switch.c 15.2 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/types.h>
19
#include <linux/jump_label.h>
20
#include <uapi/linux/psci.h>
21

22 23
#include <kvm/arm_psci.h>

24
#include <asm/cpufeature.h>
25
#include <asm/kvm_asm.h>
26
#include <asm/kvm_emulate.h>
27
#include <asm/kvm_host.h>
28
#include <asm/kvm_hyp.h>
29
#include <asm/kvm_mmu.h>
30
#include <asm/fpsimd.h>
31
#include <asm/debug-monitors.h>
32
#include <asm/processor.h>
33
#include <asm/thread_info.h>
34

35 36
/* 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)
37
{
38 39 40
	if (vcpu->arch.host_thread_info->flags & _TIF_FOREIGN_FPSTATE)
		vcpu->arch.flags &= ~(KVM_ARM64_FP_ENABLED |
				      KVM_ARM64_FP_HOST);
41

42
	return !!(vcpu->arch.flags & KVM_ARM64_FP_ENABLED);
43 44
}

45 46 47 48 49 50 51 52 53
/* 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);
}

54 55 56 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
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);
}

93
static void activate_traps_vhe(struct kvm_vcpu *vcpu)
94 95 96 97 98
{
	u64 val;

	val = read_sysreg(cpacr_el1);
	val |= CPACR_EL1_TTA;
99 100 101 102
	val &= ~CPACR_EL1_ZEN;
	if (!update_fp_enabled(vcpu))
		val &= ~CPACR_EL1_FPEN;

103 104
	write_sysreg(val, cpacr_el1);

105
	write_sysreg(kvm_get_hyp_vector(), vbar_el1);
106 107
}

108
static void __hyp_text __activate_traps_nvhe(struct kvm_vcpu *vcpu)
109 110 111
{
	u64 val;

112 113
	__activate_traps_common(vcpu);

114
	val = CPTR_EL2_DEFAULT;
115 116 117 118
	val |= CPTR_EL2_TTA | CPTR_EL2_TZ;
	if (!update_fp_enabled(vcpu))
		val |= CPTR_EL2_TFP;

119 120 121
	write_sysreg(val, cptr_el2);
}

122 123
static void __hyp_text __activate_traps(struct kvm_vcpu *vcpu)
{
124
	u64 hcr = vcpu->arch.hcr_el2;
125

126
	write_sysreg(hcr, hcr_el2);
127

128
	if (cpus_have_const_cap(ARM64_HAS_RAS_EXTN) && (hcr & HCR_VSE))
129 130
		write_sysreg_s(vcpu->arch.vsesr_el2, SYS_VSESR_EL2);

131
	__activate_traps_fpsimd32(vcpu);
132 133 134 135
	if (has_vhe())
		activate_traps_vhe(vcpu);
	else
		__activate_traps_nvhe(vcpu);
136
}
137

138
static void deactivate_traps_vhe(void)
139 140 141
{
	extern char vectors[];	/* kernel exception vectors */
	write_sysreg(HCR_HOST_VHE_FLAGS, hcr_el2);
142
	write_sysreg(CPACR_EL1_DEFAULT, cpacr_el1);
143
	write_sysreg(vectors, vbar_el1);
144 145
}

146
static void __hyp_text __deactivate_traps_nvhe(void)
147
{
148 149
	u64 mdcr_el2 = read_sysreg(mdcr_el2);

150 151
	__deactivate_traps_common();

152 153 154 155
	mdcr_el2 &= MDCR_EL2_HPMN_MASK;
	mdcr_el2 |= MDCR_EL2_E2PB_MASK << MDCR_EL2_E2PB_SHIFT;

	write_sysreg(mdcr_el2, mdcr_el2);
156
	write_sysreg(HCR_RW, hcr_el2);
157 158 159 160 161
	write_sysreg(CPTR_EL2_DEFAULT, cptr_el2);
}

static void __hyp_text __deactivate_traps(struct kvm_vcpu *vcpu)
{
162 163 164 165 166 167 168 169 170
	/*
	 * 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);

171 172 173 174
	if (has_vhe())
		deactivate_traps_vhe();
	else
		__deactivate_traps_nvhe();
175 176
}

177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194
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();
}

195
static void __hyp_text __activate_vm(struct kvm *kvm)
196 197 198 199 200 201 202 203 204
{
	write_sysreg(kvm->arch.vttbr, vttbr_el2);
}

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

205 206
/* Save VGICv3 state on non-VHE systems */
static void __hyp_text __hyp_vgic_save_state(struct kvm_vcpu *vcpu)
207
{
208
	if (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif)) {
209
		__vgic_v3_save_state(vcpu);
210 211
		__vgic_v3_deactivate_traps(vcpu);
	}
212 213
}

214 215
/* Restore VGICv3 state on non_VEH systems */
static void __hyp_text __hyp_vgic_restore_state(struct kvm_vcpu *vcpu)
216
{
217 218
	if (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif)) {
		__vgic_v3_activate_traps(vcpu);
219
		__vgic_v3_restore_state(vcpu);
220
	}
221 222
}

223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267
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 */
	*hpfar = ((tmp >> 12) & ((1UL << 36) - 1)) << 4;
	return true;
}

static bool __hyp_text __populate_fault_info(struct kvm_vcpu *vcpu)
{
268 269
	u8 ec;
	u64 esr;
270 271
	u64 hpfar, far;

272 273
	esr = vcpu->arch.fault.esr_el2;
	ec = ESR_ELx_EC(esr);
274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303

	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;
}

304 305 306 307 308
/* Skip an instruction which has been emulated. Returns true if
 * execution can continue or false if we need to exit hyp mode because
 * single-step was in effect.
 */
static bool __hyp_text __skip_instr(struct kvm_vcpu *vcpu)
309 310 311 312 313 314 315 316 317 318 319 320
{
	*vcpu_pc(vcpu) = read_sysreg_el2(elr);

	if (vcpu_mode_is_32bit(vcpu)) {
		vcpu->arch.ctxt.gp_regs.regs.pstate = read_sysreg_el2(spsr);
		kvm_skip_instr32(vcpu, kvm_vcpu_trap_il_is32bit(vcpu));
		write_sysreg_el2(vcpu->arch.ctxt.gp_regs.regs.pstate, spsr);
	} else {
		*vcpu_pc(vcpu) += 4;
	}

	write_sysreg_el2(*vcpu_pc(vcpu), elr);
321 322 323 324 325 326 327 328

	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) {
		vcpu->arch.fault.esr_el2 =
			(ESR_ELx_EC_SOFTSTP_LOW << ESR_ELx_EC_SHIFT) | 0x22;
		return false;
	} else {
		return true;
	}
329 330
}

331 332 333
void __hyp_text __hyp_switch_fpsimd(u64 esr __always_unused,
				    struct kvm_vcpu *vcpu)
{
334 335
	struct user_fpsimd_state *host_fpsimd = vcpu->arch.host_fpsimd_state;

336 337 338 339 340 341 342 343 344
	if (has_vhe())
		write_sysreg(read_sysreg(cpacr_el1) | CPACR_EL1_FPEN,
			     cpacr_el1);
	else
		write_sysreg(read_sysreg(cptr_el2) & ~(u64)CPTR_EL2_TFP,
			     cptr_el2);

	isb();

345
	if (vcpu->arch.flags & KVM_ARM64_FP_HOST) {
346 347 348 349 350 351 352 353 354 355 356 357 358 359 360
		/*
		 * In the SVE case, VHE is assumed: it is enforced by
		 * Kconfig and kvm_arch_init().
		 */
		if (system_supports_sve() &&
		    (vcpu->arch.flags & KVM_ARM64_HOST_SVE_IN_USE)) {
			struct thread_struct *thread = container_of(
				host_fpsimd,
				struct thread_struct, uw.fpsimd_state);

			sve_save_state(sve_pffr(thread), &host_fpsimd->fpsr);
		} else {
			__fpsimd_save_state(host_fpsimd);
		}

361 362 363
		vcpu->arch.flags &= ~KVM_ARM64_FP_HOST;
	}

364 365 366 367 368 369
	__fpsimd_restore_state(&vcpu->arch.ctxt.gp_regs.fp_regs);

	/* Skip restoring fpexc32 for AArch64 guests */
	if (!(read_sysreg(hcr_el2) & HCR_RW))
		write_sysreg(vcpu->arch.ctxt.sys_regs[FPEXC32_EL2],
			     fpexc32_el2);
370 371

	vcpu->arch.flags |= KVM_ARM64_FP_ENABLED;
372 373
}

374 375 376 377 378 379
/*
 * 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)
380
{
381
	if (ARM_EXCEPTION_CODE(*exit_code) != ARM_EXCEPTION_IRQ)
382
		vcpu->arch.fault.esr_el2 = read_sysreg_el2(esr);
383

384 385 386 387 388 389
	/*
	 * 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.
	 */
390 391
	if (*exit_code == ARM_EXCEPTION_TRAP && !__populate_fault_info(vcpu))
		return true;
392

393
	if (static_branch_unlikely(&vgic_v2_cpuif_trap) &&
394
	    *exit_code == ARM_EXCEPTION_TRAP) {
395 396 397 398 399 400 401 402
		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);

403 404 405
		if (valid) {
			int ret = __vgic_v2_perform_cpuif_access(vcpu);

406 407
			if (ret ==  1 && __skip_instr(vcpu))
				return true;
408 409

			if (ret == -1) {
410 411 412 413 414 415 416 417
				/* Promote an illegal access to an
				 * SError. If we would be returning
				 * due to single-step clear the SS
				 * bit so handle_exit knows what to
				 * do after dealing with the error.
				 */
				if (!__skip_instr(vcpu))
					*vcpu_cpsr(vcpu) &= ~DBG_SPSR_SS;
418
				*exit_code = ARM_EXCEPTION_EL1_SERROR;
419
			}
420 421 422
		}
	}

423
	if (static_branch_unlikely(&vgic_v3_cpuif_trap) &&
424
	    *exit_code == ARM_EXCEPTION_TRAP &&
425 426 427 428
	    (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);

429 430
		if (ret == 1 && __skip_instr(vcpu))
			return true;
431 432
	}

433 434 435 436
	/* Return to the host kernel and handle the exit */
	return false;
}

437 438 439 440 441 442 443
/* 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;

444
	host_ctxt = vcpu->arch.host_cpu_context;
445 446 447
	host_ctxt->__hyp_running_vcpu = vcpu;
	guest_ctxt = &vcpu->arch.ctxt;

448
	sysreg_save_host_state_vhe(host_ctxt);
449 450

	__activate_traps(vcpu);
451
	__activate_vm(vcpu->kvm);
452

453
	sysreg_restore_guest_state_vhe(guest_ctxt);
454 455 456 457 458 459 460 461 462
	__debug_switch_to_guest(vcpu);

	do {
		/* Jump in the fire! */
		exit_code = __guest_enter(vcpu, host_ctxt);

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

463
	sysreg_save_guest_state_vhe(guest_ctxt);
464 465 466

	__deactivate_traps(vcpu);

467
	sysreg_restore_host_state_vhe(host_ctxt);
468

469
	if (vcpu->arch.flags & KVM_ARM64_FP_ENABLED)
470
		__fpsimd_save_fpexc32(vcpu);
471 472 473 474 475 476 477 478

	__debug_switch_to_host(vcpu);

	return exit_code;
}

/* Switch to the guest for legacy non-VHE systems */
int __hyp_text __kvm_vcpu_run_nvhe(struct kvm_vcpu *vcpu)
479 480 481 482 483 484 485 486 487 488 489
{
	struct kvm_cpu_context *host_ctxt;
	struct kvm_cpu_context *guest_ctxt;
	u64 exit_code;

	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;

490
	__sysreg_save_state_nvhe(host_ctxt);
491 492

	__activate_traps(vcpu);
493
	__activate_vm(kern_hyp_va(vcpu->kvm));
494

495
	__hyp_vgic_restore_state(vcpu);
496 497 498 499 500 501 502
	__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);
503
	__sysreg_restore_state_nvhe(guest_ctxt);
504 505 506 507 508 509 510 511 512
	__debug_switch_to_guest(vcpu);

	do {
		/* Jump in the fire! */
		exit_code = __guest_enter(vcpu, host_ctxt);

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

513
	__sysreg_save_state_nvhe(guest_ctxt);
514
	__sysreg32_save_state(vcpu);
515
	__timer_disable_traps(vcpu);
516
	__hyp_vgic_save_state(vcpu);
517 518 519 520

	__deactivate_traps(vcpu);
	__deactivate_vm(vcpu);

521
	__sysreg_restore_state_nvhe(host_ctxt);
522

523
	if (vcpu->arch.flags & KVM_ARM64_FP_ENABLED)
524
		__fpsimd_save_fpexc32(vcpu);
525

526 527 528 529
	/*
	 * This must come after restoring the host sysregs, since a non-VHE
	 * system may enable SPE here and make use of the TTBRs.
	 */
530
	__debug_switch_to_host(vcpu);
531 532 533

	return exit_code;
}
M
Marc Zyngier 已提交
534 535 536

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

537
static void __hyp_text __hyp_call_panic_nvhe(u64 spsr, u64 elr, u64 par,
538
					     struct kvm_cpu_context *__host_ctxt)
M
Marc Zyngier 已提交
539
{
540
	struct kvm_vcpu *vcpu;
541
	unsigned long str_va;
542

543 544 545 546 547 548
	vcpu = __host_ctxt->__hyp_running_vcpu;

	if (read_sysreg(vttbr_el2)) {
		__timer_disable_traps(vcpu);
		__deactivate_traps(vcpu);
		__deactivate_vm(vcpu);
549
		__sysreg_restore_state_nvhe(__host_ctxt);
550 551
	}

552 553 554 555 556 557 558 559
	/*
	 * 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,
560 561
		       spsr,  elr,
		       read_sysreg(esr_el2),   read_sysreg_el2(far),
562
		       read_sysreg(hpfar_el2), par, vcpu);
563 564
}

565 566
static void __hyp_call_panic_vhe(u64 spsr, u64 elr, u64 par,
				 struct kvm_cpu_context *host_ctxt)
567
{
568 569 570 571
	struct kvm_vcpu *vcpu;
	vcpu = host_ctxt->__hyp_running_vcpu;

	__deactivate_traps(vcpu);
572
	sysreg_restore_host_state_vhe(host_ctxt);
573

574 575 576
	panic(__hyp_panic_string,
	      spsr,  elr,
	      read_sysreg_el2(esr),   read_sysreg_el2(far),
577
	      read_sysreg(hpfar_el2), par, vcpu);
578 579
}

580
void __hyp_text __noreturn hyp_panic(struct kvm_cpu_context *host_ctxt)
581 582 583
{
	u64 spsr = read_sysreg_el2(spsr);
	u64 elr = read_sysreg_el2(elr);
M
Marc Zyngier 已提交
584 585
	u64 par = read_sysreg(par_el1);

586 587 588 589
	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
590 591 592

	unreachable();
}