priv.c 24.0 KB
Newer Older
1
/*
2
 * handling privileged instructions
3
 *
4
 * Copyright IBM Corp. 2008, 2013
5 6 7 8 9 10 11 12 13 14
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License (version 2 only)
 * as published by the Free Software Foundation.
 *
 *    Author(s): Carsten Otte <cotte@de.ibm.com>
 *               Christian Borntraeger <borntraeger@de.ibm.com>
 */

#include <linux/kvm.h>
15
#include <linux/gfp.h>
16
#include <linux/errno.h>
17
#include <linux/compat.h>
18
#include <asm/asm-offsets.h>
19
#include <asm/facility.h>
20 21 22 23
#include <asm/current.h>
#include <asm/debug.h>
#include <asm/ebcdic.h>
#include <asm/sysinfo.h>
24 25 26
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/io.h>
27 28
#include <asm/ptrace.h>
#include <asm/compat.h>
29 30
#include "gaccess.h"
#include "kvm-s390.h"
31
#include "trace.h"
32

33 34 35 36 37
/* Handle SCK (SET CLOCK) interception */
static int handle_set_clock(struct kvm_vcpu *vcpu)
{
	struct kvm_vcpu *cpup;
	s64 hostclk, val;
38
	int i, rc;
39 40 41 42 43 44 45 46
	u64 op2;

	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

	op2 = kvm_s390_get_base_disp_s(vcpu);
	if (op2 & 7)	/* Operand must be on a doubleword boundary */
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
47 48 49
	rc = read_guest(vcpu, op2, &val, sizeof(val));
	if (rc)
		return kvm_s390_inject_prog_cond(vcpu, rc);
50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

	if (store_tod_clock(&hostclk)) {
		kvm_s390_set_psw_cc(vcpu, 3);
		return 0;
	}
	val = (val - hostclk) & ~0x3fUL;

	mutex_lock(&vcpu->kvm->lock);
	kvm_for_each_vcpu(i, cpup, vcpu->kvm)
		cpup->arch.sie_block->epoch = val;
	mutex_unlock(&vcpu->kvm->lock);

	kvm_s390_set_psw_cc(vcpu, 0);
	return 0;
}

66 67 68
static int handle_set_prefix(struct kvm_vcpu *vcpu)
{
	u64 operand2;
69 70
	u32 address;
	int rc;
71 72 73

	vcpu->stat.instruction_spx++;

74 75 76
	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

77
	operand2 = kvm_s390_get_base_disp_s(vcpu);
78 79

	/* must be word boundary */
80 81
	if (operand2 & 3)
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
82 83

	/* get the value */
84 85 86 87 88 89 90 91 92 93 94 95
	rc = read_guest(vcpu, operand2, &address, sizeof(address));
	if (rc)
		return kvm_s390_inject_prog_cond(vcpu, rc);

	address &= 0x7fffe000u;

	/*
	 * Make sure the new value is valid memory. We only need to check the
	 * first page, since address is 8k aligned and memory pieces are always
	 * at least 1MB aligned and have at least a size of 1MB.
	 */
	if (kvm_is_error_gpa(vcpu->kvm, address))
96
		return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
97

98
	kvm_s390_set_prefix(vcpu, address);
99 100

	VCPU_EVENT(vcpu, 5, "setting prefix to %x", address);
101
	trace_kvm_s390_handle_prefix(vcpu, 1, address);
102 103 104 105 106 107 108
	return 0;
}

static int handle_store_prefix(struct kvm_vcpu *vcpu)
{
	u64 operand2;
	u32 address;
109
	int rc;
110 111

	vcpu->stat.instruction_stpx++;
112

113 114 115
	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

116
	operand2 = kvm_s390_get_base_disp_s(vcpu);
117 118

	/* must be word boundary */
119 120
	if (operand2 & 3)
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
121 122 123 124 125

	address = vcpu->arch.sie_block->prefix;
	address = address & 0x7fffe000u;

	/* get the value */
126 127 128
	rc = write_guest(vcpu, operand2, &address, sizeof(address));
	if (rc)
		return kvm_s390_inject_prog_cond(vcpu, rc);
129 130

	VCPU_EVENT(vcpu, 5, "storing prefix to %x", address);
131
	trace_kvm_s390_handle_prefix(vcpu, 0, address);
132 133 134 135 136
	return 0;
}

static int handle_store_cpu_address(struct kvm_vcpu *vcpu)
{
137 138 139
	u16 vcpu_id = vcpu->vcpu_id;
	u64 ga;
	int rc;
140 141

	vcpu->stat.instruction_stap++;
142

143 144 145
	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

146
	ga = kvm_s390_get_base_disp_s(vcpu);
147

148
	if (ga & 1)
149
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
150

151 152 153
	rc = write_guest(vcpu, ga, &vcpu_id, sizeof(vcpu_id));
	if (rc)
		return kvm_s390_inject_prog_cond(vcpu, rc);
154

155 156
	VCPU_EVENT(vcpu, 5, "storing cpu address to %llx", ga);
	trace_kvm_s390_handle_stap(vcpu, ga);
157 158 159
	return 0;
}

160 161 162 163 164 165 166 167 168 169 170
static void __skey_check_enable(struct kvm_vcpu *vcpu)
{
	if (!(vcpu->arch.sie_block->ictl & (ICTL_ISKE | ICTL_SSKE | ICTL_RRBE)))
		return;

	s390_enable_skey();
	trace_kvm_s390_skey_related_inst(vcpu);
	vcpu->arch.sie_block->ictl &= ~(ICTL_ISKE | ICTL_SSKE | ICTL_RRBE);
}


171 172
static int handle_skey(struct kvm_vcpu *vcpu)
{
173 174
	__skey_check_enable(vcpu);

175
	vcpu->stat.instruction_storage_key++;
176 177 178 179

	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

180 181
	vcpu->arch.sie_block->gpsw.addr =
		__rewind_psw(vcpu->arch.sie_block->gpsw, 4);
182 183 184 185
	VCPU_EVENT(vcpu, 4, "%s", "retrying storage key operation");
	return 0;
}

186 187 188 189 190 191 192 193 194 195 196 197 198
static int handle_ipte_interlock(struct kvm_vcpu *vcpu)
{
	psw_t *psw = &vcpu->arch.sie_block->gpsw;

	vcpu->stat.instruction_ipte_interlock++;
	if (psw_bits(*psw).p)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);
	wait_event(vcpu->kvm->arch.ipte_wq, !ipte_lock_held(vcpu));
	psw->addr = __rewind_psw(*psw, 4);
	VCPU_EVENT(vcpu, 4, "%s", "retrying ipte interlock operation");
	return 0;
}

T
Thomas Huth 已提交
199 200 201 202 203 204 205 206 207 208 209 210
static int handle_test_block(struct kvm_vcpu *vcpu)
{
	gpa_t addr;
	int reg2;

	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

	kvm_s390_get_regs_rre(vcpu, NULL, &reg2);
	addr = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK;
	addr = kvm_s390_real_to_abs(vcpu, addr);

211
	if (kvm_is_error_gpa(vcpu->kvm, addr))
T
Thomas Huth 已提交
212 213 214 215 216
		return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
	/*
	 * We don't expect errors on modern systems, and do not care
	 * about storage keys (yet), so let's just clear the page.
	 */
217
	if (kvm_clear_guest(vcpu->kvm, addr, PAGE_SIZE))
T
Thomas Huth 已提交
218 219 220 221 222 223
		return -EFAULT;
	kvm_s390_set_psw_cc(vcpu, 0);
	vcpu->run->s.regs.gprs[0] = 0;
	return 0;
}

224
static int handle_tpi(struct kvm_vcpu *vcpu)
225
{
226
	struct kvm_s390_interrupt_info *inti;
H
Heiko Carstens 已提交
227 228 229
	unsigned long len;
	u32 tpi_data[3];
	int cc, rc;
230
	u64 addr;
231

H
Heiko Carstens 已提交
232
	rc = 0;
233
	addr = kvm_s390_get_base_disp_s(vcpu);
234 235
	if (addr & 3)
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
236
	cc = 0;
237
	inti = kvm_s390_get_io_int(vcpu->kvm, vcpu->arch.sie_block->gcr[6], 0);
238 239 240
	if (!inti)
		goto no_interrupt;
	cc = 1;
H
Heiko Carstens 已提交
241 242 243
	tpi_data[0] = inti->io.subchannel_id << 16 | inti->io.subchannel_nr;
	tpi_data[1] = inti->io.io_int_parm;
	tpi_data[2] = inti->io.io_int_word;
244 245 246 247 248
	if (addr) {
		/*
		 * Store the two-word I/O interruption code into the
		 * provided area.
		 */
H
Heiko Carstens 已提交
249 250 251 252
		len = sizeof(tpi_data) - 4;
		rc = write_guest(vcpu, addr, &tpi_data, len);
		if (rc)
			return kvm_s390_inject_prog_cond(vcpu, rc);
253 254 255 256 257
	} else {
		/*
		 * Store the three-word I/O interruption code into
		 * the appropriate lowcore area.
		 */
H
Heiko Carstens 已提交
258 259 260
		len = sizeof(tpi_data);
		if (write_guest_lc(vcpu, __LC_SUBCHANNEL_ID, &tpi_data, len))
			rc = -EFAULT;
261
	}
262 263 264 265 266 267 268 269 270
	/*
	 * If we encounter a problem storing the interruption code, the
	 * instruction is suppressed from the guest's view: reinject the
	 * interrupt.
	 */
	if (!rc)
		kfree(inti);
	else
		kvm_s390_reinject_io_int(vcpu->kvm, inti);
271
no_interrupt:
272
	/* Set condition code and we're done. */
H
Heiko Carstens 已提交
273 274 275
	if (!rc)
		kvm_s390_set_psw_cc(vcpu, cc);
	return rc ? -EFAULT : 0;
276 277
}

278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309
static int handle_tsch(struct kvm_vcpu *vcpu)
{
	struct kvm_s390_interrupt_info *inti;

	inti = kvm_s390_get_io_int(vcpu->kvm, 0,
				   vcpu->run->s.regs.gprs[1]);

	/*
	 * Prepare exit to userspace.
	 * We indicate whether we dequeued a pending I/O interrupt
	 * so that userspace can re-inject it if the instruction gets
	 * a program check. While this may re-order the pending I/O
	 * interrupts, this is no problem since the priority is kept
	 * intact.
	 */
	vcpu->run->exit_reason = KVM_EXIT_S390_TSCH;
	vcpu->run->s390_tsch.dequeued = !!inti;
	if (inti) {
		vcpu->run->s390_tsch.subchannel_id = inti->io.subchannel_id;
		vcpu->run->s390_tsch.subchannel_nr = inti->io.subchannel_nr;
		vcpu->run->s390_tsch.io_int_parm = inti->io.io_int_parm;
		vcpu->run->s390_tsch.io_int_word = inti->io.io_int_word;
	}
	vcpu->run->s390_tsch.ipb = vcpu->arch.sie_block->ipb;
	kfree(inti);
	return -EREMOTE;
}

static int handle_io_inst(struct kvm_vcpu *vcpu)
{
	VCPU_EVENT(vcpu, 4, "%s", "I/O instruction");

310 311 312
	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

313 314 315 316 317 318 319 320 321 322 323 324 325
	if (vcpu->kvm->arch.css_support) {
		/*
		 * Most I/O instructions will be handled by userspace.
		 * Exceptions are tpi and the interrupt portion of tsch.
		 */
		if (vcpu->arch.sie_block->ipa == 0xb236)
			return handle_tpi(vcpu);
		if (vcpu->arch.sie_block->ipa == 0xb235)
			return handle_tsch(vcpu);
		/* Handle in userspace. */
		return -EOPNOTSUPP;
	} else {
		/*
326
		 * Set condition code 3 to stop the guest from issuing channel
327 328
		 * I/O instructions.
		 */
329
		kvm_s390_set_psw_cc(vcpu, 3);
330 331 332 333
		return 0;
	}
}

334 335 336 337 338
static int handle_stfl(struct kvm_vcpu *vcpu)
{
	int rc;

	vcpu->stat.instruction_stfl++;
339 340 341 342

	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

343 344
	rc = write_guest_lc(vcpu, offsetof(struct _lowcore, stfl_fac_list),
			    vfacilities, 4);
345
	if (rc)
346
		return rc;
347 348 349
	VCPU_EVENT(vcpu, 5, "store facility list value %x",
		   *(unsigned int *) vfacilities);
	trace_kvm_s390_handle_stfl(vcpu, *(unsigned int *) vfacilities);
350 351 352
	return 0;
}

353 354 355 356 357 358 359 360 361
static void handle_new_psw(struct kvm_vcpu *vcpu)
{
	/* Check whether the new psw is enabled for machine checks. */
	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_MCHECK)
		kvm_s390_deliver_pending_machine_checks(vcpu);
}

#define PSW_MASK_ADDR_MODE (PSW_MASK_EA | PSW_MASK_BA)
#define PSW_MASK_UNASSIGNED 0xb80800fe7fffffffUL
362
#define PSW_ADDR_24 0x0000000000ffffffUL
363 364
#define PSW_ADDR_31 0x000000007fffffffUL

365 366 367 368 369 370 371 372 373 374 375 376 377 378
static int is_valid_psw(psw_t *psw) {
	if (psw->mask & PSW_MASK_UNASSIGNED)
		return 0;
	if ((psw->mask & PSW_MASK_ADDR_MODE) == PSW_MASK_BA) {
		if (psw->addr & ~PSW_ADDR_31)
			return 0;
	}
	if (!(psw->mask & PSW_MASK_ADDR_MODE) && (psw->addr & ~PSW_ADDR_24))
		return 0;
	if ((psw->mask & PSW_MASK_ADDR_MODE) ==  PSW_MASK_EA)
		return 0;
	return 1;
}

379 380
int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu)
{
381
	psw_t *gpsw = &vcpu->arch.sie_block->gpsw;
382
	psw_compat_t new_psw;
383
	u64 addr;
384
	int rc;
385

386
	if (gpsw->mask & PSW_MASK_PSTATE)
387 388
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

389
	addr = kvm_s390_get_base_disp_s(vcpu);
390 391
	if (addr & 7)
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
392 393 394 395

	rc = read_guest(vcpu, addr, &new_psw, sizeof(new_psw));
	if (rc)
		return kvm_s390_inject_prog_cond(vcpu, rc);
396 397
	if (!(new_psw.mask & PSW32_MASK_BASE))
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
398 399 400 401
	gpsw->mask = (new_psw.mask & ~PSW32_MASK_BASE) << 32;
	gpsw->mask |= new_psw.addr & PSW32_ADDR_AMODE;
	gpsw->addr = new_psw.addr & ~PSW32_ADDR_AMODE;
	if (!is_valid_psw(gpsw))
402
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
403 404 405 406 407 408 409
	handle_new_psw(vcpu);
	return 0;
}

static int handle_lpswe(struct kvm_vcpu *vcpu)
{
	psw_t new_psw;
410
	u64 addr;
411
	int rc;
412

413 414 415
	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

416
	addr = kvm_s390_get_base_disp_s(vcpu);
417 418
	if (addr & 7)
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
419 420 421
	rc = read_guest(vcpu, addr, &new_psw, sizeof(new_psw));
	if (rc)
		return kvm_s390_inject_prog_cond(vcpu, rc);
422 423
	vcpu->arch.sie_block->gpsw = new_psw;
	if (!is_valid_psw(&vcpu->arch.sie_block->gpsw))
424
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
425 426 427 428
	handle_new_psw(vcpu);
	return 0;
}

429 430
static int handle_stidp(struct kvm_vcpu *vcpu)
{
431
	u64 stidp_data = vcpu->arch.stidp_data;
432
	u64 operand2;
433
	int rc;
434 435

	vcpu->stat.instruction_stidp++;
436

437 438 439
	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

440
	operand2 = kvm_s390_get_base_disp_s(vcpu);
441

442 443
	if (operand2 & 7)
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
444

445 446 447
	rc = write_guest(vcpu, operand2, &stidp_data, sizeof(stidp_data));
	if (rc)
		return kvm_s390_inject_prog_cond(vcpu, rc);
448 449 450 451 452 453 454 455 456 457

	VCPU_EVENT(vcpu, 5, "%s", "store cpu id");
	return 0;
}

static void handle_stsi_3_2_2(struct kvm_vcpu *vcpu, struct sysinfo_3_2_2 *mem)
{
	int cpus = 0;
	int n;

458
	cpus = atomic_read(&vcpu->kvm->online_vcpus);
459 460

	/* deal with other level 3 hypervisors */
461
	if (stsi(mem, 3, 2, 2))
462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480
		mem->count = 0;
	if (mem->count < 8)
		mem->count++;
	for (n = mem->count - 1; n > 0 ; n--)
		memcpy(&mem->vm[n], &mem->vm[n - 1], sizeof(mem->vm[0]));

	mem->vm[0].cpus_total = cpus;
	mem->vm[0].cpus_configured = cpus;
	mem->vm[0].cpus_standby = 0;
	mem->vm[0].cpus_reserved = 0;
	mem->vm[0].caf = 1000;
	memcpy(mem->vm[0].name, "KVMguest", 8);
	ASCEBC(mem->vm[0].name, 8);
	memcpy(mem->vm[0].cpi, "KVM/Linux       ", 16);
	ASCEBC(mem->vm[0].cpi, 16);
}

static int handle_stsi(struct kvm_vcpu *vcpu)
{
481 482 483
	int fc = (vcpu->run->s.regs.gprs[0] & 0xf0000000) >> 28;
	int sel1 = vcpu->run->s.regs.gprs[0] & 0xff;
	int sel2 = vcpu->run->s.regs.gprs[1] & 0xffff;
484
	unsigned long mem = 0;
485
	u64 operand2;
486
	int rc = 0;
487 488 489 490

	vcpu->stat.instruction_stsi++;
	VCPU_EVENT(vcpu, 4, "stsi: fc: %x sel1: %x sel2: %x", fc, sel1, sel2);

491 492 493
	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

494
	if (fc > 3) {
495
		kvm_s390_set_psw_cc(vcpu, 3);
496 497
		return 0;
	}
498

499 500
	if (vcpu->run->s.regs.gprs[0] & 0x0fffff00
	    || vcpu->run->s.regs.gprs[1] & 0xffff0000)
501 502
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);

503
	if (fc == 0) {
504
		vcpu->run->s.regs.gprs[0] = 3 << 28;
505
		kvm_s390_set_psw_cc(vcpu, 0);
506
		return 0;
507 508 509 510 511 512 513 514
	}

	operand2 = kvm_s390_get_base_disp_s(vcpu);

	if (operand2 & 0xfff)
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);

	switch (fc) {
515 516 517 518
	case 1: /* same handling for 1 and 2 */
	case 2:
		mem = get_zeroed_page(GFP_KERNEL);
		if (!mem)
519
			goto out_no_data;
520
		if (stsi((void *) mem, fc, sel1, sel2))
521
			goto out_no_data;
522 523 524
		break;
	case 3:
		if (sel1 != 2 || sel2 != 2)
525
			goto out_no_data;
526 527
		mem = get_zeroed_page(GFP_KERNEL);
		if (!mem)
528
			goto out_no_data;
529 530 531 532 533
		handle_stsi_3_2_2(vcpu, (void *) mem);
		break;
	}

	if (copy_to_guest_absolute(vcpu, operand2, (void *) mem, PAGE_SIZE)) {
534
		rc = kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
535
		goto out_exception;
536
	}
537
	trace_kvm_s390_handle_stsi(vcpu, fc, sel1, sel2, operand2);
538
	free_page(mem);
539
	kvm_s390_set_psw_cc(vcpu, 0);
540
	vcpu->run->s.regs.gprs[0] = 0;
541
	return 0;
542
out_no_data:
543
	kvm_s390_set_psw_cc(vcpu, 3);
544 545
out_exception:
	free_page(mem);
546
	return rc;
547 548
}

549
static const intercept_handler_t b2_handlers[256] = {
550
	[0x02] = handle_stidp,
551
	[0x04] = handle_set_clock,
552 553 554
	[0x10] = handle_set_prefix,
	[0x11] = handle_store_prefix,
	[0x12] = handle_store_cpu_address,
555
	[0x21] = handle_ipte_interlock,
556 557 558
	[0x29] = handle_skey,
	[0x2a] = handle_skey,
	[0x2b] = handle_skey,
T
Thomas Huth 已提交
559
	[0x2c] = handle_test_block,
560 561 562 563 564 565 566 567 568 569 570 571 572
	[0x30] = handle_io_inst,
	[0x31] = handle_io_inst,
	[0x32] = handle_io_inst,
	[0x33] = handle_io_inst,
	[0x34] = handle_io_inst,
	[0x35] = handle_io_inst,
	[0x36] = handle_io_inst,
	[0x37] = handle_io_inst,
	[0x38] = handle_io_inst,
	[0x39] = handle_io_inst,
	[0x3a] = handle_io_inst,
	[0x3b] = handle_io_inst,
	[0x3c] = handle_io_inst,
573
	[0x50] = handle_ipte_interlock,
574 575 576
	[0x5f] = handle_io_inst,
	[0x74] = handle_io_inst,
	[0x76] = handle_io_inst,
577 578
	[0x7d] = handle_stsi,
	[0xb1] = handle_stfl,
579
	[0xb2] = handle_lpswe,
580 581
};

582
int kvm_s390_handle_b2(struct kvm_vcpu *vcpu)
583 584 585
{
	intercept_handler_t handler;

586
	/*
587 588 589 590
	 * A lot of B2 instructions are priviledged. Here we check for
	 * the privileged ones, that we can handle in the kernel.
	 * Anything else goes to userspace.
	 */
591
	handler = b2_handlers[vcpu->arch.sie_block->ipa & 0x00ff];
592 593 594
	if (handler)
		return handler(vcpu);

595
	return -EOPNOTSUPP;
596
}
597

598 599 600 601
static int handle_epsw(struct kvm_vcpu *vcpu)
{
	int reg1, reg2;

602
	kvm_s390_get_regs_rre(vcpu, &reg1, &reg2);
603 604

	/* This basically extracts the mask half of the psw. */
605
	vcpu->run->s.regs.gprs[reg1] &= 0xffffffff00000000UL;
606 607
	vcpu->run->s.regs.gprs[reg1] |= vcpu->arch.sie_block->gpsw.mask >> 32;
	if (reg2) {
608
		vcpu->run->s.regs.gprs[reg2] &= 0xffffffff00000000UL;
609
		vcpu->run->s.regs.gprs[reg2] |=
610
			vcpu->arch.sie_block->gpsw.mask & 0x00000000ffffffffUL;
611 612 613 614
	}
	return 0;
}

615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637
#define PFMF_RESERVED   0xfffc0101UL
#define PFMF_SK         0x00020000UL
#define PFMF_CF         0x00010000UL
#define PFMF_UI         0x00008000UL
#define PFMF_FSC        0x00007000UL
#define PFMF_NQ         0x00000800UL
#define PFMF_MR         0x00000400UL
#define PFMF_MC         0x00000200UL
#define PFMF_KEY        0x000000feUL

static int handle_pfmf(struct kvm_vcpu *vcpu)
{
	int reg1, reg2;
	unsigned long start, end;

	vcpu->stat.instruction_pfmf++;

	kvm_s390_get_regs_rre(vcpu, &reg1, &reg2);

	if (!MACHINE_HAS_PFMF)
		return kvm_s390_inject_program_int(vcpu, PGM_OPERATION);

	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
638
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);
639 640 641 642 643

	if (vcpu->run->s.regs.gprs[reg1] & PFMF_RESERVED)
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);

	/* Only provide non-quiescing support if the host supports it */
644
	if (vcpu->run->s.regs.gprs[reg1] & PFMF_NQ && !test_facility(14))
645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);

	/* No support for conditional-SSKE */
	if (vcpu->run->s.regs.gprs[reg1] & (PFMF_MR | PFMF_MC))
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);

	start = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK;
	switch (vcpu->run->s.regs.gprs[reg1] & PFMF_FSC) {
	case 0x00000000:
		end = (start + (1UL << 12)) & ~((1UL << 12) - 1);
		break;
	case 0x00001000:
		end = (start + (1UL << 20)) & ~((1UL << 20) - 1);
		break;
	/* We dont support EDAT2
	case 0x00002000:
		end = (start + (1UL << 31)) & ~((1UL << 31) - 1);
		break;*/
	default:
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
	}
	while (start < end) {
		unsigned long useraddr;

		useraddr = gmap_translate(start, vcpu->arch.gmap);
		if (IS_ERR((void *)useraddr))
			return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);

		if (vcpu->run->s.regs.gprs[reg1] & PFMF_CF) {
			if (clear_user((void __user *)useraddr, PAGE_SIZE))
				return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
		}

		if (vcpu->run->s.regs.gprs[reg1] & PFMF_SK) {
679
			__skey_check_enable(vcpu);
680 681 682 683 684 685 686 687 688 689 690 691 692
			if (set_guest_storage_key(current->mm, useraddr,
					vcpu->run->s.regs.gprs[reg1] & PFMF_KEY,
					vcpu->run->s.regs.gprs[reg1] & PFMF_NQ))
				return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
		}

		start += PAGE_SIZE;
	}
	if (vcpu->run->s.regs.gprs[reg1] & PFMF_FSC)
		vcpu->run->s.regs.gprs[reg2] = end;
	return 0;
}

693 694 695 696 697 698 699 700 701 702 703
static int handle_essa(struct kvm_vcpu *vcpu)
{
	/* entries expected to be 1FF */
	int entries = (vcpu->arch.sie_block->cbrlo & ~PAGE_MASK) >> 3;
	unsigned long *cbrlo, cbrle;
	struct gmap *gmap;
	int i;

	VCPU_EVENT(vcpu, 5, "cmma release %d pages", entries);
	gmap = vcpu->arch.gmap;
	vcpu->stat.instruction_essa++;
704
	if (!kvm_s390_cmma_enabled(vcpu->kvm))
705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732
		return kvm_s390_inject_program_int(vcpu, PGM_OPERATION);

	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

	if (((vcpu->arch.sie_block->ipb & 0xf0000000) >> 28) > 6)
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);

	/* Rewind PSW to repeat the ESSA instruction */
	vcpu->arch.sie_block->gpsw.addr =
		__rewind_psw(vcpu->arch.sie_block->gpsw, 4);
	vcpu->arch.sie_block->cbrlo &= PAGE_MASK;	/* reset nceo */
	cbrlo = phys_to_virt(vcpu->arch.sie_block->cbrlo);
	down_read(&gmap->mm->mmap_sem);
	for (i = 0; i < entries; ++i) {
		cbrle = cbrlo[i];
		if (unlikely(cbrle & ~PAGE_MASK || cbrle < 2 * PAGE_SIZE))
			/* invalid entry */
			break;
		/* try to free backing */
		__gmap_zap(cbrle, gmap);
	}
	up_read(&gmap->mm->mmap_sem);
	if (i < entries)
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
	return 0;
}

733
static const intercept_handler_t b9_handlers[256] = {
734
	[0x8a] = handle_ipte_interlock,
735
	[0x8d] = handle_epsw,
736 737
	[0x8e] = handle_ipte_interlock,
	[0x8f] = handle_ipte_interlock,
738
	[0xab] = handle_essa,
739
	[0xaf] = handle_pfmf,
740 741 742 743 744 745 746 747
};

int kvm_s390_handle_b9(struct kvm_vcpu *vcpu)
{
	intercept_handler_t handler;

	/* This is handled just as for the B2 instructions. */
	handler = b9_handlers[vcpu->arch.sie_block->ipa & 0x00ff];
748 749 750
	if (handler)
		return handler(vcpu);

751 752 753
	return -EOPNOTSUPP;
}

754 755 756 757 758 759
int kvm_s390_handle_lctl(struct kvm_vcpu *vcpu)
{
	int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4;
	int reg3 = vcpu->arch.sie_block->ipa & 0x000f;
	u32 val = 0;
	int reg, rc;
760
	u64 ga;
761 762 763 764 765 766

	vcpu->stat.instruction_lctl++;

	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

767
	ga = kvm_s390_get_base_disp_rs(vcpu);
768

769
	if (ga & 3)
770 771
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);

772 773
	VCPU_EVENT(vcpu, 5, "lctl r1:%x, r3:%x, addr:%llx", reg1, reg3, ga);
	trace_kvm_s390_handle_lctl(vcpu, 0, reg1, reg3, ga);
774 775 776

	reg = reg1;
	do {
777
		rc = read_guest(vcpu, ga, &val, sizeof(val));
778
		if (rc)
779
			return kvm_s390_inject_prog_cond(vcpu, rc);
780 781
		vcpu->arch.sie_block->gcr[reg] &= 0xffffffff00000000ul;
		vcpu->arch.sie_block->gcr[reg] |= val;
782
		ga += 4;
783 784 785 786 787 788 789 790 791 792 793 794
		if (reg == reg3)
			break;
		reg = (reg + 1) % 16;
	} while (1);

	return 0;
}

static int handle_lctlg(struct kvm_vcpu *vcpu)
{
	int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4;
	int reg3 = vcpu->arch.sie_block->ipa & 0x000f;
795
	u64 ga, val;
796 797 798 799 800 801 802
	int reg, rc;

	vcpu->stat.instruction_lctlg++;

	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

803
	ga = kvm_s390_get_base_disp_rsy(vcpu);
804

805
	if (ga & 7)
806 807 808 809
		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);

	reg = reg1;

810 811
	VCPU_EVENT(vcpu, 5, "lctlg r1:%x, r3:%x, addr:%llx", reg1, reg3, ga);
	trace_kvm_s390_handle_lctl(vcpu, 1, reg1, reg3, ga);
812 813

	do {
814
		rc = read_guest(vcpu, ga, &val, sizeof(val));
815
		if (rc)
816 817 818
			return kvm_s390_inject_prog_cond(vcpu, rc);
		vcpu->arch.sie_block->gcr[reg] = val;
		ga += 8;
819 820 821 822 823 824 825 826
		if (reg == reg3)
			break;
		reg = (reg + 1) % 16;
	} while (1);

	return 0;
}

827
static const intercept_handler_t eb_handlers[256] = {
828
	[0x2f] = handle_lctlg,
829 830
};

831
int kvm_s390_handle_eb(struct kvm_vcpu *vcpu)
832 833 834 835 836 837 838 839 840
{
	intercept_handler_t handler;

	handler = eb_handlers[vcpu->arch.sie_block->ipb & 0xff];
	if (handler)
		return handler(vcpu);
	return -EOPNOTSUPP;
}

841 842
static int handle_tprot(struct kvm_vcpu *vcpu)
{
843
	u64 address1, address2;
844
	struct vm_area_struct *vma;
845
	unsigned long user_address;
846 847 848

	vcpu->stat.instruction_tprot++;

849 850 851
	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

852 853
	kvm_s390_get_base_disp_sse(vcpu, &address1, &address2);

854 855 856 857 858 859 860 861 862 863
	/* we only handle the Linux memory detection case:
	 * access key == 0
	 * guest DAT == off
	 * everything else goes to userspace. */
	if (address2 & 0xf0)
		return -EOPNOTSUPP;
	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_DAT)
		return -EOPNOTSUPP;

	down_read(&current->mm->mmap_sem);
864 865 866
	user_address = __gmap_translate(address1, vcpu->arch.gmap);
	if (IS_ERR_VALUE(user_address))
		goto out_inject;
867
	vma = find_vma(current->mm, user_address);
868 869
	if (!vma)
		goto out_inject;
870 871 872 873 874 875 876 877
	vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44);
	if (!(vma->vm_flags & VM_WRITE) && (vma->vm_flags & VM_READ))
		vcpu->arch.sie_block->gpsw.mask |= (1ul << 44);
	if (!(vma->vm_flags & VM_WRITE) && !(vma->vm_flags & VM_READ))
		vcpu->arch.sie_block->gpsw.mask |= (2ul << 44);

	up_read(&current->mm->mmap_sem);
	return 0;
878 879 880 881

out_inject:
	up_read(&current->mm->mmap_sem);
	return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
882 883 884 885 886 887 888 889 890 891
}

int kvm_s390_handle_e5(struct kvm_vcpu *vcpu)
{
	/* For e5xx... instructions we only handle TPROT */
	if ((vcpu->arch.sie_block->ipa & 0x00ff) == 0x01)
		return handle_tprot(vcpu);
	return -EOPNOTSUPP;
}

892 893 894 895 896
static int handle_sckpf(struct kvm_vcpu *vcpu)
{
	u32 value;

	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
897
		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);
898 899 900 901 902 903 904 905 906 907 908

	if (vcpu->run->s.regs.gprs[0] & 0x00000000ffff0000)
		return kvm_s390_inject_program_int(vcpu,
						   PGM_SPECIFICATION);

	value = vcpu->run->s.regs.gprs[0] & 0x000000000000ffff;
	vcpu->arch.sie_block->todpr = value;

	return 0;
}

909
static const intercept_handler_t x01_handlers[256] = {
910 911 912 913 914 915 916 917 918 919 920 921
	[0x07] = handle_sckpf,
};

int kvm_s390_handle_01(struct kvm_vcpu *vcpu)
{
	intercept_handler_t handler;

	handler = x01_handlers[vcpu->arch.sie_block->ipa & 0x00ff];
	if (handler)
		return handler(vcpu);
	return -EOPNOTSUPP;
}