kvm-s390.c 83.2 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
#include <linux/init.h>
#include <linux/kvm.h>
#include <linux/kvm_host.h>
24
#include <linux/mman.h>
25
#include <linux/module.h>
26
#include <linux/random.h>
27
#include <linux/slab.h>
28
#include <linux/timer.h>
29
#include <linux/vmalloc.h>
30
#include <linux/bitmap.h>
31
#include <asm/asm-offsets.h>
32
#include <asm/lowcore.h>
33
#include <asm/etr.h>
34
#include <asm/pgtable.h>
35
#include <asm/gmap.h>
36
#include <asm/nmi.h>
37
#include <asm/switch_to.h>
38
#include <asm/isc.h>
39
#include <asm/sclp.h>
40 41
#include <asm/cpacf.h>
#include <asm/etr.h>
42
#include "kvm-s390.h"
43 44
#include "gaccess.h"

45 46 47 48
#define KMSG_COMPONENT "kvm-s390"
#undef pr_fmt
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt

49 50
#define CREATE_TRACE_POINTS
#include "trace.h"
51
#include "trace-s390.h"
52

53
#define MEM_OP_MAX_SIZE 65536	/* Maximum transfer size for KVM_S390_MEM_OP */
54 55 56
#define LOCAL_IRQS 32
#define VCPU_IRQS_MAX_BUF (sizeof(struct kvm_s390_irq) * \
			   (KVM_MAX_VCPUS + LOCAL_IRQS))
57

58 59 60 61
#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) },
62
	{ "exit_null", VCPU_STAT(exit_null) },
63 64 65 66
	{ "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) },
67 68 69
	{ "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) },
70
	{ "exit_operation_exception", VCPU_STAT(exit_operation_exception) },
71
	{ "halt_successful_poll", VCPU_STAT(halt_successful_poll) },
72
	{ "halt_attempted_poll", VCPU_STAT(halt_attempted_poll) },
73
	{ "halt_poll_invalid", VCPU_STAT(halt_poll_invalid) },
74
	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
75
	{ "instruction_lctlg", VCPU_STAT(instruction_lctlg) },
76
	{ "instruction_lctl", VCPU_STAT(instruction_lctl) },
77 78
	{ "instruction_stctl", VCPU_STAT(instruction_stctl) },
	{ "instruction_stctg", VCPU_STAT(instruction_stctg) },
79
	{ "deliver_emergency_signal", VCPU_STAT(deliver_emergency_signal) },
80
	{ "deliver_external_call", VCPU_STAT(deliver_external_call) },
81 82 83 84 85 86 87
	{ "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) },
88
	{ "instruction_pfmf", VCPU_STAT(instruction_pfmf) },
89 90 91 92 93
	{ "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) },
94
	{ "instruction_ipte_interlock", VCPU_STAT(instruction_ipte_interlock) },
95 96
	{ "instruction_stsch", VCPU_STAT(instruction_stsch) },
	{ "instruction_chsc", VCPU_STAT(instruction_chsc) },
97
	{ "instruction_essa", VCPU_STAT(instruction_essa) },
98 99
	{ "instruction_stsi", VCPU_STAT(instruction_stsi) },
	{ "instruction_stfl", VCPU_STAT(instruction_stfl) },
100
	{ "instruction_tprot", VCPU_STAT(instruction_tprot) },
J
Janosch Frank 已提交
101
	{ "instruction_sthyi", VCPU_STAT(instruction_sthyi) },
102
	{ "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) },
103
	{ "instruction_sigp_sense_running", VCPU_STAT(instruction_sigp_sense_running) },
104
	{ "instruction_sigp_external_call", VCPU_STAT(instruction_sigp_external_call) },
105
	{ "instruction_sigp_emergency", VCPU_STAT(instruction_sigp_emergency) },
106 107
	{ "instruction_sigp_cond_emergency", VCPU_STAT(instruction_sigp_cond_emergency) },
	{ "instruction_sigp_start", VCPU_STAT(instruction_sigp_start) },
108
	{ "instruction_sigp_stop", VCPU_STAT(instruction_sigp_stop) },
109 110
	{ "instruction_sigp_stop_store_status", VCPU_STAT(instruction_sigp_stop_store_status) },
	{ "instruction_sigp_store_status", VCPU_STAT(instruction_sigp_store_status) },
111
	{ "instruction_sigp_store_adtl_status", VCPU_STAT(instruction_sigp_store_adtl_status) },
112 113 114
	{ "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) },
115 116 117
	{ "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) },
118
	{ "diagnose_10", VCPU_STAT(diagnose_10) },
119
	{ "diagnose_44", VCPU_STAT(diagnose_44) },
120
	{ "diagnose_9c", VCPU_STAT(diagnose_9c) },
121 122 123
	{ "diagnose_258", VCPU_STAT(diagnose_258) },
	{ "diagnose_308", VCPU_STAT(diagnose_308) },
	{ "diagnose_500", VCPU_STAT(diagnose_500) },
124 125 126
	{ NULL }
};

127
/* upper facilities limit for kvm */
128 129 130
unsigned long kvm_s390_fac_list_mask[16] = {
	0xffe6000000000000UL,
	0x005e000000000000UL,
131
};
132

133
unsigned long kvm_s390_fac_list_mask_size(void)
134
{
135 136
	BUILD_BUG_ON(ARRAY_SIZE(kvm_s390_fac_list_mask) > S390_ARCH_FAC_MASK_SIZE_U64);
	return ARRAY_SIZE(kvm_s390_fac_list_mask);
137 138
}

139 140
/* available cpu features supported by kvm */
static DECLARE_BITMAP(kvm_s390_available_cpu_feat, KVM_S390_VM_CPU_FEAT_NR_BITS);
141 142
/* available subfunctions indicated via query / "test bit" */
static struct kvm_s390_vm_cpu_subfunc kvm_s390_available_subfunc;
143

144
static struct gmap_notifier gmap_notifier;
145
debug_info_t *kvm_s390_dbf;
146

147
/* Section: not file related */
148
int kvm_arch_hardware_enable(void)
149 150
{
	/* every s390 is virtualization enabled ;-) */
151
	return 0;
152 153
}

154 155
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start,
			      unsigned long end);
156

157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174
/*
 * 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;
175 176
			if (vcpu->arch.cputm_enabled)
				vcpu->arch.cputm_start += *delta;
177 178 179 180 181 182 183 184 185
		}
	}
	return NOTIFY_OK;
}

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

186 187
int kvm_arch_hardware_setup(void)
{
188
	gmap_notifier.notifier_call = kvm_gmap_notifier;
189
	gmap_register_pte_notifier(&gmap_notifier);
190 191
	atomic_notifier_chain_register(&s390_epoch_delta_notifier,
				       &kvm_clock_notifier);
192 193 194 195 196
	return 0;
}

void kvm_arch_hardware_unsetup(void)
{
197
	gmap_unregister_pte_notifier(&gmap_notifier);
198 199
	atomic_notifier_chain_unregister(&s390_epoch_delta_notifier,
					 &kvm_clock_notifier);
200 201
}

202 203 204 205 206
static void allow_cpu_feat(unsigned long nr)
{
	set_bit_inv(nr, kvm_s390_available_cpu_feat);
}

207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222
static inline int plo_test_bit(unsigned char nr)
{
	register unsigned long r0 asm("0") = (unsigned long) nr | 0x100;
	int cc = 3; /* subfunction not available */

	asm volatile(
		/* Parameter registers are ignored for "test bit" */
		"	plo	0,0,0,0(0)\n"
		"	ipm	%0\n"
		"	srl	%0,28\n"
		: "=d" (cc)
		: "d" (r0)
		: "cc");
	return cc == 0;
}

223 224
static void kvm_s390_cpu_feat_init(void)
{
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252
	int i;

	for (i = 0; i < 256; ++i) {
		if (plo_test_bit(i))
			kvm_s390_available_subfunc.plo[i >> 3] |= 0x80 >> (i & 7);
	}

	if (test_facility(28)) /* TOD-clock steering */
		etr_ptff(kvm_s390_available_subfunc.ptff, ETR_PTFF_QAF);

	if (test_facility(17)) { /* MSA */
		__cpacf_query(CPACF_KMAC, kvm_s390_available_subfunc.kmac);
		__cpacf_query(CPACF_KMC, kvm_s390_available_subfunc.kmc);
		__cpacf_query(CPACF_KM, kvm_s390_available_subfunc.km);
		__cpacf_query(CPACF_KIMD, kvm_s390_available_subfunc.kimd);
		__cpacf_query(CPACF_KLMD, kvm_s390_available_subfunc.klmd);
	}
	if (test_facility(76)) /* MSA3 */
		__cpacf_query(CPACF_PCKMO, kvm_s390_available_subfunc.pckmo);
	if (test_facility(77)) { /* MSA4 */
		__cpacf_query(CPACF_KMCTR, kvm_s390_available_subfunc.kmctr);
		__cpacf_query(CPACF_KMF, kvm_s390_available_subfunc.kmf);
		__cpacf_query(CPACF_KMO, kvm_s390_available_subfunc.kmo);
		__cpacf_query(CPACF_PCC, kvm_s390_available_subfunc.pcc);
	}
	if (test_facility(57)) /* MSA5 */
		__cpacf_query(CPACF_PPNO, kvm_s390_available_subfunc.ppno);

253 254 255 256
	if (MACHINE_HAS_ESOP)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_ESOP);
}

257 258
int kvm_arch_init(void *opaque)
{
259 260 261 262 263 264 265 266 267
	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;
	}

268 269
	kvm_s390_cpu_feat_init();

270 271
	/* Register floating interrupt controller interface. */
	return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
272 273
}

274 275 276 277 278
void kvm_arch_exit(void)
{
	debug_unregister(kvm_s390_dbf);
}

279 280 281 282 283 284 285 286 287
/* 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;
}

288
int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
289
{
290 291
	int r;

292
	switch (ext) {
293
	case KVM_CAP_S390_PSW:
294
	case KVM_CAP_S390_GMAP:
295
	case KVM_CAP_SYNC_MMU:
296 297 298
#ifdef CONFIG_KVM_S390_UCONTROL
	case KVM_CAP_S390_UCONTROL:
#endif
299
	case KVM_CAP_ASYNC_PF:
300
	case KVM_CAP_SYNC_REGS:
301
	case KVM_CAP_ONE_REG:
302
	case KVM_CAP_ENABLE_CAP:
303
	case KVM_CAP_S390_CSS_SUPPORT:
C
Cornelia Huck 已提交
304
	case KVM_CAP_IOEVENTFD:
305
	case KVM_CAP_DEVICE_CTRL:
306
	case KVM_CAP_ENABLE_CAP_VM:
307
	case KVM_CAP_S390_IRQCHIP:
308
	case KVM_CAP_VM_ATTRIBUTES:
309
	case KVM_CAP_MP_STATE:
310
	case KVM_CAP_S390_INJECT_IRQ:
311
	case KVM_CAP_S390_USER_SIGP:
312
	case KVM_CAP_S390_USER_STSI:
313
	case KVM_CAP_S390_SKEYS:
314
	case KVM_CAP_S390_IRQ_STATE:
315 316
		r = 1;
		break;
317 318 319
	case KVM_CAP_S390_MEM_OP:
		r = MEM_OP_MAX_SIZE;
		break;
320 321
	case KVM_CAP_NR_VCPUS:
	case KVM_CAP_MAX_VCPUS:
322 323 324
		r = KVM_S390_BSCA_CPU_SLOTS;
		if (sclp.has_esca && sclp.has_64bscao)
			r = KVM_S390_ESCA_CPU_SLOTS;
325
		break;
326 327 328
	case KVM_CAP_NR_MEMSLOTS:
		r = KVM_USER_MEM_SLOTS;
		break;
329
	case KVM_CAP_S390_COW:
330
		r = MACHINE_HAS_ESOP;
331
		break;
332 333 334
	case KVM_CAP_S390_VECTOR_REGISTERS:
		r = MACHINE_HAS_VX;
		break;
335 336 337
	case KVM_CAP_S390_RI:
		r = test_facility(64);
		break;
338
	default:
339
		r = 0;
340
	}
341
	return r;
342 343
}

344 345 346 347 348 349 350 351 352 353 354 355
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;

	/* 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);

356
		if (test_and_clear_guest_dirty(gmap->mm, address))
357
			mark_page_dirty(kvm, cur_gfn);
358 359
		if (fatal_signal_pending(current))
			return;
360
		cond_resched();
361 362 363
	}
}

364
/* Section: vm related */
365 366
static void sca_del_vcpu(struct kvm_vcpu *vcpu);

367 368 369 370 371 372
/*
 * 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)
{
373 374
	int r;
	unsigned long n;
375
	struct kvm_memslots *slots;
376 377 378 379 380 381 382 383 384
	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;

385 386
	slots = kvm_memslots(kvm);
	memslot = id_to_memslot(slots, log->slot);
387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404
	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;
405 406
}

407 408 409 410 411 412 413 414
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) {
415
	case KVM_CAP_S390_IRQCHIP:
416
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_IRQCHIP");
417 418 419
		kvm->arch.use_irqchip = 1;
		r = 0;
		break;
420
	case KVM_CAP_S390_USER_SIGP:
421
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_SIGP");
422 423 424
		kvm->arch.user_sigp = 1;
		r = 0;
		break;
425
	case KVM_CAP_S390_VECTOR_REGISTERS:
426
		mutex_lock(&kvm->lock);
427
		if (kvm->created_vcpus) {
428 429
			r = -EBUSY;
		} else if (MACHINE_HAS_VX) {
430 431
			set_kvm_facility(kvm->arch.model.fac_mask, 129);
			set_kvm_facility(kvm->arch.model.fac_list, 129);
432 433 434
			r = 0;
		} else
			r = -EINVAL;
435
		mutex_unlock(&kvm->lock);
436 437
		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_VECTOR_REGISTERS %s",
			 r ? "(not available)" : "(success)");
438
		break;
439 440 441
	case KVM_CAP_S390_RI:
		r = -EINVAL;
		mutex_lock(&kvm->lock);
442
		if (kvm->created_vcpus) {
443 444
			r = -EBUSY;
		} else if (test_facility(64)) {
445 446
			set_kvm_facility(kvm->arch.model.fac_mask, 64);
			set_kvm_facility(kvm->arch.model.fac_list, 64);
447 448 449 450 451 452
			r = 0;
		}
		mutex_unlock(&kvm->lock);
		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_RI %s",
			 r ? "(not available)" : "(success)");
		break;
453
	case KVM_CAP_S390_USER_STSI:
454
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_STSI");
455 456 457
		kvm->arch.user_stsi = 1;
		r = 0;
		break;
458 459 460 461 462 463 464
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

465 466 467 468 469 470 471
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;
472
		VM_EVENT(kvm, 3, "QUERY: max guest memory: %lu bytes",
473 474
			 kvm->arch.mem_limit);
		if (put_user(kvm->arch.mem_limit, (u64 __user *)attr->addr))
475 476 477 478 479 480 481 482 483 484
			ret = -EFAULT;
		break;
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *attr)
485 486 487 488 489
{
	int ret;
	unsigned int idx;
	switch (attr->attr) {
	case KVM_S390_VM_MEM_ENABLE_CMMA:
490
		ret = -ENXIO;
491
		if (!sclp.has_cmma)
492 493
			break;

494
		ret = -EBUSY;
495
		VM_EVENT(kvm, 3, "%s", "ENABLE: CMMA support");
496
		mutex_lock(&kvm->lock);
497
		if (!kvm->created_vcpus) {
498 499 500 501 502 503
			kvm->arch.use_cmma = 1;
			ret = 0;
		}
		mutex_unlock(&kvm->lock);
		break;
	case KVM_S390_VM_MEM_CLR_CMMA:
504 505 506
		ret = -ENXIO;
		if (!sclp.has_cmma)
			break;
507 508 509 510
		ret = -EINVAL;
		if (!kvm->arch.use_cmma)
			break;

511
		VM_EVENT(kvm, 3, "%s", "RESET: CMMA states");
512 513
		mutex_lock(&kvm->lock);
		idx = srcu_read_lock(&kvm->srcu);
514
		s390_reset_cmma(kvm->arch.gmap->mm);
515 516 517 518
		srcu_read_unlock(&kvm->srcu, idx);
		mutex_unlock(&kvm->lock);
		ret = 0;
		break;
519 520 521 522 523 524 525 526 527
	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;

528 529
		if (kvm->arch.mem_limit != KVM_S390_NO_MEM_LIMIT &&
		    new_limit > kvm->arch.mem_limit)
530 531
			return -E2BIG;

532 533 534
		if (!new_limit)
			return -EINVAL;

535
		/* gmap_create takes last usable address */
536 537 538
		if (new_limit != KVM_S390_NO_MEM_LIMIT)
			new_limit -= 1;

539 540
		ret = -EBUSY;
		mutex_lock(&kvm->lock);
541
		if (!kvm->created_vcpus) {
542 543
			/* gmap_create will round the limit up */
			struct gmap *new = gmap_create(current->mm, new_limit);
544 545 546 547

			if (!new) {
				ret = -ENOMEM;
			} else {
548
				gmap_remove(kvm->arch.gmap);
549 550 551 552 553 554
				new->private = kvm;
				kvm->arch.gmap = new;
				ret = 0;
			}
		}
		mutex_unlock(&kvm->lock);
555 556 557
		VM_EVENT(kvm, 3, "SET: max guest address: %lu", new_limit);
		VM_EVENT(kvm, 3, "New guest asce: 0x%pK",
			 (void *) kvm->arch.gmap->asce);
558 559
		break;
	}
560 561 562 563 564 565 566
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

567 568 569 570 571 572 573
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;

574
	if (!test_kvm_facility(kvm, 76))
575 576 577 578 579 580 581 582 583
		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;
584
		VM_EVENT(kvm, 3, "%s", "ENABLE: AES keywrapping support");
585 586 587 588 589 590
		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;
591
		VM_EVENT(kvm, 3, "%s", "ENABLE: DEA keywrapping support");
592 593 594 595 596
		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));
597
		VM_EVENT(kvm, 3, "%s", "DISABLE: AES keywrapping support");
598 599 600 601 602
		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));
603
		VM_EVENT(kvm, 3, "%s", "DISABLE: DEA keywrapping support");
604 605 606 607 608 609 610 611 612 613 614 615 616 617
		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;
}

618 619 620 621 622 623 624 625 626 627
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;
628
	VM_EVENT(kvm, 3, "SET: TOD extension: 0x%x", gtod_high);
629 630 631 632 633 634

	return 0;
}

static int kvm_s390_set_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
635
	u64 gtod;
636 637 638 639

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

640
	kvm_s390_set_tod_clock(kvm, gtod);
641
	VM_EVENT(kvm, 3, "SET: TOD base: 0x%llx", gtod);
642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
	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;
673
	VM_EVENT(kvm, 3, "QUERY: TOD extension: 0x%x", gtod_high);
674 675 676 677 678 679

	return 0;
}

static int kvm_s390_get_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
680
	u64 gtod;
681

682
	gtod = kvm_s390_get_tod_clock_fast(kvm);
683 684
	if (copy_to_user((void __user *)attr->addr, &gtod, sizeof(gtod)))
		return -EFAULT;
685
	VM_EVENT(kvm, 3, "QUERY: TOD base: 0x%llx", gtod);
686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710

	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;
}

711 712 713
static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_processor *proc;
714
	u16 lowest_ibc, unblocked_ibc;
715 716 717
	int ret = 0;

	mutex_lock(&kvm->lock);
718
	if (kvm->created_vcpus) {
719 720 721 722 723 724 725 726 727 728
		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))) {
729
		kvm->arch.model.cpuid = proc->cpuid;
730 731 732 733 734 735 736 737 738 739
		lowest_ibc = sclp.ibc >> 16 & 0xfff;
		unblocked_ibc = sclp.ibc & 0xfff;
		if (lowest_ibc) {
			if (proc->ibc > unblocked_ibc)
				kvm->arch.model.ibc = unblocked_ibc;
			else if (proc->ibc < lowest_ibc)
				kvm->arch.model.ibc = lowest_ibc;
			else
				kvm->arch.model.ibc = proc->ibc;
		}
740
		memcpy(kvm->arch.model.fac_list, proc->fac_list,
741 742 743 744 745 746 747 748 749
		       S390_ARCH_FAC_LIST_SIZE_BYTE);
	} else
		ret = -EFAULT;
	kfree(proc);
out:
	mutex_unlock(&kvm->lock);
	return ret;
}

750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772
static int kvm_s390_set_processor_feat(struct kvm *kvm,
				       struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_feat data;
	int ret = -EBUSY;

	if (copy_from_user(&data, (void __user *)attr->addr, sizeof(data)))
		return -EFAULT;
	if (!bitmap_subset((unsigned long *) data.feat,
			   kvm_s390_available_cpu_feat,
			   KVM_S390_VM_CPU_FEAT_NR_BITS))
		return -EINVAL;

	mutex_lock(&kvm->lock);
	if (!atomic_read(&kvm->online_vcpus)) {
		bitmap_copy(kvm->arch.cpu_feat, (unsigned long *) data.feat,
			    KVM_S390_VM_CPU_FEAT_NR_BITS);
		ret = 0;
	}
	mutex_unlock(&kvm->lock);
	return ret;
}

773 774 775 776 777 778 779 780 781 782
static int kvm_s390_set_processor_subfunc(struct kvm *kvm,
					  struct kvm_device_attr *attr)
{
	/*
	 * Once supported by kernel + hw, we have to store the subfunctions
	 * in kvm->arch and remember that user space configured them.
	 */
	return -ENXIO;
}

783 784 785 786 787 788 789 790
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;
791 792 793
	case KVM_S390_VM_CPU_PROCESSOR_FEAT:
		ret = kvm_s390_set_processor_feat(kvm, attr);
		break;
794 795 796
	case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
		ret = kvm_s390_set_processor_subfunc(kvm, attr);
		break;
797 798 799 800 801 802 803 804 805 806 807 808 809 810
	}
	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;
	}
811
	proc->cpuid = kvm->arch.model.cpuid;
812
	proc->ibc = kvm->arch.model.ibc;
813 814
	memcpy(&proc->fac_list, kvm->arch.model.fac_list,
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832
	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);
833
	mach->ibc = sclp.ibc;
834
	memcpy(&mach->fac_mask, kvm->arch.model.fac_mask,
835
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
836
	memcpy((unsigned long *)&mach->fac_list, S390_lowcore.stfle_fac_list,
837
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
838 839 840 841 842 843 844
	if (copy_to_user((void __user *)attr->addr, mach, sizeof(*mach)))
		ret = -EFAULT;
	kfree(mach);
out:
	return ret;
}

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
static int kvm_s390_get_processor_feat(struct kvm *kvm,
				       struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_feat data;

	bitmap_copy((unsigned long *) data.feat, kvm->arch.cpu_feat,
		    KVM_S390_VM_CPU_FEAT_NR_BITS);
	if (copy_to_user((void __user *)attr->addr, &data, sizeof(data)))
		return -EFAULT;
	return 0;
}

static int kvm_s390_get_machine_feat(struct kvm *kvm,
				     struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_feat data;

	bitmap_copy((unsigned long *) data.feat,
		    kvm_s390_available_cpu_feat,
		    KVM_S390_VM_CPU_FEAT_NR_BITS);
	if (copy_to_user((void __user *)attr->addr, &data, sizeof(data)))
		return -EFAULT;
	return 0;
}

870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888
static int kvm_s390_get_processor_subfunc(struct kvm *kvm,
					  struct kvm_device_attr *attr)
{
	/*
	 * Once we can actually configure subfunctions (kernel + hw support),
	 * we have to check if they were already set by user space, if so copy
	 * them from kvm->arch.
	 */
	return -ENXIO;
}

static int kvm_s390_get_machine_subfunc(struct kvm *kvm,
					struct kvm_device_attr *attr)
{
	if (copy_to_user((void __user *)attr->addr, &kvm_s390_available_subfunc,
	    sizeof(struct kvm_s390_vm_cpu_subfunc)))
		return -EFAULT;
	return 0;
}
889 890 891 892 893 894 895 896 897 898 899
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;
900 901 902 903 904 905
	case KVM_S390_VM_CPU_PROCESSOR_FEAT:
		ret = kvm_s390_get_processor_feat(kvm, attr);
		break;
	case KVM_S390_VM_CPU_MACHINE_FEAT:
		ret = kvm_s390_get_machine_feat(kvm, attr);
		break;
906 907 908 909 910 911
	case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
		ret = kvm_s390_get_processor_subfunc(kvm, attr);
		break;
	case KVM_S390_VM_CPU_MACHINE_SUBFUNC:
		ret = kvm_s390_get_machine_subfunc(kvm, attr);
		break;
912 913 914 915
	}
	return ret;
}

916 917 918 919 920
static int kvm_s390_vm_set_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret;

	switch (attr->group) {
921
	case KVM_S390_VM_MEM_CTRL:
922
		ret = kvm_s390_set_mem_control(kvm, attr);
923
		break;
924 925 926
	case KVM_S390_VM_TOD:
		ret = kvm_s390_set_tod(kvm, attr);
		break;
927 928 929
	case KVM_S390_VM_CPU_MODEL:
		ret = kvm_s390_set_cpu_model(kvm, attr);
		break;
930 931 932
	case KVM_S390_VM_CRYPTO:
		ret = kvm_s390_vm_set_crypto(kvm, attr);
		break;
933 934 935 936 937 938 939 940 941 942
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

static int kvm_s390_vm_get_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
943 944 945 946 947 948
	int ret;

	switch (attr->group) {
	case KVM_S390_VM_MEM_CTRL:
		ret = kvm_s390_get_mem_control(kvm, attr);
		break;
949 950 951
	case KVM_S390_VM_TOD:
		ret = kvm_s390_get_tod(kvm, attr);
		break;
952 953 954
	case KVM_S390_VM_CPU_MODEL:
		ret = kvm_s390_get_cpu_model(kvm, attr);
		break;
955 956 957 958 959 960
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
961 962 963 964 965 966 967
}

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

	switch (attr->group) {
968 969 970 971
	case KVM_S390_VM_MEM_CTRL:
		switch (attr->attr) {
		case KVM_S390_VM_MEM_ENABLE_CMMA:
		case KVM_S390_VM_MEM_CLR_CMMA:
972 973
			ret = sclp.has_cmma ? 0 : -ENXIO;
			break;
974
		case KVM_S390_VM_MEM_LIMIT_SIZE:
975 976 977 978 979 980 981
			ret = 0;
			break;
		default:
			ret = -ENXIO;
			break;
		}
		break;
982 983 984 985 986 987 988 989 990 991 992
	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;
993 994 995 996
	case KVM_S390_VM_CPU_MODEL:
		switch (attr->attr) {
		case KVM_S390_VM_CPU_PROCESSOR:
		case KVM_S390_VM_CPU_MACHINE:
997 998
		case KVM_S390_VM_CPU_PROCESSOR_FEAT:
		case KVM_S390_VM_CPU_MACHINE_FEAT:
999
		case KVM_S390_VM_CPU_MACHINE_SUBFUNC:
1000 1001
			ret = 0;
			break;
1002 1003
		/* configuring subfunctions is not supported yet */
		case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
1004 1005 1006 1007 1008
		default:
			ret = -ENXIO;
			break;
		}
		break;
1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021
	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;
1022 1023 1024 1025 1026 1027 1028 1029
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053
static long kvm_s390_get_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;

	/* 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;

1054
	down_read(&current->mm->mmap_sem);
1055 1056 1057 1058
	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
1059
			break;
1060 1061
		}

1062 1063
		r = get_guest_storage_key(current->mm, hva, &keys[i]);
		if (r)
1064
			break;
1065
	}
1066 1067 1068 1069 1070 1071 1072 1073
	up_read(&current->mm->mmap_sem);

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

	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 */
1107 1108 1109
	r = s390_enable_skey();
	if (r)
		goto out;
1110

1111
	down_read(&current->mm->mmap_sem);
1112 1113 1114 1115
	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
1116
			break;
1117 1118 1119 1120 1121
		}

		/* Lowest order bit is reserved */
		if (keys[i] & 0x01) {
			r = -EINVAL;
1122
			break;
1123 1124
		}

1125
		r = set_guest_storage_key(current->mm, hva, keys[i], 0);
1126
		if (r)
1127
			break;
1128
	}
1129
	up_read(&current->mm->mmap_sem);
1130 1131 1132 1133 1134
out:
	kvfree(keys);
	return r;
}

1135 1136 1137 1138 1139
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;
1140
	struct kvm_device_attr attr;
1141 1142 1143
	int r;

	switch (ioctl) {
1144 1145 1146 1147 1148 1149 1150 1151 1152
	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;
	}
1153 1154 1155 1156 1157 1158 1159 1160
	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;
	}
1161 1162 1163 1164 1165 1166 1167
	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));
1168
			r = kvm_set_irq_routing(kvm, &routing, 0, 0);
1169 1170 1171
		}
		break;
	}
1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192
	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;
	}
1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212
	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;
	}
1213
	default:
1214
		r = -ENOTTY;
1215 1216 1217 1218 1219
	}

	return r;
}

1220 1221 1222
static int kvm_s390_query_ap_config(u8 *config)
{
	u32 fcn_code = 0x04000000UL;
1223
	u32 cc = 0;
1224

1225
	memset(config, 0, 128);
1226 1227 1228 1229
	asm volatile(
		"lgr 0,%1\n"
		"lgr 2,%2\n"
		".long 0xb2af0000\n"		/* PQAP(QCI) */
1230
		"0: ipm %0\n"
1231
		"srl %0,28\n"
1232 1233 1234
		"1:\n"
		EX_TABLE(0b, 1b)
		: "+r" (cc)
1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246
		: "r" (fcn_code), "r" (config)
		: "cc", "0", "2", "memory"
	);

	return cc;
}

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

1247
	if (test_facility(12)) {
1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268
		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;
}

1269
static u64 kvm_s390_get_initial_cpuid(void)
1270
{
1271 1272 1273 1274 1275
	struct cpuid cpuid;

	get_cpu_id(&cpuid);
	cpuid.version = 0xff;
	return *((u64 *) &cpuid);
1276 1277
}

1278
static void kvm_s390_crypto_init(struct kvm *kvm)
1279
{
1280
	if (!test_kvm_facility(kvm, 76))
1281
		return;
1282

1283
	kvm->arch.crypto.crycb = &kvm->arch.sie_page2->crycb;
1284
	kvm_s390_set_crycb_format(kvm);
1285

1286 1287 1288 1289 1290 1291 1292
	/* 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));
1293 1294
}

1295 1296 1297
static void sca_dispose(struct kvm *kvm)
{
	if (kvm->arch.use_esca)
1298
		free_pages_exact(kvm->arch.sca, sizeof(struct esca_block));
1299 1300 1301 1302 1303
	else
		free_page((unsigned long)(kvm->arch.sca));
	kvm->arch.sca = NULL;
}

1304
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
1305
{
1306
	gfp_t alloc_flags = GFP_KERNEL;
1307
	int i, rc;
1308
	char debug_name[16];
1309
	static unsigned long sca_offset;
1310

1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321
	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

1322 1323
	rc = s390_enable_sie();
	if (rc)
1324
		goto out_err;
1325

1326 1327
	rc = -ENOMEM;

J
Janosch Frank 已提交
1328 1329
	ratelimit_state_init(&kvm->arch.sthyi_limit, 5 * HZ, 500);

1330
	kvm->arch.use_esca = 0; /* start with basic SCA */
1331 1332
	if (!sclp.has_64bscao)
		alloc_flags |= GFP_DMA;
1333
	rwlock_init(&kvm->arch.sca_lock);
1334
	kvm->arch.sca = (struct bsca_block *) get_zeroed_page(alloc_flags);
1335
	if (!kvm->arch.sca)
1336
		goto out_err;
1337
	spin_lock(&kvm_lock);
1338
	sca_offset += 16;
1339
	if (sca_offset + sizeof(struct bsca_block) > PAGE_SIZE)
1340
		sca_offset = 0;
1341 1342
	kvm->arch.sca = (struct bsca_block *)
			((char *) kvm->arch.sca + sca_offset);
1343
	spin_unlock(&kvm_lock);
1344 1345 1346

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

1347
	kvm->arch.dbf = debug_register(debug_name, 32, 1, 7 * sizeof(long));
1348
	if (!kvm->arch.dbf)
1349
		goto out_err;
1350

1351 1352 1353
	kvm->arch.sie_page2 =
	     (struct sie_page2 *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
	if (!kvm->arch.sie_page2)
1354
		goto out_err;
1355

1356
	/* Populate the facility mask initially. */
1357
	memcpy(kvm->arch.model.fac_mask, S390_lowcore.stfle_fac_list,
1358
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
1359 1360
	for (i = 0; i < S390_ARCH_FAC_LIST_SIZE_U64; i++) {
		if (i < kvm_s390_fac_list_mask_size())
1361
			kvm->arch.model.fac_mask[i] &= kvm_s390_fac_list_mask[i];
1362
		else
1363
			kvm->arch.model.fac_mask[i] = 0UL;
1364 1365
	}

1366
	/* Populate the facility list initially. */
1367 1368
	kvm->arch.model.fac_list = kvm->arch.sie_page2->fac_list;
	memcpy(kvm->arch.model.fac_list, kvm->arch.model.fac_mask,
1369 1370
	       S390_ARCH_FAC_LIST_SIZE_BYTE);

J
Janosch Frank 已提交
1371 1372 1373
	set_kvm_facility(kvm->arch.model.fac_mask, 74);
	set_kvm_facility(kvm->arch.model.fac_list, 74);

1374
	kvm->arch.model.cpuid = kvm_s390_get_initial_cpuid();
1375
	kvm->arch.model.ibc = sclp.ibc & 0x0fff;
1376

1377
	kvm_s390_crypto_init(kvm);
1378

1379
	spin_lock_init(&kvm->arch.float_int.lock);
1380 1381
	for (i = 0; i < FIRQ_LIST_COUNT; i++)
		INIT_LIST_HEAD(&kvm->arch.float_int.lists[i]);
1382
	init_waitqueue_head(&kvm->arch.ipte_wq);
1383
	mutex_init(&kvm->arch.ipte_mutex);
1384

1385
	debug_register_view(kvm->arch.dbf, &debug_sprintf_view);
1386
	VM_EVENT(kvm, 3, "vm created with type %lu", type);
1387

1388 1389
	if (type & KVM_VM_S390_UCONTROL) {
		kvm->arch.gmap = NULL;
1390
		kvm->arch.mem_limit = KVM_S390_NO_MEM_LIMIT;
1391
	} else {
1392 1393 1394 1395 1396
		if (sclp.hamax == U64_MAX)
			kvm->arch.mem_limit = TASK_MAX_SIZE;
		else
			kvm->arch.mem_limit = min_t(unsigned long, TASK_MAX_SIZE,
						    sclp.hamax + 1);
1397
		kvm->arch.gmap = gmap_create(current->mm, kvm->arch.mem_limit - 1);
1398
		if (!kvm->arch.gmap)
1399
			goto out_err;
1400
		kvm->arch.gmap->private = kvm;
1401
		kvm->arch.gmap->pfault_enabled = 0;
1402
	}
1403 1404

	kvm->arch.css_support = 0;
1405
	kvm->arch.use_irqchip = 0;
1406
	kvm->arch.epoch = 0;
1407

1408
	spin_lock_init(&kvm->arch.start_stop_lock);
1409
	KVM_EVENT(3, "vm 0x%pK created by pid %u", kvm, current->pid);
1410

1411
	return 0;
1412
out_err:
1413
	free_page((unsigned long)kvm->arch.sie_page2);
1414
	debug_unregister(kvm->arch.dbf);
1415
	sca_dispose(kvm);
1416
	KVM_EVENT(3, "creation of vm failed: %d", rc);
1417
	return rc;
1418 1419
}

1420 1421 1422
void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
{
	VCPU_EVENT(vcpu, 3, "%s", "free cpu");
1423
	trace_kvm_s390_destroy_vcpu(vcpu->vcpu_id);
1424
	kvm_s390_clear_local_irqs(vcpu);
1425
	kvm_clear_async_pf_completion_queue(vcpu);
1426
	if (!kvm_is_ucontrol(vcpu->kvm))
1427
		sca_del_vcpu(vcpu);
1428 1429

	if (kvm_is_ucontrol(vcpu->kvm))
1430
		gmap_remove(vcpu->arch.gmap);
1431

1432
	if (vcpu->kvm->arch.use_cmma)
1433
		kvm_s390_vcpu_unsetup_cmma(vcpu);
1434
	free_page((unsigned long)(vcpu->arch.sie_block));
1435

1436
	kvm_vcpu_uninit(vcpu);
1437
	kmem_cache_free(kvm_vcpu_cache, vcpu);
1438 1439 1440 1441 1442
}

static void kvm_free_vcpus(struct kvm *kvm)
{
	unsigned int i;
1443
	struct kvm_vcpu *vcpu;
1444

1445 1446 1447 1448 1449 1450 1451 1452 1453
	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);
1454 1455
}

1456 1457
void kvm_arch_destroy_vm(struct kvm *kvm)
{
1458
	kvm_free_vcpus(kvm);
1459
	sca_dispose(kvm);
1460
	debug_unregister(kvm->arch.dbf);
1461
	free_page((unsigned long)kvm->arch.sie_page2);
1462
	if (!kvm_is_ucontrol(kvm))
1463
		gmap_remove(kvm->arch.gmap);
1464
	kvm_s390_destroy_adapters(kvm);
1465
	kvm_s390_clear_float_irqs(kvm);
1466
	KVM_EVENT(3, "vm 0x%pK destroyed", kvm);
1467 1468 1469
}

/* Section: vcpu related */
1470 1471
static int __kvm_ucontrol_vcpu_init(struct kvm_vcpu *vcpu)
{
1472
	vcpu->arch.gmap = gmap_create(current->mm, -1UL);
1473 1474 1475 1476 1477 1478 1479
	if (!vcpu->arch.gmap)
		return -ENOMEM;
	vcpu->arch.gmap->private = vcpu->kvm;

	return 0;
}

1480 1481
static void sca_del_vcpu(struct kvm_vcpu *vcpu)
{
1482
	read_lock(&vcpu->kvm->arch.sca_lock);
1483 1484
	if (vcpu->kvm->arch.use_esca) {
		struct esca_block *sca = vcpu->kvm->arch.sca;
1485

1486
		clear_bit_inv(vcpu->vcpu_id, (unsigned long *) sca->mcn);
1487
		sca->cpu[vcpu->vcpu_id].sda = 0;
1488 1489 1490 1491
	} else {
		struct bsca_block *sca = vcpu->kvm->arch.sca;

		clear_bit_inv(vcpu->vcpu_id, (unsigned long *) &sca->mcn);
1492
		sca->cpu[vcpu->vcpu_id].sda = 0;
1493
	}
1494
	read_unlock(&vcpu->kvm->arch.sca_lock);
1495 1496
}

1497
static void sca_add_vcpu(struct kvm_vcpu *vcpu)
1498
{
1499 1500 1501
	read_lock(&vcpu->kvm->arch.sca_lock);
	if (vcpu->kvm->arch.use_esca) {
		struct esca_block *sca = vcpu->kvm->arch.sca;
1502

1503
		sca->cpu[vcpu->vcpu_id].sda = (__u64) vcpu->arch.sie_block;
1504 1505
		vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
		vcpu->arch.sie_block->scaol = (__u32)(__u64)sca & ~0x3fU;
1506
		vcpu->arch.sie_block->ecb2 |= 0x04U;
1507
		set_bit_inv(vcpu->vcpu_id, (unsigned long *) sca->mcn);
1508
	} else {
1509
		struct bsca_block *sca = vcpu->kvm->arch.sca;
1510

1511
		sca->cpu[vcpu->vcpu_id].sda = (__u64) vcpu->arch.sie_block;
1512 1513
		vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
		vcpu->arch.sie_block->scaol = (__u32)(__u64)sca;
1514
		set_bit_inv(vcpu->vcpu_id, (unsigned long *) &sca->mcn);
1515
	}
1516
	read_unlock(&vcpu->kvm->arch.sca_lock);
1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 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
}

/* Basic SCA to Extended SCA data copy routines */
static inline void sca_copy_entry(struct esca_entry *d, struct bsca_entry *s)
{
	d->sda = s->sda;
	d->sigp_ctrl.c = s->sigp_ctrl.c;
	d->sigp_ctrl.scn = s->sigp_ctrl.scn;
}

static void sca_copy_b_to_e(struct esca_block *d, struct bsca_block *s)
{
	int i;

	d->ipte_control = s->ipte_control;
	d->mcn[0] = s->mcn;
	for (i = 0; i < KVM_S390_BSCA_CPU_SLOTS; i++)
		sca_copy_entry(&d->cpu[i], &s->cpu[i]);
}

static int sca_switch_to_extended(struct kvm *kvm)
{
	struct bsca_block *old_sca = kvm->arch.sca;
	struct esca_block *new_sca;
	struct kvm_vcpu *vcpu;
	unsigned int vcpu_idx;
	u32 scaol, scaoh;

	new_sca = alloc_pages_exact(sizeof(*new_sca), GFP_KERNEL|__GFP_ZERO);
	if (!new_sca)
		return -ENOMEM;

	scaoh = (u32)((u64)(new_sca) >> 32);
	scaol = (u32)(u64)(new_sca) & ~0x3fU;

	kvm_s390_vcpu_block_all(kvm);
	write_lock(&kvm->arch.sca_lock);

	sca_copy_b_to_e(new_sca, old_sca);

	kvm_for_each_vcpu(vcpu_idx, vcpu, kvm) {
		vcpu->arch.sie_block->scaoh = scaoh;
		vcpu->arch.sie_block->scaol = scaol;
		vcpu->arch.sie_block->ecb2 |= 0x04U;
	}
	kvm->arch.sca = new_sca;
	kvm->arch.use_esca = 1;

	write_unlock(&kvm->arch.sca_lock);
	kvm_s390_vcpu_unblock_all(kvm);

	free_page((unsigned long)old_sca);

1570 1571
	VM_EVENT(kvm, 2, "Switched to ESCA (0x%pK -> 0x%pK)",
		 old_sca, kvm->arch.sca);
1572
	return 0;
1573 1574 1575 1576
}

static int sca_can_add_vcpu(struct kvm *kvm, unsigned int id)
{
1577 1578 1579 1580
	int rc;

	if (id < KVM_S390_BSCA_CPU_SLOTS)
		return true;
1581
	if (!sclp.has_esca || !sclp.has_64bscao)
1582 1583 1584 1585 1586 1587 1588
		return false;

	mutex_lock(&kvm->lock);
	rc = kvm->arch.use_esca ? 0 : sca_switch_to_extended(kvm);
	mutex_unlock(&kvm->lock);

	return rc == 0 && id < KVM_S390_ESCA_CPU_SLOTS;
1589 1590
}

1591 1592
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
1593 1594
	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
	kvm_clear_async_pf_completion_queue(vcpu);
1595 1596
	vcpu->run->kvm_valid_regs = KVM_SYNC_PREFIX |
				    KVM_SYNC_GPRS |
1597
				    KVM_SYNC_ACRS |
1598 1599 1600
				    KVM_SYNC_CRS |
				    KVM_SYNC_ARCH0 |
				    KVM_SYNC_PFAULT;
1601 1602
	if (test_kvm_facility(vcpu->kvm, 64))
		vcpu->run->kvm_valid_regs |= KVM_SYNC_RICCB;
1603 1604 1605 1606
	/* fprs can be synchronized via vrs, even if the guest has no vx. With
	 * MACHINE_HAS_VX, (load|store)_fpu_regs() will work with vrs format.
	 */
	if (MACHINE_HAS_VX)
1607
		vcpu->run->kvm_valid_regs |= KVM_SYNC_VRS;
1608 1609
	else
		vcpu->run->kvm_valid_regs |= KVM_SYNC_FPRS;
1610 1611 1612 1613

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

1614 1615 1616
	return 0;
}

1617 1618 1619 1620
/* needs disabled preemption to protect from TOD sync and vcpu_load/put */
static void __start_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
	WARN_ON_ONCE(vcpu->arch.cputm_start != 0);
1621
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
1622
	vcpu->arch.cputm_start = get_tod_clock_fast();
1623
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
1624 1625 1626 1627 1628 1629
}

/* needs disabled preemption to protect from TOD sync and vcpu_load/put */
static void __stop_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
	WARN_ON_ONCE(vcpu->arch.cputm_start == 0);
1630
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
1631 1632
	vcpu->arch.sie_block->cputm -= get_tod_clock_fast() - vcpu->arch.cputm_start;
	vcpu->arch.cputm_start = 0;
1633
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
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 1662 1663 1664 1665
}

/* needs disabled preemption to protect from TOD sync and vcpu_load/put */
static void __enable_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
	WARN_ON_ONCE(vcpu->arch.cputm_enabled);
	vcpu->arch.cputm_enabled = true;
	__start_cpu_timer_accounting(vcpu);
}

/* needs disabled preemption to protect from TOD sync and vcpu_load/put */
static void __disable_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
	WARN_ON_ONCE(!vcpu->arch.cputm_enabled);
	__stop_cpu_timer_accounting(vcpu);
	vcpu->arch.cputm_enabled = false;
}

static void enable_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
	preempt_disable(); /* protect from TOD sync and vcpu_load/put */
	__enable_cpu_timer_accounting(vcpu);
	preempt_enable();
}

static void disable_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
	preempt_disable(); /* protect from TOD sync and vcpu_load/put */
	__disable_cpu_timer_accounting(vcpu);
	preempt_enable();
}

1666 1667 1668
/* set the cpu timer - may only be called from the VCPU thread itself */
void kvm_s390_set_cpu_timer(struct kvm_vcpu *vcpu, __u64 cputm)
{
1669
	preempt_disable(); /* protect from TOD sync and vcpu_load/put */
1670
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
1671 1672
	if (vcpu->arch.cputm_enabled)
		vcpu->arch.cputm_start = get_tod_clock_fast();
1673
	vcpu->arch.sie_block->cputm = cputm;
1674
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
1675
	preempt_enable();
1676 1677
}

1678
/* update and get the cpu timer - can also be called from other VCPU threads */
1679 1680
__u64 kvm_s390_get_cpu_timer(struct kvm_vcpu *vcpu)
{
1681
	unsigned int seq;
1682 1683 1684 1685 1686
	__u64 value;

	if (unlikely(!vcpu->arch.cputm_enabled))
		return vcpu->arch.sie_block->cputm;

1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700
	preempt_disable(); /* protect from TOD sync and vcpu_load/put */
	do {
		seq = raw_read_seqcount(&vcpu->arch.cputm_seqcount);
		/*
		 * If the writer would ever execute a read in the critical
		 * section, e.g. in irq context, we have a deadlock.
		 */
		WARN_ON_ONCE((seq & 1) && smp_processor_id() == vcpu->cpu);
		value = vcpu->arch.sie_block->cputm;
		/* if cputm_start is 0, accounting is being started/stopped */
		if (likely(vcpu->arch.cputm_start))
			value -= get_tod_clock_fast() - vcpu->arch.cputm_start;
	} while (read_seqcount_retry(&vcpu->arch.cputm_seqcount, seq & ~1));
	preempt_enable();
1701
	return value;
1702 1703
}

1704 1705
void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
1706
	/* Save host register state */
1707
	save_fpu_regs();
1708 1709
	vcpu->arch.host_fpregs.fpc = current->thread.fpu.fpc;
	vcpu->arch.host_fpregs.regs = current->thread.fpu.regs;
1710

1711 1712 1713 1714
	if (MACHINE_HAS_VX)
		current->thread.fpu.regs = vcpu->run->s.regs.vrs;
	else
		current->thread.fpu.regs = vcpu->run->s.regs.fprs;
1715
	current->thread.fpu.fpc = vcpu->run->s.regs.fpc;
1716
	if (test_fp_ctl(current->thread.fpu.fpc))
1717
		/* User space provided an invalid FPC, let's clear it */
1718 1719 1720
		current->thread.fpu.fpc = 0;

	save_access_regs(vcpu->arch.host_acrs);
1721
	restore_access_regs(vcpu->run->s.regs.acrs);
1722
	gmap_enable(vcpu->arch.enabled_gmap);
1723
	atomic_or(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
1724
	if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
1725
		__start_cpu_timer_accounting(vcpu);
1726
	vcpu->cpu = cpu;
1727 1728 1729 1730
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
1731
	vcpu->cpu = -1;
1732
	if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
1733
		__stop_cpu_timer_accounting(vcpu);
1734
	atomic_andnot(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
1735 1736
	vcpu->arch.enabled_gmap = gmap_get_enabled();
	gmap_disable(vcpu->arch.enabled_gmap);
1737

1738
	/* Save guest register state */
1739
	save_fpu_regs();
1740
	vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
1741

1742 1743 1744
	/* Restore host register state */
	current->thread.fpu.fpc = vcpu->arch.host_fpregs.fpc;
	current->thread.fpu.regs = vcpu->arch.host_fpregs.regs;
1745 1746

	save_access_regs(vcpu->run->s.regs.acrs);
1747 1748 1749 1750 1751 1752 1753 1754
	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;
1755
	kvm_s390_set_prefix(vcpu, 0);
1756
	kvm_s390_set_cpu_timer(vcpu, 0);
1757 1758 1759 1760 1761
	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;
1762 1763 1764
	/* make sure the new fpc will be lazily loaded */
	save_fpu_regs();
	current->thread.fpu.fpc = 0;
1765
	vcpu->arch.sie_block->gbea = 1;
1766
	vcpu->arch.sie_block->pp = 0;
1767 1768
	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
	kvm_clear_async_pf_completion_queue(vcpu);
1769 1770
	if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm))
		kvm_s390_vcpu_stop(vcpu);
1771
	kvm_s390_clear_local_irqs(vcpu);
1772 1773
}

1774
void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
1775
{
1776
	mutex_lock(&vcpu->kvm->lock);
1777
	preempt_disable();
1778
	vcpu->arch.sie_block->epoch = vcpu->kvm->arch.epoch;
1779
	preempt_enable();
1780
	mutex_unlock(&vcpu->kvm->lock);
1781
	if (!kvm_is_ucontrol(vcpu->kvm)) {
1782
		vcpu->arch.gmap = vcpu->kvm->arch.gmap;
1783
		sca_add_vcpu(vcpu);
1784
	}
1785 1786
	/* make vcpu_load load the right gmap on the first trigger */
	vcpu->arch.enabled_gmap = vcpu->arch.gmap;
1787 1788
}

1789 1790
static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu)
{
1791
	if (!test_kvm_facility(vcpu->kvm, 76))
1792 1793
		return;

1794 1795 1796 1797 1798 1799 1800
	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;

1801 1802 1803
	vcpu->arch.sie_block->crycbd = vcpu->kvm->arch.crypto.crycbd;
}

1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820
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;
}

1821 1822 1823 1824 1825
static void kvm_s390_vcpu_setup_model(struct kvm_vcpu *vcpu)
{
	struct kvm_s390_cpu_model *model = &vcpu->kvm->arch.model;

	vcpu->arch.sie_block->ibc = model->ibc;
1826
	if (test_kvm_facility(vcpu->kvm, 7))
1827
		vcpu->arch.sie_block->fac = (u32)(u64) model->fac_list;
1828 1829
}

1830 1831
int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
{
1832
	int rc = 0;
1833

1834 1835
	atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH |
						    CPUSTAT_SM |
1836 1837
						    CPUSTAT_STOPPED);

1838
	if (test_kvm_facility(vcpu->kvm, 78))
1839
		atomic_or(CPUSTAT_GED2, &vcpu->arch.sie_block->cpuflags);
1840
	else if (test_kvm_facility(vcpu->kvm, 8))
1841
		atomic_or(CPUSTAT_GED, &vcpu->arch.sie_block->cpuflags);
1842

1843 1844
	kvm_s390_vcpu_setup_model(vcpu);

1845 1846 1847
	/* pgste_set_pte has special handling for !MACHINE_HAS_ESOP */
	if (MACHINE_HAS_ESOP)
		vcpu->arch.sie_block->ecb |= 0x02;
1848 1849
	if (test_kvm_facility(vcpu->kvm, 9))
		vcpu->arch.sie_block->ecb |= 0x04;
1850
	if (test_kvm_facility(vcpu->kvm, 73))
1851 1852
		vcpu->arch.sie_block->ecb |= 0x10;

1853
	if (test_kvm_facility(vcpu->kvm, 8) && sclp.has_pfmfi)
1854
		vcpu->arch.sie_block->ecb2 |= 0x08;
1855 1856 1857
	vcpu->arch.sie_block->eca = 0x1002000U;
	if (sclp.has_cei)
		vcpu->arch.sie_block->eca |= 0x80000000U;
1858 1859
	if (sclp.has_ib)
		vcpu->arch.sie_block->eca |= 0x40000000U;
1860
	if (sclp.has_siif)
1861
		vcpu->arch.sie_block->eca |= 1;
1862
	if (sclp.has_sigpif)
1863
		vcpu->arch.sie_block->eca |= 0x10000000U;
1864 1865
	if (test_kvm_facility(vcpu->kvm, 64))
		vcpu->arch.sie_block->ecb3 |= 0x01;
1866
	if (test_kvm_facility(vcpu->kvm, 129)) {
1867 1868 1869
		vcpu->arch.sie_block->eca |= 0x00020000;
		vcpu->arch.sie_block->ecd |= 0x20000000;
	}
1870
	vcpu->arch.sie_block->riccbd = (unsigned long) &vcpu->run->s.regs.riccb;
1871
	vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE;
J
Janosch Frank 已提交
1872 1873
	if (test_kvm_facility(vcpu->kvm, 74))
		vcpu->arch.sie_block->ictl |= ICTL_OPEREXC;
1874

1875
	if (vcpu->kvm->arch.use_cmma) {
1876 1877 1878
		rc = kvm_s390_vcpu_setup_cmma(vcpu);
		if (rc)
			return rc;
1879
	}
1880
	hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
1881
	vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup;
1882

1883 1884
	kvm_s390_vcpu_crypto_setup(vcpu);

1885
	return rc;
1886 1887 1888 1889 1890
}

struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
				      unsigned int id)
{
1891
	struct kvm_vcpu *vcpu;
1892
	struct sie_page *sie_page;
1893 1894
	int rc = -EINVAL;

1895
	if (!kvm_is_ucontrol(kvm) && !sca_can_add_vcpu(kvm, id))
1896 1897 1898
		goto out;

	rc = -ENOMEM;
1899

1900
	vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
1901
	if (!vcpu)
1902
		goto out;
1903

1904 1905
	sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
	if (!sie_page)
1906 1907
		goto out_free_cpu;

1908 1909 1910
	vcpu->arch.sie_block = &sie_page->sie_block;
	vcpu->arch.sie_block->itdba = (unsigned long) &sie_page->itdb;

1911 1912 1913 1914
	/* the real guest size will always be smaller than msl */
	vcpu->arch.sie_block->mso = 0;
	vcpu->arch.sie_block->msl = sclp.hamax;

1915
	vcpu->arch.sie_block->icpua = id;
1916 1917
	spin_lock_init(&vcpu->arch.local_int.lock);
	vcpu->arch.local_int.float_int = &kvm->arch.float_int;
1918
	vcpu->arch.local_int.wq = &vcpu->wq;
1919
	vcpu->arch.local_int.cpuflags = &vcpu->arch.sie_block->cpuflags;
1920
	seqcount_init(&vcpu->arch.cputm_seqcount);
1921

1922 1923
	rc = kvm_vcpu_init(vcpu, kvm, id);
	if (rc)
1924
		goto out_free_sie_block;
1925
	VM_EVENT(kvm, 3, "create cpu %d at 0x%pK, sie block at 0x%pK", id, vcpu,
1926
		 vcpu->arch.sie_block);
1927
	trace_kvm_s390_create_vcpu(id, vcpu, vcpu->arch.sie_block);
1928 1929

	return vcpu;
1930 1931
out_free_sie_block:
	free_page((unsigned long)(vcpu->arch.sie_block));
1932
out_free_cpu:
1933
	kmem_cache_free(kvm_vcpu_cache, vcpu);
1934
out:
1935 1936 1937 1938 1939
	return ERR_PTR(rc);
}

int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
{
1940
	return kvm_s390_vcpu_has_irq(vcpu, 0);
1941 1942
}

1943
void kvm_s390_vcpu_block(struct kvm_vcpu *vcpu)
1944
{
1945
	atomic_or(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
1946
	exit_sie(vcpu);
1947 1948
}

1949
void kvm_s390_vcpu_unblock(struct kvm_vcpu *vcpu)
1950
{
1951
	atomic_andnot(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
1952 1953
}

1954 1955
static void kvm_s390_vcpu_request(struct kvm_vcpu *vcpu)
{
1956
	atomic_or(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
1957
	exit_sie(vcpu);
1958 1959 1960 1961
}

static void kvm_s390_vcpu_request_handled(struct kvm_vcpu *vcpu)
{
1962
	atomic_andnot(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
1963 1964
}

1965 1966 1967 1968 1969 1970
/*
 * 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)
{
1971
	atomic_or(CPUSTAT_STOP_INT, &vcpu->arch.sie_block->cpuflags);
1972 1973 1974 1975
	while (vcpu->arch.sie_block->prog0c & PROG_IN_SIE)
		cpu_relax();
}

1976 1977
/* Kick a guest cpu out of SIE to process a request synchronously */
void kvm_s390_sync_request(int req, struct kvm_vcpu *vcpu)
1978
{
1979 1980
	kvm_make_request(req, vcpu);
	kvm_s390_vcpu_request(vcpu);
1981 1982
}

1983 1984
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start,
			      unsigned long end)
1985 1986 1987
{
	struct kvm *kvm = gmap->private;
	struct kvm_vcpu *vcpu;
1988 1989
	unsigned long prefix;
	int i;
1990

1991 1992
	if (gmap_is_shadow(gmap))
		return;
1993 1994 1995
	if (start >= 1UL << 31)
		/* We are only interested in prefix pages */
		return;
1996 1997
	kvm_for_each_vcpu(i, vcpu, kvm) {
		/* match against both prefix pages */
1998 1999 2000 2001
		prefix = kvm_s390_get_prefix(vcpu);
		if (prefix <= end && start <= prefix + 2*PAGE_SIZE - 1) {
			VCPU_EVENT(vcpu, 2, "gmap notifier for %lx-%lx",
				   start, end);
2002
			kvm_s390_sync_request(KVM_REQ_MMU_RELOAD, vcpu);
2003 2004 2005 2006
		}
	}
}

2007 2008 2009 2010 2011 2012 2013
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
	/* kvm common code refers to this, but never calls it */
	BUG();
	return 0;
}

2014 2015 2016 2017 2018 2019
static int kvm_arch_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu,
					   struct kvm_one_reg *reg)
{
	int r = -EINVAL;

	switch (reg->id) {
2020 2021 2022 2023 2024 2025 2026 2027
	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;
2028
	case KVM_REG_S390_CPU_TIMER:
2029
		r = put_user(kvm_s390_get_cpu_timer(vcpu),
2030 2031 2032 2033 2034 2035
			     (u64 __user *)reg->addr);
		break;
	case KVM_REG_S390_CLOCK_COMP:
		r = put_user(vcpu->arch.sie_block->ckc,
			     (u64 __user *)reg->addr);
		break;
2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047
	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;
2048 2049 2050 2051
	case KVM_REG_S390_PP:
		r = put_user(vcpu->arch.sie_block->pp,
			     (u64 __user *)reg->addr);
		break;
2052 2053 2054 2055
	case KVM_REG_S390_GBEA:
		r = put_user(vcpu->arch.sie_block->gbea,
			     (u64 __user *)reg->addr);
		break;
2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066
	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;
2067
	__u64 val;
2068 2069

	switch (reg->id) {
2070 2071 2072 2073 2074 2075 2076 2077
	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;
2078
	case KVM_REG_S390_CPU_TIMER:
2079 2080 2081
		r = get_user(val, (u64 __user *)reg->addr);
		if (!r)
			kvm_s390_set_cpu_timer(vcpu, val);
2082 2083 2084 2085 2086
		break;
	case KVM_REG_S390_CLOCK_COMP:
		r = get_user(vcpu->arch.sie_block->ckc,
			     (u64 __user *)reg->addr);
		break;
2087 2088 2089
	case KVM_REG_S390_PFTOKEN:
		r = get_user(vcpu->arch.pfault_token,
			     (u64 __user *)reg->addr);
2090 2091
		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
			kvm_clear_async_pf_completion_queue(vcpu);
2092 2093 2094 2095 2096 2097 2098 2099 2100
		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;
2101 2102 2103 2104
	case KVM_REG_S390_PP:
		r = get_user(vcpu->arch.sie_block->pp,
			     (u64 __user *)reg->addr);
		break;
2105 2106 2107 2108
	case KVM_REG_S390_GBEA:
		r = get_user(vcpu->arch.sie_block->gbea,
			     (u64 __user *)reg->addr);
		break;
2109 2110 2111 2112 2113 2114
	default:
		break;
	}

	return r;
}
2115

2116 2117 2118 2119 2120 2121 2122 2123
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)
{
2124
	memcpy(&vcpu->run->s.regs.gprs, &regs->gprs, sizeof(regs->gprs));
2125 2126 2127 2128 2129
	return 0;
}

int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
2130
	memcpy(&regs->gprs, &vcpu->run->s.regs.gprs, sizeof(regs->gprs));
2131 2132 2133 2134 2135 2136
	return 0;
}

int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
2137
	memcpy(&vcpu->run->s.regs.acrs, &sregs->acrs, sizeof(sregs->acrs));
2138
	memcpy(&vcpu->arch.sie_block->gcr, &sregs->crs, sizeof(sregs->crs));
2139
	restore_access_regs(vcpu->run->s.regs.acrs);
2140 2141 2142 2143 2144 2145
	return 0;
}

int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
2146
	memcpy(&sregs->acrs, &vcpu->run->s.regs.acrs, sizeof(sregs->acrs));
2147 2148 2149 2150 2151 2152
	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)
{
2153 2154
	/* make sure the new values will be lazily loaded */
	save_fpu_regs();
2155 2156
	if (test_fp_ctl(fpu->fpc))
		return -EINVAL;
2157 2158 2159 2160 2161
	current->thread.fpu.fpc = fpu->fpc;
	if (MACHINE_HAS_VX)
		convert_fp_to_vx(current->thread.fpu.vxrs, (freg_t *)fpu->fprs);
	else
		memcpy(current->thread.fpu.fprs, &fpu->fprs, sizeof(fpu->fprs));
2162 2163 2164 2165 2166
	return 0;
}

int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
2167 2168 2169 2170 2171 2172 2173
	/* make sure we have the latest values */
	save_fpu_regs();
	if (MACHINE_HAS_VX)
		convert_vx_to_fp((freg_t *)fpu->fprs, current->thread.fpu.vxrs);
	else
		memcpy(fpu->fprs, current->thread.fpu.fprs, sizeof(fpu->fprs));
	fpu->fpc = current->thread.fpu.fpc;
2174 2175 2176 2177 2178 2179 2180
	return 0;
}

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

2181
	if (!is_vcpu_stopped(vcpu))
2182
		rc = -EBUSY;
2183 2184 2185 2186
	else {
		vcpu->run->psw_mask = psw.mask;
		vcpu->run->psw_addr = psw.addr;
	}
2187 2188 2189 2190 2191 2192 2193 2194 2195
	return rc;
}

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

2196 2197 2198 2199
#define VALID_GUESTDBG_FLAGS (KVM_GUESTDBG_SINGLESTEP | \
			      KVM_GUESTDBG_USE_HW_BP | \
			      KVM_GUESTDBG_ENABLE)

J
Jan Kiszka 已提交
2200 2201
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
					struct kvm_guest_debug *dbg)
2202
{
2203 2204 2205 2206 2207
	int rc = 0;

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

2208
	if (dbg->control & ~VALID_GUESTDBG_FLAGS)
2209
		return -EINVAL;
2210 2211
	if (!sclp.has_gpere)
		return -EINVAL;
2212 2213 2214 2215

	if (dbg->control & KVM_GUESTDBG_ENABLE) {
		vcpu->guest_debug = dbg->control;
		/* enforce guest PER */
2216
		atomic_or(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
2217 2218 2219 2220

		if (dbg->control & KVM_GUESTDBG_USE_HW_BP)
			rc = kvm_s390_import_bp_data(vcpu, dbg);
	} else {
2221
		atomic_andnot(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
2222 2223 2224 2225 2226 2227
		vcpu->arch.guestdbg.last_bp = 0;
	}

	if (rc) {
		vcpu->guest_debug = 0;
		kvm_s390_clear_bp_data(vcpu);
2228
		atomic_andnot(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
2229 2230 2231
	}

	return rc;
2232 2233
}

2234 2235 2236
int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
2237 2238 2239
	/* CHECK_STOP and LOAD are not supported yet */
	return is_vcpu_stopped(vcpu) ? KVM_MP_STATE_STOPPED :
				       KVM_MP_STATE_OPERATING;
2240 2241 2242 2243 2244
}

int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264
	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;
2265 2266
}

2267 2268 2269 2270 2271
static bool ibs_enabled(struct kvm_vcpu *vcpu)
{
	return atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_IBS;
}

2272 2273
static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu)
{
2274
retry:
2275
	kvm_s390_vcpu_request_handled(vcpu);
2276 2277
	if (!vcpu->requests)
		return 0;
2278 2279
	/*
	 * We use MMU_RELOAD just to re-arm the ipte notifier for the
2280
	 * guest prefix page. gmap_mprotect_notify will wait on the ptl lock.
2281 2282 2283 2284
	 * 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.
	 */
2285
	if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) {
2286
		int rc;
2287 2288 2289
		rc = gmap_mprotect_notify(vcpu->arch.gmap,
					  kvm_s390_get_prefix(vcpu),
					  PAGE_SIZE * 2, PROT_WRITE);
2290 2291
		if (rc)
			return rc;
2292
		goto retry;
2293
	}
2294

2295 2296 2297 2298 2299
	if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) {
		vcpu->arch.sie_block->ihcpu = 0xffff;
		goto retry;
	}

2300 2301 2302
	if (kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu)) {
		if (!ibs_enabled(vcpu)) {
			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 1);
2303
			atomic_or(CPUSTAT_IBS,
2304 2305 2306
					&vcpu->arch.sie_block->cpuflags);
		}
		goto retry;
2307
	}
2308 2309 2310 2311

	if (kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu)) {
		if (ibs_enabled(vcpu)) {
			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 0);
2312
			atomic_andnot(CPUSTAT_IBS,
2313 2314 2315 2316 2317
					  &vcpu->arch.sie_block->cpuflags);
		}
		goto retry;
	}

2318 2319 2320
	/* nothing to do, just clear the request */
	clear_bit(KVM_REQ_UNHALT, &vcpu->requests);

2321 2322 2323
	return 0;
}

2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339
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);
}

2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350
/**
 * 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)
2351
{
2352 2353
	return gmap_fault(vcpu->arch.gmap, gpa,
			  writable ? FAULT_FLAG_WRITE : 0);
2354 2355
}

2356 2357 2358 2359
static void __kvm_inject_pfault_token(struct kvm_vcpu *vcpu, bool start_token,
				      unsigned long token)
{
	struct kvm_s390_interrupt inti;
2360
	struct kvm_s390_irq irq;
2361 2362

	if (start_token) {
2363 2364 2365
		irq.u.ext.ext_params2 = token;
		irq.type = KVM_S390_INT_PFAULT_INIT;
		WARN_ON_ONCE(kvm_s390_inject_vcpu(vcpu, &irq));
2366 2367
	} else {
		inti.type = KVM_S390_INT_PFAULT_DONE;
2368
		inti.parm64 = token;
2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414
		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;
2415
	if (kvm_s390_vcpu_has_irq(vcpu, 0))
2416 2417 2418 2419 2420 2421
		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
2422 2423 2424
	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))
2425 2426 2427 2428 2429 2430
		return 0;

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

2431
static int vcpu_pre_run(struct kvm_vcpu *vcpu)
2432
{
2433
	int rc, cpuflags;
2434

2435 2436 2437 2438 2439 2440 2441
	/*
	 * 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);

2442 2443
	vcpu->arch.sie_block->gg14 = vcpu->run->s.regs.gprs[14];
	vcpu->arch.sie_block->gg15 = vcpu->run->s.regs.gprs[15];
2444 2445 2446 2447

	if (need_resched())
		schedule();

2448
	if (test_cpu_flag(CIF_MCCK_PENDING))
2449 2450
		s390_handle_mcck();

2451 2452 2453 2454 2455
	if (!kvm_is_ucontrol(vcpu->kvm)) {
		rc = kvm_s390_deliver_pending_interrupts(vcpu);
		if (rc)
			return rc;
	}
C
Carsten Otte 已提交
2456

2457 2458 2459 2460
	rc = kvm_s390_handle_requests(vcpu);
	if (rc)
		return rc;

2461 2462 2463 2464 2465
	if (guestdbg_enabled(vcpu)) {
		kvm_s390_backup_guest_per_regs(vcpu);
		kvm_s390_patch_guest_per_regs(vcpu);
	}

2466
	vcpu->arch.sie_block->icptcode = 0;
2467 2468 2469
	cpuflags = atomic_read(&vcpu->arch.sie_block->cpuflags);
	VCPU_EVENT(vcpu, 6, "entering sie flags %x", cpuflags);
	trace_kvm_s390_sie_enter(vcpu, cpuflags);
2470

2471 2472 2473
	return 0;
}

2474 2475
static int vcpu_post_run_fault_in_sie(struct kvm_vcpu *vcpu)
{
2476 2477 2478 2479
	struct kvm_s390_pgm_info pgm_info = {
		.code = PGM_ADDRESSING,
	};
	u8 opcode, ilen;
2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492
	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.
	 */
2493
	rc = read_guest_instr(vcpu, &opcode, 1);
2494
	ilen = insn_length(opcode);
2495 2496 2497 2498 2499 2500 2501 2502 2503 2504
	if (rc < 0) {
		return rc;
	} else if (rc) {
		/* Instruction-Fetching Exceptions - we can't detect the ilen.
		 * Forward by arbitrary ilc, injection will take care of
		 * nullification if necessary.
		 */
		pgm_info = vcpu->arch.pgm;
		ilen = 4;
	}
2505 2506 2507
	pgm_info.flags = ilen | KVM_S390_PGM_FLAGS_ILC_VALID;
	kvm_s390_forward_psw(vcpu, ilen);
	return kvm_s390_inject_prog_irq(vcpu, &pgm_info);
2508 2509
}

2510 2511
static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason)
{
2512 2513 2514 2515
	VCPU_EVENT(vcpu, 6, "exit sie icptcode %d",
		   vcpu->arch.sie_block->icptcode);
	trace_kvm_s390_sie_exit(vcpu, vcpu->arch.sie_block->icptcode);

2516 2517 2518
	if (guestdbg_enabled(vcpu))
		kvm_s390_restore_guest_per_regs(vcpu);

2519 2520
	vcpu->run->s.regs.gprs[14] = vcpu->arch.sie_block->gg14;
	vcpu->run->s.regs.gprs[15] = vcpu->arch.sie_block->gg15;
2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534

	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;
2535 2536 2537 2538 2539
	} 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;
2540
		return -EREMOTE;
2541
	} else if (current->thread.gmap_pfault) {
2542
		trace_kvm_s390_major_guest_pfault(vcpu);
2543
		current->thread.gmap_pfault = 0;
2544 2545 2546
		if (kvm_arch_setup_async_pf(vcpu))
			return 0;
		return kvm_arch_fault_in_page(vcpu, current->thread.gmap_addr, 1);
2547
	}
2548
	return vcpu_post_run_fault_in_sie(vcpu);
2549 2550 2551 2552 2553 2554
}

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

2555 2556 2557 2558 2559 2560
	/*
	 * 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);

2561 2562 2563 2564
	do {
		rc = vcpu_pre_run(vcpu);
		if (rc)
			break;
2565

2566
		srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
2567 2568 2569 2570
		/*
		 * As PF_VCPU will be used in fault handler, between
		 * guest_enter and guest_exit should be no uaccess.
		 */
2571 2572
		local_irq_disable();
		__kvm_guest_enter();
2573
		__disable_cpu_timer_accounting(vcpu);
2574
		local_irq_enable();
2575 2576
		exit_reason = sie64a(vcpu->arch.sie_block,
				     vcpu->run->s.regs.gprs);
2577
		local_irq_disable();
2578
		__enable_cpu_timer_accounting(vcpu);
2579 2580
		__kvm_guest_exit();
		local_irq_enable();
2581
		vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
2582 2583

		rc = vcpu_post_run(vcpu, exit_reason);
2584
	} while (!signal_pending(current) && !guestdbg_exit_pending(vcpu) && !rc);
2585

2586
	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
2587
	return rc;
2588 2589
}

2590 2591 2592 2593 2594 2595 2596 2597
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);
2598 2599
		/* some control register changes require a tlb flush */
		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
2600 2601
	}
	if (kvm_run->kvm_dirty_regs & KVM_SYNC_ARCH0) {
2602
		kvm_s390_set_cpu_timer(vcpu, kvm_run->s.regs.cputm);
2603 2604 2605 2606 2607 2608 2609 2610 2611
		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;
2612 2613
		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
			kvm_clear_async_pf_completion_queue(vcpu);
2614 2615 2616 2617 2618 2619 2620 2621 2622 2623
	}
	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);
2624
	kvm_run->s.regs.cputm = kvm_s390_get_cpu_timer(vcpu);
2625 2626 2627 2628 2629 2630 2631 2632 2633
	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;
}

2634 2635
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
2636
	int rc;
2637 2638
	sigset_t sigsaved;

2639 2640 2641 2642 2643
	if (guestdbg_exit_pending(vcpu)) {
		kvm_s390_prepare_debug_exit(vcpu);
		return 0;
	}

2644 2645 2646
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);

2647 2648 2649
	if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) {
		kvm_s390_vcpu_start(vcpu);
	} else if (is_vcpu_stopped(vcpu)) {
2650
		pr_err_ratelimited("can't run stopped vcpu %d\n",
2651 2652 2653
				   vcpu->vcpu_id);
		return -EINVAL;
	}
2654

2655
	sync_regs(vcpu, kvm_run);
2656
	enable_cpu_timer_accounting(vcpu);
2657

2658
	might_fault();
2659
	rc = __vcpu_run(vcpu);
2660

2661 2662
	if (signal_pending(current) && !rc) {
		kvm_run->exit_reason = KVM_EXIT_INTR;
2663
		rc = -EINTR;
2664
	}
2665

2666 2667 2668 2669 2670
	if (guestdbg_exit_pending(vcpu) && !rc)  {
		kvm_s390_prepare_debug_exit(vcpu);
		rc = 0;
	}

2671
	if (rc == -EREMOTE) {
2672
		/* userspace support is needed, kvm_run has been prepared */
2673 2674
		rc = 0;
	}
2675

2676
	disable_cpu_timer_accounting(vcpu);
2677
	store_regs(vcpu, kvm_run);
2678

2679 2680 2681 2682
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &sigsaved, NULL);

	vcpu->stat.exit_userspace++;
2683
	return rc;
2684 2685 2686 2687 2688 2689 2690 2691
}

/*
 * 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
 */
2692
int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long gpa)
2693
{
2694
	unsigned char archmode = 1;
2695
	freg_t fprs[NUM_FPRS];
2696
	unsigned int px;
2697
	u64 clkcomp, cputm;
2698
	int rc;
2699

2700
	px = kvm_s390_get_prefix(vcpu);
2701 2702
	if (gpa == KVM_S390_STORE_STATUS_NOADDR) {
		if (write_guest_abs(vcpu, 163, &archmode, 1))
2703
			return -EFAULT;
2704
		gpa = 0;
2705 2706
	} else if (gpa == KVM_S390_STORE_STATUS_PREFIXED) {
		if (write_guest_real(vcpu, 163, &archmode, 1))
2707
			return -EFAULT;
2708 2709 2710
		gpa = px;
	} else
		gpa -= __LC_FPREGS_SAVE_AREA;
2711 2712 2713

	/* manually convert vector registers if necessary */
	if (MACHINE_HAS_VX) {
2714
		convert_vx_to_fp(fprs, (__vector128 *) vcpu->run->s.regs.vrs);
2715 2716 2717 2718
		rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
				     fprs, 128);
	} else {
		rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
2719
				     vcpu->run->s.regs.fprs, 128);
2720
	}
2721
	rc |= write_guest_abs(vcpu, gpa + __LC_GPREGS_SAVE_AREA,
2722
			      vcpu->run->s.regs.gprs, 128);
2723
	rc |= write_guest_abs(vcpu, gpa + __LC_PSW_SAVE_AREA,
2724
			      &vcpu->arch.sie_block->gpsw, 16);
2725
	rc |= write_guest_abs(vcpu, gpa + __LC_PREFIX_SAVE_AREA,
2726
			      &px, 4);
2727
	rc |= write_guest_abs(vcpu, gpa + __LC_FP_CREG_SAVE_AREA,
2728
			      &vcpu->run->s.regs.fpc, 4);
2729
	rc |= write_guest_abs(vcpu, gpa + __LC_TOD_PROGREG_SAVE_AREA,
2730
			      &vcpu->arch.sie_block->todpr, 4);
2731
	cputm = kvm_s390_get_cpu_timer(vcpu);
2732
	rc |= write_guest_abs(vcpu, gpa + __LC_CPU_TIMER_SAVE_AREA,
2733
			      &cputm, 8);
2734
	clkcomp = vcpu->arch.sie_block->ckc >> 8;
2735
	rc |= write_guest_abs(vcpu, gpa + __LC_CLOCK_COMP_SAVE_AREA,
2736
			      &clkcomp, 8);
2737
	rc |= write_guest_abs(vcpu, gpa + __LC_AREGS_SAVE_AREA,
2738
			      &vcpu->run->s.regs.acrs, 64);
2739
	rc |= write_guest_abs(vcpu, gpa + __LC_CREGS_SAVE_AREA,
2740 2741
			      &vcpu->arch.sie_block->gcr, 128);
	return rc ? -EFAULT : 0;
2742 2743
}

2744 2745 2746 2747 2748 2749 2750
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
	 */
2751
	save_fpu_regs();
2752
	vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
2753 2754 2755 2756 2757
	save_access_regs(vcpu->run->s.regs.acrs);

	return kvm_s390_store_status_unloaded(vcpu, addr);
}

E
Eric Farman 已提交
2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778
/*
 * 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
2779 2780 2781 2782 2783
	 * 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
2784
	 */
2785
	save_fpu_regs();
E
Eric Farman 已提交
2786 2787 2788 2789

	return kvm_s390_store_adtl_status_unloaded(vcpu, addr);
}

2790 2791 2792
static void __disable_ibs_on_vcpu(struct kvm_vcpu *vcpu)
{
	kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu);
2793
	kvm_s390_sync_request(KVM_REQ_DISABLE_IBS, vcpu);
2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807
}

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)
{
2808 2809
	if (!sclp.has_ibs)
		return;
2810
	kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu);
2811
	kvm_s390_sync_request(KVM_REQ_ENABLE_IBS, vcpu);
2812 2813
}

2814 2815
void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu)
{
2816 2817 2818 2819 2820
	int i, online_vcpus, started_vcpus = 0;

	if (!is_vcpu_stopped(vcpu))
		return;

2821
	trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 1);
2822
	/* Only one cpu at a time may enter/leave the STOPPED state. */
2823
	spin_lock(&vcpu->kvm->arch.start_stop_lock);
2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842
	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);
	}

2843
	atomic_andnot(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
2844 2845 2846 2847
	/*
	 * Another VCPU might have used IBS while we were offline.
	 * Let's play safe and flush the VCPU at startup.
	 */
2848
	kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
2849
	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
2850
	return;
2851 2852 2853 2854
}

void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu)
{
2855 2856 2857 2858 2859 2860
	int i, online_vcpus, started_vcpus = 0;
	struct kvm_vcpu *started_vcpu = NULL;

	if (is_vcpu_stopped(vcpu))
		return;

2861
	trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 0);
2862
	/* Only one cpu at a time may enter/leave the STOPPED state. */
2863
	spin_lock(&vcpu->kvm->arch.start_stop_lock);
2864 2865
	online_vcpus = atomic_read(&vcpu->kvm->online_vcpus);

2866
	/* SIGP STOP and SIGP STOP AND STORE STATUS has been fully processed */
2867
	kvm_s390_clear_stop_irq(vcpu);
2868

2869
	atomic_or(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886
	__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);
	}

2887
	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
2888
	return;
2889 2890
}

2891 2892 2893 2894 2895 2896 2897 2898 2899
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) {
2900 2901 2902
	case KVM_CAP_S390_CSS_SUPPORT:
		if (!vcpu->kvm->arch.css_support) {
			vcpu->kvm->arch.css_support = 1;
2903
			VM_EVENT(vcpu->kvm, 3, "%s", "ENABLE: CSS support");
2904 2905 2906 2907
			trace_kvm_s390_enable_css(vcpu->kvm);
		}
		r = 0;
		break;
2908 2909 2910 2911 2912 2913 2914
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940
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) {
2941 2942
			r = check_gva_range(vcpu, mop->gaddr, mop->ar,
					    mop->size, GACC_FETCH);
2943 2944 2945 2946 2947 2948 2949 2950 2951 2952
			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) {
2953 2954
			r = check_gva_range(vcpu, mop->gaddr, mop->ar,
					    mop->size, GACC_STORE);
2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975
			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;
}

2976 2977 2978 2979 2980
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;
2981
	int idx;
2982
	long r;
2983

2984
	switch (ioctl) {
2985 2986 2987 2988 2989 2990 2991 2992 2993
	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;
	}
2994
	case KVM_S390_INTERRUPT: {
2995
		struct kvm_s390_interrupt s390int;
2996
		struct kvm_s390_irq s390irq;
2997

2998
		r = -EFAULT;
2999
		if (copy_from_user(&s390int, argp, sizeof(s390int)))
3000
			break;
3001 3002 3003
		if (s390int_to_s390irq(&s390int, &s390irq))
			return -EINVAL;
		r = kvm_s390_inject_vcpu(vcpu, &s390irq);
3004
		break;
3005
	}
3006
	case KVM_S390_STORE_STATUS:
3007
		idx = srcu_read_lock(&vcpu->kvm->srcu);
3008
		r = kvm_s390_vcpu_store_status(vcpu, arg);
3009
		srcu_read_unlock(&vcpu->kvm->srcu, idx);
3010
		break;
3011 3012 3013
	case KVM_S390_SET_INITIAL_PSW: {
		psw_t psw;

3014
		r = -EFAULT;
3015
		if (copy_from_user(&psw, argp, sizeof(psw)))
3016 3017 3018
			break;
		r = kvm_arch_vcpu_ioctl_set_initial_psw(vcpu, psw);
		break;
3019 3020
	}
	case KVM_S390_INITIAL_RESET:
3021 3022
		r = kvm_arch_vcpu_ioctl_initial_reset(vcpu);
		break;
3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034
	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;
	}
3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070
#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
3071
	case KVM_S390_VCPU_FAULT: {
3072
		r = gmap_fault(vcpu->arch.gmap, arg, 0);
3073 3074
		break;
	}
3075 3076 3077 3078 3079 3080 3081 3082 3083
	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;
	}
3084 3085 3086 3087 3088 3089 3090 3091 3092
	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;
	}
3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124
	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;
	}
3125
	default:
3126
		r = -ENOTTY;
3127
	}
3128
	return r;
3129 3130
}

3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143
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;
}

3144 3145
int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
			    unsigned long npages)
3146 3147 3148 3149
{
	return 0;
}

3150
/* Section: memory related */
3151 3152
int kvm_arch_prepare_memory_region(struct kvm *kvm,
				   struct kvm_memory_slot *memslot,
3153
				   const struct kvm_userspace_memory_region *mem,
3154
				   enum kvm_mr_change change)
3155
{
3156 3157 3158 3159
	/* 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 */
3160

3161
	if (mem->userspace_addr & 0xffffful)
3162 3163
		return -EINVAL;

3164
	if (mem->memory_size & 0xffffful)
3165 3166
		return -EINVAL;

3167 3168 3169
	if (mem->guest_phys_addr + mem->memory_size > kvm->arch.mem_limit)
		return -EINVAL;

3170 3171 3172 3173
	return 0;
}

void kvm_arch_commit_memory_region(struct kvm *kvm,
3174
				const struct kvm_userspace_memory_region *mem,
3175
				const struct kvm_memory_slot *old,
3176
				const struct kvm_memory_slot *new,
3177
				enum kvm_mr_change change)
3178
{
3179
	int rc;
3180

3181 3182 3183 3184 3185 3186 3187 3188 3189 3190
	/* 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;
3191 3192 3193 3194

	rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr,
		mem->guest_phys_addr, mem->memory_size);
	if (rc)
3195
		pr_warn("failed to commit memory region\n");
3196
	return;
3197 3198
}

3199 3200 3201 3202 3203 3204 3205
static inline unsigned long nonhyp_mask(int i)
{
	unsigned int nonhyp_fai = (sclp.hmfai << i * 2) >> 30;

	return 0x0000ffffffffffffUL >> (nonhyp_fai << 4);
}

3206 3207 3208 3209 3210
void kvm_arch_vcpu_block_finish(struct kvm_vcpu *vcpu)
{
	vcpu->valid_wakeup = false;
}

3211 3212
static int __init kvm_s390_init(void)
{
3213 3214
	int i;

3215 3216 3217 3218 3219
	if (!sclp.has_sief2) {
		pr_info("SIE not available\n");
		return -ENODEV;
	}

3220 3221 3222 3223
	for (i = 0; i < 16; i++)
		kvm_s390_fac_list_mask[i] |=
			S390_lowcore.stfle_fac_list[i] & nonhyp_mask(i);

3224
	return kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE);
3225 3226 3227 3228 3229 3230 3231 3232 3233
}

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

module_init(kvm_s390_init);
module_exit(kvm_s390_exit);
3234 3235 3236 3237 3238 3239 3240 3241 3242

/*
 * 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");