[linux-2.6-block.git] / arch / arm64 / kvm / hyp / switch.c

/*
 * 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/>.
 */

#include <linux/types.h>
#include <linux/jump_label.h>

#include <asm/kvm_asm.h>
#include <asm/kvm_emulate.h>
#include <asm/kvm_hyp.h>
#include <asm/fpsimd.h>

static bool __hyp_text __fpsimd_enabled_nvhe(void)
{
	return !(read_sysreg(cptr_el2) & CPTR_EL2_TFP);
}

static bool __hyp_text __fpsimd_enabled_vhe(void)
{
	return !!(read_sysreg(cpacr_el1) & CPACR_EL1_FPEN);
}

static hyp_alternate_select(__fpsimd_is_enabled,
			    __fpsimd_enabled_nvhe, __fpsimd_enabled_vhe,
			    ARM64_HAS_VIRT_HOST_EXTN);

bool __hyp_text __fpsimd_enabled(void)
{
	return __fpsimd_is_enabled()();
}

static void __hyp_text __activate_traps_vhe(void)
{
	u64 val;

	val = read_sysreg(cpacr_el1);
	val |= CPACR_EL1_TTA;
	val &= ~CPACR_EL1_FPEN;
	write_sysreg(val, cpacr_el1);

	write_sysreg(__kvm_hyp_vector, vbar_el1);
}

static void __hyp_text __activate_traps_nvhe(void)
{
	u64 val;

	val = CPTR_EL2_DEFAULT;
	val |= CPTR_EL2_TTA | CPTR_EL2_TFP;
	write_sysreg(val, cptr_el2);
}

static hyp_alternate_select(__activate_traps_arch,
			    __activate_traps_nvhe, __activate_traps_vhe,
			    ARM64_HAS_VIRT_HOST_EXTN);

static void __hyp_text __activate_traps(struct kvm_vcpu *vcpu)
{
	u64 val;

	/*
	 * 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.
	 */
	val = vcpu->arch.hcr_el2;
	if (!(val & HCR_RW) && system_supports_fpsimd()) {
		write_sysreg(1 << 30, fpexc32_el2);
		isb();
	}
	write_sysreg(val, hcr_el2);
	/* Trap on AArch32 cp15 c15 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);
	__activate_traps_arch()();
}

static void __hyp_text __deactivate_traps_vhe(void)
{
	extern char vectors[];	/* kernel exception vectors */
	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);
	write_sysreg(HCR_HOST_VHE_FLAGS, hcr_el2);
	write_sysreg(CPACR_EL1_FPEN, cpacr_el1);
	write_sysreg(vectors, vbar_el1);
}

static void __hyp_text __deactivate_traps_nvhe(void)
{
	u64 mdcr_el2 = read_sysreg(mdcr_el2);

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

	write_sysreg(mdcr_el2, mdcr_el2);
	write_sysreg(HCR_RW, hcr_el2);
	write_sysreg(CPTR_EL2_DEFAULT, cptr_el2);
}

static hyp_alternate_select(__deactivate_traps_arch,
			    __deactivate_traps_nvhe, __deactivate_traps_vhe,
			    ARM64_HAS_VIRT_HOST_EXTN);

static void __hyp_text __deactivate_traps(struct kvm_vcpu *vcpu)
{
	/*
	 * 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);

	__deactivate_traps_arch()();
	write_sysreg(0, hstr_el2);
	write_sysreg(0, pmuserenr_el0);
}

static void __hyp_text __activate_vm(struct kvm_vcpu *vcpu)
{
	struct kvm *kvm = kern_hyp_va(vcpu->kvm);
	write_sysreg(kvm->arch.vttbr, vttbr_el2);
}

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

static void __hyp_text __vgic_save_state(struct kvm_vcpu *vcpu)
{
	if (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
		__vgic_v3_save_state(vcpu);
	else
		__vgic_v2_save_state(vcpu);

	write_sysreg(read_sysreg(hcr_el2) & ~HCR_INT_OVERRIDE, hcr_el2);
}

static void __hyp_text __vgic_restore_state(struct kvm_vcpu *vcpu)
{
	u64 val;

	val = read_sysreg(hcr_el2);
	val |= 	HCR_INT_OVERRIDE;
	val |= vcpu->arch.irq_lines;
	write_sysreg(val, hcr_el2);

	if (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
		__vgic_v3_restore_state(vcpu);
	else
		__vgic_v2_restore_state(vcpu);
}

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)
{
	u64 esr = read_sysreg_el2(esr);
	u8 ec = ESR_ELx_EC(esr);
	u64 hpfar, far;

	vcpu->arch.fault.esr_el2 = esr;

	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;
}

static void __hyp_text __skip_instr(struct kvm_vcpu *vcpu)
{
	*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);
}

int __hyp_text __kvm_vcpu_run(struct kvm_vcpu *vcpu)
{
	struct kvm_cpu_context *host_ctxt;
	struct kvm_cpu_context *guest_ctxt;
	bool fp_enabled;
	u64 exit_code;

	vcpu = kern_hyp_va(vcpu);
	write_sysreg(vcpu, tpidr_el2);

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

	__sysreg_save_host_state(host_ctxt);
	__debug_cond_save_host_state(vcpu);

	__activate_traps(vcpu);
	__activate_vm(vcpu);

	__vgic_restore_state(vcpu);
	__timer_restore_state(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);
	__sysreg_restore_guest_state(guest_ctxt);
	__debug_restore_state(vcpu, kern_hyp_va(vcpu->arch.debug_ptr), guest_ctxt);

	/* Jump in the fire! */
again:
	exit_code = __guest_enter(vcpu, host_ctxt);
	/* And we're baaack! */

	/*
	 * 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.
	 */
	if (exit_code == ARM_EXCEPTION_TRAP && !__populate_fault_info(vcpu))
		goto again;

	if (static_branch_unlikely(&vgic_v2_cpuif_trap) &&
	    exit_code == ARM_EXCEPTION_TRAP) {
		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);

		if (valid) {
			int ret = __vgic_v2_perform_cpuif_access(vcpu);

			if (ret == 1) {
				__skip_instr(vcpu);
				goto again;
			}

			if (ret == -1) {
				/* Promote an illegal access to an SError */
				__skip_instr(vcpu);
				exit_code = ARM_EXCEPTION_EL1_SERROR;
			}

			/* 0 falls through to be handler out of EL2 */
		}
	}

	fp_enabled = __fpsimd_enabled();

	__sysreg_save_guest_state(guest_ctxt);
	__sysreg32_save_state(vcpu);
	__timer_save_state(vcpu);
	__vgic_save_state(vcpu);

	__deactivate_traps(vcpu);
	__deactivate_vm(vcpu);

	__sysreg_restore_host_state(host_ctxt);

	if (fp_enabled) {
		__fpsimd_save_state(&guest_ctxt->gp_regs.fp_regs);
		__fpsimd_restore_state(&host_ctxt->gp_regs.fp_regs);
	}

	__debug_save_state(vcpu, kern_hyp_va(vcpu->arch.debug_ptr), guest_ctxt);
	/*
	 * This must come after restoring the host sysregs, since a non-VHE
	 * system may enable SPE here and make use of the TTBRs.
	 */
	__debug_cond_restore_host_state(vcpu);

	return exit_code;
}

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

static void __hyp_text __hyp_call_panic_nvhe(u64 spsr, u64 elr, u64 par)
{
	unsigned long str_va;

	/*
	 * 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,
		       spsr,  elr,
		       read_sysreg(esr_el2),   read_sysreg_el2(far),
		       read_sysreg(hpfar_el2), par,
		       (void *)read_sysreg(tpidr_el2));
}

static void __hyp_text __hyp_call_panic_vhe(u64 spsr, u64 elr, u64 par)
{
	panic(__hyp_panic_string,
	      spsr,  elr,
	      read_sysreg_el2(esr),   read_sysreg_el2(far),
	      read_sysreg(hpfar_el2), par,
	      (void *)read_sysreg(tpidr_el2));
}

static hyp_alternate_select(__hyp_call_panic,
			    __hyp_call_panic_nvhe, __hyp_call_panic_vhe,
			    ARM64_HAS_VIRT_HOST_EXTN);

void __hyp_text __noreturn __hyp_panic(void)
{
	u64 spsr = read_sysreg_el2(spsr);
	u64 elr = read_sysreg_el2(elr);
	u64 par = read_sysreg(par_el1);

	if (read_sysreg(vttbr_el2)) {
		struct kvm_vcpu *vcpu;
		struct kvm_cpu_context *host_ctxt;

		vcpu = (struct kvm_vcpu *)read_sysreg(tpidr_el2);
		host_ctxt = kern_hyp_va(vcpu->arch.host_cpu_context);
		__timer_save_state(vcpu);
		__deactivate_traps(vcpu);
		__deactivate_vm(vcpu);
		__sysreg_restore_host_state(host_ctxt);
	}

	/* Call panic for real */
	__hyp_call_panic()(spsr, elr, par);

	unreachable();
}
Commit	Line	Data
be901e9b MZ	1	/*
	2	* Copyright (C) 2015 - ARM Ltd
	3	* Author: Marc Zyngier <marc.zyngier@arm.com>
	4	*
	5	* This program is free software; you can redistribute it and/or modify
	6	* it under the terms of the GNU General Public License version 2 as
	7	* published by the Free Software Foundation.
	8	*
	9	* This program is distributed in the hope that it will be useful,
	10	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	11	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	12	* GNU General Public License for more details.
	13	*
	14	* You should have received a copy of the GNU General Public License
	15	* along with this program. If not, see <http://www.gnu.org/licenses/>.
	16	*/
	17
5f05a72a	18	#include <linux/types.h>
5a7a8426 VM	19	#include <linux/jump_label.h>
5a7a8426 VM	20
68908bf7	21	#include <asm/kvm_asm.h>
fb5ee369	22	#include <asm/kvm_emulate.h>
13720a56	23	#include <asm/kvm_hyp.h>
82e0191a	24	#include <asm/fpsimd.h>
be901e9b	25
32876224 MZ	26	static bool __hyp_text __fpsimd_enabled_nvhe(void)
	27	{
	28	return !(read_sysreg(cptr_el2) & CPTR_EL2_TFP);
	29	}
	30
	31	static bool __hyp_text __fpsimd_enabled_vhe(void)
	32	{
	33	return !!(read_sysreg(cpacr_el1) & CPACR_EL1_FPEN);
	34	}
	35
	36	static hyp_alternate_select(__fpsimd_is_enabled,
	37	__fpsimd_enabled_nvhe, __fpsimd_enabled_vhe,
	38	ARM64_HAS_VIRT_HOST_EXTN);
	39
	40	bool __hyp_text __fpsimd_enabled(void)
	41	{
	42	return __fpsimd_is_enabled()();
	43	}
	44
68908bf7 MZ	45	static void __hyp_text __activate_traps_vhe(void)
	46	{
	47	u64 val;
	48
	49	val = read_sysreg(cpacr_el1);
	50	val \|= CPACR_EL1_TTA;
	51	val &= ~CPACR_EL1_FPEN;
	52	write_sysreg(val, cpacr_el1);
	53
	54	write_sysreg(__kvm_hyp_vector, vbar_el1);
	55	}
	56
	57	static void __hyp_text __activate_traps_nvhe(void)
	58	{
	59	u64 val;
	60
	61	val = CPTR_EL2_DEFAULT;
	62	val \|= CPTR_EL2_TTA \| CPTR_EL2_TFP;
	63	write_sysreg(val, cptr_el2);
	64	}
	65
	66	static hyp_alternate_select(__activate_traps_arch,
	67	__activate_traps_nvhe, __activate_traps_vhe,
	68	ARM64_HAS_VIRT_HOST_EXTN);
	69
be901e9b MZ	70	static void __hyp_text __activate_traps(struct kvm_vcpu *vcpu)
	71	{
	72	u64 val;
	73
	74	/*
	75	* We are about to set CPTR_EL2.TFP to trap all floating point
	76	* register accesses to EL2, however, the ARM ARM clearly states that
	77	* traps are only taken to EL2 if the operation would not otherwise
	78	* trap to EL1. Therefore, always make sure that for 32-bit guests,
	79	* we set FPEXC.EN to prevent traps to EL1, when setting the TFP bit.
82e0191a SP	80	* If FP/ASIMD is not implemented, FPEXC is UNDEFINED and any access to
82e0191a SP	81	* it will cause an exception.
be901e9b MZ	82	*/
be901e9b MZ	83	val = vcpu->arch.hcr_el2;
82e0191a	84	if (!(val & HCR_RW) && system_supports_fpsimd()) {
be901e9b MZ	85	write_sysreg(1 << 30, fpexc32_el2);
	86	isb();
	87	}
	88	write_sysreg(val, hcr_el2);
	89	/* Trap on AArch32 cp15 c15 accesses (EL1 or EL0) */
	90	write_sysreg(1 << 15, hstr_el2);
21cbe3cc MZ	91	/*
	92	* Make sure we trap PMU access from EL0 to EL2. Also sanitize
	93	* PMSELR_EL0 to make sure it never contains the cycle
	94	* counter, which could make a PMXEVCNTR_EL0 access UNDEF at
	95	* EL1 instead of being trapped to EL2.
	96	*/
	97	write_sysreg(0, pmselr_el0);
d692b8ad	98	write_sysreg(ARMV8_PMU_USERENR_MASK, pmuserenr_el0);
68908bf7 MZ	99	write_sysreg(vcpu->arch.mdcr_el2, mdcr_el2);
	100	__activate_traps_arch()();
	101	}
a7e0ac29	102
68908bf7 MZ	103	static void __hyp_text __deactivate_traps_vhe(void)
	104	{
	105	extern char vectors[]; /* kernel exception vectors */
f85279b4	106	u64 mdcr_el2 = read_sysreg(mdcr_el2);
a7e0ac29	107
f85279b4 WD	108	mdcr_el2 &= MDCR_EL2_HPMN_MASK \|
	109	MDCR_EL2_E2PB_MASK << MDCR_EL2_E2PB_SHIFT \|
	110	MDCR_EL2_TPMS;
	111
	112	write_sysreg(mdcr_el2, mdcr_el2);
68908bf7 MZ	113	write_sysreg(HCR_HOST_VHE_FLAGS, hcr_el2);
	114	write_sysreg(CPACR_EL1_FPEN, cpacr_el1);
	115	write_sysreg(vectors, vbar_el1);
be901e9b MZ	116	}
be901e9b MZ	117
68908bf7	118	static void __hyp_text __deactivate_traps_nvhe(void)
be901e9b	119	{
f85279b4 WD	120	u64 mdcr_el2 = read_sysreg(mdcr_el2);
	121
	122	mdcr_el2 &= MDCR_EL2_HPMN_MASK;
	123	mdcr_el2 \|= MDCR_EL2_E2PB_MASK << MDCR_EL2_E2PB_SHIFT;
	124
	125	write_sysreg(mdcr_el2, mdcr_el2);
be901e9b	126	write_sysreg(HCR_RW, hcr_el2);
68908bf7 MZ	127	write_sysreg(CPTR_EL2_DEFAULT, cptr_el2);
	128	}
	129
	130	static hyp_alternate_select(__deactivate_traps_arch,
	131	__deactivate_traps_nvhe, __deactivate_traps_vhe,
	132	ARM64_HAS_VIRT_HOST_EXTN);
	133
	134	static void __hyp_text __deactivate_traps(struct kvm_vcpu *vcpu)
	135	{
44636f97 MZ	136	/*
	137	* If we pended a virtual abort, preserve it until it gets
	138	* cleared. See D1.14.3 (Virtual Interrupts) for details, but
	139	* the crucial bit is "On taking a vSError interrupt,
	140	* HCR_EL2.VSE is cleared to 0."
	141	*/
	142	if (vcpu->arch.hcr_el2 & HCR_VSE)
	143	vcpu->arch.hcr_el2 = read_sysreg(hcr_el2);
	144
68908bf7	145	__deactivate_traps_arch()();
be901e9b	146	write_sysreg(0, hstr_el2);
d692b8ad	147	write_sysreg(0, pmuserenr_el0);
be901e9b MZ	148	}
	149
	150	static void __hyp_text __activate_vm(struct kvm_vcpu *vcpu)
	151	{
	152	struct kvm *kvm = kern_hyp_va(vcpu->kvm);
	153	write_sysreg(kvm->arch.vttbr, vttbr_el2);
	154	}
	155
	156	static void __hyp_text __deactivate_vm(struct kvm_vcpu *vcpu)
	157	{
	158	write_sysreg(0, vttbr_el2);
	159	}
	160
be901e9b MZ	161	static void __hyp_text __vgic_save_state(struct kvm_vcpu *vcpu)
be901e9b MZ	162	{
5a7a8426 VM	163	if (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
	164	__vgic_v3_save_state(vcpu);
	165	else
	166	__vgic_v2_save_state(vcpu);
	167
be901e9b MZ	168	write_sysreg(read_sysreg(hcr_el2) & ~HCR_INT_OVERRIDE, hcr_el2);
	169	}
	170
	171	static void __hyp_text __vgic_restore_state(struct kvm_vcpu *vcpu)
	172	{
	173	u64 val;
	174
	175	val = read_sysreg(hcr_el2);
	176	val \|= HCR_INT_OVERRIDE;
	177	val \|= vcpu->arch.irq_lines;
	178	write_sysreg(val, hcr_el2);
	179
5a7a8426 VM	180	if (static_branch_unlikely(&kvm_vgic_global_state.gicv3_cpuif))
	181	__vgic_v3_restore_state(vcpu);
	182	else
	183	__vgic_v2_restore_state(vcpu);
be901e9b MZ	184	}
be901e9b MZ	185
5f05a72a MZ	186	static bool __hyp_text __true_value(void)
	187	{
	188	return true;
	189	}
	190
	191	static bool __hyp_text __false_value(void)
	192	{
	193	return false;
	194	}
	195
	196	static hyp_alternate_select(__check_arm_834220,
	197	__false_value, __true_value,
	198	ARM64_WORKAROUND_834220);
	199
	200	static bool __hyp_text __translate_far_to_hpfar(u64 far, u64 *hpfar)
	201	{
	202	u64 par, tmp;
	203
	204	/*
	205	* Resolve the IPA the hard way using the guest VA.
	206	*
	207	* Stage-1 translation already validated the memory access
	208	* rights. As such, we can use the EL1 translation regime, and
	209	* don't have to distinguish between EL0 and EL1 access.
	210	*
	211	* We do need to save/restore PAR_EL1 though, as we haven't
	212	* saved the guest context yet, and we may return early...
	213	*/
	214	par = read_sysreg(par_el1);
	215	asm volatile("at s1e1r, %0" : : "r" (far));
	216	isb();
	217
	218	tmp = read_sysreg(par_el1);
	219	write_sysreg(par, par_el1);
	220
	221	if (unlikely(tmp & 1))
	222	return false; /* Translation failed, back to guest */
	223
	224	/* Convert PAR to HPFAR format */
	225	*hpfar = ((tmp >> 12) & ((1UL << 36) - 1)) << 4;
	226	return true;
	227	}
	228
	229	static bool __hyp_text __populate_fault_info(struct kvm_vcpu *vcpu)
	230	{
	231	u64 esr = read_sysreg_el2(esr);
561454e2	232	u8 ec = ESR_ELx_EC(esr);
5f05a72a MZ	233	u64 hpfar, far;
	234
	235	vcpu->arch.fault.esr_el2 = esr;
	236
	237	if (ec != ESR_ELx_EC_DABT_LOW && ec != ESR_ELx_EC_IABT_LOW)
	238	return true;
	239
	240	far = read_sysreg_el2(far);
	241
	242	/*
	243	* The HPFAR can be invalid if the stage 2 fault did not
	244	* happen during a stage 1 page table walk (the ESR_EL2.S1PTW
	245	* bit is clear) and one of the two following cases are true:
	246	* 1. The fault was due to a permission fault
	247	* 2. The processor carries errata 834220
	248	*
	249	* Therefore, for all non S1PTW faults where we either have a
	250	* permission fault or the errata workaround is enabled, we
	251	* resolve the IPA using the AT instruction.
	252	*/
	253	if (!(esr & ESR_ELx_S1PTW) &&
	254	(__check_arm_834220()() \|\| (esr & ESR_ELx_FSC_TYPE) == FSC_PERM)) {
	255	if (!__translate_far_to_hpfar(far, &hpfar))
	256	return false;
	257	} else {
	258	hpfar = read_sysreg(hpfar_el2);
	259	}
	260
	261	vcpu->arch.fault.far_el2 = far;
	262	vcpu->arch.fault.hpfar_el2 = hpfar;
	263	return true;
	264	}
	265
fb5ee369 MZ	266	static void __hyp_text __skip_instr(struct kvm_vcpu *vcpu)
	267	{
	268	*vcpu_pc(vcpu) = read_sysreg_el2(elr);
	269
	270	if (vcpu_mode_is_32bit(vcpu)) {
	271	vcpu->arch.ctxt.gp_regs.regs.pstate = read_sysreg_el2(spsr);
	272	kvm_skip_instr32(vcpu, kvm_vcpu_trap_il_is32bit(vcpu));
	273	write_sysreg_el2(vcpu->arch.ctxt.gp_regs.regs.pstate, spsr);
	274	} else {
	275	*vcpu_pc(vcpu) += 4;
	276	}
	277
	278	write_sysreg_el2(*vcpu_pc(vcpu), elr);
	279	}
	280
cf0ba18a	281	int __hyp_text __kvm_vcpu_run(struct kvm_vcpu *vcpu)
be901e9b MZ	282	{
	283	struct kvm_cpu_context *host_ctxt;
	284	struct kvm_cpu_context *guest_ctxt;
c13d1683	285	bool fp_enabled;
be901e9b MZ	286	u64 exit_code;
	287
	288	vcpu = kern_hyp_va(vcpu);
	289	write_sysreg(vcpu, tpidr_el2);
	290
	291	host_ctxt = kern_hyp_va(vcpu->arch.host_cpu_context);
	292	guest_ctxt = &vcpu->arch.ctxt;
	293
edef528d	294	__sysreg_save_host_state(host_ctxt);
be901e9b MZ	295	__debug_cond_save_host_state(vcpu);
	296
	297	__activate_traps(vcpu);
	298	__activate_vm(vcpu);
	299
	300	__vgic_restore_state(vcpu);
	301	__timer_restore_state(vcpu);
	302
	303	/*
	304	* We must restore the 32-bit state before the sysregs, thanks
674e7012	305	* to erratum #852523 (Cortex-A57) or #853709 (Cortex-A72).
be901e9b MZ	306	*/
be901e9b MZ	307	__sysreg32_restore_state(vcpu);
edef528d	308	__sysreg_restore_guest_state(guest_ctxt);
be901e9b MZ	309	__debug_restore_state(vcpu, kern_hyp_va(vcpu->arch.debug_ptr), guest_ctxt);
	310
	311	/* Jump in the fire! */
5f05a72a	312	again:
be901e9b MZ	313	exit_code = __guest_enter(vcpu, host_ctxt);
	314	/* And we're baaack! */
	315
395ea79e MZ	316	/*
	317	* We're using the raw exception code in order to only process
	318	* the trap if no SError is pending. We will come back to the
	319	* same PC once the SError has been injected, and replay the
	320	* trapping instruction.
	321	*/
5f05a72a MZ	322	if (exit_code == ARM_EXCEPTION_TRAP && !__populate_fault_info(vcpu))
	323	goto again;
	324
fb5ee369 MZ	325	if (static_branch_unlikely(&vgic_v2_cpuif_trap) &&
	326	exit_code == ARM_EXCEPTION_TRAP) {
	327	bool valid;
	328
	329	valid = kvm_vcpu_trap_get_class(vcpu) == ESR_ELx_EC_DABT_LOW &&
	330	kvm_vcpu_trap_get_fault_type(vcpu) == FSC_FAULT &&
	331	kvm_vcpu_dabt_isvalid(vcpu) &&
	332	!kvm_vcpu_dabt_isextabt(vcpu) &&
	333	!kvm_vcpu_dabt_iss1tw(vcpu);
	334
3272f0d0 MZ	335	if (valid) {
	336	int ret = __vgic_v2_perform_cpuif_access(vcpu);
	337
	338	if (ret == 1) {
	339	__skip_instr(vcpu);
	340	goto again;
	341	}
	342
	343	if (ret == -1) {
	344	/* Promote an illegal access to an SError */
	345	__skip_instr(vcpu);
	346	exit_code = ARM_EXCEPTION_EL1_SERROR;
	347	}
	348
	349	/* 0 falls through to be handler out of EL2 */
fb5ee369 MZ	350	}
	351	}
	352
c13d1683 MZ	353	fp_enabled = __fpsimd_enabled();
c13d1683 MZ	354
edef528d	355	__sysreg_save_guest_state(guest_ctxt);
be901e9b MZ	356	__sysreg32_save_state(vcpu);
	357	__timer_save_state(vcpu);
	358	__vgic_save_state(vcpu);
	359
	360	__deactivate_traps(vcpu);
	361	__deactivate_vm(vcpu);
	362
edef528d	363	__sysreg_restore_host_state(host_ctxt);
be901e9b	364
c13d1683 MZ	365	if (fp_enabled) {
	366	__fpsimd_save_state(&guest_ctxt->gp_regs.fp_regs);
	367	__fpsimd_restore_state(&host_ctxt->gp_regs.fp_regs);
	368	}
	369
be901e9b	370	__debug_save_state(vcpu, kern_hyp_va(vcpu->arch.debug_ptr), guest_ctxt);
f85279b4 WD	371	/*
	372	* This must come after restoring the host sysregs, since a non-VHE
	373	* system may enable SPE here and make use of the TTBRs.
	374	*/
be901e9b MZ	375	__debug_cond_restore_host_state(vcpu);
	376
	377	return exit_code;
	378	}
53fd5b64 MZ	379
	380	static const char __hyp_panic_string[] = "HYP panic:\nPS:%08llx PC:%016llx ESR:%08llx\nFAR:%016llx HPFAR:%016llx PAR:%016llx\nVCPU:%p\n";
	381
253dcbd3	382	static void __hyp_text __hyp_call_panic_nvhe(u64 spsr, u64 elr, u64 par)
53fd5b64	383	{
cf7df13d	384	unsigned long str_va;
253dcbd3	385
cf7df13d MZ	386	/*
	387	* Force the panic string to be loaded from the literal pool,
	388	* making sure it is a kernel address and not a PC-relative
	389	* reference.
	390	*/
	391	asm volatile("ldr %0, =__hyp_panic_string" : "=r" (str_va));
	392
	393	__hyp_do_panic(str_va,
253dcbd3 MZ	394	spsr, elr,
	395	read_sysreg(esr_el2), read_sysreg_el2(far),
	396	read_sysreg(hpfar_el2), par,
	397	(void *)read_sysreg(tpidr_el2));
	398	}
	399
	400	static void __hyp_text __hyp_call_panic_vhe(u64 spsr, u64 elr, u64 par)
	401	{
	402	panic(__hyp_panic_string,
	403	spsr, elr,
	404	read_sysreg_el2(esr), read_sysreg_el2(far),
	405	read_sysreg(hpfar_el2), par,
	406	(void *)read_sysreg(tpidr_el2));
	407	}
	408
	409	static hyp_alternate_select(__hyp_call_panic,
	410	__hyp_call_panic_nvhe, __hyp_call_panic_vhe,
	411	ARM64_HAS_VIRT_HOST_EXTN);
	412
	413	void __hyp_text __noreturn __hyp_panic(void)
	414	{
	415	u64 spsr = read_sysreg_el2(spsr);
	416	u64 elr = read_sysreg_el2(elr);
53fd5b64 MZ	417	u64 par = read_sysreg(par_el1);
	418
	419	if (read_sysreg(vttbr_el2)) {
	420	struct kvm_vcpu *vcpu;
	421	struct kvm_cpu_context *host_ctxt;
	422
	423	vcpu = (struct kvm_vcpu *)read_sysreg(tpidr_el2);
	424	host_ctxt = kern_hyp_va(vcpu->arch.host_cpu_context);
e8ec032b	425	__timer_save_state(vcpu);
53fd5b64 MZ	426	__deactivate_traps(vcpu);
53fd5b64 MZ	427	__deactivate_vm(vcpu);
edef528d	428	__sysreg_restore_host_state(host_ctxt);
53fd5b64 MZ	429	}
	430
	431	/* Call panic for real */
253dcbd3	432	__hyp_call_panic()(spsr, elr, par);
53fd5b64 MZ	433
	434	unreachable();
	435	}