PPC KVM update for 5.1

There are no major new features this time, just a collection of bug
 fixes and improvements in various areas, including machine check
 handling and context switching of protection-key-related registers.
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2
 
 iQEcBAABCAAGBQJcb3lEAAoJEJ2a6ncsY3GflNwH/2ezxhHv7CRy18d2D3F+Kna+
 YQs3V/pJfBRvVdV7ZLxnR03H/NmzAK3UOzRfqGodYUtbF+gUDqSuM27lAxMKrjBv
 S87X5g/1ZdiQNnqYK7PIBn75Tx27vnw2kJAif8rXTfqbj8qLUsXcNhsziA16sJOA
 azbD5PBp9mOVzTojawyriJ3H8LYqw+vinad0idvFrApFCuNmMxv56FR6H+IBadt7
 1UJyx6AegQACdhxvy0CzmZjzzXw02z9zeFUa4lakm2sORc4fbbyyZ68CtkGURg7A
 8rt2j9SGt649ExpjfG2Cz/UihMGIMXSAOrpqTZMfyd9UPzPgHeKx2FidnxASUBc=
 =PIT8
 -----END PGP SIGNATURE-----

Merge tag 'kvm-ppc-next-5.1-1' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc into kvm-next

PPC KVM update for 5.1

There are no major new features this time, just a collection of bug
fixes and improvements in various areas, including machine check
handling and context switching of protection-key-related registers.

arch/powerpc/include/asm/kvm_host.h

@@ -99,6 +99,8 @@ struct kvm_nested_guest;
 
 struct kvm_vm_stat {
 	ulong remote_tlb_flush;
+	ulong num_2M_pages;
+	ulong num_1G_pages;
 };
 
 struct kvm_vcpu_stat {
@@ -377,6 +379,7 @@ struct kvmppc_mmu {
 	void (*slbmte)(struct kvm_vcpu *vcpu, u64 rb, u64 rs);
 	u64  (*slbmfee)(struct kvm_vcpu *vcpu, u64 slb_nr);
 	u64  (*slbmfev)(struct kvm_vcpu *vcpu, u64 slb_nr);
+	int  (*slbfee)(struct kvm_vcpu *vcpu, gva_t eaddr, ulong *ret_slb);
 	void (*slbie)(struct kvm_vcpu *vcpu, u64 slb_nr);
 	void (*slbia)(struct kvm_vcpu *vcpu);
 	/* book3s */

arch/powerpc/include/asm/kvm_ppc.h

@@ -36,6 +36,8 @@
 #endif
 #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
 #include <asm/paca.h>
+#include <asm/xive.h>
+#include <asm/cpu_has_feature.h>
 #endif
 
 /*
@@ -141,6 +143,7 @@ extern void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu);
 
 extern int kvmppc_core_prepare_to_enter(struct kvm_vcpu *vcpu);
 extern int kvmppc_core_pending_dec(struct kvm_vcpu *vcpu);
+extern void kvmppc_core_queue_machine_check(struct kvm_vcpu *vcpu, ulong flags);
 extern void kvmppc_core_queue_program(struct kvm_vcpu *vcpu, ulong flags);
 extern void kvmppc_core_queue_fpunavail(struct kvm_vcpu *vcpu);
 extern void kvmppc_core_queue_vec_unavail(struct kvm_vcpu *vcpu);
@@ -616,6 +619,18 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
 static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
 #endif /* CONFIG_KVM_XIVE */
 
+#ifdef CONFIG_PPC_POWERNV
+static inline bool xics_on_xive(void)
+{
+	return xive_enabled() && cpu_has_feature(CPU_FTR_HVMODE);
+}
+#else
+static inline bool xics_on_xive(void)
+{
+	return false;
+}
+#endif
+
 /*
  * Prototypes for functions called only from assembler code.
  * Having prototypes reduces sparse errors.
@@ -632,7 +647,7 @@ long int kvmppc_rm_h_confer(struct kvm_vcpu *vcpu, int target,
                             unsigned int yield_count);
 long kvmppc_h_random(struct kvm_vcpu *vcpu);
 void kvmhv_commence_exit(int trap);
-long kvmppc_realmode_machine_check(struct kvm_vcpu *vcpu);
+void kvmppc_realmode_machine_check(struct kvm_vcpu *vcpu);
 void kvmppc_subcore_enter_guest(void);
 void kvmppc_subcore_exit_guest(void);
 long kvmppc_realmode_hmi_handler(void);

arch/powerpc/include/asm/mce.h

@@ -209,7 +209,7 @@ extern int get_mce_event(struct machine_check_event *mce, bool release);
 extern void release_mce_event(void);
 extern void machine_check_queue_event(void);
 extern void machine_check_print_event_info(struct machine_check_event *evt,
-					   bool user_mode);
+					   bool user_mode, bool in_guest);
 #ifdef CONFIG_PPC_BOOK3S_64
 void flush_and_reload_slb(void);
 #endif /* CONFIG_PPC_BOOK3S_64 */

arch/powerpc/kernel/mce.c

@@ -301,13 +301,13 @@ static void machine_check_process_queued_event(struct irq_work *work)
 	while (__this_cpu_read(mce_queue_count) > 0) {
 		index = __this_cpu_read(mce_queue_count) - 1;
 		evt = this_cpu_ptr(&mce_event_queue[index]);
-		machine_check_print_event_info(evt, false);
+		machine_check_print_event_info(evt, false, false);
 		__this_cpu_dec(mce_queue_count);
 	}
 }
 
 void machine_check_print_event_info(struct machine_check_event *evt,
-				    bool user_mode)
+				    bool user_mode, bool in_guest)
 {
 	const char *level, *sevstr, *subtype;
 	static const char *mc_ue_types[] = {
@@ -387,7 +387,9 @@ void machine_check_print_event_info(struct machine_check_event *evt,
 	       evt->disposition == MCE_DISPOSITION_RECOVERED ?
 	       "Recovered" : "Not recovered");
 
-	if (user_mode) {
+	if (in_guest) {
+		printk("%s  Guest NIP: %016llx\n", level, evt->srr0);
+	} else if (user_mode) {
 		printk("%s  NIP: [%016llx] PID: %d Comm: %s\n", level,
 			evt->srr0, current->pid, current->comm);
 	} else {

arch/powerpc/kvm/Makefile

@@ -10,11 +10,6 @@ common-objs-y = $(KVM)/kvm_main.o $(KVM)/eventfd.o
 common-objs-$(CONFIG_KVM_VFIO) += $(KVM)/vfio.o
 common-objs-$(CONFIG_KVM_MMIO) += $(KVM)/coalesced_mmio.o
 
-CFLAGS_e500_mmu.o := -I.
-CFLAGS_e500_mmu_host.o := -I.
-CFLAGS_emulate.o  := -I.
-CFLAGS_emulate_loadstore.o  := -I.
-
 common-objs-y += powerpc.o emulate_loadstore.o
 obj-$(CONFIG_KVM_EXIT_TIMING) += timing.o
 obj-$(CONFIG_KVM_BOOK3S_HANDLER) += book3s_exports.o

arch/powerpc/kvm/book3s.c

@@ -39,6 +39,7 @@
 #include "book3s.h"
 #include "trace.h"
 
+#define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM
 #define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU
 
 /* #define EXIT_DEBUG */
@@ -71,6 +72,8 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "pthru_all",       VCPU_STAT(pthru_all) },
 	{ "pthru_host",      VCPU_STAT(pthru_host) },
 	{ "pthru_bad_aff",   VCPU_STAT(pthru_bad_aff) },
+	{ "largepages_2M",    VM_STAT(num_2M_pages) },
+	{ "largepages_1G",    VM_STAT(num_1G_pages) },
 	{ NULL }
 };
 
@@ -192,6 +195,13 @@ void kvmppc_book3s_queue_irqprio(struct kvm_vcpu *vcpu, unsigned int vec)
 }
 EXPORT_SYMBOL_GPL(kvmppc_book3s_queue_irqprio);
 
+void kvmppc_core_queue_machine_check(struct kvm_vcpu *vcpu, ulong flags)
+{
+	/* might as well deliver this straight away */
+	kvmppc_inject_interrupt(vcpu, BOOK3S_INTERRUPT_MACHINE_CHECK, flags);
+}
+EXPORT_SYMBOL_GPL(kvmppc_core_queue_machine_check);
+
 void kvmppc_core_queue_program(struct kvm_vcpu *vcpu, ulong flags)
 {
 	/* might as well deliver this straight away */
@@ -635,7 +645,7 @@ int kvmppc_get_one_reg(struct kvm_vcpu *vcpu, u64 id,
 				r = -ENXIO;
 				break;
 			}
-			if (xive_enabled())
+			if (xics_on_xive())
 				*val = get_reg_val(id, kvmppc_xive_get_icp(vcpu));
 			else
 				*val = get_reg_val(id, kvmppc_xics_get_icp(vcpu));
@@ -708,7 +718,7 @@ int kvmppc_set_one_reg(struct kvm_vcpu *vcpu, u64 id,
 				r = -ENXIO;
 				break;
 			}
-			if (xive_enabled())
+			if (xics_on_xive())
 				r = kvmppc_xive_set_icp(vcpu, set_reg_val(id, *val));
 			else
 				r = kvmppc_xics_set_icp(vcpu, set_reg_val(id, *val));
@@ -984,7 +994,7 @@ int kvmppc_book3s_hcall_implemented(struct kvm *kvm, unsigned long hcall)
 int kvm_set_irq(struct kvm *kvm, int irq_source_id, u32 irq, int level,
 		bool line_status)
 {
-	if (xive_enabled())
+	if (xics_on_xive())
 		return kvmppc_xive_set_irq(kvm, irq_source_id, irq, level,
 					   line_status);
 	else
@@ -1037,7 +1047,7 @@ static int kvmppc_book3s_init(void)
 
 #ifdef CONFIG_KVM_XICS
 #ifdef CONFIG_KVM_XIVE
-	if (xive_enabled()) {
+	if (xics_on_xive()) {
 		kvmppc_xive_init_module();
 		kvm_register_device_ops(&kvm_xive_ops, KVM_DEV_TYPE_XICS);
 	} else
@@ -1050,7 +1060,7 @@ static int kvmppc_book3s_init(void)
 static void kvmppc_book3s_exit(void)
 {
 #ifdef CONFIG_KVM_XICS
-	if (xive_enabled())
+	if (xics_on_xive())
 		kvmppc_xive_exit_module();
 #endif
 #ifdef CONFIG_KVM_BOOK3S_32_HANDLER

arch/powerpc/kvm/book3s_32_mmu.c

@@ -425,6 +425,7 @@ void kvmppc_mmu_book3s_32_init(struct kvm_vcpu *vcpu)
 	mmu->slbmte = NULL;
 	mmu->slbmfee = NULL;
 	mmu->slbmfev = NULL;
+	mmu->slbfee = NULL;
 	mmu->slbie = NULL;
 	mmu->slbia = NULL;
 }

arch/powerpc/kvm/book3s_64_mmu.c

@@ -435,6 +435,19 @@ static void kvmppc_mmu_book3s_64_slbmte(struct kvm_vcpu *vcpu, u64 rs, u64 rb)
 	kvmppc_mmu_map_segment(vcpu, esid << SID_SHIFT);
 }
 
+static int kvmppc_mmu_book3s_64_slbfee(struct kvm_vcpu *vcpu, gva_t eaddr,
+				       ulong *ret_slb)
+{
+	struct kvmppc_slb *slbe = kvmppc_mmu_book3s_64_find_slbe(vcpu, eaddr);
+
+	if (slbe) {
+		*ret_slb = slbe->origv;
+		return 0;
+	}
+	*ret_slb = 0;
+	return -ENOENT;
+}
+
 static u64 kvmppc_mmu_book3s_64_slbmfee(struct kvm_vcpu *vcpu, u64 slb_nr)
 {
 	struct kvmppc_slb *slbe;
@@ -670,6 +683,7 @@ void kvmppc_mmu_book3s_64_init(struct kvm_vcpu *vcpu)
 	mmu->slbmte = kvmppc_mmu_book3s_64_slbmte;
 	mmu->slbmfee = kvmppc_mmu_book3s_64_slbmfee;
 	mmu->slbmfev = kvmppc_mmu_book3s_64_slbmfev;
+	mmu->slbfee = kvmppc_mmu_book3s_64_slbfee;
 	mmu->slbie = kvmppc_mmu_book3s_64_slbie;
 	mmu->slbia = kvmppc_mmu_book3s_64_slbia;
 	mmu->xlate = kvmppc_mmu_book3s_64_xlate;

arch/powerpc/kvm/book3s_64_mmu_hv.c

@@ -441,6 +441,24 @@ int kvmppc_hv_emulate_mmio(struct kvm_run *run, struct kvm_vcpu *vcpu,
 {
 	u32 last_inst;
 
+	/*
+	 * Fast path - check if the guest physical address corresponds to a
+	 * device on the FAST_MMIO_BUS, if so we can avoid loading the
+	 * instruction all together, then we can just handle it and return.
+	 */
+	if (is_store) {
+		int idx, ret;
+
+		idx = srcu_read_lock(&vcpu->kvm->srcu);
+		ret = kvm_io_bus_write(vcpu, KVM_FAST_MMIO_BUS, (gpa_t) gpa, 0,
+				       NULL);
+		srcu_read_unlock(&vcpu->kvm->srcu, idx);
+		if (!ret) {
+			kvmppc_set_pc(vcpu, kvmppc_get_pc(vcpu) + 4);
+			return RESUME_GUEST;
+		}
+	}
+
 	/*
 	 * If we fail, we just return to the guest and try executing it again.
 	 */

arch/powerpc/kvm/book3s_64_mmu_radix.c

@@ -403,8 +403,13 @@ void kvmppc_unmap_pte(struct kvm *kvm, pte_t *pte, unsigned long gpa,
 		if (!memslot)
 			return;
 	}
-	if (shift)
+	if (shift) { /* 1GB or 2MB page */
 		page_size = 1ul << shift;
+		if (shift == PMD_SHIFT)
+			kvm->stat.num_2M_pages--;
+		else if (shift == PUD_SHIFT)
+			kvm->stat.num_1G_pages--;
+	}
 
 	gpa &= ~(page_size - 1);
 	hpa = old & PTE_RPN_MASK;
@@ -878,6 +883,14 @@ int kvmppc_book3s_instantiate_page(struct kvm_vcpu *vcpu,
 		put_page(page);
 	}
 
+	/* Increment number of large pages if we (successfully) inserted one */
+	if (!ret) {
+		if (level == 1)
+			kvm->stat.num_2M_pages++;
+		else if (level == 2)
+			kvm->stat.num_1G_pages++;
+	}
+
 	return ret;
 }
 

arch/powerpc/kvm/book3s_64_vio.c

@@ -133,7 +133,6 @@ extern void kvm_spapr_tce_release_iommu_group(struct kvm *kvm,
 					continue;
 
 				kref_put(&stit->kref, kvm_spapr_tce_liobn_put);
-				return;
 			}
 		}
 	}
@@ -338,14 +337,15 @@ long kvm_vm_ioctl_create_spapr_tce(struct kvm *kvm,
 		}
 	}
 
+	kvm_get_kvm(kvm);
 	if (!ret)
 		ret = anon_inode_getfd("kvm-spapr-tce", &kvm_spapr_tce_fops,
 				       stt, O_RDWR | O_CLOEXEC);
 
-	if (ret >= 0) {
+	if (ret >= 0)
 		list_add_rcu(&stt->list, &kvm->arch.spapr_tce_tables);
-		kvm_get_kvm(kvm);
+	else
-	}
+		kvm_put_kvm(kvm);
 
 	mutex_unlock(&kvm->lock);
 

arch/powerpc/kvm/book3s_emulate.c

@@ -47,6 +47,7 @@
 #define OP_31_XOP_SLBMFEV	851
 #define OP_31_XOP_EIOIO		854
 #define OP_31_XOP_SLBMFEE	915
+#define OP_31_XOP_SLBFEE	979
 
 #define OP_31_XOP_TBEGIN	654
 #define OP_31_XOP_TABORT	910
@@ -416,6 +417,23 @@ int kvmppc_core_emulate_op_pr(struct kvm_run *run, struct kvm_vcpu *vcpu,
 
 			vcpu->arch.mmu.slbia(vcpu);
 			break;
+		case OP_31_XOP_SLBFEE:
+			if (!(inst & 1) || !vcpu->arch.mmu.slbfee) {
+				return EMULATE_FAIL;
+			} else {
+				ulong b, t;
+				ulong cr = kvmppc_get_cr(vcpu) & ~CR0_MASK;
+
+				b = kvmppc_get_gpr(vcpu, rb);
+				if (!vcpu->arch.mmu.slbfee(vcpu, b, &t))
+					cr |= 2 << CR0_SHIFT;
+				kvmppc_set_gpr(vcpu, rt, t);
+				/* copy XER[SO] bit to CR0[SO] */
+				cr |= (vcpu->arch.regs.xer & 0x80000000) >>
+					(31 - CR0_SHIFT);
+				kvmppc_set_cr(vcpu, cr);
+			}
+			break;
 		case OP_31_XOP_SLBMFEE:
 			if (!vcpu->arch.mmu.slbmfee) {
 				emulated = EMULATE_FAIL;

arch/powerpc/kvm/book3s_hv.c

@@ -922,7 +922,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
 	case H_IPOLL:
 	case H_XIRR_X:
 		if (kvmppc_xics_enabled(vcpu)) {
-			if (xive_enabled()) {
+			if (xics_on_xive()) {
 				ret = H_NOT_AVAILABLE;
 				return RESUME_GUEST;
 			}
@@ -937,6 +937,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
 		ret = kvmppc_h_set_xdabr(vcpu, kvmppc_get_gpr(vcpu, 4),
 						kvmppc_get_gpr(vcpu, 5));
 		break;
+#ifdef CONFIG_SPAPR_TCE_IOMMU
 	case H_GET_TCE:
 		ret = kvmppc_h_get_tce(vcpu, kvmppc_get_gpr(vcpu, 4),
 						kvmppc_get_gpr(vcpu, 5));
@@ -966,6 +967,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
 		if (ret == H_TOO_HARD)
 			return RESUME_HOST;
 		break;
+#endif
 	case H_RANDOM:
 		if (!powernv_get_random_long(&vcpu->arch.regs.gpr[4]))
 			ret = H_HARDWARE;
@@ -1215,6 +1217,22 @@ static int kvmppc_handle_exit_hv(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		r = RESUME_GUEST;
 		break;
 	case BOOK3S_INTERRUPT_MACHINE_CHECK:
+		/* Print the MCE event to host console. */
+		machine_check_print_event_info(&vcpu->arch.mce_evt, false, true);
+
+		/*
+		 * If the guest can do FWNMI, exit to userspace so it can
+		 * deliver a FWNMI to the guest.
+		 * Otherwise we synthesize a machine check for the guest
+		 * so that it knows that the machine check occurred.
+		 */
+		if (!vcpu->kvm->arch.fwnmi_enabled) {
+			ulong flags = vcpu->arch.shregs.msr & 0x083c0000;
+			kvmppc_core_queue_machine_check(vcpu, flags);
+			r = RESUME_GUEST;
+			break;
+		}
+
 		/* Exit to guest with KVM_EXIT_NMI as exit reason */
 		run->exit_reason = KVM_EXIT_NMI;
 		run->hw.hardware_exit_reason = vcpu->arch.trap;
@@ -1227,8 +1245,6 @@ static int kvmppc_handle_exit_hv(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			run->flags |= KVM_RUN_PPC_NMI_DISP_NOT_RECOV;
 
 		r = RESUME_HOST;
-		/* Print the MCE event to host console. */
-		machine_check_print_event_info(&vcpu->arch.mce_evt, false);
 		break;
 	case BOOK3S_INTERRUPT_PROGRAM:
 	{
@@ -1392,7 +1408,7 @@ static int kvmppc_handle_nested_exit(struct kvm_run *run, struct kvm_vcpu *vcpu)
 		/* Pass the machine check to the L1 guest */
 		r = RESUME_HOST;
 		/* Print the MCE event to host console. */
-		machine_check_print_event_info(&vcpu->arch.mce_evt, false);
+		machine_check_print_event_info(&vcpu->arch.mce_evt, false, true);
 		break;
 	/*
 	 * We get these next two if the guest accesses a page which it thinks
@@ -1431,7 +1447,7 @@ static int kvmppc_handle_nested_exit(struct kvm_run *run, struct kvm_vcpu *vcpu)
 	case BOOK3S_INTERRUPT_HV_RM_HARD:
 		vcpu->arch.trap = 0;
 		r = RESUME_GUEST;
-		if (!xive_enabled())
+		if (!xics_on_xive())
 			kvmppc_xics_rm_complete(vcpu, 0);
 		break;
 	default:
@@ -3455,6 +3471,7 @@ int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
 	unsigned long host_dscr = mfspr(SPRN_DSCR);
 	unsigned long host_tidr = mfspr(SPRN_TIDR);
 	unsigned long host_iamr = mfspr(SPRN_IAMR);
+	unsigned long host_amr = mfspr(SPRN_AMR);
 	s64 dec;
 	u64 tb;
 	int trap, save_pmu;
@@ -3571,13 +3588,15 @@ int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
 
 	mtspr(SPRN_PSPB, 0);
 	mtspr(SPRN_WORT, 0);
-	mtspr(SPRN_AMR, 0);
 	mtspr(SPRN_UAMOR, 0);
 	mtspr(SPRN_DSCR, host_dscr);
 	mtspr(SPRN_TIDR, host_tidr);
 	mtspr(SPRN_IAMR, host_iamr);
 	mtspr(SPRN_PSPB, 0);
 
+	if (host_amr != vcpu->arch.amr)
+		mtspr(SPRN_AMR, host_amr);
+
 	msr_check_and_set(MSR_FP | MSR_VEC | MSR_VSX);
 	store_fp_state(&vcpu->arch.fp);
 #ifdef CONFIG_ALTIVEC
@@ -3650,7 +3669,7 @@ static void shrink_halt_poll_ns(struct kvmppc_vcore *vc)
 #ifdef CONFIG_KVM_XICS
 static inline bool xive_interrupt_pending(struct kvm_vcpu *vcpu)
 {
-	if (!xive_enabled())
+	if (!xics_on_xive())
 		return false;
 	return vcpu->arch.irq_pending || vcpu->arch.xive_saved_state.pipr <
 		vcpu->arch.xive_saved_state.cppr;
@@ -4210,7 +4229,7 @@ static int kvmppc_vcpu_run_hv(struct kvm_run *run, struct kvm_vcpu *vcpu)
 				vcpu->arch.fault_dar, vcpu->arch.fault_dsisr);
 			srcu_read_unlock(&kvm->srcu, srcu_idx);
 		} else if (r == RESUME_PASSTHROUGH) {
-			if (WARN_ON(xive_enabled()))
+			if (WARN_ON(xics_on_xive()))
 				r = H_SUCCESS;
 			else
 				r = kvmppc_xics_rm_complete(vcpu, 0);
@@ -4734,7 +4753,7 @@ static int kvmppc_core_init_vm_hv(struct kvm *kvm)
 		 * If xive is enabled, we route 0x500 interrupts directly
 		 * to the guest.
 		 */
-		if (xive_enabled())
+		if (xics_on_xive())
 			lpcr |= LPCR_LPES;
 	}
 
@@ -4970,7 +4989,7 @@ static int kvmppc_set_passthru_irq(struct kvm *kvm, int host_irq, int guest_gsi)
 	if (i == pimap->n_mapped)
 		pimap->n_mapped++;
 
-	if (xive_enabled())
+	if (xics_on_xive())
 		rc = kvmppc_xive_set_mapped(kvm, guest_gsi, desc);
 	else
 		kvmppc_xics_set_mapped(kvm, guest_gsi, desc->irq_data.hwirq);
@@ -5011,7 +5030,7 @@ static int kvmppc_clr_passthru_irq(struct kvm *kvm, int host_irq, int guest_gsi)
 		return -ENODEV;
 	}
 
-	if (xive_enabled())
+	if (xics_on_xive())
 		rc = kvmppc_xive_clr_mapped(kvm, guest_gsi, pimap->mapped[i].desc);
 	else
 		kvmppc_xics_clr_mapped(kvm, guest_gsi, pimap->mapped[i].r_hwirq);
@@ -5343,13 +5362,11 @@ static int kvm_init_subcore_bitmap(void)
 			continue;
 
 		sibling_subcore_state =
-			kmalloc_node(sizeof(struct sibling_subcore_state),
+			kzalloc_node(sizeof(struct sibling_subcore_state),
 							GFP_KERNEL, node);
 		if (!sibling_subcore_state)
 			return -ENOMEM;
 
-		memset(sibling_subcore_state, 0,
-				sizeof(struct sibling_subcore_state));
 
 		for (j = 0; j < threads_per_core; j++) {
 			int cpu = first_cpu + j;
@@ -5390,7 +5407,7 @@ static int kvmppc_book3s_init_hv(void)
 	 * indirectly, via OPAL.
 	 */
 #ifdef CONFIG_SMP
-	if (!xive_enabled() && !kvmhv_on_pseries() &&
+	if (!xics_on_xive() && !kvmhv_on_pseries() &&
 	    !local_paca->kvm_hstate.xics_phys) {
 		struct device_node *np;
 

arch/powerpc/kvm/book3s_hv_builtin.c

@@ -257,7 +257,7 @@ void kvmhv_rm_send_ipi(int cpu)
 	}
 
 	/* We should never reach this */
-	if (WARN_ON_ONCE(xive_enabled()))
+	if (WARN_ON_ONCE(xics_on_xive()))
 	    return;
 
 	/* Else poke the target with an IPI */
@@ -577,7 +577,7 @@ unsigned long kvmppc_rm_h_xirr(struct kvm_vcpu *vcpu)
 {
 	if (!kvmppc_xics_enabled(vcpu))
 		return H_TOO_HARD;
-	if (xive_enabled()) {
+	if (xics_on_xive()) {
 		if (is_rm())
 			return xive_rm_h_xirr(vcpu);
 		if (unlikely(!__xive_vm_h_xirr))
@@ -592,7 +592,7 @@ unsigned long kvmppc_rm_h_xirr_x(struct kvm_vcpu *vcpu)
 	if (!kvmppc_xics_enabled(vcpu))
 		return H_TOO_HARD;
 	vcpu->arch.regs.gpr[5] = get_tb();
-	if (xive_enabled()) {
+	if (xics_on_xive()) {
 		if (is_rm())
 			return xive_rm_h_xirr(vcpu);
 		if (unlikely(!__xive_vm_h_xirr))
@@ -606,7 +606,7 @@ unsigned long kvmppc_rm_h_ipoll(struct kvm_vcpu *vcpu, unsigned long server)
 {
 	if (!kvmppc_xics_enabled(vcpu))
 		return H_TOO_HARD;
-	if (xive_enabled()) {
+	if (xics_on_xive()) {
 		if (is_rm())
 			return xive_rm_h_ipoll(vcpu, server);
 		if (unlikely(!__xive_vm_h_ipoll))
@@ -621,7 +621,7 @@ int kvmppc_rm_h_ipi(struct kvm_vcpu *vcpu, unsigned long server,
 {
 	if (!kvmppc_xics_enabled(vcpu))
 		return H_TOO_HARD;
-	if (xive_enabled()) {
+	if (xics_on_xive()) {
 		if (is_rm())
 			return xive_rm_h_ipi(vcpu, server, mfrr);
 		if (unlikely(!__xive_vm_h_ipi))
@@ -635,7 +635,7 @@ int kvmppc_rm_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr)
 {
 	if (!kvmppc_xics_enabled(vcpu))
 		return H_TOO_HARD;
-	if (xive_enabled()) {
+	if (xics_on_xive()) {
 		if (is_rm())
 			return xive_rm_h_cppr(vcpu, cppr);
 		if (unlikely(!__xive_vm_h_cppr))
@@ -649,7 +649,7 @@ int kvmppc_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr)
 {
 	if (!kvmppc_xics_enabled(vcpu))
 		return H_TOO_HARD;
-	if (xive_enabled()) {
+	if (xics_on_xive()) {
 		if (is_rm())
 			return xive_rm_h_eoi(vcpu, xirr);
 		if (unlikely(!__xive_vm_h_eoi))

arch/powerpc/kvm/book3s_hv_ras.c

@@ -66,10 +66,8 @@ static void reload_slb(struct kvm_vcpu *vcpu)
 /*
  * On POWER7, see if we can handle a machine check that occurred inside
  * the guest in real mode, without switching to the host partition.
- *
- * Returns: 0 => exit guest, 1 => deliver machine check to guest
  */
-static long kvmppc_realmode_mc_power7(struct kvm_vcpu *vcpu)
+static void kvmppc_realmode_mc_power7(struct kvm_vcpu *vcpu)
 {
 	unsigned long srr1 = vcpu->arch.shregs.msr;
 	struct machine_check_event mce_evt;
@@ -111,52 +109,24 @@ static long kvmppc_realmode_mc_power7(struct kvm_vcpu *vcpu)
 	}
 
 	/*
-	 * See if we have already handled the condition in the linux host.
+	 * Now get the event and stash it in the vcpu struct so it can
-	 * We assume that if the condition is recovered then linux host
+	 * be handled by the primary thread in virtual mode.  We can't
-	 * will have generated an error log event that we will pick
+	 * call machine_check_queue_event() here if we are running on
-	 * up and log later.
+	 * an offline secondary thread.
-	 * Don't release mce event now. We will queue up the event so that
-	 * we can log the MCE event info on host console.
 	 */
-	if (!get_mce_event(&mce_evt, MCE_EVENT_DONTRELEASE))
+	if (get_mce_event(&mce_evt, MCE_EVENT_RELEASE)) {
-		goto out;
+		if (handled && mce_evt.version == MCE_V1)
+			mce_evt.disposition = MCE_DISPOSITION_RECOVERED;
+	} else {
+		memset(&mce_evt, 0, sizeof(mce_evt));
+	}
 
-	if (mce_evt.version == MCE_V1 &&
+	vcpu->arch.mce_evt = mce_evt;
-	    (mce_evt.severity == MCE_SEV_NO_ERROR ||
-	     mce_evt.disposition == MCE_DISPOSITION_RECOVERED))
-		handled = 1;
-
-out:
-	/*
-	 * For guest that supports FWNMI capability, hook the MCE event into
-	 * vcpu structure. We are going to exit the guest with KVM_EXIT_NMI
-	 * exit reason. On our way to exit we will pull this event from vcpu
-	 * structure and print it from thread 0 of the core/subcore.
-	 *
-	 * For guest that does not support FWNMI capability (old QEMU):
-	 * We are now going enter guest either through machine check
-	 * interrupt (for unhandled errors) or will continue from
-	 * current HSRR0 (for handled errors) in guest. Hence
-	 * queue up the event so that we can log it from host console later.
-	 */
-	if (vcpu->kvm->arch.fwnmi_enabled) {
-		/*
-		 * Hook up the mce event on to vcpu structure.
-		 * First clear the old event.
-		 */
-		memset(&vcpu->arch.mce_evt, 0, sizeof(vcpu->arch.mce_evt));
-		if (get_mce_event(&mce_evt, MCE_EVENT_RELEASE)) {
-			vcpu->arch.mce_evt = mce_evt;
-		}
-	} else
-		machine_check_queue_event();
-
-	return handled;
 }
 
-long kvmppc_realmode_machine_check(struct kvm_vcpu *vcpu)
+void kvmppc_realmode_machine_check(struct kvm_vcpu *vcpu)
 {
-	return kvmppc_realmode_mc_power7(vcpu);
+	kvmppc_realmode_mc_power7(vcpu);
 }
 
 /* Check if dynamic split is in force and return subcore size accordingly. */

arch/powerpc/kvm/book3s_hv_rm_xics.c

@@ -144,6 +144,13 @@ static void icp_rm_set_vcpu_irq(struct kvm_vcpu *vcpu,
 		return;
 	}
 
+	if (xive_enabled() && kvmhv_on_pseries()) {
+		/* No XICS access or hypercalls available, too hard */
+		this_icp->rm_action |= XICS_RM_KICK_VCPU;
+		this_icp->rm_kick_target = vcpu;
+		return;
+	}
+
 	/*
 	 * Check if the core is loaded,
 	 * if not, find an available host core to post to wake the VCPU,

arch/powerpc/kvm/book3s_hv_rmhandlers.S

@@ -58,6 +58,8 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
 #define STACK_SLOT_DAWR		(SFS-56)
 #define STACK_SLOT_DAWRX	(SFS-64)
 #define STACK_SLOT_HFSCR	(SFS-72)
+#define STACK_SLOT_AMR		(SFS-80)
+#define STACK_SLOT_UAMOR	(SFS-88)
 /* the following is used by the P9 short path */
 #define STACK_SLOT_NVGPRS	(SFS-152)	/* 18 gprs */
 
@@ -726,11 +728,9 @@ BEGIN_FTR_SECTION
 	mfspr	r5, SPRN_TIDR
 	mfspr	r6, SPRN_PSSCR
 	mfspr	r7, SPRN_PID
-	mfspr	r8, SPRN_IAMR
 	std	r5, STACK_SLOT_TID(r1)
 	std	r6, STACK_SLOT_PSSCR(r1)
 	std	r7, STACK_SLOT_PID(r1)
-	std	r8, STACK_SLOT_IAMR(r1)
 	mfspr	r5, SPRN_HFSCR
 	std	r5, STACK_SLOT_HFSCR(r1)
 END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
@@ -738,11 +738,18 @@ BEGIN_FTR_SECTION
 	mfspr	r5, SPRN_CIABR
 	mfspr	r6, SPRN_DAWR
 	mfspr	r7, SPRN_DAWRX
+	mfspr	r8, SPRN_IAMR
 	std	r5, STACK_SLOT_CIABR(r1)
 	std	r6, STACK_SLOT_DAWR(r1)
 	std	r7, STACK_SLOT_DAWRX(r1)
+	std	r8, STACK_SLOT_IAMR(r1)
 END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 
+	mfspr	r5, SPRN_AMR
+	std	r5, STACK_SLOT_AMR(r1)
+	mfspr	r6, SPRN_UAMOR
+	std	r6, STACK_SLOT_UAMOR(r1)
+
 BEGIN_FTR_SECTION
 	/* Set partition DABR */
 	/* Do this before re-enabling PMU to avoid P7 DABR corruption bug */
@@ -1631,22 +1638,25 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
 	mtspr	SPRN_PSPB, r0
 	mtspr	SPRN_WORT, r0
 BEGIN_FTR_SECTION
-	mtspr	SPRN_IAMR, r0
 	mtspr	SPRN_TCSCR, r0
 	/* Set MMCRS to 1<<31 to freeze and disable the SPMC counters */
 	li	r0, 1
 	sldi	r0, r0, 31
 	mtspr	SPRN_MMCRS, r0
 END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
-8:
 
-	/* Save and reset AMR and UAMOR before turning on the MMU */
+	/* Save and restore AMR, IAMR and UAMOR before turning on the MMU */
+	ld	r8, STACK_SLOT_IAMR(r1)
+	mtspr	SPRN_IAMR, r8
+
+8:	/* Power7 jumps back in here */
 	mfspr	r5,SPRN_AMR
 	mfspr	r6,SPRN_UAMOR
 	std	r5,VCPU_AMR(r9)
 	std	r6,VCPU_UAMOR(r9)
-	li	r6,0
+	ld	r5,STACK_SLOT_AMR(r1)
-	mtspr	SPRN_AMR,r6
+	ld	r6,STACK_SLOT_UAMOR(r1)
+	mtspr	SPRN_AMR, r5
 	mtspr	SPRN_UAMOR, r6
 
 	/* Switch DSCR back to host value */
@@ -1746,11 +1756,9 @@ BEGIN_FTR_SECTION
 	ld	r5, STACK_SLOT_TID(r1)
 	ld	r6, STACK_SLOT_PSSCR(r1)
 	ld	r7, STACK_SLOT_PID(r1)
-	ld	r8, STACK_SLOT_IAMR(r1)
 	mtspr	SPRN_TIDR, r5
 	mtspr	SPRN_PSSCR, r6
 	mtspr	SPRN_PID, r7
-	mtspr	SPRN_IAMR, r8
 END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
 
 #ifdef CONFIG_PPC_RADIX_MMU
@@ -2264,8 +2272,13 @@ hcall_real_table:
 	.long	DOTSYM(kvmppc_h_clear_mod) - hcall_real_table
 	.long	DOTSYM(kvmppc_h_clear_ref) - hcall_real_table
 	.long	DOTSYM(kvmppc_h_protect) - hcall_real_table
+#ifdef CONFIG_SPAPR_TCE_IOMMU
 	.long	DOTSYM(kvmppc_h_get_tce) - hcall_real_table
 	.long	DOTSYM(kvmppc_rm_h_put_tce) - hcall_real_table
+#else
+	.long	0		/* 0x1c */
+	.long	0		/* 0x20 */
+#endif
 	.long	0		/* 0x24 - H_SET_SPRG0 */
 	.long	DOTSYM(kvmppc_h_set_dabr) - hcall_real_table
 	.long	0		/* 0x2c */
@@ -2343,8 +2356,13 @@ hcall_real_table:
 	.long	0		/* 0x12c */
 	.long	0		/* 0x130 */
 	.long	DOTSYM(kvmppc_h_set_xdabr) - hcall_real_table
+#ifdef CONFIG_SPAPR_TCE_IOMMU
 	.long	DOTSYM(kvmppc_rm_h_stuff_tce) - hcall_real_table
 	.long	DOTSYM(kvmppc_rm_h_put_tce_indirect) - hcall_real_table
+#else
+	.long	0		/* 0x138 */
+	.long	0		/* 0x13c */
+#endif
 	.long	0		/* 0x140 */
 	.long	0		/* 0x144 */
 	.long	0		/* 0x148 */
@@ -2826,49 +2844,15 @@ kvm_cede_exit:
 #endif /* CONFIG_KVM_XICS */
 3:	b	guest_exit_cont
 
-	/* Try to handle a machine check in real mode */
+	/* Try to do machine check recovery in real mode */
 machine_check_realmode:
 	mr	r3, r9		/* get vcpu pointer */
 	bl	kvmppc_realmode_machine_check
 	nop
+	/* all machine checks go to virtual mode for further handling */
 	ld	r9, HSTATE_KVM_VCPU(r13)
 	li	r12, BOOK3S_INTERRUPT_MACHINE_CHECK
-	/*
+	b	guest_exit_cont
-	 * For the guest that is FWNMI capable, deliver all the MCE errors
-	 * (handled/unhandled) by exiting the guest with KVM_EXIT_NMI exit
-	 * reason. This new approach injects machine check errors in guest
-	 * address space to guest with additional information in the form
-	 * of RTAS event, thus enabling guest kernel to suitably handle
-	 * such errors.
-	 *
-	 * For the guest that is not FWNMI capable (old QEMU) fallback
-	 * to old behaviour for backward compatibility:
-	 * Deliver unhandled/fatal (e.g. UE) MCE errors to guest either
-	 * through machine check interrupt (set HSRR0 to 0x200).
-	 * For handled errors (no-fatal), just go back to guest execution
-	 * with current HSRR0.
-	 * if we receive machine check with MSR(RI=0) then deliver it to
-	 * guest as machine check causing guest to crash.
-	 */
-	ld	r11, VCPU_MSR(r9)
-	rldicl.	r0, r11, 64-MSR_HV_LG, 63 /* check if it happened in HV mode */
-	bne	guest_exit_cont		/* if so, exit to host */
-	/* Check if guest is capable of handling NMI exit */
-	ld	r10, VCPU_KVM(r9)
-	lbz	r10, KVM_FWNMI(r10)
-	cmpdi	r10, 1			/* FWNMI capable? */
-	beq	guest_exit_cont		/* if so, exit with KVM_EXIT_NMI. */
-
-	/* if not, fall through for backward compatibility. */
-	andi.	r10, r11, MSR_RI	/* check for unrecoverable exception */
-	beq	1f			/* Deliver a machine check to guest */
-	ld	r10, VCPU_PC(r9)
-	cmpdi	r3, 0		/* Did we handle MCE ? */
-	bne	2f	/* Continue guest execution. */
-	/* If not, deliver a machine check.  SRR0/1 are already set */
-1:	li	r10, BOOK3S_INTERRUPT_MACHINE_CHECK
-	bl	kvmppc_msr_interrupt
-2:	b	fast_interrupt_c_return
 
 /*
  * Call C code to handle a HMI in real mode.

arch/powerpc/kvm/book3s_rtas.c

@@ -33,7 +33,7 @@ static void kvm_rtas_set_xive(struct kvm_vcpu *vcpu, struct rtas_args *args)
 	server = be32_to_cpu(args->args[1]);
 	priority = be32_to_cpu(args->args[2]);
 
-	if (xive_enabled())
+	if (xics_on_xive())
 		rc = kvmppc_xive_set_xive(vcpu->kvm, irq, server, priority);
 	else
 		rc = kvmppc_xics_set_xive(vcpu->kvm, irq, server, priority);
@@ -56,7 +56,7 @@ static void kvm_rtas_get_xive(struct kvm_vcpu *vcpu, struct rtas_args *args)
 	irq = be32_to_cpu(args->args[0]);
 
 	server = priority = 0;
-	if (xive_enabled())
+	if (xics_on_xive())
 		rc = kvmppc_xive_get_xive(vcpu->kvm, irq, &server, &priority);
 	else
 		rc = kvmppc_xics_get_xive(vcpu->kvm, irq, &server, &priority);
@@ -83,7 +83,7 @@ static void kvm_rtas_int_off(struct kvm_vcpu *vcpu, struct rtas_args *args)
 
 	irq = be32_to_cpu(args->args[0]);
 
-	if (xive_enabled())
+	if (xics_on_xive())
 		rc = kvmppc_xive_int_off(vcpu->kvm, irq);
 	else
 		rc = kvmppc_xics_int_off(vcpu->kvm, irq);
@@ -105,7 +105,7 @@ static void kvm_rtas_int_on(struct kvm_vcpu *vcpu, struct rtas_args *args)
 
 	irq = be32_to_cpu(args->args[0]);
 
-	if (xive_enabled())
+	if (xics_on_xive())
 		rc = kvmppc_xive_int_on(vcpu->kvm, irq);
 	else
 		rc = kvmppc_xics_int_on(vcpu->kvm, irq);

arch/powerpc/kvm/powerpc.c

@@ -748,7 +748,7 @@ void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu)
 		kvmppc_mpic_disconnect_vcpu(vcpu->arch.mpic, vcpu);
 		break;
 	case KVMPPC_IRQ_XICS:
-		if (xive_enabled())
+		if (xics_on_xive())
 			kvmppc_xive_cleanup_vcpu(vcpu);
 		else
 			kvmppc_xics_free_icp(vcpu);
@@ -1931,7 +1931,7 @@ static int kvm_vcpu_ioctl_enable_cap(struct kvm_vcpu *vcpu,
 		r = -EPERM;
 		dev = kvm_device_from_filp(f.file);
 		if (dev) {
-			if (xive_enabled())
+			if (xics_on_xive())
 				r = kvmppc_xive_connect_vcpu(dev, vcpu, cap->args[1]);
 			else
 				r = kvmppc_xics_connect_vcpu(dev, vcpu, cap->args[1]);

arch/powerpc/platforms/powernv/opal.c

@@ -587,7 +587,7 @@ int opal_machine_check(struct pt_regs *regs)
 		       evt.version);
 		return 0;
 	}
-	machine_check_print_event_info(&evt, user_mode(regs));
+	machine_check_print_event_info(&evt, user_mode(regs), false);
 
 	if (opal_recover_mce(regs, &evt))
 		return 1;