switch.c 17.9 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 * Copyright (C) 2015 - ARM Ltd
 * Author: Marc Zyngier <marc.zyngier@arm.com>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 */

18
#include <linux/arm-smccc.h>
19
#include <linux/types.h>
20
#include <linux/jump_label.h>
21
#include <uapi/linux/psci.h>
22

23 24
#include <kvm/arm_psci.h>

25
#include <asm/arch_gicv3.h>
26
#include <asm/cpufeature.h>
27
#include <asm/kprobes.h>
28
#include <asm/kvm_asm.h>
29
#include <asm/kvm_emulate.h>
30
#include <asm/kvm_host.h>
31
#include <asm/kvm_hyp.h>
32
#include <asm/kvm_mmu.h>
33
#include <asm/fpsimd.h>
34
#include <asm/debug-monitors.h>
35
#include <asm/processor.h>
36
#include <asm/thread_info.h>
37

38 39
/* Check whether the FP regs were dirtied while in the host-side run loop: */
static bool __hyp_text update_fp_enabled(struct kvm_vcpu *vcpu)
40
{
41 42 43
	if (vcpu->arch.host_thread_info->flags & _TIF_FOREIGN_FPSTATE)
		vcpu->arch.flags &= ~(KVM_ARM64_FP_ENABLED |
				      KVM_ARM64_FP_HOST);
44

45
	return !!(vcpu->arch.flags & KVM_ARM64_FP_ENABLED);
46 47
}

48 49 50 51 52 53 54 55 56
/* Save the 32-bit only FPSIMD system register state */
static void __hyp_text __fpsimd_save_fpexc32(struct kvm_vcpu *vcpu)
{
	if (!vcpu_el1_is_32bit(vcpu))
		return;

	vcpu->arch.ctxt.sys_regs[FPEXC32_EL2] = read_sysreg(fpexc32_el2);
}

57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
static void __hyp_text __activate_traps_fpsimd32(struct kvm_vcpu *vcpu)
{
	/*
	 * We are about to set CPTR_EL2.TFP to trap all floating point
	 * register accesses to EL2, however, the ARM ARM clearly states that
	 * traps are only taken to EL2 if the operation would not otherwise
	 * trap to EL1.  Therefore, always make sure that for 32-bit guests,
	 * we set FPEXC.EN to prevent traps to EL1, when setting the TFP bit.
	 * If FP/ASIMD is not implemented, FPEXC is UNDEFINED and any access to
	 * it will cause an exception.
	 */
	if (vcpu_el1_is_32bit(vcpu) && system_supports_fpsimd()) {
		write_sysreg(1 << 30, fpexc32_el2);
		isb();
	}
}

static void __hyp_text __activate_traps_common(struct kvm_vcpu *vcpu)
{
	/* Trap on AArch32 cp15 c15 (impdef sysregs) accesses (EL1 or EL0) */
	write_sysreg(1 << 15, hstr_el2);

	/*
	 * Make sure we trap PMU access from EL0 to EL2. Also sanitize
	 * PMSELR_EL0 to make sure it never contains the cycle
	 * counter, which could make a PMXEVCNTR_EL0 access UNDEF at
	 * EL1 instead of being trapped to EL2.
	 */
	write_sysreg(0, pmselr_el0);
	write_sysreg(ARMV8_PMU_USERENR_MASK, pmuserenr_el0);
	write_sysreg(vcpu->arch.mdcr_el2, mdcr_el2);
}

static void __hyp_text __deactivate_traps_common(void)
{
	write_sysreg(0, hstr_el2);
	write_sysreg(0, pmuserenr_el0);
}

96
static void activate_traps_vhe(struct kvm_vcpu *vcpu)
97 98 99 100 101
{
	u64 val;

	val = read_sysreg(cpacr_el1);
	val |= CPACR_EL1_TTA;
102
	val &= ~CPACR_EL1_ZEN;
103 104 105 106
	if (update_fp_enabled(vcpu)) {
		if (vcpu_has_sve(vcpu))
			val |= CPACR_EL1_ZEN;
	} else {
107
		val &= ~CPACR_EL1_FPEN;
108 109
		__activate_traps_fpsimd32(vcpu);
	}
110

111 112
	write_sysreg(val, cpacr_el1);

113
	write_sysreg(kvm_get_hyp_vector(), vbar_el1);
114
}
115
NOKPROBE_SYMBOL(activate_traps_vhe);
116

117
static void __hyp_text __activate_traps_nvhe(struct kvm_vcpu *vcpu)
118 119 120
{
	u64 val;

121 122
	__activate_traps_common(vcpu);

123
	val = CPTR_EL2_DEFAULT;
124
	val |= CPTR_EL2_TTA | CPTR_EL2_TZ;
125
	if (!update_fp_enabled(vcpu)) {
126
		val |= CPTR_EL2_TFP;
127 128
		__activate_traps_fpsimd32(vcpu);
	}
129

130 131 132
	write_sysreg(val, cptr_el2);
}

133 134
static void __hyp_text __activate_traps(struct kvm_vcpu *vcpu)
{
135
	u64 hcr = vcpu->arch.hcr_el2;
136

137
	write_sysreg(hcr, hcr_el2);
138

139
	if (cpus_have_const_cap(ARM64_HAS_RAS_EXTN) && (hcr & HCR_VSE))
140 141
		write_sysreg_s(vcpu->arch.vsesr_el2, SYS_VSESR_EL2);

142 143 144 145
	if (has_vhe())
		activate_traps_vhe(vcpu);
	else
		__activate_traps_nvhe(vcpu);
146
}
147

148
static void deactivate_traps_vhe(void)
149 150 151
{
	extern char vectors[];	/* kernel exception vectors */
	write_sysreg(HCR_HOST_VHE_FLAGS, hcr_el2);
152 153 154 155 156 157 158 159

	/*
	 * ARM erratum 1165522 requires the actual execution of the above
	 * before we can switch to the EL2/EL0 translation regime used by
	 * the host.
	 */
	asm(ALTERNATIVE("nop", "isb", ARM64_WORKAROUND_1165522));

160
	write_sysreg(CPACR_EL1_DEFAULT, cpacr_el1);
161
	write_sysreg(vectors, vbar_el1);
162
}
163
NOKPROBE_SYMBOL(deactivate_traps_vhe);
164

165
static void __hyp_text __deactivate_traps_nvhe(void)
166
{
167 168
	u64 mdcr_el2 = read_sysreg(mdcr_el2);

169 170
	__deactivate_traps_common();

171 172 173 174
	mdcr_el2 &= MDCR_EL2_HPMN_MASK;
	mdcr_el2 |= MDCR_EL2_E2PB_MASK << MDCR_EL2_E2PB_SHIFT;

	write_sysreg(mdcr_el2, mdcr_el2);
175
	write_sysreg(HCR_HOST_NVHE_FLAGS, hcr_el2);
176 177 178 179 180
	write_sysreg(CPTR_EL2_DEFAULT, cptr_el2);
}

static void __hyp_text __deactivate_traps(struct kvm_vcpu *vcpu)
{
181 182 183 184 185 186 187 188 189
	/*
	 * If we pended a virtual abort, preserve it until it gets
	 * cleared. See D1.14.3 (Virtual Interrupts) for details, but
	 * the crucial bit is "On taking a vSError interrupt,
	 * HCR_EL2.VSE is cleared to 0."
	 */
	if (vcpu->arch.hcr_el2 & HCR_VSE)
		vcpu->arch.hcr_el2 = read_sysreg(hcr_el2);

190 191 192 193
	if (has_vhe())
		deactivate_traps_vhe();
	else
		__deactivate_traps_nvhe();
194 195
}

196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213
void activate_traps_vhe_load(struct kvm_vcpu *vcpu)
{
	__activate_traps_common(vcpu);
}

void deactivate_traps_vhe_put(void)
{
	u64 mdcr_el2 = read_sysreg(mdcr_el2);

	mdcr_el2 &= MDCR_EL2_HPMN_MASK |
		    MDCR_EL2_E2PB_MASK << MDCR_EL2_E2PB_SHIFT |
		    MDCR_EL2_TPMS;

	write_sysreg(mdcr_el2, mdcr_el2);

	__deactivate_traps_common();
}

214
static void __hyp_text __activate_vm(struct kvm *kvm)
215
{
216
	__load_guest_stage2(kvm);
217 218 219 220 221 222 223
}

static void __hyp_text __deactivate_vm(struct kvm_vcpu *vcpu)
{
	write_sysreg(0, vttbr_el2);
}

224 225
/* Save VGICv3 state on non-VHE systems */
static void __hyp_text __hyp_vgic_save_state(struct kvm_vcpu *vcpu)
226
{
227
	if (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif)) {
228
		__vgic_v3_save_state(vcpu);
229 230
		__vgic_v3_deactivate_traps(vcpu);
	}
231 232
}

233 234
/* Restore VGICv3 state on non_VEH systems */
static void __hyp_text __hyp_vgic_restore_state(struct kvm_vcpu *vcpu)
235
{
236 237
	if (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif)) {
		__vgic_v3_activate_traps(vcpu);
238
		__vgic_v3_restore_state(vcpu);
239
	}
240 241
}

242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280
static bool __hyp_text __true_value(void)
{
	return true;
}

static bool __hyp_text __false_value(void)
{
	return false;
}

static hyp_alternate_select(__check_arm_834220,
			    __false_value, __true_value,
			    ARM64_WORKAROUND_834220);

static bool __hyp_text __translate_far_to_hpfar(u64 far, u64 *hpfar)
{
	u64 par, tmp;

	/*
	 * Resolve the IPA the hard way using the guest VA.
	 *
	 * Stage-1 translation already validated the memory access
	 * rights. As such, we can use the EL1 translation regime, and
	 * don't have to distinguish between EL0 and EL1 access.
	 *
	 * We do need to save/restore PAR_EL1 though, as we haven't
	 * saved the guest context yet, and we may return early...
	 */
	par = read_sysreg(par_el1);
	asm volatile("at s1e1r, %0" : : "r" (far));
	isb();

	tmp = read_sysreg(par_el1);
	write_sysreg(par, par_el1);

	if (unlikely(tmp & 1))
		return false; /* Translation failed, back to guest */

	/* Convert PAR to HPFAR format */
281
	*hpfar = PAR_TO_HPFAR(tmp);
282 283 284 285 286
	return true;
}

static bool __hyp_text __populate_fault_info(struct kvm_vcpu *vcpu)
{
287 288
	u8 ec;
	u64 esr;
289 290
	u64 hpfar, far;

291 292
	esr = vcpu->arch.fault.esr_el2;
	ec = ESR_ELx_EC(esr);
293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322

	if (ec != ESR_ELx_EC_DABT_LOW && ec != ESR_ELx_EC_IABT_LOW)
		return true;

	far = read_sysreg_el2(far);

	/*
	 * The HPFAR can be invalid if the stage 2 fault did not
	 * happen during a stage 1 page table walk (the ESR_EL2.S1PTW
	 * bit is clear) and one of the two following cases are true:
	 *   1. The fault was due to a permission fault
	 *   2. The processor carries errata 834220
	 *
	 * Therefore, for all non S1PTW faults where we either have a
	 * permission fault or the errata workaround is enabled, we
	 * resolve the IPA using the AT instruction.
	 */
	if (!(esr & ESR_ELx_S1PTW) &&
	    (__check_arm_834220()() || (esr & ESR_ELx_FSC_TYPE) == FSC_PERM)) {
		if (!__translate_far_to_hpfar(far, &hpfar))
			return false;
	} else {
		hpfar = read_sysreg(hpfar_el2);
	}

	vcpu->arch.fault.far_el2 = far;
	vcpu->arch.fault.hpfar_el2 = hpfar;
	return true;
}

323 324
/* Check for an FPSIMD/SVE trap and handle as appropriate */
static bool __hyp_text __hyp_handle_fpsimd(struct kvm_vcpu *vcpu)
325
{
326 327
	bool vhe, sve_guest, sve_host;
	u8 hsr_ec;
328

329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361
	if (!system_supports_fpsimd())
		return false;

	if (system_supports_sve()) {
		sve_guest = vcpu_has_sve(vcpu);
		sve_host = vcpu->arch.flags & KVM_ARM64_HOST_SVE_IN_USE;
		vhe = true;
	} else {
		sve_guest = false;
		sve_host = false;
		vhe = has_vhe();
	}

	hsr_ec = kvm_vcpu_trap_get_class(vcpu);
	if (hsr_ec != ESR_ELx_EC_FP_ASIMD &&
	    hsr_ec != ESR_ELx_EC_SVE)
		return false;

	/* Don't handle SVE traps for non-SVE vcpus here: */
	if (!sve_guest)
		if (hsr_ec != ESR_ELx_EC_FP_ASIMD)
			return false;

	/* Valid trap.  Switch the context: */

	if (vhe) {
		u64 reg = read_sysreg(cpacr_el1) | CPACR_EL1_FPEN;

		if (sve_guest)
			reg |= CPACR_EL1_ZEN;

		write_sysreg(reg, cpacr_el1);
	} else {
362 363
		write_sysreg(read_sysreg(cptr_el2) & ~(u64)CPTR_EL2_TFP,
			     cptr_el2);
364
	}
365 366 367

	isb();

368
	if (vcpu->arch.flags & KVM_ARM64_FP_HOST) {
369 370 371 372
		/*
		 * In the SVE case, VHE is assumed: it is enforced by
		 * Kconfig and kvm_arch_init().
		 */
373
		if (sve_host) {
374
			struct thread_struct *thread = container_of(
375
				vcpu->arch.host_fpsimd_state,
376 377
				struct thread_struct, uw.fpsimd_state);

378 379
			sve_save_state(sve_pffr(thread),
				       &vcpu->arch.host_fpsimd_state->fpsr);
380
		} else {
381
			__fpsimd_save_state(vcpu->arch.host_fpsimd_state);
382 383
		}

384 385 386
		vcpu->arch.flags &= ~KVM_ARM64_FP_HOST;
	}

387 388 389 390
	if (sve_guest) {
		sve_load_state(vcpu_sve_pffr(vcpu),
			       &vcpu->arch.ctxt.gp_regs.fp_regs.fpsr,
			       sve_vq_from_vl(vcpu->arch.sve_max_vl) - 1);
391
		write_sysreg_s(vcpu->arch.ctxt.sys_regs[ZCR_EL1], SYS_ZCR_EL12);
392 393 394
	} else {
		__fpsimd_restore_state(&vcpu->arch.ctxt.gp_regs.fp_regs);
	}
395

396 397 398 399
	/* Skip restoring fpexc32 for AArch64 guests */
	if (!(read_sysreg(hcr_el2) & HCR_RW))
		write_sysreg(vcpu->arch.ctxt.sys_regs[FPEXC32_EL2],
			     fpexc32_el2);
400 401

	vcpu->arch.flags |= KVM_ARM64_FP_ENABLED;
402 403

	return true;
404 405
}

406 407 408 409 410 411
/*
 * Return true when we were able to fixup the guest exit and should return to
 * the guest, false when we should restore the host state and return to the
 * main run loop.
 */
static bool __hyp_text fixup_guest_exit(struct kvm_vcpu *vcpu, u64 *exit_code)
412
{
413
	if (ARM_EXCEPTION_CODE(*exit_code) != ARM_EXCEPTION_IRQ)
414
		vcpu->arch.fault.esr_el2 = read_sysreg_el2(esr);
415

416 417 418 419 420 421
	/*
	 * We're using the raw exception code in order to only process
	 * the trap if no SError is pending. We will come back to the
	 * same PC once the SError has been injected, and replay the
	 * trapping instruction.
	 */
422 423 424
	if (*exit_code != ARM_EXCEPTION_TRAP)
		goto exit;

425 426 427 428 429
	/*
	 * We trap the first access to the FP/SIMD to save the host context
	 * and restore the guest context lazily.
	 * If FP/SIMD is not implemented, handle the trap and inject an
	 * undefined instruction exception to the guest.
430
	 * Similarly for trapped SVE accesses.
431
	 */
432 433
	if (__hyp_handle_fpsimd(vcpu))
		return true;
434

435
	if (!__populate_fault_info(vcpu))
436
		return true;
437

438
	if (static_branch_unlikely(&vgic_v2_cpuif_trap)) {
439 440 441 442 443 444 445 446
		bool valid;

		valid = kvm_vcpu_trap_get_class(vcpu) == ESR_ELx_EC_DABT_LOW &&
			kvm_vcpu_trap_get_fault_type(vcpu) == FSC_FAULT &&
			kvm_vcpu_dabt_isvalid(vcpu) &&
			!kvm_vcpu_dabt_isextabt(vcpu) &&
			!kvm_vcpu_dabt_iss1tw(vcpu);

447 448 449
		if (valid) {
			int ret = __vgic_v2_perform_cpuif_access(vcpu);

450
			if (ret == 1)
451
				return true;
452

453 454
			/* Promote an illegal access to an SError.*/
			if (ret == -1)
455
				*exit_code = ARM_EXCEPTION_EL1_SERROR;
456 457

			goto exit;
458 459 460
		}
	}

461 462 463 464 465
	if (static_branch_unlikely(&vgic_v3_cpuif_trap) &&
	    (kvm_vcpu_trap_get_class(vcpu) == ESR_ELx_EC_SYS64 ||
	     kvm_vcpu_trap_get_class(vcpu) == ESR_ELx_EC_CP15_32)) {
		int ret = __vgic_v3_perform_cpuif_access(vcpu);

466
		if (ret == 1)
467
			return true;
468 469
	}

470
exit:
471 472 473 474
	/* Return to the host kernel and handle the exit */
	return false;
}

475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507
static inline bool __hyp_text __needs_ssbd_off(struct kvm_vcpu *vcpu)
{
	if (!cpus_have_const_cap(ARM64_SSBD))
		return false;

	return !(vcpu->arch.workaround_flags & VCPU_WORKAROUND_2_FLAG);
}

static void __hyp_text __set_guest_arch_workaround_state(struct kvm_vcpu *vcpu)
{
#ifdef CONFIG_ARM64_SSBD
	/*
	 * The host runs with the workaround always present. If the
	 * guest wants it disabled, so be it...
	 */
	if (__needs_ssbd_off(vcpu) &&
	    __hyp_this_cpu_read(arm64_ssbd_callback_required))
		arm_smccc_1_1_smc(ARM_SMCCC_ARCH_WORKAROUND_2, 0, NULL);
#endif
}

static void __hyp_text __set_host_arch_workaround_state(struct kvm_vcpu *vcpu)
{
#ifdef CONFIG_ARM64_SSBD
	/*
	 * If the guest has disabled the workaround, bring it back on.
	 */
	if (__needs_ssbd_off(vcpu) &&
	    __hyp_this_cpu_read(arm64_ssbd_callback_required))
		arm_smccc_1_1_smc(ARM_SMCCC_ARCH_WORKAROUND_2, 1, NULL);
#endif
}

508 509 510 511 512 513 514
/* Switch to the guest for VHE systems running in EL2 */
int kvm_vcpu_run_vhe(struct kvm_vcpu *vcpu)
{
	struct kvm_cpu_context *host_ctxt;
	struct kvm_cpu_context *guest_ctxt;
	u64 exit_code;

515
	host_ctxt = vcpu->arch.host_cpu_context;
516 517 518
	host_ctxt->__hyp_running_vcpu = vcpu;
	guest_ctxt = &vcpu->arch.ctxt;

519
	sysreg_save_host_state_vhe(host_ctxt);
520

521 522 523 524 525 526 527 528 529 530 531
	/*
	 * ARM erratum 1165522 requires us to configure both stage 1 and
	 * stage 2 translation for the guest context before we clear
	 * HCR_EL2.TGE.
	 *
	 * We have already configured the guest's stage 1 translation in
	 * kvm_vcpu_load_sysregs above.  We must now call __activate_vm
	 * before __activate_traps, because __activate_vm configures
	 * stage 2 translation, and __activate_traps clear HCR_EL2.TGE
	 * (among other things).
	 */
532
	__activate_vm(vcpu->kvm);
533
	__activate_traps(vcpu);
534

535
	sysreg_restore_guest_state_vhe(guest_ctxt);
536 537
	__debug_switch_to_guest(vcpu);

538 539
	__set_guest_arch_workaround_state(vcpu);

540 541 542 543 544 545 546
	do {
		/* Jump in the fire! */
		exit_code = __guest_enter(vcpu, host_ctxt);

		/* And we're baaack! */
	} while (fixup_guest_exit(vcpu, &exit_code));

547 548
	__set_host_arch_workaround_state(vcpu);

549
	sysreg_save_guest_state_vhe(guest_ctxt);
550 551 552

	__deactivate_traps(vcpu);

553
	sysreg_restore_host_state_vhe(host_ctxt);
554

555
	if (vcpu->arch.flags & KVM_ARM64_FP_ENABLED)
556
		__fpsimd_save_fpexc32(vcpu);
557 558 559 560 561

	__debug_switch_to_host(vcpu);

	return exit_code;
}
562
NOKPROBE_SYMBOL(kvm_vcpu_run_vhe);
563 564 565

/* Switch to the guest for legacy non-VHE systems */
int __hyp_text __kvm_vcpu_run_nvhe(struct kvm_vcpu *vcpu)
566 567 568
{
	struct kvm_cpu_context *host_ctxt;
	struct kvm_cpu_context *guest_ctxt;
569
	bool pmu_switch_needed;
570 571
	u64 exit_code;

572 573 574 575 576 577 578 579 580 581 582
	/*
	 * Having IRQs masked via PMR when entering the guest means the GIC
	 * will not signal the CPU of interrupts of lower priority, and the
	 * only way to get out will be via guest exceptions.
	 * Naturally, we want to avoid this.
	 */
	if (system_uses_irq_prio_masking()) {
		gic_write_pmr(GIC_PRIO_IRQON);
		dsb(sy);
	}

583 584 585 586 587 588
	vcpu = kern_hyp_va(vcpu);

	host_ctxt = kern_hyp_va(vcpu->arch.host_cpu_context);
	host_ctxt->__hyp_running_vcpu = vcpu;
	guest_ctxt = &vcpu->arch.ctxt;

589 590
	pmu_switch_needed = __pmu_switch_to_guest(host_ctxt);

591
	__sysreg_save_state_nvhe(host_ctxt);
592

593
	__activate_vm(kern_hyp_va(vcpu->kvm));
594
	__activate_traps(vcpu);
595

596
	__hyp_vgic_restore_state(vcpu);
597 598 599 600 601 602 603
	__timer_enable_traps(vcpu);

	/*
	 * We must restore the 32-bit state before the sysregs, thanks
	 * to erratum #852523 (Cortex-A57) or #853709 (Cortex-A72).
	 */
	__sysreg32_restore_state(vcpu);
604
	__sysreg_restore_state_nvhe(guest_ctxt);
605 606
	__debug_switch_to_guest(vcpu);

607 608
	__set_guest_arch_workaround_state(vcpu);

609 610 611 612 613 614 615
	do {
		/* Jump in the fire! */
		exit_code = __guest_enter(vcpu, host_ctxt);

		/* And we're baaack! */
	} while (fixup_guest_exit(vcpu, &exit_code));

616 617
	__set_host_arch_workaround_state(vcpu);

618
	__sysreg_save_state_nvhe(guest_ctxt);
619
	__sysreg32_save_state(vcpu);
620
	__timer_disable_traps(vcpu);
621
	__hyp_vgic_save_state(vcpu);
622 623 624 625

	__deactivate_traps(vcpu);
	__deactivate_vm(vcpu);

626
	__sysreg_restore_state_nvhe(host_ctxt);
627

628
	if (vcpu->arch.flags & KVM_ARM64_FP_ENABLED)
629
		__fpsimd_save_fpexc32(vcpu);
630

631 632 633 634
	/*
	 * This must come after restoring the host sysregs, since a non-VHE
	 * system may enable SPE here and make use of the TTBRs.
	 */
635
	__debug_switch_to_host(vcpu);
636

637 638 639
	if (pmu_switch_needed)
		__pmu_switch_to_host(host_ctxt);

640 641 642 643
	/* Returning to host will clear PSR.I, remask PMR if needed */
	if (system_uses_irq_prio_masking())
		gic_write_pmr(GIC_PRIO_IRQOFF);

644 645
	return exit_code;
}
M
Marc Zyngier 已提交
646 647 648

static const char __hyp_panic_string[] = "HYP panic:\nPS:%08llx PC:%016llx ESR:%08llx\nFAR:%016llx HPFAR:%016llx PAR:%016llx\nVCPU:%p\n";

649
static void __hyp_text __hyp_call_panic_nvhe(u64 spsr, u64 elr, u64 par,
650
					     struct kvm_cpu_context *__host_ctxt)
M
Marc Zyngier 已提交
651
{
652
	struct kvm_vcpu *vcpu;
653
	unsigned long str_va;
654

655 656 657 658 659 660
	vcpu = __host_ctxt->__hyp_running_vcpu;

	if (read_sysreg(vttbr_el2)) {
		__timer_disable_traps(vcpu);
		__deactivate_traps(vcpu);
		__deactivate_vm(vcpu);
661
		__sysreg_restore_state_nvhe(__host_ctxt);
662 663
	}

664 665 666 667 668 669 670 671
	/*
	 * Force the panic string to be loaded from the literal pool,
	 * making sure it is a kernel address and not a PC-relative
	 * reference.
	 */
	asm volatile("ldr %0, =__hyp_panic_string" : "=r" (str_va));

	__hyp_do_panic(str_va,
672 673
		       spsr,  elr,
		       read_sysreg(esr_el2),   read_sysreg_el2(far),
674
		       read_sysreg(hpfar_el2), par, vcpu);
675 676
}

677 678
static void __hyp_call_panic_vhe(u64 spsr, u64 elr, u64 par,
				 struct kvm_cpu_context *host_ctxt)
679
{
680 681 682 683
	struct kvm_vcpu *vcpu;
	vcpu = host_ctxt->__hyp_running_vcpu;

	__deactivate_traps(vcpu);
684
	sysreg_restore_host_state_vhe(host_ctxt);
685

686 687 688
	panic(__hyp_panic_string,
	      spsr,  elr,
	      read_sysreg_el2(esr),   read_sysreg_el2(far),
689
	      read_sysreg(hpfar_el2), par, vcpu);
690
}
691
NOKPROBE_SYMBOL(__hyp_call_panic_vhe);
692

693
void __hyp_text __noreturn hyp_panic(struct kvm_cpu_context *host_ctxt)
694 695 696
{
	u64 spsr = read_sysreg_el2(spsr);
	u64 elr = read_sysreg_el2(elr);
M
Marc Zyngier 已提交
697 698
	u64 par = read_sysreg(par_el1);

699 700 701 702
	if (!has_vhe())
		__hyp_call_panic_nvhe(spsr, elr, par, host_ctxt);
	else
		__hyp_call_panic_vhe(spsr, elr, par, host_ctxt);
M
Marc Zyngier 已提交
703 704 705

	unreachable();
}