kvm-s390.c 106.8 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
#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU
60
#define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM
61 62 63

struct kvm_stats_debugfs_item debugfs_entries[] = {
	{ "userspace_handled", VCPU_STAT(exit_userspace) },
64
	{ "exit_null", VCPU_STAT(exit_null) },
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) },
68
	{ "exit_io_request", VCPU_STAT(exit_io_request) },
69
	{ "exit_external_interrupt", VCPU_STAT(exit_external_interrupt) },
70
	{ "exit_instruction", VCPU_STAT(exit_instruction) },
71
	{ "exit_pei", VCPU_STAT(exit_pei) },
72 73
	{ "exit_program_interruption", VCPU_STAT(exit_program_interruption) },
	{ "exit_instr_and_program_int", VCPU_STAT(exit_instr_and_program) },
74
	{ "exit_operation_exception", VCPU_STAT(exit_operation_exception) },
75
	{ "halt_successful_poll", VCPU_STAT(halt_successful_poll) },
76
	{ "halt_attempted_poll", VCPU_STAT(halt_attempted_poll) },
77
	{ "halt_poll_invalid", VCPU_STAT(halt_poll_invalid) },
78
	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
79
	{ "instruction_lctlg", VCPU_STAT(instruction_lctlg) },
80
	{ "instruction_lctl", VCPU_STAT(instruction_lctl) },
81 82
	{ "instruction_stctl", VCPU_STAT(instruction_stctl) },
	{ "instruction_stctg", VCPU_STAT(instruction_stctg) },
83 84
	{ "deliver_ckc", VCPU_STAT(deliver_ckc) },
	{ "deliver_cputm", VCPU_STAT(deliver_cputm) },
85
	{ "deliver_emergency_signal", VCPU_STAT(deliver_emergency_signal) },
86
	{ "deliver_external_call", VCPU_STAT(deliver_external_call) },
87
	{ "deliver_service_signal", VCPU_STAT(deliver_service_signal) },
88
	{ "deliver_virtio", VCPU_STAT(deliver_virtio) },
89 90 91
	{ "deliver_stop_signal", VCPU_STAT(deliver_stop_signal) },
	{ "deliver_prefix_signal", VCPU_STAT(deliver_prefix_signal) },
	{ "deliver_restart_signal", VCPU_STAT(deliver_restart_signal) },
92 93
	{ "deliver_program", VCPU_STAT(deliver_program) },
	{ "deliver_io", VCPU_STAT(deliver_io) },
94
	{ "deliver_machine_check", VCPU_STAT(deliver_machine_check) },
95
	{ "exit_wait_state", VCPU_STAT(exit_wait_state) },
96 97 98 99 100 101 102 103 104 105 106 107 108 109 110
	{ "inject_ckc", VCPU_STAT(inject_ckc) },
	{ "inject_cputm", VCPU_STAT(inject_cputm) },
	{ "inject_external_call", VCPU_STAT(inject_external_call) },
	{ "inject_float_mchk", VM_STAT(inject_float_mchk) },
	{ "inject_emergency_signal", VCPU_STAT(inject_emergency_signal) },
	{ "inject_io", VM_STAT(inject_io) },
	{ "inject_mchk", VCPU_STAT(inject_mchk) },
	{ "inject_pfault_done", VM_STAT(inject_pfault_done) },
	{ "inject_program", VCPU_STAT(inject_program) },
	{ "inject_restart", VCPU_STAT(inject_restart) },
	{ "inject_service_signal", VM_STAT(inject_service_signal) },
	{ "inject_set_prefix", VCPU_STAT(inject_set_prefix) },
	{ "inject_stop_signal", VCPU_STAT(inject_stop_signal) },
	{ "inject_pfault_init", VCPU_STAT(inject_pfault_init) },
	{ "inject_virtio", VM_STAT(inject_virtio) },
111 112 113 114 115
	{ "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) },
116
	{ "instruction_pfmf", VCPU_STAT(instruction_pfmf) },
117
	{ "instruction_ptff", VCPU_STAT(instruction_ptff) },
118
	{ "instruction_stidp", VCPU_STAT(instruction_stidp) },
119 120
	{ "instruction_sck", VCPU_STAT(instruction_sck) },
	{ "instruction_sckpf", VCPU_STAT(instruction_sckpf) },
121 122 123
	{ "instruction_spx", VCPU_STAT(instruction_spx) },
	{ "instruction_stpx", VCPU_STAT(instruction_stpx) },
	{ "instruction_stap", VCPU_STAT(instruction_stap) },
124 125 126 127
	{ "instruction_iske", VCPU_STAT(instruction_iske) },
	{ "instruction_ri", VCPU_STAT(instruction_ri) },
	{ "instruction_rrbe", VCPU_STAT(instruction_rrbe) },
	{ "instruction_sske", VCPU_STAT(instruction_sske) },
128
	{ "instruction_ipte_interlock", VCPU_STAT(instruction_ipte_interlock) },
129
	{ "instruction_essa", VCPU_STAT(instruction_essa) },
130 131
	{ "instruction_stsi", VCPU_STAT(instruction_stsi) },
	{ "instruction_stfl", VCPU_STAT(instruction_stfl) },
132 133
	{ "instruction_tb", VCPU_STAT(instruction_tb) },
	{ "instruction_tpi", VCPU_STAT(instruction_tpi) },
134
	{ "instruction_tprot", VCPU_STAT(instruction_tprot) },
135
	{ "instruction_tsch", VCPU_STAT(instruction_tsch) },
J
Janosch Frank 已提交
136
	{ "instruction_sthyi", VCPU_STAT(instruction_sthyi) },
137
	{ "instruction_sie", VCPU_STAT(instruction_sie) },
138
	{ "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) },
139
	{ "instruction_sigp_sense_running", VCPU_STAT(instruction_sigp_sense_running) },
140
	{ "instruction_sigp_external_call", VCPU_STAT(instruction_sigp_external_call) },
141
	{ "instruction_sigp_emergency", VCPU_STAT(instruction_sigp_emergency) },
142 143
	{ "instruction_sigp_cond_emergency", VCPU_STAT(instruction_sigp_cond_emergency) },
	{ "instruction_sigp_start", VCPU_STAT(instruction_sigp_start) },
144
	{ "instruction_sigp_stop", VCPU_STAT(instruction_sigp_stop) },
145 146
	{ "instruction_sigp_stop_store_status", VCPU_STAT(instruction_sigp_stop_store_status) },
	{ "instruction_sigp_store_status", VCPU_STAT(instruction_sigp_store_status) },
147
	{ "instruction_sigp_store_adtl_status", VCPU_STAT(instruction_sigp_store_adtl_status) },
148 149 150
	{ "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) },
151 152 153
	{ "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) },
154 155 156 157 158 159
	{ "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) },
160
	{ "instruction_diag_other", VCPU_STAT(diagnose_other) },
161 162 163
	{ NULL }
};

164 165 166 167 168 169
struct kvm_s390_tod_clock_ext {
	__u8 epoch_idx;
	__u64 tod;
	__u8 reserved[7];
} __packed;

170 171 172 173 174
/* 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");

175

176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194
/*
 * For now we handle at most 16 double words as this is what the s390 base
 * kernel handles and stores in the prefix page. If we ever need to go beyond
 * this, this requires changes to code, but the external uapi can stay.
 */
#define SIZE_INTERNAL 16

/*
 * Base feature mask that defines default mask for facilities. Consists of the
 * defines in FACILITIES_KVM and the non-hypervisor managed bits.
 */
static unsigned long kvm_s390_fac_base[SIZE_INTERNAL] = { FACILITIES_KVM };
/*
 * Extended feature mask. Consists of the defines in FACILITIES_KVM_CPUMODEL
 * and defines the facilities that can be enabled via a cpu model.
 */
static unsigned long kvm_s390_fac_ext[SIZE_INTERNAL] = { FACILITIES_KVM_CPUMODEL };

static unsigned long kvm_s390_fac_size(void)
195
{
196 197 198 199 200 201
	BUILD_BUG_ON(SIZE_INTERNAL > S390_ARCH_FAC_MASK_SIZE_U64);
	BUILD_BUG_ON(SIZE_INTERNAL > S390_ARCH_FAC_LIST_SIZE_U64);
	BUILD_BUG_ON(SIZE_INTERNAL * sizeof(unsigned long) >
		sizeof(S390_lowcore.stfle_fac_list));

	return SIZE_INTERNAL;
202 203
}

204 205
/* available cpu features supported by kvm */
static DECLARE_BITMAP(kvm_s390_available_cpu_feat, KVM_S390_VM_CPU_FEAT_NR_BITS);
206 207
/* available subfunctions indicated via query / "test bit" */
static struct kvm_s390_vm_cpu_subfunc kvm_s390_available_subfunc;
208

209
static struct gmap_notifier gmap_notifier;
210
static struct gmap_notifier vsie_gmap_notifier;
211
debug_info_t *kvm_s390_dbf;
212

213
/* Section: not file related */
214
int kvm_arch_hardware_enable(void)
215 216
{
	/* every s390 is virtualization enabled ;-) */
217
	return 0;
218 219
}

220 221
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start,
			      unsigned long end);
222

223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244
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;
	}
}

245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260
/*
 * 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) {
261 262 263 264 265
			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;
			}
266 267
			if (vcpu->arch.cputm_enabled)
				vcpu->arch.cputm_start += *delta;
268
			if (vcpu->arch.vsie_block)
269 270
				kvm_clock_sync_scb(vcpu->arch.vsie_block,
						   *delta);
271 272 273 274 275 276 277 278 279
		}
	}
	return NOTIFY_OK;
}

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

280 281
int kvm_arch_hardware_setup(void)
{
282
	gmap_notifier.notifier_call = kvm_gmap_notifier;
283
	gmap_register_pte_notifier(&gmap_notifier);
284 285
	vsie_gmap_notifier.notifier_call = kvm_s390_vsie_gmap_notifier;
	gmap_register_pte_notifier(&vsie_gmap_notifier);
286 287
	atomic_notifier_chain_register(&s390_epoch_delta_notifier,
				       &kvm_clock_notifier);
288 289 290 291 292
	return 0;
}

void kvm_arch_hardware_unsetup(void)
{
293
	gmap_unregister_pte_notifier(&gmap_notifier);
294
	gmap_unregister_pte_notifier(&vsie_gmap_notifier);
295 296
	atomic_notifier_chain_unregister(&s390_epoch_delta_notifier,
					 &kvm_clock_notifier);
297 298
}

299 300 301 302 303
static void allow_cpu_feat(unsigned long nr)
{
	set_bit_inv(nr, kvm_s390_available_cpu_feat);
}

304 305 306
static inline int plo_test_bit(unsigned char nr)
{
	register unsigned long r0 asm("0") = (unsigned long) nr | 0x100;
307
	int cc;
308 309 310 311 312 313 314 315 316 317 318 319

	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;
}

320 321
static void kvm_s390_cpu_feat_init(void)
{
322 323 324 325 326 327 328 329
	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 */
330 331 332
		ptff(kvm_s390_available_subfunc.ptff,
		     sizeof(kvm_s390_available_subfunc.ptff),
		     PTFF_QAF);
333 334

	if (test_facility(17)) { /* MSA */
335 336 337 338 339 340 341 342 343 344
		__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);
345 346
	}
	if (test_facility(76)) /* MSA3 */
347 348
		__cpacf_query(CPACF_PCKMO, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.pckmo);
349
	if (test_facility(77)) { /* MSA4 */
350 351 352 353 354 355 356 357
		__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);
358 359
	}
	if (test_facility(57)) /* MSA5 */
360
		__cpacf_query(CPACF_PRNO, (cpacf_mask_t *)
361
			      kvm_s390_available_subfunc.ppno);
362

363 364 365 366
	if (test_facility(146)) /* MSA8 */
		__cpacf_query(CPACF_KMA, (cpacf_mask_t *)
			      kvm_s390_available_subfunc.kma);

367 368
	if (MACHINE_HAS_ESOP)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_ESOP);
369 370 371 372 373
	/*
	 * 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 ||
374
	    !test_facility(3) || !nested)
375 376
		return;
	allow_cpu_feat(KVM_S390_VM_CPU_FEAT_SIEF2);
377 378
	if (sclp.has_64bscao)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_64BSCAO);
379 380
	if (sclp.has_siif)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_SIIF);
381 382
	if (sclp.has_gpere)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_GPERE);
383 384
	if (sclp.has_gsls)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_GSLS);
385 386
	if (sclp.has_ib)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_IB);
387 388
	if (sclp.has_cei)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_CEI);
389 390
	if (sclp.has_ibs)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_IBS);
391 392
	if (sclp.has_kss)
		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_KSS);
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410
	/*
	 * 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.
	 */
411 412
}

413 414
int kvm_arch_init(void *opaque)
{
415 416 417 418 419 420 421 422 423
	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;
	}

424 425
	kvm_s390_cpu_feat_init();

426 427
	/* Register floating interrupt controller interface. */
	return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
428 429
}

430 431 432 433 434
void kvm_arch_exit(void)
{
	debug_unregister(kvm_s390_dbf);
}

435 436 437 438 439 440 441 442 443
/* 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;
}

444
int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
445
{
446 447
	int r;

448
	switch (ext) {
449
	case KVM_CAP_S390_PSW:
450
	case KVM_CAP_S390_GMAP:
451
	case KVM_CAP_SYNC_MMU:
452 453 454
#ifdef CONFIG_KVM_S390_UCONTROL
	case KVM_CAP_S390_UCONTROL:
#endif
455
	case KVM_CAP_ASYNC_PF:
456
	case KVM_CAP_SYNC_REGS:
457
	case KVM_CAP_ONE_REG:
458
	case KVM_CAP_ENABLE_CAP:
459
	case KVM_CAP_S390_CSS_SUPPORT:
C
Cornelia Huck 已提交
460
	case KVM_CAP_IOEVENTFD:
461
	case KVM_CAP_DEVICE_CTRL:
462
	case KVM_CAP_ENABLE_CAP_VM:
463
	case KVM_CAP_S390_IRQCHIP:
464
	case KVM_CAP_VM_ATTRIBUTES:
465
	case KVM_CAP_MP_STATE:
466
	case KVM_CAP_IMMEDIATE_EXIT:
467
	case KVM_CAP_S390_INJECT_IRQ:
468
	case KVM_CAP_S390_USER_SIGP:
469
	case KVM_CAP_S390_USER_STSI:
470
	case KVM_CAP_S390_SKEYS:
471
	case KVM_CAP_S390_IRQ_STATE:
472
	case KVM_CAP_S390_USER_INSTR0:
473
	case KVM_CAP_S390_CMMA_MIGRATION:
474
	case KVM_CAP_S390_AIS:
475
	case KVM_CAP_S390_AIS_MIGRATION:
476 477
		r = 1;
		break;
478 479 480
	case KVM_CAP_S390_MEM_OP:
		r = MEM_OP_MAX_SIZE;
		break;
481 482
	case KVM_CAP_NR_VCPUS:
	case KVM_CAP_MAX_VCPUS:
483
		r = KVM_S390_BSCA_CPU_SLOTS;
484 485 486
		if (!kvm_s390_use_sca_entries())
			r = KVM_MAX_VCPUS;
		else if (sclp.has_esca && sclp.has_64bscao)
487
			r = KVM_S390_ESCA_CPU_SLOTS;
488
		break;
489 490 491
	case KVM_CAP_NR_MEMSLOTS:
		r = KVM_USER_MEM_SLOTS;
		break;
492
	case KVM_CAP_S390_COW:
493
		r = MACHINE_HAS_ESOP;
494
		break;
495 496 497
	case KVM_CAP_S390_VECTOR_REGISTERS:
		r = MACHINE_HAS_VX;
		break;
498 499 500
	case KVM_CAP_S390_RI:
		r = test_facility(64);
		break;
F
Fan Zhang 已提交
501 502 503
	case KVM_CAP_S390_GS:
		r = test_facility(133);
		break;
504 505 506
	case KVM_CAP_S390_BPB:
		r = test_facility(82);
		break;
507
	default:
508
		r = 0;
509
	}
510
	return r;
511 512
}

513 514 515 516 517 518 519 520 521 522 523 524
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);

525
		if (test_and_clear_guest_dirty(gmap->mm, address))
526
			mark_page_dirty(kvm, cur_gfn);
527 528
		if (fatal_signal_pending(current))
			return;
529
		cond_resched();
530 531 532
	}
}

533
/* Section: vm related */
534 535
static void sca_del_vcpu(struct kvm_vcpu *vcpu);

536 537 538 539 540 541
/*
 * 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)
{
542 543
	int r;
	unsigned long n;
544
	struct kvm_memslots *slots;
545 546 547
	struct kvm_memory_slot *memslot;
	int is_dirty = 0;

548 549 550
	if (kvm_is_ucontrol(kvm))
		return -EINVAL;

551 552 553 554 555 556
	mutex_lock(&kvm->slots_lock);

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

557 558
	slots = kvm_memslots(kvm);
	memslot = id_to_memslot(slots, log->slot);
559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576
	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;
577 578
}

579 580 581 582 583 584 585 586 587 588
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);
	}
}

589 590 591 592 593 594 595 596
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) {
597
	case KVM_CAP_S390_IRQCHIP:
598
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_IRQCHIP");
599 600 601
		kvm->arch.use_irqchip = 1;
		r = 0;
		break;
602
	case KVM_CAP_S390_USER_SIGP:
603
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_SIGP");
604 605 606
		kvm->arch.user_sigp = 1;
		r = 0;
		break;
607
	case KVM_CAP_S390_VECTOR_REGISTERS:
608
		mutex_lock(&kvm->lock);
609
		if (kvm->created_vcpus) {
610 611
			r = -EBUSY;
		} else if (MACHINE_HAS_VX) {
612 613
			set_kvm_facility(kvm->arch.model.fac_mask, 129);
			set_kvm_facility(kvm->arch.model.fac_list, 129);
614 615 616 617
			if (test_facility(134)) {
				set_kvm_facility(kvm->arch.model.fac_mask, 134);
				set_kvm_facility(kvm->arch.model.fac_list, 134);
			}
618 619 620 621
			if (test_facility(135)) {
				set_kvm_facility(kvm->arch.model.fac_mask, 135);
				set_kvm_facility(kvm->arch.model.fac_list, 135);
			}
622 623 624
			r = 0;
		} else
			r = -EINVAL;
625
		mutex_unlock(&kvm->lock);
626 627
		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_VECTOR_REGISTERS %s",
			 r ? "(not available)" : "(success)");
628
		break;
629 630 631
	case KVM_CAP_S390_RI:
		r = -EINVAL;
		mutex_lock(&kvm->lock);
632
		if (kvm->created_vcpus) {
633 634
			r = -EBUSY;
		} else if (test_facility(64)) {
635 636
			set_kvm_facility(kvm->arch.model.fac_mask, 64);
			set_kvm_facility(kvm->arch.model.fac_list, 64);
637 638 639 640 641 642
			r = 0;
		}
		mutex_unlock(&kvm->lock);
		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_RI %s",
			 r ? "(not available)" : "(success)");
		break;
643 644 645 646 647 648 649 650 651 652 653 654 655
	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
656 657 658
	case KVM_CAP_S390_GS:
		r = -EINVAL;
		mutex_lock(&kvm->lock);
659
		if (kvm->created_vcpus) {
F
Fan Zhang 已提交
660 661 662 663 664 665 666 667 668 669
			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;
670
	case KVM_CAP_S390_USER_STSI:
671
		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_STSI");
672 673 674
		kvm->arch.user_stsi = 1;
		r = 0;
		break;
675 676 677 678 679 680
	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;
681 682 683 684 685 686 687
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

688 689 690 691 692 693 694
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;
695
		VM_EVENT(kvm, 3, "QUERY: max guest memory: %lu bytes",
696 697
			 kvm->arch.mem_limit);
		if (put_user(kvm->arch.mem_limit, (u64 __user *)attr->addr))
698 699 700 701 702 703 704 705 706 707
			ret = -EFAULT;
		break;
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *attr)
708 709 710 711 712
{
	int ret;
	unsigned int idx;
	switch (attr->attr) {
	case KVM_S390_VM_MEM_ENABLE_CMMA:
713
		ret = -ENXIO;
714
		if (!sclp.has_cmma)
715 716
			break;

717
		ret = -EBUSY;
718
		VM_EVENT(kvm, 3, "%s", "ENABLE: CMMA support");
719
		mutex_lock(&kvm->lock);
720
		if (!kvm->created_vcpus) {
721
			kvm->arch.use_cmma = 1;
722 723
			/* Not compatible with cmma. */
			kvm->arch.use_pfmfi = 0;
724 725 726 727 728
			ret = 0;
		}
		mutex_unlock(&kvm->lock);
		break;
	case KVM_S390_VM_MEM_CLR_CMMA:
729 730 731
		ret = -ENXIO;
		if (!sclp.has_cmma)
			break;
732 733 734 735
		ret = -EINVAL;
		if (!kvm->arch.use_cmma)
			break;

736
		VM_EVENT(kvm, 3, "%s", "RESET: CMMA states");
737 738
		mutex_lock(&kvm->lock);
		idx = srcu_read_lock(&kvm->srcu);
739
		s390_reset_cmma(kvm->arch.gmap->mm);
740 741 742 743
		srcu_read_unlock(&kvm->srcu, idx);
		mutex_unlock(&kvm->lock);
		ret = 0;
		break;
744 745 746 747 748 749 750 751 752
	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;

753 754
		if (kvm->arch.mem_limit != KVM_S390_NO_MEM_LIMIT &&
		    new_limit > kvm->arch.mem_limit)
755 756
			return -E2BIG;

757 758 759
		if (!new_limit)
			return -EINVAL;

760
		/* gmap_create takes last usable address */
761 762 763
		if (new_limit != KVM_S390_NO_MEM_LIMIT)
			new_limit -= 1;

764 765
		ret = -EBUSY;
		mutex_lock(&kvm->lock);
766
		if (!kvm->created_vcpus) {
767 768
			/* gmap_create will round the limit up */
			struct gmap *new = gmap_create(current->mm, new_limit);
769 770 771 772

			if (!new) {
				ret = -ENOMEM;
			} else {
773
				gmap_remove(kvm->arch.gmap);
774 775 776 777 778 779
				new->private = kvm;
				kvm->arch.gmap = new;
				ret = 0;
			}
		}
		mutex_unlock(&kvm->lock);
780 781 782
		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);
783 784
		break;
	}
785 786 787 788 789 790 791
	default:
		ret = -ENXIO;
		break;
	}
	return ret;
}

792 793 794 795 796 797 798
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;

799
	if (!test_kvm_facility(kvm, 76))
800 801 802 803 804 805 806 807 808
		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;
809
		VM_EVENT(kvm, 3, "%s", "ENABLE: AES keywrapping support");
810 811 812 813 814 815
		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;
816
		VM_EVENT(kvm, 3, "%s", "ENABLE: DEA keywrapping support");
817 818 819 820 821
		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));
822
		VM_EVENT(kvm, 3, "%s", "DISABLE: AES keywrapping support");
823 824 825 826 827
		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));
828
		VM_EVENT(kvm, 3, "%s", "DISABLE: DEA keywrapping support");
829 830 831 832 833 834 835 836 837 838 839 840 841 842
		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;
}

843 844 845 846 847 848 849 850 851 852 853
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
854
 * kvm->slots_lock to avoid races with ourselves and kvm_s390_vm_stop_migration.
855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879
 */
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) {
		/*
880 881 882 883
		 * 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.
884
		 */
885
		ms = slots->memslots;
886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909
		/* 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;
}

/*
910
 * Must be called with kvm->slots_lock to avoid races with ourselves and
911 912 913 914 915 916 917 918 919 920 921 922 923 924
 * 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);
925 926
		/* We have to wait for the essa emulation to finish */
		synchronize_srcu(&kvm->srcu);
927 928 929 930 931 932 933 934 935
		vfree(mgs->pgste_bitmap);
	}
	kfree(mgs);
	return 0;
}

static int kvm_s390_vm_set_migration(struct kvm *kvm,
				     struct kvm_device_attr *attr)
{
936
	int res = -ENXIO;
937

938
	mutex_lock(&kvm->slots_lock);
939 940 941 942 943 944 945 946 947 948
	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;
	}
949
	mutex_unlock(&kvm->slots_lock);
950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966

	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;
}

967 968 969 970 971 972 973
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;

974
	if (!test_kvm_facility(kvm, 139) && gtod.epoch_idx)
975
		return -EINVAL;
976
	kvm_s390_set_tod_clock(kvm, &gtod);
977 978 979 980 981 982 983

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

	return 0;
}

984 985 986 987 988 989 990 991 992 993
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;
994
	VM_EVENT(kvm, 3, "SET: TOD extension: 0x%x", gtod_high);
995 996 997 998 999 1000

	return 0;
}

static int kvm_s390_set_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
1001
	struct kvm_s390_vm_tod_clock gtod = { 0 };
1002

1003 1004
	if (copy_from_user(&gtod.tod, (void __user *)attr->addr,
			   sizeof(gtod.tod)))
1005 1006
		return -EFAULT;

1007 1008
	kvm_s390_set_tod_clock(kvm, &gtod);
	VM_EVENT(kvm, 3, "SET: TOD base: 0x%llx", gtod.tod);
1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019
	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) {
1020 1021 1022
	case KVM_S390_VM_TOD_EXT:
		ret = kvm_s390_set_tod_ext(kvm, attr);
		break;
1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035
	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;
}

1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072
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;
}

1073 1074 1075 1076 1077 1078 1079
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;
1080
	VM_EVENT(kvm, 3, "QUERY: TOD extension: 0x%x", gtod_high);
1081 1082 1083 1084 1085 1086

	return 0;
}

static int kvm_s390_get_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
1087
	u64 gtod;
1088

1089
	gtod = kvm_s390_get_tod_clock_fast(kvm);
1090 1091
	if (copy_to_user((void __user *)attr->addr, &gtod, sizeof(gtod)))
		return -EFAULT;
1092
	VM_EVENT(kvm, 3, "QUERY: TOD base: 0x%llx", gtod);
1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104

	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) {
1105 1106 1107
	case KVM_S390_VM_TOD_EXT:
		ret = kvm_s390_get_tod_ext(kvm, attr);
		break;
1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120
	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;
}

1121 1122 1123
static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr)
{
	struct kvm_s390_vm_cpu_processor *proc;
1124
	u16 lowest_ibc, unblocked_ibc;
1125 1126 1127
	int ret = 0;

	mutex_lock(&kvm->lock);
1128
	if (kvm->created_vcpus) {
1129 1130 1131 1132 1133 1134 1135 1136 1137 1138
		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))) {
1139
		kvm->arch.model.cpuid = proc->cpuid;
1140 1141
		lowest_ibc = sclp.ibc >> 16 & 0xfff;
		unblocked_ibc = sclp.ibc & 0xfff;
1142
		if (lowest_ibc && proc->ibc) {
1143 1144 1145 1146 1147 1148 1149
			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;
		}
1150
		memcpy(kvm->arch.model.fac_list, proc->fac_list,
1151
		       S390_ARCH_FAC_LIST_SIZE_BYTE);
1152 1153 1154 1155 1156 1157 1158
		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]);
1159 1160 1161 1162 1163 1164 1165 1166
	} else
		ret = -EFAULT;
	kfree(proc);
out:
	mutex_unlock(&kvm->lock);
	return ret;
}

1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179
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);
1180 1181 1182
	if (kvm->created_vcpus) {
		mutex_unlock(&kvm->lock);
		return -EBUSY;
1183
	}
1184 1185
	bitmap_copy(kvm->arch.cpu_feat, (unsigned long *) data.feat,
		    KVM_S390_VM_CPU_FEAT_NR_BITS);
1186
	mutex_unlock(&kvm->lock);
1187 1188 1189 1190 1191
	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;
1192 1193
}

1194 1195 1196 1197 1198 1199 1200 1201 1202 1203
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;
}

1204 1205 1206 1207 1208 1209 1210 1211
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;
1212 1213 1214
	case KVM_S390_VM_CPU_PROCESSOR_FEAT:
		ret = kvm_s390_set_processor_feat(kvm, attr);
		break;
1215 1216 1217
	case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
		ret = kvm_s390_set_processor_subfunc(kvm, attr);
		break;
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231
	}
	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;
	}
1232
	proc->cpuid = kvm->arch.model.cpuid;
1233
	proc->ibc = kvm->arch.model.ibc;
1234 1235
	memcpy(&proc->fac_list, kvm->arch.model.fac_list,
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
1236 1237 1238 1239 1240 1241 1242
	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]);
1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260
	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);
1261
	mach->ibc = sclp.ibc;
1262
	memcpy(&mach->fac_mask, kvm->arch.model.fac_mask,
1263
	       S390_ARCH_FAC_LIST_SIZE_BYTE);
1264
	memcpy((unsigned long *)&mach->fac_list, S390_lowcore.stfle_fac_list,
1265
	       sizeof(S390_lowcore.stfle_fac_list));
1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276
	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]);
1277 1278 1279 1280 1281 1282 1283
	if (copy_to_user((void __user *)attr->addr, mach, sizeof(*mach)))
		ret = -EFAULT;
	kfree(mach);
out:
	return ret;
}

1284 1285 1286 1287 1288 1289 1290 1291 1292
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;
1293 1294 1295 1296
	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]);
1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309
	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;
1310 1311 1312 1313
	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]);
1314 1315 1316
	return 0;
}

1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335
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;
}
1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346
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;
1347 1348 1349 1350 1351 1352
	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;
1353 1354 1355 1356 1357 1358
	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;
1359 1360 1361 1362
	}
	return ret;
}

1363 1364 1365 1366 1367
static int kvm_s390_vm_set_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
	int ret;

	switch (attr->group) {
1368
	case KVM_S390_VM_MEM_CTRL:
1369
		ret = kvm_s390_set_mem_control(kvm, attr);
1370
		break;
1371 1372 1373
	case KVM_S390_VM_TOD:
		ret = kvm_s390_set_tod(kvm, attr);
		break;
1374 1375 1376
	case KVM_S390_VM_CPU_MODEL:
		ret = kvm_s390_set_cpu_model(kvm, attr);
		break;
1377 1378 1379
	case KVM_S390_VM_CRYPTO:
		ret = kvm_s390_vm_set_crypto(kvm, attr);
		break;
1380 1381 1382
	case KVM_S390_VM_MIGRATION:
		ret = kvm_s390_vm_set_migration(kvm, attr);
		break;
1383 1384 1385 1386 1387 1388 1389 1390 1391 1392
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

static int kvm_s390_vm_get_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
1393 1394 1395 1396 1397 1398
	int ret;

	switch (attr->group) {
	case KVM_S390_VM_MEM_CTRL:
		ret = kvm_s390_get_mem_control(kvm, attr);
		break;
1399 1400 1401
	case KVM_S390_VM_TOD:
		ret = kvm_s390_get_tod(kvm, attr);
		break;
1402 1403 1404
	case KVM_S390_VM_CPU_MODEL:
		ret = kvm_s390_get_cpu_model(kvm, attr);
		break;
1405 1406 1407
	case KVM_S390_VM_MIGRATION:
		ret = kvm_s390_vm_get_migration(kvm, attr);
		break;
1408 1409 1410 1411 1412 1413
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
1414 1415 1416 1417 1418 1419 1420
}

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

	switch (attr->group) {
1421 1422 1423 1424
	case KVM_S390_VM_MEM_CTRL:
		switch (attr->attr) {
		case KVM_S390_VM_MEM_ENABLE_CMMA:
		case KVM_S390_VM_MEM_CLR_CMMA:
1425 1426
			ret = sclp.has_cmma ? 0 : -ENXIO;
			break;
1427
		case KVM_S390_VM_MEM_LIMIT_SIZE:
1428 1429 1430 1431 1432 1433 1434
			ret = 0;
			break;
		default:
			ret = -ENXIO;
			break;
		}
		break;
1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445
	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;
1446 1447 1448 1449
	case KVM_S390_VM_CPU_MODEL:
		switch (attr->attr) {
		case KVM_S390_VM_CPU_PROCESSOR:
		case KVM_S390_VM_CPU_MACHINE:
1450 1451
		case KVM_S390_VM_CPU_PROCESSOR_FEAT:
		case KVM_S390_VM_CPU_MACHINE_FEAT:
1452
		case KVM_S390_VM_CPU_MACHINE_SUBFUNC:
1453 1454
			ret = 0;
			break;
1455 1456
		/* configuring subfunctions is not supported yet */
		case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
1457 1458 1459 1460 1461
		default:
			ret = -ENXIO;
			break;
		}
		break;
1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474
	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;
1475 1476 1477
	case KVM_S390_VM_MIGRATION:
		ret = 0;
		break;
1478 1479 1480 1481 1482 1483 1484 1485
	default:
		ret = -ENXIO;
		break;
	}

	return ret;
}

1486 1487 1488 1489
static long kvm_s390_get_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
{
	uint8_t *keys;
	uint64_t hva;
1490
	int srcu_idx, i, r = 0;
1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502

	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;

1503
	keys = kvmalloc_array(args->count, sizeof(uint8_t), GFP_KERNEL);
1504 1505 1506
	if (!keys)
		return -ENOMEM;

1507
	down_read(&current->mm->mmap_sem);
1508
	srcu_idx = srcu_read_lock(&kvm->srcu);
1509 1510 1511 1512
	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
1513
			break;
1514 1515
		}

1516 1517
		r = get_guest_storage_key(current->mm, hva, &keys[i]);
		if (r)
1518
			break;
1519
	}
1520
	srcu_read_unlock(&kvm->srcu, srcu_idx);
1521 1522 1523 1524 1525 1526 1527
	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;
1528 1529 1530 1531 1532 1533 1534 1535 1536 1537
	}

	kvfree(keys);
	return r;
}

static long kvm_s390_set_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
{
	uint8_t *keys;
	uint64_t hva;
1538
	int srcu_idx, i, r = 0;
1539 1540 1541 1542 1543 1544 1545 1546

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

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

1547
	keys = kvmalloc_array(args->count, sizeof(uint8_t), GFP_KERNEL);
1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558
	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 */
1559 1560 1561
	r = s390_enable_skey();
	if (r)
		goto out;
1562

1563
	down_read(&current->mm->mmap_sem);
1564
	srcu_idx = srcu_read_lock(&kvm->srcu);
1565 1566 1567 1568
	for (i = 0; i < args->count; i++) {
		hva = gfn_to_hva(kvm, args->start_gfn + i);
		if (kvm_is_error_hva(hva)) {
			r = -EFAULT;
1569
			break;
1570 1571 1572 1573 1574
		}

		/* Lowest order bit is reserved */
		if (keys[i] & 0x01) {
			r = -EINVAL;
1575
			break;
1576 1577
		}

1578
		r = set_guest_storage_key(current->mm, hva, keys[i], 0);
1579
		if (r)
1580
			break;
1581
	}
1582
	srcu_read_unlock(&kvm->srcu, srcu_idx);
1583
	up_read(&current->mm->mmap_sem);
1584 1585 1586 1587 1588
out:
	kvfree(keys);
	return r;
}

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
/*
 * 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);
1628
	if (!bufsize || !kvm->mm->context.uses_cmm) {
1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670
		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 */
1671
		res[i++] = (pgstev >> 24) & 0x43;
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
		/*
		 * 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
1705
 * set and the mm->context.uses_cmm flag is set.
1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748
 */
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;
1749
		mask &= _PGSTE_GPS_USAGE_MASK | _PGSTE_GPS_NODAT;
1750 1751 1752 1753 1754
		set_pgste_bits(kvm->mm, hva, mask, pgstev);
	}
	srcu_read_unlock(&kvm->srcu, srcu_idx);
	up_read(&kvm->mm->mmap_sem);

1755
	if (!kvm->mm->context.uses_cmm) {
1756
		down_write(&kvm->mm->mmap_sem);
1757
		kvm->mm->context.uses_cmm = 1;
1758 1759 1760 1761 1762 1763 1764
		up_write(&kvm->mm->mmap_sem);
	}
out:
	vfree(bits);
	return r;
}

1765 1766 1767 1768 1769
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;
1770
	struct kvm_device_attr attr;
1771 1772 1773
	int r;

	switch (ioctl) {
1774 1775 1776 1777 1778 1779 1780 1781 1782
	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;
	}
1783 1784 1785 1786 1787 1788 1789 1790
	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;
	}
1791 1792 1793 1794 1795 1796 1797
	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));
1798
			r = kvm_set_irq_routing(kvm, &routing, 0, 0);
1799 1800 1801
		}
		break;
	}
1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822
	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;
	}
1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842
	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;
	}
1843 1844 1845 1846 1847 1848
	case KVM_S390_GET_CMMA_BITS: {
		struct kvm_s390_cmma_log args;

		r = -EFAULT;
		if (copy_from_user(&args, argp, sizeof(args)))
			break;
1849
		mutex_lock(&kvm->slots_lock);
1850
		r = kvm_s390_get_cmma_bits(kvm, &args);
1851
		mutex_unlock(&kvm->slots_lock);
1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864
		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;
1865
		mutex_lock(&kvm->slots_lock);
1866
		r = kvm_s390_set_cmma_bits(kvm, &args);
1867
		mutex_unlock(&kvm->slots_lock);
1868 1869
		break;
	}
1870
	default:
1871
		r = -ENOTTY;
1872 1873 1874 1875 1876
	}

	return r;
}

1877 1878 1879
static int kvm_s390_query_ap_config(u8 *config)
{
	u32 fcn_code = 0x04000000UL;
1880
	u32 cc = 0;
1881

1882
	memset(config, 0, 128);
1883 1884 1885 1886
	asm volatile(
		"lgr 0,%1\n"
		"lgr 2,%2\n"
		".long 0xb2af0000\n"		/* PQAP(QCI) */
1887
		"0: ipm %0\n"
1888
		"srl %0,28\n"
1889 1890 1891
		"1:\n"
		EX_TABLE(0b, 1b)
		: "+r" (cc)
1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903
		: "r" (fcn_code), "r" (config)
		: "cc", "0", "2", "memory"
	);

	return cc;
}

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

1904
	if (test_facility(12)) {
1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925
		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;
}

1926
static u64 kvm_s390_get_initial_cpuid(void)
1927
{
1928 1929 1930 1931 1932
	struct cpuid cpuid;

	get_cpu_id(&cpuid);
	cpuid.version = 0xff;
	return *((u64 *) &cpuid);
1933 1934
}

1935
static void kvm_s390_crypto_init(struct kvm *kvm)
1936
{
1937
	if (!test_kvm_facility(kvm, 76))
1938
		return;
1939

1940
	kvm->arch.crypto.crycb = &kvm->arch.sie_page2->crycb;
1941
	kvm_s390_set_crycb_format(kvm);
1942

1943 1944 1945 1946 1947 1948 1949
	/* 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));
1950 1951
}

1952 1953 1954
static void sca_dispose(struct kvm *kvm)
{
	if (kvm->arch.use_esca)
1955
		free_pages_exact(kvm->arch.sca, sizeof(struct esca_block));
1956 1957 1958 1959 1960
	else
		free_page((unsigned long)(kvm->arch.sca));
	kvm->arch.sca = NULL;
}

1961
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
1962
{
1963
	gfp_t alloc_flags = GFP_KERNEL;
1964
	int i, rc;
1965
	char debug_name[16];
1966
	static unsigned long sca_offset;
1967

1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978
	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

1979 1980
	rc = s390_enable_sie();
	if (rc)
1981
		goto out_err;
1982

1983 1984
	rc = -ENOMEM;

1985
	kvm->arch.use_esca = 0; /* start with basic SCA */
1986 1987
	if (!sclp.has_64bscao)
		alloc_flags |= GFP_DMA;
1988
	rwlock_init(&kvm->arch.sca_lock);
1989
	kvm->arch.sca = (struct bsca_block *) get_zeroed_page(alloc_flags);
1990
	if (!kvm->arch.sca)
1991
		goto out_err;
1992
	spin_lock(&kvm_lock);
1993
	sca_offset += 16;
1994
	if (sca_offset + sizeof(struct bsca_block) > PAGE_SIZE)
1995
		sca_offset = 0;
1996 1997
	kvm->arch.sca = (struct bsca_block *)
			((char *) kvm->arch.sca + sca_offset);
1998
	spin_unlock(&kvm_lock);
1999 2000 2001

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

2002
	kvm->arch.dbf = debug_register(debug_name, 32, 1, 7 * sizeof(long));
2003
	if (!kvm->arch.dbf)
2004
		goto out_err;
2005

2006
	BUILD_BUG_ON(sizeof(struct sie_page2) != 4096);
2007 2008 2009
	kvm->arch.sie_page2 =
	     (struct sie_page2 *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
	if (!kvm->arch.sie_page2)
2010
		goto out_err;
2011

2012
	kvm->arch.model.fac_list = kvm->arch.sie_page2->fac_list;
2013 2014 2015 2016 2017 2018 2019 2020

	for (i = 0; i < kvm_s390_fac_size(); i++) {
		kvm->arch.model.fac_mask[i] = S390_lowcore.stfle_fac_list[i] &
					      (kvm_s390_fac_base[i] |
					       kvm_s390_fac_ext[i]);
		kvm->arch.model.fac_list[i] = S390_lowcore.stfle_fac_list[i] &
					      kvm_s390_fac_base[i];
	}
2021

2022 2023 2024 2025
	/* 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
2026 2027
	set_kvm_facility(kvm->arch.model.fac_mask, 74);
	set_kvm_facility(kvm->arch.model.fac_list, 74);
2028 2029 2030 2031
	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
2032

2033
	kvm->arch.model.cpuid = kvm_s390_get_initial_cpuid();
2034
	kvm->arch.model.ibc = sclp.ibc & 0x0fff;
2035

2036
	kvm_s390_crypto_init(kvm);
2037

2038 2039 2040
	mutex_init(&kvm->arch.float_int.ais_lock);
	kvm->arch.float_int.simm = 0;
	kvm->arch.float_int.nimm = 0;
2041
	spin_lock_init(&kvm->arch.float_int.lock);
2042 2043
	for (i = 0; i < FIRQ_LIST_COUNT; i++)
		INIT_LIST_HEAD(&kvm->arch.float_int.lists[i]);
2044
	init_waitqueue_head(&kvm->arch.ipte_wq);
2045
	mutex_init(&kvm->arch.ipte_mutex);
2046

2047
	debug_register_view(kvm->arch.dbf, &debug_sprintf_view);
2048
	VM_EVENT(kvm, 3, "vm created with type %lu", type);
2049

2050 2051
	if (type & KVM_VM_S390_UCONTROL) {
		kvm->arch.gmap = NULL;
2052
		kvm->arch.mem_limit = KVM_S390_NO_MEM_LIMIT;
2053
	} else {
2054
		if (sclp.hamax == U64_MAX)
2055
			kvm->arch.mem_limit = TASK_SIZE_MAX;
2056
		else
2057
			kvm->arch.mem_limit = min_t(unsigned long, TASK_SIZE_MAX,
2058
						    sclp.hamax + 1);
2059
		kvm->arch.gmap = gmap_create(current->mm, kvm->arch.mem_limit - 1);
2060
		if (!kvm->arch.gmap)
2061
			goto out_err;
2062
		kvm->arch.gmap->private = kvm;
2063
		kvm->arch.gmap->pfault_enabled = 0;
2064
	}
2065 2066

	kvm->arch.css_support = 0;
2067
	kvm->arch.use_irqchip = 0;
2068
	kvm->arch.use_pfmfi = sclp.has_pfmfi;
2069
	kvm->arch.epoch = 0;
2070

2071
	spin_lock_init(&kvm->arch.start_stop_lock);
2072
	kvm_s390_vsie_init(kvm);
2073
	kvm_s390_gisa_init(kvm);
2074
	KVM_EVENT(3, "vm 0x%pK created by pid %u", kvm, current->pid);
2075

2076
	return 0;
2077
out_err:
2078
	free_page((unsigned long)kvm->arch.sie_page2);
2079
	debug_unregister(kvm->arch.dbf);
2080
	sca_dispose(kvm);
2081
	KVM_EVENT(3, "creation of vm failed: %d", rc);
2082
	return rc;
2083 2084
}

2085 2086 2087 2088 2089 2090 2091 2092 2093 2094
bool kvm_arch_has_vcpu_debugfs(void)
{
	return false;
}

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

2095 2096 2097
void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
{
	VCPU_EVENT(vcpu, 3, "%s", "free cpu");
2098
	trace_kvm_s390_destroy_vcpu(vcpu->vcpu_id);
2099
	kvm_s390_clear_local_irqs(vcpu);
2100
	kvm_clear_async_pf_completion_queue(vcpu);
2101
	if (!kvm_is_ucontrol(vcpu->kvm))
2102
		sca_del_vcpu(vcpu);
2103 2104

	if (kvm_is_ucontrol(vcpu->kvm))
2105
		gmap_remove(vcpu->arch.gmap);
2106

2107
	if (vcpu->kvm->arch.use_cmma)
2108
		kvm_s390_vcpu_unsetup_cmma(vcpu);
2109
	free_page((unsigned long)(vcpu->arch.sie_block));
2110

2111
	kvm_vcpu_uninit(vcpu);
2112
	kmem_cache_free(kvm_vcpu_cache, vcpu);
2113 2114 2115 2116 2117
}

static void kvm_free_vcpus(struct kvm *kvm)
{
	unsigned int i;
2118
	struct kvm_vcpu *vcpu;
2119

2120 2121 2122 2123 2124 2125 2126 2127 2128
	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);
2129 2130
}

2131 2132
void kvm_arch_destroy_vm(struct kvm *kvm)
{
2133
	kvm_free_vcpus(kvm);
2134
	sca_dispose(kvm);
2135
	debug_unregister(kvm->arch.dbf);
2136
	kvm_s390_gisa_destroy(kvm);
2137
	free_page((unsigned long)kvm->arch.sie_page2);
2138
	if (!kvm_is_ucontrol(kvm))
2139
		gmap_remove(kvm->arch.gmap);
2140
	kvm_s390_destroy_adapters(kvm);
2141
	kvm_s390_clear_float_irqs(kvm);
2142
	kvm_s390_vsie_destroy(kvm);
2143 2144 2145 2146
	if (kvm->arch.migration_state) {
		vfree(kvm->arch.migration_state->pgste_bitmap);
		kfree(kvm->arch.migration_state);
	}
2147
	KVM_EVENT(3, "vm 0x%pK destroyed", kvm);
2148 2149 2150
}

/* Section: vcpu related */
2151 2152
static int __kvm_ucontrol_vcpu_init(struct kvm_vcpu *vcpu)
{
2153
	vcpu->arch.gmap = gmap_create(current->mm, -1UL);
2154 2155 2156 2157 2158 2159 2160
	if (!vcpu->arch.gmap)
		return -ENOMEM;
	vcpu->arch.gmap->private = vcpu->kvm;

	return 0;
}

2161 2162
static void sca_del_vcpu(struct kvm_vcpu *vcpu)
{
2163 2164
	if (!kvm_s390_use_sca_entries())
		return;
2165
	read_lock(&vcpu->kvm->arch.sca_lock);
2166 2167
	if (vcpu->kvm->arch.use_esca) {
		struct esca_block *sca = vcpu->kvm->arch.sca;
2168

2169
		clear_bit_inv(vcpu->vcpu_id, (unsigned long *) sca->mcn);
2170
		sca->cpu[vcpu->vcpu_id].sda = 0;
2171 2172 2173 2174
	} else {
		struct bsca_block *sca = vcpu->kvm->arch.sca;

		clear_bit_inv(vcpu->vcpu_id, (unsigned long *) &sca->mcn);
2175
		sca->cpu[vcpu->vcpu_id].sda = 0;
2176
	}
2177
	read_unlock(&vcpu->kvm->arch.sca_lock);
2178 2179
}

2180
static void sca_add_vcpu(struct kvm_vcpu *vcpu)
2181
{
2182 2183 2184 2185 2186 2187
	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;
2188
		return;
2189
	}
2190 2191 2192
	read_lock(&vcpu->kvm->arch.sca_lock);
	if (vcpu->kvm->arch.use_esca) {
		struct esca_block *sca = vcpu->kvm->arch.sca;
2193

2194
		sca->cpu[vcpu->vcpu_id].sda = (__u64) vcpu->arch.sie_block;
2195 2196
		vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
		vcpu->arch.sie_block->scaol = (__u32)(__u64)sca & ~0x3fU;
2197
		vcpu->arch.sie_block->ecb2 |= ECB2_ESCA;
2198
		set_bit_inv(vcpu->vcpu_id, (unsigned long *) sca->mcn);
2199
	} else {
2200
		struct bsca_block *sca = vcpu->kvm->arch.sca;
2201

2202
		sca->cpu[vcpu->vcpu_id].sda = (__u64) vcpu->arch.sie_block;
2203 2204
		vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
		vcpu->arch.sie_block->scaol = (__u32)(__u64)sca;
2205
		set_bit_inv(vcpu->vcpu_id, (unsigned long *) &sca->mcn);
2206
	}
2207
	read_unlock(&vcpu->kvm->arch.sca_lock);
2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250
}

/* 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;
2251
		vcpu->arch.sie_block->ecb2 |= ECB2_ESCA;
2252 2253 2254 2255 2256 2257 2258 2259 2260
	}
	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);

2261 2262
	VM_EVENT(kvm, 2, "Switched to ESCA (0x%pK -> 0x%pK)",
		 old_sca, kvm->arch.sca);
2263
	return 0;
2264 2265 2266 2267
}

static int sca_can_add_vcpu(struct kvm *kvm, unsigned int id)
{
2268 2269
	int rc;

2270 2271 2272 2273 2274
	if (!kvm_s390_use_sca_entries()) {
		if (id < KVM_MAX_VCPUS)
			return true;
		return false;
	}
2275 2276
	if (id < KVM_S390_BSCA_CPU_SLOTS)
		return true;
2277
	if (!sclp.has_esca || !sclp.has_64bscao)
2278 2279 2280 2281 2282 2283 2284
		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;
2285 2286
}

2287 2288
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
2289 2290
	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
	kvm_clear_async_pf_completion_queue(vcpu);
2291 2292
	vcpu->run->kvm_valid_regs = KVM_SYNC_PREFIX |
				    KVM_SYNC_GPRS |
2293
				    KVM_SYNC_ACRS |
2294 2295 2296
				    KVM_SYNC_CRS |
				    KVM_SYNC_ARCH0 |
				    KVM_SYNC_PFAULT;
2297
	kvm_s390_set_prefix(vcpu, 0);
2298 2299
	if (test_kvm_facility(vcpu->kvm, 64))
		vcpu->run->kvm_valid_regs |= KVM_SYNC_RICCB;
2300 2301
	if (test_kvm_facility(vcpu->kvm, 82))
		vcpu->run->kvm_valid_regs |= KVM_SYNC_BPBC;
F
Fan Zhang 已提交
2302 2303
	if (test_kvm_facility(vcpu->kvm, 133))
		vcpu->run->kvm_valid_regs |= KVM_SYNC_GSCB;
2304 2305 2306 2307
	/* 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)
2308
		vcpu->run->kvm_valid_regs |= KVM_SYNC_VRS;
2309 2310
	else
		vcpu->run->kvm_valid_regs |= KVM_SYNC_FPRS;
2311 2312 2313 2314

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

2315 2316 2317
	return 0;
}

2318 2319 2320 2321
/* 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);
2322
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
2323
	vcpu->arch.cputm_start = get_tod_clock_fast();
2324
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
2325 2326 2327 2328 2329 2330
}

/* 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);
2331
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
2332 2333
	vcpu->arch.sie_block->cputm -= get_tod_clock_fast() - vcpu->arch.cputm_start;
	vcpu->arch.cputm_start = 0;
2334
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366
}

/* 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();
}

2367 2368 2369
/* 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)
{
2370
	preempt_disable(); /* protect from TOD sync and vcpu_load/put */
2371
	raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
2372 2373
	if (vcpu->arch.cputm_enabled)
		vcpu->arch.cputm_start = get_tod_clock_fast();
2374
	vcpu->arch.sie_block->cputm = cputm;
2375
	raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
2376
	preempt_enable();
2377 2378
}

2379
/* update and get the cpu timer - can also be called from other VCPU threads */
2380 2381
__u64 kvm_s390_get_cpu_timer(struct kvm_vcpu *vcpu)
{
2382
	unsigned int seq;
2383 2384 2385 2386 2387
	__u64 value;

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

2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401
	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();
2402
	return value;
2403 2404
}

2405 2406
void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
2407

2408
	gmap_enable(vcpu->arch.enabled_gmap);
2409
	kvm_s390_set_cpuflags(vcpu, CPUSTAT_RUNNING);
2410
	if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
2411
		__start_cpu_timer_accounting(vcpu);
2412
	vcpu->cpu = cpu;
2413 2414 2415 2416
}

void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
2417
	vcpu->cpu = -1;
2418
	if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
2419
		__stop_cpu_timer_accounting(vcpu);
2420
	kvm_s390_clear_cpuflags(vcpu, CPUSTAT_RUNNING);
2421 2422
	vcpu->arch.enabled_gmap = gmap_get_enabled();
	gmap_disable(vcpu->arch.enabled_gmap);
2423

2424 2425 2426 2427 2428 2429 2430
}

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;
2431
	kvm_s390_set_prefix(vcpu, 0);
2432
	kvm_s390_set_cpu_timer(vcpu, 0);
2433 2434 2435 2436 2437
	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;
2438 2439 2440
	/* make sure the new fpc will be lazily loaded */
	save_fpu_regs();
	current->thread.fpu.fpc = 0;
2441
	vcpu->arch.sie_block->gbea = 1;
2442
	vcpu->arch.sie_block->pp = 0;
2443
	vcpu->arch.sie_block->fpf &= ~FPF_BPBC;
2444 2445
	vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
	kvm_clear_async_pf_completion_queue(vcpu);
2446 2447
	if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm))
		kvm_s390_vcpu_stop(vcpu);
2448
	kvm_s390_clear_local_irqs(vcpu);
2449 2450
}

2451
void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
2452
{
2453
	mutex_lock(&vcpu->kvm->lock);
2454
	preempt_disable();
2455
	vcpu->arch.sie_block->epoch = vcpu->kvm->arch.epoch;
2456
	vcpu->arch.sie_block->epdx = vcpu->kvm->arch.epdx;
2457
	preempt_enable();
2458
	mutex_unlock(&vcpu->kvm->lock);
2459
	if (!kvm_is_ucontrol(vcpu->kvm)) {
2460
		vcpu->arch.gmap = vcpu->kvm->arch.gmap;
2461
		sca_add_vcpu(vcpu);
2462
	}
2463 2464
	if (test_kvm_facility(vcpu->kvm, 74) || vcpu->kvm->arch.user_instr0)
		vcpu->arch.sie_block->ictl |= ICTL_OPEREXC;
2465 2466
	/* make vcpu_load load the right gmap on the first trigger */
	vcpu->arch.enabled_gmap = vcpu->arch.gmap;
2467 2468
}

2469 2470
static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu)
{
2471
	if (!test_kvm_facility(vcpu->kvm, 76))
2472 2473
		return;

2474 2475 2476 2477 2478 2479 2480
	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;

2481 2482 2483
	vcpu->arch.sie_block->crycbd = vcpu->kvm->arch.crypto.crycbd;
}

2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497
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;
	return 0;
}

2498 2499 2500 2501 2502
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;
2503
	if (test_kvm_facility(vcpu->kvm, 7))
2504
		vcpu->arch.sie_block->fac = (u32)(u64) model->fac_list;
2505 2506
}

2507 2508
int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
{
2509
	int rc = 0;
2510

2511 2512
	atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH |
						    CPUSTAT_SM |
2513 2514
						    CPUSTAT_STOPPED);

2515
	if (test_kvm_facility(vcpu->kvm, 78))
2516
		kvm_s390_set_cpuflags(vcpu, CPUSTAT_GED2);
2517
	else if (test_kvm_facility(vcpu->kvm, 8))
2518
		kvm_s390_set_cpuflags(vcpu, CPUSTAT_GED);
2519

2520 2521
	kvm_s390_vcpu_setup_model(vcpu);

2522 2523
	/* pgste_set_pte has special handling for !MACHINE_HAS_ESOP */
	if (MACHINE_HAS_ESOP)
2524
		vcpu->arch.sie_block->ecb |= ECB_HOSTPROTINT;
2525
	if (test_kvm_facility(vcpu->kvm, 9))
2526
		vcpu->arch.sie_block->ecb |= ECB_SRSI;
2527
	if (test_kvm_facility(vcpu->kvm, 73))
2528
		vcpu->arch.sie_block->ecb |= ECB_TE;
2529

2530
	if (test_kvm_facility(vcpu->kvm, 8) && vcpu->kvm->arch.use_pfmfi)
2531
		vcpu->arch.sie_block->ecb2 |= ECB2_PFMFI;
2532
	if (test_kvm_facility(vcpu->kvm, 130))
2533 2534
		vcpu->arch.sie_block->ecb2 |= ECB2_IEP;
	vcpu->arch.sie_block->eca = ECA_MVPGI | ECA_PROTEXCI;
2535
	if (sclp.has_cei)
2536
		vcpu->arch.sie_block->eca |= ECA_CEI;
2537
	if (sclp.has_ib)
2538
		vcpu->arch.sie_block->eca |= ECA_IB;
2539
	if (sclp.has_siif)
2540
		vcpu->arch.sie_block->eca |= ECA_SII;
2541
	if (sclp.has_sigpif)
2542
		vcpu->arch.sie_block->eca |= ECA_SIGPI;
2543
	if (test_kvm_facility(vcpu->kvm, 129)) {
2544 2545
		vcpu->arch.sie_block->eca |= ECA_VX;
		vcpu->arch.sie_block->ecd |= ECD_HOSTREGMGMT;
2546
	}
2547 2548 2549
	if (test_kvm_facility(vcpu->kvm, 139))
		vcpu->arch.sie_block->ecd |= ECD_MEF;

2550 2551 2552 2553 2554
	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
2555 2556
	vcpu->arch.sie_block->sdnxo = ((unsigned long) &vcpu->run->s.regs.sdnx)
					| SDNXC;
2557
	vcpu->arch.sie_block->riccbd = (unsigned long) &vcpu->run->s.regs.riccb;
2558 2559

	if (sclp.has_kss)
2560
		kvm_s390_set_cpuflags(vcpu, CPUSTAT_KSS);
2561 2562
	else
		vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE;
2563

2564
	if (vcpu->kvm->arch.use_cmma) {
2565 2566 2567
		rc = kvm_s390_vcpu_setup_cmma(vcpu);
		if (rc)
			return rc;
2568
	}
2569
	hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
2570
	vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup;
2571

2572 2573
	kvm_s390_vcpu_crypto_setup(vcpu);

2574
	return rc;
2575 2576 2577 2578 2579
}

struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
				      unsigned int id)
{
2580
	struct kvm_vcpu *vcpu;
2581
	struct sie_page *sie_page;
2582 2583
	int rc = -EINVAL;

2584
	if (!kvm_is_ucontrol(kvm) && !sca_can_add_vcpu(kvm, id))
2585 2586 2587
		goto out;

	rc = -ENOMEM;
2588

2589
	vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
2590
	if (!vcpu)
2591
		goto out;
2592

2593
	BUILD_BUG_ON(sizeof(struct sie_page) != 4096);
2594 2595
	sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
	if (!sie_page)
2596 2597
		goto out_free_cpu;

2598 2599 2600
	vcpu->arch.sie_block = &sie_page->sie_block;
	vcpu->arch.sie_block->itdba = (unsigned long) &sie_page->itdb;

2601 2602 2603 2604
	/* the real guest size will always be smaller than msl */
	vcpu->arch.sie_block->mso = 0;
	vcpu->arch.sie_block->msl = sclp.hamax;

2605
	vcpu->arch.sie_block->icpua = id;
2606
	spin_lock_init(&vcpu->arch.local_int.lock);
2607
	vcpu->arch.sie_block->gd = (u32)(u64)kvm->arch.gisa;
2608 2609
	if (vcpu->arch.sie_block->gd && sclp.has_gisaf)
		vcpu->arch.sie_block->gd |= GISA_FORMAT1;
2610
	seqcount_init(&vcpu->arch.cputm_seqcount);
2611

2612 2613
	rc = kvm_vcpu_init(vcpu, kvm, id);
	if (rc)
2614
		goto out_free_sie_block;
2615
	VM_EVENT(kvm, 3, "create cpu %d at 0x%pK, sie block at 0x%pK", id, vcpu,
2616
		 vcpu->arch.sie_block);
2617
	trace_kvm_s390_create_vcpu(id, vcpu, vcpu->arch.sie_block);
2618 2619

	return vcpu;
2620 2621
out_free_sie_block:
	free_page((unsigned long)(vcpu->arch.sie_block));
2622
out_free_cpu:
2623
	kmem_cache_free(kvm_vcpu_cache, vcpu);
2624
out:
2625 2626 2627 2628 2629
	return ERR_PTR(rc);
}

int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
{
2630
	return kvm_s390_vcpu_has_irq(vcpu, 0);
2631 2632
}

2633 2634
bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
{
2635
	return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE);
2636 2637
}

2638
void kvm_s390_vcpu_block(struct kvm_vcpu *vcpu)
2639
{
2640
	atomic_or(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
2641
	exit_sie(vcpu);
2642 2643
}

2644
void kvm_s390_vcpu_unblock(struct kvm_vcpu *vcpu)
2645
{
2646
	atomic_andnot(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
2647 2648
}

2649 2650
static void kvm_s390_vcpu_request(struct kvm_vcpu *vcpu)
{
2651
	atomic_or(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
2652
	exit_sie(vcpu);
2653 2654 2655 2656
}

static void kvm_s390_vcpu_request_handled(struct kvm_vcpu *vcpu)
{
2657
	atomic_andnot(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
2658 2659
}

2660 2661 2662 2663 2664 2665
/*
 * 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)
{
2666
	kvm_s390_set_cpuflags(vcpu, CPUSTAT_STOP_INT);
2667 2668 2669 2670
	while (vcpu->arch.sie_block->prog0c & PROG_IN_SIE)
		cpu_relax();
}

2671 2672
/* Kick a guest cpu out of SIE to process a request synchronously */
void kvm_s390_sync_request(int req, struct kvm_vcpu *vcpu)
2673
{
2674 2675
	kvm_make_request(req, vcpu);
	kvm_s390_vcpu_request(vcpu);
2676 2677
}

2678 2679
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start,
			      unsigned long end)
2680 2681 2682
{
	struct kvm *kvm = gmap->private;
	struct kvm_vcpu *vcpu;
2683 2684
	unsigned long prefix;
	int i;
2685

2686 2687
	if (gmap_is_shadow(gmap))
		return;
2688 2689 2690
	if (start >= 1UL << 31)
		/* We are only interested in prefix pages */
		return;
2691 2692
	kvm_for_each_vcpu(i, vcpu, kvm) {
		/* match against both prefix pages */
2693 2694 2695 2696
		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);
2697
			kvm_s390_sync_request(KVM_REQ_MMU_RELOAD, vcpu);
2698 2699 2700 2701
		}
	}
}

2702 2703 2704 2705 2706 2707 2708
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
	/* kvm common code refers to this, but never calls it */
	BUG();
	return 0;
}

2709 2710 2711 2712 2713 2714
static int kvm_arch_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu,
					   struct kvm_one_reg *reg)
{
	int r = -EINVAL;

	switch (reg->id) {
2715 2716 2717 2718 2719 2720 2721 2722
	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;
2723
	case KVM_REG_S390_CPU_TIMER:
2724
		r = put_user(kvm_s390_get_cpu_timer(vcpu),
2725 2726 2727 2728 2729 2730
			     (u64 __user *)reg->addr);
		break;
	case KVM_REG_S390_CLOCK_COMP:
		r = put_user(vcpu->arch.sie_block->ckc,
			     (u64 __user *)reg->addr);
		break;
2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742
	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;
2743 2744 2745 2746
	case KVM_REG_S390_PP:
		r = put_user(vcpu->arch.sie_block->pp,
			     (u64 __user *)reg->addr);
		break;
2747 2748 2749 2750
	case KVM_REG_S390_GBEA:
		r = put_user(vcpu->arch.sie_block->gbea,
			     (u64 __user *)reg->addr);
		break;
2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761
	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;
2762
	__u64 val;
2763 2764

	switch (reg->id) {
2765 2766 2767 2768 2769 2770 2771 2772
	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;
2773
	case KVM_REG_S390_CPU_TIMER:
2774 2775 2776
		r = get_user(val, (u64 __user *)reg->addr);
		if (!r)
			kvm_s390_set_cpu_timer(vcpu, val);
2777 2778 2779 2780 2781
		break;
	case KVM_REG_S390_CLOCK_COMP:
		r = get_user(vcpu->arch.sie_block->ckc,
			     (u64 __user *)reg->addr);
		break;
2782 2783 2784
	case KVM_REG_S390_PFTOKEN:
		r = get_user(vcpu->arch.pfault_token,
			     (u64 __user *)reg->addr);
2785 2786
		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
			kvm_clear_async_pf_completion_queue(vcpu);
2787 2788 2789 2790 2791 2792 2793 2794 2795
		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;
2796 2797 2798 2799
	case KVM_REG_S390_PP:
		r = get_user(vcpu->arch.sie_block->pp,
			     (u64 __user *)reg->addr);
		break;
2800 2801 2802 2803
	case KVM_REG_S390_GBEA:
		r = get_user(vcpu->arch.sie_block->gbea,
			     (u64 __user *)reg->addr);
		break;
2804 2805 2806 2807 2808 2809
	default:
		break;
	}

	return r;
}
2810

2811 2812 2813 2814 2815 2816 2817 2818
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)
{
2819
	vcpu_load(vcpu);
2820
	memcpy(&vcpu->run->s.regs.gprs, &regs->gprs, sizeof(regs->gprs));
2821
	vcpu_put(vcpu);
2822 2823 2824 2825 2826
	return 0;
}

int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
2827
	vcpu_load(vcpu);
2828
	memcpy(&regs->gprs, &vcpu->run->s.regs.gprs, sizeof(regs->gprs));
2829
	vcpu_put(vcpu);
2830 2831 2832 2833 2834 2835
	return 0;
}

int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
2836 2837
	vcpu_load(vcpu);

2838
	memcpy(&vcpu->run->s.regs.acrs, &sregs->acrs, sizeof(sregs->acrs));
2839
	memcpy(&vcpu->arch.sie_block->gcr, &sregs->crs, sizeof(sregs->crs));
2840 2841

	vcpu_put(vcpu);
2842 2843 2844 2845 2846 2847
	return 0;
}

int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
				  struct kvm_sregs *sregs)
{
2848 2849
	vcpu_load(vcpu);

2850
	memcpy(&sregs->acrs, &vcpu->run->s.regs.acrs, sizeof(sregs->acrs));
2851
	memcpy(&sregs->crs, &vcpu->arch.sie_block->gcr, sizeof(sregs->crs));
2852 2853

	vcpu_put(vcpu);
2854 2855 2856 2857 2858
	return 0;
}

int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
2859 2860 2861 2862 2863 2864 2865 2866
	int ret = 0;

	vcpu_load(vcpu);

	if (test_fp_ctl(fpu->fpc)) {
		ret = -EINVAL;
		goto out;
	}
2867
	vcpu->run->s.regs.fpc = fpu->fpc;
2868
	if (MACHINE_HAS_VX)
2869 2870
		convert_fp_to_vx((__vector128 *) vcpu->run->s.regs.vrs,
				 (freg_t *) fpu->fprs);
2871
	else
2872
		memcpy(vcpu->run->s.regs.fprs, &fpu->fprs, sizeof(fpu->fprs));
2873 2874 2875 2876

out:
	vcpu_put(vcpu);
	return ret;
2877 2878 2879 2880
}

int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
2881 2882
	vcpu_load(vcpu);

2883 2884 2885
	/* make sure we have the latest values */
	save_fpu_regs();
	if (MACHINE_HAS_VX)
2886 2887
		convert_vx_to_fp((freg_t *) fpu->fprs,
				 (__vector128 *) vcpu->run->s.regs.vrs);
2888
	else
2889
		memcpy(fpu->fprs, vcpu->run->s.regs.fprs, sizeof(fpu->fprs));
2890
	fpu->fpc = vcpu->run->s.regs.fpc;
2891 2892

	vcpu_put(vcpu);
2893 2894 2895 2896 2897 2898 2899
	return 0;
}

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

2900
	if (!is_vcpu_stopped(vcpu))
2901
		rc = -EBUSY;
2902 2903 2904 2905
	else {
		vcpu->run->psw_mask = psw.mask;
		vcpu->run->psw_addr = psw.addr;
	}
2906 2907 2908 2909 2910 2911 2912 2913 2914
	return rc;
}

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

2915 2916 2917 2918
#define VALID_GUESTDBG_FLAGS (KVM_GUESTDBG_SINGLESTEP | \
			      KVM_GUESTDBG_USE_HW_BP | \
			      KVM_GUESTDBG_ENABLE)

J
Jan Kiszka 已提交
2919 2920
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
					struct kvm_guest_debug *dbg)
2921
{
2922 2923
	int rc = 0;

2924 2925
	vcpu_load(vcpu);

2926 2927 2928
	vcpu->guest_debug = 0;
	kvm_s390_clear_bp_data(vcpu);

2929 2930 2931 2932 2933 2934 2935 2936
	if (dbg->control & ~VALID_GUESTDBG_FLAGS) {
		rc = -EINVAL;
		goto out;
	}
	if (!sclp.has_gpere) {
		rc = -EINVAL;
		goto out;
	}
2937 2938 2939 2940

	if (dbg->control & KVM_GUESTDBG_ENABLE) {
		vcpu->guest_debug = dbg->control;
		/* enforce guest PER */
2941
		kvm_s390_set_cpuflags(vcpu, CPUSTAT_P);
2942 2943 2944 2945

		if (dbg->control & KVM_GUESTDBG_USE_HW_BP)
			rc = kvm_s390_import_bp_data(vcpu, dbg);
	} else {
2946
		kvm_s390_clear_cpuflags(vcpu, CPUSTAT_P);
2947 2948 2949 2950 2951 2952
		vcpu->arch.guestdbg.last_bp = 0;
	}

	if (rc) {
		vcpu->guest_debug = 0;
		kvm_s390_clear_bp_data(vcpu);
2953
		kvm_s390_clear_cpuflags(vcpu, CPUSTAT_P);
2954 2955
	}

2956 2957
out:
	vcpu_put(vcpu);
2958
	return rc;
2959 2960
}

2961 2962 2963
int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
2964 2965 2966 2967
	int ret;

	vcpu_load(vcpu);

2968
	/* CHECK_STOP and LOAD are not supported yet */
2969 2970 2971 2972 2973
	ret = is_vcpu_stopped(vcpu) ? KVM_MP_STATE_STOPPED :
				      KVM_MP_STATE_OPERATING;

	vcpu_put(vcpu);
	return ret;
2974 2975 2976 2977 2978
}

int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
				    struct kvm_mp_state *mp_state)
{
2979 2980
	int rc = 0;

2981 2982
	vcpu_load(vcpu);

2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999
	/* 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;
	}

3000
	vcpu_put(vcpu);
3001
	return rc;
3002 3003
}

3004 3005
static bool ibs_enabled(struct kvm_vcpu *vcpu)
{
3006
	return kvm_s390_test_cpuflags(vcpu, CPUSTAT_IBS);
3007 3008
}

3009 3010
static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu)
{
3011
retry:
3012
	kvm_s390_vcpu_request_handled(vcpu);
R
Radim Krčmář 已提交
3013
	if (!kvm_request_pending(vcpu))
3014
		return 0;
3015 3016
	/*
	 * We use MMU_RELOAD just to re-arm the ipte notifier for the
3017
	 * guest prefix page. gmap_mprotect_notify will wait on the ptl lock.
3018 3019 3020 3021
	 * 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.
	 */
3022
	if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) {
3023
		int rc;
3024 3025 3026
		rc = gmap_mprotect_notify(vcpu->arch.gmap,
					  kvm_s390_get_prefix(vcpu),
					  PAGE_SIZE * 2, PROT_WRITE);
3027 3028
		if (rc) {
			kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu);
3029
			return rc;
3030
		}
3031
		goto retry;
3032
	}
3033

3034 3035 3036 3037 3038
	if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) {
		vcpu->arch.sie_block->ihcpu = 0xffff;
		goto retry;
	}

3039 3040 3041
	if (kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu)) {
		if (!ibs_enabled(vcpu)) {
			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 1);
3042
			kvm_s390_set_cpuflags(vcpu, CPUSTAT_IBS);
3043 3044
		}
		goto retry;
3045
	}
3046 3047 3048 3049

	if (kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu)) {
		if (ibs_enabled(vcpu)) {
			trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 0);
3050
			kvm_s390_clear_cpuflags(vcpu, CPUSTAT_IBS);
3051 3052 3053 3054
		}
		goto retry;
	}

3055 3056 3057 3058 3059
	if (kvm_check_request(KVM_REQ_ICPT_OPEREXC, vcpu)) {
		vcpu->arch.sie_block->ictl |= ICTL_OPEREXC;
		goto retry;
	}

3060 3061
	if (kvm_check_request(KVM_REQ_START_MIGRATION, vcpu)) {
		/*
3062
		 * Disable CMM virtualization; we will emulate the ESSA
3063 3064 3065 3066 3067 3068 3069 3070 3071
		 * 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)) {
		/*
3072 3073
		 * Re-enable CMM virtualization if CMMA is available and
		 * CMM has been used.
3074 3075
		 */
		if ((vcpu->kvm->arch.use_cmma) &&
3076
		    (vcpu->kvm->mm->context.uses_cmm))
3077 3078 3079 3080
			vcpu->arch.sie_block->ecb2 |= ECB2_CMMA;
		goto retry;
	}

3081
	/* nothing to do, just clear the request */
3082
	kvm_clear_request(KVM_REQ_UNHALT, vcpu);
3083

3084 3085 3086
	return 0;
}

3087 3088
void kvm_s390_set_tod_clock(struct kvm *kvm,
			    const struct kvm_s390_vm_tod_clock *gtod)
3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099
{
	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;
3100 3101 3102 3103 3104 3105
	kvm->arch.epdx = 0;
	if (test_kvm_facility(kvm, 139)) {
		kvm->arch.epdx = gtod->epoch_idx - htod.epoch_idx;
		if (kvm->arch.epoch > gtod->tod)
			kvm->arch.epdx -= 1;
	}
3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117

	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);
}

3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128
/**
 * 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)
3129
{
3130 3131
	return gmap_fault(vcpu->arch.gmap, gpa,
			  writable ? FAULT_FLAG_WRITE : 0);
3132 3133
}

3134 3135 3136 3137
static void __kvm_inject_pfault_token(struct kvm_vcpu *vcpu, bool start_token,
				      unsigned long token)
{
	struct kvm_s390_interrupt inti;
3138
	struct kvm_s390_irq irq;
3139 3140

	if (start_token) {
3141 3142 3143
		irq.u.ext.ext_params2 = token;
		irq.type = KVM_S390_INT_PFAULT_INIT;
		WARN_ON_ONCE(kvm_s390_inject_vcpu(vcpu, &irq));
3144 3145
	} else {
		inti.type = KVM_S390_INT_PFAULT_DONE;
3146
		inti.parm64 = token;
3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192
		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;
3193
	if (kvm_s390_vcpu_has_irq(vcpu, 0))
3194 3195 3196 3197 3198 3199
		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
3200 3201 3202
	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))
3203 3204 3205 3206 3207 3208
		return 0;

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

3209
static int vcpu_pre_run(struct kvm_vcpu *vcpu)
3210
{
3211
	int rc, cpuflags;
3212

3213 3214 3215 3216 3217 3218 3219
	/*
	 * 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);

3220 3221
	vcpu->arch.sie_block->gg14 = vcpu->run->s.regs.gprs[14];
	vcpu->arch.sie_block->gg15 = vcpu->run->s.regs.gprs[15];
3222 3223 3224 3225

	if (need_resched())
		schedule();

3226
	if (test_cpu_flag(CIF_MCCK_PENDING))
3227 3228
		s390_handle_mcck();

3229 3230 3231 3232 3233
	if (!kvm_is_ucontrol(vcpu->kvm)) {
		rc = kvm_s390_deliver_pending_interrupts(vcpu);
		if (rc)
			return rc;
	}
C
Carsten Otte 已提交
3234

3235 3236 3237 3238
	rc = kvm_s390_handle_requests(vcpu);
	if (rc)
		return rc;

3239 3240 3241 3242 3243
	if (guestdbg_enabled(vcpu)) {
		kvm_s390_backup_guest_per_regs(vcpu);
		kvm_s390_patch_guest_per_regs(vcpu);
	}

3244
	vcpu->arch.sie_block->icptcode = 0;
3245 3246 3247
	cpuflags = atomic_read(&vcpu->arch.sie_block->cpuflags);
	VCPU_EVENT(vcpu, 6, "entering sie flags %x", cpuflags);
	trace_kvm_s390_sie_enter(vcpu, cpuflags);
3248

3249 3250 3251
	return 0;
}

3252 3253
static int vcpu_post_run_fault_in_sie(struct kvm_vcpu *vcpu)
{
3254 3255 3256 3257
	struct kvm_s390_pgm_info pgm_info = {
		.code = PGM_ADDRESSING,
	};
	u8 opcode, ilen;
3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270
	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.
	 */
3271
	rc = read_guest_instr(vcpu, vcpu->arch.sie_block->gpsw.addr, &opcode, 1);
3272
	ilen = insn_length(opcode);
3273 3274 3275 3276 3277 3278 3279 3280 3281 3282
	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;
	}
3283 3284 3285
	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);
3286 3287
}

3288 3289
static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason)
{
3290 3291 3292
	struct mcck_volatile_info *mcck_info;
	struct sie_page *sie_page;

3293 3294 3295 3296
	VCPU_EVENT(vcpu, 6, "exit sie icptcode %d",
		   vcpu->arch.sie_block->icptcode);
	trace_kvm_s390_sie_exit(vcpu, vcpu->arch.sie_block->icptcode);

3297 3298 3299
	if (guestdbg_enabled(vcpu))
		kvm_s390_restore_guest_per_regs(vcpu);

3300 3301
	vcpu->run->s.regs.gprs[14] = vcpu->arch.sie_block->gg14;
	vcpu->run->s.regs.gprs[15] = vcpu->arch.sie_block->gg15;
3302

3303 3304 3305 3306 3307 3308 3309 3310 3311
	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;
	}

3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324
	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;
3325 3326 3327 3328 3329
	} 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;
3330
		return -EREMOTE;
3331
	} else if (current->thread.gmap_pfault) {
3332
		trace_kvm_s390_major_guest_pfault(vcpu);
3333
		current->thread.gmap_pfault = 0;
3334 3335 3336
		if (kvm_arch_setup_async_pf(vcpu))
			return 0;
		return kvm_arch_fault_in_page(vcpu, current->thread.gmap_addr, 1);
3337
	}
3338
	return vcpu_post_run_fault_in_sie(vcpu);
3339 3340 3341 3342 3343 3344
}

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

3345 3346 3347 3348 3349 3350
	/*
	 * 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);

3351 3352 3353 3354
	do {
		rc = vcpu_pre_run(vcpu);
		if (rc)
			break;
3355

3356
		srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
3357 3358 3359 3360
		/*
		 * As PF_VCPU will be used in fault handler, between
		 * guest_enter and guest_exit should be no uaccess.
		 */
3361
		local_irq_disable();
3362
		guest_enter_irqoff();
3363
		__disable_cpu_timer_accounting(vcpu);
3364
		local_irq_enable();
3365 3366
		exit_reason = sie64a(vcpu->arch.sie_block,
				     vcpu->run->s.regs.gprs);
3367
		local_irq_disable();
3368
		__enable_cpu_timer_accounting(vcpu);
3369
		guest_exit_irqoff();
3370
		local_irq_enable();
3371
		vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
3372 3373

		rc = vcpu_post_run(vcpu, exit_reason);
3374
	} while (!signal_pending(current) && !guestdbg_exit_pending(vcpu) && !rc);
3375

3376
	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
3377
	return rc;
3378 3379
}

3380 3381
static void sync_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
3382
	struct runtime_instr_cb *riccb;
F
Fan Zhang 已提交
3383
	struct gs_cb *gscb;
3384 3385

	riccb = (struct runtime_instr_cb *) &kvm_run->s.regs.riccb;
F
Fan Zhang 已提交
3386
	gscb = (struct gs_cb *) &kvm_run->s.regs.gscb;
3387 3388 3389 3390 3391 3392
	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);
3393 3394
		/* some control register changes require a tlb flush */
		kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
3395 3396
	}
	if (kvm_run->kvm_dirty_regs & KVM_SYNC_ARCH0) {
3397
		kvm_s390_set_cpu_timer(vcpu, kvm_run->s.regs.cputm);
3398 3399 3400 3401 3402 3403 3404 3405 3406
		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;
3407 3408
		if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
			kvm_clear_async_pf_completion_queue(vcpu);
3409
	}
F
Fan Zhang 已提交
3410 3411 3412 3413 3414
	/*
	 * 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) &&
3415
	    test_kvm_facility(vcpu->kvm, 64) &&
3416
	    riccb->v &&
3417
	    !(vcpu->arch.sie_block->ecb3 & ECB3_RI)) {
3418
		VCPU_EVENT(vcpu, 3, "%s", "ENABLE: RI (sync_regs)");
3419
		vcpu->arch.sie_block->ecb3 |= ECB3_RI;
F
Fan Zhang 已提交
3420
	}
F
Fan Zhang 已提交
3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432
	/*
	 * 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
3433
	}
3434 3435 3436 3437 3438
	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;
	}
3439 3440
	save_access_regs(vcpu->arch.host_acrs);
	restore_access_regs(vcpu->run->s.regs.acrs);
3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452
	/* 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
3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466
	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
3467

3468 3469 3470 3471 3472 3473 3474 3475 3476
	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);
3477
	kvm_run->s.regs.cputm = kvm_s390_get_cpu_timer(vcpu);
3478 3479 3480 3481 3482 3483 3484
	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;
3485
	kvm_run->s.regs.bpbc = (vcpu->arch.sie_block->fpf & FPF_BPBC) == FPF_BPBC;
3486 3487
	save_access_regs(vcpu->run->s.regs.acrs);
	restore_access_regs(vcpu->arch.host_acrs);
3488 3489 3490 3491 3492 3493
	/* 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
3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505
	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;
	}
3506

3507 3508
}

3509 3510
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
3511
	int rc;
3512

3513 3514 3515
	if (kvm_run->immediate_exit)
		return -EINTR;

3516 3517
	vcpu_load(vcpu);

3518 3519
	if (guestdbg_exit_pending(vcpu)) {
		kvm_s390_prepare_debug_exit(vcpu);
3520 3521
		rc = 0;
		goto out;
3522 3523
	}

3524
	kvm_sigset_activate(vcpu);
3525

3526 3527 3528
	if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) {
		kvm_s390_vcpu_start(vcpu);
	} else if (is_vcpu_stopped(vcpu)) {
3529
		pr_err_ratelimited("can't run stopped vcpu %d\n",
3530
				   vcpu->vcpu_id);
3531 3532
		rc = -EINVAL;
		goto out;
3533
	}
3534

3535
	sync_regs(vcpu, kvm_run);
3536
	enable_cpu_timer_accounting(vcpu);
3537

3538
	might_fault();
3539
	rc = __vcpu_run(vcpu);
3540

3541 3542
	if (signal_pending(current) && !rc) {
		kvm_run->exit_reason = KVM_EXIT_INTR;
3543
		rc = -EINTR;
3544
	}
3545

3546 3547 3548 3549 3550
	if (guestdbg_exit_pending(vcpu) && !rc)  {
		kvm_s390_prepare_debug_exit(vcpu);
		rc = 0;
	}

3551
	if (rc == -EREMOTE) {
3552
		/* userspace support is needed, kvm_run has been prepared */
3553 3554
		rc = 0;
	}
3555

3556
	disable_cpu_timer_accounting(vcpu);
3557
	store_regs(vcpu, kvm_run);
3558

3559
	kvm_sigset_deactivate(vcpu);
3560 3561

	vcpu->stat.exit_userspace++;
3562 3563
out:
	vcpu_put(vcpu);
3564
	return rc;
3565 3566 3567 3568 3569 3570 3571 3572
}

/*
 * 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
 */
3573
int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long gpa)
3574
{
3575
	unsigned char archmode = 1;
3576
	freg_t fprs[NUM_FPRS];
3577
	unsigned int px;
3578
	u64 clkcomp, cputm;
3579
	int rc;
3580

3581
	px = kvm_s390_get_prefix(vcpu);
3582 3583
	if (gpa == KVM_S390_STORE_STATUS_NOADDR) {
		if (write_guest_abs(vcpu, 163, &archmode, 1))
3584
			return -EFAULT;
3585
		gpa = 0;
3586 3587
	} else if (gpa == KVM_S390_STORE_STATUS_PREFIXED) {
		if (write_guest_real(vcpu, 163, &archmode, 1))
3588
			return -EFAULT;
3589 3590 3591
		gpa = px;
	} else
		gpa -= __LC_FPREGS_SAVE_AREA;
3592 3593 3594

	/* manually convert vector registers if necessary */
	if (MACHINE_HAS_VX) {
3595
		convert_vx_to_fp(fprs, (__vector128 *) vcpu->run->s.regs.vrs);
3596 3597 3598 3599
		rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
				     fprs, 128);
	} else {
		rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
3600
				     vcpu->run->s.regs.fprs, 128);
3601
	}
3602
	rc |= write_guest_abs(vcpu, gpa + __LC_GPREGS_SAVE_AREA,
3603
			      vcpu->run->s.regs.gprs, 128);
3604
	rc |= write_guest_abs(vcpu, gpa + __LC_PSW_SAVE_AREA,
3605
			      &vcpu->arch.sie_block->gpsw, 16);
3606
	rc |= write_guest_abs(vcpu, gpa + __LC_PREFIX_SAVE_AREA,
3607
			      &px, 4);
3608
	rc |= write_guest_abs(vcpu, gpa + __LC_FP_CREG_SAVE_AREA,
3609
			      &vcpu->run->s.regs.fpc, 4);
3610
	rc |= write_guest_abs(vcpu, gpa + __LC_TOD_PROGREG_SAVE_AREA,
3611
			      &vcpu->arch.sie_block->todpr, 4);
3612
	cputm = kvm_s390_get_cpu_timer(vcpu);
3613
	rc |= write_guest_abs(vcpu, gpa + __LC_CPU_TIMER_SAVE_AREA,
3614
			      &cputm, 8);
3615
	clkcomp = vcpu->arch.sie_block->ckc >> 8;
3616
	rc |= write_guest_abs(vcpu, gpa + __LC_CLOCK_COMP_SAVE_AREA,
3617
			      &clkcomp, 8);
3618
	rc |= write_guest_abs(vcpu, gpa + __LC_AREGS_SAVE_AREA,
3619
			      &vcpu->run->s.regs.acrs, 64);
3620
	rc |= write_guest_abs(vcpu, gpa + __LC_CREGS_SAVE_AREA,
3621 3622
			      &vcpu->arch.sie_block->gcr, 128);
	return rc ? -EFAULT : 0;
3623 3624
}

3625 3626 3627 3628
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
3629
	 * switch in the run ioctl. Let's update our copies before we save
3630 3631
	 * it into the save area
	 */
3632
	save_fpu_regs();
3633
	vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
3634 3635 3636 3637 3638
	save_access_regs(vcpu->run->s.regs.acrs);

	return kvm_s390_store_status_unloaded(vcpu, addr);
}

3639 3640 3641
static void __disable_ibs_on_vcpu(struct kvm_vcpu *vcpu)
{
	kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu);
3642
	kvm_s390_sync_request(KVM_REQ_DISABLE_IBS, vcpu);
3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656
}

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)
{
3657 3658
	if (!sclp.has_ibs)
		return;
3659
	kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu);
3660
	kvm_s390_sync_request(KVM_REQ_ENABLE_IBS, vcpu);
3661 3662
}

3663 3664
void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu)
{
3665 3666 3667 3668 3669
	int i, online_vcpus, started_vcpus = 0;

	if (!is_vcpu_stopped(vcpu))
		return;

3670
	trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 1);
3671
	/* Only one cpu at a time may enter/leave the STOPPED state. */
3672
	spin_lock(&vcpu->kvm->arch.start_stop_lock);
3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691
	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);
	}

3692
	kvm_s390_clear_cpuflags(vcpu, CPUSTAT_STOPPED);
3693 3694 3695 3696
	/*
	 * Another VCPU might have used IBS while we were offline.
	 * Let's play safe and flush the VCPU at startup.
	 */
3697
	kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
3698
	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
3699
	return;
3700 3701 3702 3703
}

void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu)
{
3704 3705 3706 3707 3708 3709
	int i, online_vcpus, started_vcpus = 0;
	struct kvm_vcpu *started_vcpu = NULL;

	if (is_vcpu_stopped(vcpu))
		return;

3710
	trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 0);
3711
	/* Only one cpu at a time may enter/leave the STOPPED state. */
3712
	spin_lock(&vcpu->kvm->arch.start_stop_lock);
3713 3714
	online_vcpus = atomic_read(&vcpu->kvm->online_vcpus);

3715
	/* SIGP STOP and SIGP STOP AND STORE STATUS has been fully processed */
3716
	kvm_s390_clear_stop_irq(vcpu);
3717

3718
	kvm_s390_set_cpuflags(vcpu, CPUSTAT_STOPPED);
3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735
	__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);
	}

3736
	spin_unlock(&vcpu->kvm->arch.start_stop_lock);
3737
	return;
3738 3739
}

3740 3741 3742 3743 3744 3745 3746 3747 3748
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) {
3749 3750 3751
	case KVM_CAP_S390_CSS_SUPPORT:
		if (!vcpu->kvm->arch.css_support) {
			vcpu->kvm->arch.css_support = 1;
3752
			VM_EVENT(vcpu->kvm, 3, "%s", "ENABLE: CSS support");
3753 3754 3755 3756
			trace_kvm_s390_enable_css(vcpu->kvm);
		}
		r = 0;
		break;
3757 3758 3759 3760 3761 3762 3763
	default:
		r = -EINVAL;
		break;
	}
	return r;
}

3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789
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) {
3790 3791
			r = check_gva_range(vcpu, mop->gaddr, mop->ar,
					    mop->size, GACC_FETCH);
3792 3793 3794 3795 3796 3797 3798 3799 3800 3801
			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) {
3802 3803
			r = check_gva_range(vcpu, mop->gaddr, mop->ar,
					    mop->size, GACC_STORE);
3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824
			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;
}

3825 3826
long kvm_arch_vcpu_async_ioctl(struct file *filp,
			       unsigned int ioctl, unsigned long arg)
3827 3828 3829 3830
{
	struct kvm_vcpu *vcpu = filp->private_data;
	void __user *argp = (void __user *)arg;

3831
	switch (ioctl) {
3832 3833 3834 3835
	case KVM_S390_IRQ: {
		struct kvm_s390_irq s390irq;

		if (copy_from_user(&s390irq, argp, sizeof(s390irq)))
3836 3837
			return -EFAULT;
		return kvm_s390_inject_vcpu(vcpu, &s390irq);
3838
	}
3839
	case KVM_S390_INTERRUPT: {
3840
		struct kvm_s390_interrupt s390int;
3841
		struct kvm_s390_irq s390irq;
3842 3843

		if (copy_from_user(&s390int, argp, sizeof(s390int)))
3844
			return -EFAULT;
3845 3846
		if (s390int_to_s390irq(&s390int, &s390irq))
			return -EINVAL;
3847
		return kvm_s390_inject_vcpu(vcpu, &s390irq);
3848
	}
3849
	}
3850 3851 3852 3853 3854 3855 3856 3857 3858 3859
	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;
3860 3861 3862 3863

	vcpu_load(vcpu);

	switch (ioctl) {
3864
	case KVM_S390_STORE_STATUS:
3865
		idx = srcu_read_lock(&vcpu->kvm->srcu);
3866
		r = kvm_s390_vcpu_store_status(vcpu, arg);
3867
		srcu_read_unlock(&vcpu->kvm->srcu, idx);
3868
		break;
3869 3870 3871
	case KVM_S390_SET_INITIAL_PSW: {
		psw_t psw;

3872
		r = -EFAULT;
3873
		if (copy_from_user(&psw, argp, sizeof(psw)))
3874 3875 3876
			break;
		r = kvm_arch_vcpu_ioctl_set_initial_psw(vcpu, psw);
		break;
3877 3878
	}
	case KVM_S390_INITIAL_RESET:
3879 3880
		r = kvm_arch_vcpu_ioctl_initial_reset(vcpu);
		break;
3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892
	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;
	}
3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928
#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
3929
	case KVM_S390_VCPU_FAULT: {
3930
		r = gmap_fault(vcpu->arch.gmap, arg, 0);
3931 3932
		break;
	}
3933 3934 3935 3936 3937 3938 3939 3940 3941
	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;
	}
3942 3943 3944 3945 3946 3947 3948 3949 3950
	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;
	}
3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962
	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;
		}
3963
		/* do not use irq_state.flags, it will break old QEMUs */
3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978
		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;
		}
3979
		/* do not use irq_state.flags, it will break old QEMUs */
3980 3981 3982 3983 3984
		r = kvm_s390_get_irq_state(vcpu,
					   (__u8 __user *)  irq_state.buf,
					   irq_state.len);
		break;
	}
3985
	default:
3986
		r = -ENOTTY;
3987
	}
3988 3989

	vcpu_put(vcpu);
3990
	return r;
3991 3992
}

3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005
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;
}

4006 4007
int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
			    unsigned long npages)
4008 4009 4010 4011
{
	return 0;
}

4012
/* Section: memory related */
4013 4014
int kvm_arch_prepare_memory_region(struct kvm *kvm,
				   struct kvm_memory_slot *memslot,
4015
				   const struct kvm_userspace_memory_region *mem,
4016
				   enum kvm_mr_change change)
4017
{
4018 4019 4020 4021
	/* 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 */
4022

4023
	if (mem->userspace_addr & 0xffffful)
4024 4025
		return -EINVAL;

4026
	if (mem->memory_size & 0xffffful)
4027 4028
		return -EINVAL;

4029 4030 4031
	if (mem->guest_phys_addr + mem->memory_size > kvm->arch.mem_limit)
		return -EINVAL;

4032 4033 4034 4035
	return 0;
}

void kvm_arch_commit_memory_region(struct kvm *kvm,
4036
				const struct kvm_userspace_memory_region *mem,
4037
				const struct kvm_memory_slot *old,
4038
				const struct kvm_memory_slot *new,
4039
				enum kvm_mr_change change)
4040
{
4041
	int rc;
4042

4043 4044 4045 4046 4047 4048 4049 4050 4051 4052
	/* 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;
4053 4054 4055 4056

	rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr,
		mem->guest_phys_addr, mem->memory_size);
	if (rc)
4057
		pr_warn("failed to commit memory region\n");
4058
	return;
4059 4060
}

4061 4062 4063 4064 4065 4066 4067
static inline unsigned long nonhyp_mask(int i)
{
	unsigned int nonhyp_fai = (sclp.hmfai << i * 2) >> 30;

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

4068 4069 4070 4071 4072
void kvm_arch_vcpu_block_finish(struct kvm_vcpu *vcpu)
{
	vcpu->valid_wakeup = false;
}

4073 4074
static int __init kvm_s390_init(void)
{
4075 4076
	int i;

4077 4078 4079 4080 4081
	if (!sclp.has_sief2) {
		pr_info("SIE not available\n");
		return -ENODEV;
	}

4082
	for (i = 0; i < 16; i++)
4083
		kvm_s390_fac_base[i] |=
4084 4085
			S390_lowcore.stfle_fac_list[i] & nonhyp_mask(i);

4086
	return kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE);
4087 4088 4089 4090 4091 4092 4093 4094 4095
}

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

module_init(kvm_s390_init);
module_exit(kvm_s390_exit);
4096 4097 4098 4099 4100 4101 4102 4103 4104

/*
 * 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");