Merge tag 'kvm-ppc-next-5.5-1' of...

Merge tag 'kvm-ppc-next-5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc into HEAD

KVM PPC update for 5.5

* Add capability to tell userspace whether we can single-step the guest.

* Improve the allocation of XIVE virtual processor IDs, to reduce the
  risk of running out of IDs when running many VMs on POWER9.

* Rewrite interrupt synthesis code to deliver interrupts in virtual
  mode when appropriate.

* Minor cleanups and improvements.

Documentation/virt/kvm/api.txt

@@ -2982,6 +2982,9 @@ can be determined by querying the KVM_CAP_GUEST_DEBUG_HW_BPS and
 KVM_CAP_GUEST_DEBUG_HW_WPS capabilities which return a positive number
 indicating the number of supported registers.
 
+For ppc, the KVM_CAP_PPC_GUEST_DEBUG_SSTEP capability indicates whether
+the single-step debug event (KVM_GUESTDBG_SINGLESTEP) is supported.
+
 When debug events exit the main run loop with the reason
 KVM_EXIT_DEBUG with the kvm_debug_exit_arch part of the kvm_run
 structure containing architecture specific debug information.

Documentation/virt/kvm/devices/xics.txt

@@ -3,9 +3,19 @@ XICS interrupt controller
 Device type supported: KVM_DEV_TYPE_XICS
 
 Groups:
-  KVM_DEV_XICS_SOURCES
+  1. KVM_DEV_XICS_GRP_SOURCES
   Attributes: One per interrupt source, indexed by the source number.
 
+  2. KVM_DEV_XICS_GRP_CTRL
+  Attributes:
+    2.1 KVM_DEV_XICS_NR_SERVERS (write only)
+  The kvm_device_attr.addr points to a __u32 value which is the number of
+  interrupt server numbers (ie, highest possible vcpu id plus one).
+  Errors:
+    -EINVAL: Value greater than KVM_MAX_VCPU_ID.
+    -EFAULT: Invalid user pointer for attr->addr.
+    -EBUSY:  A vcpu is already connected to the device.
+
 This device emulates the XICS (eXternal Interrupt Controller
 Specification) defined in PAPR.  The XICS has a set of interrupt
 sources, each identified by a 20-bit source number, and a set of
@@ -38,7 +48,7 @@ least-significant end of the word:
 
 Each source has 64 bits of state that can be read and written using
 the KVM_GET_DEVICE_ATTR and KVM_SET_DEVICE_ATTR ioctls, specifying the
-KVM_DEV_XICS_SOURCES attribute group, with the attribute number being
+KVM_DEV_XICS_GRP_SOURCES attribute group, with the attribute number being
 the interrupt source number.  The 64 bit state word has the following
 bitfields, starting from the least-significant end of the word:
 

Documentation/virt/kvm/devices/xive.txt

@@ -78,6 +78,14 @@ the legacy interrupt mode, referred as XICS (POWER7/8).
     migrating the VM.
     Errors: none
 
+    1.3 KVM_DEV_XIVE_NR_SERVERS (write only)
+    The kvm_device_attr.addr points to a __u32 value which is the number of
+    interrupt server numbers (ie, highest possible vcpu id plus one).
+    Errors:
+      -EINVAL: Value greater than KVM_MAX_VCPU_ID.
+      -EFAULT: Invalid user pointer for attr->addr.
+      -EBUSY:  A vCPU is already connected to the device.
+
   2. KVM_DEV_XIVE_GRP_SOURCE (write only)
   Initializes a new source in the XIVE device and mask it.
   Attributes:

arch/powerpc/include/asm/kvm_host.h

@@ -401,7 +401,6 @@ struct kvmppc_mmu {
 	u32  (*mfsrin)(struct kvm_vcpu *vcpu, u32 srnum);
 	int  (*xlate)(struct kvm_vcpu *vcpu, gva_t eaddr,
 		      struct kvmppc_pte *pte, bool data, bool iswrite);
-	void (*reset_msr)(struct kvm_vcpu *vcpu);
 	void (*tlbie)(struct kvm_vcpu *vcpu, ulong addr, bool large);
 	int  (*esid_to_vsid)(struct kvm_vcpu *vcpu, ulong esid, u64 *vsid);
 	u64  (*ea_to_vp)(struct kvm_vcpu *vcpu, gva_t eaddr, bool data);

arch/powerpc/include/asm/kvm_ppc.h

@@ -271,6 +271,7 @@ struct kvmppc_ops {
 			   union kvmppc_one_reg *val);
 	void (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu);
 	void (*vcpu_put)(struct kvm_vcpu *vcpu);
+	void (*inject_interrupt)(struct kvm_vcpu *vcpu, int vec, u64 srr1_flags);
 	void (*set_msr)(struct kvm_vcpu *vcpu, u64 msr);
 	int (*vcpu_run)(struct kvm_run *run, struct kvm_vcpu *vcpu);
 	struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned int id);

arch/powerpc/include/asm/reg.h

@@ -748,6 +748,18 @@
 #define SPRN_USPRG7	0x107	/* SPRG7 userspace read */
 #define SPRN_SRR0	0x01A	/* Save/Restore Register 0 */
 #define SPRN_SRR1	0x01B	/* Save/Restore Register 1 */
+
+#ifdef CONFIG_PPC_BOOK3S
+/*
+ * Bits loaded from MSR upon interrupt.
+ * PPC (64-bit) bits 33-36,42-47 are interrupt dependent, the others are
+ * loaded from MSR. The exception is that SRESET and MCE do not always load
+ * bit 62 (RI) from MSR. Don't use PPC_BITMASK for this because 32-bit uses
+ * it.
+ */
+#define   SRR1_MSR_BITS		(~0x783f0000UL)
+#endif
+
 #define   SRR1_ISI_NOPT		0x40000000 /* ISI: Not found in hash */
 #define   SRR1_ISI_N_OR_G	0x10000000 /* ISI: Access is no-exec or G */
 #define   SRR1_ISI_PROT		0x08000000 /* ISI: Other protection fault */

arch/powerpc/include/uapi/asm/kvm.h

@@ -667,6 +667,8 @@ struct kvm_ppc_cpu_char {
 
 /* PPC64 eXternal Interrupt Controller Specification */
 #define KVM_DEV_XICS_GRP_SOURCES	1	/* 64-bit source attributes */
+#define KVM_DEV_XICS_GRP_CTRL		2
+#define   KVM_DEV_XICS_NR_SERVERS	1
 
 /* Layout of 64-bit source attribute values */
 #define  KVM_XICS_DESTINATION_SHIFT	0
@@ -683,6 +685,7 @@ struct kvm_ppc_cpu_char {
 #define KVM_DEV_XIVE_GRP_CTRL		1
 #define   KVM_DEV_XIVE_RESET		1
 #define   KVM_DEV_XIVE_EQ_SYNC		2
+#define   KVM_DEV_XIVE_NR_SERVERS	3
 #define KVM_DEV_XIVE_GRP_SOURCE		2	/* 64-bit source identifier */
 #define KVM_DEV_XIVE_GRP_SOURCE_CONFIG	3	/* 64-bit source identifier */
 #define KVM_DEV_XIVE_GRP_EQ_CONFIG	4	/* 64-bit EQ identifier */

arch/powerpc/kvm/book3s.c

@@ -74,27 +74,6 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ NULL }
 };
 
-void kvmppc_unfixup_split_real(struct kvm_vcpu *vcpu)
-{
-	if (vcpu->arch.hflags & BOOK3S_HFLAG_SPLIT_HACK) {
-		ulong pc = kvmppc_get_pc(vcpu);
-		ulong lr = kvmppc_get_lr(vcpu);
-		if ((pc & SPLIT_HACK_MASK) == SPLIT_HACK_OFFS)
-			kvmppc_set_pc(vcpu, pc & ~SPLIT_HACK_MASK);
-		if ((lr & SPLIT_HACK_MASK) == SPLIT_HACK_OFFS)
-			kvmppc_set_lr(vcpu, lr & ~SPLIT_HACK_MASK);
-		vcpu->arch.hflags &= ~BOOK3S_HFLAG_SPLIT_HACK;
-	}
-}
-EXPORT_SYMBOL_GPL(kvmppc_unfixup_split_real);
-
-static inline unsigned long kvmppc_interrupt_offset(struct kvm_vcpu *vcpu)
-{
-	if (!is_kvmppc_hv_enabled(vcpu->kvm))
-		return to_book3s(vcpu)->hior;
-	return 0;
-}
-
 static inline void kvmppc_update_int_pending(struct kvm_vcpu *vcpu,
 			unsigned long pending_now, unsigned long old_pending)
 {
@@ -134,11 +113,7 @@ static inline bool kvmppc_critical_section(struct kvm_vcpu *vcpu)
 
 void kvmppc_inject_interrupt(struct kvm_vcpu *vcpu, int vec, u64 flags)
 {
-	kvmppc_unfixup_split_real(vcpu);
-	kvmppc_set_srr0(vcpu, kvmppc_get_pc(vcpu));
-	kvmppc_set_srr1(vcpu, (kvmppc_get_msr(vcpu) & ~0x783f0000ul) | flags);
-	kvmppc_set_pc(vcpu, kvmppc_interrupt_offset(vcpu) + vec);
-	vcpu->arch.mmu.reset_msr(vcpu);
+	vcpu->kvm->arch.kvm_ops->inject_interrupt(vcpu, vec, flags);
 }
 
 static int kvmppc_book3s_vec2irqprio(unsigned int vec)

arch/powerpc/kvm/book3s.h

@@ -32,4 +32,7 @@ extern void kvmppc_emulate_tabort(struct kvm_vcpu *vcpu, int ra_val);
 static inline void kvmppc_emulate_tabort(struct kvm_vcpu *vcpu, int ra_val) {}
 #endif
 
+extern void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr);
+extern void kvmppc_inject_interrupt_hv(struct kvm_vcpu *vcpu, int vec, u64 srr1_flags);
+
 #endif

arch/powerpc/kvm/book3s_32_mmu.c

@@ -90,11 +90,6 @@ static u64 kvmppc_mmu_book3s_32_ea_to_vp(struct kvm_vcpu *vcpu, gva_t eaddr,
 	return (((u64)eaddr >> 12) & 0xffff) | (vsid << 16);
 }
 
-static void kvmppc_mmu_book3s_32_reset_msr(struct kvm_vcpu *vcpu)
-{
-	kvmppc_set_msr(vcpu, 0);
-}
-
 static hva_t kvmppc_mmu_book3s_32_get_pteg(struct kvm_vcpu *vcpu,
 				      u32 sre, gva_t eaddr,
 				      bool primary)
@@ -406,7 +401,6 @@ void kvmppc_mmu_book3s_32_init(struct kvm_vcpu *vcpu)
 	mmu->mtsrin = kvmppc_mmu_book3s_32_mtsrin;
 	mmu->mfsrin = kvmppc_mmu_book3s_32_mfsrin;
 	mmu->xlate = kvmppc_mmu_book3s_32_xlate;
-	mmu->reset_msr = kvmppc_mmu_book3s_32_reset_msr;
 	mmu->tlbie = kvmppc_mmu_book3s_32_tlbie;
 	mmu->esid_to_vsid = kvmppc_mmu_book3s_32_esid_to_vsid;
 	mmu->ea_to_vp = kvmppc_mmu_book3s_32_ea_to_vp;

arch/powerpc/kvm/book3s_64_mmu.c

@@ -24,20 +24,6 @@
 #define dprintk(X...) do { } while(0)
 #endif
 
-static void kvmppc_mmu_book3s_64_reset_msr(struct kvm_vcpu *vcpu)
-{
-	unsigned long msr = vcpu->arch.intr_msr;
-	unsigned long cur_msr = kvmppc_get_msr(vcpu);
-
-	/* If transactional, change to suspend mode on IRQ delivery */
-	if (MSR_TM_TRANSACTIONAL(cur_msr))
-		msr |= MSR_TS_S;
-	else
-		msr |= cur_msr & MSR_TS_MASK;
-
-	kvmppc_set_msr(vcpu, msr);
-}
-
 static struct kvmppc_slb *kvmppc_mmu_book3s_64_find_slbe(
 				struct kvm_vcpu *vcpu,
 				gva_t eaddr)
@@ -676,7 +662,6 @@ void kvmppc_mmu_book3s_64_init(struct kvm_vcpu *vcpu)
 	mmu->slbie = kvmppc_mmu_book3s_64_slbie;
 	mmu->slbia = kvmppc_mmu_book3s_64_slbia;
 	mmu->xlate = kvmppc_mmu_book3s_64_xlate;
-	mmu->reset_msr = kvmppc_mmu_book3s_64_reset_msr;
 	mmu->tlbie = kvmppc_mmu_book3s_64_tlbie;
 	mmu->esid_to_vsid = kvmppc_mmu_book3s_64_esid_to_vsid;
 	mmu->ea_to_vp = kvmppc_mmu_book3s_64_ea_to_vp;

arch/powerpc/kvm/book3s_64_mmu_hv.c

@@ -275,18 +275,6 @@ int kvmppc_mmu_hv_init(void)
 	return 0;
 }
 
-static void kvmppc_mmu_book3s_64_hv_reset_msr(struct kvm_vcpu *vcpu)
-{
-	unsigned long msr = vcpu->arch.intr_msr;
-
-	/* If transactional, change to suspend mode on IRQ delivery */
-	if (MSR_TM_TRANSACTIONAL(vcpu->arch.shregs.msr))
-		msr |= MSR_TS_S;
-	else
-		msr |= vcpu->arch.shregs.msr & MSR_TS_MASK;
-	kvmppc_set_msr(vcpu, msr);
-}
-
 static long kvmppc_virtmode_do_h_enter(struct kvm *kvm, unsigned long flags,
 				long pte_index, unsigned long pteh,
 				unsigned long ptel, unsigned long *pte_idx_ret)
@@ -508,6 +496,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	struct vm_area_struct *vma;
 	unsigned long rcbits;
 	long mmio_update;
+	struct mm_struct *mm;
 
 	if (kvm_is_radix(kvm))
 		return kvmppc_book3s_radix_page_fault(run, vcpu, ea, dsisr);
@@ -584,6 +573,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	is_ci = false;
 	pfn = 0;
 	page = NULL;
+	mm = current->mm;
 	pte_size = PAGE_SIZE;
 	writing = (dsisr & DSISR_ISSTORE) != 0;
 	/* If writing != 0, then the HPTE must allow writing, if we get here */
@@ -592,8 +582,8 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	npages = get_user_pages_fast(hva, 1, writing ? FOLL_WRITE : 0, pages);
 	if (npages < 1) {
 		/* Check if it's an I/O mapping */
-		down_read(&current->mm->mmap_sem);
-		vma = find_vma(current->mm, hva);
+		down_read(&mm->mmap_sem);
+		vma = find_vma(mm, hva);
 		if (vma && vma->vm_start <= hva && hva + psize <= vma->vm_end &&
 		    (vma->vm_flags & VM_PFNMAP)) {
 			pfn = vma->vm_pgoff +
@@ -602,7 +592,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			is_ci = pte_ci(__pte((pgprot_val(vma->vm_page_prot))));
 			write_ok = vma->vm_flags & VM_WRITE;
 		}
-		up_read(&current->mm->mmap_sem);
+		up_read(&mm->mmap_sem);
 		if (!pfn)
 			goto out_put;
 	} else {
@@ -621,8 +611,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			 * hugepage split and collapse.
 			 */
 			local_irq_save(flags);
-			ptep = find_current_mm_pte(current->mm->pgd,
-						   hva, NULL, NULL);
+			ptep = find_current_mm_pte(mm->pgd, hva, NULL, NULL);
 			if (ptep) {
 				pte = kvmppc_read_update_linux_pte(ptep, 1);
 				if (__pte_write(pte))
@@ -2161,7 +2150,6 @@ void kvmppc_mmu_book3s_hv_init(struct kvm_vcpu *vcpu)
 	vcpu->arch.slb_nr = 32;		/* POWER7/POWER8 */
 
 	mmu->xlate = kvmppc_mmu_book3s_64_hv_xlate;
-	mmu->reset_msr = kvmppc_mmu_book3s_64_hv_reset_msr;
 
 	vcpu->arch.hflags |= BOOK3S_HFLAG_SLB;
 }

arch/powerpc/kvm/book3s_hv.c

@@ -133,7 +133,6 @@ static inline bool nesting_enabled(struct kvm *kvm)
 /* If set, the threads on each CPU core have to be in the same MMU mode */
 static bool no_mixing_hpt_and_radix;
 
-static void kvmppc_end_cede(struct kvm_vcpu *vcpu);
 static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu);
 
 /*
@@ -338,18 +337,6 @@ static void kvmppc_core_vcpu_put_hv(struct kvm_vcpu *vcpu)
 	spin_unlock_irqrestore(&vcpu->arch.tbacct_lock, flags);
 }
 
-static void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr)
-{
-	/*
-	 * Check for illegal transactional state bit combination
-	 * and if we find it, force the TS field to a safe state.
-	 */
-	if ((msr & MSR_TS_MASK) == MSR_TS_MASK)
-		msr &= ~MSR_TS_MASK;
-	vcpu->arch.shregs.msr = msr;
-	kvmppc_end_cede(vcpu);
-}
-
 static void kvmppc_set_pvr_hv(struct kvm_vcpu *vcpu, u32 pvr)
 {
 	vcpu->arch.pvr = pvr;
@@ -792,6 +779,11 @@ static int kvmppc_h_set_mode(struct kvm_vcpu *vcpu, unsigned long mflags,
 		vcpu->arch.dawr  = value1;
 		vcpu->arch.dawrx = value2;
 		return H_SUCCESS;
+	case H_SET_MODE_RESOURCE_ADDR_TRANS_MODE:
+		/* KVM does not support mflags=2 (AIL=2) */
+		if (mflags != 0 && mflags != 3)
+			return H_UNSUPPORTED_FLAG_START;
+		return H_TOO_HARD;
 	default:
 		return H_TOO_HARD;
 	}
@@ -2454,15 +2446,6 @@ static void kvmppc_set_timer(struct kvm_vcpu *vcpu)
 	vcpu->arch.timer_running = 1;
 }
 
-static void kvmppc_end_cede(struct kvm_vcpu *vcpu)
-{
-	vcpu->arch.ceded = 0;
-	if (vcpu->arch.timer_running) {
-		hrtimer_try_to_cancel(&vcpu->arch.dec_timer);
-		vcpu->arch.timer_running = 0;
-	}
-}
-
 extern int __kvmppc_vcore_entry(void);
 
 static void kvmppc_remove_runnable(struct kvmppc_vcore *vc,
@@ -5401,6 +5384,7 @@ static struct kvmppc_ops kvm_ops_hv = {
 	.set_one_reg = kvmppc_set_one_reg_hv,
 	.vcpu_load   = kvmppc_core_vcpu_load_hv,
 	.vcpu_put    = kvmppc_core_vcpu_put_hv,
+	.inject_interrupt = kvmppc_inject_interrupt_hv,
 	.set_msr     = kvmppc_set_msr_hv,
 	.vcpu_run    = kvmppc_vcpu_run_hv,
 	.vcpu_create = kvmppc_core_vcpu_create_hv,

arch/powerpc/kvm/book3s_hv_builtin.c

@@ -755,6 +755,71 @@ void kvmhv_p9_restore_lpcr(struct kvm_split_mode *sip)
 	local_paca->kvm_hstate.kvm_split_mode = NULL;
 }
 
+static void kvmppc_end_cede(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.ceded = 0;
+	if (vcpu->arch.timer_running) {
+		hrtimer_try_to_cancel(&vcpu->arch.dec_timer);
+		vcpu->arch.timer_running = 0;
+	}
+}
+
+void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr)
+{
+	/*
+	 * Check for illegal transactional state bit combination
+	 * and if we find it, force the TS field to a safe state.
+	 */
+	if ((msr & MSR_TS_MASK) == MSR_TS_MASK)
+		msr &= ~MSR_TS_MASK;
+	vcpu->arch.shregs.msr = msr;
+	kvmppc_end_cede(vcpu);
+}
+EXPORT_SYMBOL_GPL(kvmppc_set_msr_hv);
+
+static void inject_interrupt(struct kvm_vcpu *vcpu, int vec, u64 srr1_flags)
+{
+	unsigned long msr, pc, new_msr, new_pc;
+
+	msr = kvmppc_get_msr(vcpu);
+	pc = kvmppc_get_pc(vcpu);
+	new_msr = vcpu->arch.intr_msr;
+	new_pc = vec;
+
+	/* If transactional, change to suspend mode on IRQ delivery */
+	if (MSR_TM_TRANSACTIONAL(msr))
+		new_msr |= MSR_TS_S;
+	else
+		new_msr |= msr & MSR_TS_MASK;
+
+	/*
+	 * Perform MSR and PC adjustment for LPCR[AIL]=3 if it is set and
+	 * applicable. AIL=2 is not supported.
+	 *
+	 * AIL does not apply to SRESET, MCE, or HMI (which is never
+	 * delivered to the guest), and does not apply if IR=0 or DR=0.
+	 */
+	if (vec != BOOK3S_INTERRUPT_SYSTEM_RESET &&
+	    vec != BOOK3S_INTERRUPT_MACHINE_CHECK &&
+	    (vcpu->arch.vcore->lpcr & LPCR_AIL) == LPCR_AIL_3 &&
+	    (msr & (MSR_IR|MSR_DR)) == (MSR_IR|MSR_DR) ) {
+		new_msr |= MSR_IR | MSR_DR;
+		new_pc += 0xC000000000004000ULL;
+	}
+
+	kvmppc_set_srr0(vcpu, pc);
+	kvmppc_set_srr1(vcpu, (msr & SRR1_MSR_BITS) | srr1_flags);
+	kvmppc_set_pc(vcpu, new_pc);
+	vcpu->arch.shregs.msr = new_msr;
+}
+
+void kvmppc_inject_interrupt_hv(struct kvm_vcpu *vcpu, int vec, u64 srr1_flags)
+{
+	inject_interrupt(vcpu, vec, srr1_flags);
+	kvmppc_end_cede(vcpu);
+}
+EXPORT_SYMBOL_GPL(kvmppc_inject_interrupt_hv);
+
 /*
  * Is there a PRIV_DOORBELL pending for the guest (on POWER9)?
  * Can we inject a Decrementer or a External interrupt?
@@ -762,7 +827,6 @@ void kvmhv_p9_restore_lpcr(struct kvm_split_mode *sip)
 void kvmppc_guest_entry_inject_int(struct kvm_vcpu *vcpu)
 {
 	int ext;
-	unsigned long vec = 0;
 	unsigned long lpcr;
 
 	/* Insert EXTERNAL bit into LPCR at the MER bit position */
@@ -774,26 +838,16 @@ void kvmppc_guest_entry_inject_int(struct kvm_vcpu *vcpu)
 
 	if (vcpu->arch.shregs.msr & MSR_EE) {
 		if (ext) {
-			vec = BOOK3S_INTERRUPT_EXTERNAL;
+			inject_interrupt(vcpu, BOOK3S_INTERRUPT_EXTERNAL, 0);
 		} else {
 			long int dec = mfspr(SPRN_DEC);
 			if (!(lpcr & LPCR_LD))
 				dec = (int) dec;
 			if (dec < 0)
-				vec = BOOK3S_INTERRUPT_DECREMENTER;
+				inject_interrupt(vcpu,
+					BOOK3S_INTERRUPT_DECREMENTER, 0);
 		}
 	}
-	if (vec) {
-		unsigned long msr, old_msr = vcpu->arch.shregs.msr;
-
-		kvmppc_set_srr0(vcpu, kvmppc_get_pc(vcpu));
-		kvmppc_set_srr1(vcpu, old_msr);
-		kvmppc_set_pc(vcpu, vec);
-		msr = vcpu->arch.intr_msr;
-		if (MSR_TM_ACTIVE(old_msr))
-			msr |= MSR_TS_S;
-		vcpu->arch.shregs.msr = msr;
-	}
 
 	if (vcpu->arch.doorbell_request) {
 		mtspr(SPRN_DPDES, 1);

arch/powerpc/kvm/book3s_hv_nested.c

@@ -1186,7 +1186,7 @@ static int kvmhv_translate_addr_nested(struct kvm_vcpu *vcpu,
 forward_to_l1:
 	vcpu->arch.fault_dsisr = flags;
 	if (vcpu->arch.trap == BOOK3S_INTERRUPT_H_INST_STORAGE) {
-		vcpu->arch.shregs.msr &= ~0x783f0000ul;
+		vcpu->arch.shregs.msr &= SRR1_MSR_BITS;
 		vcpu->arch.shregs.msr |= flags;
 	}
 	return RESUME_HOST;

arch/powerpc/kvm/book3s_pr.c

@@ -90,7 +90,43 @@ static void kvmppc_fixup_split_real(struct kvm_vcpu *vcpu)
 	kvmppc_set_pc(vcpu, pc | SPLIT_HACK_OFFS);
 }
 
-void kvmppc_unfixup_split_real(struct kvm_vcpu *vcpu);
+static void kvmppc_unfixup_split_real(struct kvm_vcpu *vcpu)
+{
+	if (vcpu->arch.hflags & BOOK3S_HFLAG_SPLIT_HACK) {
+		ulong pc = kvmppc_get_pc(vcpu);
+		ulong lr = kvmppc_get_lr(vcpu);
+		if ((pc & SPLIT_HACK_MASK) == SPLIT_HACK_OFFS)
+			kvmppc_set_pc(vcpu, pc & ~SPLIT_HACK_MASK);
+		if ((lr & SPLIT_HACK_MASK) == SPLIT_HACK_OFFS)
+			kvmppc_set_lr(vcpu, lr & ~SPLIT_HACK_MASK);
+		vcpu->arch.hflags &= ~BOOK3S_HFLAG_SPLIT_HACK;
+	}
+}
+
+static void kvmppc_inject_interrupt_pr(struct kvm_vcpu *vcpu, int vec, u64 srr1_flags)
+{
+	unsigned long msr, pc, new_msr, new_pc;
+
+	kvmppc_unfixup_split_real(vcpu);
+
+	msr = kvmppc_get_msr(vcpu);
+	pc = kvmppc_get_pc(vcpu);
+	new_msr = vcpu->arch.intr_msr;
+	new_pc = to_book3s(vcpu)->hior + vec;
+
+#ifdef CONFIG_PPC_BOOK3S_64
+	/* If transactional, change to suspend mode on IRQ delivery */
+	if (MSR_TM_TRANSACTIONAL(msr))
+		new_msr |= MSR_TS_S;
+	else
+		new_msr |= msr & MSR_TS_MASK;
+#endif
+
+	kvmppc_set_srr0(vcpu, pc);
+	kvmppc_set_srr1(vcpu, (msr & SRR1_MSR_BITS) | srr1_flags);
+	kvmppc_set_pc(vcpu, new_pc);
+	kvmppc_set_msr(vcpu, new_msr);
+}
 
 static void kvmppc_core_vcpu_load_pr(struct kvm_vcpu *vcpu, int cpu)
 {
@@ -1761,6 +1797,7 @@ static struct kvm_vcpu *kvmppc_core_vcpu_create_pr(struct kvm *kvm,
 #else
 	/* default to book3s_32 (750) */
 	vcpu->arch.pvr = 0x84202;
+	vcpu->arch.intr_msr = 0;
 #endif
 	kvmppc_set_pvr_pr(vcpu, vcpu->arch.pvr);
 	vcpu->arch.slb_nr = 64;
@@ -2058,6 +2095,7 @@ static struct kvmppc_ops kvm_ops_pr = {
 	.set_one_reg = kvmppc_set_one_reg_pr,
 	.vcpu_load   = kvmppc_core_vcpu_load_pr,
 	.vcpu_put    = kvmppc_core_vcpu_put_pr,
+	.inject_interrupt = kvmppc_inject_interrupt_pr,
 	.set_msr     = kvmppc_set_msr_pr,
 	.vcpu_run    = kvmppc_vcpu_run_pr,
 	.vcpu_create = kvmppc_core_vcpu_create_pr,

arch/powerpc/kvm/book3s_xive.c

@@ -1211,6 +1211,45 @@ void kvmppc_xive_cleanup_vcpu(struct kvm_vcpu *vcpu)
 	vcpu->arch.xive_vcpu = NULL;
 }
 
+static bool kvmppc_xive_vcpu_id_valid(struct kvmppc_xive *xive, u32 cpu)
+{
+	/* We have a block of xive->nr_servers VPs. We just need to check
+	 * raw vCPU ids are below the expected limit for this guest's
+	 * core stride ; kvmppc_pack_vcpu_id() will pack them down to an
+	 * index that can be safely used to compute a VP id that belongs
+	 * to the VP block.
+	 */
+	return cpu < xive->nr_servers * xive->kvm->arch.emul_smt_mode;
+}
+
+int kvmppc_xive_compute_vp_id(struct kvmppc_xive *xive, u32 cpu, u32 *vp)
+{
+	u32 vp_id;
+
+	if (!kvmppc_xive_vcpu_id_valid(xive, cpu)) {
+		pr_devel("Out of bounds !\n");
+		return -EINVAL;
+	}
+
+	if (xive->vp_base == XIVE_INVALID_VP) {
+		xive->vp_base = xive_native_alloc_vp_block(xive->nr_servers);
+		pr_devel("VP_Base=%x nr_servers=%d\n", xive->vp_base, xive->nr_servers);
+
+		if (xive->vp_base == XIVE_INVALID_VP)
+			return -ENOSPC;
+	}
+
+	vp_id = kvmppc_xive_vp(xive, cpu);
+	if (kvmppc_xive_vp_in_use(xive->kvm, vp_id)) {
+		pr_devel("Duplicate !\n");
+		return -EEXIST;
+	}
+
+	*vp = vp_id;
+
+	return 0;
+}
+
 int kvmppc_xive_connect_vcpu(struct kvm_device *dev,
 			     struct kvm_vcpu *vcpu, u32 cpu)
 {
@@ -1229,20 +1268,13 @@ int kvmppc_xive_connect_vcpu(struct kvm_device *dev,
 		return -EPERM;
 	if (vcpu->arch.irq_type != KVMPPC_IRQ_DEFAULT)
 		return -EBUSY;
-	if (cpu >= (KVM_MAX_VCPUS * vcpu->kvm->arch.emul_smt_mode)) {
-		pr_devel("Out of bounds !\n");
-		return -EINVAL;
-	}
 
 	/* We need to synchronize with queue provisioning */
 	mutex_lock(&xive->lock);
 
-	vp_id = kvmppc_xive_vp(xive, cpu);
-	if (kvmppc_xive_vp_in_use(xive->kvm, vp_id)) {
-		pr_devel("Duplicate !\n");
-		r = -EEXIST;
+	r = kvmppc_xive_compute_vp_id(xive, cpu, &vp_id);
+	if (r)
 		goto bail;
-	}
 
 	xc = kzalloc(sizeof(*xc), GFP_KERNEL);
 	if (!xc) {
@@ -1834,6 +1866,43 @@ int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq, int level,
 	return 0;
 }
 
+int kvmppc_xive_set_nr_servers(struct kvmppc_xive *xive, u64 addr)
+{
+	u32 __user *ubufp = (u32 __user *) addr;
+	u32 nr_servers;
+	int rc = 0;
+
+	if (get_user(nr_servers, ubufp))
+		return -EFAULT;
+
+	pr_devel("%s nr_servers=%u\n", __func__, nr_servers);
+
+	if (!nr_servers || nr_servers > KVM_MAX_VCPU_ID)
+		return -EINVAL;
+
+	mutex_lock(&xive->lock);
+	if (xive->vp_base != XIVE_INVALID_VP)
+		/* The VP block is allocated once and freed when the device
+		 * is released. Better not allow to change its size since its
+		 * used by connect_vcpu to validate vCPU ids are valid (eg,
+		 * setting it back to a higher value could allow connect_vcpu
+		 * to come up with a VP id that goes beyond the VP block, which
+		 * is likely to cause a crash in OPAL).
+		 */
+		rc = -EBUSY;
+	else if (nr_servers > KVM_MAX_VCPUS)
+		/* We don't need more servers. Higher vCPU ids get packed
+		 * down below KVM_MAX_VCPUS by kvmppc_pack_vcpu_id().
+		 */
+		xive->nr_servers = KVM_MAX_VCPUS;
+	else
+		xive->nr_servers = nr_servers;
+
+	mutex_unlock(&xive->lock);
+
+	return rc;
+}
+
 static int xive_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr)
 {
 	struct kvmppc_xive *xive = dev->private;
@@ -1842,6 +1911,11 @@ static int xive_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr)
 	switch (attr->group) {
 	case KVM_DEV_XICS_GRP_SOURCES:
 		return xive_set_source(xive, attr->attr, attr->addr);
+	case KVM_DEV_XICS_GRP_CTRL:
+		switch (attr->attr) {
+		case KVM_DEV_XICS_NR_SERVERS:
+			return kvmppc_xive_set_nr_servers(xive, attr->addr);
+		}
 	}
 	return -ENXIO;
 }
@@ -1867,6 +1941,11 @@ static int xive_has_attr(struct kvm_device *dev, struct kvm_device_attr *attr)
 		    attr->attr < KVMPPC_XICS_NR_IRQS)
 			return 0;
 		break;
+	case KVM_DEV_XICS_GRP_CTRL:
+		switch (attr->attr) {
+		case KVM_DEV_XICS_NR_SERVERS:
+			return 0;
+		}
 	}
 	return -ENXIO;
 }
@@ -2001,10 +2080,13 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
 {
 	struct kvmppc_xive *xive;
 	struct kvm *kvm = dev->kvm;
-	int ret = 0;
 
 	pr_devel("Creating xive for partition\n");
 
+	/* Already there ? */
+	if (kvm->arch.xive)
+		return -EEXIST;
+
 	xive = kvmppc_xive_get_device(kvm, type);
 	if (!xive)
 		return -ENOMEM;
@@ -2014,12 +2096,6 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
 	xive->kvm = kvm;
 	mutex_init(&xive->lock);
 
-	/* Already there ? */
-	if (kvm->arch.xive)
-		ret = -EEXIST;
-	else
-		kvm->arch.xive = xive;
-
 	/* We use the default queue size set by the host */
 	xive->q_order = xive_native_default_eq_shift();
 	if (xive->q_order < PAGE_SHIFT)
@@ -2027,18 +2103,16 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
 	else
 		xive->q_page_order = xive->q_order - PAGE_SHIFT;
 
-	/* Allocate a bunch of VPs */
-	xive->vp_base = xive_native_alloc_vp_block(KVM_MAX_VCPUS);
-	pr_devel("VP_Base=%x\n", xive->vp_base);
-
-	if (xive->vp_base == XIVE_INVALID_VP)
-		ret = -ENOMEM;
+	/* VP allocation is delayed to the first call to connect_vcpu */
+	xive->vp_base = XIVE_INVALID_VP;
+	/* KVM_MAX_VCPUS limits the number of VMs to roughly 64 per sockets
+	 * on a POWER9 system.
+	 */
+	xive->nr_servers = KVM_MAX_VCPUS;
 
 	xive->single_escalation = xive_native_has_single_escalation();
 
-	if (ret)
-		return ret;
-
+	kvm->arch.xive = xive;
 	return 0;
 }
 
@@ -2108,9 +2182,9 @@ static int xive_debug_show(struct seq_file *m, void *private)
 		if (!xc)
 			continue;
 
-		seq_printf(m, "cpu server %#x CPPR:%#x HWCPPR:%#x"
+		seq_printf(m, "cpu server %#x VP:%#x CPPR:%#x HWCPPR:%#x"
 			   " MFRR:%#x PEND:%#x h_xirr: R=%lld V=%lld\n",
-			   xc->server_num, xc->cppr, xc->hw_cppr,
+			   xc->server_num, xc->vp_id, xc->cppr, xc->hw_cppr,
 			   xc->mfrr, xc->pending,
 			   xc->stat_rm_h_xirr, xc->stat_vm_h_xirr);
 

arch/powerpc/kvm/book3s_xive.h

@@ -135,6 +135,9 @@ struct kvmppc_xive {
 	/* Flags */
 	u8	single_escalation;
 
+	/* Number of entries in the VP block */
+	u32	nr_servers;
+
 	struct kvmppc_xive_ops *ops;
 	struct address_space   *mapping;
 	struct mutex mapping_lock;
@@ -296,6 +299,8 @@ int kvmppc_xive_attach_escalation(struct kvm_vcpu *vcpu, u8 prio,
 struct kvmppc_xive *kvmppc_xive_get_device(struct kvm *kvm, u32 type);
 void xive_cleanup_single_escalation(struct kvm_vcpu *vcpu,
 				    struct kvmppc_xive_vcpu *xc, int irq);
+int kvmppc_xive_compute_vp_id(struct kvmppc_xive *xive, u32 cpu, u32 *vp);
+int kvmppc_xive_set_nr_servers(struct kvmppc_xive *xive, u64 addr);
 
 #endif /* CONFIG_KVM_XICS */
 #endif /* _KVM_PPC_BOOK3S_XICS_H */

arch/powerpc/kvm/book3s_xive_native.c

@@ -118,19 +118,12 @@ int kvmppc_xive_native_connect_vcpu(struct kvm_device *dev,
 		return -EPERM;
 	if (vcpu->arch.irq_type != KVMPPC_IRQ_DEFAULT)
 		return -EBUSY;
-	if (server_num >= (KVM_MAX_VCPUS * vcpu->kvm->arch.emul_smt_mode)) {
-		pr_devel("Out of bounds !\n");
-		return -EINVAL;
-	}
 
 	mutex_lock(&xive->lock);
 
-	vp_id = kvmppc_xive_vp(xive, server_num);
-	if (kvmppc_xive_vp_in_use(xive->kvm, vp_id)) {
-		pr_devel("Duplicate !\n");
-		rc = -EEXIST;
+	rc = kvmppc_xive_compute_vp_id(xive, server_num, &vp_id);
+	if (rc)
 		goto bail;
-	}
 
 	xc = kzalloc(sizeof(*xc), GFP_KERNEL);
 	if (!xc) {
@@ -928,6 +921,8 @@ static int kvmppc_xive_native_set_attr(struct kvm_device *dev,
 			return kvmppc_xive_reset(xive);
 		case KVM_DEV_XIVE_EQ_SYNC:
 			return kvmppc_xive_native_eq_sync(xive);
+		case KVM_DEV_XIVE_NR_SERVERS:
+			return kvmppc_xive_set_nr_servers(xive, attr->addr);
 		}
 		break;
 	case KVM_DEV_XIVE_GRP_SOURCE:
@@ -967,6 +962,7 @@ static int kvmppc_xive_native_has_attr(struct kvm_device *dev,
 		switch (attr->attr) {
 		case KVM_DEV_XIVE_RESET:
 		case KVM_DEV_XIVE_EQ_SYNC:
+		case KVM_DEV_XIVE_NR_SERVERS:
 			return 0;
 		}
 		break;
@@ -1067,7 +1063,6 @@ static int kvmppc_xive_native_create(struct kvm_device *dev, u32 type)
 {
 	struct kvmppc_xive *xive;
 	struct kvm *kvm = dev->kvm;
-	int ret = 0;
 
 	pr_devel("Creating xive native device\n");
 
@@ -1081,27 +1076,20 @@ static int kvmppc_xive_native_create(struct kvm_device *dev, u32 type)
 	dev->private = xive;
 	xive->dev = dev;
 	xive->kvm = kvm;
-	kvm->arch.xive = xive;
 	mutex_init(&xive->mapping_lock);
 	mutex_init(&xive->lock);
 
-	/*
-	 * Allocate a bunch of VPs. KVM_MAX_VCPUS is a large value for
-	 * a default. Getting the max number of CPUs the VM was
-	 * configured with would improve our usage of the XIVE VP space.
+	/* VP allocation is delayed to the first call to connect_vcpu */
+	xive->vp_base = XIVE_INVALID_VP;
+	/* KVM_MAX_VCPUS limits the number of VMs to roughly 64 per sockets
+	 * on a POWER9 system.
 	 */
-	xive->vp_base = xive_native_alloc_vp_block(KVM_MAX_VCPUS);
-	pr_devel("VP_Base=%x\n", xive->vp_base);
-
-	if (xive->vp_base == XIVE_INVALID_VP)
-		ret = -ENXIO;
+	xive->nr_servers = KVM_MAX_VCPUS;
 
 	xive->single_escalation = xive_native_has_single_escalation();
 	xive->ops = &kvmppc_xive_native_ops;
 
-	if (ret)
-		return ret;
-
+	kvm->arch.xive = xive;
 	return 0;
 }
 
@@ -1204,8 +1192,8 @@ static int xive_native_debug_show(struct seq_file *m, void *private)
 		if (!xc)
 			continue;
 
-		seq_printf(m, "cpu server %#x NSR=%02x CPPR=%02x IBP=%02x PIPR=%02x w01=%016llx w2=%08x\n",
-			   xc->server_num,
+		seq_printf(m, "cpu server %#x VP=%#x NSR=%02x CPPR=%02x IBP=%02x PIPR=%02x w01=%016llx w2=%08x\n",
+			   xc->server_num, xc->vp_id,
 			   vcpu->arch.xive_saved_state.nsr,
 			   vcpu->arch.xive_saved_state.cppr,
 			   vcpu->arch.xive_saved_state.ipb,

arch/powerpc/kvm/e500_mmu_host.c

@@ -355,9 +355,9 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500,
 
 	if (tlbsel == 1) {
 		struct vm_area_struct *vma;
-		down_read(&current->mm->mmap_sem);
+		down_read(&kvm->mm->mmap_sem);
 
-		vma = find_vma(current->mm, hva);
+		vma = find_vma(kvm->mm, hva);
 		if (vma && hva >= vma->vm_start &&
 		    (vma->vm_flags & VM_PFNMAP)) {
 			/*
@@ -441,7 +441,7 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500,
 			tsize = max(BOOK3E_PAGESZ_4K, tsize & ~1);
 		}
 
-		up_read(&current->mm->mmap_sem);
+		up_read(&kvm->mm->mmap_sem);
 	}
 
 	if (likely(!pfnmap)) {

arch/powerpc/kvm/powerpc.c

@@ -522,6 +522,8 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_IMMEDIATE_EXIT:
 		r = 1;
 		break;
+	case KVM_CAP_PPC_GUEST_DEBUG_SSTEP:
+		/* fall through */
 	case KVM_CAP_PPC_PAIRED_SINGLES:
 	case KVM_CAP_PPC_OSI:
 	case KVM_CAP_PPC_GET_PVINFO:

include/uapi/linux/kvm.h

@@ -1000,6 +1000,7 @@ struct kvm_ppc_resize_hpt {
 #define KVM_CAP_PMU_EVENT_FILTER 173
 #define KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 174
 #define KVM_CAP_HYPERV_DIRECT_TLBFLUSH 175
+#define KVM_CAP_PPC_GUEST_DEBUG_SSTEP 176
 
 #ifdef KVM_CAP_IRQ_ROUTING