5 files changed, 168 insertions, 170 deletions

diff --git a/arch/arm64/kvm/hyp/include/hyp/switch.h b/arch/arm64/kvm/hyp/include/hyp/switch.h
index 76dfda116e56..2ad57b117385 100644
--- a/arch/arm64/kvm/hyp/include/hyp/switch.h
+++ b/arch/arm64/kvm/hyp/include/hyp/switch.h
@@ -65,6 +65,136 @@ static inline void __activate_traps_fpsimd32(struct kvm_vcpu *vcpu)
 	}
 }
 
+static inline void __activate_cptr_traps_nvhe(struct kvm_vcpu *vcpu)
+{
+	u64 val = CPTR_NVHE_EL2_RES1 | CPTR_EL2_TAM | CPTR_EL2_TTA;
+
+	/*
+	 * Always trap SME since it's not supported in KVM.
+	 * TSM is RES1 if SME isn't implemented.
+	 */
+	val |= CPTR_EL2_TSM;
+
+	if (!vcpu_has_sve(vcpu) || !guest_owns_fp_regs())
+		val |= CPTR_EL2_TZ;
+
+	if (!guest_owns_fp_regs())
+		val |= CPTR_EL2_TFP;
+
+	write_sysreg(val, cptr_el2);
+}
+
+static inline void __activate_cptr_traps_vhe(struct kvm_vcpu *vcpu)
+{
+	/*
+	 * With VHE (HCR.E2H == 1), accesses to CPACR_EL1 are routed to
+	 * CPTR_EL2. In general, CPACR_EL1 has the same layout as CPTR_EL2,
+	 * except for some missing controls, such as TAM.
+	 * In this case, CPTR_EL2.TAM has the same position with or without
+	 * VHE (HCR.E2H == 1) which allows us to use here the CPTR_EL2.TAM
+	 * shift value for trapping the AMU accesses.
+	 */
+	u64 val = CPTR_EL2_TAM | CPACR_EL1_TTA;
+	u64 cptr;
+
+	if (guest_owns_fp_regs()) {
+		val |= CPACR_EL1_FPEN;
+		if (vcpu_has_sve(vcpu))
+			val |= CPACR_EL1_ZEN;
+	}
+
+	if (!vcpu_has_nv(vcpu))
+		goto write;
+
+	/*
+	 * The architecture is a bit crap (what a surprise): an EL2 guest
+	 * writing to CPTR_EL2 via CPACR_EL1 can't set any of TCPAC or TTA,
+	 * as they are RES0 in the guest's view. To work around it, trap the
+	 * sucker using the very same bit it can't set...
+	 */
+	if (vcpu_el2_e2h_is_set(vcpu) && is_hyp_ctxt(vcpu))
+		val |= CPTR_EL2_TCPAC;
+
+	/*
+	 * Layer the guest hypervisor's trap configuration on top of our own if
+	 * we're in a nested context.
+	 */
+	if (is_hyp_ctxt(vcpu))
+		goto write;
+
+	cptr = vcpu_sanitised_cptr_el2(vcpu);
+
+	/*
+	 * Pay attention, there's some interesting detail here.
+	 *
+	 * The CPTR_EL2.xEN fields are 2 bits wide, although there are only two
+	 * meaningful trap states when HCR_EL2.TGE = 0 (running a nested guest):
+	 *
+	 *  - CPTR_EL2.xEN = x0, traps are enabled
+	 *  - CPTR_EL2.xEN = x1, traps are disabled
+	 *
+	 * In other words, bit[0] determines if guest accesses trap or not. In
+	 * the interest of simplicity, clear the entire field if the guest
+	 * hypervisor has traps enabled to dispel any illusion of something more
+	 * complicated taking place.
+	 */
+	if (!(SYS_FIELD_GET(CPACR_EL1, FPEN, cptr) & BIT(0)))
+		val &= ~CPACR_EL1_FPEN;
+	if (!(SYS_FIELD_GET(CPACR_EL1, ZEN, cptr) & BIT(0)))
+		val &= ~CPACR_EL1_ZEN;
+
+	if (kvm_has_feat(vcpu->kvm, ID_AA64MMFR3_EL1, S2POE, IMP))
+		val |= cptr & CPACR_EL1_E0POE;
+
+	val |= cptr & CPTR_EL2_TCPAC;
+
+write:
+	write_sysreg(val, cpacr_el1);
+}
+
+static inline void __activate_cptr_traps(struct kvm_vcpu *vcpu)
+{
+	if (!guest_owns_fp_regs())
+		__activate_traps_fpsimd32(vcpu);
+
+	if (has_vhe() || has_hvhe())
+		__activate_cptr_traps_vhe(vcpu);
+	else
+		__activate_cptr_traps_nvhe(vcpu);
+}
+
+static inline void __deactivate_cptr_traps_nvhe(struct kvm_vcpu *vcpu)
+{
+	u64 val = CPTR_NVHE_EL2_RES1;
+
+	if (!cpus_have_final_cap(ARM64_SVE))
+		val |= CPTR_EL2_TZ;
+	if (!cpus_have_final_cap(ARM64_SME))
+		val |= CPTR_EL2_TSM;
+
+	write_sysreg(val, cptr_el2);
+}
+
+static inline void __deactivate_cptr_traps_vhe(struct kvm_vcpu *vcpu)
+{
+	u64 val = CPACR_EL1_FPEN;
+
+	if (cpus_have_final_cap(ARM64_SVE))
+		val |= CPACR_EL1_ZEN;
+	if (cpus_have_final_cap(ARM64_SME))
+		val |= CPACR_EL1_SMEN;
+
+	write_sysreg(val, cpacr_el1);
+}
+
+static inline void __deactivate_cptr_traps(struct kvm_vcpu *vcpu)
+{
+	if (has_vhe() || has_hvhe())
+		__deactivate_cptr_traps_vhe(vcpu);
+	else
+		__deactivate_cptr_traps_nvhe(vcpu);
+}
+
 #define reg_to_fgt_masks(reg)						\
 	({								\
 		struct fgt_masks *m;					\
@@ -486,11 +616,6 @@ static void kvm_hyp_save_fpsimd_host(struct kvm_vcpu *vcpu)
 	 */
 	if (system_supports_sve()) {
 		__hyp_sve_save_host();
-
-		/* Re-enable SVE traps if not supported for the guest vcpu. */
-		if (!vcpu_has_sve(vcpu))
-			cpacr_clear_set(CPACR_EL1_ZEN, 0);
-
 	} else {
 		__fpsimd_save_state(host_data_ptr(host_ctxt.fp_regs));
 	}
@@ -541,10 +666,7 @@ static inline bool kvm_hyp_handle_fpsimd(struct kvm_vcpu *vcpu, u64 *exit_code)
 	/* Valid trap.  Switch the context: */
 
 	/* First disable enough traps to allow us to update the registers */
-	if (sve_guest || (is_protected_kvm_enabled() && system_supports_sve()))
-		cpacr_clear_set(0, CPACR_EL1_FPEN | CPACR_EL1_ZEN);
-	else
-		cpacr_clear_set(0, CPACR_EL1_FPEN);
+	__deactivate_cptr_traps(vcpu);
 	isb();
 
 	/* Write out the host state if it's in the registers */
@@ -566,6 +688,13 @@ static inline bool kvm_hyp_handle_fpsimd(struct kvm_vcpu *vcpu, u64 *exit_code)
 
 	*host_data_ptr(fp_owner) = FP_STATE_GUEST_OWNED;
 
+	/*
+	 * Re-enable traps necessary for the current state of the guest, e.g.
+	 * those enabled by a guest hypervisor. The ERET to the guest will
+	 * provide the necessary context synchronization.
+	 */
+	__activate_cptr_traps(vcpu);
+
 	return true;
 }
 
diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-main.c b/arch/arm64/kvm/hyp/nvhe/hyp-main.c
index e9198e56e784..3206b2c07f82 100644
--- a/arch/arm64/kvm/hyp/nvhe/hyp-main.c
+++ b/arch/arm64/kvm/hyp/nvhe/hyp-main.c
@@ -69,7 +69,10 @@ static void fpsimd_sve_sync(struct kvm_vcpu *vcpu)
 	if (!guest_owns_fp_regs())
 		return;
 
-	cpacr_clear_set(0, CPACR_EL1_FPEN | CPACR_EL1_ZEN);
+	/*
+	 * Traps have been disabled by __deactivate_cptr_traps(), but there
+	 * hasn't necessarily been a context synchronization event yet.
+	 */
 	isb();
 
 	if (vcpu_has_sve(vcpu))
diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
index 95d7534c9679..8957734d6183 100644
--- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c
+++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
@@ -479,6 +479,7 @@ static int host_stage2_adjust_range(u64 addr, struct kvm_mem_range *range)
 {
 	struct kvm_mem_range cur;
 	kvm_pte_t pte;
+	u64 granule;
 	s8 level;
 	int ret;
 
@@ -496,18 +497,21 @@ static int host_stage2_adjust_range(u64 addr, struct kvm_mem_range *range)
 		return -EPERM;
 	}
 
-	do {
-		u64 granule = kvm_granule_size(level);
+	for (; level <= KVM_PGTABLE_LAST_LEVEL; level++) {
+		if (!kvm_level_supports_block_mapping(level))
+			continue;
+		granule = kvm_granule_size(level);
 		cur.start = ALIGN_DOWN(addr, granule);
 		cur.end = cur.start + granule;
-		level++;
-	} while ((level <= KVM_PGTABLE_LAST_LEVEL) &&
-			!(kvm_level_supports_block_mapping(level) &&
-			  range_included(&cur, range)));
+		if (!range_included(&cur, range))
+			continue;
+		*range = cur;
+		return 0;
+	}
 
-	*range = cur;
+	WARN_ON(1);
 
-	return 0;
+	return -EINVAL;
 }
 
 int host_stage2_idmap_locked(phys_addr_t addr, u64 size,
diff --git a/arch/arm64/kvm/hyp/nvhe/switch.c b/arch/arm64/kvm/hyp/nvhe/switch.c
index 73affe1333a4..0e752b515d0f 100644
--- a/arch/arm64/kvm/hyp/nvhe/switch.c
+++ b/arch/arm64/kvm/hyp/nvhe/switch.c
@@ -47,65 +47,6 @@ struct fgt_masks hdfgwtr2_masks;
 
 extern void kvm_nvhe_prepare_backtrace(unsigned long fp, unsigned long pc);
 
-static void __activate_cptr_traps(struct kvm_vcpu *vcpu)
-{
-	u64 val = CPTR_EL2_TAM;	/* Same bit irrespective of E2H */
-
-	if (!guest_owns_fp_regs())
-		__activate_traps_fpsimd32(vcpu);
-
-	if (has_hvhe()) {
-		val |= CPACR_EL1_TTA;
-
-		if (guest_owns_fp_regs()) {
-			val |= CPACR_EL1_FPEN;
-			if (vcpu_has_sve(vcpu))
-				val |= CPACR_EL1_ZEN;
-		}
-
-		write_sysreg(val, cpacr_el1);
-	} else {
-		val |= CPTR_EL2_TTA | CPTR_NVHE_EL2_RES1;
-
-		/*
-		 * Always trap SME since it's not supported in KVM.
-		 * TSM is RES1 if SME isn't implemented.
-		 */
-		val |= CPTR_EL2_TSM;
-
-		if (!vcpu_has_sve(vcpu) || !guest_owns_fp_regs())
-			val |= CPTR_EL2_TZ;
-
-		if (!guest_owns_fp_regs())
-			val |= CPTR_EL2_TFP;
-
-		write_sysreg(val, cptr_el2);
-	}
-}
-
-static void __deactivate_cptr_traps(struct kvm_vcpu *vcpu)
-{
-	if (has_hvhe()) {
-		u64 val = CPACR_EL1_FPEN;
-
-		if (cpus_have_final_cap(ARM64_SVE))
-			val |= CPACR_EL1_ZEN;
-		if (cpus_have_final_cap(ARM64_SME))
-			val |= CPACR_EL1_SMEN;
-
-		write_sysreg(val, cpacr_el1);
-	} else {
-		u64 val = CPTR_NVHE_EL2_RES1;
-
-		if (!cpus_have_final_cap(ARM64_SVE))
-			val |= CPTR_EL2_TZ;
-		if (!cpus_have_final_cap(ARM64_SME))
-			val |= CPTR_EL2_TSM;
-
-		write_sysreg(val, cptr_el2);
-	}
-}
-
 static void __activate_traps(struct kvm_vcpu *vcpu)
 {
 	___activate_traps(vcpu, vcpu->arch.hcr_el2);
diff --git a/arch/arm64/kvm/hyp/vhe/switch.c b/arch/arm64/kvm/hyp/vhe/switch.c
index 09df2b42bc1b..477f1580ffea 100644
--- a/arch/arm64/kvm/hyp/vhe/switch.c
+++ b/arch/arm64/kvm/hyp/vhe/switch.c
@@ -90,87 +90,6 @@ static u64 __compute_hcr(struct kvm_vcpu *vcpu)
 	return hcr | (guest_hcr & ~NV_HCR_GUEST_EXCLUDE);
 }
 
-static void __activate_cptr_traps(struct kvm_vcpu *vcpu)
-{
-	u64 cptr;
-
-	/*
-	 * With VHE (HCR.E2H == 1), accesses to CPACR_EL1 are routed to
-	 * CPTR_EL2. In general, CPACR_EL1 has the same layout as CPTR_EL2,
-	 * except for some missing controls, such as TAM.
-	 * In this case, CPTR_EL2.TAM has the same position with or without
-	 * VHE (HCR.E2H == 1) which allows us to use here the CPTR_EL2.TAM
-	 * shift value for trapping the AMU accesses.
-	 */
-	u64 val = CPACR_EL1_TTA | CPTR_EL2_TAM;
-
-	if (guest_owns_fp_regs()) {
-		val |= CPACR_EL1_FPEN;
-		if (vcpu_has_sve(vcpu))
-			val |= CPACR_EL1_ZEN;
-	} else {
-		__activate_traps_fpsimd32(vcpu);
-	}
-
-	if (!vcpu_has_nv(vcpu))
-		goto write;
-
-	/*
-	 * The architecture is a bit crap (what a surprise): an EL2 guest
-	 * writing to CPTR_EL2 via CPACR_EL1 can't set any of TCPAC or TTA,
-	 * as they are RES0 in the guest's view. To work around it, trap the
-	 * sucker using the very same bit it can't set...
-	 */
-	if (vcpu_el2_e2h_is_set(vcpu) && is_hyp_ctxt(vcpu))
-		val |= CPTR_EL2_TCPAC;
-
-	/*
-	 * Layer the guest hypervisor's trap configuration on top of our own if
-	 * we're in a nested context.
-	 */
-	if (is_hyp_ctxt(vcpu))
-		goto write;
-
-	cptr = vcpu_sanitised_cptr_el2(vcpu);
-
-	/*
-	 * Pay attention, there's some interesting detail here.
-	 *
-	 * The CPTR_EL2.xEN fields are 2 bits wide, although there are only two
-	 * meaningful trap states when HCR_EL2.TGE = 0 (running a nested guest):
-	 *
-	 *  - CPTR_EL2.xEN = x0, traps are enabled
-	 *  - CPTR_EL2.xEN = x1, traps are disabled
-	 *
-	 * In other words, bit[0] determines if guest accesses trap or not. In
-	 * the interest of simplicity, clear the entire field if the guest
-	 * hypervisor has traps enabled to dispel any illusion of something more
-	 * complicated taking place.
-	 */
-	if (!(SYS_FIELD_GET(CPACR_EL1, FPEN, cptr) & BIT(0)))
-		val &= ~CPACR_EL1_FPEN;
-	if (!(SYS_FIELD_GET(CPACR_EL1, ZEN, cptr) & BIT(0)))
-		val &= ~CPACR_EL1_ZEN;
-
-	if (kvm_has_feat(vcpu->kvm, ID_AA64MMFR3_EL1, S2POE, IMP))
-		val |= cptr & CPACR_EL1_E0POE;
-
-	val |= cptr & CPTR_EL2_TCPAC;
-
-write:
-	write_sysreg(val, cpacr_el1);
-}
-
-static void __deactivate_cptr_traps(struct kvm_vcpu *vcpu)
-{
-	u64 val = CPACR_EL1_FPEN | CPACR_EL1_ZEN_EL1EN;
-
-	if (cpus_have_final_cap(ARM64_SME))
-		val |= CPACR_EL1_SMEN_EL1EN;
-
-	write_sysreg(val, cpacr_el1);
-}
-
 static void __activate_traps(struct kvm_vcpu *vcpu)
 {
 	u64 val;
@@ -639,10 +558,10 @@ static int __kvm_vcpu_run_vhe(struct kvm_vcpu *vcpu)
 	host_ctxt = host_data_ptr(host_ctxt);
 	guest_ctxt = &vcpu->arch.ctxt;
 
-	sysreg_save_host_state_vhe(host_ctxt);
-
 	fpsimd_lazy_switch_to_guest(vcpu);
 
+	sysreg_save_host_state_vhe(host_ctxt);
+
 	/*
 	 * Note that ARM erratum 1165522 requires us to configure both stage 1
 	 * and stage 2 translation for the guest context before we clear
@@ -667,15 +586,23 @@ static int __kvm_vcpu_run_vhe(struct kvm_vcpu *vcpu)
 
 	__deactivate_traps(vcpu);
 
-	fpsimd_lazy_switch_to_host(vcpu);
-
 	sysreg_restore_host_state_vhe(host_ctxt);
 
+	__debug_switch_to_host(vcpu);
+
+	/*
+	 * Ensure that all system register writes above have taken effect
+	 * before returning to the host. In VHE mode, CPTR traps for
+	 * FPSIMD/SVE/SME also apply to EL2, so FPSIMD/SVE/SME state must be
+	 * manipulated after the ISB.
+	 */
+	isb();
+
+	fpsimd_lazy_switch_to_host(vcpu);
+
 	if (guest_owns_fp_regs())
 		__fpsimd_save_fpexc32(vcpu);
 
-	__debug_switch_to_host(vcpu);
-
 	return exit_code;
 }
 NOKPROBE_SYMBOL(__kvm_vcpu_run_vhe);
@@ -705,12 +632,6 @@ int __kvm_vcpu_run(struct kvm_vcpu *vcpu)
 	 */
 	local_daif_restore(DAIF_PROCCTX_NOIRQ);
 
-	/*
-	 * When we exit from the guest we change a number of CPU configuration
-	 * parameters, such as traps.  We rely on the isb() in kvm_call_hyp*()
-	 * to make sure these changes take effect before running the host or
-	 * additional guests.
-	 */
 	return ret;
 }