kvm-s390.c 86.0 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/stp.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
#include <asm/cpacf.h>
41
#include <asm/timex.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
	{ "exit_instruction", VCPU_STAT(exit_instruction) },
68
	{ "exit_pei", VCPU_STAT(exit_pei) },
69 70
	{ "exit_program_interruption", VCPU_STAT(exit_program_interruption) },
	{ "exit_instr_and_program_int", VCPU_STAT(exit_instr_and_program) },
71
	{ "exit_operation_exception", VCPU_STAT(exit_operation_exception) },
72
	{ "halt_successful_poll", VCPU_STAT(halt_successful_poll) },
73
	{ "halt_attempted_poll", VCPU_STAT(halt_attempted_poll) },
74
	{ "halt_poll_invalid", VCPU_STAT(halt_poll_invalid) },
75
	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
76
	{ "instruction_lctlg", VCPU_STAT(instruction_lctlg) },
77
	{ "instruction_lctl", VCPU_STAT(instruction_lctl) },
78 79
	{ "instruction_stctl", VCPU_STAT(instruction_stctl) },
	{ "instruction_stctg", VCPU_STAT(instruction_stctg) },
80
	{ "deliver_emergency_signal", VCPU_STAT(deliver_emergency_signal) },
81
	{ "deliver_external_call", VCPU_STAT(deliver_external_call) },
82 83 84 85 86 87 88
	{ "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) },
89
	{ "instruction_pfmf", VCPU_STAT(instruction_pfmf) },
90 91 92 93 94
	{ "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) },
95
	{ "instruction_ipte_interlock", VCPU_STAT(instruction_ipte_interlock) },
96 97
	{ "instruction_stsch", VCPU_STAT(instruction_stsch) },
	{ "instruction_chsc", VCPU_STAT(instruction_chsc) },
98
	{ "instruction_essa", VCPU_STAT(instruction_essa) },
99 100
	{ "instruction_stsi", VCPU_STAT(instruction_stsi) },
	{ "instruction_stfl", VCPU_STAT(instruction_stfl) },
101
	{ "instruction_tprot", VCPU_STAT(instruction_tprot) },
J
Janosch Frank 已提交
102
	{ "instruction_sthyi", VCPU_STAT(instruction_sthyi) },
103
	{ "instruction_sie", VCPU_STAT(instruction_sie) },
104
	{ "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) },
105
	{ "instruction_sigp_sense_running", VCPU_STAT(instruction_sigp_sense_running) },
106
	{ "instruction_sigp_external_call", VCPU_STAT(instruction_sigp_external_call) },
107
	{ "instruction_sigp_emergency", VCPU_STAT(instruction_sigp_emergency) },
108 109
	{ "instruction_sigp_cond_emergency", VCPU_STAT(instruction_sigp_cond_emergency) },
	{ "instruction_sigp_start", VCPU_STAT(instruction_sigp_start) },
110
	{ "instruction_sigp_stop", VCPU_STAT(instruction_sigp_stop) },
111 112
	{ "instruction_sigp_stop_store_status", VCPU_STAT(instruction_sigp_stop_store_status) },
	{ "instruction_sigp_store_status", VCPU_STAT(instruction_sigp_store_status) },
113
	{ "instruction_sigp_store_adtl_status", VCPU_STAT(instruction_sigp_store_adtl_status) },
114 115 116
	{ "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) },
117 118 119
	{ "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) },
120
	{ "diagnose_10", VCPU_STAT(diagnose_10) },
121
	{ "diagnose_44", VCPU_STAT(diagnose_44) },
122
	{ "diagnose_9c", VCPU_STAT(diagnose_9c) },
123 124 125
	{ "diagnose_258", VCPU_STAT(diagnose_258) },
	{ "diagnose_308", VCPU_STAT(diagnose_308) },
	{ "diagnose_500", VCPU_STAT(diagnose_500) },
126 127 128
	{ NULL }
};

129 130 131 132 133
/* allow nested virtualization in KVM (if enabled by user space) */
static int nested;
module_param(nested, int, S_IRUGO);
MODULE_PARM_DESC(nested, "Nested virtualization support");

134
/* upper facilities limit for kvm */
135 136 137
unsigned long kvm_s390_fac_list_mask[16] = {
	0xffe6000000000000UL,
	0x005e000000000000UL,
138
};
139

140
unsigned long kvm_s390_fac_list_mask_size(void)
141
{
142 143
	BUILD_BUG_ON(ARRAY_SIZE(kvm_s390_fac_list_mask) > S390_ARCH_FAC_MASK_SIZE_U64);
	return ARRAY_SIZE(kvm_s390_fac_list_mask);
144 145
}

146 147
/* available cpu features supported by kvm */
static DECLARE_BITMAP(kvm_s390_available_cpu_feat, KVM_S390_VM_CPU_FEAT_NR_BITS);
148 149
/* available subfunctions indicated via query / "test bit" */
static struct kvm_s390_vm_cpu_subfunc kvm_s390_available_subfunc;
150

151
static struct gmap_notifier gmap_notifier;
152
static struct gmap_notifier vsie_gmap_notifier;
153
debug_info_t *kvm_s390_dbf;
154

155
/* Section: not file related */
156
int kvm_arch_hardware_enable(void)
157 158
{
	/* every s390 is virtualization enabled ;-) */
159
	return 0;
160 161
}

162 163
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start,
			      unsigned long end);
164

165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182
/*
 * 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;
183 184
			if (vcpu->arch.cputm_enabled)
				vcpu->arch.cputm_start += *delta;
185 186
			if (vcpu->arch.vsie_block)
				vcpu->arch.vsie_block->epoch -= *delta;
187 188 189 190 191 192 193 194 195
		}
	}
	return NOTIFY_OK;
}

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

196 197
int kvm_arch_hardware_setup(void)
{
198
	gmap_notifier.notifier_call = kvm_gmap_notifier;
199
	gmap_register_pte_notifier(&gmap_notifier);
200 201
	vsie_gmap_notifier.notifier_call = kvm_s390_vsie_gmap_notifier;
	gmap_register_pte_notifier(&vsie_gmap_notifier);
202 203
	atomic_notifier_chain_register(&s390_epoch_delta_notifier,
				       &kvm_clock_notifier);
204 205 206 207 208
	return 0;
}

void kvm_arch_hardware_unsetup(void)
{
209
	gmap_unregister_pte_notifier(&gmap_notifier);
210
	gmap_unregister_pte_notifier(&vsie_gmap_notifier);
211 212
	atomic_notifier_chain_unregister(&s390_epoch_delta_notifier,
					 &kvm_clock_notifier);
213 214
}

215 216 217 218 219
static void allow_cpu_feat(unsigned long nr)
{
	set_bit_inv(nr, kvm_s390_available_cpu_feat);
}

220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
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;
}

236 237
static void kvm_s390_cpu_feat_init(void)
{
238 239 240 241 242 243 244 245
	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 */
246 247 248
		ptff(kvm_s390_available_subfunc.ptff,
		     sizeof(kvm_s390_available_subfunc.ptff),
		     PTFF_QAF);
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267

	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);

268 269
	if (MACHINE_HAS_ESOP)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_ESOP);
270 271 272 273 274
	/*
	 * We need SIE support, ESOP (PROT_READ protection for gmap_shadow),
	 * 64bit SCAO (SCA passthrough) and IDTE (for gmap_shadow unshadowing).
	 */
	if (!sclp.has_sief2 || !MACHINE_HAS_ESOP || !sclp.has_64bscao ||
275
	    !test_facility(3) || !nested)
276 277
		return;
	allow_cpu_feat(KVM_S390_VM_CPU_FEAT_SIEF2);
278 279
	if (sclp.has_64bscao)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_64BSCAO);
280 281
	if (sclp.has_siif)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_SIIF);
282 283
	if (sclp.has_gpere)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_GPERE);
284 285
	if (sclp.has_gsls)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_GSLS);
286 287
	if (sclp.has_ib)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_IB);
288 289
	if (sclp.has_cei)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_CEI);
290 291
	if (sclp.has_ibs)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_IBS);
292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309
	/*
	 * KVM_S390_VM_CPU_FEAT_SKEY: Wrong shadow of PTE.I bits will make
	 * all skey handling functions read/set the skey from the PGSTE
	 * instead of the real storage key.
	 *
	 * KVM_S390_VM_CPU_FEAT_CMMA: Wrong shadow of PTE.I bits will make
	 * pages being detected as preserved although they are resident.
	 *
	 * KVM_S390_VM_CPU_FEAT_PFMFI: Wrong shadow of PTE.I bits will
	 * have the same effect as for KVM_S390_VM_CPU_FEAT_SKEY.
	 *
	 * For KVM_S390_VM_CPU_FEAT_SKEY, KVM_S390_VM_CPU_FEAT_CMMA and
	 * KVM_S390_VM_CPU_FEAT_PFMFI, all PTE.I and PGSTE bits have to be
	 * correctly shadowed. We can do that for the PGSTE but not for PTE.I.
	 *
	 * KVM_S390_VM_CPU_FEAT_SIGPIF: Wrong SCB addresses in the SCA. We
	 * cannot easily shadow the SCA because of the ipte lock.
	 */
310 311
}

312 313
int kvm_arch_init(void *opaque)
{
314 315 316 317 318 319 320 321 322
	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;
	}

323 324
	kvm_s390_cpu_feat_init();

325 326
	/* Register floating interrupt controller interface. */
	return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
327 328
}

329 330 331 332 333
void kvm_arch_exit(void)
{
	debug_unregister(kvm_s390_dbf);
}

334 335 336 337 338 339 340 341 342
/* 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;
}

343
int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
344
{
345 346
	int r;

347
	switch (ext) {
348
	case KVM_CAP_S390_PSW:
349
	case KVM_CAP_S390_GMAP:
350
	case KVM_CAP_SYNC_MMU:
351 352 353
#ifdef CONFIG_KVM_S390_UCONTROL
	case KVM_CAP_S390_UCONTROL:
#endif
354
	case KVM_CAP_ASYNC_PF:
355
	case KVM_CAP_SYNC_REGS:
356
	case KVM_CAP_ONE_REG:
357
	case KVM_CAP_ENABLE_CAP:
358
	case KVM_CAP_S390_CSS_SUPPORT:
C
Cornelia Huck 已提交
359
	case KVM_CAP_IOEVENTFD:
360
	case KVM_CAP_DEVICE_CTRL:
361
	case KVM_CAP_ENABLE_CAP_VM:
362
	case KVM_CAP_S390_IRQCHIP:
363
	case KVM_CAP_VM_ATTRIBUTES:
364
	case KVM_CAP_MP_STATE:
365
	case KVM_CAP_S390_INJECT_IRQ:
366
	case KVM_CAP_S390_USER_SIGP:
367
	case KVM_CAP_S390_USER_STSI:
368
	case KVM_CAP_S390_SKEYS:
369
	case KVM_CAP_S390_IRQ_STATE:
370
	case KVM_CAP_S390_USER_INSTR0:
371 372
		r = 1;
		break;
373 374 375
	case KVM_CAP_S390_MEM_OP:
		r = MEM_OP_MAX_SIZE;
		break;
376 377
	case KVM_CAP_NR_VCPUS:
	case KVM_CAP_MAX_VCPUS:
378
		r = KVM_S390_BSCA_CPU_SLOTS;
379 380 381
		if (!kvm_s390_use_sca_entries())
			r = KVM_MAX_VCPUS;
		else if (sclp.has_esca && sclp.has_64bscao)
382
			r = KVM_S390_ESCA_CPU_SLOTS;
383
		break;
384 385 386
	case KVM_CAP_NR_MEMSLOTS:
		r = KVM_USER_MEM_SLOTS;
		break;
387
	case KVM_CAP_S390_COW:
388
		r = MACHINE_HAS_ESOP;
389
		break;
390 391 392
	case KVM_CAP_S390_VECTOR_REGISTERS:
		r = MACHINE_HAS_VX;
		break;
393 394 395
	case KVM_CAP_S390_RI:
		r = test_facility(64);
		break;
396
	default:
397
		r = 0;
398
	}
399
	return r;
400 401
}

402 403 404 405 406 407 408 409 410 411 412 413
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);

414
		if (test_and_clear_guest_dirty(gmap->mm, address))
415
			mark_page_dirty(kvm, cur_gfn);
416 417
		if (fatal_signal_pending(current))
			return;
418
		cond_resched();
419 420 421
	}
}

422
/* Section: vm related */
423 424
static void sca_del_vcpu(struct kvm_vcpu *vcpu);

425 426 427 428 429 430
/*
 * 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)
{
431 432
	int r;
	unsigned long n;
433
	struct kvm_memslots *slots;
434 435 436 437 438 439 440 441 442
	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;

443 444
	slots = kvm_memslots(kvm);
	memslot = id_to_memslot(slots, log->slot);
445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462
	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;
463 464
}

465 466 467 468 469 470 471 472 473 474
static void icpt_operexc_on_all_vcpus(struct kvm *kvm)
{
	unsigned int i;
	struct kvm_vcpu *vcpu;

	kvm_for_each_vcpu(i, vcpu, kvm) {
		kvm_s390_sync_request(KVM_REQ_ICPT_OPEREXC, vcpu);
	}
}

475 476 477 478 479 480 481 482
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) {
483
	case KVM_CAP_S390_IRQCHIP:
484
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_IRQCHIP");
485 486 487
		kvm->arch.use_irqchip = 1;
		r = 0;
		break;
488
	case KVM_CAP_S390_USER_SIGP:
489
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_SIGP");
490 491 492
		kvm->arch.user_sigp = 1;
		r = 0;
		break;
493
	case KVM_CAP_S390_VECTOR_REGISTERS:
494
		mutex_lock(&kvm->lock);
495
		if (kvm->created_vcpus) {
496 497
			r = -EBUSY;
		} else if (MACHINE_HAS_VX) {
498 499
			set_kvm_facility(kvm->arch.model.fac_mask, 129);
			set_kvm_facility(kvm->arch.model.fac_list, 129);
500 501 502
			r = 0;
		} else
			r = -EINVAL;
503
		mutex_unlock(&kvm->lock);
504 505
		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_VECTOR_REGISTERS %s",
			 r ? "(not available)" : "(success)");
506
		break;
507 508 509
	case KVM_CAP_S390_RI:
		r = -EINVAL;
		mutex_lock(&kvm->lock);
510
		if (kvm->created_vcpus) {
511 512
			r = -EBUSY;
		} else if (test_facility(64)) {
513 514
			set_kvm_facility(kvm->arch.model.fac_mask, 64);
			set_kvm_facility(kvm->arch.model.fac_list, 64);
515 516 517 518 519 520
			r = 0;
		}
		mutex_unlock(&kvm->lock);
		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_RI %s",
			 r ? "(not available)" : "(success)");
		break;
521
	case KVM_CAP_S390_USER_STSI:
522
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_STSI");
523 524 525
		kvm->arch.user_stsi = 1;
		r = 0;
		break;
526 527 528 529 530 531
	case KVM_CAP_S390_USER_INSTR0:
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_INSTR0");
		kvm->arch.user_instr0 = 1;
		icpt_operexc_on_all_vcpus(kvm);
		r = 0;
		break;
532 533 534 535 536 537 538
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

539 540 541 542 543 544 545
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;
546
		VM_EVENT(kvm, 3, "QUERY: max guest memory: %lu bytes",
547 548
			 kvm->arch.mem_limit);
		if (put_user(kvm->arch.mem_limit, (u64 __user *)attr->addr))
549 550 551 552 553 554 555 556 557 558
			ret = -EFAULT;
		break;
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *attr)
559 560 561 562 563
{
	int ret;
	unsigned int idx;
	switch (attr->attr) {
	case KVM_S390_VM_MEM_ENABLE_CMMA:
564
		ret = -ENXIO;
565
		if (!sclp.has_cmma)
566 567
			break;

568
		ret = -EBUSY;
569
		VM_EVENT(kvm, 3, "%s", "ENABLE: CMMA support");
570
		mutex_lock(&kvm->lock);
571
		if (!kvm->created_vcpus) {
572 573 574 575 576 577
			kvm->arch.use_cmma = 1;
			ret = 0;
		}
		mutex_unlock(&kvm->lock);
		break;
	case KVM_S390_VM_MEM_CLR_CMMA:
578 579 580
		ret = -ENXIO;
		if (!sclp.has_cmma)
			break;
581 582 583 584
		ret = -EINVAL;
		if (!kvm->arch.use_cmma)
			break;

585
		VM_EVENT(kvm, 3, "%s", "RESET: CMMA states");
586 587
		mutex_lock(&kvm->lock);
		idx = srcu_read_lock(&kvm->srcu);
588
		s390_reset_cmma(kvm->arch.gmap->mm);
589 590 591 592
		srcu_read_unlock(&kvm->srcu, idx);
		mutex_unlock(&kvm->lock);
		ret = 0;
		break;
593 594 595 596 597 598 599 600 601
	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;

602 603
		if (kvm->arch.mem_limit != KVM_S390_NO_MEM_LIMIT &&
		    new_limit > kvm->arch.mem_limit)
604 605
			return -E2BIG;

606 607 608
		if (!new_limit)
			return -EINVAL;

609
		/* gmap_create takes last usable address */
610 611 612
		if (new_limit != KVM_S390_NO_MEM_LIMIT)
			new_limit -= 1;

613 614
		ret = -EBUSY;
		mutex_lock(&kvm->lock);
615
		if (!kvm->created_vcpus) {
616 617
			/* gmap_create will round the limit up */
			struct gmap *new = gmap_create(current->mm, new_limit);
618 619 620 621

			if (!new) {
				ret = -ENOMEM;
			} else {
622
				gmap_remove(kvm->arch.gmap);
623 624 625 626 627 628
				new->private = kvm;
				kvm->arch.gmap = new;
				ret = 0;
			}
		}
		mutex_unlock(&kvm->lock);
629 630 631
		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);
632 633
		break;
	}
634 635 636 637 638 639 640
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

641 642 643 644 645 646 647
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;

648
	if (!test_kvm_facility(kvm, 76))
649 650 651 652 653 654 655 656 657
		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;
658
		VM_EVENT(kvm, 3, "%s", "ENABLE: AES keywrapping support");
659 660 661 662 663 664
		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;
665
		VM_EVENT(kvm, 3, "%s", "ENABLE: DEA keywrapping support");
666 667 668 669 670
		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));
671
		VM_EVENT(kvm, 3, "%s", "DISABLE: AES keywrapping support");
672 673 674 675 676
		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));
677
		VM_EVENT(kvm, 3, "%s", "DISABLE: DEA keywrapping support");
678 679 680 681 682 683 684 685 686 687 688 689 690 691
		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;
}

692 693 694 695 696 697 698 699 700 701
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;
702
	VM_EVENT(kvm, 3, "SET: TOD extension: 0x%x", gtod_high);
703 704 705 706 707 708

	return 0;
}

static int kvm_s390_set_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
709
	u64 gtod;
710 711 712 713

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

714
	kvm_s390_set_tod_clock(kvm, gtod);
715
	VM_EVENT(kvm, 3, "SET: TOD base: 0x%llx", gtod);
716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746
	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;
747
	VM_EVENT(kvm, 3, "QUERY: TOD extension: 0x%x", gtod_high);
748 749 750 751 752 753

	return 0;
}

static int kvm_s390_get_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
754
	u64 gtod;
755

756
	gtod = kvm_s390_get_tod_clock_fast(kvm);
757 758
	if (copy_to_user((void __user *)attr->addr, &gtod, sizeof(gtod)))
		return -EFAULT;
759
	VM_EVENT(kvm, 3, "QUERY: TOD base: 0x%llx", gtod);
760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784

	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;
}

785 786 787
static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_processor *proc;
788
	u16 lowest_ibc, unblocked_ibc;
789 790 791
	int ret = 0;

	mutex_lock(&kvm->lock);
792
	if (kvm->created_vcpus) {
793 794 795 796 797 798 799 800 801 802
		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))) {
803
		kvm->arch.model.cpuid = proc->cpuid;
804 805
		lowest_ibc = sclp.ibc >> 16 & 0xfff;
		unblocked_ibc = sclp.ibc & 0xfff;
806
		if (lowest_ibc && proc->ibc) {
807 808 809 810 811 812 813
			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;
		}
814
		memcpy(kvm->arch.model.fac_list, proc->fac_list,
815 816 817 818 819 820 821 822 823
		       S390_ARCH_FAC_LIST_SIZE_BYTE);
	} else
		ret = -EFAULT;
	kfree(proc);
out:
	mutex_unlock(&kvm->lock);
	return ret;
}

824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846
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;
}

847 848 849 850 851 852 853 854 855 856
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;
}

857 858 859 860 861 862 863 864
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;
865 866 867
	case KVM_S390_VM_CPU_PROCESSOR_FEAT:
		ret = kvm_s390_set_processor_feat(kvm, attr);
		break;
868 869 870
	case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
		ret = kvm_s390_set_processor_subfunc(kvm, attr);
		break;
871 872 873 874 875 876 877 878 879 880 881 882 883 884
	}
	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;
	}
885
	proc->cpuid = kvm->arch.model.cpuid;
886
	proc->ibc = kvm->arch.model.ibc;
887 888
	memcpy(&proc->fac_list, kvm->arch.model.fac_list,
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906
	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);
907
	mach->ibc = sclp.ibc;
908
	memcpy(&mach->fac_mask, kvm->arch.model.fac_mask,
909
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
910
	memcpy((unsigned long *)&mach->fac_list, S390_lowcore.stfle_fac_list,
911
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
912 913 914 915 916 917 918
	if (copy_to_user((void __user *)attr->addr, mach, sizeof(*mach)))
		ret = -EFAULT;
	kfree(mach);
out:
	return ret;
}

919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943
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;
}

944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962
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;
}
963 964 965 966 967 968 969 970 971 972 973
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;
974 975 976 977 978 979
	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;
980 981 982 983 984 985
	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;
986 987 988 989
	}
	return ret;
}

990 991 992 993 994
static int kvm_s390_vm_set_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret;

	switch (attr->group) {
995
	case KVM_S390_VM_MEM_CTRL:
996
		ret = kvm_s390_set_mem_control(kvm, attr);
997
		break;
998 999 1000
	case KVM_S390_VM_TOD:
		ret = kvm_s390_set_tod(kvm, attr);
		break;
1001 1002 1003
	case KVM_S390_VM_CPU_MODEL:
		ret = kvm_s390_set_cpu_model(kvm, attr);
		break;
1004 1005 1006
	case KVM_S390_VM_CRYPTO:
		ret = kvm_s390_vm_set_crypto(kvm, attr);
		break;
1007 1008 1009 1010 1011 1012 1013 1014 1015 1016
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

static int kvm_s390_vm_get_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
1017 1018 1019 1020 1021 1022
	int ret;

	switch (attr->group) {
	case KVM_S390_VM_MEM_CTRL:
		ret = kvm_s390_get_mem_control(kvm, attr);
		break;
1023 1024 1025
	case KVM_S390_VM_TOD:
		ret = kvm_s390_get_tod(kvm, attr);
		break;
1026 1027 1028
	case KVM_S390_VM_CPU_MODEL:
		ret = kvm_s390_get_cpu_model(kvm, attr);
		break;
1029 1030 1031 1032 1033 1034
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
1035 1036 1037 1038 1039 1040 1041
}

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

	switch (attr->group) {
1042 1043 1044 1045
	case KVM_S390_VM_MEM_CTRL:
		switch (attr->attr) {
		case KVM_S390_VM_MEM_ENABLE_CMMA:
		case KVM_S390_VM_MEM_CLR_CMMA:
1046 1047
			ret = sclp.has_cmma ? 0 : -ENXIO;
			break;
1048
		case KVM_S390_VM_MEM_LIMIT_SIZE:
1049 1050 1051 1052 1053 1054 1055
			ret = 0;
			break;
		default:
			ret = -ENXIO;
			break;
		}
		break;
1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066
	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;
1067 1068 1069 1070
	case KVM_S390_VM_CPU_MODEL:
		switch (attr->attr) {
		case KVM_S390_VM_CPU_PROCESSOR:
		case KVM_S390_VM_CPU_MACHINE:
1071 1072
		case KVM_S390_VM_CPU_PROCESSOR_FEAT:
		case KVM_S390_VM_CPU_MACHINE_FEAT:
1073
		case KVM_S390_VM_CPU_MACHINE_SUBFUNC:
1074 1075
			ret = 0;
			break;
1076 1077
		/* configuring subfunctions is not supported yet */
		case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
1078 1079 1080 1081 1082
		default:
			ret = -ENXIO;
			break;
		}
		break;
1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095
	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;
1096 1097 1098 1099 1100 1101 1102 1103
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127
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;

1128
	down_read(&current->mm->mmap_sem);
1129 1130 1131 1132
	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
1133
			break;
1134 1135
		}

1136 1137
		r = get_guest_storage_key(current->mm, hva, &keys[i]);
		if (r)
1138
			break;
1139
	}
1140 1141 1142 1143 1144 1145 1146
	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;
1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180
	}

	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 */
1181 1182 1183
	r = s390_enable_skey();
	if (r)
		goto out;
1184

1185
	down_read(&current->mm->mmap_sem);
1186 1187 1188 1189
	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
1190
			break;
1191 1192 1193 1194 1195
		}

		/* Lowest order bit is reserved */
		if (keys[i] & 0x01) {
			r = -EINVAL;
1196
			break;
1197 1198
		}

1199
		r = set_guest_storage_key(current->mm, hva, keys[i], 0);
1200
		if (r)
1201
			break;
1202
	}
1203
	up_read(&current->mm->mmap_sem);
1204 1205 1206 1207 1208
out:
	kvfree(keys);
	return r;
}

1209 1210 1211 1212 1213
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;
1214
	struct kvm_device_attr attr;
1215 1216 1217
	int r;

	switch (ioctl) {
1218 1219 1220 1221 1222 1223 1224 1225 1226
	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;
	}
1227 1228 1229 1230 1231 1232 1233 1234
	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;
	}
1235 1236 1237 1238 1239 1240 1241
	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));
1242
			r = kvm_set_irq_routing(kvm, &routing, 0, 0);
1243 1244 1245
		}
		break;
	}
1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266
	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;
	}
1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286
	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;
	}
1287
	default:
1288
		r = -ENOTTY;
1289 1290 1291 1292 1293
	}

	return r;
}

1294 1295 1296
static int kvm_s390_query_ap_config(u8 *config)
{
	u32 fcn_code = 0x04000000UL;
1297
	u32 cc = 0;
1298

1299
	memset(config, 0, 128);
1300 1301 1302 1303
	asm volatile(
		"lgr 0,%1\n"
		"lgr 2,%2\n"
		".long 0xb2af0000\n"		/* PQAP(QCI) */
1304
		"0: ipm %0\n"
1305
		"srl %0,28\n"
1306 1307 1308
		"1:\n"
		EX_TABLE(0b, 1b)
		: "+r" (cc)
1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320
		: "r" (fcn_code), "r" (config)
		: "cc", "0", "2", "memory"
	);

	return cc;
}

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

1321
	if (test_facility(12)) {
1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342
		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;
}

1343
static u64 kvm_s390_get_initial_cpuid(void)
1344
{
1345 1346 1347 1348 1349
	struct cpuid cpuid;

	get_cpu_id(&cpuid);
	cpuid.version = 0xff;
	return *((u64 *) &cpuid);
1350 1351
}

1352
static void kvm_s390_crypto_init(struct kvm *kvm)
1353
{
1354
	if (!test_kvm_facility(kvm, 76))
1355
		return;
1356

1357
	kvm->arch.crypto.crycb = &kvm->arch.sie_page2->crycb;
1358
	kvm_s390_set_crycb_format(kvm);
1359

1360 1361 1362 1363 1364 1365 1366
	/* 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));
1367 1368
}

1369 1370 1371
static void sca_dispose(struct kvm *kvm)
{
	if (kvm->arch.use_esca)
1372
		free_pages_exact(kvm->arch.sca, sizeof(struct esca_block));
1373 1374 1375 1376 1377
	else
		free_page((unsigned long)(kvm->arch.sca));
	kvm->arch.sca = NULL;
}

1378
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
1379
{
1380
	gfp_t alloc_flags = GFP_KERNEL;
1381
	int i, rc;
1382
	char debug_name[16];
1383
	static unsigned long sca_offset;
1384

1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395
	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

1396 1397
	rc = s390_enable_sie();
	if (rc)
1398
		goto out_err;
1399

1400 1401
	rc = -ENOMEM;

J
Janosch Frank 已提交
1402 1403
	ratelimit_state_init(&kvm->arch.sthyi_limit, 5 * HZ, 500);

1404
	kvm->arch.use_esca = 0; /* start with basic SCA */
1405 1406
	if (!sclp.has_64bscao)
		alloc_flags |= GFP_DMA;
1407
	rwlock_init(&kvm->arch.sca_lock);
1408
	kvm->arch.sca = (struct bsca_block *) get_zeroed_page(alloc_flags);
1409
	if (!kvm->arch.sca)
1410
		goto out_err;
1411
	spin_lock(&kvm_lock);
1412
	sca_offset += 16;
1413
	if (sca_offset + sizeof(struct bsca_block) > PAGE_SIZE)
1414
		sca_offset = 0;
1415 1416
	kvm->arch.sca = (struct bsca_block *)
			((char *) kvm->arch.sca + sca_offset);
1417
	spin_unlock(&kvm_lock);
1418 1419 1420

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

1421
	kvm->arch.dbf = debug_register(debug_name, 32, 1, 7 * sizeof(long));
1422
	if (!kvm->arch.dbf)
1423
		goto out_err;
1424

1425 1426 1427
	kvm->arch.sie_page2 =
	     (struct sie_page2 *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
	if (!kvm->arch.sie_page2)
1428
		goto out_err;
1429

1430
	/* Populate the facility mask initially. */
1431
	memcpy(kvm->arch.model.fac_mask, S390_lowcore.stfle_fac_list,
1432
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
1433 1434
	for (i = 0; i < S390_ARCH_FAC_LIST_SIZE_U64; i++) {
		if (i < kvm_s390_fac_list_mask_size())
1435
			kvm->arch.model.fac_mask[i] &= kvm_s390_fac_list_mask[i];
1436
		else
1437
			kvm->arch.model.fac_mask[i] = 0UL;
1438 1439
	}

1440
	/* Populate the facility list initially. */
1441 1442
	kvm->arch.model.fac_list = kvm->arch.sie_page2->fac_list;
	memcpy(kvm->arch.model.fac_list, kvm->arch.model.fac_mask,
1443 1444
	       S390_ARCH_FAC_LIST_SIZE_BYTE);

J
Janosch Frank 已提交
1445 1446 1447
	set_kvm_facility(kvm->arch.model.fac_mask, 74);
	set_kvm_facility(kvm->arch.model.fac_list, 74);

1448
	kvm->arch.model.cpuid = kvm_s390_get_initial_cpuid();
1449
	kvm->arch.model.ibc = sclp.ibc & 0x0fff;
1450

1451
	kvm_s390_crypto_init(kvm);
1452

1453
	spin_lock_init(&kvm->arch.float_int.lock);
1454 1455
	for (i = 0; i < FIRQ_LIST_COUNT; i++)
		INIT_LIST_HEAD(&kvm->arch.float_int.lists[i]);
1456
	init_waitqueue_head(&kvm->arch.ipte_wq);
1457
	mutex_init(&kvm->arch.ipte_mutex);
1458

1459
	debug_register_view(kvm->arch.dbf, &debug_sprintf_view);
1460
	VM_EVENT(kvm, 3, "vm created with type %lu", type);
1461

1462 1463
	if (type & KVM_VM_S390_UCONTROL) {
		kvm->arch.gmap = NULL;
1464
		kvm->arch.mem_limit = KVM_S390_NO_MEM_LIMIT;
1465
	} else {
1466 1467 1468 1469 1470
		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);
1471
		kvm->arch.gmap = gmap_create(current->mm, kvm->arch.mem_limit - 1);
1472
		if (!kvm->arch.gmap)
1473
			goto out_err;
1474
		kvm->arch.gmap->private = kvm;
1475
		kvm->arch.gmap->pfault_enabled = 0;
1476
	}
1477 1478

	kvm->arch.css_support = 0;
1479
	kvm->arch.use_irqchip = 0;
1480
	kvm->arch.epoch = 0;
1481

1482
	spin_lock_init(&kvm->arch.start_stop_lock);
1483
	kvm_s390_vsie_init(kvm);
1484
	KVM_EVENT(3, "vm 0x%pK created by pid %u", kvm, current->pid);
1485

1486
	return 0;
1487
out_err:
1488
	free_page((unsigned long)kvm->arch.sie_page2);
1489
	debug_unregister(kvm->arch.dbf);
1490
	sca_dispose(kvm);
1491
	KVM_EVENT(3, "creation of vm failed: %d", rc);
1492
	return rc;
1493 1494
}

1495 1496 1497
void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
{
	VCPU_EVENT(vcpu, 3, "%s", "free cpu");
1498
	trace_kvm_s390_destroy_vcpu(vcpu->vcpu_id);
1499
	kvm_s390_clear_local_irqs(vcpu);
1500
	kvm_clear_async_pf_completion_queue(vcpu);
1501
	if (!kvm_is_ucontrol(vcpu->kvm))
1502
		sca_del_vcpu(vcpu);
1503 1504

	if (kvm_is_ucontrol(vcpu->kvm))
1505
		gmap_remove(vcpu->arch.gmap);
1506

1507
	if (vcpu->kvm->arch.use_cmma)
1508
		kvm_s390_vcpu_unsetup_cmma(vcpu);
1509
	free_page((unsigned long)(vcpu->arch.sie_block));
1510

1511
	kvm_vcpu_uninit(vcpu);
1512
	kmem_cache_free(kvm_vcpu_cache, vcpu);
1513 1514 1515 1516 1517
}

static void kvm_free_vcpus(struct kvm *kvm)
{
	unsigned int i;
1518
	struct kvm_vcpu *vcpu;
1519

1520 1521 1522 1523 1524 1525 1526 1527 1528
	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);
1529 1530
}

1531 1532
void kvm_arch_destroy_vm(struct kvm *kvm)
{
1533
	kvm_free_vcpus(kvm);
1534
	sca_dispose(kvm);
1535
	debug_unregister(kvm->arch.dbf);
1536
	free_page((unsigned long)kvm->arch.sie_page2);
1537
	if (!kvm_is_ucontrol(kvm))
1538
		gmap_remove(kvm->arch.gmap);
1539
	kvm_s390_destroy_adapters(kvm);
1540
	kvm_s390_clear_float_irqs(kvm);
1541
	kvm_s390_vsie_destroy(kvm);
1542
	KVM_EVENT(3, "vm 0x%pK destroyed", kvm);
1543 1544 1545
}

/* Section: vcpu related */
1546 1547
static int __kvm_ucontrol_vcpu_init(struct kvm_vcpu *vcpu)
{
1548
	vcpu->arch.gmap = gmap_create(current->mm, -1UL);
1549 1550 1551 1552 1553 1554 1555
	if (!vcpu->arch.gmap)
		return -ENOMEM;
	vcpu->arch.gmap->private = vcpu->kvm;

	return 0;
}

1556 1557
static void sca_del_vcpu(struct kvm_vcpu *vcpu)
{
1558 1559
	if (!kvm_s390_use_sca_entries())
		return;
1560
	read_lock(&vcpu->kvm->arch.sca_lock);
1561 1562
	if (vcpu->kvm->arch.use_esca) {
		struct esca_block *sca = vcpu->kvm->arch.sca;
1563

1564
		clear_bit_inv(vcpu->vcpu_id, (unsigned long *) sca->mcn);
1565
		sca->cpu[vcpu->vcpu_id].sda = 0;
1566 1567 1568 1569
	} else {
		struct bsca_block *sca = vcpu->kvm->arch.sca;

		clear_bit_inv(vcpu->vcpu_id, (unsigned long *) &sca->mcn);
1570
		sca->cpu[vcpu->vcpu_id].sda = 0;
1571
	}
1572
	read_unlock(&vcpu->kvm->arch.sca_lock);
1573 1574
}

1575
static void sca_add_vcpu(struct kvm_vcpu *vcpu)
1576
{
1577 1578 1579 1580 1581 1582 1583
	if (!kvm_s390_use_sca_entries()) {
		struct bsca_block *sca = vcpu->kvm->arch.sca;

		/* we still need the basic sca for the ipte control */
		vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
		vcpu->arch.sie_block->scaol = (__u32)(__u64)sca;
	}
1584 1585 1586
	read_lock(&vcpu->kvm->arch.sca_lock);
	if (vcpu->kvm->arch.use_esca) {
		struct esca_block *sca = vcpu->kvm->arch.sca;
1587

1588
		sca->cpu[vcpu->vcpu_id].sda = (__u64) vcpu->arch.sie_block;
1589 1590
		vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
		vcpu->arch.sie_block->scaol = (__u32)(__u64)sca & ~0x3fU;
1591
		vcpu->arch.sie_block->ecb2 |= 0x04U;
1592
		set_bit_inv(vcpu->vcpu_id, (unsigned long *) sca->mcn);
1593
	} else {
1594
		struct bsca_block *sca = vcpu->kvm->arch.sca;
1595

1596
		sca->cpu[vcpu->vcpu_id].sda = (__u64) vcpu->arch.sie_block;
1597 1598
		vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
		vcpu->arch.sie_block->scaol = (__u32)(__u64)sca;
1599
		set_bit_inv(vcpu->vcpu_id, (unsigned long *) &sca->mcn);
1600
	}
1601
	read_unlock(&vcpu->kvm->arch.sca_lock);
1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654
}

/* 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);

1655 1656
	VM_EVENT(kvm, 2, "Switched to ESCA (0x%pK -> 0x%pK)",
		 old_sca, kvm->arch.sca);
1657
	return 0;
1658 1659 1660 1661
}

static int sca_can_add_vcpu(struct kvm *kvm, unsigned int id)
{
1662 1663
	int rc;

1664 1665 1666 1667 1668
	if (!kvm_s390_use_sca_entries()) {
		if (id < KVM_MAX_VCPUS)
			return true;
		return false;
	}
1669 1670
	if (id < KVM_S390_BSCA_CPU_SLOTS)
		return true;
1671
	if (!sclp.has_esca || !sclp.has_64bscao)
1672 1673 1674 1675 1676 1677 1678
		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;
1679 1680
}

1681 1682
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
1683 1684
	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
	kvm_clear_async_pf_completion_queue(vcpu);
1685 1686
	vcpu->run->kvm_valid_regs = KVM_SYNC_PREFIX |
				    KVM_SYNC_GPRS |
1687
				    KVM_SYNC_ACRS |
1688 1689 1690
				    KVM_SYNC_CRS |
				    KVM_SYNC_ARCH0 |
				    KVM_SYNC_PFAULT;
1691
	kvm_s390_set_prefix(vcpu, 0);
1692 1693
	if (test_kvm_facility(vcpu->kvm, 64))
		vcpu->run->kvm_valid_regs |= KVM_SYNC_RICCB;
1694 1695 1696 1697
	/* 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)
1698
		vcpu->run->kvm_valid_regs |= KVM_SYNC_VRS;
1699 1700
	else
		vcpu->run->kvm_valid_regs |= KVM_SYNC_FPRS;
1701 1702 1703 1704

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

1705 1706 1707
	return 0;
}

1708 1709 1710 1711
/* 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);
1712
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
1713
	vcpu->arch.cputm_start = get_tod_clock_fast();
1714
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
1715 1716 1717 1718 1719 1720
}

/* 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);
1721
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
1722 1723
	vcpu->arch.sie_block->cputm -= get_tod_clock_fast() - vcpu->arch.cputm_start;
	vcpu->arch.cputm_start = 0;
1724
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756
}

/* 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();
}

1757 1758 1759
/* 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)
{
1760
	preempt_disable(); /* protect from TOD sync and vcpu_load/put */
1761
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
1762 1763
	if (vcpu->arch.cputm_enabled)
		vcpu->arch.cputm_start = get_tod_clock_fast();
1764
	vcpu->arch.sie_block->cputm = cputm;
1765
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
1766
	preempt_enable();
1767 1768
}

1769
/* update and get the cpu timer - can also be called from other VCPU threads */
1770 1771
__u64 kvm_s390_get_cpu_timer(struct kvm_vcpu *vcpu)
{
1772
	unsigned int seq;
1773 1774 1775 1776 1777
	__u64 value;

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

1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791
	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();
1792
	return value;
1793 1794
}

1795 1796
void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
1797
	/* Save host register state */
1798
	save_fpu_regs();
1799 1800
	vcpu->arch.host_fpregs.fpc = current->thread.fpu.fpc;
	vcpu->arch.host_fpregs.regs = current->thread.fpu.regs;
1801

1802 1803 1804 1805
	if (MACHINE_HAS_VX)
		current->thread.fpu.regs = vcpu->run->s.regs.vrs;
	else
		current->thread.fpu.regs = vcpu->run->s.regs.fprs;
1806
	current->thread.fpu.fpc = vcpu->run->s.regs.fpc;
1807
	if (test_fp_ctl(current->thread.fpu.fpc))
1808
		/* User space provided an invalid FPC, let's clear it */
1809 1810 1811
		current->thread.fpu.fpc = 0;

	save_access_regs(vcpu->arch.host_acrs);
1812
	restore_access_regs(vcpu->run->s.regs.acrs);
1813
	gmap_enable(vcpu->arch.enabled_gmap);
1814
	atomic_or(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
1815
	if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
1816
		__start_cpu_timer_accounting(vcpu);
1817
	vcpu->cpu = cpu;
1818 1819 1820 1821
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
1822
	vcpu->cpu = -1;
1823
	if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
1824
		__stop_cpu_timer_accounting(vcpu);
1825
	atomic_andnot(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
1826 1827
	vcpu->arch.enabled_gmap = gmap_get_enabled();
	gmap_disable(vcpu->arch.enabled_gmap);
1828

1829
	/* Save guest register state */
1830
	save_fpu_regs();
1831
	vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
1832

1833 1834 1835
	/* Restore host register state */
	current->thread.fpu.fpc = vcpu->arch.host_fpregs.fpc;
	current->thread.fpu.regs = vcpu->arch.host_fpregs.regs;
1836 1837

	save_access_regs(vcpu->run->s.regs.acrs);
1838 1839 1840 1841 1842 1843 1844 1845
	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;
1846
	kvm_s390_set_prefix(vcpu, 0);
1847
	kvm_s390_set_cpu_timer(vcpu, 0);
1848 1849 1850 1851 1852
	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;
1853 1854 1855
	/* make sure the new fpc will be lazily loaded */
	save_fpu_regs();
	current->thread.fpu.fpc = 0;
1856
	vcpu->arch.sie_block->gbea = 1;
1857
	vcpu->arch.sie_block->pp = 0;
1858 1859
	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
	kvm_clear_async_pf_completion_queue(vcpu);
1860 1861
	if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm))
		kvm_s390_vcpu_stop(vcpu);
1862
	kvm_s390_clear_local_irqs(vcpu);
1863 1864
}

1865
void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
1866
{
1867
	mutex_lock(&vcpu->kvm->lock);
1868
	preempt_disable();
1869
	vcpu->arch.sie_block->epoch = vcpu->kvm->arch.epoch;
1870
	preempt_enable();
1871
	mutex_unlock(&vcpu->kvm->lock);
1872
	if (!kvm_is_ucontrol(vcpu->kvm)) {
1873
		vcpu->arch.gmap = vcpu->kvm->arch.gmap;
1874
		sca_add_vcpu(vcpu);
1875
	}
1876 1877
	if (test_kvm_facility(vcpu->kvm, 74) || vcpu->kvm->arch.user_instr0)
		vcpu->arch.sie_block->ictl |= ICTL_OPEREXC;
1878 1879
	/* make vcpu_load load the right gmap on the first trigger */
	vcpu->arch.enabled_gmap = vcpu->arch.gmap;
1880 1881
}

1882 1883
static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu)
{
1884
	if (!test_kvm_facility(vcpu->kvm, 76))
1885 1886
		return;

1887 1888 1889 1890 1891 1892 1893
	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;

1894 1895 1896
	vcpu->arch.sie_block->crycbd = vcpu->kvm->arch.crypto.crycbd;
}

1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913
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;
}

1914 1915 1916 1917 1918
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;
1919
	if (test_kvm_facility(vcpu->kvm, 7))
1920
		vcpu->arch.sie_block->fac = (u32)(u64) model->fac_list;
1921 1922
}

1923 1924
int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
{
1925
	int rc = 0;
1926

1927 1928
	atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH |
						    CPUSTAT_SM |
1929 1930
						    CPUSTAT_STOPPED);

1931
	if (test_kvm_facility(vcpu->kvm, 78))
1932
		atomic_or(CPUSTAT_GED2, &vcpu->arch.sie_block->cpuflags);
1933
	else if (test_kvm_facility(vcpu->kvm, 8))
1934
		atomic_or(CPUSTAT_GED, &vcpu->arch.sie_block->cpuflags);
1935

1936 1937
	kvm_s390_vcpu_setup_model(vcpu);

1938 1939 1940
	/* pgste_set_pte has special handling for !MACHINE_HAS_ESOP */
	if (MACHINE_HAS_ESOP)
		vcpu->arch.sie_block->ecb |= 0x02;
1941 1942
	if (test_kvm_facility(vcpu->kvm, 9))
		vcpu->arch.sie_block->ecb |= 0x04;
1943
	if (test_kvm_facility(vcpu->kvm, 73))
1944 1945
		vcpu->arch.sie_block->ecb |= 0x10;

1946
	if (test_kvm_facility(vcpu->kvm, 8) && sclp.has_pfmfi)
1947
		vcpu->arch.sie_block->ecb2 |= 0x08;
1948 1949 1950
	vcpu->arch.sie_block->eca = 0x1002000U;
	if (sclp.has_cei)
		vcpu->arch.sie_block->eca |= 0x80000000U;
1951 1952
	if (sclp.has_ib)
		vcpu->arch.sie_block->eca |= 0x40000000U;
1953
	if (sclp.has_siif)
1954
		vcpu->arch.sie_block->eca |= 1;
1955
	if (sclp.has_sigpif)
1956
		vcpu->arch.sie_block->eca |= 0x10000000U;
1957
	if (test_kvm_facility(vcpu->kvm, 129)) {
1958 1959 1960
		vcpu->arch.sie_block->eca |= 0x00020000;
		vcpu->arch.sie_block->ecd |= 0x20000000;
	}
1961
	vcpu->arch.sie_block->riccbd = (unsigned long) &vcpu->run->s.regs.riccb;
1962
	vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE;
1963

1964
	if (vcpu->kvm->arch.use_cmma) {
1965 1966 1967
		rc = kvm_s390_vcpu_setup_cmma(vcpu);
		if (rc)
			return rc;
1968
	}
1969
	hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
1970
	vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup;
1971

1972 1973
	kvm_s390_vcpu_crypto_setup(vcpu);

1974
	return rc;
1975 1976 1977 1978 1979
}

struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
				      unsigned int id)
{
1980
	struct kvm_vcpu *vcpu;
1981
	struct sie_page *sie_page;
1982 1983
	int rc = -EINVAL;

1984
	if (!kvm_is_ucontrol(kvm) && !sca_can_add_vcpu(kvm, id))
1985 1986 1987
		goto out;

	rc = -ENOMEM;
1988

1989
	vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
1990
	if (!vcpu)
1991
		goto out;
1992

1993 1994
	sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
	if (!sie_page)
1995 1996
		goto out_free_cpu;

1997 1998 1999
	vcpu->arch.sie_block = &sie_page->sie_block;
	vcpu->arch.sie_block->itdba = (unsigned long) &sie_page->itdb;

2000 2001 2002 2003
	/* the real guest size will always be smaller than msl */
	vcpu->arch.sie_block->mso = 0;
	vcpu->arch.sie_block->msl = sclp.hamax;

2004
	vcpu->arch.sie_block->icpua = id;
2005 2006
	spin_lock_init(&vcpu->arch.local_int.lock);
	vcpu->arch.local_int.float_int = &kvm->arch.float_int;
2007
	vcpu->arch.local_int.wq = &vcpu->wq;
2008
	vcpu->arch.local_int.cpuflags = &vcpu->arch.sie_block->cpuflags;
2009
	seqcount_init(&vcpu->arch.cputm_seqcount);
2010

2011 2012
	rc = kvm_vcpu_init(vcpu, kvm, id);
	if (rc)
2013
		goto out_free_sie_block;
2014
	VM_EVENT(kvm, 3, "create cpu %d at 0x%pK, sie block at 0x%pK", id, vcpu,
2015
		 vcpu->arch.sie_block);
2016
	trace_kvm_s390_create_vcpu(id, vcpu, vcpu->arch.sie_block);
2017 2018

	return vcpu;
2019 2020
out_free_sie_block:
	free_page((unsigned long)(vcpu->arch.sie_block));
2021
out_free_cpu:
2022
	kmem_cache_free(kvm_vcpu_cache, vcpu);
2023
out:
2024 2025 2026 2027 2028
	return ERR_PTR(rc);
}

int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
{
2029
	return kvm_s390_vcpu_has_irq(vcpu, 0);
2030 2031
}

2032
void kvm_s390_vcpu_block(struct kvm_vcpu *vcpu)
2033
{
2034
	atomic_or(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
2035
	exit_sie(vcpu);
2036 2037
}

2038
void kvm_s390_vcpu_unblock(struct kvm_vcpu *vcpu)
2039
{
2040
	atomic_andnot(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
2041 2042
}

2043 2044
static void kvm_s390_vcpu_request(struct kvm_vcpu *vcpu)
{
2045
	atomic_or(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
2046
	exit_sie(vcpu);
2047 2048 2049 2050
}

static void kvm_s390_vcpu_request_handled(struct kvm_vcpu *vcpu)
{
2051
	atomic_andnot(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
2052 2053
}

2054 2055 2056 2057 2058 2059
/*
 * 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)
{
2060
	atomic_or(CPUSTAT_STOP_INT, &vcpu->arch.sie_block->cpuflags);
2061 2062 2063 2064
	while (vcpu->arch.sie_block->prog0c & PROG_IN_SIE)
		cpu_relax();
}

2065 2066
/* Kick a guest cpu out of SIE to process a request synchronously */
void kvm_s390_sync_request(int req, struct kvm_vcpu *vcpu)
2067
{
2068 2069
	kvm_make_request(req, vcpu);
	kvm_s390_vcpu_request(vcpu);
2070 2071
}

2072 2073
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start,
			      unsigned long end)
2074 2075 2076
{
	struct kvm *kvm = gmap->private;
	struct kvm_vcpu *vcpu;
2077 2078
	unsigned long prefix;
	int i;
2079

2080 2081
	if (gmap_is_shadow(gmap))
		return;
2082 2083 2084
	if (start >= 1UL << 31)
		/* We are only interested in prefix pages */
		return;
2085 2086
	kvm_for_each_vcpu(i, vcpu, kvm) {
		/* match against both prefix pages */
2087 2088 2089 2090
		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);
2091
			kvm_s390_sync_request(KVM_REQ_MMU_RELOAD, vcpu);
2092 2093 2094 2095
		}
	}
}

2096 2097 2098 2099 2100 2101 2102
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
	/* kvm common code refers to this, but never calls it */
	BUG();
	return 0;
}

2103 2104 2105 2106 2107 2108
static int kvm_arch_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu,
					   struct kvm_one_reg *reg)
{
	int r = -EINVAL;

	switch (reg->id) {
2109 2110 2111 2112 2113 2114 2115 2116
	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;
2117
	case KVM_REG_S390_CPU_TIMER:
2118
		r = put_user(kvm_s390_get_cpu_timer(vcpu),
2119 2120 2121 2122 2123 2124
			     (u64 __user *)reg->addr);
		break;
	case KVM_REG_S390_CLOCK_COMP:
		r = put_user(vcpu->arch.sie_block->ckc,
			     (u64 __user *)reg->addr);
		break;
2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136
	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;
2137 2138 2139 2140
	case KVM_REG_S390_PP:
		r = put_user(vcpu->arch.sie_block->pp,
			     (u64 __user *)reg->addr);
		break;
2141 2142 2143 2144
	case KVM_REG_S390_GBEA:
		r = put_user(vcpu->arch.sie_block->gbea,
			     (u64 __user *)reg->addr);
		break;
2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155
	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;
2156
	__u64 val;
2157 2158

	switch (reg->id) {
2159 2160 2161 2162 2163 2164 2165 2166
	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;
2167
	case KVM_REG_S390_CPU_TIMER:
2168 2169 2170
		r = get_user(val, (u64 __user *)reg->addr);
		if (!r)
			kvm_s390_set_cpu_timer(vcpu, val);
2171 2172 2173 2174 2175
		break;
	case KVM_REG_S390_CLOCK_COMP:
		r = get_user(vcpu->arch.sie_block->ckc,
			     (u64 __user *)reg->addr);
		break;
2176 2177 2178
	case KVM_REG_S390_PFTOKEN:
		r = get_user(vcpu->arch.pfault_token,
			     (u64 __user *)reg->addr);
2179 2180
		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
			kvm_clear_async_pf_completion_queue(vcpu);
2181 2182 2183 2184 2185 2186 2187 2188 2189
		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;
2190 2191 2192 2193
	case KVM_REG_S390_PP:
		r = get_user(vcpu->arch.sie_block->pp,
			     (u64 __user *)reg->addr);
		break;
2194 2195 2196 2197
	case KVM_REG_S390_GBEA:
		r = get_user(vcpu->arch.sie_block->gbea,
			     (u64 __user *)reg->addr);
		break;
2198 2199 2200 2201 2202 2203
	default:
		break;
	}

	return r;
}
2204

2205 2206 2207 2208 2209 2210 2211 2212
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)
{
2213
	memcpy(&vcpu->run->s.regs.gprs, &regs->gprs, sizeof(regs->gprs));
2214 2215 2216 2217 2218
	return 0;
}

int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
2219
	memcpy(&regs->gprs, &vcpu->run->s.regs.gprs, sizeof(regs->gprs));
2220 2221 2222 2223 2224 2225
	return 0;
}

int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
2226
	memcpy(&vcpu->run->s.regs.acrs, &sregs->acrs, sizeof(sregs->acrs));
2227
	memcpy(&vcpu->arch.sie_block->gcr, &sregs->crs, sizeof(sregs->crs));
2228
	restore_access_regs(vcpu->run->s.regs.acrs);
2229 2230 2231 2232 2233 2234
	return 0;
}

int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
2235
	memcpy(&sregs->acrs, &vcpu->run->s.regs.acrs, sizeof(sregs->acrs));
2236 2237 2238 2239 2240 2241
	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)
{
2242 2243
	/* make sure the new values will be lazily loaded */
	save_fpu_regs();
2244 2245
	if (test_fp_ctl(fpu->fpc))
		return -EINVAL;
2246 2247 2248 2249 2250
	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));
2251 2252 2253 2254 2255
	return 0;
}

int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
2256 2257 2258 2259 2260 2261 2262
	/* 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;
2263 2264 2265 2266 2267 2268 2269
	return 0;
}

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

2270
	if (!is_vcpu_stopped(vcpu))
2271
		rc = -EBUSY;
2272 2273 2274 2275
	else {
		vcpu->run->psw_mask = psw.mask;
		vcpu->run->psw_addr = psw.addr;
	}
2276 2277 2278 2279 2280 2281 2282 2283 2284
	return rc;
}

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

2285 2286 2287 2288
#define VALID_GUESTDBG_FLAGS (KVM_GUESTDBG_SINGLESTEP | \
			      KVM_GUESTDBG_USE_HW_BP | \
			      KVM_GUESTDBG_ENABLE)

J
Jan Kiszka 已提交
2289 2290
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
					struct kvm_guest_debug *dbg)
2291
{
2292 2293 2294 2295 2296
	int rc = 0;

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

2297
	if (dbg->control & ~VALID_GUESTDBG_FLAGS)
2298
		return -EINVAL;
2299 2300
	if (!sclp.has_gpere)
		return -EINVAL;
2301 2302 2303 2304

	if (dbg->control & KVM_GUESTDBG_ENABLE) {
		vcpu->guest_debug = dbg->control;
		/* enforce guest PER */
2305
		atomic_or(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
2306 2307 2308 2309

		if (dbg->control & KVM_GUESTDBG_USE_HW_BP)
			rc = kvm_s390_import_bp_data(vcpu, dbg);
	} else {
2310
		atomic_andnot(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
2311 2312 2313 2314 2315 2316
		vcpu->arch.guestdbg.last_bp = 0;
	}

	if (rc) {
		vcpu->guest_debug = 0;
		kvm_s390_clear_bp_data(vcpu);
2317
		atomic_andnot(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
2318 2319 2320
	}

	return rc;
2321 2322
}

2323 2324 2325
int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
2326 2327 2328
	/* CHECK_STOP and LOAD are not supported yet */
	return is_vcpu_stopped(vcpu) ? KVM_MP_STATE_STOPPED :
				       KVM_MP_STATE_OPERATING;
2329 2330 2331 2332 2333
}

int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353
	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;
2354 2355
}

2356 2357 2358 2359 2360
static bool ibs_enabled(struct kvm_vcpu *vcpu)
{
	return atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_IBS;
}

2361 2362
static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu)
{
2363
retry:
2364
	kvm_s390_vcpu_request_handled(vcpu);
2365 2366
	if (!vcpu->requests)
		return 0;
2367 2368
	/*
	 * We use MMU_RELOAD just to re-arm the ipte notifier for the
2369
	 * guest prefix page. gmap_mprotect_notify will wait on the ptl lock.
2370 2371 2372 2373
	 * 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.
	 */
2374
	if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) {
2375
		int rc;
2376 2377 2378
		rc = gmap_mprotect_notify(vcpu->arch.gmap,
					  kvm_s390_get_prefix(vcpu),
					  PAGE_SIZE * 2, PROT_WRITE);
2379 2380
		if (rc) {
			kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu);
2381
			return rc;
2382
		}
2383
		goto retry;
2384
	}
2385

2386 2387 2388 2389 2390
	if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) {
		vcpu->arch.sie_block->ihcpu = 0xffff;
		goto retry;
	}

2391 2392 2393
	if (kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu)) {
		if (!ibs_enabled(vcpu)) {
			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 1);
2394
			atomic_or(CPUSTAT_IBS,
2395 2396 2397
					&vcpu->arch.sie_block->cpuflags);
		}
		goto retry;
2398
	}
2399 2400 2401 2402

	if (kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu)) {
		if (ibs_enabled(vcpu)) {
			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 0);
2403
			atomic_andnot(CPUSTAT_IBS,
2404 2405 2406 2407 2408
					  &vcpu->arch.sie_block->cpuflags);
		}
		goto retry;
	}

2409 2410 2411 2412 2413
	if (kvm_check_request(KVM_REQ_ICPT_OPEREXC, vcpu)) {
		vcpu->arch.sie_block->ictl |= ICTL_OPEREXC;
		goto retry;
	}

2414 2415 2416
	/* nothing to do, just clear the request */
	clear_bit(KVM_REQ_UNHALT, &vcpu->requests);

2417 2418 2419
	return 0;
}

2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435
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);
}

2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446
/**
 * 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)
2447
{
2448 2449
	return gmap_fault(vcpu->arch.gmap, gpa,
			  writable ? FAULT_FLAG_WRITE : 0);
2450 2451
}

2452 2453 2454 2455
static void __kvm_inject_pfault_token(struct kvm_vcpu *vcpu, bool start_token,
				      unsigned long token)
{
	struct kvm_s390_interrupt inti;
2456
	struct kvm_s390_irq irq;
2457 2458

	if (start_token) {
2459 2460 2461
		irq.u.ext.ext_params2 = token;
		irq.type = KVM_S390_INT_PFAULT_INIT;
		WARN_ON_ONCE(kvm_s390_inject_vcpu(vcpu, &irq));
2462 2463
	} else {
		inti.type = KVM_S390_INT_PFAULT_DONE;
2464
		inti.parm64 = token;
2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510
		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;
2511
	if (kvm_s390_vcpu_has_irq(vcpu, 0))
2512 2513 2514 2515 2516 2517
		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
2518 2519 2520
	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))
2521 2522 2523 2524 2525 2526
		return 0;

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

2527
static int vcpu_pre_run(struct kvm_vcpu *vcpu)
2528
{
2529
	int rc, cpuflags;
2530

2531 2532 2533 2534 2535 2536 2537
	/*
	 * 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);

2538 2539
	vcpu->arch.sie_block->gg14 = vcpu->run->s.regs.gprs[14];
	vcpu->arch.sie_block->gg15 = vcpu->run->s.regs.gprs[15];
2540 2541 2542 2543

	if (need_resched())
		schedule();

2544
	if (test_cpu_flag(CIF_MCCK_PENDING))
2545 2546
		s390_handle_mcck();

2547 2548 2549 2550 2551
	if (!kvm_is_ucontrol(vcpu->kvm)) {
		rc = kvm_s390_deliver_pending_interrupts(vcpu);
		if (rc)
			return rc;
	}
C
Carsten Otte 已提交
2552

2553 2554 2555 2556
	rc = kvm_s390_handle_requests(vcpu);
	if (rc)
		return rc;

2557 2558 2559 2560 2561
	if (guestdbg_enabled(vcpu)) {
		kvm_s390_backup_guest_per_regs(vcpu);
		kvm_s390_patch_guest_per_regs(vcpu);
	}

2562
	vcpu->arch.sie_block->icptcode = 0;
2563 2564 2565
	cpuflags = atomic_read(&vcpu->arch.sie_block->cpuflags);
	VCPU_EVENT(vcpu, 6, "entering sie flags %x", cpuflags);
	trace_kvm_s390_sie_enter(vcpu, cpuflags);
2566

2567 2568 2569
	return 0;
}

2570 2571
static int vcpu_post_run_fault_in_sie(struct kvm_vcpu *vcpu)
{
2572 2573 2574 2575
	struct kvm_s390_pgm_info pgm_info = {
		.code = PGM_ADDRESSING,
	};
	u8 opcode, ilen;
2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588
	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.
	 */
2589
	rc = read_guest_instr(vcpu, &opcode, 1);
2590
	ilen = insn_length(opcode);
2591 2592 2593 2594 2595 2596 2597 2598 2599 2600
	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;
	}
2601 2602 2603
	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);
2604 2605
}

2606 2607
static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason)
{
2608 2609 2610 2611
	VCPU_EVENT(vcpu, 6, "exit sie icptcode %d",
		   vcpu->arch.sie_block->icptcode);
	trace_kvm_s390_sie_exit(vcpu, vcpu->arch.sie_block->icptcode);

2612 2613 2614
	if (guestdbg_enabled(vcpu))
		kvm_s390_restore_guest_per_regs(vcpu);

2615 2616
	vcpu->run->s.regs.gprs[14] = vcpu->arch.sie_block->gg14;
	vcpu->run->s.regs.gprs[15] = vcpu->arch.sie_block->gg15;
2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630

	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;
2631 2632 2633 2634 2635
	} 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;
2636
		return -EREMOTE;
2637
	} else if (current->thread.gmap_pfault) {
2638
		trace_kvm_s390_major_guest_pfault(vcpu);
2639
		current->thread.gmap_pfault = 0;
2640 2641 2642
		if (kvm_arch_setup_async_pf(vcpu))
			return 0;
		return kvm_arch_fault_in_page(vcpu, current->thread.gmap_addr, 1);
2643
	}
2644
	return vcpu_post_run_fault_in_sie(vcpu);
2645 2646 2647 2648 2649 2650
}

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

2651 2652 2653 2654 2655 2656
	/*
	 * 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);

2657 2658 2659 2660
	do {
		rc = vcpu_pre_run(vcpu);
		if (rc)
			break;
2661

2662
		srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
2663 2664 2665 2666
		/*
		 * As PF_VCPU will be used in fault handler, between
		 * guest_enter and guest_exit should be no uaccess.
		 */
2667
		local_irq_disable();
2668
		guest_enter_irqoff();
2669
		__disable_cpu_timer_accounting(vcpu);
2670
		local_irq_enable();
2671 2672
		exit_reason = sie64a(vcpu->arch.sie_block,
				     vcpu->run->s.regs.gprs);
2673
		local_irq_disable();
2674
		__enable_cpu_timer_accounting(vcpu);
2675
		guest_exit_irqoff();
2676
		local_irq_enable();
2677
		vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
2678 2679

		rc = vcpu_post_run(vcpu, exit_reason);
2680
	} while (!signal_pending(current) && !guestdbg_exit_pending(vcpu) && !rc);
2681

2682
	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
2683
	return rc;
2684 2685
}

2686 2687 2688 2689 2690 2691 2692 2693
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);
2694 2695
		/* some control register changes require a tlb flush */
		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
2696 2697
	}
	if (kvm_run->kvm_dirty_regs & KVM_SYNC_ARCH0) {
2698
		kvm_s390_set_cpu_timer(vcpu, kvm_run->s.regs.cputm);
2699 2700 2701 2702 2703 2704 2705 2706 2707
		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;
2708 2709
		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
			kvm_clear_async_pf_completion_queue(vcpu);
2710
	}
F
Fan Zhang 已提交
2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723
	/*
	 * If userspace sets the riccb (e.g. after migration) to a valid state,
	 * we should enable RI here instead of doing the lazy enablement.
	 */
	if ((kvm_run->kvm_dirty_regs & KVM_SYNC_RICCB) &&
	    test_kvm_facility(vcpu->kvm, 64)) {
		struct runtime_instr_cb *riccb =
			(struct runtime_instr_cb *) &kvm_run->s.regs.riccb;

		if (riccb->valid)
			vcpu->arch.sie_block->ecb3 |= 0x01;
	}

2724 2725 2726 2727 2728 2729 2730 2731 2732
	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);
2733
	kvm_run->s.regs.cputm = kvm_s390_get_cpu_timer(vcpu);
2734 2735 2736 2737 2738 2739 2740 2741 2742
	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;
}

2743 2744
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
2745
	int rc;
2746 2747
	sigset_t sigsaved;

2748 2749 2750 2751 2752
	if (guestdbg_exit_pending(vcpu)) {
		kvm_s390_prepare_debug_exit(vcpu);
		return 0;
	}

2753 2754 2755
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);

2756 2757 2758
	if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) {
		kvm_s390_vcpu_start(vcpu);
	} else if (is_vcpu_stopped(vcpu)) {
2759
		pr_err_ratelimited("can't run stopped vcpu %d\n",
2760 2761 2762
				   vcpu->vcpu_id);
		return -EINVAL;
	}
2763

2764
	sync_regs(vcpu, kvm_run);
2765
	enable_cpu_timer_accounting(vcpu);
2766

2767
	might_fault();
2768
	rc = __vcpu_run(vcpu);
2769

2770 2771
	if (signal_pending(current) && !rc) {
		kvm_run->exit_reason = KVM_EXIT_INTR;
2772
		rc = -EINTR;
2773
	}
2774

2775 2776 2777 2778 2779
	if (guestdbg_exit_pending(vcpu) && !rc)  {
		kvm_s390_prepare_debug_exit(vcpu);
		rc = 0;
	}

2780
	if (rc == -EREMOTE) {
2781
		/* userspace support is needed, kvm_run has been prepared */
2782 2783
		rc = 0;
	}
2784

2785
	disable_cpu_timer_accounting(vcpu);
2786
	store_regs(vcpu, kvm_run);
2787

2788 2789 2790 2791
	if (vcpu->sigset_active)
		sigprocmask(SIG_SETMASK, &sigsaved, NULL);

	vcpu->stat.exit_userspace++;
2792
	return rc;
2793 2794 2795 2796 2797 2798 2799 2800
}

/*
 * 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
 */
2801
int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long gpa)
2802
{
2803
	unsigned char archmode = 1;
2804
	freg_t fprs[NUM_FPRS];
2805
	unsigned int px;
2806
	u64 clkcomp, cputm;
2807
	int rc;
2808

2809
	px = kvm_s390_get_prefix(vcpu);
2810 2811
	if (gpa == KVM_S390_STORE_STATUS_NOADDR) {
		if (write_guest_abs(vcpu, 163, &archmode, 1))
2812
			return -EFAULT;
2813
		gpa = 0;
2814 2815
	} else if (gpa == KVM_S390_STORE_STATUS_PREFIXED) {
		if (write_guest_real(vcpu, 163, &archmode, 1))
2816
			return -EFAULT;
2817 2818 2819
		gpa = px;
	} else
		gpa -= __LC_FPREGS_SAVE_AREA;
2820 2821 2822

	/* manually convert vector registers if necessary */
	if (MACHINE_HAS_VX) {
2823
		convert_vx_to_fp(fprs, (__vector128 *) vcpu->run->s.regs.vrs);
2824 2825 2826 2827
		rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
				     fprs, 128);
	} else {
		rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
2828
				     vcpu->run->s.regs.fprs, 128);
2829
	}
2830
	rc |= write_guest_abs(vcpu, gpa + __LC_GPREGS_SAVE_AREA,
2831
			      vcpu->run->s.regs.gprs, 128);
2832
	rc |= write_guest_abs(vcpu, gpa + __LC_PSW_SAVE_AREA,
2833
			      &vcpu->arch.sie_block->gpsw, 16);
2834
	rc |= write_guest_abs(vcpu, gpa + __LC_PREFIX_SAVE_AREA,
2835
			      &px, 4);
2836
	rc |= write_guest_abs(vcpu, gpa + __LC_FP_CREG_SAVE_AREA,
2837
			      &vcpu->run->s.regs.fpc, 4);
2838
	rc |= write_guest_abs(vcpu, gpa + __LC_TOD_PROGREG_SAVE_AREA,
2839
			      &vcpu->arch.sie_block->todpr, 4);
2840
	cputm = kvm_s390_get_cpu_timer(vcpu);
2841
	rc |= write_guest_abs(vcpu, gpa + __LC_CPU_TIMER_SAVE_AREA,
2842
			      &cputm, 8);
2843
	clkcomp = vcpu->arch.sie_block->ckc >> 8;
2844
	rc |= write_guest_abs(vcpu, gpa + __LC_CLOCK_COMP_SAVE_AREA,
2845
			      &clkcomp, 8);
2846
	rc |= write_guest_abs(vcpu, gpa + __LC_AREGS_SAVE_AREA,
2847
			      &vcpu->run->s.regs.acrs, 64);
2848
	rc |= write_guest_abs(vcpu, gpa + __LC_CREGS_SAVE_AREA,
2849 2850
			      &vcpu->arch.sie_block->gcr, 128);
	return rc ? -EFAULT : 0;
2851 2852
}

2853 2854 2855 2856 2857 2858 2859
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
	 */
2860
	save_fpu_regs();
2861
	vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
2862 2863 2864 2865 2866
	save_access_regs(vcpu->run->s.regs.acrs);

	return kvm_s390_store_status_unloaded(vcpu, addr);
}

2867 2868 2869
static void __disable_ibs_on_vcpu(struct kvm_vcpu *vcpu)
{
	kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu);
2870
	kvm_s390_sync_request(KVM_REQ_DISABLE_IBS, vcpu);
2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884
}

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)
{
2885 2886
	if (!sclp.has_ibs)
		return;
2887
	kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu);
2888
	kvm_s390_sync_request(KVM_REQ_ENABLE_IBS, vcpu);
2889 2890
}

2891 2892
void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu)
{
2893 2894 2895 2896 2897
	int i, online_vcpus, started_vcpus = 0;

	if (!is_vcpu_stopped(vcpu))
		return;

2898
	trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 1);
2899
	/* Only one cpu at a time may enter/leave the STOPPED state. */
2900
	spin_lock(&vcpu->kvm->arch.start_stop_lock);
2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919
	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);
	}

2920
	atomic_andnot(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
2921 2922 2923 2924
	/*
	 * Another VCPU might have used IBS while we were offline.
	 * Let's play safe and flush the VCPU at startup.
	 */
2925
	kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
2926
	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
2927
	return;
2928 2929 2930 2931
}

void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu)
{
2932 2933 2934 2935 2936 2937
	int i, online_vcpus, started_vcpus = 0;
	struct kvm_vcpu *started_vcpu = NULL;

	if (is_vcpu_stopped(vcpu))
		return;

2938
	trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 0);
2939
	/* Only one cpu at a time may enter/leave the STOPPED state. */
2940
	spin_lock(&vcpu->kvm->arch.start_stop_lock);
2941 2942
	online_vcpus = atomic_read(&vcpu->kvm->online_vcpus);

2943
	/* SIGP STOP and SIGP STOP AND STORE STATUS has been fully processed */
2944
	kvm_s390_clear_stop_irq(vcpu);
2945

2946
	atomic_or(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963
	__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);
	}

2964
	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
2965
	return;
2966 2967
}

2968 2969 2970 2971 2972 2973 2974 2975 2976
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) {
2977 2978 2979
	case KVM_CAP_S390_CSS_SUPPORT:
		if (!vcpu->kvm->arch.css_support) {
			vcpu->kvm->arch.css_support = 1;
2980
			VM_EVENT(vcpu->kvm, 3, "%s", "ENABLE: CSS support");
2981 2982 2983 2984
			trace_kvm_s390_enable_css(vcpu->kvm);
		}
		r = 0;
		break;
2985 2986 2987 2988 2989 2990 2991
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017
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) {
3018 3019
			r = check_gva_range(vcpu, mop->gaddr, mop->ar,
					    mop->size, GACC_FETCH);
3020 3021 3022 3023 3024 3025 3026 3027 3028 3029
			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) {
3030 3031
			r = check_gva_range(vcpu, mop->gaddr, mop->ar,
					    mop->size, GACC_STORE);
3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052
			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;
}

3053 3054 3055 3056 3057
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;
3058
	int idx;
3059
	long r;
3060

3061
	switch (ioctl) {
3062 3063 3064 3065 3066 3067 3068 3069 3070
	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;
	}
3071
	case KVM_S390_INTERRUPT: {
3072
		struct kvm_s390_interrupt s390int;
3073
		struct kvm_s390_irq s390irq;
3074

3075
		r = -EFAULT;
3076
		if (copy_from_user(&s390int, argp, sizeof(s390int)))
3077
			break;
3078 3079 3080
		if (s390int_to_s390irq(&s390int, &s390irq))
			return -EINVAL;
		r = kvm_s390_inject_vcpu(vcpu, &s390irq);
3081
		break;
3082
	}
3083
	case KVM_S390_STORE_STATUS:
3084
		idx = srcu_read_lock(&vcpu->kvm->srcu);
3085
		r = kvm_s390_vcpu_store_status(vcpu, arg);
3086
		srcu_read_unlock(&vcpu->kvm->srcu, idx);
3087
		break;
3088 3089 3090
	case KVM_S390_SET_INITIAL_PSW: {
		psw_t psw;

3091
		r = -EFAULT;
3092
		if (copy_from_user(&psw, argp, sizeof(psw)))
3093 3094 3095
			break;
		r = kvm_arch_vcpu_ioctl_set_initial_psw(vcpu, psw);
		break;
3096 3097
	}
	case KVM_S390_INITIAL_RESET:
3098 3099
		r = kvm_arch_vcpu_ioctl_initial_reset(vcpu);
		break;
3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111
	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;
	}
3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147
#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
3148
	case KVM_S390_VCPU_FAULT: {
3149
		r = gmap_fault(vcpu->arch.gmap, arg, 0);
3150 3151
		break;
	}
3152 3153 3154 3155 3156 3157 3158 3159 3160
	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;
	}
3161 3162 3163 3164 3165 3166 3167 3168 3169
	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;
	}
3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201
	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;
	}
3202
	default:
3203
		r = -ENOTTY;
3204
	}
3205
	return r;
3206 3207
}

3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220
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;
}

3221 3222
int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
			    unsigned long npages)
3223 3224 3225 3226
{
	return 0;
}

3227
/* Section: memory related */
3228 3229
int kvm_arch_prepare_memory_region(struct kvm *kvm,
				   struct kvm_memory_slot *memslot,
3230
				   const struct kvm_userspace_memory_region *mem,
3231
				   enum kvm_mr_change change)
3232
{
3233 3234 3235 3236
	/* 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 */
3237

3238
	if (mem->userspace_addr & 0xffffful)
3239 3240
		return -EINVAL;

3241
	if (mem->memory_size & 0xffffful)
3242 3243
		return -EINVAL;

3244 3245 3246
	if (mem->guest_phys_addr + mem->memory_size > kvm->arch.mem_limit)
		return -EINVAL;

3247 3248 3249 3250
	return 0;
}

void kvm_arch_commit_memory_region(struct kvm *kvm,
3251
				const struct kvm_userspace_memory_region *mem,
3252
				const struct kvm_memory_slot *old,
3253
				const struct kvm_memory_slot *new,
3254
				enum kvm_mr_change change)
3255
{
3256
	int rc;
3257

3258 3259 3260 3261 3262 3263 3264 3265 3266 3267
	/* 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;
3268 3269 3270 3271

	rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr,
		mem->guest_phys_addr, mem->memory_size);
	if (rc)
3272
		pr_warn("failed to commit memory region\n");
3273
	return;
3274 3275
}

3276 3277 3278 3279 3280 3281 3282
static inline unsigned long nonhyp_mask(int i)
{
	unsigned int nonhyp_fai = (sclp.hmfai << i * 2) >> 30;

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

3283 3284 3285 3286 3287
void kvm_arch_vcpu_block_finish(struct kvm_vcpu *vcpu)
{
	vcpu->valid_wakeup = false;
}

3288 3289
static int __init kvm_s390_init(void)
{
3290 3291
	int i;

3292 3293 3294 3295 3296
	if (!sclp.has_sief2) {
		pr_info("SIE not available\n");
		return -ENODEV;
	}

3297 3298 3299 3300
	for (i = 0; i < 16; i++)
		kvm_s390_fac_list_mask[i] |=
			S390_lowcore.stfle_fac_list[i] & nonhyp_mask(i);

3301
	return kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE);
3302 3303 3304 3305 3306 3307 3308 3309 3310
}

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

module_init(kvm_s390_init);
module_exit(kvm_s390_exit);
3311 3312 3313 3314 3315 3316 3317 3318 3319

/*
 * 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");