kvm-s390.c 105.5 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0
2
/*
3
 * hosting IBM Z kernel virtual machines (s390x)
4
 *
5
 * Copyright IBM Corp. 2008, 2018
6 7 8 9
 *
 *    Author(s): Carsten Otte <cotte@de.ibm.com>
 *               Christian Borntraeger <borntraeger@de.ibm.com>
 *               Heiko Carstens <heiko.carstens@de.ibm.com>
10
 *               Christian Ehrhardt <ehrhardt@de.ibm.com>
11
 *               Jason J. Herne <jjherne@us.ibm.com>
12 13 14 15 16
 */

#include <linux/compiler.h>
#include <linux/err.h>
#include <linux/fs.h>
17
#include <linux/hrtimer.h>
18 19 20
#include <linux/init.h>
#include <linux/kvm.h>
#include <linux/kvm_host.h>
21
#include <linux/mman.h>
22
#include <linux/module.h>
23
#include <linux/moduleparam.h>
24
#include <linux/random.h>
25
#include <linux/slab.h>
26
#include <linux/timer.h>
27
#include <linux/vmalloc.h>
28
#include <linux/bitmap.h>
29
#include <linux/sched/signal.h>
30
#include <linux/string.h>
31

32
#include <asm/asm-offsets.h>
33
#include <asm/lowcore.h>
34
#include <asm/stp.h>
35
#include <asm/pgtable.h>
36
#include <asm/gmap.h>
37
#include <asm/nmi.h>
38
#include <asm/switch_to.h>
39
#include <asm/isc.h>
40
#include <asm/sclp.h>
41
#include <asm/cpacf.h>
42
#include <asm/timex.h>
43
#include "kvm-s390.h"
44 45
#include "gaccess.h"

46 47 48 49
#define KMSG_COMPONENT "kvm-s390"
#undef pr_fmt
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt

50 51
#define CREATE_TRACE_POINTS
#include "trace.h"
52
#include "trace-s390.h"
53

54
#define MEM_OP_MAX_SIZE 65536	/* Maximum transfer size for KVM_S390_MEM_OP */
55 56 57
#define LOCAL_IRQS 32
#define VCPU_IRQS_MAX_BUF (sizeof(struct kvm_s390_irq) * \
			   (KVM_MAX_VCPUS + LOCAL_IRQS))
58

59 60 61 62
#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) },
63
	{ "exit_null", VCPU_STAT(exit_null) },
64 65 66 67
	{ "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) },
68
	{ "exit_instruction", VCPU_STAT(exit_instruction) },
69
	{ "exit_pei", VCPU_STAT(exit_pei) },
70 71
	{ "exit_program_interruption", VCPU_STAT(exit_program_interruption) },
	{ "exit_instr_and_program_int", VCPU_STAT(exit_instr_and_program) },
72
	{ "exit_operation_exception", VCPU_STAT(exit_operation_exception) },
73
	{ "halt_successful_poll", VCPU_STAT(halt_successful_poll) },
74
	{ "halt_attempted_poll", VCPU_STAT(halt_attempted_poll) },
75
	{ "halt_poll_invalid", VCPU_STAT(halt_poll_invalid) },
76
	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
77
	{ "instruction_lctlg", VCPU_STAT(instruction_lctlg) },
78
	{ "instruction_lctl", VCPU_STAT(instruction_lctl) },
79 80
	{ "instruction_stctl", VCPU_STAT(instruction_stctl) },
	{ "instruction_stctg", VCPU_STAT(instruction_stctg) },
81
	{ "deliver_emergency_signal", VCPU_STAT(deliver_emergency_signal) },
82
	{ "deliver_external_call", VCPU_STAT(deliver_external_call) },
83 84 85 86 87 88 89
	{ "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) },
90 91 92 93 94
	{ "instruction_epsw", VCPU_STAT(instruction_epsw) },
	{ "instruction_gs", VCPU_STAT(instruction_gs) },
	{ "instruction_io_other", VCPU_STAT(instruction_io_other) },
	{ "instruction_lpsw", VCPU_STAT(instruction_lpsw) },
	{ "instruction_lpswe", VCPU_STAT(instruction_lpswe) },
95
	{ "instruction_pfmf", VCPU_STAT(instruction_pfmf) },
96
	{ "instruction_ptff", VCPU_STAT(instruction_ptff) },
97
	{ "instruction_stidp", VCPU_STAT(instruction_stidp) },
98 99
	{ "instruction_sck", VCPU_STAT(instruction_sck) },
	{ "instruction_sckpf", VCPU_STAT(instruction_sckpf) },
100 101 102
	{ "instruction_spx", VCPU_STAT(instruction_spx) },
	{ "instruction_stpx", VCPU_STAT(instruction_stpx) },
	{ "instruction_stap", VCPU_STAT(instruction_stap) },
103 104 105 106
	{ "instruction_iske", VCPU_STAT(instruction_iske) },
	{ "instruction_ri", VCPU_STAT(instruction_ri) },
	{ "instruction_rrbe", VCPU_STAT(instruction_rrbe) },
	{ "instruction_sske", VCPU_STAT(instruction_sske) },
107
	{ "instruction_ipte_interlock", VCPU_STAT(instruction_ipte_interlock) },
108
	{ "instruction_essa", VCPU_STAT(instruction_essa) },
109 110
	{ "instruction_stsi", VCPU_STAT(instruction_stsi) },
	{ "instruction_stfl", VCPU_STAT(instruction_stfl) },
111 112
	{ "instruction_tb", VCPU_STAT(instruction_tb) },
	{ "instruction_tpi", VCPU_STAT(instruction_tpi) },
113
	{ "instruction_tprot", VCPU_STAT(instruction_tprot) },
114
	{ "instruction_tsch", VCPU_STAT(instruction_tsch) },
J
Janosch Frank 已提交
115
	{ "instruction_sthyi", VCPU_STAT(instruction_sthyi) },
116
	{ "instruction_sie", VCPU_STAT(instruction_sie) },
117
	{ "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) },
118
	{ "instruction_sigp_sense_running", VCPU_STAT(instruction_sigp_sense_running) },
119
	{ "instruction_sigp_external_call", VCPU_STAT(instruction_sigp_external_call) },
120
	{ "instruction_sigp_emergency", VCPU_STAT(instruction_sigp_emergency) },
121 122
	{ "instruction_sigp_cond_emergency", VCPU_STAT(instruction_sigp_cond_emergency) },
	{ "instruction_sigp_start", VCPU_STAT(instruction_sigp_start) },
123
	{ "instruction_sigp_stop", VCPU_STAT(instruction_sigp_stop) },
124 125
	{ "instruction_sigp_stop_store_status", VCPU_STAT(instruction_sigp_stop_store_status) },
	{ "instruction_sigp_store_status", VCPU_STAT(instruction_sigp_store_status) },
126
	{ "instruction_sigp_store_adtl_status", VCPU_STAT(instruction_sigp_store_adtl_status) },
127 128 129
	{ "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) },
130 131 132
	{ "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) },
133 134 135 136 137 138
	{ "instruction_diag_10", VCPU_STAT(diagnose_10) },
	{ "instruction_diag_44", VCPU_STAT(diagnose_44) },
	{ "instruction_diag_9c", VCPU_STAT(diagnose_9c) },
	{ "instruction_diag_258", VCPU_STAT(diagnose_258) },
	{ "instruction_diag_308", VCPU_STAT(diagnose_308) },
	{ "instruction_diag_500", VCPU_STAT(diagnose_500) },
139
	{ "instruction_diag_other", VCPU_STAT(diagnose_other) },
140 141 142
	{ NULL }
};

143 144 145 146 147 148
struct kvm_s390_tod_clock_ext {
	__u8 epoch_idx;
	__u64 tod;
	__u8 reserved[7];
} __packed;

149 150 151 152 153
/* 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");

154
/* upper facilities limit for kvm */
155
unsigned long kvm_s390_fac_list_mask[16] = { FACILITIES_KVM };
156

157
unsigned long kvm_s390_fac_list_mask_size(void)
158
{
159 160
	BUILD_BUG_ON(ARRAY_SIZE(kvm_s390_fac_list_mask) > S390_ARCH_FAC_MASK_SIZE_U64);
	return ARRAY_SIZE(kvm_s390_fac_list_mask);
161 162
}

163 164
/* available cpu features supported by kvm */
static DECLARE_BITMAP(kvm_s390_available_cpu_feat, KVM_S390_VM_CPU_FEAT_NR_BITS);
165 166
/* available subfunctions indicated via query / "test bit" */
static struct kvm_s390_vm_cpu_subfunc kvm_s390_available_subfunc;
167

168
static struct gmap_notifier gmap_notifier;
169
static struct gmap_notifier vsie_gmap_notifier;
170
debug_info_t *kvm_s390_dbf;
171

172
/* Section: not file related */
173
int kvm_arch_hardware_enable(void)
174 175
{
	/* every s390 is virtualization enabled ;-) */
176
	return 0;
177 178
}

179 180
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start,
			      unsigned long end);
181

182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203
static void kvm_clock_sync_scb(struct kvm_s390_sie_block *scb, u64 delta)
{
	u8 delta_idx = 0;

	/*
	 * The TOD jumps by delta, we have to compensate this by adding
	 * -delta to the epoch.
	 */
	delta = -delta;

	/* sign-extension - we're adding to signed values below */
	if ((s64)delta < 0)
		delta_idx = -1;

	scb->epoch += delta;
	if (scb->ecd & ECD_MEF) {
		scb->epdx += delta_idx;
		if (scb->epoch < delta)
			scb->epdx += 1;
	}
}

204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219
/*
 * 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_for_each_vcpu(i, vcpu, kvm) {
220 221 222 223 224
			kvm_clock_sync_scb(vcpu->arch.sie_block, *delta);
			if (i == 0) {
				kvm->arch.epoch = vcpu->arch.sie_block->epoch;
				kvm->arch.epdx = vcpu->arch.sie_block->epdx;
			}
225 226
			if (vcpu->arch.cputm_enabled)
				vcpu->arch.cputm_start += *delta;
227
			if (vcpu->arch.vsie_block)
228 229
				kvm_clock_sync_scb(vcpu->arch.vsie_block,
						   *delta);
230 231 232 233 234 235 236 237 238
		}
	}
	return NOTIFY_OK;
}

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

239 240
int kvm_arch_hardware_setup(void)
{
241
	gmap_notifier.notifier_call = kvm_gmap_notifier;
242
	gmap_register_pte_notifier(&gmap_notifier);
243 244
	vsie_gmap_notifier.notifier_call = kvm_s390_vsie_gmap_notifier;
	gmap_register_pte_notifier(&vsie_gmap_notifier);
245 246
	atomic_notifier_chain_register(&s390_epoch_delta_notifier,
				       &kvm_clock_notifier);
247 248 249 250 251
	return 0;
}

void kvm_arch_hardware_unsetup(void)
{
252
	gmap_unregister_pte_notifier(&gmap_notifier);
253
	gmap_unregister_pte_notifier(&vsie_gmap_notifier);
254 255
	atomic_notifier_chain_unregister(&s390_epoch_delta_notifier,
					 &kvm_clock_notifier);
256 257
}

258 259 260 261 262
static void allow_cpu_feat(unsigned long nr)
{
	set_bit_inv(nr, kvm_s390_available_cpu_feat);
}

263 264 265
static inline int plo_test_bit(unsigned char nr)
{
	register unsigned long r0 asm("0") = (unsigned long) nr | 0x100;
266
	int cc;
267 268 269 270 271 272 273 274 275 276 277 278

	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;
}

279 280
static void kvm_s390_cpu_feat_init(void)
{
281 282 283 284 285 286 287 288
	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 */
289 290 291
		ptff(kvm_s390_available_subfunc.ptff,
		     sizeof(kvm_s390_available_subfunc.ptff),
		     PTFF_QAF);
292 293

	if (test_facility(17)) { /* MSA */
294 295 296 297 298 299 300 301 302 303
		__cpacf_query(CPACF_KMAC, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.kmac);
		__cpacf_query(CPACF_KMC, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.kmc);
		__cpacf_query(CPACF_KM, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.km);
		__cpacf_query(CPACF_KIMD, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.kimd);
		__cpacf_query(CPACF_KLMD, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.klmd);
304 305
	}
	if (test_facility(76)) /* MSA3 */
306 307
		__cpacf_query(CPACF_PCKMO, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.pckmo);
308
	if (test_facility(77)) { /* MSA4 */
309 310 311 312 313 314 315 316
		__cpacf_query(CPACF_KMCTR, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.kmctr);
		__cpacf_query(CPACF_KMF, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.kmf);
		__cpacf_query(CPACF_KMO, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.kmo);
		__cpacf_query(CPACF_PCC, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.pcc);
317 318
	}
	if (test_facility(57)) /* MSA5 */
319
		__cpacf_query(CPACF_PRNO, (cpacf_mask_t *)
320
			      kvm_s390_available_subfunc.ppno);
321

322 323 324 325
	if (test_facility(146)) /* MSA8 */
		__cpacf_query(CPACF_KMA, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.kma);

326 327
	if (MACHINE_HAS_ESOP)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_ESOP);
328 329 330 331 332
	/*
	 * 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 ||
333
	    !test_facility(3) || !nested)
334 335
		return;
	allow_cpu_feat(KVM_S390_VM_CPU_FEAT_SIEF2);
336 337
	if (sclp.has_64bscao)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_64BSCAO);
338 339
	if (sclp.has_siif)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_SIIF);
340 341
	if (sclp.has_gpere)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_GPERE);
342 343
	if (sclp.has_gsls)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_GSLS);
344 345
	if (sclp.has_ib)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_IB);
346 347
	if (sclp.has_cei)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_CEI);
348 349
	if (sclp.has_ibs)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_IBS);
350 351
	if (sclp.has_kss)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_KSS);
352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
	/*
	 * 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.
	 */
370 371
}

372 373
int kvm_arch_init(void *opaque)
{
374 375 376 377 378 379 380 381 382
	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;
	}

383 384
	kvm_s390_cpu_feat_init();

385 386
	/* Register floating interrupt controller interface. */
	return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
387 388
}

389 390 391 392 393
void kvm_arch_exit(void)
{
	debug_unregister(kvm_s390_dbf);
}

394 395 396 397 398 399 400 401 402
/* 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;
}

403
int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
404
{
405 406
	int r;

407
	switch (ext) {
408
	case KVM_CAP_S390_PSW:
409
	case KVM_CAP_S390_GMAP:
410
	case KVM_CAP_SYNC_MMU:
411 412 413
#ifdef CONFIG_KVM_S390_UCONTROL
	case KVM_CAP_S390_UCONTROL:
#endif
414
	case KVM_CAP_ASYNC_PF:
415
	case KVM_CAP_SYNC_REGS:
416
	case KVM_CAP_ONE_REG:
417
	case KVM_CAP_ENABLE_CAP:
418
	case KVM_CAP_S390_CSS_SUPPORT:
C
Cornelia Huck 已提交
419
	case KVM_CAP_IOEVENTFD:
420
	case KVM_CAP_DEVICE_CTRL:
421
	case KVM_CAP_ENABLE_CAP_VM:
422
	case KVM_CAP_S390_IRQCHIP:
423
	case KVM_CAP_VM_ATTRIBUTES:
424
	case KVM_CAP_MP_STATE:
425
	case KVM_CAP_IMMEDIATE_EXIT:
426
	case KVM_CAP_S390_INJECT_IRQ:
427
	case KVM_CAP_S390_USER_SIGP:
428
	case KVM_CAP_S390_USER_STSI:
429
	case KVM_CAP_S390_SKEYS:
430
	case KVM_CAP_S390_IRQ_STATE:
431
	case KVM_CAP_S390_USER_INSTR0:
432
	case KVM_CAP_S390_CMMA_MIGRATION:
433
	case KVM_CAP_S390_AIS:
434
	case KVM_CAP_S390_AIS_MIGRATION:
435 436
		r = 1;
		break;
437 438 439
	case KVM_CAP_S390_MEM_OP:
		r = MEM_OP_MAX_SIZE;
		break;
440 441
	case KVM_CAP_NR_VCPUS:
	case KVM_CAP_MAX_VCPUS:
442
		r = KVM_S390_BSCA_CPU_SLOTS;
443 444 445
		if (!kvm_s390_use_sca_entries())
			r = KVM_MAX_VCPUS;
		else if (sclp.has_esca && sclp.has_64bscao)
446
			r = KVM_S390_ESCA_CPU_SLOTS;
447
		break;
448 449 450
	case KVM_CAP_NR_MEMSLOTS:
		r = KVM_USER_MEM_SLOTS;
		break;
451
	case KVM_CAP_S390_COW:
452
		r = MACHINE_HAS_ESOP;
453
		break;
454 455 456
	case KVM_CAP_S390_VECTOR_REGISTERS:
		r = MACHINE_HAS_VX;
		break;
457 458 459
	case KVM_CAP_S390_RI:
		r = test_facility(64);
		break;
F
Fan Zhang 已提交
460 461 462
	case KVM_CAP_S390_GS:
		r = test_facility(133);
		break;
463 464 465
	case KVM_CAP_S390_BPB:
		r = test_facility(82);
		break;
466
	default:
467
		r = 0;
468
	}
469
	return r;
470 471
}

472 473 474 475 476 477 478 479 480 481 482 483
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);

484
		if (test_and_clear_guest_dirty(gmap->mm, address))
485
			mark_page_dirty(kvm, cur_gfn);
486 487
		if (fatal_signal_pending(current))
			return;
488
		cond_resched();
489 490 491
	}
}

492
/* Section: vm related */
493 494
static void sca_del_vcpu(struct kvm_vcpu *vcpu);

495 496 497 498 499 500
/*
 * 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)
{
501 502
	int r;
	unsigned long n;
503
	struct kvm_memslots *slots;
504 505 506
	struct kvm_memory_slot *memslot;
	int is_dirty = 0;

507 508 509
	if (kvm_is_ucontrol(kvm))
		return -EINVAL;

510 511 512 513 514 515
	mutex_lock(&kvm->slots_lock);

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

516 517
	slots = kvm_memslots(kvm);
	memslot = id_to_memslot(slots, log->slot);
518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535
	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;
536 537
}

538 539 540 541 542 543 544 545 546 547
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);
	}
}

548 549 550 551 552 553 554 555
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) {
556
	case KVM_CAP_S390_IRQCHIP:
557
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_IRQCHIP");
558 559 560
		kvm->arch.use_irqchip = 1;
		r = 0;
		break;
561
	case KVM_CAP_S390_USER_SIGP:
562
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_SIGP");
563 564 565
		kvm->arch.user_sigp = 1;
		r = 0;
		break;
566
	case KVM_CAP_S390_VECTOR_REGISTERS:
567
		mutex_lock(&kvm->lock);
568
		if (kvm->created_vcpus) {
569 570
			r = -EBUSY;
		} else if (MACHINE_HAS_VX) {
571 572
			set_kvm_facility(kvm->arch.model.fac_mask, 129);
			set_kvm_facility(kvm->arch.model.fac_list, 129);
573 574 575 576
			if (test_facility(134)) {
				set_kvm_facility(kvm->arch.model.fac_mask, 134);
				set_kvm_facility(kvm->arch.model.fac_list, 134);
			}
577 578 579 580
			if (test_facility(135)) {
				set_kvm_facility(kvm->arch.model.fac_mask, 135);
				set_kvm_facility(kvm->arch.model.fac_list, 135);
			}
581 582 583
			r = 0;
		} else
			r = -EINVAL;
584
		mutex_unlock(&kvm->lock);
585 586
		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_VECTOR_REGISTERS %s",
			 r ? "(not available)" : "(success)");
587
		break;
588 589 590
	case KVM_CAP_S390_RI:
		r = -EINVAL;
		mutex_lock(&kvm->lock);
591
		if (kvm->created_vcpus) {
592 593
			r = -EBUSY;
		} else if (test_facility(64)) {
594 595
			set_kvm_facility(kvm->arch.model.fac_mask, 64);
			set_kvm_facility(kvm->arch.model.fac_list, 64);
596 597 598 599 600 601
			r = 0;
		}
		mutex_unlock(&kvm->lock);
		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_RI %s",
			 r ? "(not available)" : "(success)");
		break;
602 603 604 605 606 607 608 609 610 611 612 613 614
	case KVM_CAP_S390_AIS:
		mutex_lock(&kvm->lock);
		if (kvm->created_vcpus) {
			r = -EBUSY;
		} else {
			set_kvm_facility(kvm->arch.model.fac_mask, 72);
			set_kvm_facility(kvm->arch.model.fac_list, 72);
			r = 0;
		}
		mutex_unlock(&kvm->lock);
		VM_EVENT(kvm, 3, "ENABLE: AIS %s",
			 r ? "(not available)" : "(success)");
		break;
F
Fan Zhang 已提交
615 616 617
	case KVM_CAP_S390_GS:
		r = -EINVAL;
		mutex_lock(&kvm->lock);
618
		if (kvm->created_vcpus) {
F
Fan Zhang 已提交
619 620 621 622 623 624 625 626 627 628
			r = -EBUSY;
		} else if (test_facility(133)) {
			set_kvm_facility(kvm->arch.model.fac_mask, 133);
			set_kvm_facility(kvm->arch.model.fac_list, 133);
			r = 0;
		}
		mutex_unlock(&kvm->lock);
		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_GS %s",
			 r ? "(not available)" : "(success)");
		break;
629
	case KVM_CAP_S390_USER_STSI:
630
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_STSI");
631 632 633
		kvm->arch.user_stsi = 1;
		r = 0;
		break;
634 635 636 637 638 639
	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;
640 641 642 643 644 645 646
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

647 648 649 650 651 652 653
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;
654
		VM_EVENT(kvm, 3, "QUERY: max guest memory: %lu bytes",
655 656
			 kvm->arch.mem_limit);
		if (put_user(kvm->arch.mem_limit, (u64 __user *)attr->addr))
657 658 659 660 661 662 663 664 665 666
			ret = -EFAULT;
		break;
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *attr)
667 668 669 670 671
{
	int ret;
	unsigned int idx;
	switch (attr->attr) {
	case KVM_S390_VM_MEM_ENABLE_CMMA:
672
		ret = -ENXIO;
673
		if (!sclp.has_cmma)
674 675
			break;

676
		ret = -EBUSY;
677
		VM_EVENT(kvm, 3, "%s", "ENABLE: CMMA support");
678
		mutex_lock(&kvm->lock);
679
		if (!kvm->created_vcpus) {
680 681 682 683 684 685
			kvm->arch.use_cmma = 1;
			ret = 0;
		}
		mutex_unlock(&kvm->lock);
		break;
	case KVM_S390_VM_MEM_CLR_CMMA:
686 687 688
		ret = -ENXIO;
		if (!sclp.has_cmma)
			break;
689 690 691 692
		ret = -EINVAL;
		if (!kvm->arch.use_cmma)
			break;

693
		VM_EVENT(kvm, 3, "%s", "RESET: CMMA states");
694 695
		mutex_lock(&kvm->lock);
		idx = srcu_read_lock(&kvm->srcu);
696
		s390_reset_cmma(kvm->arch.gmap->mm);
697 698 699 700
		srcu_read_unlock(&kvm->srcu, idx);
		mutex_unlock(&kvm->lock);
		ret = 0;
		break;
701 702 703 704 705 706 707 708 709
	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;

710 711
		if (kvm->arch.mem_limit != KVM_S390_NO_MEM_LIMIT &&
		    new_limit > kvm->arch.mem_limit)
712 713
			return -E2BIG;

714 715 716
		if (!new_limit)
			return -EINVAL;

717
		/* gmap_create takes last usable address */
718 719 720
		if (new_limit != KVM_S390_NO_MEM_LIMIT)
			new_limit -= 1;

721 722
		ret = -EBUSY;
		mutex_lock(&kvm->lock);
723
		if (!kvm->created_vcpus) {
724 725
			/* gmap_create will round the limit up */
			struct gmap *new = gmap_create(current->mm, new_limit);
726 727 728 729

			if (!new) {
				ret = -ENOMEM;
			} else {
730
				gmap_remove(kvm->arch.gmap);
731 732 733 734 735 736
				new->private = kvm;
				kvm->arch.gmap = new;
				ret = 0;
			}
		}
		mutex_unlock(&kvm->lock);
737 738 739
		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);
740 741
		break;
	}
742 743 744 745 746 747 748
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

749 750 751 752 753 754 755
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;

756
	if (!test_kvm_facility(kvm, 76))
757 758 759 760 761 762 763 764 765
		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;
766
		VM_EVENT(kvm, 3, "%s", "ENABLE: AES keywrapping support");
767 768 769 770 771 772
		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;
773
		VM_EVENT(kvm, 3, "%s", "ENABLE: DEA keywrapping support");
774 775 776 777 778
		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));
779
		VM_EVENT(kvm, 3, "%s", "DISABLE: AES keywrapping support");
780 781 782 783 784
		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));
785
		VM_EVENT(kvm, 3, "%s", "DISABLE: DEA keywrapping support");
786 787 788 789 790 791 792 793 794 795 796 797 798 799
		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;
}

800 801 802 803 804 805 806 807 808 809 810
static void kvm_s390_sync_request_broadcast(struct kvm *kvm, int req)
{
	int cx;
	struct kvm_vcpu *vcpu;

	kvm_for_each_vcpu(cx, vcpu, kvm)
		kvm_s390_sync_request(req, vcpu);
}

/*
 * Must be called with kvm->srcu held to avoid races on memslots, and with
811
 * kvm->slots_lock to avoid races with ourselves and kvm_s390_vm_stop_migration.
812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836
 */
static int kvm_s390_vm_start_migration(struct kvm *kvm)
{
	struct kvm_s390_migration_state *mgs;
	struct kvm_memory_slot *ms;
	/* should be the only one */
	struct kvm_memslots *slots;
	unsigned long ram_pages;
	int slotnr;

	/* migration mode already enabled */
	if (kvm->arch.migration_state)
		return 0;

	slots = kvm_memslots(kvm);
	if (!slots || !slots->used_slots)
		return -EINVAL;

	mgs = kzalloc(sizeof(*mgs), GFP_KERNEL);
	if (!mgs)
		return -ENOMEM;
	kvm->arch.migration_state = mgs;

	if (kvm->arch.use_cmma) {
		/*
837 838 839 840
		 * Get the first slot. They are reverse sorted by base_gfn, so
		 * the first slot is also the one at the end of the address
		 * space. We have verified above that at least one slot is
		 * present.
841
		 */
842
		ms = slots->memslots;
843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866
		/* round up so we only use full longs */
		ram_pages = roundup(ms->base_gfn + ms->npages, BITS_PER_LONG);
		/* allocate enough bytes to store all the bits */
		mgs->pgste_bitmap = vmalloc(ram_pages / 8);
		if (!mgs->pgste_bitmap) {
			kfree(mgs);
			kvm->arch.migration_state = NULL;
			return -ENOMEM;
		}

		mgs->bitmap_size = ram_pages;
		atomic64_set(&mgs->dirty_pages, ram_pages);
		/* mark all the pages in active slots as dirty */
		for (slotnr = 0; slotnr < slots->used_slots; slotnr++) {
			ms = slots->memslots + slotnr;
			bitmap_set(mgs->pgste_bitmap, ms->base_gfn, ms->npages);
		}

		kvm_s390_sync_request_broadcast(kvm, KVM_REQ_START_MIGRATION);
	}
	return 0;
}

/*
867
 * Must be called with kvm->slots_lock to avoid races with ourselves and
868 869 870 871 872 873 874 875 876 877 878 879 880 881
 * kvm_s390_vm_start_migration.
 */
static int kvm_s390_vm_stop_migration(struct kvm *kvm)
{
	struct kvm_s390_migration_state *mgs;

	/* migration mode already disabled */
	if (!kvm->arch.migration_state)
		return 0;
	mgs = kvm->arch.migration_state;
	kvm->arch.migration_state = NULL;

	if (kvm->arch.use_cmma) {
		kvm_s390_sync_request_broadcast(kvm, KVM_REQ_STOP_MIGRATION);
882 883
		/* We have to wait for the essa emulation to finish */
		synchronize_srcu(&kvm->srcu);
884 885 886 887 888 889 890 891 892
		vfree(mgs->pgste_bitmap);
	}
	kfree(mgs);
	return 0;
}

static int kvm_s390_vm_set_migration(struct kvm *kvm,
				     struct kvm_device_attr *attr)
{
893
	int res = -ENXIO;
894

895
	mutex_lock(&kvm->slots_lock);
896 897 898 899 900 901 902 903 904 905
	switch (attr->attr) {
	case KVM_S390_VM_MIGRATION_START:
		res = kvm_s390_vm_start_migration(kvm);
		break;
	case KVM_S390_VM_MIGRATION_STOP:
		res = kvm_s390_vm_stop_migration(kvm);
		break;
	default:
		break;
	}
906
	mutex_unlock(&kvm->slots_lock);
907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923

	return res;
}

static int kvm_s390_vm_get_migration(struct kvm *kvm,
				     struct kvm_device_attr *attr)
{
	u64 mig = (kvm->arch.migration_state != NULL);

	if (attr->attr != KVM_S390_VM_MIGRATION_STATUS)
		return -ENXIO;

	if (copy_to_user((void __user *)attr->addr, &mig, sizeof(mig)))
		return -EFAULT;
	return 0;
}

924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943
static int kvm_s390_set_tod_ext(struct kvm *kvm, struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_tod_clock gtod;

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

	if (test_kvm_facility(kvm, 139))
		kvm_s390_set_tod_clock_ext(kvm, &gtod);
	else if (gtod.epoch_idx == 0)
		kvm_s390_set_tod_clock(kvm, gtod.tod);
	else
		return -EINVAL;

	VM_EVENT(kvm, 3, "SET: TOD extension: 0x%x, TOD base: 0x%llx",
		gtod.epoch_idx, gtod.tod);

	return 0;
}

944 945 946 947 948 949 950 951 952 953
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;
954
	VM_EVENT(kvm, 3, "SET: TOD extension: 0x%x", gtod_high);
955 956 957 958 959 960

	return 0;
}

static int kvm_s390_set_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
961
	u64 gtod;
962 963 964 965

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

966
	kvm_s390_set_tod_clock(kvm, gtod);
967
	VM_EVENT(kvm, 3, "SET: TOD base: 0x%llx", gtod);
968 969 970 971 972 973 974 975 976 977 978
	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) {
979 980 981
	case KVM_S390_VM_TOD_EXT:
		ret = kvm_s390_set_tod_ext(kvm, attr);
		break;
982 983 984 985 986 987 988 989 990 991 992 993 994
	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;
}

995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031
static void kvm_s390_get_tod_clock_ext(struct kvm *kvm,
					struct kvm_s390_vm_tod_clock *gtod)
{
	struct kvm_s390_tod_clock_ext htod;

	preempt_disable();

	get_tod_clock_ext((char *)&htod);

	gtod->tod = htod.tod + kvm->arch.epoch;
	gtod->epoch_idx = htod.epoch_idx + kvm->arch.epdx;

	if (gtod->tod < htod.tod)
		gtod->epoch_idx += 1;

	preempt_enable();
}

static int kvm_s390_get_tod_ext(struct kvm *kvm, struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_tod_clock gtod;

	memset(&gtod, 0, sizeof(gtod));

	if (test_kvm_facility(kvm, 139))
		kvm_s390_get_tod_clock_ext(kvm, &gtod);
	else
		gtod.tod = kvm_s390_get_tod_clock_fast(kvm);

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

	VM_EVENT(kvm, 3, "QUERY: TOD extension: 0x%x, TOD base: 0x%llx",
		gtod.epoch_idx, gtod.tod);
	return 0;
}

1032 1033 1034 1035 1036 1037 1038
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;
1039
	VM_EVENT(kvm, 3, "QUERY: TOD extension: 0x%x", gtod_high);
1040 1041 1042 1043 1044 1045

	return 0;
}

static int kvm_s390_get_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
1046
	u64 gtod;
1047

1048
	gtod = kvm_s390_get_tod_clock_fast(kvm);
1049 1050
	if (copy_to_user((void __user *)attr->addr, &gtod, sizeof(gtod)))
		return -EFAULT;
1051
	VM_EVENT(kvm, 3, "QUERY: TOD base: 0x%llx", gtod);
1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063

	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) {
1064 1065 1066
	case KVM_S390_VM_TOD_EXT:
		ret = kvm_s390_get_tod_ext(kvm, attr);
		break;
1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079
	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;
}

1080 1081 1082
static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_processor *proc;
1083
	u16 lowest_ibc, unblocked_ibc;
1084 1085 1086
	int ret = 0;

	mutex_lock(&kvm->lock);
1087
	if (kvm->created_vcpus) {
1088 1089 1090 1091 1092 1093 1094 1095 1096 1097
		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))) {
1098
		kvm->arch.model.cpuid = proc->cpuid;
1099 1100
		lowest_ibc = sclp.ibc >> 16 & 0xfff;
		unblocked_ibc = sclp.ibc & 0xfff;
1101
		if (lowest_ibc && proc->ibc) {
1102 1103 1104 1105 1106 1107 1108
			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;
		}
1109
		memcpy(kvm->arch.model.fac_list, proc->fac_list,
1110
		       S390_ARCH_FAC_LIST_SIZE_BYTE);
1111 1112 1113 1114 1115 1116 1117
		VM_EVENT(kvm, 3, "SET: guest ibc: 0x%4.4x, guest cpuid: 0x%16.16llx",
			 kvm->arch.model.ibc,
			 kvm->arch.model.cpuid);
		VM_EVENT(kvm, 3, "SET: guest faclist: 0x%16.16llx.%16.16llx.%16.16llx",
			 kvm->arch.model.fac_list[0],
			 kvm->arch.model.fac_list[1],
			 kvm->arch.model.fac_list[2]);
1118 1119 1120 1121 1122 1123 1124 1125
	} else
		ret = -EFAULT;
	kfree(proc);
out:
	mutex_unlock(&kvm->lock);
	return ret;
}

1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138
static int kvm_s390_set_processor_feat(struct kvm *kvm,
				       struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_feat data;

	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);
1139 1140 1141
	if (kvm->created_vcpus) {
		mutex_unlock(&kvm->lock);
		return -EBUSY;
1142
	}
1143 1144
	bitmap_copy(kvm->arch.cpu_feat, (unsigned long *) data.feat,
		    KVM_S390_VM_CPU_FEAT_NR_BITS);
1145
	mutex_unlock(&kvm->lock);
1146 1147 1148 1149 1150
	VM_EVENT(kvm, 3, "SET: guest feat: 0x%16.16llx.0x%16.16llx.0x%16.16llx",
			 data.feat[0],
			 data.feat[1],
			 data.feat[2]);
	return 0;
1151 1152
}

1153 1154 1155 1156 1157 1158 1159 1160 1161 1162
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;
}

1163 1164 1165 1166 1167 1168 1169 1170
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;
1171 1172 1173
	case KVM_S390_VM_CPU_PROCESSOR_FEAT:
		ret = kvm_s390_set_processor_feat(kvm, attr);
		break;
1174 1175 1176
	case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
		ret = kvm_s390_set_processor_subfunc(kvm, attr);
		break;
1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190
	}
	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;
	}
1191
	proc->cpuid = kvm->arch.model.cpuid;
1192
	proc->ibc = kvm->arch.model.ibc;
1193 1194
	memcpy(&proc->fac_list, kvm->arch.model.fac_list,
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
1195 1196 1197 1198 1199 1200 1201
	VM_EVENT(kvm, 3, "GET: guest ibc: 0x%4.4x, guest cpuid: 0x%16.16llx",
		 kvm->arch.model.ibc,
		 kvm->arch.model.cpuid);
	VM_EVENT(kvm, 3, "GET: guest faclist: 0x%16.16llx.%16.16llx.%16.16llx",
		 kvm->arch.model.fac_list[0],
		 kvm->arch.model.fac_list[1],
		 kvm->arch.model.fac_list[2]);
1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219
	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);
1220
	mach->ibc = sclp.ibc;
1221
	memcpy(&mach->fac_mask, kvm->arch.model.fac_mask,
1222
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
1223
	memcpy((unsigned long *)&mach->fac_list, S390_lowcore.stfle_fac_list,
1224
	       sizeof(S390_lowcore.stfle_fac_list));
1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235
	VM_EVENT(kvm, 3, "GET: host ibc:  0x%4.4x, host cpuid:  0x%16.16llx",
		 kvm->arch.model.ibc,
		 kvm->arch.model.cpuid);
	VM_EVENT(kvm, 3, "GET: host facmask:  0x%16.16llx.%16.16llx.%16.16llx",
		 mach->fac_mask[0],
		 mach->fac_mask[1],
		 mach->fac_mask[2]);
	VM_EVENT(kvm, 3, "GET: host faclist:  0x%16.16llx.%16.16llx.%16.16llx",
		 mach->fac_list[0],
		 mach->fac_list[1],
		 mach->fac_list[2]);
1236 1237 1238 1239 1240 1241 1242
	if (copy_to_user((void __user *)attr->addr, mach, sizeof(*mach)))
		ret = -EFAULT;
	kfree(mach);
out:
	return ret;
}

1243 1244 1245 1246 1247 1248 1249 1250 1251
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;
1252 1253 1254 1255
	VM_EVENT(kvm, 3, "GET: guest feat: 0x%16.16llx.0x%16.16llx.0x%16.16llx",
			 data.feat[0],
			 data.feat[1],
			 data.feat[2]);
1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268
	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;
1269 1270 1271 1272
	VM_EVENT(kvm, 3, "GET: host feat:  0x%16.16llx.0x%16.16llx.0x%16.16llx",
			 data.feat[0],
			 data.feat[1],
			 data.feat[2]);
1273 1274 1275
	return 0;
}

1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294
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;
}
1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305
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;
1306 1307 1308 1309 1310 1311
	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;
1312 1313 1314 1315 1316 1317
	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;
1318 1319 1320 1321
	}
	return ret;
}

1322 1323 1324 1325 1326
static int kvm_s390_vm_set_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret;

	switch (attr->group) {
1327
	case KVM_S390_VM_MEM_CTRL:
1328
		ret = kvm_s390_set_mem_control(kvm, attr);
1329
		break;
1330 1331 1332
	case KVM_S390_VM_TOD:
		ret = kvm_s390_set_tod(kvm, attr);
		break;
1333 1334 1335
	case KVM_S390_VM_CPU_MODEL:
		ret = kvm_s390_set_cpu_model(kvm, attr);
		break;
1336 1337 1338
	case KVM_S390_VM_CRYPTO:
		ret = kvm_s390_vm_set_crypto(kvm, attr);
		break;
1339 1340 1341
	case KVM_S390_VM_MIGRATION:
		ret = kvm_s390_vm_set_migration(kvm, attr);
		break;
1342 1343 1344 1345 1346 1347 1348 1349 1350 1351
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

static int kvm_s390_vm_get_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
1352 1353 1354 1355 1356 1357
	int ret;

	switch (attr->group) {
	case KVM_S390_VM_MEM_CTRL:
		ret = kvm_s390_get_mem_control(kvm, attr);
		break;
1358 1359 1360
	case KVM_S390_VM_TOD:
		ret = kvm_s390_get_tod(kvm, attr);
		break;
1361 1362 1363
	case KVM_S390_VM_CPU_MODEL:
		ret = kvm_s390_get_cpu_model(kvm, attr);
		break;
1364 1365 1366
	case KVM_S390_VM_MIGRATION:
		ret = kvm_s390_vm_get_migration(kvm, attr);
		break;
1367 1368 1369 1370 1371 1372
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
1373 1374 1375 1376 1377 1378 1379
}

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

	switch (attr->group) {
1380 1381 1382 1383
	case KVM_S390_VM_MEM_CTRL:
		switch (attr->attr) {
		case KVM_S390_VM_MEM_ENABLE_CMMA:
		case KVM_S390_VM_MEM_CLR_CMMA:
1384 1385
			ret = sclp.has_cmma ? 0 : -ENXIO;
			break;
1386
		case KVM_S390_VM_MEM_LIMIT_SIZE:
1387 1388 1389 1390 1391 1392 1393
			ret = 0;
			break;
		default:
			ret = -ENXIO;
			break;
		}
		break;
1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404
	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;
1405 1406 1407 1408
	case KVM_S390_VM_CPU_MODEL:
		switch (attr->attr) {
		case KVM_S390_VM_CPU_PROCESSOR:
		case KVM_S390_VM_CPU_MACHINE:
1409 1410
		case KVM_S390_VM_CPU_PROCESSOR_FEAT:
		case KVM_S390_VM_CPU_MACHINE_FEAT:
1411
		case KVM_S390_VM_CPU_MACHINE_SUBFUNC:
1412 1413
			ret = 0;
			break;
1414 1415
		/* configuring subfunctions is not supported yet */
		case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
1416 1417 1418 1419 1420
		default:
			ret = -ENXIO;
			break;
		}
		break;
1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433
	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;
1434 1435 1436
	case KVM_S390_VM_MIGRATION:
		ret = 0;
		break;
1437 1438 1439 1440 1441 1442 1443 1444
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

1445 1446 1447 1448
static long kvm_s390_get_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
{
	uint8_t *keys;
	uint64_t hva;
1449
	int srcu_idx, i, r = 0;
1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461

	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;

1462
	keys = kvmalloc_array(args->count, sizeof(uint8_t), GFP_KERNEL);
1463 1464 1465
	if (!keys)
		return -ENOMEM;

1466
	down_read(&current->mm->mmap_sem);
1467
	srcu_idx = srcu_read_lock(&kvm->srcu);
1468 1469 1470 1471
	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
1472
			break;
1473 1474
		}

1475 1476
		r = get_guest_storage_key(current->mm, hva, &keys[i]);
		if (r)
1477
			break;
1478
	}
1479
	srcu_read_unlock(&kvm->srcu, srcu_idx);
1480 1481 1482 1483 1484 1485 1486
	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;
1487 1488 1489 1490 1491 1492 1493 1494 1495 1496
	}

	kvfree(keys);
	return r;
}

static long kvm_s390_set_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
{
	uint8_t *keys;
	uint64_t hva;
1497
	int srcu_idx, i, r = 0;
1498 1499 1500 1501 1502 1503 1504 1505

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

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

1506
	keys = kvmalloc_array(args->count, sizeof(uint8_t), GFP_KERNEL);
1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517
	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 */
1518 1519 1520
	r = s390_enable_skey();
	if (r)
		goto out;
1521

1522
	down_read(&current->mm->mmap_sem);
1523
	srcu_idx = srcu_read_lock(&kvm->srcu);
1524 1525 1526 1527
	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
1528
			break;
1529 1530 1531 1532 1533
		}

		/* Lowest order bit is reserved */
		if (keys[i] & 0x01) {
			r = -EINVAL;
1534
			break;
1535 1536
		}

1537
		r = set_guest_storage_key(current->mm, hva, keys[i], 0);
1538
		if (r)
1539
			break;
1540
	}
1541
	srcu_read_unlock(&kvm->srcu, srcu_idx);
1542
	up_read(&current->mm->mmap_sem);
1543 1544 1545 1546 1547
out:
	kvfree(keys);
	return r;
}

1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 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
/*
 * Base address and length must be sent at the start of each block, therefore
 * it's cheaper to send some clean data, as long as it's less than the size of
 * two longs.
 */
#define KVM_S390_MAX_BIT_DISTANCE (2 * sizeof(void *))
/* for consistency */
#define KVM_S390_CMMA_SIZE_MAX ((u32)KVM_S390_SKEYS_MAX)

/*
 * This function searches for the next page with dirty CMMA attributes, and
 * saves the attributes in the buffer up to either the end of the buffer or
 * until a block of at least KVM_S390_MAX_BIT_DISTANCE clean bits is found;
 * no trailing clean bytes are saved.
 * In case no dirty bits were found, or if CMMA was not enabled or used, the
 * output buffer will indicate 0 as length.
 */
static int kvm_s390_get_cmma_bits(struct kvm *kvm,
				  struct kvm_s390_cmma_log *args)
{
	struct kvm_s390_migration_state *s = kvm->arch.migration_state;
	unsigned long bufsize, hva, pgstev, i, next, cur;
	int srcu_idx, peek, r = 0, rr;
	u8 *res;

	cur = args->start_gfn;
	i = next = pgstev = 0;

	if (unlikely(!kvm->arch.use_cmma))
		return -ENXIO;
	/* Invalid/unsupported flags were specified */
	if (args->flags & ~KVM_S390_CMMA_PEEK)
		return -EINVAL;
	/* Migration mode query, and we are not doing a migration */
	peek = !!(args->flags & KVM_S390_CMMA_PEEK);
	if (!peek && !s)
		return -EINVAL;
	/* CMMA is disabled or was not used, or the buffer has length zero */
	bufsize = min(args->count, KVM_S390_CMMA_SIZE_MAX);
	if (!bufsize || !kvm->mm->context.use_cmma) {
		memset(args, 0, sizeof(*args));
		return 0;
	}

	if (!peek) {
		/* We are not peeking, and there are no dirty pages */
		if (!atomic64_read(&s->dirty_pages)) {
			memset(args, 0, sizeof(*args));
			return 0;
		}
		cur = find_next_bit(s->pgste_bitmap, s->bitmap_size,
				    args->start_gfn);
		if (cur >= s->bitmap_size)	/* nothing found, loop back */
			cur = find_next_bit(s->pgste_bitmap, s->bitmap_size, 0);
		if (cur >= s->bitmap_size) {	/* again! (very unlikely) */
			memset(args, 0, sizeof(*args));
			return 0;
		}
		next = find_next_bit(s->pgste_bitmap, s->bitmap_size, cur + 1);
	}

	res = vmalloc(bufsize);
	if (!res)
		return -ENOMEM;

	args->start_gfn = cur;

	down_read(&kvm->mm->mmap_sem);
	srcu_idx = srcu_read_lock(&kvm->srcu);
	while (i < bufsize) {
		hva = gfn_to_hva(kvm, cur);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
			break;
		}
		/* decrement only if we actually flipped the bit to 0 */
		if (!peek && test_and_clear_bit(cur, s->pgste_bitmap))
			atomic64_dec(&s->dirty_pages);
		r = get_pgste(kvm->mm, hva, &pgstev);
		if (r < 0)
			pgstev = 0;
		/* save the value */
1630
		res[i++] = (pgstev >> 24) & 0x43;
1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707
		/*
		 * if the next bit is too far away, stop.
		 * if we reached the previous "next", find the next one
		 */
		if (!peek) {
			if (next > cur + KVM_S390_MAX_BIT_DISTANCE)
				break;
			if (cur == next)
				next = find_next_bit(s->pgste_bitmap,
						     s->bitmap_size, cur + 1);
		/* reached the end of the bitmap or of the buffer, stop */
			if ((next >= s->bitmap_size) ||
			    (next >= args->start_gfn + bufsize))
				break;
		}
		cur++;
	}
	srcu_read_unlock(&kvm->srcu, srcu_idx);
	up_read(&kvm->mm->mmap_sem);
	args->count = i;
	args->remaining = s ? atomic64_read(&s->dirty_pages) : 0;

	rr = copy_to_user((void __user *)args->values, res, args->count);
	if (rr)
		r = -EFAULT;

	vfree(res);
	return r;
}

/*
 * This function sets the CMMA attributes for the given pages. If the input
 * buffer has zero length, no action is taken, otherwise the attributes are
 * set and the mm->context.use_cmma flag is set.
 */
static int kvm_s390_set_cmma_bits(struct kvm *kvm,
				  const struct kvm_s390_cmma_log *args)
{
	unsigned long hva, mask, pgstev, i;
	uint8_t *bits;
	int srcu_idx, r = 0;

	mask = args->mask;

	if (!kvm->arch.use_cmma)
		return -ENXIO;
	/* invalid/unsupported flags */
	if (args->flags != 0)
		return -EINVAL;
	/* Enforce sane limit on memory allocation */
	if (args->count > KVM_S390_CMMA_SIZE_MAX)
		return -EINVAL;
	/* Nothing to do */
	if (args->count == 0)
		return 0;

	bits = vmalloc(sizeof(*bits) * args->count);
	if (!bits)
		return -ENOMEM;

	r = copy_from_user(bits, (void __user *)args->values, args->count);
	if (r) {
		r = -EFAULT;
		goto out;
	}

	down_read(&kvm->mm->mmap_sem);
	srcu_idx = srcu_read_lock(&kvm->srcu);
	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
			break;
		}

		pgstev = bits[i];
		pgstev = pgstev << 24;
1708
		mask &= _PGSTE_GPS_USAGE_MASK | _PGSTE_GPS_NODAT;
1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723
		set_pgste_bits(kvm->mm, hva, mask, pgstev);
	}
	srcu_read_unlock(&kvm->srcu, srcu_idx);
	up_read(&kvm->mm->mmap_sem);

	if (!kvm->mm->context.use_cmma) {
		down_write(&kvm->mm->mmap_sem);
		kvm->mm->context.use_cmma = 1;
		up_write(&kvm->mm->mmap_sem);
	}
out:
	vfree(bits);
	return r;
}

1724 1725 1726 1727 1728
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;
1729
	struct kvm_device_attr attr;
1730 1731 1732
	int r;

	switch (ioctl) {
1733 1734 1735 1736 1737 1738 1739 1740 1741
	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;
	}
1742 1743 1744 1745 1746 1747 1748 1749
	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;
	}
1750 1751 1752 1753 1754 1755 1756
	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));
1757
			r = kvm_set_irq_routing(kvm, &routing, 0, 0);
1758 1759 1760
		}
		break;
	}
1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781
	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;
	}
1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801
	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;
	}
1802 1803 1804 1805 1806 1807
	case KVM_S390_GET_CMMA_BITS: {
		struct kvm_s390_cmma_log args;

		r = -EFAULT;
		if (copy_from_user(&args, argp, sizeof(args)))
			break;
1808
		mutex_lock(&kvm->slots_lock);
1809
		r = kvm_s390_get_cmma_bits(kvm, &args);
1810
		mutex_unlock(&kvm->slots_lock);
1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823
		if (!r) {
			r = copy_to_user(argp, &args, sizeof(args));
			if (r)
				r = -EFAULT;
		}
		break;
	}
	case KVM_S390_SET_CMMA_BITS: {
		struct kvm_s390_cmma_log args;

		r = -EFAULT;
		if (copy_from_user(&args, argp, sizeof(args)))
			break;
1824
		mutex_lock(&kvm->slots_lock);
1825
		r = kvm_s390_set_cmma_bits(kvm, &args);
1826
		mutex_unlock(&kvm->slots_lock);
1827 1828
		break;
	}
1829
	default:
1830
		r = -ENOTTY;
1831 1832 1833 1834 1835
	}

	return r;
}

1836 1837 1838
static int kvm_s390_query_ap_config(u8 *config)
{
	u32 fcn_code = 0x04000000UL;
1839
	u32 cc = 0;
1840

1841
	memset(config, 0, 128);
1842 1843 1844 1845
	asm volatile(
		"lgr 0,%1\n"
		"lgr 2,%2\n"
		".long 0xb2af0000\n"		/* PQAP(QCI) */
1846
		"0: ipm %0\n"
1847
		"srl %0,28\n"
1848 1849 1850
		"1:\n"
		EX_TABLE(0b, 1b)
		: "+r" (cc)
1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862
		: "r" (fcn_code), "r" (config)
		: "cc", "0", "2", "memory"
	);

	return cc;
}

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

1863
	if (test_facility(12)) {
1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884
		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;
}

1885
static u64 kvm_s390_get_initial_cpuid(void)
1886
{
1887 1888 1889 1890 1891
	struct cpuid cpuid;

	get_cpu_id(&cpuid);
	cpuid.version = 0xff;
	return *((u64 *) &cpuid);
1892 1893
}

1894
static void kvm_s390_crypto_init(struct kvm *kvm)
1895
{
1896
	if (!test_kvm_facility(kvm, 76))
1897
		return;
1898

1899
	kvm->arch.crypto.crycb = &kvm->arch.sie_page2->crycb;
1900
	kvm_s390_set_crycb_format(kvm);
1901

1902 1903 1904 1905 1906 1907 1908
	/* 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));
1909 1910
}

1911 1912 1913
static void sca_dispose(struct kvm *kvm)
{
	if (kvm->arch.use_esca)
1914
		free_pages_exact(kvm->arch.sca, sizeof(struct esca_block));
1915 1916 1917 1918 1919
	else
		free_page((unsigned long)(kvm->arch.sca));
	kvm->arch.sca = NULL;
}

1920
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
1921
{
1922
	gfp_t alloc_flags = GFP_KERNEL;
1923
	int i, rc;
1924
	char debug_name[16];
1925
	static unsigned long sca_offset;
1926

1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937
	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

1938 1939
	rc = s390_enable_sie();
	if (rc)
1940
		goto out_err;
1941

1942 1943
	rc = -ENOMEM;

1944
	kvm->arch.use_esca = 0; /* start with basic SCA */
1945 1946
	if (!sclp.has_64bscao)
		alloc_flags |= GFP_DMA;
1947
	rwlock_init(&kvm->arch.sca_lock);
1948
	kvm->arch.sca = (struct bsca_block *) get_zeroed_page(alloc_flags);
1949
	if (!kvm->arch.sca)
1950
		goto out_err;
1951
	spin_lock(&kvm_lock);
1952
	sca_offset += 16;
1953
	if (sca_offset + sizeof(struct bsca_block) > PAGE_SIZE)
1954
		sca_offset = 0;
1955 1956
	kvm->arch.sca = (struct bsca_block *)
			((char *) kvm->arch.sca + sca_offset);
1957
	spin_unlock(&kvm_lock);
1958 1959 1960

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

1961
	kvm->arch.dbf = debug_register(debug_name, 32, 1, 7 * sizeof(long));
1962
	if (!kvm->arch.dbf)
1963
		goto out_err;
1964

1965
	BUILD_BUG_ON(sizeof(struct sie_page2) != 4096);
1966 1967 1968
	kvm->arch.sie_page2 =
	     (struct sie_page2 *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
	if (!kvm->arch.sie_page2)
1969
		goto out_err;
1970

1971
	/* Populate the facility mask initially. */
1972
	memcpy(kvm->arch.model.fac_mask, S390_lowcore.stfle_fac_list,
1973
	       sizeof(S390_lowcore.stfle_fac_list));
1974 1975
	for (i = 0; i < S390_ARCH_FAC_LIST_SIZE_U64; i++) {
		if (i < kvm_s390_fac_list_mask_size())
1976
			kvm->arch.model.fac_mask[i] &= kvm_s390_fac_list_mask[i];
1977
		else
1978
			kvm->arch.model.fac_mask[i] = 0UL;
1979 1980
	}

1981
	/* Populate the facility list initially. */
1982 1983
	kvm->arch.model.fac_list = kvm->arch.sie_page2->fac_list;
	memcpy(kvm->arch.model.fac_list, kvm->arch.model.fac_mask,
1984 1985
	       S390_ARCH_FAC_LIST_SIZE_BYTE);

1986 1987 1988 1989
	/* we are always in czam mode - even on pre z14 machines */
	set_kvm_facility(kvm->arch.model.fac_mask, 138);
	set_kvm_facility(kvm->arch.model.fac_list, 138);
	/* we emulate STHYI in kvm */
J
Janosch Frank 已提交
1990 1991
	set_kvm_facility(kvm->arch.model.fac_mask, 74);
	set_kvm_facility(kvm->arch.model.fac_list, 74);
1992 1993 1994 1995
	if (MACHINE_HAS_TLB_GUEST) {
		set_kvm_facility(kvm->arch.model.fac_mask, 147);
		set_kvm_facility(kvm->arch.model.fac_list, 147);
	}
J
Janosch Frank 已提交
1996

1997
	kvm->arch.model.cpuid = kvm_s390_get_initial_cpuid();
1998
	kvm->arch.model.ibc = sclp.ibc & 0x0fff;
1999

2000
	kvm_s390_crypto_init(kvm);
2001

2002 2003 2004
	mutex_init(&kvm->arch.float_int.ais_lock);
	kvm->arch.float_int.simm = 0;
	kvm->arch.float_int.nimm = 0;
2005
	spin_lock_init(&kvm->arch.float_int.lock);
2006 2007
	for (i = 0; i < FIRQ_LIST_COUNT; i++)
		INIT_LIST_HEAD(&kvm->arch.float_int.lists[i]);
2008
	init_waitqueue_head(&kvm->arch.ipte_wq);
2009
	mutex_init(&kvm->arch.ipte_mutex);
2010

2011
	debug_register_view(kvm->arch.dbf, &debug_sprintf_view);
2012
	VM_EVENT(kvm, 3, "vm created with type %lu", type);
2013

2014 2015
	if (type & KVM_VM_S390_UCONTROL) {
		kvm->arch.gmap = NULL;
2016
		kvm->arch.mem_limit = KVM_S390_NO_MEM_LIMIT;
2017
	} else {
2018
		if (sclp.hamax == U64_MAX)
2019
			kvm->arch.mem_limit = TASK_SIZE_MAX;
2020
		else
2021
			kvm->arch.mem_limit = min_t(unsigned long, TASK_SIZE_MAX,
2022
						    sclp.hamax + 1);
2023
		kvm->arch.gmap = gmap_create(current->mm, kvm->arch.mem_limit - 1);
2024
		if (!kvm->arch.gmap)
2025
			goto out_err;
2026
		kvm->arch.gmap->private = kvm;
2027
		kvm->arch.gmap->pfault_enabled = 0;
2028
	}
2029 2030

	kvm->arch.css_support = 0;
2031
	kvm->arch.use_irqchip = 0;
2032
	kvm->arch.epoch = 0;
2033

2034
	spin_lock_init(&kvm->arch.start_stop_lock);
2035
	kvm_s390_vsie_init(kvm);
2036
	kvm_s390_gisa_init(kvm);
2037
	KVM_EVENT(3, "vm 0x%pK created by pid %u", kvm, current->pid);
2038

2039
	return 0;
2040
out_err:
2041
	free_page((unsigned long)kvm->arch.sie_page2);
2042
	debug_unregister(kvm->arch.dbf);
2043
	sca_dispose(kvm);
2044
	KVM_EVENT(3, "creation of vm failed: %d", rc);
2045
	return rc;
2046 2047
}

2048 2049 2050 2051 2052 2053 2054 2055 2056 2057
bool kvm_arch_has_vcpu_debugfs(void)
{
	return false;
}

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

2058 2059 2060
void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
{
	VCPU_EVENT(vcpu, 3, "%s", "free cpu");
2061
	trace_kvm_s390_destroy_vcpu(vcpu->vcpu_id);
2062
	kvm_s390_clear_local_irqs(vcpu);
2063
	kvm_clear_async_pf_completion_queue(vcpu);
2064
	if (!kvm_is_ucontrol(vcpu->kvm))
2065
		sca_del_vcpu(vcpu);
2066 2067

	if (kvm_is_ucontrol(vcpu->kvm))
2068
		gmap_remove(vcpu->arch.gmap);
2069

2070
	if (vcpu->kvm->arch.use_cmma)
2071
		kvm_s390_vcpu_unsetup_cmma(vcpu);
2072
	free_page((unsigned long)(vcpu->arch.sie_block));
2073

2074
	kvm_vcpu_uninit(vcpu);
2075
	kmem_cache_free(kvm_vcpu_cache, vcpu);
2076 2077 2078 2079 2080
}

static void kvm_free_vcpus(struct kvm *kvm)
{
	unsigned int i;
2081
	struct kvm_vcpu *vcpu;
2082

2083 2084 2085 2086 2087 2088 2089 2090 2091
	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);
2092 2093
}

2094 2095
void kvm_arch_destroy_vm(struct kvm *kvm)
{
2096
	kvm_free_vcpus(kvm);
2097
	sca_dispose(kvm);
2098
	debug_unregister(kvm->arch.dbf);
2099
	kvm_s390_gisa_destroy(kvm);
2100
	free_page((unsigned long)kvm->arch.sie_page2);
2101
	if (!kvm_is_ucontrol(kvm))
2102
		gmap_remove(kvm->arch.gmap);
2103
	kvm_s390_destroy_adapters(kvm);
2104
	kvm_s390_clear_float_irqs(kvm);
2105
	kvm_s390_vsie_destroy(kvm);
2106 2107 2108 2109
	if (kvm->arch.migration_state) {
		vfree(kvm->arch.migration_state->pgste_bitmap);
		kfree(kvm->arch.migration_state);
	}
2110
	KVM_EVENT(3, "vm 0x%pK destroyed", kvm);
2111 2112 2113
}

/* Section: vcpu related */
2114 2115
static int __kvm_ucontrol_vcpu_init(struct kvm_vcpu *vcpu)
{
2116
	vcpu->arch.gmap = gmap_create(current->mm, -1UL);
2117 2118 2119 2120 2121 2122 2123
	if (!vcpu->arch.gmap)
		return -ENOMEM;
	vcpu->arch.gmap->private = vcpu->kvm;

	return 0;
}

2124 2125
static void sca_del_vcpu(struct kvm_vcpu *vcpu)
{
2126 2127
	if (!kvm_s390_use_sca_entries())
		return;
2128
	read_lock(&vcpu->kvm->arch.sca_lock);
2129 2130
	if (vcpu->kvm->arch.use_esca) {
		struct esca_block *sca = vcpu->kvm->arch.sca;
2131

2132
		clear_bit_inv(vcpu->vcpu_id, (unsigned long *) sca->mcn);
2133
		sca->cpu[vcpu->vcpu_id].sda = 0;
2134 2135 2136 2137
	} else {
		struct bsca_block *sca = vcpu->kvm->arch.sca;

		clear_bit_inv(vcpu->vcpu_id, (unsigned long *) &sca->mcn);
2138
		sca->cpu[vcpu->vcpu_id].sda = 0;
2139
	}
2140
	read_unlock(&vcpu->kvm->arch.sca_lock);
2141 2142
}

2143
static void sca_add_vcpu(struct kvm_vcpu *vcpu)
2144
{
2145 2146 2147 2148 2149 2150 2151
	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;
	}
2152 2153 2154
	read_lock(&vcpu->kvm->arch.sca_lock);
	if (vcpu->kvm->arch.use_esca) {
		struct esca_block *sca = vcpu->kvm->arch.sca;
2155

2156
		sca->cpu[vcpu->vcpu_id].sda = (__u64) vcpu->arch.sie_block;
2157 2158
		vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
		vcpu->arch.sie_block->scaol = (__u32)(__u64)sca & ~0x3fU;
2159
		vcpu->arch.sie_block->ecb2 |= ECB2_ESCA;
2160
		set_bit_inv(vcpu->vcpu_id, (unsigned long *) sca->mcn);
2161
	} else {
2162
		struct bsca_block *sca = vcpu->kvm->arch.sca;
2163

2164
		sca->cpu[vcpu->vcpu_id].sda = (__u64) vcpu->arch.sie_block;
2165 2166
		vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
		vcpu->arch.sie_block->scaol = (__u32)(__u64)sca;
2167
		set_bit_inv(vcpu->vcpu_id, (unsigned long *) &sca->mcn);
2168
	}
2169
	read_unlock(&vcpu->kvm->arch.sca_lock);
2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212
}

/* 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;
2213
		vcpu->arch.sie_block->ecb2 |= ECB2_ESCA;
2214 2215 2216 2217 2218 2219 2220 2221 2222
	}
	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);

2223 2224
	VM_EVENT(kvm, 2, "Switched to ESCA (0x%pK -> 0x%pK)",
		 old_sca, kvm->arch.sca);
2225
	return 0;
2226 2227 2228 2229
}

static int sca_can_add_vcpu(struct kvm *kvm, unsigned int id)
{
2230 2231
	int rc;

2232 2233 2234 2235 2236
	if (!kvm_s390_use_sca_entries()) {
		if (id < KVM_MAX_VCPUS)
			return true;
		return false;
	}
2237 2238
	if (id < KVM_S390_BSCA_CPU_SLOTS)
		return true;
2239
	if (!sclp.has_esca || !sclp.has_64bscao)
2240 2241 2242 2243 2244 2245 2246
		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;
2247 2248
}

2249 2250
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
2251 2252
	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
	kvm_clear_async_pf_completion_queue(vcpu);
2253 2254
	vcpu->run->kvm_valid_regs = KVM_SYNC_PREFIX |
				    KVM_SYNC_GPRS |
2255
				    KVM_SYNC_ACRS |
2256 2257 2258
				    KVM_SYNC_CRS |
				    KVM_SYNC_ARCH0 |
				    KVM_SYNC_PFAULT;
2259
	kvm_s390_set_prefix(vcpu, 0);
2260 2261
	if (test_kvm_facility(vcpu->kvm, 64))
		vcpu->run->kvm_valid_regs |= KVM_SYNC_RICCB;
2262 2263
	if (test_kvm_facility(vcpu->kvm, 82))
		vcpu->run->kvm_valid_regs |= KVM_SYNC_BPBC;
F
Fan Zhang 已提交
2264 2265
	if (test_kvm_facility(vcpu->kvm, 133))
		vcpu->run->kvm_valid_regs |= KVM_SYNC_GSCB;
2266 2267 2268 2269
	/* 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)
2270
		vcpu->run->kvm_valid_regs |= KVM_SYNC_VRS;
2271 2272
	else
		vcpu->run->kvm_valid_regs |= KVM_SYNC_FPRS;
2273 2274 2275 2276

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

2277 2278 2279
	return 0;
}

2280 2281 2282 2283
/* 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);
2284
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
2285
	vcpu->arch.cputm_start = get_tod_clock_fast();
2286
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
2287 2288 2289 2290 2291 2292
}

/* 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);
2293
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
2294 2295
	vcpu->arch.sie_block->cputm -= get_tod_clock_fast() - vcpu->arch.cputm_start;
	vcpu->arch.cputm_start = 0;
2296
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328
}

/* 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();
}

2329 2330 2331
/* 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)
{
2332
	preempt_disable(); /* protect from TOD sync and vcpu_load/put */
2333
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
2334 2335
	if (vcpu->arch.cputm_enabled)
		vcpu->arch.cputm_start = get_tod_clock_fast();
2336
	vcpu->arch.sie_block->cputm = cputm;
2337
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
2338
	preempt_enable();
2339 2340
}

2341
/* update and get the cpu timer - can also be called from other VCPU threads */
2342 2343
__u64 kvm_s390_get_cpu_timer(struct kvm_vcpu *vcpu)
{
2344
	unsigned int seq;
2345 2346 2347 2348 2349
	__u64 value;

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

2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363
	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();
2364
	return value;
2365 2366
}

2367 2368
void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
2369

2370
	gmap_enable(vcpu->arch.enabled_gmap);
2371
	kvm_s390_set_cpuflags(vcpu, CPUSTAT_RUNNING);
2372
	if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
2373
		__start_cpu_timer_accounting(vcpu);
2374
	vcpu->cpu = cpu;
2375 2376 2377 2378
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
2379
	vcpu->cpu = -1;
2380
	if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
2381
		__stop_cpu_timer_accounting(vcpu);
2382
	kvm_s390_clear_cpuflags(vcpu, CPUSTAT_RUNNING);
2383 2384
	vcpu->arch.enabled_gmap = gmap_get_enabled();
	gmap_disable(vcpu->arch.enabled_gmap);
2385

2386 2387 2388 2389 2390 2391 2392
}

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;
2393
	kvm_s390_set_prefix(vcpu, 0);
2394
	kvm_s390_set_cpu_timer(vcpu, 0);
2395 2396 2397 2398 2399
	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;
2400 2401 2402
	/* make sure the new fpc will be lazily loaded */
	save_fpu_regs();
	current->thread.fpu.fpc = 0;
2403
	vcpu->arch.sie_block->gbea = 1;
2404
	vcpu->arch.sie_block->pp = 0;
2405
	vcpu->arch.sie_block->fpf &= ~FPF_BPBC;
2406 2407
	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
	kvm_clear_async_pf_completion_queue(vcpu);
2408 2409
	if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm))
		kvm_s390_vcpu_stop(vcpu);
2410
	kvm_s390_clear_local_irqs(vcpu);
2411 2412
}

2413
void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
2414
{
2415
	mutex_lock(&vcpu->kvm->lock);
2416
	preempt_disable();
2417
	vcpu->arch.sie_block->epoch = vcpu->kvm->arch.epoch;
2418
	vcpu->arch.sie_block->epdx = vcpu->kvm->arch.epdx;
2419
	preempt_enable();
2420
	mutex_unlock(&vcpu->kvm->lock);
2421
	if (!kvm_is_ucontrol(vcpu->kvm)) {
2422
		vcpu->arch.gmap = vcpu->kvm->arch.gmap;
2423
		sca_add_vcpu(vcpu);
2424
	}
2425 2426
	if (test_kvm_facility(vcpu->kvm, 74) || vcpu->kvm->arch.user_instr0)
		vcpu->arch.sie_block->ictl |= ICTL_OPEREXC;
2427 2428
	/* make vcpu_load load the right gmap on the first trigger */
	vcpu->arch.enabled_gmap = vcpu->arch.gmap;
2429 2430
}

2431 2432
static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu)
{
2433
	if (!test_kvm_facility(vcpu->kvm, 76))
2434 2435
		return;

2436 2437 2438 2439 2440 2441 2442
	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;

2443 2444 2445
	vcpu->arch.sie_block->crycbd = vcpu->kvm->arch.crypto.crycbd;
}

2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457
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;

2458
	vcpu->arch.sie_block->ecb2 &= ~ECB2_PFMFI;
2459 2460 2461
	return 0;
}

2462 2463 2464 2465 2466
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;
2467
	if (test_kvm_facility(vcpu->kvm, 7))
2468
		vcpu->arch.sie_block->fac = (u32)(u64) model->fac_list;
2469 2470
}

2471 2472
int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
{
2473
	int rc = 0;
2474

2475 2476
	atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH |
						    CPUSTAT_SM |
2477 2478
						    CPUSTAT_STOPPED);

2479
	if (test_kvm_facility(vcpu->kvm, 78))
2480
		kvm_s390_set_cpuflags(vcpu, CPUSTAT_GED2);
2481
	else if (test_kvm_facility(vcpu->kvm, 8))
2482
		kvm_s390_set_cpuflags(vcpu, CPUSTAT_GED);
2483

2484 2485
	kvm_s390_vcpu_setup_model(vcpu);

2486 2487
	/* pgste_set_pte has special handling for !MACHINE_HAS_ESOP */
	if (MACHINE_HAS_ESOP)
2488
		vcpu->arch.sie_block->ecb |= ECB_HOSTPROTINT;
2489
	if (test_kvm_facility(vcpu->kvm, 9))
2490
		vcpu->arch.sie_block->ecb |= ECB_SRSI;
2491
	if (test_kvm_facility(vcpu->kvm, 73))
2492
		vcpu->arch.sie_block->ecb |= ECB_TE;
2493

2494
	if (test_kvm_facility(vcpu->kvm, 8) && sclp.has_pfmfi)
2495
		vcpu->arch.sie_block->ecb2 |= ECB2_PFMFI;
2496
	if (test_kvm_facility(vcpu->kvm, 130))
2497 2498
		vcpu->arch.sie_block->ecb2 |= ECB2_IEP;
	vcpu->arch.sie_block->eca = ECA_MVPGI | ECA_PROTEXCI;
2499
	if (sclp.has_cei)
2500
		vcpu->arch.sie_block->eca |= ECA_CEI;
2501
	if (sclp.has_ib)
2502
		vcpu->arch.sie_block->eca |= ECA_IB;
2503
	if (sclp.has_siif)
2504
		vcpu->arch.sie_block->eca |= ECA_SII;
2505
	if (sclp.has_sigpif)
2506
		vcpu->arch.sie_block->eca |= ECA_SIGPI;
2507
	if (test_kvm_facility(vcpu->kvm, 129)) {
2508 2509
		vcpu->arch.sie_block->eca |= ECA_VX;
		vcpu->arch.sie_block->ecd |= ECD_HOSTREGMGMT;
2510
	}
2511 2512 2513
	if (test_kvm_facility(vcpu->kvm, 139))
		vcpu->arch.sie_block->ecd |= ECD_MEF;

2514 2515 2516 2517 2518
	if (vcpu->arch.sie_block->gd) {
		vcpu->arch.sie_block->eca |= ECA_AIV;
		VCPU_EVENT(vcpu, 3, "AIV gisa format-%u enabled for cpu %03u",
			   vcpu->arch.sie_block->gd & 0x3, vcpu->vcpu_id);
	}
F
Fan Zhang 已提交
2519 2520
	vcpu->arch.sie_block->sdnxo = ((unsigned long) &vcpu->run->s.regs.sdnx)
					| SDNXC;
2521
	vcpu->arch.sie_block->riccbd = (unsigned long) &vcpu->run->s.regs.riccb;
2522 2523

	if (sclp.has_kss)
2524
		kvm_s390_set_cpuflags(vcpu, CPUSTAT_KSS);
2525 2526
	else
		vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE;
2527

2528
	if (vcpu->kvm->arch.use_cmma) {
2529 2530 2531
		rc = kvm_s390_vcpu_setup_cmma(vcpu);
		if (rc)
			return rc;
2532
	}
2533
	hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
2534
	vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup;
2535

2536 2537
	kvm_s390_vcpu_crypto_setup(vcpu);

2538
	return rc;
2539 2540 2541 2542 2543
}

struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
				      unsigned int id)
{
2544
	struct kvm_vcpu *vcpu;
2545
	struct sie_page *sie_page;
2546 2547
	int rc = -EINVAL;

2548
	if (!kvm_is_ucontrol(kvm) && !sca_can_add_vcpu(kvm, id))
2549 2550 2551
		goto out;

	rc = -ENOMEM;
2552

2553
	vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
2554
	if (!vcpu)
2555
		goto out;
2556

2557
	BUILD_BUG_ON(sizeof(struct sie_page) != 4096);
2558 2559
	sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
	if (!sie_page)
2560 2561
		goto out_free_cpu;

2562 2563 2564
	vcpu->arch.sie_block = &sie_page->sie_block;
	vcpu->arch.sie_block->itdba = (unsigned long) &sie_page->itdb;

2565 2566 2567 2568
	/* the real guest size will always be smaller than msl */
	vcpu->arch.sie_block->mso = 0;
	vcpu->arch.sie_block->msl = sclp.hamax;

2569
	vcpu->arch.sie_block->icpua = id;
2570
	spin_lock_init(&vcpu->arch.local_int.lock);
2571
	vcpu->arch.sie_block->gd = (u32)(u64)kvm->arch.gisa;
2572 2573
	if (vcpu->arch.sie_block->gd && sclp.has_gisaf)
		vcpu->arch.sie_block->gd |= GISA_FORMAT1;
2574
	seqcount_init(&vcpu->arch.cputm_seqcount);
2575

2576 2577
	rc = kvm_vcpu_init(vcpu, kvm, id);
	if (rc)
2578
		goto out_free_sie_block;
2579
	VM_EVENT(kvm, 3, "create cpu %d at 0x%pK, sie block at 0x%pK", id, vcpu,
2580
		 vcpu->arch.sie_block);
2581
	trace_kvm_s390_create_vcpu(id, vcpu, vcpu->arch.sie_block);
2582 2583

	return vcpu;
2584 2585
out_free_sie_block:
	free_page((unsigned long)(vcpu->arch.sie_block));
2586
out_free_cpu:
2587
	kmem_cache_free(kvm_vcpu_cache, vcpu);
2588
out:
2589 2590 2591 2592 2593
	return ERR_PTR(rc);
}

int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
{
2594
	return kvm_s390_vcpu_has_irq(vcpu, 0);
2595 2596
}

2597 2598
bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
{
2599
	return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE);
2600 2601
}

2602
void kvm_s390_vcpu_block(struct kvm_vcpu *vcpu)
2603
{
2604
	atomic_or(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
2605
	exit_sie(vcpu);
2606 2607
}

2608
void kvm_s390_vcpu_unblock(struct kvm_vcpu *vcpu)
2609
{
2610
	atomic_andnot(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
2611 2612
}

2613 2614
static void kvm_s390_vcpu_request(struct kvm_vcpu *vcpu)
{
2615
	atomic_or(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
2616
	exit_sie(vcpu);
2617 2618 2619 2620
}

static void kvm_s390_vcpu_request_handled(struct kvm_vcpu *vcpu)
{
2621
	atomic_andnot(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
2622 2623
}

2624 2625 2626 2627 2628 2629
/*
 * 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)
{
2630
	kvm_s390_set_cpuflags(vcpu, CPUSTAT_STOP_INT);
2631 2632 2633 2634
	while (vcpu->arch.sie_block->prog0c & PROG_IN_SIE)
		cpu_relax();
}

2635 2636
/* Kick a guest cpu out of SIE to process a request synchronously */
void kvm_s390_sync_request(int req, struct kvm_vcpu *vcpu)
2637
{
2638 2639
	kvm_make_request(req, vcpu);
	kvm_s390_vcpu_request(vcpu);
2640 2641
}

2642 2643
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start,
			      unsigned long end)
2644 2645 2646
{
	struct kvm *kvm = gmap->private;
	struct kvm_vcpu *vcpu;
2647 2648
	unsigned long prefix;
	int i;
2649

2650 2651
	if (gmap_is_shadow(gmap))
		return;
2652 2653 2654
	if (start >= 1UL << 31)
		/* We are only interested in prefix pages */
		return;
2655 2656
	kvm_for_each_vcpu(i, vcpu, kvm) {
		/* match against both prefix pages */
2657 2658 2659 2660
		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);
2661
			kvm_s390_sync_request(KVM_REQ_MMU_RELOAD, vcpu);
2662 2663 2664 2665
		}
	}
}

2666 2667 2668 2669 2670 2671 2672
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
	/* kvm common code refers to this, but never calls it */
	BUG();
	return 0;
}

2673 2674 2675 2676 2677 2678
static int kvm_arch_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu,
					   struct kvm_one_reg *reg)
{
	int r = -EINVAL;

	switch (reg->id) {
2679 2680 2681 2682 2683 2684 2685 2686
	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;
2687
	case KVM_REG_S390_CPU_TIMER:
2688
		r = put_user(kvm_s390_get_cpu_timer(vcpu),
2689 2690 2691 2692 2693 2694
			     (u64 __user *)reg->addr);
		break;
	case KVM_REG_S390_CLOCK_COMP:
		r = put_user(vcpu->arch.sie_block->ckc,
			     (u64 __user *)reg->addr);
		break;
2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706
	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;
2707 2708 2709 2710
	case KVM_REG_S390_PP:
		r = put_user(vcpu->arch.sie_block->pp,
			     (u64 __user *)reg->addr);
		break;
2711 2712 2713 2714
	case KVM_REG_S390_GBEA:
		r = put_user(vcpu->arch.sie_block->gbea,
			     (u64 __user *)reg->addr);
		break;
2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725
	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;
2726
	__u64 val;
2727 2728

	switch (reg->id) {
2729 2730 2731 2732 2733 2734 2735 2736
	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;
2737
	case KVM_REG_S390_CPU_TIMER:
2738 2739 2740
		r = get_user(val, (u64 __user *)reg->addr);
		if (!r)
			kvm_s390_set_cpu_timer(vcpu, val);
2741 2742 2743 2744 2745
		break;
	case KVM_REG_S390_CLOCK_COMP:
		r = get_user(vcpu->arch.sie_block->ckc,
			     (u64 __user *)reg->addr);
		break;
2746 2747 2748
	case KVM_REG_S390_PFTOKEN:
		r = get_user(vcpu->arch.pfault_token,
			     (u64 __user *)reg->addr);
2749 2750
		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
			kvm_clear_async_pf_completion_queue(vcpu);
2751 2752 2753 2754 2755 2756 2757 2758 2759
		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;
2760 2761 2762 2763
	case KVM_REG_S390_PP:
		r = get_user(vcpu->arch.sie_block->pp,
			     (u64 __user *)reg->addr);
		break;
2764 2765 2766 2767
	case KVM_REG_S390_GBEA:
		r = get_user(vcpu->arch.sie_block->gbea,
			     (u64 __user *)reg->addr);
		break;
2768 2769 2770 2771 2772 2773
	default:
		break;
	}

	return r;
}
2774

2775 2776 2777 2778 2779 2780 2781 2782
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)
{
2783
	vcpu_load(vcpu);
2784
	memcpy(&vcpu->run->s.regs.gprs, &regs->gprs, sizeof(regs->gprs));
2785
	vcpu_put(vcpu);
2786 2787 2788 2789 2790
	return 0;
}

int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
2791
	vcpu_load(vcpu);
2792
	memcpy(&regs->gprs, &vcpu->run->s.regs.gprs, sizeof(regs->gprs));
2793
	vcpu_put(vcpu);
2794 2795 2796 2797 2798 2799
	return 0;
}

int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
2800 2801
	vcpu_load(vcpu);

2802
	memcpy(&vcpu->run->s.regs.acrs, &sregs->acrs, sizeof(sregs->acrs));
2803
	memcpy(&vcpu->arch.sie_block->gcr, &sregs->crs, sizeof(sregs->crs));
2804 2805

	vcpu_put(vcpu);
2806 2807 2808 2809 2810 2811
	return 0;
}

int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
2812 2813
	vcpu_load(vcpu);

2814
	memcpy(&sregs->acrs, &vcpu->run->s.regs.acrs, sizeof(sregs->acrs));
2815
	memcpy(&sregs->crs, &vcpu->arch.sie_block->gcr, sizeof(sregs->crs));
2816 2817

	vcpu_put(vcpu);
2818 2819 2820 2821 2822
	return 0;
}

int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
2823 2824 2825 2826 2827 2828 2829 2830
	int ret = 0;

	vcpu_load(vcpu);

	if (test_fp_ctl(fpu->fpc)) {
		ret = -EINVAL;
		goto out;
	}
2831
	vcpu->run->s.regs.fpc = fpu->fpc;
2832
	if (MACHINE_HAS_VX)
2833 2834
		convert_fp_to_vx((__vector128 *) vcpu->run->s.regs.vrs,
				 (freg_t *) fpu->fprs);
2835
	else
2836
		memcpy(vcpu->run->s.regs.fprs, &fpu->fprs, sizeof(fpu->fprs));
2837 2838 2839 2840

out:
	vcpu_put(vcpu);
	return ret;
2841 2842 2843 2844
}

int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
2845 2846
	vcpu_load(vcpu);

2847 2848 2849
	/* make sure we have the latest values */
	save_fpu_regs();
	if (MACHINE_HAS_VX)
2850 2851
		convert_vx_to_fp((freg_t *) fpu->fprs,
				 (__vector128 *) vcpu->run->s.regs.vrs);
2852
	else
2853
		memcpy(fpu->fprs, vcpu->run->s.regs.fprs, sizeof(fpu->fprs));
2854
	fpu->fpc = vcpu->run->s.regs.fpc;
2855 2856

	vcpu_put(vcpu);
2857 2858 2859 2860 2861 2862 2863
	return 0;
}

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

2864
	if (!is_vcpu_stopped(vcpu))
2865
		rc = -EBUSY;
2866 2867 2868 2869
	else {
		vcpu->run->psw_mask = psw.mask;
		vcpu->run->psw_addr = psw.addr;
	}
2870 2871 2872 2873 2874 2875 2876 2877 2878
	return rc;
}

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

2879 2880 2881 2882
#define VALID_GUESTDBG_FLAGS (KVM_GUESTDBG_SINGLESTEP | \
			      KVM_GUESTDBG_USE_HW_BP | \
			      KVM_GUESTDBG_ENABLE)

J
Jan Kiszka 已提交
2883 2884
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
					struct kvm_guest_debug *dbg)
2885
{
2886 2887
	int rc = 0;

2888 2889
	vcpu_load(vcpu);

2890 2891 2892
	vcpu->guest_debug = 0;
	kvm_s390_clear_bp_data(vcpu);

2893 2894 2895 2896 2897 2898 2899 2900
	if (dbg->control & ~VALID_GUESTDBG_FLAGS) {
		rc = -EINVAL;
		goto out;
	}
	if (!sclp.has_gpere) {
		rc = -EINVAL;
		goto out;
	}
2901 2902 2903 2904

	if (dbg->control & KVM_GUESTDBG_ENABLE) {
		vcpu->guest_debug = dbg->control;
		/* enforce guest PER */
2905
		kvm_s390_set_cpuflags(vcpu, CPUSTAT_P);
2906 2907 2908 2909

		if (dbg->control & KVM_GUESTDBG_USE_HW_BP)
			rc = kvm_s390_import_bp_data(vcpu, dbg);
	} else {
2910
		kvm_s390_clear_cpuflags(vcpu, CPUSTAT_P);
2911 2912 2913 2914 2915 2916
		vcpu->arch.guestdbg.last_bp = 0;
	}

	if (rc) {
		vcpu->guest_debug = 0;
		kvm_s390_clear_bp_data(vcpu);
2917
		kvm_s390_clear_cpuflags(vcpu, CPUSTAT_P);
2918 2919
	}

2920 2921
out:
	vcpu_put(vcpu);
2922
	return rc;
2923 2924
}

2925 2926 2927
int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
2928 2929 2930 2931
	int ret;

	vcpu_load(vcpu);

2932
	/* CHECK_STOP and LOAD are not supported yet */
2933 2934 2935 2936 2937
	ret = is_vcpu_stopped(vcpu) ? KVM_MP_STATE_STOPPED :
				      KVM_MP_STATE_OPERATING;

	vcpu_put(vcpu);
	return ret;
2938 2939 2940 2941 2942
}

int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
2943 2944
	int rc = 0;

2945 2946
	vcpu_load(vcpu);

2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963
	/* 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;
	}

2964
	vcpu_put(vcpu);
2965
	return rc;
2966 2967
}

2968 2969
static bool ibs_enabled(struct kvm_vcpu *vcpu)
{
2970
	return kvm_s390_test_cpuflags(vcpu, CPUSTAT_IBS);
2971 2972
}

2973 2974
static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu)
{
2975
retry:
2976
	kvm_s390_vcpu_request_handled(vcpu);
R
Radim Krčmář 已提交
2977
	if (!kvm_request_pending(vcpu))
2978
		return 0;
2979 2980
	/*
	 * We use MMU_RELOAD just to re-arm the ipte notifier for the
2981
	 * guest prefix page. gmap_mprotect_notify will wait on the ptl lock.
2982 2983 2984 2985
	 * 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.
	 */
2986
	if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) {
2987
		int rc;
2988 2989 2990
		rc = gmap_mprotect_notify(vcpu->arch.gmap,
					  kvm_s390_get_prefix(vcpu),
					  PAGE_SIZE * 2, PROT_WRITE);
2991 2992
		if (rc) {
			kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu);
2993
			return rc;
2994
		}
2995
		goto retry;
2996
	}
2997

2998 2999 3000 3001 3002
	if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) {
		vcpu->arch.sie_block->ihcpu = 0xffff;
		goto retry;
	}

3003 3004 3005
	if (kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu)) {
		if (!ibs_enabled(vcpu)) {
			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 1);
3006
			kvm_s390_set_cpuflags(vcpu, CPUSTAT_IBS);
3007 3008
		}
		goto retry;
3009
	}
3010 3011 3012 3013

	if (kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu)) {
		if (ibs_enabled(vcpu)) {
			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 0);
3014
			kvm_s390_clear_cpuflags(vcpu, CPUSTAT_IBS);
3015 3016 3017 3018
		}
		goto retry;
	}

3019 3020 3021 3022 3023
	if (kvm_check_request(KVM_REQ_ICPT_OPEREXC, vcpu)) {
		vcpu->arch.sie_block->ictl |= ICTL_OPEREXC;
		goto retry;
	}

3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044
	if (kvm_check_request(KVM_REQ_START_MIGRATION, vcpu)) {
		/*
		 * Disable CMMA virtualization; we will emulate the ESSA
		 * instruction manually, in order to provide additional
		 * functionalities needed for live migration.
		 */
		vcpu->arch.sie_block->ecb2 &= ~ECB2_CMMA;
		goto retry;
	}

	if (kvm_check_request(KVM_REQ_STOP_MIGRATION, vcpu)) {
		/*
		 * Re-enable CMMA virtualization if CMMA is available and
		 * was used.
		 */
		if ((vcpu->kvm->arch.use_cmma) &&
		    (vcpu->kvm->mm->context.use_cmma))
			vcpu->arch.sie_block->ecb2 |= ECB2_CMMA;
		goto retry;
	}

3045
	/* nothing to do, just clear the request */
3046
	kvm_clear_request(KVM_REQ_UNHALT, vcpu);
3047

3048 3049 3050
	return 0;
}

3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079
void kvm_s390_set_tod_clock_ext(struct kvm *kvm,
				 const struct kvm_s390_vm_tod_clock *gtod)
{
	struct kvm_vcpu *vcpu;
	struct kvm_s390_tod_clock_ext htod;
	int i;

	mutex_lock(&kvm->lock);
	preempt_disable();

	get_tod_clock_ext((char *)&htod);

	kvm->arch.epoch = gtod->tod - htod.tod;
	kvm->arch.epdx = gtod->epoch_idx - htod.epoch_idx;

	if (kvm->arch.epoch > gtod->tod)
		kvm->arch.epdx -= 1;

	kvm_s390_vcpu_block_all(kvm);
	kvm_for_each_vcpu(i, vcpu, kvm) {
		vcpu->arch.sie_block->epoch = kvm->arch.epoch;
		vcpu->arch.sie_block->epdx  = kvm->arch.epdx;
	}

	kvm_s390_vcpu_unblock_all(kvm);
	preempt_enable();
	mutex_unlock(&kvm->lock);
}

3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095
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);
}

3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106
/**
 * 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)
3107
{
3108 3109
	return gmap_fault(vcpu->arch.gmap, gpa,
			  writable ? FAULT_FLAG_WRITE : 0);
3110 3111
}

3112 3113 3114 3115
static void __kvm_inject_pfault_token(struct kvm_vcpu *vcpu, bool start_token,
				      unsigned long token)
{
	struct kvm_s390_interrupt inti;
3116
	struct kvm_s390_irq irq;
3117 3118

	if (start_token) {
3119 3120 3121
		irq.u.ext.ext_params2 = token;
		irq.type = KVM_S390_INT_PFAULT_INIT;
		WARN_ON_ONCE(kvm_s390_inject_vcpu(vcpu, &irq));
3122 3123
	} else {
		inti.type = KVM_S390_INT_PFAULT_DONE;
3124
		inti.parm64 = token;
3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170
		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;
3171
	if (kvm_s390_vcpu_has_irq(vcpu, 0))
3172 3173 3174 3175 3176 3177
		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
3178 3179 3180
	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))
3181 3182 3183 3184 3185 3186
		return 0;

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

3187
static int vcpu_pre_run(struct kvm_vcpu *vcpu)
3188
{
3189
	int rc, cpuflags;
3190

3191 3192 3193 3194 3195 3196 3197
	/*
	 * 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);

3198 3199
	vcpu->arch.sie_block->gg14 = vcpu->run->s.regs.gprs[14];
	vcpu->arch.sie_block->gg15 = vcpu->run->s.regs.gprs[15];
3200 3201 3202 3203

	if (need_resched())
		schedule();

3204
	if (test_cpu_flag(CIF_MCCK_PENDING))
3205 3206
		s390_handle_mcck();

3207 3208 3209 3210 3211
	if (!kvm_is_ucontrol(vcpu->kvm)) {
		rc = kvm_s390_deliver_pending_interrupts(vcpu);
		if (rc)
			return rc;
	}
C
Carsten Otte 已提交
3212

3213 3214 3215 3216
	rc = kvm_s390_handle_requests(vcpu);
	if (rc)
		return rc;

3217 3218 3219 3220 3221
	if (guestdbg_enabled(vcpu)) {
		kvm_s390_backup_guest_per_regs(vcpu);
		kvm_s390_patch_guest_per_regs(vcpu);
	}

3222
	vcpu->arch.sie_block->icptcode = 0;
3223 3224 3225
	cpuflags = atomic_read(&vcpu->arch.sie_block->cpuflags);
	VCPU_EVENT(vcpu, 6, "entering sie flags %x", cpuflags);
	trace_kvm_s390_sie_enter(vcpu, cpuflags);
3226

3227 3228 3229
	return 0;
}

3230 3231
static int vcpu_post_run_fault_in_sie(struct kvm_vcpu *vcpu)
{
3232 3233 3234 3235
	struct kvm_s390_pgm_info pgm_info = {
		.code = PGM_ADDRESSING,
	};
	u8 opcode, ilen;
3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248
	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.
	 */
3249
	rc = read_guest_instr(vcpu, vcpu->arch.sie_block->gpsw.addr, &opcode, 1);
3250
	ilen = insn_length(opcode);
3251 3252 3253 3254 3255 3256 3257 3258 3259 3260
	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;
	}
3261 3262 3263
	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);
3264 3265
}

3266 3267
static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason)
{
3268 3269 3270
	struct mcck_volatile_info *mcck_info;
	struct sie_page *sie_page;

3271 3272 3273 3274
	VCPU_EVENT(vcpu, 6, "exit sie icptcode %d",
		   vcpu->arch.sie_block->icptcode);
	trace_kvm_s390_sie_exit(vcpu, vcpu->arch.sie_block->icptcode);

3275 3276 3277
	if (guestdbg_enabled(vcpu))
		kvm_s390_restore_guest_per_regs(vcpu);

3278 3279
	vcpu->run->s.regs.gprs[14] = vcpu->arch.sie_block->gg14;
	vcpu->run->s.regs.gprs[15] = vcpu->arch.sie_block->gg15;
3280

3281 3282 3283 3284 3285 3286 3287 3288 3289
	if (exit_reason == -EINTR) {
		VCPU_EVENT(vcpu, 3, "%s", "machine check");
		sie_page = container_of(vcpu->arch.sie_block,
					struct sie_page, sie_block);
		mcck_info = &sie_page->mcck_info;
		kvm_s390_reinject_machine_check(vcpu, mcck_info);
		return 0;
	}

3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302
	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;
3303 3304 3305 3306 3307
	} 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;
3308
		return -EREMOTE;
3309
	} else if (current->thread.gmap_pfault) {
3310
		trace_kvm_s390_major_guest_pfault(vcpu);
3311
		current->thread.gmap_pfault = 0;
3312 3313 3314
		if (kvm_arch_setup_async_pf(vcpu))
			return 0;
		return kvm_arch_fault_in_page(vcpu, current->thread.gmap_addr, 1);
3315
	}
3316
	return vcpu_post_run_fault_in_sie(vcpu);
3317 3318 3319 3320 3321 3322
}

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

3323 3324 3325 3326 3327 3328
	/*
	 * 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);

3329 3330 3331 3332
	do {
		rc = vcpu_pre_run(vcpu);
		if (rc)
			break;
3333

3334
		srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
3335 3336 3337 3338
		/*
		 * As PF_VCPU will be used in fault handler, between
		 * guest_enter and guest_exit should be no uaccess.
		 */
3339
		local_irq_disable();
3340
		guest_enter_irqoff();
3341
		__disable_cpu_timer_accounting(vcpu);
3342
		local_irq_enable();
3343 3344
		exit_reason = sie64a(vcpu->arch.sie_block,
				     vcpu->run->s.regs.gprs);
3345
		local_irq_disable();
3346
		__enable_cpu_timer_accounting(vcpu);
3347
		guest_exit_irqoff();
3348
		local_irq_enable();
3349
		vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
3350 3351

		rc = vcpu_post_run(vcpu, exit_reason);
3352
	} while (!signal_pending(current) && !guestdbg_exit_pending(vcpu) && !rc);
3353

3354
	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
3355
	return rc;
3356 3357
}

3358 3359
static void sync_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
3360
	struct runtime_instr_cb *riccb;
F
Fan Zhang 已提交
3361
	struct gs_cb *gscb;
3362 3363

	riccb = (struct runtime_instr_cb *) &kvm_run->s.regs.riccb;
F
Fan Zhang 已提交
3364
	gscb = (struct gs_cb *) &kvm_run->s.regs.gscb;
3365 3366 3367 3368 3369 3370
	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);
3371 3372
		/* some control register changes require a tlb flush */
		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
3373 3374
	}
	if (kvm_run->kvm_dirty_regs & KVM_SYNC_ARCH0) {
3375
		kvm_s390_set_cpu_timer(vcpu, kvm_run->s.regs.cputm);
3376 3377 3378 3379 3380 3381 3382 3383 3384
		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;
3385 3386
		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
			kvm_clear_async_pf_completion_queue(vcpu);
3387
	}
F
Fan Zhang 已提交
3388 3389 3390 3391 3392
	/*
	 * 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) &&
3393
	    test_kvm_facility(vcpu->kvm, 64) &&
3394
	    riccb->v &&
3395
	    !(vcpu->arch.sie_block->ecb3 & ECB3_RI)) {
3396
		VCPU_EVENT(vcpu, 3, "%s", "ENABLE: RI (sync_regs)");
3397
		vcpu->arch.sie_block->ecb3 |= ECB3_RI;
F
Fan Zhang 已提交
3398
	}
F
Fan Zhang 已提交
3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410
	/*
	 * If userspace sets the gscb (e.g. after migration) to non-zero,
	 * we should enable GS here instead of doing the lazy enablement.
	 */
	if ((kvm_run->kvm_dirty_regs & KVM_SYNC_GSCB) &&
	    test_kvm_facility(vcpu->kvm, 133) &&
	    gscb->gssm &&
	    !vcpu->arch.gs_enabled) {
		VCPU_EVENT(vcpu, 3, "%s", "ENABLE: GS (sync_regs)");
		vcpu->arch.sie_block->ecb |= ECB_GS;
		vcpu->arch.sie_block->ecd |= ECD_HOSTREGMGMT;
		vcpu->arch.gs_enabled = 1;
F
Fan Zhang 已提交
3411
	}
3412 3413 3414 3415 3416
	if ((kvm_run->kvm_dirty_regs & KVM_SYNC_BPBC) &&
	    test_kvm_facility(vcpu->kvm, 82)) {
		vcpu->arch.sie_block->fpf &= ~FPF_BPBC;
		vcpu->arch.sie_block->fpf |= kvm_run->s.regs.bpbc ? FPF_BPBC : 0;
	}
3417 3418
	save_access_regs(vcpu->arch.host_acrs);
	restore_access_regs(vcpu->run->s.regs.acrs);
3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430
	/* save host (userspace) fprs/vrs */
	save_fpu_regs();
	vcpu->arch.host_fpregs.fpc = current->thread.fpu.fpc;
	vcpu->arch.host_fpregs.regs = current->thread.fpu.regs;
	if (MACHINE_HAS_VX)
		current->thread.fpu.regs = vcpu->run->s.regs.vrs;
	else
		current->thread.fpu.regs = vcpu->run->s.regs.fprs;
	current->thread.fpu.fpc = vcpu->run->s.regs.fpc;
	if (test_fp_ctl(current->thread.fpu.fpc))
		/* User space provided an invalid FPC, let's clear it */
		current->thread.fpu.fpc = 0;
F
Fan Zhang 已提交
3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444
	if (MACHINE_HAS_GS) {
		preempt_disable();
		__ctl_set_bit(2, 4);
		if (current->thread.gs_cb) {
			vcpu->arch.host_gscb = current->thread.gs_cb;
			save_gs_cb(vcpu->arch.host_gscb);
		}
		if (vcpu->arch.gs_enabled) {
			current->thread.gs_cb = (struct gs_cb *)
						&vcpu->run->s.regs.gscb;
			restore_gs_cb(current->thread.gs_cb);
		}
		preempt_enable();
	}
F
Fan Zhang 已提交
3445

3446 3447 3448 3449 3450 3451 3452 3453 3454
	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);
3455
	kvm_run->s.regs.cputm = kvm_s390_get_cpu_timer(vcpu);
3456 3457 3458 3459 3460 3461 3462
	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;
3463
	kvm_run->s.regs.bpbc = (vcpu->arch.sie_block->fpf & FPF_BPBC) == FPF_BPBC;
3464 3465
	save_access_regs(vcpu->run->s.regs.acrs);
	restore_access_regs(vcpu->arch.host_acrs);
3466 3467 3468 3469 3470 3471
	/* Save guest register state */
	save_fpu_regs();
	vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
	/* Restore will be done lazily at return */
	current->thread.fpu.fpc = vcpu->arch.host_fpregs.fpc;
	current->thread.fpu.regs = vcpu->arch.host_fpregs.regs;
F
Fan Zhang 已提交
3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483
	if (MACHINE_HAS_GS) {
		__ctl_set_bit(2, 4);
		if (vcpu->arch.gs_enabled)
			save_gs_cb(current->thread.gs_cb);
		preempt_disable();
		current->thread.gs_cb = vcpu->arch.host_gscb;
		restore_gs_cb(vcpu->arch.host_gscb);
		preempt_enable();
		if (!vcpu->arch.host_gscb)
			__ctl_clear_bit(2, 4);
		vcpu->arch.host_gscb = NULL;
	}
3484

3485 3486
}

3487 3488
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
3489
	int rc;
3490

3491 3492 3493
	if (kvm_run->immediate_exit)
		return -EINTR;

3494 3495
	vcpu_load(vcpu);

3496 3497
	if (guestdbg_exit_pending(vcpu)) {
		kvm_s390_prepare_debug_exit(vcpu);
3498 3499
		rc = 0;
		goto out;
3500 3501
	}

3502
	kvm_sigset_activate(vcpu);
3503

3504 3505 3506
	if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) {
		kvm_s390_vcpu_start(vcpu);
	} else if (is_vcpu_stopped(vcpu)) {
3507
		pr_err_ratelimited("can't run stopped vcpu %d\n",
3508
				   vcpu->vcpu_id);
3509 3510
		rc = -EINVAL;
		goto out;
3511
	}
3512

3513
	sync_regs(vcpu, kvm_run);
3514
	enable_cpu_timer_accounting(vcpu);
3515

3516
	might_fault();
3517
	rc = __vcpu_run(vcpu);
3518

3519 3520
	if (signal_pending(current) && !rc) {
		kvm_run->exit_reason = KVM_EXIT_INTR;
3521
		rc = -EINTR;
3522
	}
3523

3524 3525 3526 3527 3528
	if (guestdbg_exit_pending(vcpu) && !rc)  {
		kvm_s390_prepare_debug_exit(vcpu);
		rc = 0;
	}

3529
	if (rc == -EREMOTE) {
3530
		/* userspace support is needed, kvm_run has been prepared */
3531 3532
		rc = 0;
	}
3533

3534
	disable_cpu_timer_accounting(vcpu);
3535
	store_regs(vcpu, kvm_run);
3536

3537
	kvm_sigset_deactivate(vcpu);
3538 3539

	vcpu->stat.exit_userspace++;
3540 3541
out:
	vcpu_put(vcpu);
3542
	return rc;
3543 3544 3545 3546 3547 3548 3549 3550
}

/*
 * 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
 */
3551
int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long gpa)
3552
{
3553
	unsigned char archmode = 1;
3554
	freg_t fprs[NUM_FPRS];
3555
	unsigned int px;
3556
	u64 clkcomp, cputm;
3557
	int rc;
3558

3559
	px = kvm_s390_get_prefix(vcpu);
3560 3561
	if (gpa == KVM_S390_STORE_STATUS_NOADDR) {
		if (write_guest_abs(vcpu, 163, &archmode, 1))
3562
			return -EFAULT;
3563
		gpa = 0;
3564 3565
	} else if (gpa == KVM_S390_STORE_STATUS_PREFIXED) {
		if (write_guest_real(vcpu, 163, &archmode, 1))
3566
			return -EFAULT;
3567 3568 3569
		gpa = px;
	} else
		gpa -= __LC_FPREGS_SAVE_AREA;
3570 3571 3572

	/* manually convert vector registers if necessary */
	if (MACHINE_HAS_VX) {
3573
		convert_vx_to_fp(fprs, (__vector128 *) vcpu->run->s.regs.vrs);
3574 3575 3576 3577
		rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
				     fprs, 128);
	} else {
		rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
3578
				     vcpu->run->s.regs.fprs, 128);
3579
	}
3580
	rc |= write_guest_abs(vcpu, gpa + __LC_GPREGS_SAVE_AREA,
3581
			      vcpu->run->s.regs.gprs, 128);
3582
	rc |= write_guest_abs(vcpu, gpa + __LC_PSW_SAVE_AREA,
3583
			      &vcpu->arch.sie_block->gpsw, 16);
3584
	rc |= write_guest_abs(vcpu, gpa + __LC_PREFIX_SAVE_AREA,
3585
			      &px, 4);
3586
	rc |= write_guest_abs(vcpu, gpa + __LC_FP_CREG_SAVE_AREA,
3587
			      &vcpu->run->s.regs.fpc, 4);
3588
	rc |= write_guest_abs(vcpu, gpa + __LC_TOD_PROGREG_SAVE_AREA,
3589
			      &vcpu->arch.sie_block->todpr, 4);
3590
	cputm = kvm_s390_get_cpu_timer(vcpu);
3591
	rc |= write_guest_abs(vcpu, gpa + __LC_CPU_TIMER_SAVE_AREA,
3592
			      &cputm, 8);
3593
	clkcomp = vcpu->arch.sie_block->ckc >> 8;
3594
	rc |= write_guest_abs(vcpu, gpa + __LC_CLOCK_COMP_SAVE_AREA,
3595
			      &clkcomp, 8);
3596
	rc |= write_guest_abs(vcpu, gpa + __LC_AREGS_SAVE_AREA,
3597
			      &vcpu->run->s.regs.acrs, 64);
3598
	rc |= write_guest_abs(vcpu, gpa + __LC_CREGS_SAVE_AREA,
3599 3600
			      &vcpu->arch.sie_block->gcr, 128);
	return rc ? -EFAULT : 0;
3601 3602
}

3603 3604 3605 3606
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
3607
	 * switch in the run ioctl. Let's update our copies before we save
3608 3609
	 * it into the save area
	 */
3610
	save_fpu_regs();
3611
	vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
3612 3613 3614 3615 3616
	save_access_regs(vcpu->run->s.regs.acrs);

	return kvm_s390_store_status_unloaded(vcpu, addr);
}

3617 3618 3619
static void __disable_ibs_on_vcpu(struct kvm_vcpu *vcpu)
{
	kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu);
3620
	kvm_s390_sync_request(KVM_REQ_DISABLE_IBS, vcpu);
3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634
}

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)
{
3635 3636
	if (!sclp.has_ibs)
		return;
3637
	kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu);
3638
	kvm_s390_sync_request(KVM_REQ_ENABLE_IBS, vcpu);
3639 3640
}

3641 3642
void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu)
{
3643 3644 3645 3646 3647
	int i, online_vcpus, started_vcpus = 0;

	if (!is_vcpu_stopped(vcpu))
		return;

3648
	trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 1);
3649
	/* Only one cpu at a time may enter/leave the STOPPED state. */
3650
	spin_lock(&vcpu->kvm->arch.start_stop_lock);
3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669
	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);
	}

3670
	kvm_s390_clear_cpuflags(vcpu, CPUSTAT_STOPPED);
3671 3672 3673 3674
	/*
	 * Another VCPU might have used IBS while we were offline.
	 * Let's play safe and flush the VCPU at startup.
	 */
3675
	kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
3676
	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
3677
	return;
3678 3679 3680 3681
}

void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu)
{
3682 3683 3684 3685 3686 3687
	int i, online_vcpus, started_vcpus = 0;
	struct kvm_vcpu *started_vcpu = NULL;

	if (is_vcpu_stopped(vcpu))
		return;

3688
	trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 0);
3689
	/* Only one cpu at a time may enter/leave the STOPPED state. */
3690
	spin_lock(&vcpu->kvm->arch.start_stop_lock);
3691 3692
	online_vcpus = atomic_read(&vcpu->kvm->online_vcpus);

3693
	/* SIGP STOP and SIGP STOP AND STORE STATUS has been fully processed */
3694
	kvm_s390_clear_stop_irq(vcpu);
3695

3696
	kvm_s390_set_cpuflags(vcpu, CPUSTAT_STOPPED);
3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713
	__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);
	}

3714
	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
3715
	return;
3716 3717
}

3718 3719 3720 3721 3722 3723 3724 3725 3726
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) {
3727 3728 3729
	case KVM_CAP_S390_CSS_SUPPORT:
		if (!vcpu->kvm->arch.css_support) {
			vcpu->kvm->arch.css_support = 1;
3730
			VM_EVENT(vcpu->kvm, 3, "%s", "ENABLE: CSS support");
3731 3732 3733 3734
			trace_kvm_s390_enable_css(vcpu->kvm);
		}
		r = 0;
		break;
3735 3736 3737 3738 3739 3740 3741
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767
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) {
3768 3769
			r = check_gva_range(vcpu, mop->gaddr, mop->ar,
					    mop->size, GACC_FETCH);
3770 3771 3772 3773 3774 3775 3776 3777 3778 3779
			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) {
3780 3781
			r = check_gva_range(vcpu, mop->gaddr, mop->ar,
					    mop->size, GACC_STORE);
3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802
			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;
}

3803 3804
long kvm_arch_vcpu_async_ioctl(struct file *filp,
			       unsigned int ioctl, unsigned long arg)
3805 3806 3807 3808
{
	struct kvm_vcpu *vcpu = filp->private_data;
	void __user *argp = (void __user *)arg;

3809
	switch (ioctl) {
3810 3811 3812 3813
	case KVM_S390_IRQ: {
		struct kvm_s390_irq s390irq;

		if (copy_from_user(&s390irq, argp, sizeof(s390irq)))
3814 3815
			return -EFAULT;
		return kvm_s390_inject_vcpu(vcpu, &s390irq);
3816
	}
3817
	case KVM_S390_INTERRUPT: {
3818
		struct kvm_s390_interrupt s390int;
3819
		struct kvm_s390_irq s390irq;
3820 3821

		if (copy_from_user(&s390int, argp, sizeof(s390int)))
3822
			return -EFAULT;
3823 3824
		if (s390int_to_s390irq(&s390int, &s390irq))
			return -EINVAL;
3825
		return kvm_s390_inject_vcpu(vcpu, &s390irq);
3826
	}
3827
	}
3828 3829 3830 3831 3832 3833 3834 3835 3836 3837
	return -ENOIOCTLCMD;
}

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;
	int idx;
	long r;
3838 3839 3840 3841

	vcpu_load(vcpu);

	switch (ioctl) {
3842
	case KVM_S390_STORE_STATUS:
3843
		idx = srcu_read_lock(&vcpu->kvm->srcu);
3844
		r = kvm_s390_vcpu_store_status(vcpu, arg);
3845
		srcu_read_unlock(&vcpu->kvm->srcu, idx);
3846
		break;
3847 3848 3849
	case KVM_S390_SET_INITIAL_PSW: {
		psw_t psw;

3850
		r = -EFAULT;
3851
		if (copy_from_user(&psw, argp, sizeof(psw)))
3852 3853 3854
			break;
		r = kvm_arch_vcpu_ioctl_set_initial_psw(vcpu, psw);
		break;
3855 3856
	}
	case KVM_S390_INITIAL_RESET:
3857 3858
		r = kvm_arch_vcpu_ioctl_initial_reset(vcpu);
		break;
3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870
	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;
	}
3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906
#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
3907
	case KVM_S390_VCPU_FAULT: {
3908
		r = gmap_fault(vcpu->arch.gmap, arg, 0);
3909 3910
		break;
	}
3911 3912 3913 3914 3915 3916 3917 3918 3919
	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;
	}
3920 3921 3922 3923 3924 3925 3926 3927 3928
	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;
	}
3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940
	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;
		}
3941
		/* do not use irq_state.flags, it will break old QEMUs */
3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956
		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;
		}
3957
		/* do not use irq_state.flags, it will break old QEMUs */
3958 3959 3960 3961 3962
		r = kvm_s390_get_irq_state(vcpu,
					   (__u8 __user *)  irq_state.buf,
					   irq_state.len);
		break;
	}
3963
	default:
3964
		r = -ENOTTY;
3965
	}
3966 3967

	vcpu_put(vcpu);
3968
	return r;
3969 3970
}

3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983
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;
}

3984 3985
int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
			    unsigned long npages)
3986 3987 3988 3989
{
	return 0;
}

3990
/* Section: memory related */
3991 3992
int kvm_arch_prepare_memory_region(struct kvm *kvm,
				   struct kvm_memory_slot *memslot,
3993
				   const struct kvm_userspace_memory_region *mem,
3994
				   enum kvm_mr_change change)
3995
{
3996 3997 3998 3999
	/* 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 */
4000

4001
	if (mem->userspace_addr & 0xffffful)
4002 4003
		return -EINVAL;

4004
	if (mem->memory_size & 0xffffful)
4005 4006
		return -EINVAL;

4007 4008 4009
	if (mem->guest_phys_addr + mem->memory_size > kvm->arch.mem_limit)
		return -EINVAL;

4010 4011 4012 4013
	return 0;
}

void kvm_arch_commit_memory_region(struct kvm *kvm,
4014
				const struct kvm_userspace_memory_region *mem,
4015
				const struct kvm_memory_slot *old,
4016
				const struct kvm_memory_slot *new,
4017
				enum kvm_mr_change change)
4018
{
4019
	int rc;
4020

4021 4022 4023 4024 4025 4026 4027 4028 4029 4030
	/* 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;
4031 4032 4033 4034

	rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr,
		mem->guest_phys_addr, mem->memory_size);
	if (rc)
4035
		pr_warn("failed to commit memory region\n");
4036
	return;
4037 4038
}

4039 4040 4041 4042 4043 4044 4045
static inline unsigned long nonhyp_mask(int i)
{
	unsigned int nonhyp_fai = (sclp.hmfai << i * 2) >> 30;

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

4046 4047 4048 4049 4050
void kvm_arch_vcpu_block_finish(struct kvm_vcpu *vcpu)
{
	vcpu->valid_wakeup = false;
}

4051 4052
static int __init kvm_s390_init(void)
{
4053 4054
	int i;

4055 4056 4057 4058 4059
	if (!sclp.has_sief2) {
		pr_info("SIE not available\n");
		return -ENODEV;
	}

4060 4061 4062 4063
	for (i = 0; i < 16; i++)
		kvm_s390_fac_list_mask[i] |=
			S390_lowcore.stfle_fac_list[i] & nonhyp_mask(i);

4064
	return kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE);
4065 4066 4067 4068 4069 4070 4071 4072 4073
}

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

module_init(kvm_s390_init);
module_exit(kvm_s390_exit);
4074 4075 4076 4077 4078 4079 4080 4081 4082

/*
 * 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");