interrupt.c 53.0 KB
Newer Older
1
/*
2
 * handling kvm guest interrupts
3
 *
4
 * Copyright IBM Corp. 2008, 2015
5 6 7 8 9 10 11 12
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License (version 2 only)
 * as published by the Free Software Foundation.
 *
 *    Author(s): Carsten Otte <cotte@de.ibm.com>
 */

13
#include <linux/interrupt.h>
14
#include <linux/kvm_host.h>
15
#include <linux/hrtimer.h>
16
#include <linux/mmu_context.h>
17
#include <linux/signal.h>
18
#include <linux/slab.h>
19
#include <linux/bitmap.h>
20
#include <asm/asm-offsets.h>
21
#include <asm/dis.h>
22
#include <asm/uaccess.h>
23
#include <asm/sclp.h>
24 25
#include "kvm-s390.h"
#include "gaccess.h"
26
#include "trace-s390.h"
27

28 29 30 31
#define IOINT_SCHID_MASK 0x0000ffff
#define IOINT_SSID_MASK 0x00030000
#define IOINT_CSSID_MASK 0x03fc0000
#define IOINT_AI_MASK 0x04000000
32
#define PFAULT_INIT 0x0600
33 34
#define PFAULT_DONE 0x0680
#define VIRTIO_PARAM 0x0d00
35 36 37 38 39 40

static int is_ioint(u64 type)
{
	return ((type & 0xfffe0000u) != 0xfffe0000u);
}

41
int psw_extint_disabled(struct kvm_vcpu *vcpu)
42 43 44 45
{
	return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_EXT);
}

46 47 48 49 50
static int psw_ioint_disabled(struct kvm_vcpu *vcpu)
{
	return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_IO);
}

51 52 53 54 55
static int psw_mchk_disabled(struct kvm_vcpu *vcpu)
{
	return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_MCHECK);
}

56 57 58 59 60 61 62 63 64
static int psw_interrupts_disabled(struct kvm_vcpu *vcpu)
{
	if ((vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PER) ||
	    (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_IO) ||
	    (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_EXT))
		return 0;
	return 1;
}

65 66 67 68 69
static int ckc_interrupts_enabled(struct kvm_vcpu *vcpu)
{
	if (psw_extint_disabled(vcpu) ||
	    !(vcpu->arch.sie_block->gcr[0] & 0x800ul))
		return 0;
70 71 72
	if (guestdbg_enabled(vcpu) && guestdbg_sstep_enabled(vcpu))
		/* No timer interrupts when single stepping */
		return 0;
73 74 75
	return 1;
}

C
Cornelia Huck 已提交
76 77 78 79 80 81 82
static u64 int_word_to_isc_bits(u32 int_word)
{
	u8 isc = (int_word & 0x38000000) >> 27;

	return (0x80 >> isc) << 24;
}

83
static int __must_check __interrupt_is_deliverable(struct kvm_vcpu *vcpu,
84
				      struct kvm_s390_interrupt_info *inti)
85 86
{
	switch (inti->type) {
87 88 89 90 91
	case KVM_S390_INT_EXTERNAL_CALL:
		if (psw_extint_disabled(vcpu))
			return 0;
		if (vcpu->arch.sie_block->gcr[0] & 0x2000ul)
			return 1;
92
		return 0;
93 94 95 96 97 98
	case KVM_S390_INT_EMERGENCY:
		if (psw_extint_disabled(vcpu))
			return 0;
		if (vcpu->arch.sie_block->gcr[0] & 0x4000ul)
			return 1;
		return 0;
99 100 101 102 103 104 105 106
	case KVM_S390_INT_CLOCK_COMP:
		return ckc_interrupts_enabled(vcpu);
	case KVM_S390_INT_CPU_TIMER:
		if (psw_extint_disabled(vcpu))
			return 0;
		if (vcpu->arch.sie_block->gcr[0] & 0x400ul)
			return 1;
		return 0;
107
	case KVM_S390_INT_SERVICE:
108 109
	case KVM_S390_INT_PFAULT_INIT:
	case KVM_S390_INT_PFAULT_DONE:
110 111 112 113 114 115 116 117 118 119 120
	case KVM_S390_INT_VIRTIO:
		if (psw_extint_disabled(vcpu))
			return 0;
		if (vcpu->arch.sie_block->gcr[0] & 0x200ul)
			return 1;
		return 0;
	case KVM_S390_PROGRAM_INT:
	case KVM_S390_SIGP_STOP:
	case KVM_S390_SIGP_SET_PREFIX:
	case KVM_S390_RESTART:
		return 1;
121 122 123 124 125 126
	case KVM_S390_MCHK:
		if (psw_mchk_disabled(vcpu))
			return 0;
		if (vcpu->arch.sie_block->gcr[14] & inti->mchk.cr14)
			return 1;
		return 0;
127 128 129
	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
		if (psw_ioint_disabled(vcpu))
			return 0;
C
Cornelia Huck 已提交
130 131
		if (vcpu->arch.sie_block->gcr[6] &
		    int_word_to_isc_bits(inti->io.io_int_word))
132 133
			return 1;
		return 0;
134
	default:
135 136
		printk(KERN_WARNING "illegal interrupt type %llx\n",
		       inti->type);
137 138 139 140 141
		BUG();
	}
	return 0;
}

142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163
static inline unsigned long pending_local_irqs(struct kvm_vcpu *vcpu)
{
	return vcpu->arch.local_int.pending_irqs;
}

static unsigned long deliverable_local_irqs(struct kvm_vcpu *vcpu)
{
	unsigned long active_mask = pending_local_irqs(vcpu);

	if (psw_extint_disabled(vcpu))
		active_mask &= ~IRQ_PEND_EXT_MASK;
	if (!(vcpu->arch.sie_block->gcr[0] & 0x2000ul))
		__clear_bit(IRQ_PEND_EXT_EXTERNAL, &active_mask);
	if (!(vcpu->arch.sie_block->gcr[0] & 0x4000ul))
		__clear_bit(IRQ_PEND_EXT_EMERGENCY, &active_mask);
	if (!(vcpu->arch.sie_block->gcr[0] & 0x800ul))
		__clear_bit(IRQ_PEND_EXT_CLOCK_COMP, &active_mask);
	if (!(vcpu->arch.sie_block->gcr[0] & 0x400ul))
		__clear_bit(IRQ_PEND_EXT_CPU_TIMER, &active_mask);
	if (psw_mchk_disabled(vcpu))
		active_mask &= ~IRQ_PEND_MCHK_MASK;

164 165 166 167 168 169
	/*
	 * STOP irqs will never be actively delivered. They are triggered via
	 * intercept requests and cleared when the stop intercept is performed.
	 */
	__clear_bit(IRQ_PEND_SIGP_STOP, &active_mask);

170 171 172
	return active_mask;
}

173 174 175 176 177 178 179 180 181 182 183 184 185 186
static void __set_cpu_idle(struct kvm_vcpu *vcpu)
{
	atomic_set_mask(CPUSTAT_WAIT, &vcpu->arch.sie_block->cpuflags);
	set_bit(vcpu->vcpu_id, vcpu->arch.local_int.float_int->idle_mask);
}

static void __unset_cpu_idle(struct kvm_vcpu *vcpu)
{
	atomic_clear_mask(CPUSTAT_WAIT, &vcpu->arch.sie_block->cpuflags);
	clear_bit(vcpu->vcpu_id, vcpu->arch.local_int.float_int->idle_mask);
}

static void __reset_intercept_indicators(struct kvm_vcpu *vcpu)
{
187 188
	atomic_clear_mask(CPUSTAT_IO_INT | CPUSTAT_EXT_INT | CPUSTAT_STOP_INT,
			  &vcpu->arch.sie_block->cpuflags);
189
	vcpu->arch.sie_block->lctl = 0x0000;
190 191 192 193 194 195 196
	vcpu->arch.sie_block->ictl &= ~(ICTL_LPSW | ICTL_STCTL | ICTL_PINT);

	if (guestdbg_enabled(vcpu)) {
		vcpu->arch.sie_block->lctl |= (LCTL_CR0 | LCTL_CR9 |
					       LCTL_CR10 | LCTL_CR11);
		vcpu->arch.sie_block->ictl |= (ICTL_STCTL | ICTL_PINT);
	}
197 198 199 200 201 202 203
}

static void __set_cpuflag(struct kvm_vcpu *vcpu, u32 flag)
{
	atomic_set_mask(flag, &vcpu->arch.sie_block->cpuflags);
}

204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223
static void set_intercept_indicators_ext(struct kvm_vcpu *vcpu)
{
	if (!(pending_local_irqs(vcpu) & IRQ_PEND_EXT_MASK))
		return;
	if (psw_extint_disabled(vcpu))
		__set_cpuflag(vcpu, CPUSTAT_EXT_INT);
	else
		vcpu->arch.sie_block->lctl |= LCTL_CR0;
}

static void set_intercept_indicators_mchk(struct kvm_vcpu *vcpu)
{
	if (!(pending_local_irqs(vcpu) & IRQ_PEND_MCHK_MASK))
		return;
	if (psw_mchk_disabled(vcpu))
		vcpu->arch.sie_block->ictl |= ICTL_LPSW;
	else
		vcpu->arch.sie_block->lctl |= LCTL_CR14;
}

224 225 226 227 228 229
static void set_intercept_indicators_stop(struct kvm_vcpu *vcpu)
{
	if (kvm_s390_is_stop_irq_pending(vcpu))
		__set_cpuflag(vcpu, CPUSTAT_STOP_INT);
}

230 231 232 233 234
/* Set interception request for non-deliverable local interrupts */
static void set_intercept_indicators_local(struct kvm_vcpu *vcpu)
{
	set_intercept_indicators_ext(vcpu);
	set_intercept_indicators_mchk(vcpu);
235
	set_intercept_indicators_stop(vcpu);
236 237
}

238
static void __set_intercept_indicator(struct kvm_vcpu *vcpu,
239
				      struct kvm_s390_interrupt_info *inti)
240 241 242
{
	switch (inti->type) {
	case KVM_S390_INT_SERVICE:
243
	case KVM_S390_INT_PFAULT_DONE:
244 245 246 247 248 249
	case KVM_S390_INT_VIRTIO:
		if (psw_extint_disabled(vcpu))
			__set_cpuflag(vcpu, CPUSTAT_EXT_INT);
		else
			vcpu->arch.sie_block->lctl |= LCTL_CR0;
		break;
250 251 252 253 254 255
	case KVM_S390_MCHK:
		if (psw_mchk_disabled(vcpu))
			vcpu->arch.sie_block->ictl |= ICTL_LPSW;
		else
			vcpu->arch.sie_block->lctl |= LCTL_CR14;
		break;
256 257 258 259 260 261
	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
		if (psw_ioint_disabled(vcpu))
			__set_cpuflag(vcpu, CPUSTAT_IO_INT);
		else
			vcpu->arch.sie_block->lctl |= LCTL_CR6;
		break;
262 263 264 265 266
	default:
		BUG();
	}
}

267 268 269 270 271 272 273 274 275
static u16 get_ilc(struct kvm_vcpu *vcpu)
{
	switch (vcpu->arch.sie_block->icptcode) {
	case ICPT_INST:
	case ICPT_INSTPROGI:
	case ICPT_OPEREXC:
	case ICPT_PARTEXEC:
	case ICPT_IOINST:
		/* last instruction only stored for these icptcodes */
276
		return insn_length(vcpu->arch.sie_block->ipa >> 8);
277 278 279 280 281 282 283
	case ICPT_PROGI:
		return vcpu->arch.sie_block->pgmilc;
	default:
		return 0;
	}
}

284 285
static int __must_check __deliver_cpu_timer(struct kvm_vcpu *vcpu)
{
286
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
287 288 289 290 291 292 293
	int rc;

	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_INT_CPU_TIMER,
					 0, 0);

	rc  = put_guest_lc(vcpu, EXT_IRQ_CPU_TIMER,
			   (u16 *)__LC_EXT_INT_CODE);
294
	rc |= put_guest_lc(vcpu, 0, (u16 *)__LC_EXT_CPU_ADDR);
295 296 297 298
	rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
299
	clear_bit(IRQ_PEND_EXT_CPU_TIMER, &li->pending_irqs);
300
	return rc ? -EFAULT : 0;
301 302 303 304
}

static int __must_check __deliver_ckc(struct kvm_vcpu *vcpu)
{
305
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
306 307 308 309 310 311 312
	int rc;

	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_INT_CLOCK_COMP,
					 0, 0);

	rc  = put_guest_lc(vcpu, EXT_IRQ_CLK_COMP,
			   (u16 __user *)__LC_EXT_INT_CODE);
313
	rc |= put_guest_lc(vcpu, 0, (u16 *)__LC_EXT_CPU_ADDR);
314 315 316 317
	rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
318
	clear_bit(IRQ_PEND_EXT_CLOCK_COMP, &li->pending_irqs);
319
	return rc ? -EFAULT : 0;
320 321
}

322
static int __must_check __deliver_pfault_init(struct kvm_vcpu *vcpu)
323
{
324 325
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
	struct kvm_s390_ext_info ext;
326 327
	int rc;

328 329 330 331 332 333
	spin_lock(&li->lock);
	ext = li->irq.ext;
	clear_bit(IRQ_PEND_PFAULT_INIT, &li->pending_irqs);
	li->irq.ext.ext_params2 = 0;
	spin_unlock(&li->lock);

334
	VCPU_EVENT(vcpu, 4, "interrupt: pfault init parm:%x,parm64:%llx",
335
		   0, ext.ext_params2);
336 337
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
					 KVM_S390_INT_PFAULT_INIT,
338
					 0, ext.ext_params2);
339 340 341 342 343 344 345

	rc  = put_guest_lc(vcpu, EXT_IRQ_CP_SERVICE, (u16 *) __LC_EXT_INT_CODE);
	rc |= put_guest_lc(vcpu, PFAULT_INIT, (u16 *) __LC_EXT_CPU_ADDR);
	rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
346
	rc |= put_guest_lc(vcpu, ext.ext_params2, (u64 *) __LC_EXT_PARAMS2);
347
	return rc ? -EFAULT : 0;
348 349
}

350
static int __must_check __deliver_machine_check(struct kvm_vcpu *vcpu)
351
{
352 353
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
	struct kvm_s390_mchk_info mchk;
E
Eric Farman 已提交
354
	unsigned long adtl_status_addr;
355 356
	int rc;

357 358 359 360 361 362 363 364 365 366 367 368
	spin_lock(&li->lock);
	mchk = li->irq.mchk;
	/*
	 * If there was an exigent machine check pending, then any repressible
	 * machine checks that might have been pending are indicated along
	 * with it, so always clear both bits
	 */
	clear_bit(IRQ_PEND_MCHK_EX, &li->pending_irqs);
	clear_bit(IRQ_PEND_MCHK_REP, &li->pending_irqs);
	memset(&li->irq.mchk, 0, sizeof(mchk));
	spin_unlock(&li->lock);

369
	VCPU_EVENT(vcpu, 4, "interrupt: machine check mcic=%llx",
370
		   mchk.mcic);
371
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_MCHK,
372
					 mchk.cr14, mchk.mcic);
373 374

	rc  = kvm_s390_vcpu_store_status(vcpu, KVM_S390_STORE_STATUS_PREFIXED);
E
Eric Farman 已提交
375 376 377
	rc |= read_guest_lc(vcpu, __LC_VX_SAVE_AREA_ADDR,
			    &adtl_status_addr, sizeof(unsigned long));
	rc |= kvm_s390_vcpu_store_adtl_status(vcpu, adtl_status_addr);
378
	rc |= put_guest_lc(vcpu, mchk.mcic,
379
			   (u64 __user *) __LC_MCCK_CODE);
380
	rc |= put_guest_lc(vcpu, mchk.failing_storage_address,
381 382
			   (u64 __user *) __LC_MCCK_FAIL_STOR_ADDR);
	rc |= write_guest_lc(vcpu, __LC_PSW_SAVE_AREA,
383
			     &mchk.fixed_logout, sizeof(mchk.fixed_logout));
384 385 386 387
	rc |= write_guest_lc(vcpu, __LC_MCK_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_MCK_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
388
	return rc ? -EFAULT : 0;
389 390 391 392
}

static int __must_check __deliver_restart(struct kvm_vcpu *vcpu)
{
393
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
394 395 396 397 398 399 400 401 402 403 404
	int rc;

	VCPU_EVENT(vcpu, 4, "%s", "interrupt: cpu restart");
	vcpu->stat.deliver_restart_signal++;
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_RESTART, 0, 0);

	rc  = write_guest_lc(vcpu,
			     offsetof(struct _lowcore, restart_old_psw),
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, offsetof(struct _lowcore, restart_psw),
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
405
	clear_bit(IRQ_PEND_RESTART, &li->pending_irqs);
406
	return rc ? -EFAULT : 0;
407 408
}

409
static int __must_check __deliver_set_prefix(struct kvm_vcpu *vcpu)
410
{
411 412 413 414 415 416 417 418
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
	struct kvm_s390_prefix_info prefix;

	spin_lock(&li->lock);
	prefix = li->irq.prefix;
	li->irq.prefix.address = 0;
	clear_bit(IRQ_PEND_SET_PREFIX, &li->pending_irqs);
	spin_unlock(&li->lock);
419

420
	VCPU_EVENT(vcpu, 4, "interrupt: set prefix to %x", prefix.address);
421 422 423
	vcpu->stat.deliver_prefix_signal++;
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
					 KVM_S390_SIGP_SET_PREFIX,
424
					 prefix.address, 0);
425

426
	kvm_s390_set_prefix(vcpu, prefix.address);
427 428 429
	return 0;
}

430
static int __must_check __deliver_emergency_signal(struct kvm_vcpu *vcpu)
431
{
432
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
433
	int rc;
434 435 436 437 438 439 440 441
	int cpu_addr;

	spin_lock(&li->lock);
	cpu_addr = find_first_bit(li->sigp_emerg_pending, KVM_MAX_VCPUS);
	clear_bit(cpu_addr, li->sigp_emerg_pending);
	if (bitmap_empty(li->sigp_emerg_pending, KVM_MAX_VCPUS))
		clear_bit(IRQ_PEND_EXT_EMERGENCY, &li->pending_irqs);
	spin_unlock(&li->lock);
442 443 444

	VCPU_EVENT(vcpu, 4, "%s", "interrupt: sigp emerg");
	vcpu->stat.deliver_emergency_signal++;
445 446
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_INT_EMERGENCY,
					 cpu_addr, 0);
447 448 449

	rc  = put_guest_lc(vcpu, EXT_IRQ_EMERGENCY_SIG,
			   (u16 *)__LC_EXT_INT_CODE);
450
	rc |= put_guest_lc(vcpu, cpu_addr, (u16 *)__LC_EXT_CPU_ADDR);
451 452 453 454
	rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
455
	return rc ? -EFAULT : 0;
456 457
}

458
static int __must_check __deliver_external_call(struct kvm_vcpu *vcpu)
459
{
460 461
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
	struct kvm_s390_extcall_info extcall;
462 463
	int rc;

464 465 466 467 468 469
	spin_lock(&li->lock);
	extcall = li->irq.extcall;
	li->irq.extcall.code = 0;
	clear_bit(IRQ_PEND_EXT_EXTERNAL, &li->pending_irqs);
	spin_unlock(&li->lock);

470 471 472 473
	VCPU_EVENT(vcpu, 4, "%s", "interrupt: sigp ext call");
	vcpu->stat.deliver_external_call++;
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
					 KVM_S390_INT_EXTERNAL_CALL,
474
					 extcall.code, 0);
475 476 477

	rc  = put_guest_lc(vcpu, EXT_IRQ_EXTERNAL_CALL,
			   (u16 *)__LC_EXT_INT_CODE);
478
	rc |= put_guest_lc(vcpu, extcall.code, (u16 *)__LC_EXT_CPU_ADDR);
479 480 481 482
	rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, &vcpu->arch.sie_block->gpsw,
			    sizeof(psw_t));
483
	return rc ? -EFAULT : 0;
484 485
}

486
static int __must_check __deliver_prog(struct kvm_vcpu *vcpu)
487
{
488 489
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
	struct kvm_s390_pgm_info pgm_info;
490
	int rc = 0, nullifying = false;
491
	u16 ilc = get_ilc(vcpu);
492

493 494 495 496 497 498
	spin_lock(&li->lock);
	pgm_info = li->irq.pgm;
	clear_bit(IRQ_PEND_PROG, &li->pending_irqs);
	memset(&li->irq.pgm, 0, sizeof(pgm_info));
	spin_unlock(&li->lock);

499
	VCPU_EVENT(vcpu, 4, "interrupt: pgm check code:%x, ilc:%x",
500
		   pgm_info.code, ilc);
501 502
	vcpu->stat.deliver_program_int++;
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_PROGRAM_INT,
503
					 pgm_info.code, 0);
504

505
	switch (pgm_info.code & ~PGM_PER) {
506 507 508 509 510 511 512 513 514
	case PGM_AFX_TRANSLATION:
	case PGM_ASX_TRANSLATION:
	case PGM_EX_TRANSLATION:
	case PGM_LFX_TRANSLATION:
	case PGM_LSTE_SEQUENCE:
	case PGM_LSX_TRANSLATION:
	case PGM_LX_TRANSLATION:
	case PGM_PRIMARY_AUTHORITY:
	case PGM_SECONDARY_AUTHORITY:
515 516
		nullifying = true;
		/* fall through */
517
	case PGM_SPACE_SWITCH:
518
		rc = put_guest_lc(vcpu, pgm_info.trans_exc_code,
519 520 521 522 523 524 525 526
				  (u64 *)__LC_TRANS_EXC_CODE);
		break;
	case PGM_ALEN_TRANSLATION:
	case PGM_ALE_SEQUENCE:
	case PGM_ASTE_INSTANCE:
	case PGM_ASTE_SEQUENCE:
	case PGM_ASTE_VALIDITY:
	case PGM_EXTENDED_AUTHORITY:
527
		rc = put_guest_lc(vcpu, pgm_info.exc_access_id,
528
				  (u8 *)__LC_EXC_ACCESS_ID);
529
		nullifying = true;
530 531 532 533 534 535 536
		break;
	case PGM_ASCE_TYPE:
	case PGM_PAGE_TRANSLATION:
	case PGM_REGION_FIRST_TRANS:
	case PGM_REGION_SECOND_TRANS:
	case PGM_REGION_THIRD_TRANS:
	case PGM_SEGMENT_TRANSLATION:
537
		rc = put_guest_lc(vcpu, pgm_info.trans_exc_code,
538
				  (u64 *)__LC_TRANS_EXC_CODE);
539
		rc |= put_guest_lc(vcpu, pgm_info.exc_access_id,
540
				   (u8 *)__LC_EXC_ACCESS_ID);
541
		rc |= put_guest_lc(vcpu, pgm_info.op_access_id,
542
				   (u8 *)__LC_OP_ACCESS_ID);
543
		nullifying = true;
544 545
		break;
	case PGM_MONITOR:
546
		rc = put_guest_lc(vcpu, pgm_info.mon_class_nr,
547
				  (u16 *)__LC_MON_CLASS_NR);
548
		rc |= put_guest_lc(vcpu, pgm_info.mon_code,
549 550
				   (u64 *)__LC_MON_CODE);
		break;
E
Eric Farman 已提交
551
	case PGM_VECTOR_PROCESSING:
552
	case PGM_DATA:
553
		rc = put_guest_lc(vcpu, pgm_info.data_exc_code,
554 555 556
				  (u32 *)__LC_DATA_EXC_CODE);
		break;
	case PGM_PROTECTION:
557
		rc = put_guest_lc(vcpu, pgm_info.trans_exc_code,
558
				  (u64 *)__LC_TRANS_EXC_CODE);
559
		rc |= put_guest_lc(vcpu, pgm_info.exc_access_id,
560 561
				   (u8 *)__LC_EXC_ACCESS_ID);
		break;
562 563 564 565 566 567 568 569 570
	case PGM_STACK_FULL:
	case PGM_STACK_EMPTY:
	case PGM_STACK_SPECIFICATION:
	case PGM_STACK_TYPE:
	case PGM_STACK_OPERATION:
	case PGM_TRACE_TABEL:
	case PGM_CRYPTO_OPERATION:
		nullifying = true;
		break;
571 572
	}

573 574
	if (pgm_info.code & PGM_PER) {
		rc |= put_guest_lc(vcpu, pgm_info.per_code,
575
				   (u8 *) __LC_PER_CODE);
576
		rc |= put_guest_lc(vcpu, pgm_info.per_atmid,
577
				   (u8 *)__LC_PER_ATMID);
578
		rc |= put_guest_lc(vcpu, pgm_info.per_address,
579
				   (u64 *) __LC_PER_ADDRESS);
580
		rc |= put_guest_lc(vcpu, pgm_info.per_access_id,
581 582 583
				   (u8 *) __LC_PER_ACCESS_ID);
	}

584 585 586
	if (nullifying && vcpu->arch.sie_block->icptcode == ICPT_INST)
		kvm_s390_rewind_psw(vcpu, ilc);

587
	rc |= put_guest_lc(vcpu, ilc, (u16 *) __LC_PGM_ILC);
588
	rc |= put_guest_lc(vcpu, pgm_info.code,
589 590 591 592 593
			   (u16 *)__LC_PGM_INT_CODE);
	rc |= write_guest_lc(vcpu, __LC_PGM_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_PGM_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
594
	return rc ? -EFAULT : 0;
595 596 597 598 599 600 601 602 603 604 605 606 607 608
}

static int __must_check __deliver_service(struct kvm_vcpu *vcpu,
					  struct kvm_s390_interrupt_info *inti)
{
	int rc;

	VCPU_EVENT(vcpu, 4, "interrupt: sclp parm:%x",
		   inti->ext.ext_params);
	vcpu->stat.deliver_service_signal++;
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type,
					 inti->ext.ext_params, 0);

	rc  = put_guest_lc(vcpu, EXT_IRQ_SERVICE_SIG, (u16 *)__LC_EXT_INT_CODE);
609
	rc |= put_guest_lc(vcpu, 0, (u16 *)__LC_EXT_CPU_ADDR);
610 611 612 613 614 615
	rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= put_guest_lc(vcpu, inti->ext.ext_params,
			   (u32 *)__LC_EXT_PARAMS);
616
	return rc ? -EFAULT : 0;
617 618 619 620 621 622 623 624 625 626
}

static int __must_check __deliver_pfault_done(struct kvm_vcpu *vcpu,
					   struct kvm_s390_interrupt_info *inti)
{
	int rc;

	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
					 KVM_S390_INT_PFAULT_DONE, 0,
					 inti->ext.ext_params2);
627

628 629 630 631 632 633 634 635
	rc  = put_guest_lc(vcpu, EXT_IRQ_CP_SERVICE, (u16 *)__LC_EXT_INT_CODE);
	rc |= put_guest_lc(vcpu, PFAULT_DONE, (u16 *)__LC_EXT_CPU_ADDR);
	rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= put_guest_lc(vcpu, inti->ext.ext_params2,
			   (u64 *)__LC_EXT_PARAMS2);
636
	return rc ? -EFAULT : 0;
637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660
}

static int __must_check __deliver_virtio(struct kvm_vcpu *vcpu,
					 struct kvm_s390_interrupt_info *inti)
{
	int rc;

	VCPU_EVENT(vcpu, 4, "interrupt: virtio parm:%x,parm64:%llx",
		   inti->ext.ext_params, inti->ext.ext_params2);
	vcpu->stat.deliver_virtio_interrupt++;
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type,
					 inti->ext.ext_params,
					 inti->ext.ext_params2);

	rc  = put_guest_lc(vcpu, EXT_IRQ_CP_SERVICE, (u16 *)__LC_EXT_INT_CODE);
	rc |= put_guest_lc(vcpu, VIRTIO_PARAM, (u16 *)__LC_EXT_CPU_ADDR);
	rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= put_guest_lc(vcpu, inti->ext.ext_params,
			   (u32 *)__LC_EXT_PARAMS);
	rc |= put_guest_lc(vcpu, inti->ext.ext_params2,
			   (u64 *)__LC_EXT_PARAMS2);
661
	return rc ? -EFAULT : 0;
662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688
}

static int __must_check __deliver_io(struct kvm_vcpu *vcpu,
				     struct kvm_s390_interrupt_info *inti)
{
	int rc;

	VCPU_EVENT(vcpu, 4, "interrupt: I/O %llx", inti->type);
	vcpu->stat.deliver_io_int++;
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type,
					 ((__u32)inti->io.subchannel_id << 16) |
						inti->io.subchannel_nr,
					 ((__u64)inti->io.io_int_parm << 32) |
						inti->io.io_int_word);

	rc  = put_guest_lc(vcpu, inti->io.subchannel_id,
			   (u16 *)__LC_SUBCHANNEL_ID);
	rc |= put_guest_lc(vcpu, inti->io.subchannel_nr,
			   (u16 *)__LC_SUBCHANNEL_NR);
	rc |= put_guest_lc(vcpu, inti->io.io_int_parm,
			   (u32 *)__LC_IO_INT_PARM);
	rc |= put_guest_lc(vcpu, inti->io.io_int_word,
			   (u32 *)__LC_IO_INT_WORD);
	rc |= write_guest_lc(vcpu, __LC_IO_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_IO_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
689
	return rc ? -EFAULT : 0;
690 691
}

692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713
static int __must_check __deliver_mchk_floating(struct kvm_vcpu *vcpu,
					   struct kvm_s390_interrupt_info *inti)
{
	struct kvm_s390_mchk_info *mchk = &inti->mchk;
	int rc;

	VCPU_EVENT(vcpu, 4, "interrupt: machine check mcic=%llx",
		   mchk->mcic);
	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_MCHK,
					 mchk->cr14, mchk->mcic);

	rc  = kvm_s390_vcpu_store_status(vcpu, KVM_S390_STORE_STATUS_PREFIXED);
	rc |= put_guest_lc(vcpu, mchk->mcic,
			(u64 __user *) __LC_MCCK_CODE);
	rc |= put_guest_lc(vcpu, mchk->failing_storage_address,
			(u64 __user *) __LC_MCCK_FAIL_STOR_ADDR);
	rc |= write_guest_lc(vcpu, __LC_PSW_SAVE_AREA,
			     &mchk->fixed_logout, sizeof(mchk->fixed_logout));
	rc |= write_guest_lc(vcpu, __LC_MCK_OLD_PSW,
			     &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
	rc |= read_guest_lc(vcpu, __LC_MCK_NEW_PSW,
			    &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
714
	return rc ? -EFAULT : 0;
715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732
}

typedef int (*deliver_irq_t)(struct kvm_vcpu *vcpu);

static const deliver_irq_t deliver_irq_funcs[] = {
	[IRQ_PEND_MCHK_EX]        = __deliver_machine_check,
	[IRQ_PEND_PROG]           = __deliver_prog,
	[IRQ_PEND_EXT_EMERGENCY]  = __deliver_emergency_signal,
	[IRQ_PEND_EXT_EXTERNAL]   = __deliver_external_call,
	[IRQ_PEND_EXT_CLOCK_COMP] = __deliver_ckc,
	[IRQ_PEND_EXT_CPU_TIMER]  = __deliver_cpu_timer,
	[IRQ_PEND_RESTART]        = __deliver_restart,
	[IRQ_PEND_SET_PREFIX]     = __deliver_set_prefix,
	[IRQ_PEND_PFAULT_INIT]    = __deliver_pfault_init,
};

static int __must_check __deliver_floating_interrupt(struct kvm_vcpu *vcpu,
					   struct kvm_s390_interrupt_info *inti)
733
{
734
	int rc;
735 736 737

	switch (inti->type) {
	case KVM_S390_INT_SERVICE:
738
		rc = __deliver_service(vcpu, inti);
739
		break;
740
	case KVM_S390_INT_PFAULT_DONE:
741
		rc = __deliver_pfault_done(vcpu, inti);
742
		break;
743
	case KVM_S390_INT_VIRTIO:
744
		rc = __deliver_virtio(vcpu, inti);
745
		break;
746
	case KVM_S390_MCHK:
747
		rc = __deliver_mchk_floating(vcpu, inti);
748
		break;
749
	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
750
		rc = __deliver_io(vcpu, inti);
751
		break;
752 753 754
	default:
		BUG();
	}
755 756

	return rc;
757 758
}

759 760
/* Check whether an external call is pending (deliverable or not) */
int kvm_s390_ext_call_pending(struct kvm_vcpu *vcpu)
761
{
762 763
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
	uint8_t sigp_ctrl = vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sigp_ctrl;
764

765 766
	if (!sclp_has_sigpif())
		return test_bit(IRQ_PEND_EXT_EXTERNAL, &li->pending_irqs);
767

768 769
	return (sigp_ctrl & SIGP_CTRL_C) &&
	       (atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_ECALL_PEND);
770 771
}

772
int kvm_s390_vcpu_has_irq(struct kvm_vcpu *vcpu, int exclude_stop)
773
{
774 775
	struct kvm_s390_float_interrupt *fi = vcpu->arch.local_int.float_int;
	struct kvm_s390_interrupt_info  *inti;
776
	int rc;
777

778
	rc = !!deliverable_local_irqs(vcpu);
779 780

	if ((!rc) && atomic_read(&fi->active)) {
781
		spin_lock(&fi->lock);
782 783 784 785 786
		list_for_each_entry(inti, &fi->list, list)
			if (__interrupt_is_deliverable(vcpu, inti)) {
				rc = 1;
				break;
			}
787
		spin_unlock(&fi->lock);
788 789
	}

790 791
	if (!rc && kvm_cpu_has_pending_timer(vcpu))
		rc = 1;
792

793 794 795 796
	/* external call pending and deliverable */
	if (!rc && kvm_s390_ext_call_pending(vcpu) &&
	    !psw_extint_disabled(vcpu) &&
	    (vcpu->arch.sie_block->gcr[0] & 0x2000ul))
797 798
		rc = 1;

799
	if (!rc && !exclude_stop && kvm_s390_is_stop_irq_pending(vcpu))
800 801
		rc = 1;

802 803 804
	return rc;
}

805 806
int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
{
807 808 809 810 811 812
	if (!(vcpu->arch.sie_block->ckc <
	      get_tod_clock_fast() + vcpu->arch.sie_block->epoch))
		return 0;
	if (!ckc_interrupts_enabled(vcpu))
		return 0;
	return 1;
813 814
}

815 816 817 818 819 820
int kvm_s390_handle_wait(struct kvm_vcpu *vcpu)
{
	u64 now, sltime;

	vcpu->stat.exit_wait_state++;

821 822 823
	/* fast path */
	if (kvm_cpu_has_pending_timer(vcpu) || kvm_arch_vcpu_runnable(vcpu))
		return 0;
824

825 826
	if (psw_interrupts_disabled(vcpu)) {
		VCPU_EVENT(vcpu, 3, "%s", "disabled wait");
827
		return -EOPNOTSUPP; /* disabled wait */
828 829
	}

830
	if (!ckc_interrupts_enabled(vcpu)) {
831
		VCPU_EVENT(vcpu, 3, "%s", "enabled wait w/o timer");
832
		__set_cpu_idle(vcpu);
833 834 835
		goto no_timer;
	}

836
	now = get_tod_clock_fast() + vcpu->arch.sie_block->epoch;
837
	sltime = tod_to_ns(vcpu->arch.sie_block->ckc - now);
838 839 840 841 842 843

	/* underflow */
	if (vcpu->arch.sie_block->ckc < now)
		return 0;

	__set_cpu_idle(vcpu);
844 845
	hrtimer_start(&vcpu->arch.ckc_timer, ktime_set (0, sltime) , HRTIMER_MODE_REL);
	VCPU_EVENT(vcpu, 5, "enabled wait via clock comparator: %llx ns", sltime);
846
no_timer:
847
	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
848
	kvm_vcpu_block(vcpu);
849
	__unset_cpu_idle(vcpu);
850 851
	vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);

852
	hrtimer_cancel(&vcpu->arch.ckc_timer);
853 854 855
	return 0;
}

856 857 858 859 860 861 862 863 864
void kvm_s390_vcpu_wakeup(struct kvm_vcpu *vcpu)
{
	if (waitqueue_active(&vcpu->wq)) {
		/*
		 * The vcpu gave up the cpu voluntarily, mark it as a good
		 * yield-candidate.
		 */
		vcpu->preempted = true;
		wake_up_interruptible(&vcpu->wq);
865
		vcpu->stat.halt_wakeup++;
866 867 868
	}
}

869 870 871
enum hrtimer_restart kvm_s390_idle_wakeup(struct hrtimer *timer)
{
	struct kvm_vcpu *vcpu;
872
	u64 now, sltime;
873 874

	vcpu = container_of(timer, struct kvm_vcpu, arch.ckc_timer);
875 876
	now = get_tod_clock_fast() + vcpu->arch.sie_block->epoch;
	sltime = tod_to_ns(vcpu->arch.sie_block->ckc - now);
877

878 879 880 881 882 883 884 885
	/*
	 * If the monotonic clock runs faster than the tod clock we might be
	 * woken up too early and have to go back to sleep to avoid deadlocks.
	 */
	if (vcpu->arch.sie_block->ckc > now &&
	    hrtimer_forward_now(timer, ns_to_ktime(sltime)))
		return HRTIMER_RESTART;
	kvm_s390_vcpu_wakeup(vcpu);
886 887
	return HRTIMER_NORESTART;
}
888

889 890 891 892
void kvm_s390_clear_local_irqs(struct kvm_vcpu *vcpu)
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;

893
	spin_lock(&li->lock);
894 895 896
	li->pending_irqs = 0;
	bitmap_zero(li->sigp_emerg_pending, KVM_MAX_VCPUS);
	memset(&li->irq, 0, sizeof(li->irq));
897
	spin_unlock(&li->lock);
898 899

	/* clear pending external calls set by sigp interpretation facility */
900
	atomic_clear_mask(CPUSTAT_ECALL_PEND, li->cpuflags);
901
	vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sigp_ctrl = 0;
902 903
}

904
int __must_check kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu)
905
{
906 907 908
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
	struct kvm_s390_float_interrupt *fi = vcpu->arch.local_int.float_int;
	struct kvm_s390_interrupt_info  *n, *inti = NULL;
909
	deliver_irq_t func;
910
	int deliver;
911
	int rc = 0;
912 913
	unsigned long irq_type;
	unsigned long deliverable_irqs;
914 915 916

	__reset_intercept_indicators(vcpu);

917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937
	/* pending ckc conditions might have been invalidated */
	clear_bit(IRQ_PEND_EXT_CLOCK_COMP, &li->pending_irqs);
	if (kvm_cpu_has_pending_timer(vcpu))
		set_bit(IRQ_PEND_EXT_CLOCK_COMP, &li->pending_irqs);

	do {
		deliverable_irqs = deliverable_local_irqs(vcpu);
		/* bits are in the order of interrupt priority */
		irq_type = find_first_bit(&deliverable_irqs, IRQ_PEND_COUNT);
		if (irq_type == IRQ_PEND_COUNT)
			break;
		func = deliver_irq_funcs[irq_type];
		if (!func) {
			WARN_ON_ONCE(func == NULL);
			clear_bit(irq_type, &li->pending_irqs);
			continue;
		}
		rc = func(vcpu);
	} while (!rc && irq_type != IRQ_PEND_COUNT);

	set_intercept_indicators_local(vcpu);
938

939
	if (!rc && atomic_read(&fi->active)) {
940 941
		do {
			deliver = 0;
942
			spin_lock(&fi->lock);
943 944 945
			list_for_each_entry_safe(inti, n, &fi->list, list) {
				if (__interrupt_is_deliverable(vcpu, inti)) {
					list_del(&inti->list);
J
Jens Freimann 已提交
946
					fi->irq_count--;
947 948 949 950 951 952 953
					deliver = 1;
					break;
				}
				__set_intercept_indicator(vcpu, inti);
			}
			if (list_empty(&fi->list))
				atomic_set(&fi->active, 0);
954
			spin_unlock(&fi->lock);
955
			if (deliver) {
956
				rc = __deliver_floating_interrupt(vcpu, inti);
957 958
				kfree(inti);
			}
959
		} while (!rc && deliver);
960
	}
961 962

	return rc;
963 964
}

965
static int __inject_prog(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
966 967 968
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;

969
	li->irq.pgm = irq->u.pgm;
970
	set_bit(IRQ_PEND_PROG, &li->pending_irqs);
971 972 973
	return 0;
}

974 975
int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code)
{
976
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
977
	struct kvm_s390_irq irq;
978 979

	VCPU_EVENT(vcpu, 3, "inject: program check %d (from kernel)", code);
980 981
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_PROGRAM_INT, code,
				   0, 1);
982
	spin_lock(&li->lock);
983 984
	irq.u.pgm.code = code;
	__inject_prog(vcpu, &irq);
985
	BUG_ON(waitqueue_active(li->wq));
986
	spin_unlock(&li->lock);
987 988 989 990 991 992 993
	return 0;
}

int kvm_s390_inject_prog_irq(struct kvm_vcpu *vcpu,
			     struct kvm_s390_pgm_info *pgm_info)
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
994
	struct kvm_s390_irq irq;
995
	int rc;
996 997 998 999 1000

	VCPU_EVENT(vcpu, 3, "inject: prog irq %d (from kernel)",
		   pgm_info->code);
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_PROGRAM_INT,
				   pgm_info->code, 0, 1);
1001
	spin_lock(&li->lock);
1002 1003
	irq.u.pgm = *pgm_info;
	rc = __inject_prog(vcpu, &irq);
1004
	BUG_ON(waitqueue_active(li->wq));
1005
	spin_unlock(&li->lock);
1006 1007 1008
	return rc;
}

1009
static int __inject_pfault_init(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
1010 1011 1012
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;

1013 1014 1015 1016 1017 1018 1019 1020
	VCPU_EVENT(vcpu, 3, "inject: external irq params:%x, params2:%llx",
		   irq->u.ext.ext_params, irq->u.ext.ext_params2);
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_PFAULT_INIT,
				   irq->u.ext.ext_params,
				   irq->u.ext.ext_params2, 2);

	li->irq.ext = irq->u.ext;
	set_bit(IRQ_PEND_PFAULT_INIT, &li->pending_irqs);
1021 1022 1023 1024
	atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
	return 0;
}

1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039
static int __inject_extcall_sigpif(struct kvm_vcpu *vcpu, uint16_t src_id)
{
	unsigned char new_val, old_val;
	uint8_t *sigp_ctrl = &vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sigp_ctrl;

	new_val = SIGP_CTRL_C | (src_id & SIGP_CTRL_SCN_MASK);
	old_val = *sigp_ctrl & ~SIGP_CTRL_C;
	if (cmpxchg(sigp_ctrl, old_val, new_val) != old_val) {
		/* another external call is pending */
		return -EBUSY;
	}
	atomic_set_mask(CPUSTAT_ECALL_PEND, &vcpu->arch.sie_block->cpuflags);
	return 0;
}

1040
static int __inject_extcall(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
1041 1042
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
1043
	struct kvm_s390_extcall_info *extcall = &li->irq.extcall;
1044
	uint16_t src_id = irq->u.extcall.code;
1045 1046

	VCPU_EVENT(vcpu, 3, "inject: external call source-cpu:%u",
1047
		   src_id);
1048
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_EXTERNAL_CALL,
1049 1050 1051 1052 1053 1054 1055 1056 1057
				   src_id, 0, 2);

	/* sending vcpu invalid */
	if (src_id >= KVM_MAX_VCPUS ||
	    kvm_get_vcpu(vcpu->kvm, src_id) == NULL)
		return -EINVAL;

	if (sclp_has_sigpif())
		return __inject_extcall_sigpif(vcpu, src_id);
1058

1059 1060
	if (!test_and_set_bit(IRQ_PEND_EXT_EXTERNAL, &li->pending_irqs))
		return -EBUSY;
1061
	*extcall = irq->u.extcall;
1062 1063 1064 1065
	atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
	return 0;
}

1066
static int __inject_set_prefix(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
1067 1068
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
1069
	struct kvm_s390_prefix_info *prefix = &li->irq.prefix;
1070 1071

	VCPU_EVENT(vcpu, 3, "inject: set prefix to %x (from user)",
1072
		   irq->u.prefix.address);
1073
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_SIGP_SET_PREFIX,
1074
				   irq->u.prefix.address, 0, 2);
1075

1076 1077 1078
	if (!is_vcpu_stopped(vcpu))
		return -EBUSY;

1079 1080
	*prefix = irq->u.prefix;
	set_bit(IRQ_PEND_SET_PREFIX, &li->pending_irqs);
1081 1082 1083
	return 0;
}

1084
#define KVM_S390_STOP_SUPP_FLAGS (KVM_S390_STOP_FLAG_STORE_STATUS)
1085
static int __inject_sigp_stop(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
1086 1087
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
1088
	struct kvm_s390_stop_info *stop = &li->irq.stop;
1089
	int rc = 0;
1090

1091 1092
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_SIGP_STOP, 0, 0, 2);

1093 1094 1095
	if (irq->u.stop.flags & ~KVM_S390_STOP_SUPP_FLAGS)
		return -EINVAL;

1096 1097 1098 1099 1100 1101 1102 1103 1104
	if (is_vcpu_stopped(vcpu)) {
		if (irq->u.stop.flags & KVM_S390_STOP_FLAG_STORE_STATUS)
			rc = kvm_s390_store_status_unloaded(vcpu,
						KVM_S390_STORE_STATUS_NOADDR);
		return rc;
	}

	if (test_and_set_bit(IRQ_PEND_SIGP_STOP, &li->pending_irqs))
		return -EBUSY;
1105
	stop->flags = irq->u.stop.flags;
1106
	__set_cpuflag(vcpu, CPUSTAT_STOP_INT);
1107 1108 1109 1110
	return 0;
}

static int __inject_sigp_restart(struct kvm_vcpu *vcpu,
1111
				 struct kvm_s390_irq *irq)
1112 1113 1114
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;

1115 1116 1117 1118
	VCPU_EVENT(vcpu, 3, "inject: restart type %llx", irq->type);
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_RESTART, 0, 0, 2);

	set_bit(IRQ_PEND_RESTART, &li->pending_irqs);
1119 1120 1121 1122
	return 0;
}

static int __inject_sigp_emergency(struct kvm_vcpu *vcpu,
1123
				   struct kvm_s390_irq *irq)
1124 1125 1126
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;

1127 1128 1129
	VCPU_EVENT(vcpu, 3, "inject: emergency %u\n",
		   irq->u.emerg.code);
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_EMERGENCY,
1130
				   irq->u.emerg.code, 0, 2);
1131

1132
	set_bit(irq->u.emerg.code, li->sigp_emerg_pending);
1133
	set_bit(IRQ_PEND_EXT_EMERGENCY, &li->pending_irqs);
1134 1135 1136 1137
	atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
	return 0;
}

1138
static int __inject_mchk(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
1139 1140
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
1141
	struct kvm_s390_mchk_info *mchk = &li->irq.mchk;
1142 1143

	VCPU_EVENT(vcpu, 5, "inject: machine check parm64:%llx",
1144
		   irq->u.mchk.mcic);
1145
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_MCHK, 0,
1146
				   irq->u.mchk.mcic, 2);
1147 1148

	/*
1149 1150 1151 1152 1153 1154
	 * Because repressible machine checks can be indicated along with
	 * exigent machine checks (PoP, Chapter 11, Interruption action)
	 * we need to combine cr14, mcic and external damage code.
	 * Failing storage address and the logout area should not be or'ed
	 * together, we just indicate the last occurrence of the corresponding
	 * machine check
1155
	 */
1156
	mchk->cr14 |= irq->u.mchk.cr14;
1157
	mchk->mcic |= irq->u.mchk.mcic;
1158 1159 1160 1161
	mchk->ext_damage_code |= irq->u.mchk.ext_damage_code;
	mchk->failing_storage_address = irq->u.mchk.failing_storage_address;
	memcpy(&mchk->fixed_logout, &irq->u.mchk.fixed_logout,
	       sizeof(mchk->fixed_logout));
1162 1163 1164 1165
	if (mchk->mcic & MCHK_EX_MASK)
		set_bit(IRQ_PEND_MCHK_EX, &li->pending_irqs);
	else if (mchk->mcic & MCHK_REP_MASK)
		set_bit(IRQ_PEND_MCHK_REP,  &li->pending_irqs);
1166 1167 1168
	return 0;
}

1169
static int __inject_ckc(struct kvm_vcpu *vcpu)
1170 1171 1172
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;

1173 1174 1175 1176 1177
	VCPU_EVENT(vcpu, 3, "inject: type %x", KVM_S390_INT_CLOCK_COMP);
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_CLOCK_COMP,
				   0, 0, 2);

	set_bit(IRQ_PEND_EXT_CLOCK_COMP, &li->pending_irqs);
1178 1179 1180 1181
	atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
	return 0;
}

1182
static int __inject_cpu_timer(struct kvm_vcpu *vcpu)
1183 1184 1185
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;

1186 1187 1188 1189 1190
	VCPU_EVENT(vcpu, 3, "inject: type %x", KVM_S390_INT_CPU_TIMER);
	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_CPU_TIMER,
				   0, 0, 2);

	set_bit(IRQ_PEND_EXT_CPU_TIMER, &li->pending_irqs);
1191
	atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
1192 1193 1194
	return 0;
}

1195

1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209
struct kvm_s390_interrupt_info *kvm_s390_get_io_int(struct kvm *kvm,
						    u64 cr6, u64 schid)
{
	struct kvm_s390_float_interrupt *fi;
	struct kvm_s390_interrupt_info *inti, *iter;

	if ((!schid && !cr6) || (schid && cr6))
		return NULL;
	fi = &kvm->arch.float_int;
	spin_lock(&fi->lock);
	inti = NULL;
	list_for_each_entry(iter, &fi->list, list) {
		if (!is_ioint(iter->type))
			continue;
C
Cornelia Huck 已提交
1210 1211
		if (cr6 &&
		    ((cr6 & int_word_to_isc_bits(iter->io.io_int_word)) == 0))
1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223
			continue;
		if (schid) {
			if (((schid & 0x00000000ffff0000) >> 16) !=
			    iter->io.subchannel_id)
				continue;
			if ((schid & 0x000000000000ffff) !=
			    iter->io.subchannel_nr)
				continue;
		}
		inti = iter;
		break;
	}
J
Jens Freimann 已提交
1224
	if (inti) {
1225
		list_del_init(&inti->list);
J
Jens Freimann 已提交
1226 1227
		fi->irq_count--;
	}
1228 1229 1230 1231 1232
	if (list_empty(&fi->list))
		atomic_set(&fi->active, 0);
	spin_unlock(&fi->lock);
	return inti;
}
1233

J
Jens Freimann 已提交
1234
static int __inject_vm(struct kvm *kvm, struct kvm_s390_interrupt_info *inti)
1235
{
1236 1237
	struct kvm_s390_local_interrupt *li;
	struct kvm_s390_float_interrupt *fi;
1238
	struct kvm_s390_interrupt_info *iter;
1239
	struct kvm_vcpu *dst_vcpu = NULL;
1240
	int sigcpu;
J
Jens Freimann 已提交
1241
	int rc = 0;
1242

1243 1244
	fi = &kvm->arch.float_int;
	spin_lock(&fi->lock);
J
Jens Freimann 已提交
1245 1246 1247 1248 1249
	if (fi->irq_count >= KVM_S390_MAX_FLOAT_IRQS) {
		rc = -EINVAL;
		goto unlock_fi;
	}
	fi->irq_count++;
1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266
	if (!is_ioint(inti->type)) {
		list_add_tail(&inti->list, &fi->list);
	} else {
		u64 isc_bits = int_word_to_isc_bits(inti->io.io_int_word);

		/* Keep I/O interrupts sorted in isc order. */
		list_for_each_entry(iter, &fi->list, list) {
			if (!is_ioint(iter->type))
				continue;
			if (int_word_to_isc_bits(iter->io.io_int_word)
			    <= isc_bits)
				continue;
			break;
		}
		list_add_tail(&inti->list, &iter->list);
	}
	atomic_set(&fi->active, 1);
1267 1268
	if (atomic_read(&kvm->online_vcpus) == 0)
		goto unlock_fi;
1269 1270 1271 1272 1273 1274
	sigcpu = find_first_bit(fi->idle_mask, KVM_MAX_VCPUS);
	if (sigcpu == KVM_MAX_VCPUS) {
		do {
			sigcpu = fi->next_rr_cpu++;
			if (sigcpu == KVM_MAX_VCPUS)
				sigcpu = fi->next_rr_cpu = 0;
1275
		} while (kvm_get_vcpu(kvm, sigcpu) == NULL);
1276
	}
1277 1278
	dst_vcpu = kvm_get_vcpu(kvm, sigcpu);
	li = &dst_vcpu->arch.local_int;
1279
	spin_lock(&li->lock);
1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290
	switch (inti->type) {
	case KVM_S390_MCHK:
		atomic_set_mask(CPUSTAT_STOP_INT, li->cpuflags);
		break;
	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
		atomic_set_mask(CPUSTAT_IO_INT, li->cpuflags);
		break;
	default:
		atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
		break;
	}
1291
	spin_unlock(&li->lock);
1292
	kvm_s390_vcpu_wakeup(kvm_get_vcpu(kvm, sigcpu));
J
Jens Freimann 已提交
1293
unlock_fi:
1294
	spin_unlock(&fi->lock);
J
Jens Freimann 已提交
1295
	return rc;
1296 1297 1298 1299 1300 1301
}

int kvm_s390_inject_vm(struct kvm *kvm,
		       struct kvm_s390_interrupt *s390int)
{
	struct kvm_s390_interrupt_info *inti;
1302
	int rc;
1303

1304 1305 1306 1307
	inti = kzalloc(sizeof(*inti), GFP_KERNEL);
	if (!inti)
		return -ENOMEM;

1308 1309
	inti->type = s390int->type;
	switch (inti->type) {
1310
	case KVM_S390_INT_VIRTIO:
1311
		VM_EVENT(kvm, 5, "inject: virtio parm:%x,parm64:%llx",
1312 1313 1314 1315 1316 1317 1318 1319
			 s390int->parm, s390int->parm64);
		inti->ext.ext_params = s390int->parm;
		inti->ext.ext_params2 = s390int->parm64;
		break;
	case KVM_S390_INT_SERVICE:
		VM_EVENT(kvm, 5, "inject: sclp parm:%x", s390int->parm);
		inti->ext.ext_params = s390int->parm;
		break;
1320 1321 1322
	case KVM_S390_INT_PFAULT_DONE:
		inti->ext.ext_params2 = s390int->parm64;
		break;
1323 1324 1325 1326 1327 1328
	case KVM_S390_MCHK:
		VM_EVENT(kvm, 5, "inject: machine check parm64:%llx",
			 s390int->parm64);
		inti->mchk.cr14 = s390int->parm; /* upper bits are not used */
		inti->mchk.mcic = s390int->parm64;
		break;
1329
	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
1330
		if (inti->type & IOINT_AI_MASK)
1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341
			VM_EVENT(kvm, 5, "%s", "inject: I/O (AI)");
		else
			VM_EVENT(kvm, 5, "inject: I/O css %x ss %x schid %04x",
				 s390int->type & IOINT_CSSID_MASK,
				 s390int->type & IOINT_SSID_MASK,
				 s390int->type & IOINT_SCHID_MASK);
		inti->io.subchannel_id = s390int->parm >> 16;
		inti->io.subchannel_nr = s390int->parm & 0x0000ffffu;
		inti->io.io_int_parm = s390int->parm64 >> 32;
		inti->io.io_int_word = s390int->parm64 & 0x00000000ffffffffull;
		break;
1342 1343 1344 1345
	default:
		kfree(inti);
		return -EINVAL;
	}
1346 1347
	trace_kvm_s390_inject_vm(s390int->type, s390int->parm, s390int->parm64,
				 2);
1348

1349 1350 1351 1352
	rc = __inject_vm(kvm, inti);
	if (rc)
		kfree(inti);
	return rc;
1353 1354
}

1355
int kvm_s390_reinject_io_int(struct kvm *kvm,
1356 1357
			      struct kvm_s390_interrupt_info *inti)
{
1358
	return __inject_vm(kvm, inti);
1359 1360
}

1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373
int s390int_to_s390irq(struct kvm_s390_interrupt *s390int,
		       struct kvm_s390_irq *irq)
{
	irq->type = s390int->type;
	switch (irq->type) {
	case KVM_S390_PROGRAM_INT:
		if (s390int->parm & 0xffff0000)
			return -EINVAL;
		irq->u.pgm.code = s390int->parm;
		break;
	case KVM_S390_SIGP_SET_PREFIX:
		irq->u.prefix.address = s390int->parm;
		break;
1374 1375 1376
	case KVM_S390_SIGP_STOP:
		irq->u.stop.flags = s390int->parm;
		break;
1377
	case KVM_S390_INT_EXTERNAL_CALL:
1378
		if (s390int->parm & 0xffff0000)
1379 1380 1381 1382
			return -EINVAL;
		irq->u.extcall.code = s390int->parm;
		break;
	case KVM_S390_INT_EMERGENCY:
1383
		if (s390int->parm & 0xffff0000)
1384 1385 1386 1387 1388 1389 1390 1391 1392 1393
			return -EINVAL;
		irq->u.emerg.code = s390int->parm;
		break;
	case KVM_S390_MCHK:
		irq->u.mchk.mcic = s390int->parm64;
		break;
	}
	return 0;
}

1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410
int kvm_s390_is_stop_irq_pending(struct kvm_vcpu *vcpu)
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;

	return test_bit(IRQ_PEND_SIGP_STOP, &li->pending_irqs);
}

void kvm_s390_clear_stop_irq(struct kvm_vcpu *vcpu)
{
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;

	spin_lock(&li->lock);
	li->irq.stop.flags = 0;
	clear_bit(IRQ_PEND_SIGP_STOP, &li->pending_irqs);
	spin_unlock(&li->lock);
}

1411
int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
1412
{
1413 1414
	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
	int rc;
1415

1416
	spin_lock(&li->lock);
1417
	switch (irq->type) {
1418 1419
	case KVM_S390_PROGRAM_INT:
		VCPU_EVENT(vcpu, 3, "inject: program check %d (from user)",
1420 1421
			   irq->u.pgm.code);
		rc = __inject_prog(vcpu, irq);
1422
		break;
1423
	case KVM_S390_SIGP_SET_PREFIX:
1424
		rc = __inject_set_prefix(vcpu, irq);
1425
		break;
1426
	case KVM_S390_SIGP_STOP:
1427
		rc = __inject_sigp_stop(vcpu, irq);
1428
		break;
1429
	case KVM_S390_RESTART:
1430
		rc = __inject_sigp_restart(vcpu, irq);
1431
		break;
1432
	case KVM_S390_INT_CLOCK_COMP:
1433
		rc = __inject_ckc(vcpu);
1434
		break;
1435
	case KVM_S390_INT_CPU_TIMER:
1436
		rc = __inject_cpu_timer(vcpu);
1437
		break;
1438
	case KVM_S390_INT_EXTERNAL_CALL:
1439
		rc = __inject_extcall(vcpu, irq);
1440
		break;
1441
	case KVM_S390_INT_EMERGENCY:
1442
		rc = __inject_sigp_emergency(vcpu, irq);
1443
		break;
1444
	case KVM_S390_MCHK:
1445
		rc = __inject_mchk(vcpu, irq);
1446
		break;
1447
	case KVM_S390_INT_PFAULT_INIT:
1448
		rc = __inject_pfault_init(vcpu, irq);
1449
		break;
1450 1451
	case KVM_S390_INT_VIRTIO:
	case KVM_S390_INT_SERVICE:
1452
	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
1453
	default:
1454
		rc = -EINVAL;
1455
	}
1456
	spin_unlock(&li->lock);
1457 1458 1459
	if (!rc)
		kvm_s390_vcpu_wakeup(vcpu);
	return rc;
1460
}
1461

1462
void kvm_s390_clear_float_irqs(struct kvm *kvm)
1463 1464 1465 1466 1467 1468 1469 1470 1471 1472
{
	struct kvm_s390_float_interrupt *fi;
	struct kvm_s390_interrupt_info	*n, *inti = NULL;

	fi = &kvm->arch.float_int;
	spin_lock(&fi->lock);
	list_for_each_entry_safe(inti, n, &fi->list, list) {
		list_del(&inti->list);
		kfree(inti);
	}
J
Jens Freimann 已提交
1473
	fi->irq_count = 0;
1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485
	atomic_set(&fi->active, 0);
	spin_unlock(&fi->lock);
}

static inline int copy_irq_to_user(struct kvm_s390_interrupt_info *inti,
				   u8 *addr)
{
	struct kvm_s390_irq __user *uptr = (struct kvm_s390_irq __user *) addr;
	struct kvm_s390_irq irq = {0};

	irq.type = inti->type;
	switch (inti->type) {
1486 1487
	case KVM_S390_INT_PFAULT_INIT:
	case KVM_S390_INT_PFAULT_DONE:
1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564
	case KVM_S390_INT_VIRTIO:
	case KVM_S390_INT_SERVICE:
		irq.u.ext = inti->ext;
		break;
	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
		irq.u.io = inti->io;
		break;
	case KVM_S390_MCHK:
		irq.u.mchk = inti->mchk;
		break;
	default:
		return -EINVAL;
	}

	if (copy_to_user(uptr, &irq, sizeof(irq)))
		return -EFAULT;

	return 0;
}

static int get_all_floating_irqs(struct kvm *kvm, __u8 *buf, __u64 len)
{
	struct kvm_s390_interrupt_info *inti;
	struct kvm_s390_float_interrupt *fi;
	int ret = 0;
	int n = 0;

	fi = &kvm->arch.float_int;
	spin_lock(&fi->lock);

	list_for_each_entry(inti, &fi->list, list) {
		if (len < sizeof(struct kvm_s390_irq)) {
			/* signal userspace to try again */
			ret = -ENOMEM;
			break;
		}
		ret = copy_irq_to_user(inti, buf);
		if (ret)
			break;
		buf += sizeof(struct kvm_s390_irq);
		len -= sizeof(struct kvm_s390_irq);
		n++;
	}

	spin_unlock(&fi->lock);

	return ret < 0 ? ret : n;
}

static int flic_get_attr(struct kvm_device *dev, struct kvm_device_attr *attr)
{
	int r;

	switch (attr->group) {
	case KVM_DEV_FLIC_GET_ALL_IRQS:
		r = get_all_floating_irqs(dev->kvm, (u8 *) attr->addr,
					  attr->attr);
		break;
	default:
		r = -EINVAL;
	}

	return r;
}

static inline int copy_irq_from_user(struct kvm_s390_interrupt_info *inti,
				     u64 addr)
{
	struct kvm_s390_irq __user *uptr = (struct kvm_s390_irq __user *) addr;
	void *target = NULL;
	void __user *source;
	u64 size;

	if (get_user(inti->type, (u64 __user *)addr))
		return -EFAULT;

	switch (inti->type) {
1565 1566
	case KVM_S390_INT_PFAULT_INIT:
	case KVM_S390_INT_PFAULT_DONE:
1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614
	case KVM_S390_INT_VIRTIO:
	case KVM_S390_INT_SERVICE:
		target = (void *) &inti->ext;
		source = &uptr->u.ext;
		size = sizeof(inti->ext);
		break;
	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
		target = (void *) &inti->io;
		source = &uptr->u.io;
		size = sizeof(inti->io);
		break;
	case KVM_S390_MCHK:
		target = (void *) &inti->mchk;
		source = &uptr->u.mchk;
		size = sizeof(inti->mchk);
		break;
	default:
		return -EINVAL;
	}

	if (copy_from_user(target, source, size))
		return -EFAULT;

	return 0;
}

static int enqueue_floating_irq(struct kvm_device *dev,
				struct kvm_device_attr *attr)
{
	struct kvm_s390_interrupt_info *inti = NULL;
	int r = 0;
	int len = attr->attr;

	if (len % sizeof(struct kvm_s390_irq) != 0)
		return -EINVAL;
	else if (len > KVM_S390_FLIC_MAX_BUFFER)
		return -EINVAL;

	while (len >= sizeof(struct kvm_s390_irq)) {
		inti = kzalloc(sizeof(*inti), GFP_KERNEL);
		if (!inti)
			return -ENOMEM;

		r = copy_irq_from_user(inti, attr->addr);
		if (r) {
			kfree(inti);
			return r;
		}
J
Jens Freimann 已提交
1615 1616 1617 1618 1619
		r = __inject_vm(dev->kvm, inti);
		if (r) {
			kfree(inti);
			return r;
		}
1620 1621 1622 1623 1624 1625 1626
		len -= sizeof(struct kvm_s390_irq);
		attr->addr += sizeof(struct kvm_s390_irq);
	}

	return r;
}

1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692
static struct s390_io_adapter *get_io_adapter(struct kvm *kvm, unsigned int id)
{
	if (id >= MAX_S390_IO_ADAPTERS)
		return NULL;
	return kvm->arch.adapters[id];
}

static int register_io_adapter(struct kvm_device *dev,
			       struct kvm_device_attr *attr)
{
	struct s390_io_adapter *adapter;
	struct kvm_s390_io_adapter adapter_info;

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

	if ((adapter_info.id >= MAX_S390_IO_ADAPTERS) ||
	    (dev->kvm->arch.adapters[adapter_info.id] != NULL))
		return -EINVAL;

	adapter = kzalloc(sizeof(*adapter), GFP_KERNEL);
	if (!adapter)
		return -ENOMEM;

	INIT_LIST_HEAD(&adapter->maps);
	init_rwsem(&adapter->maps_lock);
	atomic_set(&adapter->nr_maps, 0);
	adapter->id = adapter_info.id;
	adapter->isc = adapter_info.isc;
	adapter->maskable = adapter_info.maskable;
	adapter->masked = false;
	adapter->swap = adapter_info.swap;
	dev->kvm->arch.adapters[adapter->id] = adapter;

	return 0;
}

int kvm_s390_mask_adapter(struct kvm *kvm, unsigned int id, bool masked)
{
	int ret;
	struct s390_io_adapter *adapter = get_io_adapter(kvm, id);

	if (!adapter || !adapter->maskable)
		return -EINVAL;
	ret = adapter->masked;
	adapter->masked = masked;
	return ret;
}

static int kvm_s390_adapter_map(struct kvm *kvm, unsigned int id, __u64 addr)
{
	struct s390_io_adapter *adapter = get_io_adapter(kvm, id);
	struct s390_map_info *map;
	int ret;

	if (!adapter || !addr)
		return -EINVAL;

	map = kzalloc(sizeof(*map), GFP_KERNEL);
	if (!map) {
		ret = -ENOMEM;
		goto out;
	}
	INIT_LIST_HEAD(&map->list);
	map->guest_addr = addr;
1693
	map->addr = gmap_translate(kvm->arch.gmap, addr);
1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 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 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791
	if (map->addr == -EFAULT) {
		ret = -EFAULT;
		goto out;
	}
	ret = get_user_pages_fast(map->addr, 1, 1, &map->page);
	if (ret < 0)
		goto out;
	BUG_ON(ret != 1);
	down_write(&adapter->maps_lock);
	if (atomic_inc_return(&adapter->nr_maps) < MAX_S390_ADAPTER_MAPS) {
		list_add_tail(&map->list, &adapter->maps);
		ret = 0;
	} else {
		put_page(map->page);
		ret = -EINVAL;
	}
	up_write(&adapter->maps_lock);
out:
	if (ret)
		kfree(map);
	return ret;
}

static int kvm_s390_adapter_unmap(struct kvm *kvm, unsigned int id, __u64 addr)
{
	struct s390_io_adapter *adapter = get_io_adapter(kvm, id);
	struct s390_map_info *map, *tmp;
	int found = 0;

	if (!adapter || !addr)
		return -EINVAL;

	down_write(&adapter->maps_lock);
	list_for_each_entry_safe(map, tmp, &adapter->maps, list) {
		if (map->guest_addr == addr) {
			found = 1;
			atomic_dec(&adapter->nr_maps);
			list_del(&map->list);
			put_page(map->page);
			kfree(map);
			break;
		}
	}
	up_write(&adapter->maps_lock);

	return found ? 0 : -EINVAL;
}

void kvm_s390_destroy_adapters(struct kvm *kvm)
{
	int i;
	struct s390_map_info *map, *tmp;

	for (i = 0; i < MAX_S390_IO_ADAPTERS; i++) {
		if (!kvm->arch.adapters[i])
			continue;
		list_for_each_entry_safe(map, tmp,
					 &kvm->arch.adapters[i]->maps, list) {
			list_del(&map->list);
			put_page(map->page);
			kfree(map);
		}
		kfree(kvm->arch.adapters[i]);
	}
}

static int modify_io_adapter(struct kvm_device *dev,
			     struct kvm_device_attr *attr)
{
	struct kvm_s390_io_adapter_req req;
	struct s390_io_adapter *adapter;
	int ret;

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

	adapter = get_io_adapter(dev->kvm, req.id);
	if (!adapter)
		return -EINVAL;
	switch (req.type) {
	case KVM_S390_IO_ADAPTER_MASK:
		ret = kvm_s390_mask_adapter(dev->kvm, req.id, req.mask);
		if (ret > 0)
			ret = 0;
		break;
	case KVM_S390_IO_ADAPTER_MAP:
		ret = kvm_s390_adapter_map(dev->kvm, req.id, req.addr);
		break;
	case KVM_S390_IO_ADAPTER_UNMAP:
		ret = kvm_s390_adapter_unmap(dev->kvm, req.id, req.addr);
		break;
	default:
		ret = -EINVAL;
	}

	return ret;
}

1792 1793 1794
static int flic_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr)
{
	int r = 0;
1795 1796
	unsigned int i;
	struct kvm_vcpu *vcpu;
1797 1798 1799 1800 1801 1802

	switch (attr->group) {
	case KVM_DEV_FLIC_ENQUEUE:
		r = enqueue_floating_irq(dev, attr);
		break;
	case KVM_DEV_FLIC_CLEAR_IRQS:
1803
		kvm_s390_clear_float_irqs(dev->kvm);
1804
		break;
1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818
	case KVM_DEV_FLIC_APF_ENABLE:
		dev->kvm->arch.gmap->pfault_enabled = 1;
		break;
	case KVM_DEV_FLIC_APF_DISABLE_WAIT:
		dev->kvm->arch.gmap->pfault_enabled = 0;
		/*
		 * Make sure no async faults are in transition when
		 * clearing the queues. So we don't need to worry
		 * about late coming workers.
		 */
		synchronize_srcu(&dev->kvm->srcu);
		kvm_for_each_vcpu(i, vcpu, dev->kvm)
			kvm_clear_async_pf_completion_queue(vcpu);
		break;
1819 1820 1821 1822 1823 1824
	case KVM_DEV_FLIC_ADAPTER_REGISTER:
		r = register_io_adapter(dev, attr);
		break;
	case KVM_DEV_FLIC_ADAPTER_MODIFY:
		r = modify_io_adapter(dev, attr);
		break;
1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855
	default:
		r = -EINVAL;
	}

	return r;
}

static int flic_create(struct kvm_device *dev, u32 type)
{
	if (!dev)
		return -EINVAL;
	if (dev->kvm->arch.flic)
		return -EINVAL;
	dev->kvm->arch.flic = dev;
	return 0;
}

static void flic_destroy(struct kvm_device *dev)
{
	dev->kvm->arch.flic = NULL;
	kfree(dev);
}

/* s390 floating irq controller (flic) */
struct kvm_device_ops kvm_flic_ops = {
	.name = "kvm-flic",
	.get_attr = flic_get_attr,
	.set_attr = flic_set_attr,
	.create = flic_create,
	.destroy = flic_destroy,
};
1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947

static unsigned long get_ind_bit(__u64 addr, unsigned long bit_nr, bool swap)
{
	unsigned long bit;

	bit = bit_nr + (addr % PAGE_SIZE) * 8;

	return swap ? (bit ^ (BITS_PER_LONG - 1)) : bit;
}

static struct s390_map_info *get_map_info(struct s390_io_adapter *adapter,
					  u64 addr)
{
	struct s390_map_info *map;

	if (!adapter)
		return NULL;

	list_for_each_entry(map, &adapter->maps, list) {
		if (map->guest_addr == addr)
			return map;
	}
	return NULL;
}

static int adapter_indicators_set(struct kvm *kvm,
				  struct s390_io_adapter *adapter,
				  struct kvm_s390_adapter_int *adapter_int)
{
	unsigned long bit;
	int summary_set, idx;
	struct s390_map_info *info;
	void *map;

	info = get_map_info(adapter, adapter_int->ind_addr);
	if (!info)
		return -1;
	map = page_address(info->page);
	bit = get_ind_bit(info->addr, adapter_int->ind_offset, adapter->swap);
	set_bit(bit, map);
	idx = srcu_read_lock(&kvm->srcu);
	mark_page_dirty(kvm, info->guest_addr >> PAGE_SHIFT);
	set_page_dirty_lock(info->page);
	info = get_map_info(adapter, adapter_int->summary_addr);
	if (!info) {
		srcu_read_unlock(&kvm->srcu, idx);
		return -1;
	}
	map = page_address(info->page);
	bit = get_ind_bit(info->addr, adapter_int->summary_offset,
			  adapter->swap);
	summary_set = test_and_set_bit(bit, map);
	mark_page_dirty(kvm, info->guest_addr >> PAGE_SHIFT);
	set_page_dirty_lock(info->page);
	srcu_read_unlock(&kvm->srcu, idx);
	return summary_set ? 0 : 1;
}

/*
 * < 0 - not injected due to error
 * = 0 - coalesced, summary indicator already active
 * > 0 - injected interrupt
 */
static int set_adapter_int(struct kvm_kernel_irq_routing_entry *e,
			   struct kvm *kvm, int irq_source_id, int level,
			   bool line_status)
{
	int ret;
	struct s390_io_adapter *adapter;

	/* We're only interested in the 0->1 transition. */
	if (!level)
		return 0;
	adapter = get_io_adapter(kvm, e->adapter.adapter_id);
	if (!adapter)
		return -1;
	down_read(&adapter->maps_lock);
	ret = adapter_indicators_set(kvm, adapter, &e->adapter);
	up_read(&adapter->maps_lock);
	if ((ret > 0) && !adapter->masked) {
		struct kvm_s390_interrupt s390int = {
			.type = KVM_S390_INT_IO(1, 0, 0, 0),
			.parm = 0,
			.parm64 = (adapter->isc << 27) | 0x80000000,
		};
		ret = kvm_s390_inject_vm(kvm, &s390int);
		if (ret == 0)
			ret = 1;
	}
	return ret;
}

1948
int kvm_set_routing_entry(struct kvm_kernel_irq_routing_entry *e,
1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974
			  const struct kvm_irq_routing_entry *ue)
{
	int ret;

	switch (ue->type) {
	case KVM_IRQ_ROUTING_S390_ADAPTER:
		e->set = set_adapter_int;
		e->adapter.summary_addr = ue->u.adapter.summary_addr;
		e->adapter.ind_addr = ue->u.adapter.ind_addr;
		e->adapter.summary_offset = ue->u.adapter.summary_offset;
		e->adapter.ind_offset = ue->u.adapter.ind_offset;
		e->adapter.adapter_id = ue->u.adapter.adapter_id;
		ret = 0;
		break;
	default:
		ret = -EINVAL;
	}

	return ret;
}

int kvm_set_msi(struct kvm_kernel_irq_routing_entry *e, struct kvm *kvm,
		int irq_source_id, int level, bool line_status)
{
	return -EINVAL;
}