}

	if (rc) {
		vcpu->guest_debug = 0;
		kvm_s390_clear_bp_data(vcpu);
1802
		atomic_andnot(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
1803 1804 1805
	}

	return rc;
1806 1807
}

1808 1809 1810
int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
1811 1812 1813
	/* CHECK_STOP and LOAD are not supported yet */
	return is_vcpu_stopped(vcpu) ? KVM_MP_STATE_STOPPED :
				       KVM_MP_STATE_OPERATING;
1814 1815 1816 1817 1818
}

int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838
	int rc = 0;

	/* user space knows about this interface - let it control the state */
	vcpu->kvm->arch.user_cpu_state_ctrl = 1;

	switch (mp_state->mp_state) {
	case KVM_MP_STATE_STOPPED:
		kvm_s390_vcpu_stop(vcpu);
		break;
	case KVM_MP_STATE_OPERATING:
		kvm_s390_vcpu_start(vcpu);
		break;
	case KVM_MP_STATE_LOAD:
	case KVM_MP_STATE_CHECK_STOP:
		/* fall through - CHECK_STOP and LOAD are not supported yet */
	default:
		rc = -ENXIO;
	}

	return rc;
1839 1840
}

1841 1842 1843 1844 1845
static bool ibs_enabled(struct kvm_vcpu *vcpu)
{
	return atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_IBS;
}

1846 1847
static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu)
{
1848
retry:
1849
	kvm_s390_vcpu_request_handled(vcpu);
1850 1851
	if (!vcpu->requests)
		return 0;
1852 1853 1854 1855 1856 1857 1858
	/*
	 * We use MMU_RELOAD just to re-arm the ipte notifier for the
	 * guest prefix page. gmap_ipte_notify will wait on the ptl lock.
	 * This ensures that the ipte instruction for this request has
	 * already finished. We might race against a second unmapper that
	 * wants to set the blocking bit. Lets just retry the request loop.
	 */
1859
	if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) {
1860 1861
		int rc;
		rc = gmap_ipte_notify(vcpu->arch.gmap,
1862
				      kvm_s390_get_prefix(vcpu),
1863 1864 1865
				      PAGE_SIZE * 2);
		if (rc)
			return rc;
1866
		goto retry;
1867
	}
1868

1869 1870 1871 1872 1873
	if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) {
		vcpu->arch.sie_block->ihcpu = 0xffff;
		goto retry;
	}

1874 1875 1876
	if (kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu)) {
		if (!ibs_enabled(vcpu)) {
			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 1);
1877
			atomic_or(CPUSTAT_IBS,
1878 1879 1880
					&vcpu->arch.sie_block->cpuflags);
		}
		goto retry;
1881
	}
1882 1883 1884 1885

	if (kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu)) {
		if (ibs_enabled(vcpu)) {
			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 0);
1886
			atomic_andnot(CPUSTAT_IBS,
1887 1888 1889 1890 1891
					  &vcpu->arch.sie_block->cpuflags);
		}
		goto retry;
	}

1892 1893 1894
	/* nothing to do, just clear the request */
	clear_bit(KVM_REQ_UNHALT, &vcpu->requests);

1895 1896 1897
	return 0;
}

1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913
void kvm_s390_set_tod_clock(struct kvm *kvm, u64 tod)
{
	struct kvm_vcpu *vcpu;
	int i;

	mutex_lock(&kvm->lock);
	preempt_disable();
	kvm->arch.epoch = tod - get_tod_clock();
	kvm_s390_vcpu_block_all(kvm);
	kvm_for_each_vcpu(i, vcpu, kvm)
		vcpu->arch.sie_block->epoch = kvm->arch.epoch;
	kvm_s390_vcpu_unblock_all(kvm);
	preempt_enable();
	mutex_unlock(&kvm->lock);
}

1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924
/**
 * kvm_arch_fault_in_page - fault-in guest page if necessary
 * @vcpu: The corresponding virtual cpu
 * @gpa: Guest physical address
 * @writable: Whether the page should be writable or not
 *
 * Make sure that a guest page has been faulted-in on the host.
 *
 * Return: Zero on success, negative error code otherwise.
 */
long kvm_arch_fault_in_page(struct kvm_vcpu *vcpu, gpa_t gpa, int writable)
1925
{
1926 1927
	return gmap_fault(vcpu->arch.gmap, gpa,
			  writable ? FAULT_FLAG_WRITE : 0);
1928 1929
}

1930 1931 1932 1933
static void __kvm_inject_pfault_token(struct kvm_vcpu *vcpu, bool start_token,
				      unsigned long token)
{
	struct kvm_s390_interrupt inti;
1934
	struct kvm_s390_irq irq;
1935 1936

	if (start_token) {
1937 1938 1939
		irq.u.ext.ext_params2 = token;
		irq.type = KVM_S390_INT_PFAULT_INIT;
		WARN_ON_ONCE(kvm_s390_inject_vcpu(vcpu, &irq));
1940 1941
	} else {
		inti.type = KVM_S390_INT_PFAULT_DONE;
1942
		inti.parm64 = token;
1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988
		WARN_ON_ONCE(kvm_s390_inject_vm(vcpu->kvm, &inti));
	}
}

void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu,
				     struct kvm_async_pf *work)
{
	trace_kvm_s390_pfault_init(vcpu, work->arch.pfault_token);
	__kvm_inject_pfault_token(vcpu, true, work->arch.pfault_token);
}

void kvm_arch_async_page_present(struct kvm_vcpu *vcpu,
				 struct kvm_async_pf *work)
{
	trace_kvm_s390_pfault_done(vcpu, work->arch.pfault_token);
	__kvm_inject_pfault_token(vcpu, false, work->arch.pfault_token);
}

void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu,
			       struct kvm_async_pf *work)
{
	/* s390 will always inject the page directly */
}

bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu)
{
	/*
	 * s390 will always inject the page directly,
	 * but we still want check_async_completion to cleanup
	 */
	return true;
}

static int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu)
{
	hva_t hva;
	struct kvm_arch_async_pf arch;
	int rc;

	if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
		return 0;
	if ((vcpu->arch.sie_block->gpsw.mask & vcpu->arch.pfault_select) !=
	    vcpu->arch.pfault_compare)
		return 0;
	if (psw_extint_disabled(vcpu))
		return 0;
1989
	if (kvm_s390_vcpu_has_irq(vcpu, 0))
1990 1991 1992 1993 1994 1995
		return 0;
	if (!(vcpu->arch.sie_block->gcr[0] & 0x200ul))
		return 0;
	if (!vcpu->arch.gmap->pfault_enabled)
		return 0;

H
Heiko Carstens 已提交
1996 1997 1998
	hva = gfn_to_hva(vcpu->kvm, gpa_to_gfn(current->thread.gmap_addr));
	hva += current->thread.gmap_addr & ~PAGE_MASK;
	if (read_guest_real(vcpu, vcpu->arch.pfault_token, &arch.pfault_token, 8))
1999 2000 2001 2002 2003 2004
		return 0;

	rc = kvm_setup_async_pf(vcpu, current->thread.gmap_addr, hva, &arch);
	return rc;
}

2005
static int vcpu_pre_run(struct kvm_vcpu *vcpu)
2006
{
2007
	int rc, cpuflags;
2008

2009 2010 2011 2012 2013 2014 2015
	/*
	 * On s390 notifications for arriving pages will be delivered directly
	 * to the guest but the house keeping for completed pfaults is
	 * handled outside the worker.
	 */
	kvm_check_async_pf_completion(vcpu);

2016
	memcpy(&vcpu->arch.sie_block->gg14, &vcpu->run->s.regs.gprs[14], 16);
2017 2018 2019 2020

	if (need_resched())
		schedule();

2021
	if (test_cpu_flag(CIF_MCCK_PENDING))
2022 2023
		s390_handle_mcck();

2024 2025 2026 2027 2028
	if (!kvm_is_ucontrol(vcpu->kvm)) {
		rc = kvm_s390_deliver_pending_interrupts(vcpu);
		if (rc)
			return rc;
	}
C
Carsten Otte 已提交
2029

2030 2031 2032 2033
	rc = kvm_s390_handle_requests(vcpu);
	if (rc)
		return rc;

2034 2035 2036 2037 2038
	if (guestdbg_enabled(vcpu)) {
		kvm_s390_backup_guest_per_regs(vcpu);
		kvm_s390_patch_guest_per_regs(vcpu);
	}

2039
	vcpu->arch.sie_block->icptcode = 0;
2040 2041 2042
	cpuflags = atomic_read(&vcpu->arch.sie_block->cpuflags);
	VCPU_EVENT(vcpu, 6, "entering sie flags %x", cpuflags);
	trace_kvm_s390_sie_enter(vcpu, cpuflags);
2043

2044 2045 2046
	return 0;
}

2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063
static int vcpu_post_run_fault_in_sie(struct kvm_vcpu *vcpu)
{
	psw_t *psw = &vcpu->arch.sie_block->gpsw;
	u8 opcode;
	int rc;

	VCPU_EVENT(vcpu, 3, "%s", "fault in sie instruction");
	trace_kvm_s390_sie_fault(vcpu);

	/*
	 * We want to inject an addressing exception, which is defined as a
	 * suppressing or terminating exception. However, since we came here
	 * by a DAT access exception, the PSW still points to the faulting
	 * instruction since DAT exceptions are nullifying. So we've got
	 * to look up the current opcode to get the length of the instruction
	 * to be able to forward the PSW.
	 */
2064
	rc = read_guest(vcpu, psw->addr, 0, &opcode, 1);
2065 2066 2067 2068 2069 2070 2071
	if (rc)
		return kvm_s390_inject_prog_cond(vcpu, rc);
	psw->addr = __rewind_psw(*psw, -insn_length(opcode));

	return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
}

2072 2073
static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason)
{
2074 2075 2076 2077
	VCPU_EVENT(vcpu, 6, "exit sie icptcode %d",
		   vcpu->arch.sie_block->icptcode);
	trace_kvm_s390_sie_exit(vcpu, vcpu->arch.sie_block->icptcode);

2078 2079 2080
	if (guestdbg_enabled(vcpu))
		kvm_s390_restore_guest_per_regs(vcpu);

2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095
	memcpy(&vcpu->run->s.regs.gprs[14], &vcpu->arch.sie_block->gg14, 16);

	if (vcpu->arch.sie_block->icptcode > 0) {
		int rc = kvm_handle_sie_intercept(vcpu);

		if (rc != -EOPNOTSUPP)
			return rc;
		vcpu->run->exit_reason = KVM_EXIT_S390_SIEIC;
		vcpu->run->s390_sieic.icptcode = vcpu->arch.sie_block->icptcode;
		vcpu->run->s390_sieic.ipa = vcpu->arch.sie_block->ipa;
		vcpu->run->s390_sieic.ipb = vcpu->arch.sie_block->ipb;
		return -EREMOTE;
	} else if (exit_reason != -EFAULT) {
		vcpu->stat.exit_null++;
		return 0;
2096 2097 2098 2099 2100
	} else if (kvm_is_ucontrol(vcpu->kvm)) {
		vcpu->run->exit_reason = KVM_EXIT_S390_UCONTROL;
		vcpu->run->s390_ucontrol.trans_exc_code =
						current->thread.gmap_addr;
		vcpu->run->s390_ucontrol.pgm_code = 0x10;
2101
		return -EREMOTE;
2102
	} else if (current->thread.gmap_pfault) {
2103
		trace_kvm_s390_major_guest_pfault(vcpu);
2104
		current->thread.gmap_pfault = 0;
2105 2106 2107
		if (kvm_arch_setup_async_pf(vcpu))
			return 0;
		return kvm_arch_fault_in_page(vcpu, current->thread.gmap_addr, 1);
2108
	}
2109
	return vcpu_post_run_fault_in_sie(vcpu);
2110 2111 2112 2113 2114 2115
}

static int __vcpu_run(struct kvm_vcpu *vcpu)
{
	int rc, exit_reason;

2116 2117 2118 2119 2120 2121
	/*
	 * We try to hold kvm->srcu during most of vcpu_run (except when run-
	 * ning the guest), so that memslots (and other stuff) are protected
	 */
	vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);

2122 2123 2124 2125
	do {
		rc = vcpu_pre_run(vcpu);
		if (rc)
			break;
2126

2127
		srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
2128 2129 2130 2131
		/*
		 * As PF_VCPU will be used in fault handler, between
		 * guest_enter and guest_exit should be no uaccess.
		 */
2132 2133 2134
		local_irq_disable();
		__kvm_guest_enter();
		local_irq_enable();
2135 2136
		exit_reason = sie64a(vcpu->arch.sie_block,
				     vcpu->run->s.regs.gprs);
2137 2138 2139
		local_irq_disable();
		__kvm_guest_exit();
		local_irq_enable();
2140
		vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
2141 2142

		rc = vcpu_post_run(vcpu, exit_reason);
2143
	} while (!signal_pending(current) && !guestdbg_exit_pending(vcpu) && !rc);
2144

2145
	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
2146
	return rc;
2147 2148
}

2149 2150 2151 2152 2153 2154 2155 2156
static void sync_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
	vcpu->arch.sie_block->gpsw.mask = kvm_run->psw_mask;
	vcpu->arch.sie_block->gpsw.addr = kvm_run->psw_addr;
	if (kvm_run->kvm_dirty_regs & KVM_SYNC_PREFIX)
		kvm_s390_set_prefix(vcpu, kvm_run->s.regs.prefix);
	if (kvm_run->kvm_dirty_regs & KVM_SYNC_CRS) {
		memcpy(&vcpu->arch.sie_block->gcr, &kvm_run->s.regs.crs, 128);
2157 2158
		/* some control register changes require a tlb flush */
		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170
	}
	if (kvm_run->kvm_dirty_regs & KVM_SYNC_ARCH0) {
		vcpu->arch.sie_block->cputm = kvm_run->s.regs.cputm;
		vcpu->arch.sie_block->ckc = kvm_run->s.regs.ckc;
		vcpu->arch.sie_block->todpr = kvm_run->s.regs.todpr;
		vcpu->arch.sie_block->pp = kvm_run->s.regs.pp;
		vcpu->arch.sie_block->gbea = kvm_run->s.regs.gbea;
	}
	if (kvm_run->kvm_dirty_regs & KVM_SYNC_PFAULT) {
		vcpu->arch.pfault_token = kvm_run->s.regs.pft;
		vcpu->arch.pfault_select = kvm_run->s.regs.pfs;
		vcpu->arch.pfault_compare = kvm_run->s.regs.pfc;
2171 2172
		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
			kvm_clear_async_pf_completion_queue(vcpu);
2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192
	}
	kvm_run->kvm_dirty_regs = 0;
}

static void store_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
	kvm_run->psw_mask = vcpu->arch.sie_block->gpsw.mask;
	kvm_run->psw_addr = vcpu->arch.sie_block->gpsw.addr;
	kvm_run->s.regs.prefix = kvm_s390_get_prefix(vcpu);
	memcpy(&kvm_run->s.regs.crs, &vcpu->arch.sie_block->gcr, 128);
	kvm_run->s.regs.cputm = vcpu->arch.sie_block->cputm;
	kvm_run->s.regs.ckc = vcpu->arch.sie_block->ckc;
	kvm_run->s.regs.todpr = vcpu->arch.sie_block->todpr;
	kvm_run->s.regs.pp = vcpu->arch.sie_block->pp;
	kvm_run->s.regs.gbea = vcpu->arch.sie_block->gbea;
	kvm_run->s.regs.pft = vcpu->arch.pfault_token;
	kvm_run->s.regs.pfs = vcpu->arch.pfault_select;
	kvm_run->s.regs.pfc = vcpu->arch.pfault_compare;
}

2193 2194
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
2195
	int rc;
2196 2197
	sigset_t sigsaved;

2198 2199 2200 2201 2202
	if (guestdbg_exit_pending(vcpu)) {
		kvm_s390_prepare_debug_exit(vcpu);
		return 0;
	}

2203 2204 2205
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);

2206 2207 2208
	if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) {
		kvm_s390_vcpu_start(vcpu);
	} else if (is_vcpu_stopped(vcpu)) {
2209
		pr_err_ratelimited("can't run stopped vcpu %d\n",
2210 2211 2212
				   vcpu->vcpu_id);
		return -EINVAL;
	}
2213

2214
	sync_regs(vcpu, kvm_run);
2215

2216
	might_fault();
2217
	rc = __vcpu_run(vcpu);
2218

2219 2220
	if (signal_pending(current) && !rc) {
		kvm_run->exit_reason = KVM_EXIT_INTR;
2221
		rc = -EINTR;
2222
	}
2223

2224 2225 2226 2227 2228
	if (guestdbg_exit_pending(vcpu) && !rc)  {
		kvm_s390_prepare_debug_exit(vcpu);
		rc = 0;
	}

2229
	if (rc == -EREMOTE) {
2230
		/* userspace support is needed, kvm_run has been prepared */
2231 2232
		rc = 0;
	}
2233

2234
	store_regs(vcpu, kvm_run);
2235

2236 2237 2238 2239
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &sigsaved, NULL);

	vcpu->stat.exit_userspace++;
2240
	return rc;
2241 2242 2243 2244 2245 2246 2247 2248
}

/*
 * store status at address
 * we use have two special cases:
 * KVM_S390_STORE_STATUS_NOADDR: -> 0x1200 on 64 bit
 * KVM_S390_STORE_STATUS_PREFIXED: -> prefix
 */
2249
int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long gpa)
2250
{
2251
	unsigned char archmode = 1;
2252
	unsigned int px;
2253
	u64 clkcomp;
2254
	int rc;
2255

2256 2257
	if (gpa == KVM_S390_STORE_STATUS_NOADDR) {
		if (write_guest_abs(vcpu, 163, &archmode, 1))
2258
			return -EFAULT;
2259 2260 2261
		gpa = SAVE_AREA_BASE;
	} else if (gpa == KVM_S390_STORE_STATUS_PREFIXED) {
		if (write_guest_real(vcpu, 163, &archmode, 1))
2262
			return -EFAULT;
2263 2264 2265 2266 2267 2268 2269 2270
		gpa = kvm_s390_real_to_abs(vcpu, SAVE_AREA_BASE);
	}
	rc = write_guest_abs(vcpu, gpa + offsetof(struct save_area, fp_regs),
			     vcpu->arch.guest_fpregs.fprs, 128);
	rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, gp_regs),
			      vcpu->run->s.regs.gprs, 128);
	rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, psw),
			      &vcpu->arch.sie_block->gpsw, 16);
2271
	px = kvm_s390_get_prefix(vcpu);
2272
	rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, pref_reg),
2273
			      &px, 4);
2274 2275 2276 2277 2278 2279 2280
	rc |= write_guest_abs(vcpu,
			      gpa + offsetof(struct save_area, fp_ctrl_reg),
			      &vcpu->arch.guest_fpregs.fpc, 4);
	rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, tod_reg),
			      &vcpu->arch.sie_block->todpr, 4);
	rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, timer),
			      &vcpu->arch.sie_block->cputm, 8);
2281
	clkcomp = vcpu->arch.sie_block->ckc >> 8;
2282 2283 2284 2285 2286 2287 2288
	rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, clk_cmp),
			      &clkcomp, 8);
	rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, acc_regs),
			      &vcpu->run->s.regs.acrs, 64);
	rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, ctrl_regs),
			      &vcpu->arch.sie_block->gcr, 128);
	return rc ? -EFAULT : 0;
2289 2290
}

2291 2292 2293 2294 2295 2296 2297
int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr)
{
	/*
	 * The guest FPRS and ACRS are in the host FPRS/ACRS due to the lazy
	 * copying in vcpu load/put. Lets update our copies before we save
	 * it into the save area
	 */
2298
	save_fpu_regs();
2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311
	if (test_kvm_facility(vcpu->kvm, 129)) {
		/*
		 * If the vector extension is available, the vector registers
		 * which overlaps with floating-point registers are saved in
		 * the SIE-control block.  Hence, extract the floating-point
		 * registers and the FPC value and store them in the
		 * guest_fpregs structure.
		 */
		vcpu->arch.guest_fpregs.fpc = current->thread.fpu.fpc;
		convert_vx_to_fp(vcpu->arch.guest_fpregs.fprs,
				 current->thread.fpu.vxrs);
	} else
		save_fpu_to(&vcpu->arch.guest_fpregs);
2312 2313 2314 2315 2316
	save_access_regs(vcpu->run->s.regs.acrs);

	return kvm_s390_store_status_unloaded(vcpu, addr);
}

E
Eric Farman 已提交
2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337
/*
 * store additional status at address
 */
int kvm_s390_store_adtl_status_unloaded(struct kvm_vcpu *vcpu,
					unsigned long gpa)
{
	/* Only bits 0-53 are used for address formation */
	if (!(gpa & ~0x3ff))
		return 0;

	return write_guest_abs(vcpu, gpa & ~0x3ff,
			       (void *)&vcpu->run->s.regs.vrs, 512);
}

int kvm_s390_vcpu_store_adtl_status(struct kvm_vcpu *vcpu, unsigned long addr)
{
	if (!test_kvm_facility(vcpu->kvm, 129))
		return 0;

	/*
	 * The guest VXRS are in the host VXRs due to the lazy
2338 2339 2340 2341 2342
	 * copying in vcpu load/put. We can simply call save_fpu_regs()
	 * to save the current register state because we are in the
	 * middle of a load/put cycle.
	 *
	 * Let's update our copies before we save it into the save area.
E
Eric Farman 已提交
2343
	 */
2344
	save_fpu_regs();
E
Eric Farman 已提交
2345 2346 2347 2348

	return kvm_s390_store_adtl_status_unloaded(vcpu, addr);
}

2349 2350 2351
static void __disable_ibs_on_vcpu(struct kvm_vcpu *vcpu)
{
	kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu);
2352
	kvm_s390_sync_request(KVM_REQ_DISABLE_IBS, vcpu);
2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367
}

static void __disable_ibs_on_all_vcpus(struct kvm *kvm)
{
	unsigned int i;
	struct kvm_vcpu *vcpu;

	kvm_for_each_vcpu(i, vcpu, kvm) {
		__disable_ibs_on_vcpu(vcpu);
	}
}

static void __enable_ibs_on_vcpu(struct kvm_vcpu *vcpu)
{
	kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu);
2368
	kvm_s390_sync_request(KVM_REQ_ENABLE_IBS, vcpu);
2369 2370
}

2371 2372
void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu)
{
2373 2374 2375 2376 2377
	int i, online_vcpus, started_vcpus = 0;

	if (!is_vcpu_stopped(vcpu))
		return;

2378
	trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 1);
2379
	/* Only one cpu at a time may enter/leave the STOPPED state. */
2380
	spin_lock(&vcpu->kvm->arch.start_stop_lock);
2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399
	online_vcpus = atomic_read(&vcpu->kvm->online_vcpus);

	for (i = 0; i < online_vcpus; i++) {
		if (!is_vcpu_stopped(vcpu->kvm->vcpus[i]))
			started_vcpus++;
	}

	if (started_vcpus == 0) {
		/* we're the only active VCPU -> speed it up */
		__enable_ibs_on_vcpu(vcpu);
	} else if (started_vcpus == 1) {
		/*
		 * As we are starting a second VCPU, we have to disable
		 * the IBS facility on all VCPUs to remove potentially
		 * oustanding ENABLE requests.
		 */
		__disable_ibs_on_all_vcpus(vcpu->kvm);
	}

2400
	atomic_andnot(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
2401 2402 2403 2404
	/*
	 * Another VCPU might have used IBS while we were offline.
	 * Let's play safe and flush the VCPU at startup.
	 */
2405
	kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
2406
	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
2407
	return;
2408 2409 2410 2411
}

void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu)
{
2412 2413 2414 2415 2416 2417
	int i, online_vcpus, started_vcpus = 0;
	struct kvm_vcpu *started_vcpu = NULL;

	if (is_vcpu_stopped(vcpu))
		return;

2418
	trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 0);
2419
	/* Only one cpu at a time may enter/leave the STOPPED state. */
2420
	spin_lock(&vcpu->kvm->arch.start_stop_lock);
2421 2422
	online_vcpus = atomic_read(&vcpu->kvm->online_vcpus);

2423
	/* SIGP STOP and SIGP STOP AND STORE STATUS has been fully processed */
2424
	kvm_s390_clear_stop_irq(vcpu);
2425

2426
	atomic_or(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443
	__disable_ibs_on_vcpu(vcpu);

	for (i = 0; i < online_vcpus; i++) {
		if (!is_vcpu_stopped(vcpu->kvm->vcpus[i])) {
			started_vcpus++;
			started_vcpu = vcpu->kvm->vcpus[i];
		}
	}

	if (started_vcpus == 1) {
		/*
		 * As we only have one VCPU left, we want to enable the
		 * IBS facility for that VCPU to speed it up.
		 */
		__enable_ibs_on_vcpu(started_vcpu);
	}

2444
	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
2445
	return;
2446 2447
}

2448 2449 2450 2451 2452 2453 2454 2455 2456
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) {
2457 2458 2459
	case KVM_CAP_S390_CSS_SUPPORT:
		if (!vcpu->kvm->arch.css_support) {
			vcpu->kvm->arch.css_support = 1;
2460
			VM_EVENT(vcpu->kvm, 3, "%s", "ENABLE: CSS support");
2461 2462 2463 2464
			trace_kvm_s390_enable_css(vcpu->kvm);
		}
		r = 0;
		break;
2465 2466 2467 2468 2469 2470 2471
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530
static long kvm_s390_guest_mem_op(struct kvm_vcpu *vcpu,
				  struct kvm_s390_mem_op *mop)
{
	void __user *uaddr = (void __user *)mop->buf;
	void *tmpbuf = NULL;
	int r, srcu_idx;
	const u64 supported_flags = KVM_S390_MEMOP_F_INJECT_EXCEPTION
				    | KVM_S390_MEMOP_F_CHECK_ONLY;

	if (mop->flags & ~supported_flags)
		return -EINVAL;

	if (mop->size > MEM_OP_MAX_SIZE)
		return -E2BIG;

	if (!(mop->flags & KVM_S390_MEMOP_F_CHECK_ONLY)) {
		tmpbuf = vmalloc(mop->size);
		if (!tmpbuf)
			return -ENOMEM;
	}

	srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);

	switch (mop->op) {
	case KVM_S390_MEMOP_LOGICAL_READ:
		if (mop->flags & KVM_S390_MEMOP_F_CHECK_ONLY) {
			r = check_gva_range(vcpu, mop->gaddr, mop->ar, mop->size, false);
			break;
		}
		r = read_guest(vcpu, mop->gaddr, mop->ar, tmpbuf, mop->size);
		if (r == 0) {
			if (copy_to_user(uaddr, tmpbuf, mop->size))
				r = -EFAULT;
		}
		break;
	case KVM_S390_MEMOP_LOGICAL_WRITE:
		if (mop->flags & KVM_S390_MEMOP_F_CHECK_ONLY) {
			r = check_gva_range(vcpu, mop->gaddr, mop->ar, mop->size, true);
			break;
		}
		if (copy_from_user(tmpbuf, uaddr, mop->size)) {
			r = -EFAULT;
			break;
		}
		r = write_guest(vcpu, mop->gaddr, mop->ar, tmpbuf, mop->size);
		break;
	default:
		r = -EINVAL;
	}

	srcu_read_unlock(&vcpu->kvm->srcu, srcu_idx);

	if (r > 0 && (mop->flags & KVM_S390_MEMOP_F_INJECT_EXCEPTION) != 0)
		kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm);

	vfree(tmpbuf);
	return r;
}

2531 2532 2533 2534 2535
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;
2536
	int idx;
2537
	long r;
2538

2539
	switch (ioctl) {
2540 2541 2542 2543 2544 2545 2546 2547 2548
	case KVM_S390_IRQ: {
		struct kvm_s390_irq s390irq;

		r = -EFAULT;
		if (copy_from_user(&s390irq, argp, sizeof(s390irq)))
			break;
		r = kvm_s390_inject_vcpu(vcpu, &s390irq);
		break;
	}
2549
	case KVM_S390_INTERRUPT: {
2550
		struct kvm_s390_interrupt s390int;
2551
		struct kvm_s390_irq s390irq;
2552

2553
		r = -EFAULT;
2554
		if (copy_from_user(&s390int, argp, sizeof(s390int)))
2555
			break;
2556 2557 2558
		if (s390int_to_s390irq(&s390int, &s390irq))
			return -EINVAL;
		r = kvm_s390_inject_vcpu(vcpu, &s390irq);
2559
		break;
2560
	}
2561
	case KVM_S390_STORE_STATUS:
2562
		idx = srcu_read_lock(&vcpu->kvm->srcu);
2563
		r = kvm_s390_vcpu_store_status(vcpu, arg);
2564
		srcu_read_unlock(&vcpu->kvm->srcu, idx);
2565
		break;
2566 2567 2568
	case KVM_S390_SET_INITIAL_PSW: {
		psw_t psw;

2569
		r = -EFAULT;
2570
		if (copy_from_user(&psw, argp, sizeof(psw)))
2571 2572 2573
			break;
		r = kvm_arch_vcpu_ioctl_set_initial_psw(vcpu, psw);
		break;
2574 2575
	}
	case KVM_S390_INITIAL_RESET:
2576 2577
		r = kvm_arch_vcpu_ioctl_initial_reset(vcpu);
		break;
2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589
	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)))
			break;
		if (ioctl == KVM_SET_ONE_REG)
			r = kvm_arch_vcpu_ioctl_set_one_reg(vcpu, &reg);
		else
			r = kvm_arch_vcpu_ioctl_get_one_reg(vcpu, &reg);
		break;
	}
2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625
#ifdef CONFIG_KVM_S390_UCONTROL
	case KVM_S390_UCAS_MAP: {
		struct kvm_s390_ucas_mapping ucasmap;

		if (copy_from_user(&ucasmap, argp, sizeof(ucasmap))) {
			r = -EFAULT;
			break;
		}

		if (!kvm_is_ucontrol(vcpu->kvm)) {
			r = -EINVAL;
			break;
		}

		r = gmap_map_segment(vcpu->arch.gmap, ucasmap.user_addr,
				     ucasmap.vcpu_addr, ucasmap.length);
		break;
	}
	case KVM_S390_UCAS_UNMAP: {
		struct kvm_s390_ucas_mapping ucasmap;

		if (copy_from_user(&ucasmap, argp, sizeof(ucasmap))) {
			r = -EFAULT;
			break;
		}

		if (!kvm_is_ucontrol(vcpu->kvm)) {
			r = -EINVAL;
			break;
		}

		r = gmap_unmap_segment(vcpu->arch.gmap, ucasmap.vcpu_addr,
			ucasmap.length);
		break;
	}
#endif
2626
	case KVM_S390_VCPU_FAULT: {
2627
		r = gmap_fault(vcpu->arch.gmap, arg, 0);
2628 2629
		break;
	}
2630 2631 2632 2633 2634 2635 2636 2637 2638
	case KVM_ENABLE_CAP:
	{
		struct kvm_enable_cap cap;
		r = -EFAULT;
		if (copy_from_user(&cap, argp, sizeof(cap)))
			break;
		r = kvm_vcpu_ioctl_enable_cap(vcpu, &cap);
		break;
	}
2639 2640 2641 2642 2643 2644 2645 2646 2647
	case KVM_S390_MEM_OP: {
		struct kvm_s390_mem_op mem_op;

		if (copy_from_user(&mem_op, argp, sizeof(mem_op)) == 0)
			r = kvm_s390_guest_mem_op(vcpu, &mem_op);
		else
			r = -EFAULT;
		break;
	}
2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679
	case KVM_S390_SET_IRQ_STATE: {
		struct kvm_s390_irq_state irq_state;

		r = -EFAULT;
		if (copy_from_user(&irq_state, argp, sizeof(irq_state)))
			break;
		if (irq_state.len > VCPU_IRQS_MAX_BUF ||
		    irq_state.len == 0 ||
		    irq_state.len % sizeof(struct kvm_s390_irq) > 0) {
			r = -EINVAL;
			break;
		}
		r = kvm_s390_set_irq_state(vcpu,
					   (void __user *) irq_state.buf,
					   irq_state.len);
		break;
	}
	case KVM_S390_GET_IRQ_STATE: {
		struct kvm_s390_irq_state irq_state;

		r = -EFAULT;
		if (copy_from_user(&irq_state, argp, sizeof(irq_state)))
			break;
		if (irq_state.len == 0) {
			r = -EINVAL;
			break;
		}
		r = kvm_s390_get_irq_state(vcpu,
					   (__u8 __user *)  irq_state.buf,
					   irq_state.len);
		break;
	}
2680
	default:
2681
		r = -ENOTTY;
2682
	}
2683
	return r;
2684 2685
}

2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698
int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
{
#ifdef CONFIG_KVM_S390_UCONTROL
	if ((vmf->pgoff == KVM_S390_SIE_PAGE_OFFSET)
		 && (kvm_is_ucontrol(vcpu->kvm))) {
		vmf->page = virt_to_page(vcpu->arch.sie_block);
		get_page(vmf->page);
		return 0;
	}
#endif
	return VM_FAULT_SIGBUS;
}

2699 2700
int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
			    unsigned long npages)
2701 2702 2703 2704
{
	return 0;
}

2705
/* Section: memory related */
2706 2707
int kvm_arch_prepare_memory_region(struct kvm *kvm,
				   struct kvm_memory_slot *memslot,
2708
				   const struct kvm_userspace_memory_region *mem,
2709
				   enum kvm_mr_change change)
2710
{
2711 2712 2713 2714
	/* A few sanity checks. We can have memory slots which have to be
	   located/ended at a segment boundary (1MB). The memory in userland is
	   ok to be fragmented into various different vmas. It is okay to mmap()
	   and munmap() stuff in this slot after doing this call at any time */
2715

2716
	if (mem->userspace_addr & 0xffffful)
2717 2718
		return -EINVAL;

2719
	if (mem->memory_size & 0xffffful)
2720 2721
		return -EINVAL;

2722 2723 2724 2725
	return 0;
}

void kvm_arch_commit_memory_region(struct kvm *kvm,
2726
				const struct kvm_userspace_memory_region *mem,
2727
				const struct kvm_memory_slot *old,
2728
				const struct kvm_memory_slot *new,
2729
				enum kvm_mr_change change)
2730
{
2731
	int rc;
2732

2733 2734 2735 2736 2737 2738 2739 2740 2741 2742
	/* If the basics of the memslot do not change, we do not want
	 * to update the gmap. Every update causes several unnecessary
	 * segment translation exceptions. This is usually handled just
	 * fine by the normal fault handler + gmap, but it will also
	 * cause faults on the prefix page of running guest CPUs.
	 */
	if (old->userspace_addr == mem->userspace_addr &&
	    old->base_gfn * PAGE_SIZE == mem->guest_phys_addr &&
	    old->npages * PAGE_SIZE == mem->memory_size)
		return;
2743 2744 2745 2746

	rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr,
		mem->guest_phys_addr, mem->memory_size);
	if (rc)
2747
		pr_warn("failed to commit memory region\n");
2748
	return;
2749 2750 2751 2752
}

static int __init kvm_s390_init(void)
{
2753
	return kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE);
2754 2755 2756 2757 2758 2759 2760 2761 2762
}

static void __exit kvm_s390_exit(void)
{
	kvm_exit();
}

module_init(kvm_s390_init);
module_exit(kvm_s390_exit);
2763 2764 2765 2766 2767 2768 2769 2770 2771

/*
 * Enable autoloading of the kvm module.
 * Note that we add the module alias here instead of virt/kvm/kvm_main.c
 * since x86 takes a different approach.
 */
#include <linux/miscdevice.h>
MODULE_ALIAS_MISCDEV(KVM_MINOR);
MODULE_ALIAS("devname:kvm");