Adding upstream version 6.7.7.upstream/6.7.7

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

1 files changed, 170 insertions, 30 deletions

diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 0429488ba1..b5c6af0bef 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -391,9 +391,27 @@ static void kvmppc_set_pvr_hv(struct kvm_vcpu *vcpu, u32 pvr)
 /* Dummy value used in computing PCR value below */
 #define PCR_ARCH_31    (PCR_ARCH_300 << 1)
 
+static inline unsigned long map_pcr_to_cap(unsigned long pcr)
+{
+	unsigned long cap = 0;
+
+	switch (pcr) {
+	case PCR_ARCH_300:
+		cap = H_GUEST_CAP_POWER9;
+		break;
+	case PCR_ARCH_31:
+		cap = H_GUEST_CAP_POWER10;
+		break;
+	default:
+		break;
+	}
+
+	return cap;
+}
+
 static int kvmppc_set_arch_compat(struct kvm_vcpu *vcpu, u32 arch_compat)
 {
-	unsigned long host_pcr_bit = 0, guest_pcr_bit = 0;
+	unsigned long host_pcr_bit = 0, guest_pcr_bit = 0, cap = 0;
 	struct kvmppc_vcore *vc = vcpu->arch.vcore;
 
 	/* We can (emulate) our own architecture version and anything older */
@@ -437,8 +455,20 @@ static int kvmppc_set_arch_compat(struct kvm_vcpu *vcpu, u32 arch_compat)
 	if (guest_pcr_bit > host_pcr_bit)
 		return -EINVAL;
 
+	if (kvmhv_on_pseries() && kvmhv_is_nestedv2()) {
+		/*
+		 * 'arch_compat == 0' would mean the guest should default to
+		 * L1's compatibility. In this case, the guest would pick
+		 * host's PCR and evaluate the corresponding capabilities.
+		 */
+		cap = map_pcr_to_cap(guest_pcr_bit);
+		if (!(cap & nested_capabilities))
+			return -EINVAL;
+	}
+
 	spin_lock(&vc->lock);
 	vc->arch_compat = arch_compat;
+	kvmhv_nestedv2_mark_dirty(vcpu, KVMPPC_GSID_LOGICAL_PVR);
 	/*
 	 * Set all PCR bits for which guest_pcr_bit <= bit < host_pcr_bit
 	 * Also set all reserved PCR bits
@@ -794,7 +824,7 @@ static void kvmppc_update_vpa_dispatch(struct kvm_vcpu *vcpu,
 
 	vpa->enqueue_dispatch_tb = cpu_to_be64(be64_to_cpu(vpa->enqueue_dispatch_tb) + stolen);
 
-	__kvmppc_create_dtl_entry(vcpu, vpa, vc->pcpu, now + vc->tb_offset, stolen);
+	__kvmppc_create_dtl_entry(vcpu, vpa, vc->pcpu, now + kvmppc_get_tb_offset(vcpu), stolen);
 
 	vcpu->arch.vpa.dirty = true;
 }
@@ -845,9 +875,9 @@ static bool kvmppc_doorbell_pending(struct kvm_vcpu *vcpu)
 
 static bool kvmppc_power8_compatible(struct kvm_vcpu *vcpu)
 {
-	if (vcpu->arch.vcore->arch_compat >= PVR_ARCH_207)
+	if (kvmppc_get_arch_compat(vcpu) >= PVR_ARCH_207)
 		return true;
-	if ((!vcpu->arch.vcore->arch_compat) &&
+	if ((!kvmppc_get_arch_compat(vcpu)) &&
 	    cpu_has_feature(CPU_FTR_ARCH_207S))
 		return true;
 	return false;
@@ -1267,10 +1297,14 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
 			return RESUME_HOST;
 		break;
 #endif
-	case H_RANDOM:
-		if (!arch_get_random_seed_longs(&vcpu->arch.regs.gpr[4], 1))
+	case H_RANDOM: {
+		unsigned long rand;
+
+		if (!arch_get_random_seed_longs(&rand, 1))
 			ret = H_HARDWARE;
+		kvmppc_set_gpr(vcpu, 4, rand);
 		break;
+	}
 	case H_RPT_INVALIDATE:
 		ret = kvmppc_h_rpt_invalidate(vcpu, kvmppc_get_gpr(vcpu, 4),
 					      kvmppc_get_gpr(vcpu, 5),
@@ -2183,6 +2217,7 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
 	}
 
 	vc->lpcr = new_lpcr;
+	kvmhv_nestedv2_mark_dirty(vcpu, KVMPPC_GSID_LPCR);
 
 	spin_unlock(&vc->lock);
 }
@@ -2279,7 +2314,7 @@ static int kvmppc_get_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 			*val = get_reg_val(id, vcpu->arch.vcore->dpdes);
 		break;
 	case KVM_REG_PPC_VTB:
-		*val = get_reg_val(id, vcpu->arch.vcore->vtb);
+		*val = get_reg_val(id, kvmppc_get_vtb(vcpu));
 		break;
 	case KVM_REG_PPC_DAWR:
 		*val = get_reg_val(id, kvmppc_get_dawr0_hv(vcpu));
@@ -2306,7 +2341,7 @@ static int kvmppc_get_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 		*val = get_reg_val(id, vcpu->arch.tcscr);
 		break;
 	case KVM_REG_PPC_PID:
-		*val = get_reg_val(id, vcpu->arch.pid);
+		*val = get_reg_val(id, kvmppc_get_pid(vcpu));
 		break;
 	case KVM_REG_PPC_ACOP:
 		*val = get_reg_val(id, vcpu->arch.acop);
@@ -2338,11 +2373,11 @@ static int kvmppc_get_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 		spin_unlock(&vcpu->arch.vpa_update_lock);
 		break;
 	case KVM_REG_PPC_TB_OFFSET:
-		*val = get_reg_val(id, vcpu->arch.vcore->tb_offset);
+		*val = get_reg_val(id, kvmppc_get_tb_offset(vcpu));
 		break;
 	case KVM_REG_PPC_LPCR:
 	case KVM_REG_PPC_LPCR_64:
-		*val = get_reg_val(id, vcpu->arch.vcore->lpcr);
+		*val = get_reg_val(id, kvmppc_get_lpcr(vcpu));
 		break;
 	case KVM_REG_PPC_PPR:
 		*val = get_reg_val(id, kvmppc_get_ppr_hv(vcpu));
@@ -2414,10 +2449,10 @@ static int kvmppc_get_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 		break;
 #endif
 	case KVM_REG_PPC_ARCH_COMPAT:
-		*val = get_reg_val(id, vcpu->arch.vcore->arch_compat);
+		*val = get_reg_val(id, kvmppc_get_arch_compat(vcpu));
 		break;
 	case KVM_REG_PPC_DEC_EXPIRY:
-		*val = get_reg_val(id, vcpu->arch.dec_expires);
+		*val = get_reg_val(id, kvmppc_get_dec_expires(vcpu));
 		break;
 	case KVM_REG_PPC_ONLINE:
 		*val = get_reg_val(id, vcpu->arch.online);
@@ -2522,7 +2557,7 @@ static int kvmppc_set_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 			vcpu->arch.vcore->dpdes = set_reg_val(id, *val);
 		break;
 	case KVM_REG_PPC_VTB:
-		vcpu->arch.vcore->vtb = set_reg_val(id, *val);
+		kvmppc_set_vtb(vcpu, set_reg_val(id, *val));
 		break;
 	case KVM_REG_PPC_DAWR:
 		kvmppc_set_dawr0_hv(vcpu, set_reg_val(id, *val));
@@ -2552,7 +2587,7 @@ static int kvmppc_set_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 		vcpu->arch.tcscr = set_reg_val(id, *val);
 		break;
 	case KVM_REG_PPC_PID:
-		vcpu->arch.pid = set_reg_val(id, *val);
+		kvmppc_set_pid(vcpu, set_reg_val(id, *val));
 		break;
 	case KVM_REG_PPC_ACOP:
 		vcpu->arch.acop = set_reg_val(id, *val);
@@ -2605,10 +2640,11 @@ static int kvmppc_set_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 		 * decrementer, which is better than a large one that
 		 * causes a hang.
 		 */
-		if (!vcpu->arch.dec_expires && tb_offset)
-			vcpu->arch.dec_expires = get_tb() + tb_offset;
+		kvmppc_set_tb_offset(vcpu, tb_offset);
+		if (!kvmppc_get_dec_expires(vcpu) && tb_offset)
+			kvmppc_set_dec_expires(vcpu, get_tb() + tb_offset);
 
-		vcpu->arch.vcore->tb_offset = tb_offset;
+		kvmppc_set_tb_offset(vcpu, tb_offset);
 		break;
 	}
 	case KVM_REG_PPC_LPCR:
@@ -2689,7 +2725,7 @@ static int kvmppc_set_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 		r = kvmppc_set_arch_compat(vcpu, set_reg_val(id, *val));
 		break;
 	case KVM_REG_PPC_DEC_EXPIRY:
-		vcpu->arch.dec_expires = set_reg_val(id, *val);
+		kvmppc_set_dec_expires(vcpu, set_reg_val(id, *val));
 		break;
 	case KVM_REG_PPC_ONLINE:
 		i = set_reg_val(id, *val);
@@ -2922,8 +2958,14 @@ static int kvmppc_core_vcpu_create_hv(struct kvm_vcpu *vcpu)
 	vcpu->arch.shared_big_endian = false;
 #endif
 #endif
-	kvmppc_set_mmcr_hv(vcpu, 0, MMCR0_FC);
 
+	if (kvmhv_is_nestedv2()) {
+		err = kvmhv_nestedv2_vcpu_create(vcpu, &vcpu->arch.nestedv2_io);
+		if (err < 0)
+			return err;
+	}
+
+	kvmppc_set_mmcr_hv(vcpu, 0, MMCR0_FC);
 	if (cpu_has_feature(CPU_FTR_ARCH_31)) {
 		kvmppc_set_mmcr_hv(vcpu, 0, kvmppc_get_mmcr_hv(vcpu, 0) | MMCR0_PMCCEXT);
 		kvmppc_set_mmcra_hv(vcpu, MMCRA_BHRB_DISABLE);
@@ -3079,6 +3121,8 @@ static void kvmppc_core_vcpu_free_hv(struct kvm_vcpu *vcpu)
 	unpin_vpa(vcpu->kvm, &vcpu->arch.slb_shadow);
 	unpin_vpa(vcpu->kvm, &vcpu->arch.vpa);
 	spin_unlock(&vcpu->arch.vpa_update_lock);
+	if (kvmhv_is_nestedv2())
+		kvmhv_nestedv2_vcpu_free(vcpu, &vcpu->arch.nestedv2_io);
 }
 
 static int kvmppc_core_check_requests_hv(struct kvm_vcpu *vcpu)
@@ -4043,10 +4087,58 @@ static void vcpu_vpa_increment_dispatch(struct kvm_vcpu *vcpu)
 	}
 }
 
+static int kvmhv_vcpu_entry_nestedv2(struct kvm_vcpu *vcpu, u64 time_limit,
+				     unsigned long lpcr, u64 *tb)
+{
+	struct kvmhv_nestedv2_io *io;
+	unsigned long msr, i;
+	int trap;
+	long rc;
+
+	io = &vcpu->arch.nestedv2_io;
+
+	msr = mfmsr();
+	kvmppc_msr_hard_disable_set_facilities(vcpu, msr);
+	if (lazy_irq_pending())
+		return 0;
+
+	rc = kvmhv_nestedv2_flush_vcpu(vcpu, time_limit);
+	if (rc < 0)
+		return -EINVAL;
+
+	accumulate_time(vcpu, &vcpu->arch.in_guest);
+	rc = plpar_guest_run_vcpu(0, vcpu->kvm->arch.lpid, vcpu->vcpu_id,
+				  &trap, &i);
+
+	if (rc != H_SUCCESS) {
+		pr_err("KVM Guest Run VCPU hcall failed\n");
+		if (rc == H_INVALID_ELEMENT_ID)
+			pr_err("KVM: Guest Run VCPU invalid element id at %ld\n", i);
+		else if (rc == H_INVALID_ELEMENT_SIZE)
+			pr_err("KVM: Guest Run VCPU invalid element size at %ld\n", i);
+		else if (rc == H_INVALID_ELEMENT_VALUE)
+			pr_err("KVM: Guest Run VCPU invalid element value at %ld\n", i);
+		return -EINVAL;
+	}
+	accumulate_time(vcpu, &vcpu->arch.guest_exit);
+
+	*tb = mftb();
+	kvmppc_gsm_reset(io->vcpu_message);
+	kvmppc_gsm_reset(io->vcore_message);
+	kvmppc_gsbm_zero(&io->valids);
+
+	rc = kvmhv_nestedv2_parse_output(vcpu);
+	if (rc < 0)
+		return -EINVAL;
+
+	timer_rearm_host_dec(*tb);
+
+	return trap;
+}
+
 /* call our hypervisor to load up HV regs and go */
 static int kvmhv_vcpu_entry_p9_nested(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpcr, u64 *tb)
 {
-	struct kvmppc_vcore *vc = vcpu->arch.vcore;
 	unsigned long host_psscr;
 	unsigned long msr;
 	struct hv_guest_state hvregs;
@@ -4126,7 +4218,7 @@ static int kvmhv_vcpu_entry_p9_nested(struct kvm_vcpu *vcpu, u64 time_limit, uns
 	if (!(lpcr & LPCR_LD)) /* Sign extend if not using large decrementer */
 		dec = (s32) dec;
 	*tb = mftb();
-	vcpu->arch.dec_expires = dec + (*tb + vc->tb_offset);
+	vcpu->arch.dec_expires = dec + (*tb + kvmppc_get_tb_offset(vcpu));
 
 	timer_rearm_host_dec(*tb);
 
@@ -4161,7 +4253,10 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
 	vcpu_vpa_increment_dispatch(vcpu);
 
 	if (kvmhv_on_pseries()) {
-		trap = kvmhv_vcpu_entry_p9_nested(vcpu, time_limit, lpcr, tb);
+		if (kvmhv_is_nestedv1())
+			trap = kvmhv_vcpu_entry_p9_nested(vcpu, time_limit, lpcr, tb);
+		else
+			trap = kvmhv_vcpu_entry_nestedv2(vcpu, time_limit, lpcr, tb);
 
 		/* H_CEDE has to be handled now, not later */
 		if (trap == BOOK3S_INTERRUPT_SYSCALL && !nested &&
@@ -4691,7 +4786,7 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
 
 	tb = mftb();
 
-	kvmppc_update_vpa_dispatch_p9(vcpu, vc, tb + vc->tb_offset);
+	kvmppc_update_vpa_dispatch_p9(vcpu, vc, tb + kvmppc_get_tb_offset(vcpu));
 
 	trace_kvm_guest_enter(vcpu);
 
@@ -5147,6 +5242,14 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
 		if (++cores_done >= kvm->arch.online_vcores)
 			break;
 	}
+
+	if (kvmhv_is_nestedv2()) {
+		struct kvm_vcpu *vcpu;
+
+		kvm_for_each_vcpu(i, vcpu, kvm) {
+			kvmhv_nestedv2_mark_dirty(vcpu, KVMPPC_GSID_LPCR);
+		}
+	}
 }
 
 void kvmppc_setup_partition_table(struct kvm *kvm)
@@ -5413,15 +5516,43 @@ static int kvmppc_core_init_vm_hv(struct kvm *kvm)
 
 	/* Allocate the guest's logical partition ID */
 
-	lpid = kvmppc_alloc_lpid();
-	if ((long)lpid < 0)
-		return -ENOMEM;
-	kvm->arch.lpid = lpid;
+	if (!kvmhv_is_nestedv2()) {
+		lpid = kvmppc_alloc_lpid();
+		if ((long)lpid < 0)
+			return -ENOMEM;
+		kvm->arch.lpid = lpid;
+	}
 
 	kvmppc_alloc_host_rm_ops();
 
 	kvmhv_vm_nested_init(kvm);
 
+	if (kvmhv_is_nestedv2()) {
+		long rc;
+		unsigned long guest_id;
+
+		rc = plpar_guest_create(0, &guest_id);
+
+		if (rc != H_SUCCESS)
+			pr_err("KVM: Create Guest hcall failed, rc=%ld\n", rc);
+
+		switch (rc) {
+		case H_PARAMETER:
+		case H_FUNCTION:
+		case H_STATE:
+			return -EINVAL;
+		case H_NOT_ENOUGH_RESOURCES:
+		case H_ABORTED:
+			return -ENOMEM;
+		case H_AUTHORITY:
+			return -EPERM;
+		case H_NOT_AVAILABLE:
+			return -EBUSY;
+		}
+		kvm->arch.lpid = guest_id;
+	}
+
+
 	/*
 	 * Since we don't flush the TLB when tearing down a VM,
 	 * and this lpid might have previously been used,
@@ -5491,7 +5622,10 @@ static int kvmppc_core_init_vm_hv(struct kvm *kvm)
 			lpcr |= LPCR_HAIL;
 		ret = kvmppc_init_vm_radix(kvm);
 		if (ret) {
-			kvmppc_free_lpid(kvm->arch.lpid);
+			if (kvmhv_is_nestedv2())
+				plpar_guest_delete(0, kvm->arch.lpid);
+			else
+				kvmppc_free_lpid(kvm->arch.lpid);
 			return ret;
 		}
 		kvmppc_setup_partition_table(kvm);
@@ -5581,10 +5715,14 @@ static void kvmppc_core_destroy_vm_hv(struct kvm *kvm)
 		kvm->arch.process_table = 0;
 		if (kvm->arch.secure_guest)
 			uv_svm_terminate(kvm->arch.lpid);
-		kvmhv_set_ptbl_entry(kvm->arch.lpid, 0, 0);
+		if (!kvmhv_is_nestedv2())
+			kvmhv_set_ptbl_entry(kvm->arch.lpid, 0, 0);
 	}
 
-	kvmppc_free_lpid(kvm->arch.lpid);
+	if (kvmhv_is_nestedv2())
+		plpar_guest_delete(0, kvm->arch.lpid);
+	else
+		kvmppc_free_lpid(kvm->arch.lpid);
 
 	kvmppc_free_pimap(kvm);
 }
@@ -5996,6 +6134,8 @@ static int kvmhv_enable_nested(struct kvm *kvm)
 		return -ENODEV;
 	if (!radix_enabled())
 		return -ENODEV;
+	if (kvmhv_is_nestedv2())
+		return -ENODEV;
 
 	/* kvm == NULL means the caller is testing if the capability exists */
 	if (kvm)