Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull KVM updates from Paolo Bonzini:
 "First batch of KVM changes for 4.1

  The most interesting bit here is irqfd/ioeventfd support for ARM and
  ARM64.

  Summary:

  ARM/ARM64:
     fixes for live migration, irqfd and ioeventfd support (enabling
     vhost, too), page aging

  s390:
     interrupt handling rework, allowing to inject all local interrupts
     via new ioctl and to get/set the full local irq state for migration
     and introspection.  New ioctls to access memory by virtual address,
     and to get/set the guest storage keys.  SIMD support.

  MIPS:
     FPU and MIPS SIMD Architecture (MSA) support.  Includes some
     patches from Ralf Baechle's MIPS tree.

  x86:
     bugfixes (notably for pvclock, the others are small) and cleanups.
     Another small latency improvement for the TSC deadline timer"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (146 commits)
  KVM: use slowpath for cross page cached accesses
  kvm: mmu: lazy collapse small sptes into large sptes
  KVM: x86: Clear CR2 on VCPU reset
  KVM: x86: DR0-DR3 are not clear on reset
  KVM: x86: BSP in MSR_IA32_APICBASE is writable
  KVM: x86: simplify kvm_apic_map
  KVM: x86: avoid logical_map when it is invalid
  KVM: x86: fix mixed APIC mode broadcast
  KVM: x86: use MDA for interrupt matching
  kvm/ppc/mpic: drop unused IRQ_testbit
  KVM: nVMX: remove unnecessary double caching of MAXPHYADDR
  KVM: nVMX: checks for address bits beyond MAXPHYADDR on VM-entry
  KVM: x86: cache maxphyaddr CPUID leaf in struct kvm_vcpu
  KVM: vmx: pass error code with internal error #2
  x86: vdso: fix pvclock races with task migration
  KVM: remove kvm_read_hva and kvm_read_hva_atomic
  KVM: x86: optimize delivery of TSC deadline timer interrupt
  KVM: x86: extract blocking logic from __vcpu_run
  kvm: x86: fix x86 eflags fixed bit
  KVM: s390: migrate vcpu interrupt state
  ...

Documentation/virtual/kvm/devices/s390_flic.txt

@@ -27,6 +27,9 @@ Groups:
     Copies all floating interrupts into a buffer provided by userspace.
     When the buffer is too small it returns -ENOMEM, which is the indication
     for userspace to try again with a bigger buffer.
+    -ENOBUFS is returned when the allocation of a kernelspace buffer has
+    failed.
+    -EFAULT is returned when copying data to userspace failed.
     All interrupts remain pending, i.e. are not deleted from the list of
     currently pending interrupts.
     attr->addr contains the userspace address of the buffer into which all

MAINTAINERS

@@ -5591,6 +5591,8 @@ S:	Supported
 F:	Documentation/*/kvm*.txt
 F:	Documentation/virtual/kvm/
 F:	arch/*/kvm/
+F:	arch/x86/kernel/kvm.c
+F:	arch/x86/kernel/kvmclock.c
 F:	arch/*/include/asm/kvm*
 F:	include/linux/kvm*
 F:	include/uapi/linux/kvm*

arch/arm/include/asm/kvm_arm.h

@@ -185,6 +185,7 @@
 #define HSR_COND	(0xfU << HSR_COND_SHIFT)
 
 #define FSC_FAULT	(0x04)
+#define FSC_ACCESS	(0x08)
 #define FSC_PERM	(0x0c)
 
 /* Hyp Prefetch Fault Address Register (HPFAR/HDFAR) */

arch/arm/include/asm/kvm_host.h

@@ -27,6 +27,8 @@
 #include <asm/fpstate.h>
 #include <kvm/arm_arch_timer.h>
 
+#define __KVM_HAVE_ARCH_INTC_INITIALIZED
+
 #if defined(CONFIG_KVM_ARM_MAX_VCPUS)
 #define KVM_MAX_VCPUS CONFIG_KVM_ARM_MAX_VCPUS
 #else
@@ -165,19 +167,10 @@ void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
 
 unsigned long kvm_arm_num_regs(struct kvm_vcpu *vcpu);
 int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
+int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end);
+int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
 
 /* We do not have shadow page tables, hence the empty hooks */
-static inline int kvm_age_hva(struct kvm *kvm, unsigned long start,
-			      unsigned long end)
-{
-	return 0;
-}
-
-static inline int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
-{
-	return 0;
-}
-
 static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
 							 unsigned long address)
 {

arch/arm/include/asm/kvm_mmio.h

@@ -28,28 +28,6 @@ struct kvm_decode {
 	bool sign_extend;
 };
 
-/*
- * The in-kernel MMIO emulation code wants to use a copy of run->mmio,
- * which is an anonymous type. Use our own type instead.
- */
-struct kvm_exit_mmio {
-	phys_addr_t	phys_addr;
-	u8		data[8];
-	u32		len;
-	bool		is_write;
-	void		*private;
-};
-
-static inline void kvm_prepare_mmio(struct kvm_run *run,
-				    struct kvm_exit_mmio *mmio)
-{
-	run->mmio.phys_addr	= mmio->phys_addr;
-	run->mmio.len		= mmio->len;
-	run->mmio.is_write	= mmio->is_write;
-	memcpy(run->mmio.data, mmio->data, mmio->len);
-	run->exit_reason	= KVM_EXIT_MMIO;
-}
-
 int kvm_handle_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run);
 int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		 phys_addr_t fault_ipa);

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

@@ -198,6 +198,9 @@ struct kvm_arch_memory_slot {
 /* Highest supported SPI, from VGIC_NR_IRQS */
 #define KVM_ARM_IRQ_GIC_MAX		127
 
+/* One single KVM irqchip, ie. the VGIC */
+#define KVM_NR_IRQCHIPS          1
+
 /* PSCI interface */
 #define KVM_PSCI_FN_BASE		0x95c1ba5e
 #define KVM_PSCI_FN(n)			(KVM_PSCI_FN_BASE + (n))

arch/arm/kernel/asm-offsets.c

@@ -190,7 +190,6 @@ int main(void)
   DEFINE(VCPU_HxFAR,		offsetof(struct kvm_vcpu, arch.fault.hxfar));
   DEFINE(VCPU_HPFAR,		offsetof(struct kvm_vcpu, arch.fault.hpfar));
   DEFINE(VCPU_HYP_PC,		offsetof(struct kvm_vcpu, arch.fault.hyp_pc));
-#ifdef CONFIG_KVM_ARM_VGIC
   DEFINE(VCPU_VGIC_CPU,		offsetof(struct kvm_vcpu, arch.vgic_cpu));
   DEFINE(VGIC_V2_CPU_HCR,	offsetof(struct vgic_cpu, vgic_v2.vgic_hcr));
   DEFINE(VGIC_V2_CPU_VMCR,	offsetof(struct vgic_cpu, vgic_v2.vgic_vmcr));
@@ -200,14 +199,11 @@ int main(void)
   DEFINE(VGIC_V2_CPU_APR,	offsetof(struct vgic_cpu, vgic_v2.vgic_apr));
   DEFINE(VGIC_V2_CPU_LR,	offsetof(struct vgic_cpu, vgic_v2.vgic_lr));
   DEFINE(VGIC_CPU_NR_LR,	offsetof(struct vgic_cpu, nr_lr));
-#ifdef CONFIG_KVM_ARM_TIMER
   DEFINE(VCPU_TIMER_CNTV_CTL,	offsetof(struct kvm_vcpu, arch.timer_cpu.cntv_ctl));
   DEFINE(VCPU_TIMER_CNTV_CVAL,	offsetof(struct kvm_vcpu, arch.timer_cpu.cntv_cval));
   DEFINE(KVM_TIMER_CNTVOFF,	offsetof(struct kvm, arch.timer.cntvoff));
   DEFINE(KVM_TIMER_ENABLED,	offsetof(struct kvm, arch.timer.enabled));
-#endif
   DEFINE(KVM_VGIC_VCTRL,	offsetof(struct kvm, arch.vgic.vctrl_base));
-#endif
   DEFINE(KVM_VTTBR,		offsetof(struct kvm, arch.vttbr));
 #endif
   return 0; 

arch/arm/kvm/Kconfig

@@ -18,6 +18,7 @@ if VIRTUALIZATION
 
 config KVM
 	bool "Kernel-based Virtual Machine (KVM) support"
+	depends on MMU && OF
 	select PREEMPT_NOTIFIERS
 	select ANON_INODES
 	select HAVE_KVM_CPU_RELAX_INTERCEPT
@@ -26,10 +27,12 @@ config KVM
 	select KVM_ARM_HOST
 	select KVM_GENERIC_DIRTYLOG_READ_PROTECT
 	select SRCU
-	depends on ARM_VIRT_EXT && ARM_LPAE
+	select MMU_NOTIFIER
+	select HAVE_KVM_EVENTFD
+	select HAVE_KVM_IRQFD
+	depends on ARM_VIRT_EXT && ARM_LPAE && ARM_ARCH_TIMER
 	---help---
-	  Support hosting virtualized guest machines. You will also
-	  need to select one or more of the processor modules below.
+	  Support hosting virtualized guest machines.
 
 	  This module provides access to the hardware capabilities through
 	  a character device node named /dev/kvm.
@@ -37,10 +40,7 @@ config KVM
 	  If unsure, say N.
 
 config KVM_ARM_HOST
-	bool "KVM host support for ARM cpus."
-	depends on KVM
-	depends on MMU
-	select	MMU_NOTIFIER
+	bool
 	---help---
 	  Provides host support for ARM processors.
 
@@ -55,20 +55,4 @@ config KVM_ARM_MAX_VCPUS
 	  large, so only choose a reasonable number that you expect to
 	  actually use.
 
-config KVM_ARM_VGIC
-	bool "KVM support for Virtual GIC"
-	depends on KVM_ARM_HOST && OF
-	select HAVE_KVM_IRQCHIP
-	default y
-	---help---
-	  Adds support for a hardware assisted, in-kernel GIC emulation.
-
-config KVM_ARM_TIMER
-	bool "KVM support for Architected Timers"
-	depends on KVM_ARM_VGIC && ARM_ARCH_TIMER
-	select HAVE_KVM_IRQCHIP
-	default y
-	---help---
-	  Adds support for the Architected Timers in virtual machines
-
 endif # VIRTUALIZATION

arch/arm/kvm/Makefile

@@ -7,7 +7,7 @@ ifeq ($(plus_virt),+virt)
 	plus_virt_def := -DREQUIRES_VIRT=1
 endif
 
-ccflags-y += -Ivirt/kvm -Iarch/arm/kvm
+ccflags-y += -Iarch/arm/kvm
 CFLAGS_arm.o := -I. $(plus_virt_def)
 CFLAGS_mmu.o := -I.
 
@@ -15,12 +15,12 @@ AFLAGS_init.o := -Wa,-march=armv7-a$(plus_virt)
 AFLAGS_interrupts.o := -Wa,-march=armv7-a$(plus_virt)
 
 KVM := ../../../virt/kvm
-kvm-arm-y = $(KVM)/kvm_main.o $(KVM)/coalesced_mmio.o
+kvm-arm-y = $(KVM)/kvm_main.o $(KVM)/coalesced_mmio.o $(KVM)/eventfd.o
 
 obj-y += kvm-arm.o init.o interrupts.o
 obj-y += arm.o handle_exit.o guest.o mmu.o emulate.o reset.o
 obj-y += coproc.o coproc_a15.o coproc_a7.o mmio.o psci.o perf.o
-obj-$(CONFIG_KVM_ARM_VGIC) += $(KVM)/arm/vgic.o
-obj-$(CONFIG_KVM_ARM_VGIC) += $(KVM)/arm/vgic-v2.o
-obj-$(CONFIG_KVM_ARM_VGIC) += $(KVM)/arm/vgic-v2-emul.o
-obj-$(CONFIG_KVM_ARM_TIMER) += $(KVM)/arm/arch_timer.o
+obj-y += $(KVM)/arm/vgic.o
+obj-y += $(KVM)/arm/vgic-v2.o
+obj-y += $(KVM)/arm/vgic-v2-emul.o
+obj-y += $(KVM)/arm/arch_timer.o

arch/arm/kvm/arm.c

@@ -61,8 +61,6 @@ static atomic64_t kvm_vmid_gen = ATOMIC64_INIT(1);
 static u8 kvm_next_vmid;
 static DEFINE_SPINLOCK(kvm_vmid_lock);
 
-static bool vgic_present;
-
 static void kvm_arm_set_running_vcpu(struct kvm_vcpu *vcpu)
 {
 	BUG_ON(preemptible());
@@ -173,8 +171,8 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	int r;
 	switch (ext) {
 	case KVM_CAP_IRQCHIP:
-		r = vgic_present;
-		break;
+	case KVM_CAP_IRQFD:
+	case KVM_CAP_IOEVENTFD:
 	case KVM_CAP_DEVICE_CTRL:
 	case KVM_CAP_USER_MEMORY:
 	case KVM_CAP_SYNC_MMU:
@@ -183,6 +181,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_ARM_PSCI:
 	case KVM_CAP_ARM_PSCI_0_2:
 	case KVM_CAP_READONLY_MEM:
+	case KVM_CAP_MP_STATE:
 		r = 1;
 		break;
 	case KVM_CAP_COALESCED_MMIO:
@@ -268,7 +267,7 @@ void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
 
 int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
 {
-	return 0;
+	return kvm_timer_should_fire(vcpu);
 }
 
 int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
@@ -313,13 +312,29 @@ int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
 int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
 				    struct kvm_mp_state *mp_state)
 {
-	return -EINVAL;
+	if (vcpu->arch.pause)
+		mp_state->mp_state = KVM_MP_STATE_STOPPED;
+	else
+		mp_state->mp_state = KVM_MP_STATE_RUNNABLE;
+
+	return 0;
 }
 
 int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
 				    struct kvm_mp_state *mp_state)
 {
+	switch (mp_state->mp_state) {
+	case KVM_MP_STATE_RUNNABLE:
+		vcpu->arch.pause = false;
+		break;
+	case KVM_MP_STATE_STOPPED:
+		vcpu->arch.pause = true;
+		break;
+	default:
 		return -EINVAL;
+	}
+
+	return 0;
 }
 
 /**
@@ -452,6 +467,11 @@ static int kvm_vcpu_first_run_init(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
+bool kvm_arch_intc_initialized(struct kvm *kvm)
+{
+	return vgic_initialized(kvm);
+}
+
 static void vcpu_pause(struct kvm_vcpu *vcpu)
 {
 	wait_queue_head_t *wq = kvm_arch_vcpu_wq(vcpu);
@@ -831,8 +851,6 @@ static int kvm_vm_ioctl_set_device_addr(struct kvm *kvm,
 
 	switch (dev_id) {
 	case KVM_ARM_DEVICE_VGIC_V2:
-		if (!vgic_present)
-			return -ENXIO;
 		return kvm_vgic_addr(kvm, type, &dev_addr->addr, true);
 	default:
 		return -ENODEV;
@@ -847,10 +865,7 @@ long kvm_arch_vm_ioctl(struct file *filp,
 
 	switch (ioctl) {
 	case KVM_CREATE_IRQCHIP: {
-		if (vgic_present)
 		return kvm_vgic_create(kvm, KVM_DEV_TYPE_ARM_VGIC_V2);
-		else
-			return -ENXIO;
 	}
 	case KVM_ARM_SET_DEVICE_ADDR: {
 		struct kvm_arm_device_addr dev_addr;
@@ -1035,10 +1050,6 @@ static int init_hyp_mode(void)
 	if (err)
 		goto out_free_context;
 
-#ifdef CONFIG_KVM_ARM_VGIC
-		vgic_present = true;
-#endif
-
 	/*
 	 * Init HYP architected timer support
 	 */

arch/arm/kvm/guest.c

@@ -109,22 +109,6 @@ int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
 	return -EINVAL;
 }
 
-#ifndef CONFIG_KVM_ARM_TIMER
-
-#define NUM_TIMER_REGS 0
-
-static int copy_timer_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
-{
-	return 0;
-}
-
-static bool is_timer_reg(u64 index)
-{
-	return false;
-}
-
-#else
-
 #define NUM_TIMER_REGS 3
 
 static bool is_timer_reg(u64 index)
@@ -152,8 +136,6 @@ static int copy_timer_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
 	return 0;
 }
 
-#endif
-
 static int set_timer_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
 {
 	void __user *uaddr = (void __user *)(long)reg->addr;

arch/arm/kvm/interrupts_head.S

@@ -402,7 +402,6 @@ vcpu	.req	r0		@ vcpu pointer always in r0
  * Assumes vcpu pointer in vcpu reg
  */
 .macro save_vgic_state
-#ifdef CONFIG_KVM_ARM_VGIC
 	/* Get VGIC VCTRL base into r2 */
 	ldr	r2, [vcpu, #VCPU_KVM]
 	ldr	r2, [r2, #KVM_VGIC_VCTRL]
@@ -460,7 +459,6 @@ ARM_BE8(rev	r6, r6	)
 	subs	r4, r4, #1
 	bne	1b
 2:
-#endif
 .endm
 
 /*
@@ -469,7 +467,6 @@ ARM_BE8(rev	r6, r6	)
  * Assumes vcpu pointer in vcpu reg
  */
 .macro restore_vgic_state
-#ifdef CONFIG_KVM_ARM_VGIC
 	/* Get VGIC VCTRL base into r2 */
 	ldr	r2, [vcpu, #VCPU_KVM]
 	ldr	r2, [r2, #KVM_VGIC_VCTRL]
@@ -501,7 +498,6 @@ ARM_BE8(rev	r6, r6  )
 	subs	r4, r4, #1
 	bne	1b
 2:
-#endif
 .endm
 
 #define CNTHCTL_PL1PCTEN	(1 << 0)
@@ -515,7 +511,6 @@ ARM_BE8(rev	r6, r6  )
  * Clobbers r2-r5
  */
 .macro save_timer_state
-#ifdef CONFIG_KVM_ARM_TIMER
 	ldr	r4, [vcpu, #VCPU_KVM]
 	ldr	r2, [r4, #KVM_TIMER_ENABLED]
 	cmp	r2, #0
@@ -537,7 +532,6 @@ ARM_BE8(rev	r6, r6  )
 	mcrr	p15, 4, r2, r2, c14	@ CNTVOFF
 
 1:
-#endif
 	@ Allow physical timer/counter access for the host
 	mrc	p15, 4, r2, c14, c1, 0	@ CNTHCTL
 	orr	r2, r2, #(CNTHCTL_PL1PCEN | CNTHCTL_PL1PCTEN)
@@ -559,7 +553,6 @@ ARM_BE8(rev	r6, r6  )
 	bic	r2, r2, #CNTHCTL_PL1PCEN
 	mcr	p15, 4, r2, c14, c1, 0	@ CNTHCTL
 
-#ifdef CONFIG_KVM_ARM_TIMER
 	ldr	r4, [vcpu, #VCPU_KVM]
 	ldr	r2, [r4, #KVM_TIMER_ENABLED]
 	cmp	r2, #0
@@ -579,7 +572,6 @@ ARM_BE8(rev	r6, r6  )
 	and	r2, r2, #3
 	mcr	p15, 0, r2, c14, c3, 1	@ CNTV_CTL
 1:
-#endif
 .endm
 
 .equ vmentry,	0

arch/arm/kvm/mmio.c

@@ -121,12 +121,11 @@ int kvm_handle_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	return 0;
 }
 
-static int decode_hsr(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
-		      struct kvm_exit_mmio *mmio)
+static int decode_hsr(struct kvm_vcpu *vcpu, bool *is_write, int *len)
 {
 	unsigned long rt;
-	int len;
-	bool is_write, sign_extend;
+	int access_size;
+	bool sign_extend;
 
 	if (kvm_vcpu_dabt_isextabt(vcpu)) {
 		/* cache operation on I/O addr, tell guest unsupported */
@@ -140,17 +139,15 @@ static int decode_hsr(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 		return 1;
 	}
 
-	len = kvm_vcpu_dabt_get_as(vcpu);
-	if (unlikely(len < 0))
-		return len;
+	access_size = kvm_vcpu_dabt_get_as(vcpu);
+	if (unlikely(access_size < 0))
+		return access_size;
 
-	is_write = kvm_vcpu_dabt_iswrite(vcpu);
+	*is_write = kvm_vcpu_dabt_iswrite(vcpu);
 	sign_extend = kvm_vcpu_dabt_issext(vcpu);
 	rt = kvm_vcpu_dabt_get_rd(vcpu);
 
-	mmio->is_write = is_write;
-	mmio->phys_addr = fault_ipa;
-	mmio->len = len;
+	*len = access_size;
 	vcpu->arch.mmio_decode.sign_extend = sign_extend;
 	vcpu->arch.mmio_decode.rt = rt;
 
@@ -165,20 +162,20 @@ static int decode_hsr(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		 phys_addr_t fault_ipa)
 {
-	struct kvm_exit_mmio mmio;
 	unsigned long data;
 	unsigned long rt;
 	int ret;
+	bool is_write;
+	int len;
+	u8 data_buf[8];
 
 	/*
-	 * Prepare MMIO operation. First stash it in a private
-	 * structure that we can use for in-kernel emulation. If the
-	 * kernel can't handle it, copy it into run->mmio and let user
-	 * space do its magic.
+	 * Prepare MMIO operation. First decode the syndrome data we get
+	 * from the CPU. Then try if some in-kernel emulation feels
+	 * responsible, otherwise let user space do its magic.
 	 */
-
 	if (kvm_vcpu_dabt_isvalid(vcpu)) {
-		ret = decode_hsr(vcpu, fault_ipa, &mmio);
+		ret = decode_hsr(vcpu, &is_write, &len);
 		if (ret)
 			return ret;
 	} else {
@@ -188,21 +185,34 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 
 	rt = vcpu->arch.mmio_decode.rt;
 
-	if (mmio.is_write) {
-		data = vcpu_data_guest_to_host(vcpu, *vcpu_reg(vcpu, rt),
-					       mmio.len);
+	if (is_write) {
+		data = vcpu_data_guest_to_host(vcpu, *vcpu_reg(vcpu, rt), len);
+
+		trace_kvm_mmio(KVM_TRACE_MMIO_WRITE, len, fault_ipa, data);
+		mmio_write_buf(data_buf, len, data);
 
-		trace_kvm_mmio(KVM_TRACE_MMIO_WRITE, mmio.len,
-			       fault_ipa, data);
-		mmio_write_buf(mmio.data, mmio.len, data);
+		ret = kvm_io_bus_write(vcpu, KVM_MMIO_BUS, fault_ipa, len,
+				       data_buf);
 	} else {
-		trace_kvm_mmio(KVM_TRACE_MMIO_READ_UNSATISFIED, mmio.len,
+		trace_kvm_mmio(KVM_TRACE_MMIO_READ_UNSATISFIED, len,
 			       fault_ipa, 0);
+
+		ret = kvm_io_bus_read(vcpu, KVM_MMIO_BUS, fault_ipa, len,
+				      data_buf);
 	}
 
-	if (vgic_handle_mmio(vcpu, run, &mmio))
+	/* Now prepare kvm_run for the potential return to userland. */
+	run->mmio.is_write	= is_write;
+	run->mmio.phys_addr	= fault_ipa;
+	run->mmio.len		= len;
+	memcpy(run->mmio.data, data_buf, len);
+
+	if (!ret) {
+		/* We handled the access successfully in the kernel. */
+		kvm_handle_mmio_return(vcpu, run);
 		return 1;
+	}
 
-	kvm_prepare_mmio(run, &mmio);
+	run->exit_reason	= KVM_EXIT_MMIO;
 	return 0;
 }

arch/arm/kvm/mmu.c

@@ -1330,10 +1330,51 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 
 out_unlock:
 	spin_unlock(&kvm->mmu_lock);
+	kvm_set_pfn_accessed(pfn);
 	kvm_release_pfn_clean(pfn);
 	return ret;
 }
 
+/*
+ * Resolve the access fault by making the page young again.
+ * Note that because the faulting entry is guaranteed not to be
+ * cached in the TLB, we don't need to invalidate anything.
+ */
+static void handle_access_fault(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa)
+{
+	pmd_t *pmd;
+	pte_t *pte;
+	pfn_t pfn;
+	bool pfn_valid = false;
+
+	trace_kvm_access_fault(fault_ipa);
+
+	spin_lock(&vcpu->kvm->mmu_lock);
+
+	pmd = stage2_get_pmd(vcpu->kvm, NULL, fault_ipa);
+	if (!pmd || pmd_none(*pmd))	/* Nothing there */
+		goto out;
+
+	if (kvm_pmd_huge(*pmd)) {	/* THP, HugeTLB */
+		*pmd = pmd_mkyoung(*pmd);
+		pfn = pmd_pfn(*pmd);
+		pfn_valid = true;
+		goto out;
+	}
+
+	pte = pte_offset_kernel(pmd, fault_ipa);
+	if (pte_none(*pte))		/* Nothing there either */
+		goto out;
+
+	*pte = pte_mkyoung(*pte);	/* Just a page... */
+	pfn = pte_pfn(*pte);
+	pfn_valid = true;
+out:
+	spin_unlock(&vcpu->kvm->mmu_lock);
+	if (pfn_valid)
+		kvm_set_pfn_accessed(pfn);
+}
+
 /**
  * kvm_handle_guest_abort - handles all 2nd stage aborts
  * @vcpu:	the VCPU pointer
@@ -1364,7 +1405,8 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
 
 	/* Check the stage-2 fault is trans. fault or write fault */
 	fault_status = kvm_vcpu_trap_get_fault_type(vcpu);
-	if (fault_status != FSC_FAULT && fault_status != FSC_PERM) {
+	if (fault_status != FSC_FAULT && fault_status != FSC_PERM &&
+	    fault_status != FSC_ACCESS) {
 		kvm_err("Unsupported FSC: EC=%#x xFSC=%#lx ESR_EL2=%#lx\n",
 			kvm_vcpu_trap_get_class(vcpu),
 			(unsigned long)kvm_vcpu_trap_get_fault(vcpu),
@@ -1400,6 +1442,12 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	/* Userspace should not be able to register out-of-bounds IPAs */
 	VM_BUG_ON(fault_ipa >= KVM_PHYS_SIZE);
 
+	if (fault_status == FSC_ACCESS) {
+		handle_access_fault(vcpu, fault_ipa);
+		ret = 1;
+		goto out_unlock;
+	}
+
 	ret = user_mem_abort(vcpu, fault_ipa, memslot, hva, fault_status);
 	if (ret == 0)
 		ret = 1;
@@ -1408,15 +1456,16 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	return ret;
 }
 
-static void handle_hva_to_gpa(struct kvm *kvm,
+static int handle_hva_to_gpa(struct kvm *kvm,
 			     unsigned long start,
 			     unsigned long end,
-			      void (*handler)(struct kvm *kvm,
+			     int (*handler)(struct kvm *kvm,
 					    gpa_t gpa, void *data),
 			     void *data)
 {
 	struct kvm_memslots *slots;
 	struct kvm_memory_slot *memslot;
+	int ret = 0;
 
 	slots = kvm_memslots(kvm);
 
@@ -1440,14 +1489,17 @@ static void handle_hva_to_gpa(struct kvm *kvm,
 
 		for (; gfn < gfn_end; ++gfn) {
 			gpa_t gpa = gfn << PAGE_SHIFT;
-			handler(kvm, gpa, data);
+			ret |= handler(kvm, gpa, data);
 		}
 	}
+
+	return ret;
 }
 
-static void kvm_unmap_hva_handler(struct kvm *kvm, gpa_t gpa, void *data)
+static int kvm_unmap_hva_handler(struct kvm *kvm, gpa_t gpa, void *data)
 {
 	unmap_stage2_range(kvm, gpa, PAGE_SIZE);
+	return 0;
 }
 
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva)
@@ -1473,7 +1525,7 @@ int kvm_unmap_hva_range(struct kvm *kvm,
 	return 0;
 }
 
-static void kvm_set_spte_handler(struct kvm *kvm, gpa_t gpa, void *data)
+static int kvm_set_spte_handler(struct kvm *kvm, gpa_t gpa, void *data)
 {
 	pte_t *pte = (pte_t *)data;
 
@@ -1485,6 +1537,7 @@ static void kvm_set_spte_handler(struct kvm *kvm, gpa_t gpa, void *data)
 	 * through this calling path.
 	 */
 	stage2_set_pte(kvm, NULL, gpa, pte, 0);
+	return 0;
 }
 
 
@@ -1501,6 +1554,67 @@ void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte)
 	handle_hva_to_gpa(kvm, hva, end, &kvm_set_spte_handler, &stage2_pte);
 }
 
+static int kvm_age_hva_handler(struct kvm *kvm, gpa_t gpa, void *data)
+{
+	pmd_t *pmd;
+	pte_t *pte;
+
+	pmd = stage2_get_pmd(kvm, NULL, gpa);
+	if (!pmd || pmd_none(*pmd))	/* Nothing there */
+		return 0;
+
+	if (kvm_pmd_huge(*pmd)) {	/* THP, HugeTLB */
+		if (pmd_young(*pmd)) {
+			*pmd = pmd_mkold(*pmd);
+			return 1;
+		}
+
+		return 0;
+	}
+
+	pte = pte_offset_kernel(pmd, gpa);
+	if (pte_none(*pte))
+		return 0;
+
+	if (pte_young(*pte)) {
+		*pte = pte_mkold(*pte);	/* Just a page... */
+		return 1;
+	}
+
+	return 0;
+}
+
+static int kvm_test_age_hva_handler(struct kvm *kvm, gpa_t gpa, void *data)
+{
+	pmd_t *pmd;
+	pte_t *pte;
+
+	pmd = stage2_get_pmd(kvm, NULL, gpa);
+	if (!pmd || pmd_none(*pmd))	/* Nothing there */
+		return 0;
+
+	if (kvm_pmd_huge(*pmd))		/* THP, HugeTLB */
+		return pmd_young(*pmd);
+
+	pte = pte_offset_kernel(pmd, gpa);
+	if (!pte_none(*pte))		/* Just a page... */
+		return pte_young(*pte);
+
+	return 0;
+}
+
+int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end)
+{
+	trace_kvm_age_hva(start, end);
+	return handle_hva_to_gpa(kvm, start, end, kvm_age_hva_handler, NULL);
+}
+
+int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
+{
+	trace_kvm_test_age_hva(hva);
+	return handle_hva_to_gpa(kvm, hva, hva, kvm_test_age_hva_handler, NULL);
+}
+
 void kvm_mmu_free_memory_caches(struct kvm_vcpu *vcpu)
 {
 	mmu_free_memory_cache(&vcpu->arch.mmu_page_cache);

arch/arm/kvm/trace.h

@@ -68,6 +68,21 @@ TRACE_EVENT(kvm_guest_fault,
 		  __entry->hxfar, __entry->vcpu_pc)
 );
 
+TRACE_EVENT(kvm_access_fault,
+	TP_PROTO(unsigned long ipa),
+	TP_ARGS(ipa),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	ipa		)
+	),
+
+	TP_fast_assign(
+		__entry->ipa		= ipa;
+	),
+
+	TP_printk("IPA: %lx", __entry->ipa)
+);
+
 TRACE_EVENT(kvm_irq_line,
 	TP_PROTO(unsigned int type, int vcpu_idx, int irq_num, int level),
 	TP_ARGS(type, vcpu_idx, irq_num, level),
@@ -210,6 +225,39 @@ TRACE_EVENT(kvm_set_spte_hva,
 	TP_printk("mmu notifier set pte hva: %#08lx", __entry->hva)
 );
 
+TRACE_EVENT(kvm_age_hva,
+	TP_PROTO(unsigned long start, unsigned long end),
+	TP_ARGS(start, end),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	start		)
+		__field(	unsigned long,	end		)
+	),
+
+	TP_fast_assign(
+		__entry->start		= start;
+		__entry->end		= end;
+	),
+
+	TP_printk("mmu notifier age hva: %#08lx -- %#08lx",
+		  __entry->start, __entry->end)
+);
+
+TRACE_EVENT(kvm_test_age_hva,
+	TP_PROTO(unsigned long hva),
+	TP_ARGS(hva),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	hva		)
+	),
+
+	TP_fast_assign(
+		__entry->hva		= hva;
+	),
+
+	TP_printk("mmu notifier test age hva: %#08lx", __entry->hva)
+);
+
 TRACE_EVENT(kvm_hvc,
 	TP_PROTO(unsigned long vcpu_pc, unsigned long r0, unsigned long imm),
 	TP_ARGS(vcpu_pc, r0, imm),

arch/arm64/include/asm/esr.h

@@ -90,6 +90,7 @@
 #define ESR_ELx_FSC		(0x3F)
 #define ESR_ELx_FSC_TYPE	(0x3C)
 #define ESR_ELx_FSC_EXTABT	(0x10)
+#define ESR_ELx_FSC_ACCESS	(0x08)
 #define ESR_ELx_FSC_FAULT	(0x04)
 #define ESR_ELx_FSC_PERM	(0x0C)
 #define ESR_ELx_CV		(UL(1) << 24)

arch/arm64/include/asm/kvm_arm.h

@@ -188,6 +188,7 @@
 
 /* For compatibility with fault code shared with 32-bit */
 #define FSC_FAULT	ESR_ELx_FSC_FAULT
+#define FSC_ACCESS	ESR_ELx_FSC_ACCESS
 #define FSC_PERM	ESR_ELx_FSC_PERM
 
 /* Hyp Prefetch Fault Address Register (HPFAR/HDFAR) */

arch/arm64/include/asm/kvm_host.h

@@ -28,6 +28,8 @@
 #include <asm/kvm_asm.h>
 #include <asm/kvm_mmio.h>
 
+#define __KVM_HAVE_ARCH_INTC_INITIALIZED
+
 #if defined(CONFIG_KVM_ARM_MAX_VCPUS)
 #define KVM_MAX_VCPUS CONFIG_KVM_ARM_MAX_VCPUS
 #else
@@ -177,19 +179,10 @@ int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
 int kvm_unmap_hva_range(struct kvm *kvm,
 			unsigned long start, unsigned long end);
 void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
+int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end);
+int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
 
 /* We do not have shadow page tables, hence the empty hooks */
-static inline int kvm_age_hva(struct kvm *kvm, unsigned long start,
-			      unsigned long end)
-{
-	return 0;
-}
-
-static inline int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
-{
-	return 0;
-}
-
 static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
 							 unsigned long address)
 {

arch/arm64/include/asm/kvm_mmio.h

@@ -31,28 +31,6 @@ struct kvm_decode {
 	bool sign_extend;
 };
 
-/*
- * The in-kernel MMIO emulation code wants to use a copy of run->mmio,
- * which is an anonymous type. Use our own type instead.
- */
-struct kvm_exit_mmio {
-	phys_addr_t	phys_addr;
-	u8		data[8];
-	u32		len;
-	bool		is_write;
-	void		*private;
-};
-
-static inline void kvm_prepare_mmio(struct kvm_run *run,
-				    struct kvm_exit_mmio *mmio)
-{
-	run->mmio.phys_addr	= mmio->phys_addr;
-	run->mmio.len		= mmio->len;
-	run->mmio.is_write	= mmio->is_write;
-	memcpy(run->mmio.data, mmio->data, mmio->len);
-	run->exit_reason	= KVM_EXIT_MMIO;
-}
-
 int kvm_handle_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run);
 int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		 phys_addr_t fault_ipa);

arch/arm64/include/uapi/asm/kvm.h

@@ -191,6 +191,9 @@ struct kvm_arch_memory_slot {
 /* Highest supported SPI, from VGIC_NR_IRQS */
 #define KVM_ARM_IRQ_GIC_MAX		127
 
+/* One single KVM irqchip, ie. the VGIC */
+#define KVM_NR_IRQCHIPS          1
+
 /* PSCI interface */
 #define KVM_PSCI_FN_BASE		0x95c1ba5e
 #define KVM_PSCI_FN(n)			(KVM_PSCI_FN_BASE + (n))

arch/arm64/kvm/Kconfig

@@ -18,6 +18,7 @@ if VIRTUALIZATION
 
 config KVM
 	bool "Kernel-based Virtual Machine (KVM) support"
+	depends on OF
 	select MMU_NOTIFIER
 	select PREEMPT_NOTIFIERS
 	select ANON_INODES
@@ -25,10 +26,10 @@ config KVM
 	select HAVE_KVM_ARCH_TLB_FLUSH_ALL
 	select KVM_MMIO
 	select KVM_ARM_HOST
-	select KVM_ARM_VGIC
-	select KVM_ARM_TIMER
 	select KVM_GENERIC_DIRTYLOG_READ_PROTECT
 	select SRCU
+	select HAVE_KVM_EVENTFD
+	select HAVE_KVM_IRQFD
 	---help---
 	  Support hosting virtualized guest machines.
 
@@ -50,17 +51,4 @@ config KVM_ARM_MAX_VCPUS
 	  large, so only choose a reasonable number that you expect to
 	  actually use.
 
-config KVM_ARM_VGIC
-	bool
-	depends on KVM_ARM_HOST && OF
-	select HAVE_KVM_IRQCHIP
-	---help---
-	  Adds support for a hardware assisted, in-kernel GIC emulation.
-
-config KVM_ARM_TIMER
-	bool
-	depends on KVM_ARM_VGIC
-	---help---
-	  Adds support for the Architected Timers in virtual machines.
-
 endif # VIRTUALIZATION

arch/arm64/kvm/Makefile

@@ -2,7 +2,7 @@
 # Makefile for Kernel-based Virtual Machine module
 #
 
-ccflags-y += -Ivirt/kvm -Iarch/arm64/kvm
+ccflags-y += -Iarch/arm64/kvm
 CFLAGS_arm.o := -I.
 CFLAGS_mmu.o := -I.
 
@@ -11,7 +11,7 @@ ARM=../../../arch/arm/kvm
 
 obj-$(CONFIG_KVM_ARM_HOST) += kvm.o
 
-kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/kvm_main.o $(KVM)/coalesced_mmio.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/kvm_main.o $(KVM)/coalesced_mmio.o $(KVM)/eventfd.o
 kvm-$(CONFIG_KVM_ARM_HOST) += $(ARM)/arm.o $(ARM)/mmu.o $(ARM)/mmio.o
 kvm-$(CONFIG_KVM_ARM_HOST) += $(ARM)/psci.o $(ARM)/perf.o
 
@@ -19,11 +19,11 @@ kvm-$(CONFIG_KVM_ARM_HOST) += emulate.o inject_fault.o regmap.o
 kvm-$(CONFIG_KVM_ARM_HOST) += hyp.o hyp-init.o handle_exit.o
 kvm-$(CONFIG_KVM_ARM_HOST) += guest.o reset.o sys_regs.o sys_regs_generic_v8.o
 
-kvm-$(CONFIG_KVM_ARM_VGIC) += $(KVM)/arm/vgic.o
-kvm-$(CONFIG_KVM_ARM_VGIC) += $(KVM)/arm/vgic-v2.o
-kvm-$(CONFIG_KVM_ARM_VGIC) += $(KVM)/arm/vgic-v2-emul.o
-kvm-$(CONFIG_KVM_ARM_VGIC) += vgic-v2-switch.o
-kvm-$(CONFIG_KVM_ARM_VGIC) += $(KVM)/arm/vgic-v3.o
-kvm-$(CONFIG_KVM_ARM_VGIC) += $(KVM)/arm/vgic-v3-emul.o
-kvm-$(CONFIG_KVM_ARM_VGIC) += vgic-v3-switch.o
-kvm-$(CONFIG_KVM_ARM_TIMER) += $(KVM)/arm/arch_timer.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic-v2.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic-v2-emul.o
+kvm-$(CONFIG_KVM_ARM_HOST) += vgic-v2-switch.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic-v3.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic-v3-emul.o
+kvm-$(CONFIG_KVM_ARM_HOST) += vgic-v3-switch.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/arch_timer.o

arch/mips/include/asm/asmmacro-32.h

@@ -16,38 +16,38 @@
 	.set push
 	SET_HARDFLOAT
 	cfc1	\tmp,  fcr31
-	swc1	$f0,  THREAD_FPR0_LS64(\thread)
-	swc1	$f1,  THREAD_FPR1_LS64(\thread)
-	swc1	$f2,  THREAD_FPR2_LS64(\thread)
-	swc1	$f3,  THREAD_FPR3_LS64(\thread)
-	swc1	$f4,  THREAD_FPR4_LS64(\thread)
-	swc1	$f5,  THREAD_FPR5_LS64(\thread)
-	swc1	$f6,  THREAD_FPR6_LS64(\thread)
-	swc1	$f7,  THREAD_FPR7_LS64(\thread)
-	swc1	$f8,  THREAD_FPR8_LS64(\thread)
-	swc1	$f9,  THREAD_FPR9_LS64(\thread)
-	swc1	$f10, THREAD_FPR10_LS64(\thread)
-	swc1	$f11, THREAD_FPR11_LS64(\thread)
-	swc1	$f12, THREAD_FPR12_LS64(\thread)
-	swc1	$f13, THREAD_FPR13_LS64(\thread)
-	swc1	$f14, THREAD_FPR14_LS64(\thread)
-	swc1	$f15, THREAD_FPR15_LS64(\thread)
-	swc1	$f16, THREAD_FPR16_LS64(\thread)
-	swc1	$f17, THREAD_FPR17_LS64(\thread)
-	swc1	$f18, THREAD_FPR18_LS64(\thread)
-	swc1	$f19, THREAD_FPR19_LS64(\thread)
-	swc1	$f20, THREAD_FPR20_LS64(\thread)
-	swc1	$f21, THREAD_FPR21_LS64(\thread)
-	swc1	$f22, THREAD_FPR22_LS64(\thread)
-	swc1	$f23, THREAD_FPR23_LS64(\thread)
-	swc1	$f24, THREAD_FPR24_LS64(\thread)
-	swc1	$f25, THREAD_FPR25_LS64(\thread)
-	swc1	$f26, THREAD_FPR26_LS64(\thread)
-	swc1	$f27, THREAD_FPR27_LS64(\thread)
-	swc1	$f28, THREAD_FPR28_LS64(\thread)
-	swc1	$f29, THREAD_FPR29_LS64(\thread)
-	swc1	$f30, THREAD_FPR30_LS64(\thread)
-	swc1	$f31, THREAD_FPR31_LS64(\thread)
+	swc1	$f0,  THREAD_FPR0(\thread)
+	swc1	$f1,  THREAD_FPR1(\thread)
+	swc1	$f2,  THREAD_FPR2(\thread)
+	swc1	$f3,  THREAD_FPR3(\thread)
+	swc1	$f4,  THREAD_FPR4(\thread)
+	swc1	$f5,  THREAD_FPR5(\thread)
+	swc1	$f6,  THREAD_FPR6(\thread)
+	swc1	$f7,  THREAD_FPR7(\thread)
+	swc1	$f8,  THREAD_FPR8(\thread)
+	swc1	$f9,  THREAD_FPR9(\thread)
+	swc1	$f10, THREAD_FPR10(\thread)
+	swc1	$f11, THREAD_FPR11(\thread)
+	swc1	$f12, THREAD_FPR12(\thread)
+	swc1	$f13, THREAD_FPR13(\thread)
+	swc1	$f14, THREAD_FPR14(\thread)
+	swc1	$f15, THREAD_FPR15(\thread)
+	swc1	$f16, THREAD_FPR16(\thread)
+	swc1	$f17, THREAD_FPR17(\thread)
+	swc1	$f18, THREAD_FPR18(\thread)
+	swc1	$f19, THREAD_FPR19(\thread)
+	swc1	$f20, THREAD_FPR20(\thread)
+	swc1	$f21, THREAD_FPR21(\thread)
+	swc1	$f22, THREAD_FPR22(\thread)
+	swc1	$f23, THREAD_FPR23(\thread)
+	swc1	$f24, THREAD_FPR24(\thread)
+	swc1	$f25, THREAD_FPR25(\thread)
+	swc1	$f26, THREAD_FPR26(\thread)
+	swc1	$f27, THREAD_FPR27(\thread)
+	swc1	$f28, THREAD_FPR28(\thread)
+	swc1	$f29, THREAD_FPR29(\thread)
+	swc1	$f30, THREAD_FPR30(\thread)
+	swc1	$f31, THREAD_FPR31(\thread)
 	sw	\tmp, THREAD_FCR31(\thread)
 	.set pop
 	.endm
@@ -56,38 +56,38 @@
 	.set push
 	SET_HARDFLOAT
 	lw	\tmp, THREAD_FCR31(\thread)
-	lwc1	$f0,  THREAD_FPR0_LS64(\thread)
-	lwc1	$f1,  THREAD_FPR1_LS64(\thread)
-	lwc1	$f2,  THREAD_FPR2_LS64(\thread)
-	lwc1	$f3,  THREAD_FPR3_LS64(\thread)
-	lwc1	$f4,  THREAD_FPR4_LS64(\thread)
-	lwc1	$f5,  THREAD_FPR5_LS64(\thread)
-	lwc1	$f6,  THREAD_FPR6_LS64(\thread)
-	lwc1	$f7,  THREAD_FPR7_LS64(\thread)
-	lwc1	$f8,  THREAD_FPR8_LS64(\thread)
-	lwc1	$f9,  THREAD_FPR9_LS64(\thread)
-	lwc1	$f10, THREAD_FPR10_LS64(\thread)
-	lwc1	$f11, THREAD_FPR11_LS64(\thread)
-	lwc1	$f12, THREAD_FPR12_LS64(\thread)
-	lwc1	$f13, THREAD_FPR13_LS64(\thread)
-	lwc1	$f14, THREAD_FPR14_LS64(\thread)
-	lwc1	$f15, THREAD_FPR15_LS64(\thread)
-	lwc1	$f16, THREAD_FPR16_LS64(\thread)
-	lwc1	$f17, THREAD_FPR17_LS64(\thread)
-	lwc1	$f18, THREAD_FPR18_LS64(\thread)
-	lwc1	$f19, THREAD_FPR19_LS64(\thread)
-	lwc1	$f20, THREAD_FPR20_LS64(\thread)
-	lwc1	$f21, THREAD_FPR21_LS64(\thread)
-	lwc1	$f22, THREAD_FPR22_LS64(\thread)
-	lwc1	$f23, THREAD_FPR23_LS64(\thread)
-	lwc1	$f24, THREAD_FPR24_LS64(\thread)
-	lwc1	$f25, THREAD_FPR25_LS64(\thread)
-	lwc1	$f26, THREAD_FPR26_LS64(\thread)
-	lwc1	$f27, THREAD_FPR27_LS64(\thread)
-	lwc1	$f28, THREAD_FPR28_LS64(\thread)
-	lwc1	$f29, THREAD_FPR29_LS64(\thread)
-	lwc1	$f30, THREAD_FPR30_LS64(\thread)
-	lwc1	$f31, THREAD_FPR31_LS64(\thread)
+	lwc1	$f0,  THREAD_FPR0(\thread)
+	lwc1	$f1,  THREAD_FPR1(\thread)
+	lwc1	$f2,  THREAD_FPR2(\thread)
+	lwc1	$f3,  THREAD_FPR3(\thread)
+	lwc1	$f4,  THREAD_FPR4(\thread)
+	lwc1	$f5,  THREAD_FPR5(\thread)
+	lwc1	$f6,  THREAD_FPR6(\thread)
+	lwc1	$f7,  THREAD_FPR7(\thread)
+	lwc1	$f8,  THREAD_FPR8(\thread)
+	lwc1	$f9,  THREAD_FPR9(\thread)
+	lwc1	$f10, THREAD_FPR10(\thread)
+	lwc1	$f11, THREAD_FPR11(\thread)
+	lwc1	$f12, THREAD_FPR12(\thread)
+	lwc1	$f13, THREAD_FPR13(\thread)
+	lwc1	$f14, THREAD_FPR14(\thread)
+	lwc1	$f15, THREAD_FPR15(\thread)
+	lwc1	$f16, THREAD_FPR16(\thread)
+	lwc1	$f17, THREAD_FPR17(\thread)
+	lwc1	$f18, THREAD_FPR18(\thread)
+	lwc1	$f19, THREAD_FPR19(\thread)
+	lwc1	$f20, THREAD_FPR20(\thread)
+	lwc1	$f21, THREAD_FPR21(\thread)
+	lwc1	$f22, THREAD_FPR22(\thread)
+	lwc1	$f23, THREAD_FPR23(\thread)
+	lwc1	$f24, THREAD_FPR24(\thread)
+	lwc1	$f25, THREAD_FPR25(\thread)
+	lwc1	$f26, THREAD_FPR26(\thread)
+	lwc1	$f27, THREAD_FPR27(\thread)
+	lwc1	$f28, THREAD_FPR28(\thread)
+	lwc1	$f29, THREAD_FPR29(\thread)
+	lwc1	$f30, THREAD_FPR30(\thread)
+	lwc1	$f31, THREAD_FPR31(\thread)
 	ctc1	\tmp, fcr31
 	.set pop
 	.endm

arch/mips/include/asm/asmmacro.h

@@ -60,22 +60,22 @@
 	.set	push
 	SET_HARDFLOAT
 	cfc1	\tmp, fcr31
-	sdc1	$f0,  THREAD_FPR0_LS64(\thread)
-	sdc1	$f2,  THREAD_FPR2_LS64(\thread)
-	sdc1	$f4,  THREAD_FPR4_LS64(\thread)
-	sdc1	$f6,  THREAD_FPR6_LS64(\thread)
-	sdc1	$f8,  THREAD_FPR8_LS64(\thread)
-	sdc1	$f10, THREAD_FPR10_LS64(\thread)
-	sdc1	$f12, THREAD_FPR12_LS64(\thread)
-	sdc1	$f14, THREAD_FPR14_LS64(\thread)
-	sdc1	$f16, THREAD_FPR16_LS64(\thread)
-	sdc1	$f18, THREAD_FPR18_LS64(\thread)
-	sdc1	$f20, THREAD_FPR20_LS64(\thread)
-	sdc1	$f22, THREAD_FPR22_LS64(\thread)
-	sdc1	$f24, THREAD_FPR24_LS64(\thread)
-	sdc1	$f26, THREAD_FPR26_LS64(\thread)
-	sdc1	$f28, THREAD_FPR28_LS64(\thread)
-	sdc1	$f30, THREAD_FPR30_LS64(\thread)
+	sdc1	$f0,  THREAD_FPR0(\thread)
+	sdc1	$f2,  THREAD_FPR2(\thread)
+	sdc1	$f4,  THREAD_FPR4(\thread)
+	sdc1	$f6,  THREAD_FPR6(\thread)
+	sdc1	$f8,  THREAD_FPR8(\thread)
+	sdc1	$f10, THREAD_FPR10(\thread)
+	sdc1	$f12, THREAD_FPR12(\thread)
+	sdc1	$f14, THREAD_FPR14(\thread)
+	sdc1	$f16, THREAD_FPR16(\thread)
+	sdc1	$f18, THREAD_FPR18(\thread)
+	sdc1	$f20, THREAD_FPR20(\thread)
+	sdc1	$f22, THREAD_FPR22(\thread)
+	sdc1	$f24, THREAD_FPR24(\thread)
+	sdc1	$f26, THREAD_FPR26(\thread)
+	sdc1	$f28, THREAD_FPR28(\thread)
+	sdc1	$f30, THREAD_FPR30(\thread)
 	sw	\tmp, THREAD_FCR31(\thread)
 	.set	pop
 	.endm
@@ -84,22 +84,22 @@
 	.set	push
 	.set	mips64r2
 	SET_HARDFLOAT
-	sdc1	$f1,  THREAD_FPR1_LS64(\thread)
-	sdc1	$f3,  THREAD_FPR3_LS64(\thread)
-	sdc1	$f5,  THREAD_FPR5_LS64(\thread)
-	sdc1	$f7,  THREAD_FPR7_LS64(\thread)
-	sdc1	$f9,  THREAD_FPR9_LS64(\thread)
-	sdc1	$f11, THREAD_FPR11_LS64(\thread)
-	sdc1	$f13, THREAD_FPR13_LS64(\thread)
-	sdc1	$f15, THREAD_FPR15_LS64(\thread)
-	sdc1	$f17, THREAD_FPR17_LS64(\thread)
-	sdc1	$f19, THREAD_FPR19_LS64(\thread)
-	sdc1	$f21, THREAD_FPR21_LS64(\thread)
-	sdc1	$f23, THREAD_FPR23_LS64(\thread)
-	sdc1	$f25, THREAD_FPR25_LS64(\thread)
-	sdc1	$f27, THREAD_FPR27_LS64(\thread)
-	sdc1	$f29, THREAD_FPR29_LS64(\thread)
-	sdc1	$f31, THREAD_FPR31_LS64(\thread)
+	sdc1	$f1,  THREAD_FPR1(\thread)
+	sdc1	$f3,  THREAD_FPR3(\thread)
+	sdc1	$f5,  THREAD_FPR5(\thread)
+	sdc1	$f7,  THREAD_FPR7(\thread)
+	sdc1	$f9,  THREAD_FPR9(\thread)
+	sdc1	$f11, THREAD_FPR11(\thread)
+	sdc1	$f13, THREAD_FPR13(\thread)
+	sdc1	$f15, THREAD_FPR15(\thread)
+	sdc1	$f17, THREAD_FPR17(\thread)
+	sdc1	$f19, THREAD_FPR19(\thread)
+	sdc1	$f21, THREAD_FPR21(\thread)
+	sdc1	$f23, THREAD_FPR23(\thread)
+	sdc1	$f25, THREAD_FPR25(\thread)
+	sdc1	$f27, THREAD_FPR27(\thread)
+	sdc1	$f29, THREAD_FPR29(\thread)
+	sdc1	$f31, THREAD_FPR31(\thread)
 	.set	pop
 	.endm
 
@@ -118,22 +118,22 @@
 	.set	push
 	SET_HARDFLOAT
 	lw	\tmp, THREAD_FCR31(\thread)
-	ldc1	$f0,  THREAD_FPR0_LS64(\thread)
-	ldc1	$f2,  THREAD_FPR2_LS64(\thread)
-	ldc1	$f4,  THREAD_FPR4_LS64(\thread)
-	ldc1	$f6,  THREAD_FPR6_LS64(\thread)
-	ldc1	$f8,  THREAD_FPR8_LS64(\thread)
-	ldc1	$f10, THREAD_FPR10_LS64(\thread)
-	ldc1	$f12, THREAD_FPR12_LS64(\thread)
-	ldc1	$f14, THREAD_FPR14_LS64(\thread)
-	ldc1	$f16, THREAD_FPR16_LS64(\thread)
-	ldc1	$f18, THREAD_FPR18_LS64(\thread)
-	ldc1	$f20, THREAD_FPR20_LS64(\thread)
-	ldc1	$f22, THREAD_FPR22_LS64(\thread)
-	ldc1	$f24, THREAD_FPR24_LS64(\thread)
-	ldc1	$f26, THREAD_FPR26_LS64(\thread)
-	ldc1	$f28, THREAD_FPR28_LS64(\thread)
-	ldc1	$f30, THREAD_FPR30_LS64(\thread)
+	ldc1	$f0,  THREAD_FPR0(\thread)
+	ldc1	$f2,  THREAD_FPR2(\thread)
+	ldc1	$f4,  THREAD_FPR4(\thread)
+	ldc1	$f6,  THREAD_FPR6(\thread)
+	ldc1	$f8,  THREAD_FPR8(\thread)
+	ldc1	$f10, THREAD_FPR10(\thread)
+	ldc1	$f12, THREAD_FPR12(\thread)
+	ldc1	$f14, THREAD_FPR14(\thread)
+	ldc1	$f16, THREAD_FPR16(\thread)
+	ldc1	$f18, THREAD_FPR18(\thread)
+	ldc1	$f20, THREAD_FPR20(\thread)
+	ldc1	$f22, THREAD_FPR22(\thread)
+	ldc1	$f24, THREAD_FPR24(\thread)
+	ldc1	$f26, THREAD_FPR26(\thread)
+	ldc1	$f28, THREAD_FPR28(\thread)
+	ldc1	$f30, THREAD_FPR30(\thread)
 	ctc1	\tmp, fcr31
 	.endm
 
@@ -141,22 +141,22 @@
 	.set	push
 	.set	mips64r2
 	SET_HARDFLOAT
-	ldc1	$f1,  THREAD_FPR1_LS64(\thread)
-	ldc1	$f3,  THREAD_FPR3_LS64(\thread)
-	ldc1	$f5,  THREAD_FPR5_LS64(\thread)
-	ldc1	$f7,  THREAD_FPR7_LS64(\thread)
-	ldc1	$f9,  THREAD_FPR9_LS64(\thread)
-	ldc1	$f11, THREAD_FPR11_LS64(\thread)
-	ldc1	$f13, THREAD_FPR13_LS64(\thread)
-	ldc1	$f15, THREAD_FPR15_LS64(\thread)
-	ldc1	$f17, THREAD_FPR17_LS64(\thread)
-	ldc1	$f19, THREAD_FPR19_LS64(\thread)
-	ldc1	$f21, THREAD_FPR21_LS64(\thread)
-	ldc1	$f23, THREAD_FPR23_LS64(\thread)
-	ldc1	$f25, THREAD_FPR25_LS64(\thread)
-	ldc1	$f27, THREAD_FPR27_LS64(\thread)
-	ldc1	$f29, THREAD_FPR29_LS64(\thread)
-	ldc1	$f31, THREAD_FPR31_LS64(\thread)
+	ldc1	$f1,  THREAD_FPR1(\thread)
+	ldc1	$f3,  THREAD_FPR3(\thread)
+	ldc1	$f5,  THREAD_FPR5(\thread)
+	ldc1	$f7,  THREAD_FPR7(\thread)
+	ldc1	$f9,  THREAD_FPR9(\thread)
+	ldc1	$f11, THREAD_FPR11(\thread)
+	ldc1	$f13, THREAD_FPR13(\thread)
+	ldc1	$f15, THREAD_FPR15(\thread)
+	ldc1	$f17, THREAD_FPR17(\thread)
+	ldc1	$f19, THREAD_FPR19(\thread)
+	ldc1	$f21, THREAD_FPR21(\thread)
+	ldc1	$f23, THREAD_FPR23(\thread)
+	ldc1	$f25, THREAD_FPR25(\thread)
+	ldc1	$f27, THREAD_FPR27(\thread)
+	ldc1	$f29, THREAD_FPR29(\thread)
+	ldc1	$f31, THREAD_FPR31(\thread)
 	.set	pop
 	.endm
 
@@ -211,6 +211,22 @@
 	.endm
 
 #ifdef TOOLCHAIN_SUPPORTS_MSA
+	.macro	_cfcmsa	rd, cs
+	.set	push
+	.set	mips32r2
+	.set	msa
+	cfcmsa	\rd, $\cs
+	.set	pop
+	.endm
+
+	.macro	_ctcmsa	cd, rs
+	.set	push
+	.set	mips32r2
+	.set	msa
+	ctcmsa	$\cd, \rs
+	.set	pop
+	.endm
+
 	.macro	ld_d	wd, off, base
 	.set	push
 	.set	mips32r2
@@ -227,35 +243,35 @@
 	.set	pop
 	.endm
 
-	.macro	copy_u_w	rd, ws, n
+	.macro	copy_u_w	ws, n
 	.set	push
 	.set	mips32r2
 	.set	msa
-	copy_u.w \rd, $w\ws[\n]
+	copy_u.w $1, $w\ws[\n]
 	.set	pop
 	.endm
 
-	.macro	copy_u_d	rd, ws, n
+	.macro	copy_u_d	ws, n
 	.set	push
 	.set	mips64r2
 	.set	msa
-	copy_u.d \rd, $w\ws[\n]
+	copy_u.d $1, $w\ws[\n]
 	.set	pop
 	.endm
 
-	.macro	insert_w	wd, n, rs
+	.macro	insert_w	wd, n
 	.set	push
 	.set	mips32r2
 	.set	msa
-	insert.w $w\wd[\n], \rs
+	insert.w $w\wd[\n], $1
 	.set	pop
 	.endm
 
-	.macro	insert_d	wd, n, rs
+	.macro	insert_d	wd, n
 	.set	push
 	.set	mips64r2
 	.set	msa
-	insert.d $w\wd[\n], \rs
+	insert.d $w\wd[\n], $1
 	.set	pop
 	.endm
 #else
@@ -283,7 +299,7 @@
 	/*
 	 * Temporary until all toolchains in use include MSA support.
 	 */
-	.macro	cfcmsa	rd, cs
+	.macro	_cfcmsa	rd, cs
 	.set	push
 	.set	noat
 	SET_HARDFLOAT
@@ -293,7 +309,7 @@
 	.set	pop
 	.endm
 
-	.macro	ctcmsa	cd, rs
+	.macro	_ctcmsa	cd, rs
 	.set	push
 	.set	noat
 	SET_HARDFLOAT
@@ -320,44 +336,36 @@
 	.set	pop
 	.endm
 
-	.macro	copy_u_w	rd, ws, n
+	.macro	copy_u_w	ws, n
 	.set	push
 	.set	noat
 	SET_HARDFLOAT
 	.insn
 	.word	COPY_UW_MSA_INSN | (\n << 16) | (\ws << 11)
-	/* move triggers an assembler bug... */
-	or	\rd, $1, zero
 	.set	pop
 	.endm
 
-	.macro	copy_u_d	rd, ws, n
+	.macro	copy_u_d	ws, n
 	.set	push
 	.set	noat
 	SET_HARDFLOAT
 	.insn
 	.word	COPY_UD_MSA_INSN | (\n << 16) | (\ws << 11)
-	/* move triggers an assembler bug... */
-	or	\rd, $1, zero
 	.set	pop
 	.endm
 
-	.macro	insert_w	wd, n, rs
+	.macro	insert_w	wd, n
 	.set	push
 	.set	noat
 	SET_HARDFLOAT
-	/* move triggers an assembler bug... */
-	or	$1, \rs, zero
 	.word	INSERT_W_MSA_INSN | (\n << 16) | (\wd << 6)
 	.set	pop
 	.endm
 
-	.macro	insert_d	wd, n, rs
+	.macro	insert_d	wd, n
 	.set	push
 	.set	noat
 	SET_HARDFLOAT
-	/* move triggers an assembler bug... */
-	or	$1, \rs, zero
 	.word	INSERT_D_MSA_INSN | (\n << 16) | (\wd << 6)
 	.set	pop
 	.endm
@@ -399,7 +407,7 @@
 	.set	push
 	.set	noat
 	SET_HARDFLOAT
-	cfcmsa	$1, MSA_CSR
+	_cfcmsa	$1, MSA_CSR
 	sw	$1, THREAD_MSA_CSR(\thread)
 	.set	pop
 	.endm
@@ -409,7 +417,7 @@
 	.set	noat
 	SET_HARDFLOAT
 	lw	$1, THREAD_MSA_CSR(\thread)
-	ctcmsa	MSA_CSR, $1
+	_ctcmsa	MSA_CSR, $1
 	.set	pop
 	ld_d	0, THREAD_FPR0, \thread
 	ld_d	1, THREAD_FPR1, \thread
@@ -452,9 +460,6 @@
 	insert_w \wd, 2
 	insert_w \wd, 3
 #endif
-	.if	31-\wd
-	msa_init_upper	(\wd+1)
-	.endif
 	.endm
 
 	.macro	msa_init_all_upper
@@ -463,6 +468,37 @@
 	SET_HARDFLOAT
 	not	$1, zero
 	msa_init_upper	0
+	msa_init_upper	1
+	msa_init_upper	2
+	msa_init_upper	3
+	msa_init_upper	4
+	msa_init_upper	5
+	msa_init_upper	6
+	msa_init_upper	7
+	msa_init_upper	8
+	msa_init_upper	9
+	msa_init_upper	10
+	msa_init_upper	11
+	msa_init_upper	12
+	msa_init_upper	13
+	msa_init_upper	14
+	msa_init_upper	15
+	msa_init_upper	16
+	msa_init_upper	17
+	msa_init_upper	18
+	msa_init_upper	19
+	msa_init_upper	20
+	msa_init_upper	21
+	msa_init_upper	22
+	msa_init_upper	23
+	msa_init_upper	24
+	msa_init_upper	25
+	msa_init_upper	26
+	msa_init_upper	27
+	msa_init_upper	28
+	msa_init_upper	29
+	msa_init_upper	30
+	msa_init_upper	31
 	.set	pop
 	.endm
 

arch/mips/include/asm/fpu.h

@@ -48,6 +48,12 @@ enum fpu_mode {
 #define FPU_FR_MASK		0x1
 };
 
+#define __disable_fpu()							\
+do {									\
+	clear_c0_status(ST0_CU1);					\
+	disable_fpu_hazard();						\
+} while (0)
+
 static inline int __enable_fpu(enum fpu_mode mode)
 {
 	int fr;
@@ -86,7 +92,12 @@ static inline int __enable_fpu(enum fpu_mode mode)
 		enable_fpu_hazard();
 
 		/* check FR has the desired value */
-		return (!!(read_c0_status() & ST0_FR) == !!fr) ? 0 : SIGFPE;
+		if (!!(read_c0_status() & ST0_FR) == !!fr)
+			return 0;
+
+		/* unsupported FR value */
+		__disable_fpu();
+		return SIGFPE;
 
 	default:
 		BUG();
@@ -95,12 +106,6 @@ static inline int __enable_fpu(enum fpu_mode mode)
 	return SIGFPE;
 }
 
-#define __disable_fpu()							\
-do {									\
-	clear_c0_status(ST0_CU1);					\
-	disable_fpu_hazard();						\
-} while (0)
-
 #define clear_fpu_owner()	clear_thread_flag(TIF_USEDFPU)
 
 static inline int __is_fpu_owner(void)
@@ -170,6 +175,7 @@ static inline void lose_fpu(int save)
 		}
 		disable_msa();
 		clear_thread_flag(TIF_USEDMSA);
+		__disable_fpu();
 	} else if (is_fpu_owner()) {
 		if (save)
 			_save_fp(current);

arch/mips/include/asm/kdebug.h

@@ -10,7 +10,8 @@ enum die_val {
 	DIE_RI,
 	DIE_PAGE_FAULT,
 	DIE_BREAK,
-	DIE_SSTEPBP
+	DIE_SSTEPBP,
+	DIE_MSAFP
 };
 
 #endif /* _ASM_MIPS_KDEBUG_H */

arch/mips/include/asm/kvm_host.h

@@ -21,10 +21,10 @@
 
 /* MIPS KVM register ids */
 #define MIPS_CP0_32(_R, _S)					\
-	(KVM_REG_MIPS | KVM_REG_SIZE_U32 | 0x10000 | (8 * (_R) + (_S)))
+	(KVM_REG_MIPS_CP0 | KVM_REG_SIZE_U32 | (8 * (_R) + (_S)))
 
 #define MIPS_CP0_64(_R, _S)					\
-	(KVM_REG_MIPS | KVM_REG_SIZE_U64 | 0x10000 | (8 * (_R) + (_S)))
+	(KVM_REG_MIPS_CP0 | KVM_REG_SIZE_U64 | (8 * (_R) + (_S)))
 
 #define KVM_REG_MIPS_CP0_INDEX		MIPS_CP0_32(0, 0)
 #define KVM_REG_MIPS_CP0_ENTRYLO0	MIPS_CP0_64(2, 0)
@@ -42,11 +42,14 @@
 #define KVM_REG_MIPS_CP0_STATUS		MIPS_CP0_32(12, 0)
 #define KVM_REG_MIPS_CP0_CAUSE		MIPS_CP0_32(13, 0)
 #define KVM_REG_MIPS_CP0_EPC		MIPS_CP0_64(14, 0)
+#define KVM_REG_MIPS_CP0_PRID		MIPS_CP0_32(15, 0)
 #define KVM_REG_MIPS_CP0_EBASE		MIPS_CP0_64(15, 1)
 #define KVM_REG_MIPS_CP0_CONFIG		MIPS_CP0_32(16, 0)
 #define KVM_REG_MIPS_CP0_CONFIG1	MIPS_CP0_32(16, 1)
 #define KVM_REG_MIPS_CP0_CONFIG2	MIPS_CP0_32(16, 2)
 #define KVM_REG_MIPS_CP0_CONFIG3	MIPS_CP0_32(16, 3)
+#define KVM_REG_MIPS_CP0_CONFIG4	MIPS_CP0_32(16, 4)
+#define KVM_REG_MIPS_CP0_CONFIG5	MIPS_CP0_32(16, 5)
 #define KVM_REG_MIPS_CP0_CONFIG7	MIPS_CP0_32(16, 7)
 #define KVM_REG_MIPS_CP0_XCONTEXT	MIPS_CP0_64(20, 0)
 #define KVM_REG_MIPS_CP0_ERROREPC	MIPS_CP0_64(30, 0)
@@ -119,6 +122,10 @@ struct kvm_vcpu_stat {
 	u32 syscall_exits;
 	u32 resvd_inst_exits;
 	u32 break_inst_exits;
+	u32 trap_inst_exits;
+	u32 msa_fpe_exits;
+	u32 fpe_exits;
+	u32 msa_disabled_exits;
 	u32 flush_dcache_exits;
 	u32 halt_successful_poll;
 	u32 halt_wakeup;
@@ -138,6 +145,10 @@ enum kvm_mips_exit_types {
 	SYSCALL_EXITS,
 	RESVD_INST_EXITS,
 	BREAK_INST_EXITS,
+	TRAP_INST_EXITS,
+	MSA_FPE_EXITS,
+	FPE_EXITS,
+	MSA_DISABLED_EXITS,
 	FLUSH_DCACHE_EXITS,
 	MAX_KVM_MIPS_EXIT_TYPES
 };
@@ -206,6 +217,8 @@ struct mips_coproc {
 #define MIPS_CP0_CONFIG1_SEL	1
 #define MIPS_CP0_CONFIG2_SEL	2
 #define MIPS_CP0_CONFIG3_SEL	3
+#define MIPS_CP0_CONFIG4_SEL	4
+#define MIPS_CP0_CONFIG5_SEL	5
 
 /* Config0 register bits */
 #define CP0C0_M			31
@@ -262,31 +275,6 @@ struct mips_coproc {
 #define CP0C3_SM		1
 #define CP0C3_TL		0
 
-/* Have config1, Cacheable, noncoherent, write-back, write allocate*/
-#define MIPS_CONFIG0						\
-  ((1 << CP0C0_M) | (0x3 << CP0C0_K0))
-
-/* Have config2, no coprocessor2 attached, no MDMX support attached,
-   no performance counters, watch registers present,
-   no code compression, EJTAG present, no FPU, no watch registers */
-#define MIPS_CONFIG1						\
-((1 << CP0C1_M) |						\
- (0 << CP0C1_C2) | (0 << CP0C1_MD) | (0 << CP0C1_PC) |		\
- (0 << CP0C1_WR) | (0 << CP0C1_CA) | (1 << CP0C1_EP) |		\
- (0 << CP0C1_FP))
-
-/* Have config3, no tertiary/secondary caches implemented */
-#define MIPS_CONFIG2						\
-((1 << CP0C2_M))
-
-/* No config4, no DSP ASE, no large physaddr (PABITS),
-   no external interrupt controller, no vectored interrupts,
-   no 1kb pages, no SmartMIPS ASE, no trace logic */
-#define MIPS_CONFIG3						\
-((0 << CP0C3_M) | (0 << CP0C3_DSPP) | (0 << CP0C3_LPA) |	\
- (0 << CP0C3_VEIC) | (0 << CP0C3_VInt) | (0 << CP0C3_SP) |	\
- (0 << CP0C3_SM) | (0 << CP0C3_TL))
-
 /* MMU types, the first four entries have the same layout as the
    CP0C0_MT field.  */
 enum mips_mmu_types {
@@ -321,7 +309,9 @@ enum mips_mmu_types {
  */
 #define T_TRAP			13	/* Trap instruction */
 #define T_VCEI			14	/* Virtual coherency exception */
+#define T_MSAFPE		14	/* MSA floating point exception */
 #define T_FPE			15	/* Floating point exception */
+#define T_MSADIS		21	/* MSA disabled exception */
 #define T_WATCH			23	/* Watch address reference */
 #define T_VCED			31	/* Virtual coherency data */
 
@@ -374,6 +364,9 @@ struct kvm_mips_tlb {
 	long tlb_lo1;
 };
 
+#define KVM_MIPS_FPU_FPU	0x1
+#define KVM_MIPS_FPU_MSA	0x2
+
 #define KVM_MIPS_GUEST_TLB_SIZE	64
 struct kvm_vcpu_arch {
 	void *host_ebase, *guest_ebase;
@@ -395,6 +388,8 @@ struct kvm_vcpu_arch {
 
 	/* FPU State */
 	struct mips_fpu_struct fpu;
+	/* Which FPU state is loaded (KVM_MIPS_FPU_*) */
+	unsigned int fpu_inuse;
 
 	/* COP0 State */
 	struct mips_coproc *cop0;
@@ -441,6 +436,9 @@ struct kvm_vcpu_arch {
 
 	/* WAIT executed */
 	int wait;
+
+	u8 fpu_enabled;
+	u8 msa_enabled;
 };
 
 
@@ -482,11 +480,15 @@ struct kvm_vcpu_arch {
 #define kvm_read_c0_guest_config1(cop0)		(cop0->reg[MIPS_CP0_CONFIG][1])
 #define kvm_read_c0_guest_config2(cop0)		(cop0->reg[MIPS_CP0_CONFIG][2])
 #define kvm_read_c0_guest_config3(cop0)		(cop0->reg[MIPS_CP0_CONFIG][3])
+#define kvm_read_c0_guest_config4(cop0)		(cop0->reg[MIPS_CP0_CONFIG][4])
+#define kvm_read_c0_guest_config5(cop0)		(cop0->reg[MIPS_CP0_CONFIG][5])
 #define kvm_read_c0_guest_config7(cop0)		(cop0->reg[MIPS_CP0_CONFIG][7])
 #define kvm_write_c0_guest_config(cop0, val)	(cop0->reg[MIPS_CP0_CONFIG][0] = (val))
 #define kvm_write_c0_guest_config1(cop0, val)	(cop0->reg[MIPS_CP0_CONFIG][1] = (val))
 #define kvm_write_c0_guest_config2(cop0, val)	(cop0->reg[MIPS_CP0_CONFIG][2] = (val))
 #define kvm_write_c0_guest_config3(cop0, val)	(cop0->reg[MIPS_CP0_CONFIG][3] = (val))
+#define kvm_write_c0_guest_config4(cop0, val)	(cop0->reg[MIPS_CP0_CONFIG][4] = (val))
+#define kvm_write_c0_guest_config5(cop0, val)	(cop0->reg[MIPS_CP0_CONFIG][5] = (val))
 #define kvm_write_c0_guest_config7(cop0, val)	(cop0->reg[MIPS_CP0_CONFIG][7] = (val))
 #define kvm_read_c0_guest_errorepc(cop0)	(cop0->reg[MIPS_CP0_ERROR_PC][0])
 #define kvm_write_c0_guest_errorepc(cop0, val)	(cop0->reg[MIPS_CP0_ERROR_PC][0] = (val))
@@ -567,6 +569,31 @@ static inline void _kvm_atomic_change_c0_guest_reg(unsigned long *reg,
 	kvm_set_c0_guest_ebase(cop0, ((val) & (change)));		\
 }
 
+/* Helpers */
+
+static inline bool kvm_mips_guest_can_have_fpu(struct kvm_vcpu_arch *vcpu)
+{
+	return (!__builtin_constant_p(cpu_has_fpu) || cpu_has_fpu) &&
+		vcpu->fpu_enabled;
+}
+
+static inline bool kvm_mips_guest_has_fpu(struct kvm_vcpu_arch *vcpu)
+{
+	return kvm_mips_guest_can_have_fpu(vcpu) &&
+		kvm_read_c0_guest_config1(vcpu->cop0) & MIPS_CONF1_FP;
+}
+
+static inline bool kvm_mips_guest_can_have_msa(struct kvm_vcpu_arch *vcpu)
+{
+	return (!__builtin_constant_p(cpu_has_msa) || cpu_has_msa) &&
+		vcpu->msa_enabled;
+}
+
+static inline bool kvm_mips_guest_has_msa(struct kvm_vcpu_arch *vcpu)
+{
+	return kvm_mips_guest_can_have_msa(vcpu) &&
+		kvm_read_c0_guest_config3(vcpu->cop0) & MIPS_CONF3_MSA;
+}
 
 struct kvm_mips_callbacks {
 	int (*handle_cop_unusable)(struct kvm_vcpu *vcpu);
@@ -578,6 +605,10 @@ struct kvm_mips_callbacks {
 	int (*handle_syscall)(struct kvm_vcpu *vcpu);
 	int (*handle_res_inst)(struct kvm_vcpu *vcpu);
 	int (*handle_break)(struct kvm_vcpu *vcpu);
+	int (*handle_trap)(struct kvm_vcpu *vcpu);
+	int (*handle_msa_fpe)(struct kvm_vcpu *vcpu);
+	int (*handle_fpe)(struct kvm_vcpu *vcpu);
+	int (*handle_msa_disabled)(struct kvm_vcpu *vcpu);
 	int (*vm_init)(struct kvm *kvm);
 	int (*vcpu_init)(struct kvm_vcpu *vcpu);
 	int (*vcpu_setup)(struct kvm_vcpu *vcpu);
@@ -596,6 +627,8 @@ struct kvm_mips_callbacks {
 			   const struct kvm_one_reg *reg, s64 *v);
 	int (*set_one_reg)(struct kvm_vcpu *vcpu,
 			   const struct kvm_one_reg *reg, s64 v);
+	int (*vcpu_get_regs)(struct kvm_vcpu *vcpu);
+	int (*vcpu_set_regs)(struct kvm_vcpu *vcpu);
 };
 extern struct kvm_mips_callbacks *kvm_mips_callbacks;
 int kvm_mips_emulation_init(struct kvm_mips_callbacks **install_callbacks);
@@ -606,6 +639,19 @@ int kvm_arch_vcpu_dump_regs(struct kvm_vcpu *vcpu);
 /* Trampoline ASM routine to start running in "Guest" context */
 extern int __kvm_mips_vcpu_run(struct kvm_run *run, struct kvm_vcpu *vcpu);
 
+/* FPU/MSA context management */
+void __kvm_save_fpu(struct kvm_vcpu_arch *vcpu);
+void __kvm_restore_fpu(struct kvm_vcpu_arch *vcpu);
+void __kvm_restore_fcsr(struct kvm_vcpu_arch *vcpu);
+void __kvm_save_msa(struct kvm_vcpu_arch *vcpu);
+void __kvm_restore_msa(struct kvm_vcpu_arch *vcpu);
+void __kvm_restore_msa_upper(struct kvm_vcpu_arch *vcpu);
+void __kvm_restore_msacsr(struct kvm_vcpu_arch *vcpu);
+void kvm_own_fpu(struct kvm_vcpu *vcpu);
+void kvm_own_msa(struct kvm_vcpu *vcpu);
+void kvm_drop_fpu(struct kvm_vcpu *vcpu);
+void kvm_lose_fpu(struct kvm_vcpu *vcpu);
+
 /* TLB handling */
 uint32_t kvm_get_kernel_asid(struct kvm_vcpu *vcpu);
 
@@ -711,6 +757,26 @@ extern enum emulation_result kvm_mips_emulate_bp_exc(unsigned long cause,
 						     struct kvm_run *run,
 						     struct kvm_vcpu *vcpu);
 
+extern enum emulation_result kvm_mips_emulate_trap_exc(unsigned long cause,
+						       uint32_t *opc,
+						       struct kvm_run *run,
+						       struct kvm_vcpu *vcpu);
+
+extern enum emulation_result kvm_mips_emulate_msafpe_exc(unsigned long cause,
+							 uint32_t *opc,
+							 struct kvm_run *run,
+							 struct kvm_vcpu *vcpu);
+
+extern enum emulation_result kvm_mips_emulate_fpe_exc(unsigned long cause,
+						      uint32_t *opc,
+						      struct kvm_run *run,
+						      struct kvm_vcpu *vcpu);
+
+extern enum emulation_result kvm_mips_emulate_msadis_exc(unsigned long cause,
+							 uint32_t *opc,
+							 struct kvm_run *run,
+							 struct kvm_vcpu *vcpu);
+
 extern enum emulation_result kvm_mips_complete_mmio_load(struct kvm_vcpu *vcpu,
 							 struct kvm_run *run);
 
@@ -749,6 +815,11 @@ enum emulation_result kvm_mips_emulate_load(uint32_t inst,
 					    struct kvm_run *run,
 					    struct kvm_vcpu *vcpu);
 
+unsigned int kvm_mips_config1_wrmask(struct kvm_vcpu *vcpu);
+unsigned int kvm_mips_config3_wrmask(struct kvm_vcpu *vcpu);
+unsigned int kvm_mips_config4_wrmask(struct kvm_vcpu *vcpu);
+unsigned int kvm_mips_config5_wrmask(struct kvm_vcpu *vcpu);
+
 /* Dynamic binary translation */
 extern int kvm_mips_trans_cache_index(uint32_t inst, uint32_t *opc,
 				      struct kvm_vcpu *vcpu);

arch/mips/include/asm/processor.h

@@ -105,7 +105,7 @@ union fpureg {
 #ifdef CONFIG_CPU_LITTLE_ENDIAN
 # define FPR_IDX(width, idx)	(idx)
 #else
-# define FPR_IDX(width, idx)	((FPU_REG_WIDTH / (width)) - 1 - (idx))
+# define FPR_IDX(width, idx)	((idx) ^ ((64 / (width)) - 1))
 #endif
 
 #define BUILD_FPR_ACCESS(width) \

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

@@ -36,77 +36,85 @@ struct kvm_regs {
 
 /*
  * for KVM_GET_FPU and KVM_SET_FPU
- *
- * If Status[FR] is zero (32-bit FPU), the upper 32-bits of the FPRs
- * are zero filled.
  */
 struct kvm_fpu {
-	__u64 fpr[32];
-	__u32 fir;
-	__u32 fccr;
-	__u32 fexr;
-	__u32 fenr;
-	__u32 fcsr;
-	__u32 pad;
 };
 
 
 /*
- * For MIPS, we use KVM_SET_ONE_REG and KVM_GET_ONE_REG to access CP0
+ * For MIPS, we use KVM_SET_ONE_REG and KVM_GET_ONE_REG to access various
  * registers.  The id field is broken down as follows:
  *
- *  bits[2..0]   - Register 'sel' index.
- *  bits[7..3]   - Register 'rd'  index.
- *  bits[15..8]  - Must be zero.
- *  bits[31..16] - 1 -> CP0 registers.
- *  bits[51..32] - Must be zero.
  *  bits[63..52] - As per linux/kvm.h
+ *  bits[51..32] - Must be zero.
+ *  bits[31..16] - Register set.
+ *
+ * Register set = 0: GP registers from kvm_regs (see definitions below).
+ *
+ * Register set = 1: CP0 registers.
+ *  bits[15..8]  - Must be zero.
+ *  bits[7..3]   - Register 'rd'  index.
+ *  bits[2..0]   - Register 'sel' index.
+ *
+ * Register set = 2: KVM specific registers (see definitions below).
+ *
+ * Register set = 3: FPU / MSA registers (see definitions below).
  *
  * Other sets registers may be added in the future.  Each set would
  * have its own identifier in bits[31..16].
- *
- * The registers defined in struct kvm_regs are also accessible, the
- * id values for these are below.
  */
 
-#define KVM_REG_MIPS_R0 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 0)
-#define KVM_REG_MIPS_R1 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 1)
-#define KVM_REG_MIPS_R2 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 2)
-#define KVM_REG_MIPS_R3 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 3)
-#define KVM_REG_MIPS_R4 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 4)
-#define KVM_REG_MIPS_R5 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 5)
-#define KVM_REG_MIPS_R6 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 6)
-#define KVM_REG_MIPS_R7 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 7)
-#define KVM_REG_MIPS_R8 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 8)
-#define KVM_REG_MIPS_R9 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 9)
-#define KVM_REG_MIPS_R10 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 10)
-#define KVM_REG_MIPS_R11 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 11)
-#define KVM_REG_MIPS_R12 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 12)
-#define KVM_REG_MIPS_R13 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 13)
-#define KVM_REG_MIPS_R14 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 14)
-#define KVM_REG_MIPS_R15 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 15)
-#define KVM_REG_MIPS_R16 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 16)
-#define KVM_REG_MIPS_R17 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 17)
-#define KVM_REG_MIPS_R18 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 18)
-#define KVM_REG_MIPS_R19 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 19)
-#define KVM_REG_MIPS_R20 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 20)
-#define KVM_REG_MIPS_R21 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 21)
-#define KVM_REG_MIPS_R22 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 22)
-#define KVM_REG_MIPS_R23 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 23)
-#define KVM_REG_MIPS_R24 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 24)
-#define KVM_REG_MIPS_R25 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 25)
-#define KVM_REG_MIPS_R26 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 26)
-#define KVM_REG_MIPS_R27 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 27)
-#define KVM_REG_MIPS_R28 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 28)
-#define KVM_REG_MIPS_R29 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 29)
-#define KVM_REG_MIPS_R30 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 30)
-#define KVM_REG_MIPS_R31 (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 31)
-
-#define KVM_REG_MIPS_HI (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 32)
-#define KVM_REG_MIPS_LO (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 33)
-#define KVM_REG_MIPS_PC (KVM_REG_MIPS | KVM_REG_SIZE_U64 | 34)
-
-/* KVM specific control registers */
+#define KVM_REG_MIPS_GP		(KVM_REG_MIPS | 0x0000000000000000ULL)
+#define KVM_REG_MIPS_CP0	(KVM_REG_MIPS | 0x0000000000010000ULL)
+#define KVM_REG_MIPS_KVM	(KVM_REG_MIPS | 0x0000000000020000ULL)
+#define KVM_REG_MIPS_FPU	(KVM_REG_MIPS | 0x0000000000030000ULL)
+
+
+/*
+ * KVM_REG_MIPS_GP - General purpose registers from kvm_regs.
+ */
+
+#define KVM_REG_MIPS_R0		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 |  0)
+#define KVM_REG_MIPS_R1		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 |  1)
+#define KVM_REG_MIPS_R2		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 |  2)
+#define KVM_REG_MIPS_R3		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 |  3)
+#define KVM_REG_MIPS_R4		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 |  4)
+#define KVM_REG_MIPS_R5		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 |  5)
+#define KVM_REG_MIPS_R6		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 |  6)
+#define KVM_REG_MIPS_R7		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 |  7)
+#define KVM_REG_MIPS_R8		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 |  8)
+#define KVM_REG_MIPS_R9		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 |  9)
+#define KVM_REG_MIPS_R10	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 10)
+#define KVM_REG_MIPS_R11	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 11)
+#define KVM_REG_MIPS_R12	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 12)
+#define KVM_REG_MIPS_R13	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 13)
+#define KVM_REG_MIPS_R14	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 14)
+#define KVM_REG_MIPS_R15	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 15)
+#define KVM_REG_MIPS_R16	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 16)
+#define KVM_REG_MIPS_R17	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 17)
+#define KVM_REG_MIPS_R18	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 18)
+#define KVM_REG_MIPS_R19	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 19)
+#define KVM_REG_MIPS_R20	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 20)
+#define KVM_REG_MIPS_R21	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 21)
+#define KVM_REG_MIPS_R22	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 22)
+#define KVM_REG_MIPS_R23	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 23)
+#define KVM_REG_MIPS_R24	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 24)
+#define KVM_REG_MIPS_R25	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 25)
+#define KVM_REG_MIPS_R26	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 26)
+#define KVM_REG_MIPS_R27	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 27)
+#define KVM_REG_MIPS_R28	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 28)
+#define KVM_REG_MIPS_R29	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 29)
+#define KVM_REG_MIPS_R30	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 30)
+#define KVM_REG_MIPS_R31	(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 31)
+
+#define KVM_REG_MIPS_HI		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 32)
+#define KVM_REG_MIPS_LO		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 33)
+#define KVM_REG_MIPS_PC		(KVM_REG_MIPS_GP | KVM_REG_SIZE_U64 | 34)
+
+
+/*
+ * KVM_REG_MIPS_KVM - KVM specific control registers.
+ */
 
 /*
  * CP0_Count control
@@ -118,8 +126,7 @@ struct kvm_fpu {
  *        safely without losing time or guest timer interrupts.
  * Other: Reserved, do not change.
  */
-#define KVM_REG_MIPS_COUNT_CTL		(KVM_REG_MIPS | KVM_REG_SIZE_U64 | \
-					 0x20000 | 0)
+#define KVM_REG_MIPS_COUNT_CTL	    (KVM_REG_MIPS_KVM | KVM_REG_SIZE_U64 | 0)
 #define KVM_REG_MIPS_COUNT_CTL_DC	0x00000001
 
 /*
@@ -131,15 +138,46 @@ struct kvm_fpu {
  * emulated.
  * Modifications to times in the future are rejected.
  */
-#define KVM_REG_MIPS_COUNT_RESUME	(KVM_REG_MIPS | KVM_REG_SIZE_U64 | \
-					 0x20000 | 1)
+#define KVM_REG_MIPS_COUNT_RESUME   (KVM_REG_MIPS_KVM | KVM_REG_SIZE_U64 | 1)
 /*
  * CP0_Count rate in Hz
  * Specifies the rate of the CP0_Count timer in Hz. Modifications occur without
  * discontinuities in CP0_Count.
  */
-#define KVM_REG_MIPS_COUNT_HZ		(KVM_REG_MIPS | KVM_REG_SIZE_U64 | \
-					 0x20000 | 2)
+#define KVM_REG_MIPS_COUNT_HZ	    (KVM_REG_MIPS_KVM | KVM_REG_SIZE_U64 | 2)
+
+
+/*
+ * KVM_REG_MIPS_FPU - Floating Point and MIPS SIMD Architecture (MSA) registers.
+ *
+ *  bits[15..8]  - Register subset (see definitions below).
+ *  bits[7..5]   - Must be zero.
+ *  bits[4..0]   - Register number within register subset.
+ */
+
+#define KVM_REG_MIPS_FPR	(KVM_REG_MIPS_FPU | 0x0000000000000000ULL)
+#define KVM_REG_MIPS_FCR	(KVM_REG_MIPS_FPU | 0x0000000000000100ULL)
+#define KVM_REG_MIPS_MSACR	(KVM_REG_MIPS_FPU | 0x0000000000000200ULL)
+
+/*
+ * KVM_REG_MIPS_FPR - Floating point / Vector registers.
+ */
+#define KVM_REG_MIPS_FPR_32(n)	(KVM_REG_MIPS_FPR | KVM_REG_SIZE_U32  | (n))
+#define KVM_REG_MIPS_FPR_64(n)	(KVM_REG_MIPS_FPR | KVM_REG_SIZE_U64  | (n))
+#define KVM_REG_MIPS_VEC_128(n)	(KVM_REG_MIPS_FPR | KVM_REG_SIZE_U128 | (n))
+
+/*
+ * KVM_REG_MIPS_FCR - Floating point control registers.
+ */
+#define KVM_REG_MIPS_FCR_IR	(KVM_REG_MIPS_FCR | KVM_REG_SIZE_U32 |  0)
+#define KVM_REG_MIPS_FCR_CSR	(KVM_REG_MIPS_FCR | KVM_REG_SIZE_U32 | 31)
+
+/*
+ * KVM_REG_MIPS_MSACR - MIPS SIMD Architecture (MSA) control registers.
+ */
+#define KVM_REG_MIPS_MSA_IR	 (KVM_REG_MIPS_MSACR | KVM_REG_SIZE_U32 |  0)
+#define KVM_REG_MIPS_MSA_CSR	 (KVM_REG_MIPS_MSACR | KVM_REG_SIZE_U32 |  1)
+
 
 /*
  * KVM MIPS specific structures and definitions

arch/mips/kernel/asm-offsets.c

@@ -167,72 +167,6 @@ void output_thread_fpu_defines(void)
 	OFFSET(THREAD_FPR30, task_struct, thread.fpu.fpr[30]);
 	OFFSET(THREAD_FPR31, task_struct, thread.fpu.fpr[31]);
 
-	/* the least significant 64 bits of each FP register */
-	OFFSET(THREAD_FPR0_LS64, task_struct,
-	       thread.fpu.fpr[0].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR1_LS64, task_struct,
-	       thread.fpu.fpr[1].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR2_LS64, task_struct,
-	       thread.fpu.fpr[2].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR3_LS64, task_struct,
-	       thread.fpu.fpr[3].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR4_LS64, task_struct,
-	       thread.fpu.fpr[4].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR5_LS64, task_struct,
-	       thread.fpu.fpr[5].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR6_LS64, task_struct,
-	       thread.fpu.fpr[6].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR7_LS64, task_struct,
-	       thread.fpu.fpr[7].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR8_LS64, task_struct,
-	       thread.fpu.fpr[8].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR9_LS64, task_struct,
-	       thread.fpu.fpr[9].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR10_LS64, task_struct,
-	       thread.fpu.fpr[10].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR11_LS64, task_struct,
-	       thread.fpu.fpr[11].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR12_LS64, task_struct,
-	       thread.fpu.fpr[12].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR13_LS64, task_struct,
-	       thread.fpu.fpr[13].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR14_LS64, task_struct,
-	       thread.fpu.fpr[14].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR15_LS64, task_struct,
-	       thread.fpu.fpr[15].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR16_LS64, task_struct,
-	       thread.fpu.fpr[16].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR17_LS64, task_struct,
-	       thread.fpu.fpr[17].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR18_LS64, task_struct,
-	       thread.fpu.fpr[18].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR19_LS64, task_struct,
-	       thread.fpu.fpr[19].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR20_LS64, task_struct,
-	       thread.fpu.fpr[20].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR21_LS64, task_struct,
-	       thread.fpu.fpr[21].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR22_LS64, task_struct,
-	       thread.fpu.fpr[22].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR23_LS64, task_struct,
-	       thread.fpu.fpr[23].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR24_LS64, task_struct,
-	       thread.fpu.fpr[24].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR25_LS64, task_struct,
-	       thread.fpu.fpr[25].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR26_LS64, task_struct,
-	       thread.fpu.fpr[26].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR27_LS64, task_struct,
-	       thread.fpu.fpr[27].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR28_LS64, task_struct,
-	       thread.fpu.fpr[28].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR29_LS64, task_struct,
-	       thread.fpu.fpr[29].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR30_LS64, task_struct,
-	       thread.fpu.fpr[30].val64[FPR_IDX(64, 0)]);
-	OFFSET(THREAD_FPR31_LS64, task_struct,
-	       thread.fpu.fpr[31].val64[FPR_IDX(64, 0)]);
-
 	OFFSET(THREAD_FCR31, task_struct, thread.fpu.fcr31);
 	OFFSET(THREAD_MSA_CSR, task_struct, thread.fpu.msacsr);
 	BLANK();
@@ -470,6 +404,45 @@ void output_kvm_defines(void)
 	OFFSET(VCPU_LO, kvm_vcpu_arch, lo);
 	OFFSET(VCPU_HI, kvm_vcpu_arch, hi);
 	OFFSET(VCPU_PC, kvm_vcpu_arch, pc);
+	BLANK();
+
+	OFFSET(VCPU_FPR0, kvm_vcpu_arch, fpu.fpr[0]);
+	OFFSET(VCPU_FPR1, kvm_vcpu_arch, fpu.fpr[1]);
+	OFFSET(VCPU_FPR2, kvm_vcpu_arch, fpu.fpr[2]);
+	OFFSET(VCPU_FPR3, kvm_vcpu_arch, fpu.fpr[3]);
+	OFFSET(VCPU_FPR4, kvm_vcpu_arch, fpu.fpr[4]);
+	OFFSET(VCPU_FPR5, kvm_vcpu_arch, fpu.fpr[5]);
+	OFFSET(VCPU_FPR6, kvm_vcpu_arch, fpu.fpr[6]);
+	OFFSET(VCPU_FPR7, kvm_vcpu_arch, fpu.fpr[7]);
+	OFFSET(VCPU_FPR8, kvm_vcpu_arch, fpu.fpr[8]);
+	OFFSET(VCPU_FPR9, kvm_vcpu_arch, fpu.fpr[9]);
+	OFFSET(VCPU_FPR10, kvm_vcpu_arch, fpu.fpr[10]);
+	OFFSET(VCPU_FPR11, kvm_vcpu_arch, fpu.fpr[11]);
+	OFFSET(VCPU_FPR12, kvm_vcpu_arch, fpu.fpr[12]);
+	OFFSET(VCPU_FPR13, kvm_vcpu_arch, fpu.fpr[13]);
+	OFFSET(VCPU_FPR14, kvm_vcpu_arch, fpu.fpr[14]);
+	OFFSET(VCPU_FPR15, kvm_vcpu_arch, fpu.fpr[15]);
+	OFFSET(VCPU_FPR16, kvm_vcpu_arch, fpu.fpr[16]);
+	OFFSET(VCPU_FPR17, kvm_vcpu_arch, fpu.fpr[17]);
+	OFFSET(VCPU_FPR18, kvm_vcpu_arch, fpu.fpr[18]);
+	OFFSET(VCPU_FPR19, kvm_vcpu_arch, fpu.fpr[19]);
+	OFFSET(VCPU_FPR20, kvm_vcpu_arch, fpu.fpr[20]);
+	OFFSET(VCPU_FPR21, kvm_vcpu_arch, fpu.fpr[21]);
+	OFFSET(VCPU_FPR22, kvm_vcpu_arch, fpu.fpr[22]);
+	OFFSET(VCPU_FPR23, kvm_vcpu_arch, fpu.fpr[23]);
+	OFFSET(VCPU_FPR24, kvm_vcpu_arch, fpu.fpr[24]);
+	OFFSET(VCPU_FPR25, kvm_vcpu_arch, fpu.fpr[25]);
+	OFFSET(VCPU_FPR26, kvm_vcpu_arch, fpu.fpr[26]);
+	OFFSET(VCPU_FPR27, kvm_vcpu_arch, fpu.fpr[27]);
+	OFFSET(VCPU_FPR28, kvm_vcpu_arch, fpu.fpr[28]);
+	OFFSET(VCPU_FPR29, kvm_vcpu_arch, fpu.fpr[29]);
+	OFFSET(VCPU_FPR30, kvm_vcpu_arch, fpu.fpr[30]);
+	OFFSET(VCPU_FPR31, kvm_vcpu_arch, fpu.fpr[31]);
+
+	OFFSET(VCPU_FCR31, kvm_vcpu_arch, fpu.fcr31);
+	OFFSET(VCPU_MSA_CSR, kvm_vcpu_arch, fpu.msacsr);
+	BLANK();
+
 	OFFSET(VCPU_COP0, kvm_vcpu_arch, cop0);
 	OFFSET(VCPU_GUEST_KERNEL_ASID, kvm_vcpu_arch, guest_kernel_asid);
 	OFFSET(VCPU_GUEST_USER_ASID, kvm_vcpu_arch, guest_user_asid);

arch/mips/kernel/genex.S

@@ -360,12 +360,15 @@ NESTED(nmi_handler, PT_SIZE, sp)
 	.set	mips1
 	SET_HARDFLOAT
 	cfc1	a1, fcr31
-	li	a2, ~(0x3f << 12)
-	and	a2, a1
-	ctc1	a2, fcr31
 	.set	pop
-	TRACE_IRQS_ON
-	STI
+	CLI
+	TRACE_IRQS_OFF
+	.endm
+
+	.macro	__build_clear_msa_fpe
+	_cfcmsa	a1, MSA_CSR
+	CLI
+	TRACE_IRQS_OFF
 	.endm
 
 	.macro	__build_clear_ade
@@ -426,7 +429,7 @@ NESTED(nmi_handler, PT_SIZE, sp)
 	BUILD_HANDLER cpu cpu sti silent		/* #11 */
 	BUILD_HANDLER ov ov sti silent			/* #12 */
 	BUILD_HANDLER tr tr sti silent			/* #13 */
-	BUILD_HANDLER msa_fpe msa_fpe sti silent	/* #14 */
+	BUILD_HANDLER msa_fpe msa_fpe msa_fpe silent	/* #14 */
 	BUILD_HANDLER fpe fpe fpe silent		/* #15 */
 	BUILD_HANDLER ftlb ftlb none silent		/* #16 */
 	BUILD_HANDLER msa msa sti silent		/* #21 */

arch/mips/kernel/ptrace.c

@@ -46,6 +46,26 @@
 #define CREATE_TRACE_POINTS
 #include <trace/events/syscalls.h>
 
+static void init_fp_ctx(struct task_struct *target)
+{
+	/* If FP has been used then the target already has context */
+	if (tsk_used_math(target))
+		return;
+
+	/* Begin with data registers set to all 1s... */
+	memset(&target->thread.fpu.fpr, ~0, sizeof(target->thread.fpu.fpr));
+
+	/* ...and FCSR zeroed */
+	target->thread.fpu.fcr31 = 0;
+
+	/*
+	 * Record that the target has "used" math, such that the context
+	 * just initialised, and any modifications made by the caller,
+	 * aren't discarded.
+	 */
+	set_stopped_child_used_math(target);
+}
+
 /*
  * Called by kernel/ptrace.c when detaching..
  *
@@ -142,6 +162,7 @@ int ptrace_setfpregs(struct task_struct *child, __u32 __user *data)
 	if (!access_ok(VERIFY_READ, data, 33 * 8))
 		return -EIO;
 
+	init_fp_ctx(child);
 	fregs = get_fpu_regs(child);
 
 	for (i = 0; i < 32; i++) {
@@ -439,6 +460,8 @@ static int fpr_set(struct task_struct *target,
 
 	/* XXX fcr31  */
 
+	init_fp_ctx(target);
+
 	if (sizeof(target->thread.fpu.fpr[i]) == sizeof(elf_fpreg_t))
 		return user_regset_copyin(&pos, &count, &kbuf, &ubuf,
 					  &target->thread.fpu,
@@ -660,12 +683,7 @@ long arch_ptrace(struct task_struct *child, long request,
 		case FPR_BASE ... FPR_BASE + 31: {
 			union fpureg *fregs = get_fpu_regs(child);
 
-			if (!tsk_used_math(child)) {
-				/* FP not yet used  */
-				memset(&child->thread.fpu, ~0,
-				       sizeof(child->thread.fpu));
-				child->thread.fpu.fcr31 = 0;
-			}
+			init_fp_ctx(child);
 #ifdef CONFIG_32BIT
 			if (test_thread_flag(TIF_32BIT_FPREGS)) {
 				/*

arch/mips/kernel/r4k_fpu.S

@@ -34,7 +34,6 @@
 	.endm
 
 	.set	noreorder
-	.set	MIPS_ISA_ARCH_LEVEL_RAW
 
 LEAF(_save_fp_context)
 	.set	push
@@ -103,6 +102,7 @@ LEAF(_save_fp_context)
 	/* Save 32-bit process floating point context */
 LEAF(_save_fp_context32)
 	.set push
+	.set MIPS_ISA_ARCH_LEVEL_RAW
 	SET_HARDFLOAT
 	cfc1	t1, fcr31
 

arch/mips/kernel/traps.c

@@ -701,6 +701,13 @@ asmlinkage void do_ov(struct pt_regs *regs)
 
 int process_fpemu_return(int sig, void __user *fault_addr)
 {
+	/*
+	 * We can't allow the emulated instruction to leave any of the cause
+	 * bits set in FCSR. If they were then the kernel would take an FP
+	 * exception when restoring FP context.
+	 */
+	current->thread.fpu.fcr31 &= ~FPU_CSR_ALL_X;
+
 	if (sig == SIGSEGV || sig == SIGBUS) {
 		struct siginfo si = {0};
 		si.si_addr = fault_addr;
@@ -781,6 +788,11 @@ asmlinkage void do_fpe(struct pt_regs *regs, unsigned long fcr31)
 	if (notify_die(DIE_FP, "FP exception", regs, 0, regs_to_trapnr(regs),
 		       SIGFPE) == NOTIFY_STOP)
 		goto out;
+
+	/* Clear FCSR.Cause before enabling interrupts */
+	write_32bit_cp1_register(CP1_STATUS, fcr31 & ~FPU_CSR_ALL_X);
+	local_irq_enable();
+
 	die_if_kernel("FP exception in kernel code", regs);
 
 	if (fcr31 & FPU_CSR_UNI_X) {
@@ -804,18 +816,12 @@ asmlinkage void do_fpe(struct pt_regs *regs, unsigned long fcr31)
 		sig = fpu_emulator_cop1Handler(regs, &current->thread.fpu, 1,
 					       &fault_addr);
 
-		/*
-		 * We can't allow the emulated instruction to leave any of
-		 * the cause bit set in $fcr31.
-		 */
-		current->thread.fpu.fcr31 &= ~FPU_CSR_ALL_X;
+		/* If something went wrong, signal */
+		process_fpemu_return(sig, fault_addr);
 
 		/* Restore the hardware register state */
 		own_fpu(1);	/* Using the FPU again.	 */
 
-		/* If something went wrong, signal */
-		process_fpemu_return(sig, fault_addr);
-
 		goto out;
 	} else if (fcr31 & FPU_CSR_INV_X)
 		info.si_code = FPE_FLTINV;
@@ -1392,13 +1398,22 @@ asmlinkage void do_cpu(struct pt_regs *regs)
 	exception_exit(prev_state);
 }
 
-asmlinkage void do_msa_fpe(struct pt_regs *regs)
+asmlinkage void do_msa_fpe(struct pt_regs *regs, unsigned int msacsr)
 {
 	enum ctx_state prev_state;
 
 	prev_state = exception_enter();
+	if (notify_die(DIE_MSAFP, "MSA FP exception", regs, 0,
+		       regs_to_trapnr(regs), SIGFPE) == NOTIFY_STOP)
+		goto out;
+
+	/* Clear MSACSR.Cause before enabling interrupts */
+	write_msa_csr(msacsr & ~MSA_CSR_CAUSEF);
+	local_irq_enable();
+
 	die_if_kernel("do_msa_fpe invoked from kernel context!", regs);
 	force_sig(SIGFPE, current);
+out:
 	exception_exit(prev_state);
 }
 

arch/mips/kvm/Makefile

 # Makefile for KVM support for MIPS
 #
 
-common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o coalesced_mmio.o)
+common-objs-y = $(addprefix ../../../virt/kvm/, kvm_main.o coalesced_mmio.o)
 
 EXTRA_CFLAGS += -Ivirt/kvm -Iarch/mips/kvm
 
-kvm-objs := $(common-objs) mips.o emulate.o locore.o \
+common-objs-$(CONFIG_CPU_HAS_MSA) += msa.o
+
+kvm-objs := $(common-objs-y) mips.o emulate.o locore.o \
 	    interrupt.o stats.o commpage.o \
-	    dyntrans.o trap_emul.o
+	    dyntrans.o trap_emul.o fpu.o
 
 obj-$(CONFIG_KVM)	+= kvm.o
 obj-y			+= callback.o tlb.o

arch/mips/kvm/emulate.c

@@ -884,6 +884,84 @@ enum emulation_result kvm_mips_emul_tlbp(struct kvm_vcpu *vcpu)
 	return EMULATE_DONE;
 }
 
+/**
+ * kvm_mips_config1_wrmask() - Find mask of writable bits in guest Config1
+ * @vcpu:	Virtual CPU.
+ *
+ * Finds the mask of bits which are writable in the guest's Config1 CP0
+ * register, by userland (currently read-only to the guest).
+ */
+unsigned int kvm_mips_config1_wrmask(struct kvm_vcpu *vcpu)
+{
+	unsigned int mask = 0;
+
+	/* Permit FPU to be present if FPU is supported */
+	if (kvm_mips_guest_can_have_fpu(&vcpu->arch))
+		mask |= MIPS_CONF1_FP;
+
+	return mask;
+}
+
+/**
+ * kvm_mips_config3_wrmask() - Find mask of writable bits in guest Config3
+ * @vcpu:	Virtual CPU.
+ *
+ * Finds the mask of bits which are writable in the guest's Config3 CP0
+ * register, by userland (currently read-only to the guest).
+ */
+unsigned int kvm_mips_config3_wrmask(struct kvm_vcpu *vcpu)
+{
+	/* Config4 is optional */
+	unsigned int mask = MIPS_CONF_M;
+
+	/* Permit MSA to be present if MSA is supported */
+	if (kvm_mips_guest_can_have_msa(&vcpu->arch))
+		mask |= MIPS_CONF3_MSA;
+
+	return mask;
+}
+
+/**
+ * kvm_mips_config4_wrmask() - Find mask of writable bits in guest Config4
+ * @vcpu:	Virtual CPU.
+ *
+ * Finds the mask of bits which are writable in the guest's Config4 CP0
+ * register, by userland (currently read-only to the guest).
+ */
+unsigned int kvm_mips_config4_wrmask(struct kvm_vcpu *vcpu)
+{
+	/* Config5 is optional */
+	return MIPS_CONF_M;
+}
+
+/**
+ * kvm_mips_config5_wrmask() - Find mask of writable bits in guest Config5
+ * @vcpu:	Virtual CPU.
+ *
+ * Finds the mask of bits which are writable in the guest's Config5 CP0
+ * register, by the guest itself.
+ */
+unsigned int kvm_mips_config5_wrmask(struct kvm_vcpu *vcpu)
+{
+	unsigned int mask = 0;
+
+	/* Permit MSAEn changes if MSA supported and enabled */
+	if (kvm_mips_guest_has_msa(&vcpu->arch))
+		mask |= MIPS_CONF5_MSAEN;
+
+	/*
+	 * Permit guest FPU mode changes if FPU is enabled and the relevant
+	 * feature exists according to FIR register.
+	 */
+	if (kvm_mips_guest_has_fpu(&vcpu->arch)) {
+		if (cpu_has_fre)
+			mask |= MIPS_CONF5_FRE;
+		/* We don't support UFR or UFE */
+	}
+
+	return mask;
+}
+
 enum emulation_result kvm_mips_emulate_CP0(uint32_t inst, uint32_t *opc,
 					   uint32_t cause, struct kvm_run *run,
 					   struct kvm_vcpu *vcpu)
@@ -1021,18 +1099,114 @@ enum emulation_result kvm_mips_emulate_CP0(uint32_t inst, uint32_t *opc,
 				kvm_mips_write_compare(vcpu,
 						       vcpu->arch.gprs[rt]);
 			} else if ((rd == MIPS_CP0_STATUS) && (sel == 0)) {
-				kvm_write_c0_guest_status(cop0,
-							  vcpu->arch.gprs[rt]);
+				unsigned int old_val, val, change;
+
+				old_val = kvm_read_c0_guest_status(cop0);
+				val = vcpu->arch.gprs[rt];
+				change = val ^ old_val;
+
+				/* Make sure that the NMI bit is never set */
+				val &= ~ST0_NMI;
+
+				/*
+				 * Don't allow CU1 or FR to be set unless FPU
+				 * capability enabled and exists in guest
+				 * configuration.
+				 */
+				if (!kvm_mips_guest_has_fpu(&vcpu->arch))
+					val &= ~(ST0_CU1 | ST0_FR);
+
+				/*
+				 * Also don't allow FR to be set if host doesn't
+				 * support it.
+				 */
+				if (!(current_cpu_data.fpu_id & MIPS_FPIR_F64))
+					val &= ~ST0_FR;
+
+
+				/* Handle changes in FPU mode */
+				preempt_disable();
+
+				/*
+				 * FPU and Vector register state is made
+				 * UNPREDICTABLE by a change of FR, so don't
+				 * even bother saving it.
+				 */
+				if (change & ST0_FR)
+					kvm_drop_fpu(vcpu);
+
+				/*
+				 * If MSA state is already live, it is undefined
+				 * how it interacts with FR=0 FPU state, and we
+				 * don't want to hit reserved instruction
+				 * exceptions trying to save the MSA state later
+				 * when CU=1 && FR=1, so play it safe and save
+				 * it first.
+				 */
+				if (change & ST0_CU1 && !(val & ST0_FR) &&
+				    vcpu->arch.fpu_inuse & KVM_MIPS_FPU_MSA)
+					kvm_lose_fpu(vcpu);
+
 				/*
-				 * Make sure that CU1 and NMI bits are
-				 * never set
+				 * Propagate CU1 (FPU enable) changes
+				 * immediately if the FPU context is already
+				 * loaded. When disabling we leave the context
+				 * loaded so it can be quickly enabled again in
+				 * the near future.
 				 */
-				kvm_clear_c0_guest_status(cop0,
-							  (ST0_CU1 | ST0_NMI));
+				if (change & ST0_CU1 &&
+				    vcpu->arch.fpu_inuse & KVM_MIPS_FPU_FPU)
+					change_c0_status(ST0_CU1, val);
+
+				preempt_enable();
+
+				kvm_write_c0_guest_status(cop0, val);
 
 #ifdef CONFIG_KVM_MIPS_DYN_TRANS
+				/*
+				 * If FPU present, we need CU1/FR bits to take
+				 * effect fairly soon.
+				 */
+				if (!kvm_mips_guest_has_fpu(&vcpu->arch))
 					kvm_mips_trans_mtc0(inst, opc, vcpu);
 #endif
+			} else if ((rd == MIPS_CP0_CONFIG) && (sel == 5)) {
+				unsigned int old_val, val, change, wrmask;
+
+				old_val = kvm_read_c0_guest_config5(cop0);
+				val = vcpu->arch.gprs[rt];
+
+				/* Only a few bits are writable in Config5 */
+				wrmask = kvm_mips_config5_wrmask(vcpu);
+				change = (val ^ old_val) & wrmask;
+				val = old_val ^ change;
+
+
+				/* Handle changes in FPU/MSA modes */
+				preempt_disable();
+
+				/*
+				 * Propagate FRE changes immediately if the FPU
+				 * context is already loaded.
+				 */
+				if (change & MIPS_CONF5_FRE &&
+				    vcpu->arch.fpu_inuse & KVM_MIPS_FPU_FPU)
+					change_c0_config5(MIPS_CONF5_FRE, val);
+
+				/*
+				 * Propagate MSAEn changes immediately if the
+				 * MSA context is already loaded. When disabling
+				 * we leave the context loaded so it can be
+				 * quickly enabled again in the near future.
+				 */
+				if (change & MIPS_CONF5_MSAEN &&
+				    vcpu->arch.fpu_inuse & KVM_MIPS_FPU_MSA)
+					change_c0_config5(MIPS_CONF5_MSAEN,
+							  val);
+
+				preempt_enable();
+
+				kvm_write_c0_guest_config5(cop0, val);
 			} else if ((rd == MIPS_CP0_CAUSE) && (sel == 0)) {
 				uint32_t old_cause, new_cause;
 
@@ -1970,6 +2144,146 @@ enum emulation_result kvm_mips_emulate_bp_exc(unsigned long cause,
 	return er;
 }
 
+enum emulation_result kvm_mips_emulate_trap_exc(unsigned long cause,
+						uint32_t *opc,
+						struct kvm_run *run,
+						struct kvm_vcpu *vcpu)
+{
+	struct mips_coproc *cop0 = vcpu->arch.cop0;
+	struct kvm_vcpu_arch *arch = &vcpu->arch;
+	enum emulation_result er = EMULATE_DONE;
+
+	if ((kvm_read_c0_guest_status(cop0) & ST0_EXL) == 0) {
+		/* save old pc */
+		kvm_write_c0_guest_epc(cop0, arch->pc);
+		kvm_set_c0_guest_status(cop0, ST0_EXL);
+
+		if (cause & CAUSEF_BD)
+			kvm_set_c0_guest_cause(cop0, CAUSEF_BD);
+		else
+			kvm_clear_c0_guest_cause(cop0, CAUSEF_BD);
+
+		kvm_debug("Delivering TRAP @ pc %#lx\n", arch->pc);
+
+		kvm_change_c0_guest_cause(cop0, (0xff),
+					  (T_TRAP << CAUSEB_EXCCODE));
+
+		/* Set PC to the exception entry point */
+		arch->pc = KVM_GUEST_KSEG0 + 0x180;
+
+	} else {
+		kvm_err("Trying to deliver TRAP when EXL is already set\n");
+		er = EMULATE_FAIL;
+	}
+
+	return er;
+}
+
+enum emulation_result kvm_mips_emulate_msafpe_exc(unsigned long cause,
+						  uint32_t *opc,
+						  struct kvm_run *run,
+						  struct kvm_vcpu *vcpu)
+{
+	struct mips_coproc *cop0 = vcpu->arch.cop0;
+	struct kvm_vcpu_arch *arch = &vcpu->arch;
+	enum emulation_result er = EMULATE_DONE;
+
+	if ((kvm_read_c0_guest_status(cop0) & ST0_EXL) == 0) {
+		/* save old pc */
+		kvm_write_c0_guest_epc(cop0, arch->pc);
+		kvm_set_c0_guest_status(cop0, ST0_EXL);
+
+		if (cause & CAUSEF_BD)
+			kvm_set_c0_guest_cause(cop0, CAUSEF_BD);
+		else
+			kvm_clear_c0_guest_cause(cop0, CAUSEF_BD);
+
+		kvm_debug("Delivering MSAFPE @ pc %#lx\n", arch->pc);
+
+		kvm_change_c0_guest_cause(cop0, (0xff),
+					  (T_MSAFPE << CAUSEB_EXCCODE));
+
+		/* Set PC to the exception entry point */
+		arch->pc = KVM_GUEST_KSEG0 + 0x180;
+
+	} else {
+		kvm_err("Trying to deliver MSAFPE when EXL is already set\n");
+		er = EMULATE_FAIL;
+	}
+
+	return er;
+}
+
+enum emulation_result kvm_mips_emulate_fpe_exc(unsigned long cause,
+					       uint32_t *opc,
+					       struct kvm_run *run,
+					       struct kvm_vcpu *vcpu)
+{
+	struct mips_coproc *cop0 = vcpu->arch.cop0;
+	struct kvm_vcpu_arch *arch = &vcpu->arch;
+	enum emulation_result er = EMULATE_DONE;
+
+	if ((kvm_read_c0_guest_status(cop0) & ST0_EXL) == 0) {
+		/* save old pc */
+		kvm_write_c0_guest_epc(cop0, arch->pc);
+		kvm_set_c0_guest_status(cop0, ST0_EXL);
+
+		if (cause & CAUSEF_BD)
+			kvm_set_c0_guest_cause(cop0, CAUSEF_BD);
+		else
+			kvm_clear_c0_guest_cause(cop0, CAUSEF_BD);
+
+		kvm_debug("Delivering FPE @ pc %#lx\n", arch->pc);
+
+		kvm_change_c0_guest_cause(cop0, (0xff),
+					  (T_FPE << CAUSEB_EXCCODE));
+
+		/* Set PC to the exception entry point */
+		arch->pc = KVM_GUEST_KSEG0 + 0x180;
+
+	} else {
+		kvm_err("Trying to deliver FPE when EXL is already set\n");
+		er = EMULATE_FAIL;
+	}
+
+	return er;
+}
+
+enum emulation_result kvm_mips_emulate_msadis_exc(unsigned long cause,
+						  uint32_t *opc,
+						  struct kvm_run *run,
+						  struct kvm_vcpu *vcpu)
+{
+	struct mips_coproc *cop0 = vcpu->arch.cop0;
+	struct kvm_vcpu_arch *arch = &vcpu->arch;
+	enum emulation_result er = EMULATE_DONE;
+
+	if ((kvm_read_c0_guest_status(cop0) & ST0_EXL) == 0) {
+		/* save old pc */
+		kvm_write_c0_guest_epc(cop0, arch->pc);
+		kvm_set_c0_guest_status(cop0, ST0_EXL);
+
+		if (cause & CAUSEF_BD)
+			kvm_set_c0_guest_cause(cop0, CAUSEF_BD);
+		else
+			kvm_clear_c0_guest_cause(cop0, CAUSEF_BD);
+
+		kvm_debug("Delivering MSADIS @ pc %#lx\n", arch->pc);
+
+		kvm_change_c0_guest_cause(cop0, (0xff),
+					  (T_MSADIS << CAUSEB_EXCCODE));
+
+		/* Set PC to the exception entry point */
+		arch->pc = KVM_GUEST_KSEG0 + 0x180;
+
+	} else {
+		kvm_err("Trying to deliver MSADIS when EXL is already set\n");
+		er = EMULATE_FAIL;
+	}
+
+	return er;
+}
+
 /* ll/sc, rdhwr, sync emulation */
 
 #define OPCODE 0xfc000000
@@ -2176,6 +2490,10 @@ enum emulation_result kvm_mips_check_privilege(unsigned long cause,
 		case T_SYSCALL:
 		case T_BREAK:
 		case T_RES_INST:
+		case T_TRAP:
+		case T_MSAFPE:
+		case T_FPE:
+		case T_MSADIS:
 			break;
 
 		case T_COP_UNUSABLE:

arch/mips/kvm/fpu.S

+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * FPU context handling code for KVM.
+ *
+ * Copyright (C) 2015 Imagination Technologies Ltd.
+ */
+
+#include <asm/asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/fpregdef.h>
+#include <asm/mipsregs.h>
+#include <asm/regdef.h>
+
+	.set	noreorder
+	.set	noat
+
+LEAF(__kvm_save_fpu)
+	.set	push
+	.set	mips64r2
+	SET_HARDFLOAT
+	mfc0	t0, CP0_STATUS
+	sll     t0, t0, 5			# is Status.FR set?
+	bgez    t0, 1f				# no: skip odd doubles
+	 nop
+	sdc1	$f1,  VCPU_FPR1(a0)
+	sdc1	$f3,  VCPU_FPR3(a0)
+	sdc1	$f5,  VCPU_FPR5(a0)
+	sdc1	$f7,  VCPU_FPR7(a0)
+	sdc1	$f9,  VCPU_FPR9(a0)
+	sdc1	$f11, VCPU_FPR11(a0)
+	sdc1	$f13, VCPU_FPR13(a0)
+	sdc1	$f15, VCPU_FPR15(a0)
+	sdc1	$f17, VCPU_FPR17(a0)
+	sdc1	$f19, VCPU_FPR19(a0)
+	sdc1	$f21, VCPU_FPR21(a0)
+	sdc1	$f23, VCPU_FPR23(a0)
+	sdc1	$f25, VCPU_FPR25(a0)
+	sdc1	$f27, VCPU_FPR27(a0)
+	sdc1	$f29, VCPU_FPR29(a0)
+	sdc1	$f31, VCPU_FPR31(a0)
+1:	sdc1	$f0,  VCPU_FPR0(a0)
+	sdc1	$f2,  VCPU_FPR2(a0)
+	sdc1	$f4,  VCPU_FPR4(a0)
+	sdc1	$f6,  VCPU_FPR6(a0)
+	sdc1	$f8,  VCPU_FPR8(a0)
+	sdc1	$f10, VCPU_FPR10(a0)
+	sdc1	$f12, VCPU_FPR12(a0)
+	sdc1	$f14, VCPU_FPR14(a0)
+	sdc1	$f16, VCPU_FPR16(a0)
+	sdc1	$f18, VCPU_FPR18(a0)
+	sdc1	$f20, VCPU_FPR20(a0)
+	sdc1	$f22, VCPU_FPR22(a0)
+	sdc1	$f24, VCPU_FPR24(a0)
+	sdc1	$f26, VCPU_FPR26(a0)
+	sdc1	$f28, VCPU_FPR28(a0)
+	jr	ra
+	 sdc1	$f30, VCPU_FPR30(a0)
+	.set	pop
+	END(__kvm_save_fpu)
+
+LEAF(__kvm_restore_fpu)
+	.set	push
+	.set	mips64r2
+	SET_HARDFLOAT
+	mfc0	t0, CP0_STATUS
+	sll     t0, t0, 5			# is Status.FR set?
+	bgez    t0, 1f				# no: skip odd doubles
+	 nop
+	ldc1	$f1,  VCPU_FPR1(a0)
+	ldc1	$f3,  VCPU_FPR3(a0)
+	ldc1	$f5,  VCPU_FPR5(a0)
+	ldc1	$f7,  VCPU_FPR7(a0)
+	ldc1	$f9,  VCPU_FPR9(a0)
+	ldc1	$f11, VCPU_FPR11(a0)
+	ldc1	$f13, VCPU_FPR13(a0)
+	ldc1	$f15, VCPU_FPR15(a0)
+	ldc1	$f17, VCPU_FPR17(a0)
+	ldc1	$f19, VCPU_FPR19(a0)
+	ldc1	$f21, VCPU_FPR21(a0)
+	ldc1	$f23, VCPU_FPR23(a0)
+	ldc1	$f25, VCPU_FPR25(a0)
+	ldc1	$f27, VCPU_FPR27(a0)
+	ldc1	$f29, VCPU_FPR29(a0)
+	ldc1	$f31, VCPU_FPR31(a0)
+1:	ldc1	$f0,  VCPU_FPR0(a0)
+	ldc1	$f2,  VCPU_FPR2(a0)
+	ldc1	$f4,  VCPU_FPR4(a0)
+	ldc1	$f6,  VCPU_FPR6(a0)
+	ldc1	$f8,  VCPU_FPR8(a0)
+	ldc1	$f10, VCPU_FPR10(a0)
+	ldc1	$f12, VCPU_FPR12(a0)
+	ldc1	$f14, VCPU_FPR14(a0)
+	ldc1	$f16, VCPU_FPR16(a0)
+	ldc1	$f18, VCPU_FPR18(a0)
+	ldc1	$f20, VCPU_FPR20(a0)
+	ldc1	$f22, VCPU_FPR22(a0)
+	ldc1	$f24, VCPU_FPR24(a0)
+	ldc1	$f26, VCPU_FPR26(a0)
+	ldc1	$f28, VCPU_FPR28(a0)
+	jr	ra
+	 ldc1	$f30, VCPU_FPR30(a0)
+	.set	pop
+	END(__kvm_restore_fpu)
+
+LEAF(__kvm_restore_fcsr)
+	.set	push
+	SET_HARDFLOAT
+	lw	t0, VCPU_FCR31(a0)
+	/*
+	 * The ctc1 must stay at this offset in __kvm_restore_fcsr.
+	 * See kvm_mips_csr_die_notify() which handles t0 containing a value
+	 * which triggers an FP Exception, which must be stepped over and
+	 * ignored since the set cause bits must remain there for the guest.
+	 */
+	ctc1	t0, fcr31
+	jr	ra
+	 nop
+	.set	pop
+	END(__kvm_restore_fcsr)

arch/mips/kvm/locore.S

@@ -36,6 +36,8 @@
 #define PT_HOST_USERLOCAL   PT_EPC
 
 #define CP0_DDATA_LO        $28,3
+#define CP0_CONFIG3         $16,3
+#define CP0_CONFIG5         $16,5
 #define CP0_EBASE           $15,1
 
 #define CP0_INTCTL          $12,1
@@ -353,6 +355,42 @@ NESTED (MIPSX(GuestException), CALLFRAME_SIZ, ra)
 	LONG_L	k0, VCPU_HOST_EBASE(k1)
 	mtc0	k0,CP0_EBASE
 
+	/*
+	 * If FPU is enabled, save FCR31 and clear it so that later ctc1's don't
+	 * trigger FPE for pending exceptions.
+	 */
+	.set	at
+	and	v1, v0, ST0_CU1
+	beqz	v1, 1f
+	 nop
+	.set	push
+	SET_HARDFLOAT
+	cfc1	t0, fcr31
+	sw	t0, VCPU_FCR31(k1)
+	ctc1	zero,fcr31
+	.set	pop
+	.set	noat
+1:
+
+#ifdef CONFIG_CPU_HAS_MSA
+	/*
+	 * If MSA is enabled, save MSACSR and clear it so that later
+	 * instructions don't trigger MSAFPE for pending exceptions.
+	 */
+	mfc0	t0, CP0_CONFIG3
+	ext	t0, t0, 28, 1 /* MIPS_CONF3_MSAP */
+	beqz	t0, 1f
+	 nop
+	mfc0	t0, CP0_CONFIG5
+	ext	t0, t0, 27, 1 /* MIPS_CONF5_MSAEN */
+	beqz	t0, 1f
+	 nop
+	_cfcmsa	t0, MSA_CSR
+	sw	t0, VCPU_MSA_CSR(k1)
+	_ctcmsa	MSA_CSR, zero
+1:
+#endif
+
 	/* Now that the new EBASE has been loaded, unset BEV and KSU_USER */
 	.set	at
 	and	v0, v0, ~(ST0_EXL | KSU_USER | ST0_IE)

arch/mips/kvm/msa.S

+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * MIPS SIMD Architecture (MSA) context handling code for KVM.
+ *
+ * Copyright (C) 2015 Imagination Technologies Ltd.
+ */
+
+#include <asm/asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/asmmacro.h>
+#include <asm/regdef.h>
+
+	.set	noreorder
+	.set	noat
+
+LEAF(__kvm_save_msa)
+	st_d	0,  VCPU_FPR0,  a0
+	st_d	1,  VCPU_FPR1,  a0
+	st_d	2,  VCPU_FPR2,  a0
+	st_d	3,  VCPU_FPR3,  a0
+	st_d	4,  VCPU_FPR4,  a0
+	st_d	5,  VCPU_FPR5,  a0
+	st_d	6,  VCPU_FPR6,  a0
+	st_d	7,  VCPU_FPR7,  a0
+	st_d	8,  VCPU_FPR8,  a0
+	st_d	9,  VCPU_FPR9,  a0
+	st_d	10, VCPU_FPR10, a0
+	st_d	11, VCPU_FPR11, a0
+	st_d	12, VCPU_FPR12, a0
+	st_d	13, VCPU_FPR13, a0
+	st_d	14, VCPU_FPR14, a0
+	st_d	15, VCPU_FPR15, a0
+	st_d	16, VCPU_FPR16, a0
+	st_d	17, VCPU_FPR17, a0
+	st_d	18, VCPU_FPR18, a0
+	st_d	19, VCPU_FPR19, a0
+	st_d	20, VCPU_FPR20, a0
+	st_d	21, VCPU_FPR21, a0
+	st_d	22, VCPU_FPR22, a0
+	st_d	23, VCPU_FPR23, a0
+	st_d	24, VCPU_FPR24, a0
+	st_d	25, VCPU_FPR25, a0
+	st_d	26, VCPU_FPR26, a0
+	st_d	27, VCPU_FPR27, a0
+	st_d	28, VCPU_FPR28, a0
+	st_d	29, VCPU_FPR29, a0
+	st_d	30, VCPU_FPR30, a0
+	st_d	31, VCPU_FPR31, a0
+	jr	ra
+	 nop
+	END(__kvm_save_msa)
+
+LEAF(__kvm_restore_msa)
+	ld_d	0,  VCPU_FPR0,  a0
+	ld_d	1,  VCPU_FPR1,  a0
+	ld_d	2,  VCPU_FPR2,  a0
+	ld_d	3,  VCPU_FPR3,  a0
+	ld_d	4,  VCPU_FPR4,  a0
+	ld_d	5,  VCPU_FPR5,  a0
+	ld_d	6,  VCPU_FPR6,  a0
+	ld_d	7,  VCPU_FPR7,  a0
+	ld_d	8,  VCPU_FPR8,  a0
+	ld_d	9,  VCPU_FPR9,  a0
+	ld_d	10, VCPU_FPR10, a0
+	ld_d	11, VCPU_FPR11, a0
+	ld_d	12, VCPU_FPR12, a0
+	ld_d	13, VCPU_FPR13, a0
+	ld_d	14, VCPU_FPR14, a0
+	ld_d	15, VCPU_FPR15, a0
+	ld_d	16, VCPU_FPR16, a0
+	ld_d	17, VCPU_FPR17, a0
+	ld_d	18, VCPU_FPR18, a0
+	ld_d	19, VCPU_FPR19, a0
+	ld_d	20, VCPU_FPR20, a0
+	ld_d	21, VCPU_FPR21, a0
+	ld_d	22, VCPU_FPR22, a0
+	ld_d	23, VCPU_FPR23, a0
+	ld_d	24, VCPU_FPR24, a0
+	ld_d	25, VCPU_FPR25, a0
+	ld_d	26, VCPU_FPR26, a0
+	ld_d	27, VCPU_FPR27, a0
+	ld_d	28, VCPU_FPR28, a0
+	ld_d	29, VCPU_FPR29, a0
+	ld_d	30, VCPU_FPR30, a0
+	ld_d	31, VCPU_FPR31, a0
+	jr	ra
+	 nop
+	END(__kvm_restore_msa)
+
+	.macro	kvm_restore_msa_upper	wr, off, base
+	.set	push
+	.set	noat
+#ifdef CONFIG_64BIT
+	ld	$1, \off(\base)
+	insert_d \wr, 1
+#elif defined(CONFIG_CPU_LITTLE_ENDIAN)
+	lw	$1, \off(\base)
+	insert_w \wr, 2
+	lw	$1, (\off+4)(\base)
+	insert_w \wr, 3
+#else /* CONFIG_CPU_BIG_ENDIAN */
+	lw	$1, (\off+4)(\base)
+	insert_w \wr, 2
+	lw	$1, \off(\base)
+	insert_w \wr, 3
+#endif
+	.set	pop
+	.endm
+
+LEAF(__kvm_restore_msa_upper)
+	kvm_restore_msa_upper	0,  VCPU_FPR0 +8, a0
+	kvm_restore_msa_upper	1,  VCPU_FPR1 +8, a0
+	kvm_restore_msa_upper	2,  VCPU_FPR2 +8, a0
+	kvm_restore_msa_upper	3,  VCPU_FPR3 +8, a0
+	kvm_restore_msa_upper	4,  VCPU_FPR4 +8, a0
+	kvm_restore_msa_upper	5,  VCPU_FPR5 +8, a0
+	kvm_restore_msa_upper	6,  VCPU_FPR6 +8, a0
+	kvm_restore_msa_upper	7,  VCPU_FPR7 +8, a0
+	kvm_restore_msa_upper	8,  VCPU_FPR8 +8, a0
+	kvm_restore_msa_upper	9,  VCPU_FPR9 +8, a0
+	kvm_restore_msa_upper	10, VCPU_FPR10+8, a0
+	kvm_restore_msa_upper	11, VCPU_FPR11+8, a0
+	kvm_restore_msa_upper	12, VCPU_FPR12+8, a0
+	kvm_restore_msa_upper	13, VCPU_FPR13+8, a0
+	kvm_restore_msa_upper	14, VCPU_FPR14+8, a0
+	kvm_restore_msa_upper	15, VCPU_FPR15+8, a0
+	kvm_restore_msa_upper	16, VCPU_FPR16+8, a0
+	kvm_restore_msa_upper	17, VCPU_FPR17+8, a0
+	kvm_restore_msa_upper	18, VCPU_FPR18+8, a0
+	kvm_restore_msa_upper	19, VCPU_FPR19+8, a0
+	kvm_restore_msa_upper	20, VCPU_FPR20+8, a0
+	kvm_restore_msa_upper	21, VCPU_FPR21+8, a0
+	kvm_restore_msa_upper	22, VCPU_FPR22+8, a0
+	kvm_restore_msa_upper	23, VCPU_FPR23+8, a0
+	kvm_restore_msa_upper	24, VCPU_FPR24+8, a0
+	kvm_restore_msa_upper	25, VCPU_FPR25+8, a0
+	kvm_restore_msa_upper	26, VCPU_FPR26+8, a0
+	kvm_restore_msa_upper	27, VCPU_FPR27+8, a0
+	kvm_restore_msa_upper	28, VCPU_FPR28+8, a0
+	kvm_restore_msa_upper	29, VCPU_FPR29+8, a0
+	kvm_restore_msa_upper	30, VCPU_FPR30+8, a0
+	kvm_restore_msa_upper	31, VCPU_FPR31+8, a0
+	jr	ra
+	 nop
+	END(__kvm_restore_msa_upper)
+
+LEAF(__kvm_restore_msacsr)
+	lw	t0, VCPU_MSA_CSR(a0)
+	/*
+	 * The ctcmsa must stay at this offset in __kvm_restore_msacsr.
+	 * See kvm_mips_csr_die_notify() which handles t0 containing a value
+	 * which triggers an MSA FP Exception, which must be stepped over and
+	 * ignored since the set cause bits must remain there for the guest.
+	 */
+	_ctcmsa	MSA_CSR, t0
+	jr	ra
+	 nop
+	END(__kvm_restore_msacsr)

arch/mips/kvm/stats.c

@@ -25,6 +25,10 @@ char *kvm_mips_exit_types_str[MAX_KVM_MIPS_EXIT_TYPES] = {
 	"System Call",
 	"Reserved Inst",
 	"Break Inst",
+	"Trap Inst",
+	"MSA FPE",
+	"FPE",
+	"MSA Disabled",
 	"D-Cache Flushes",
 };
 

arch/mips/kvm/tlb.c

@@ -733,6 +733,9 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 		}
 	}
 
+	/* restore guest state to registers */
+	kvm_mips_callbacks->vcpu_set_regs(vcpu);
+
 	local_irq_restore(flags);
 
 }
@@ -751,6 +754,9 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 	vcpu->arch.preempt_entryhi = read_c0_entryhi();
 	vcpu->arch.last_sched_cpu = cpu;
 
+	/* save guest state in registers */
+	kvm_mips_callbacks->vcpu_get_regs(vcpu);
+
 	if (((cpu_context(cpu, current->mm) ^ asid_cache(cpu)) &
 	     ASID_VERSION_MASK)) {
 		kvm_debug("%s: Dropping MMU Context:  %#lx\n", __func__,

arch/mips/kvm/trap_emul.c

@@ -39,16 +39,30 @@ static gpa_t kvm_trap_emul_gva_to_gpa_cb(gva_t gva)
 
 static int kvm_trap_emul_handle_cop_unusable(struct kvm_vcpu *vcpu)
 {
+	struct mips_coproc *cop0 = vcpu->arch.cop0;
 	struct kvm_run *run = vcpu->run;
 	uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
 	unsigned long cause = vcpu->arch.host_cp0_cause;
 	enum emulation_result er = EMULATE_DONE;
 	int ret = RESUME_GUEST;
 
-	if (((cause & CAUSEF_CE) >> CAUSEB_CE) == 1)
+	if (((cause & CAUSEF_CE) >> CAUSEB_CE) == 1) {
+		/* FPU Unusable */
+		if (!kvm_mips_guest_has_fpu(&vcpu->arch) ||
+		    (kvm_read_c0_guest_status(cop0) & ST0_CU1) == 0) {
+			/*
+			 * Unusable/no FPU in guest:
+			 * deliver guest COP1 Unusable Exception
+			 */
 			er = kvm_mips_emulate_fpu_exc(cause, opc, run, vcpu);
-	else
+		} else {
+			/* Restore FPU state */
+			kvm_own_fpu(vcpu);
+			er = EMULATE_DONE;
+		}
+	} else {
 		er = kvm_mips_emulate_inst(cause, opc, run, vcpu);
+	}
 
 	switch (er) {
 	case EMULATE_DONE:
@@ -330,6 +344,107 @@ static int kvm_trap_emul_handle_break(struct kvm_vcpu *vcpu)
 	return ret;
 }
 
+static int kvm_trap_emul_handle_trap(struct kvm_vcpu *vcpu)
+{
+	struct kvm_run *run = vcpu->run;
+	uint32_t __user *opc = (uint32_t __user *)vcpu->arch.pc;
+	unsigned long cause = vcpu->arch.host_cp0_cause;
+	enum emulation_result er = EMULATE_DONE;
+	int ret = RESUME_GUEST;
+
+	er = kvm_mips_emulate_trap_exc(cause, opc, run, vcpu);
+	if (er == EMULATE_DONE) {
+		ret = RESUME_GUEST;
+	} else {
+		run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
+		ret = RESUME_HOST;
+	}
+	return ret;
+}
+
+static int kvm_trap_emul_handle_msa_fpe(struct kvm_vcpu *vcpu)
+{
+	struct kvm_run *run = vcpu->run;
+	uint32_t __user *opc = (uint32_t __user *)vcpu->arch.pc;
+	unsigned long cause = vcpu->arch.host_cp0_cause;
+	enum emulation_result er = EMULATE_DONE;
+	int ret = RESUME_GUEST;
+
+	er = kvm_mips_emulate_msafpe_exc(cause, opc, run, vcpu);
+	if (er == EMULATE_DONE) {
+		ret = RESUME_GUEST;
+	} else {
+		run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
+		ret = RESUME_HOST;
+	}
+	return ret;
+}
+
+static int kvm_trap_emul_handle_fpe(struct kvm_vcpu *vcpu)
+{
+	struct kvm_run *run = vcpu->run;
+	uint32_t __user *opc = (uint32_t __user *)vcpu->arch.pc;
+	unsigned long cause = vcpu->arch.host_cp0_cause;
+	enum emulation_result er = EMULATE_DONE;
+	int ret = RESUME_GUEST;
+
+	er = kvm_mips_emulate_fpe_exc(cause, opc, run, vcpu);
+	if (er == EMULATE_DONE) {
+		ret = RESUME_GUEST;
+	} else {
+		run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
+		ret = RESUME_HOST;
+	}
+	return ret;
+}
+
+/**
+ * kvm_trap_emul_handle_msa_disabled() - Guest used MSA while disabled in root.
+ * @vcpu:	Virtual CPU context.
+ *
+ * Handle when the guest attempts to use MSA when it is disabled.
+ */
+static int kvm_trap_emul_handle_msa_disabled(struct kvm_vcpu *vcpu)
+{
+	struct mips_coproc *cop0 = vcpu->arch.cop0;
+	struct kvm_run *run = vcpu->run;
+	uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
+	unsigned long cause = vcpu->arch.host_cp0_cause;
+	enum emulation_result er = EMULATE_DONE;
+	int ret = RESUME_GUEST;
+
+	if (!kvm_mips_guest_has_msa(&vcpu->arch) ||
+	    (kvm_read_c0_guest_status(cop0) & (ST0_CU1 | ST0_FR)) == ST0_CU1) {
+		/*
+		 * No MSA in guest, or FPU enabled and not in FR=1 mode,
+		 * guest reserved instruction exception
+		 */
+		er = kvm_mips_emulate_ri_exc(cause, opc, run, vcpu);
+	} else if (!(kvm_read_c0_guest_config5(cop0) & MIPS_CONF5_MSAEN)) {
+		/* MSA disabled by guest, guest MSA disabled exception */
+		er = kvm_mips_emulate_msadis_exc(cause, opc, run, vcpu);
+	} else {
+		/* Restore MSA/FPU state */
+		kvm_own_msa(vcpu);
+		er = EMULATE_DONE;
+	}
+
+	switch (er) {
+	case EMULATE_DONE:
+		ret = RESUME_GUEST;
+		break;
+
+	case EMULATE_FAIL:
+		run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
+		ret = RESUME_HOST;
+		break;
+
+	default:
+		BUG();
+	}
+	return ret;
+}
+
 static int kvm_trap_emul_vm_init(struct kvm *kvm)
 {
 	return 0;
@@ -351,8 +466,9 @@ static int kvm_trap_emul_vcpu_setup(struct kvm_vcpu *vcpu)
 	 * guest will come up as expected, for now we simulate a MIPS 24kc
 	 */
 	kvm_write_c0_guest_prid(cop0, 0x00019300);
-	kvm_write_c0_guest_config(cop0,
-				  MIPS_CONFIG0 | (0x1 << CP0C0_AR) |
+	/* Have config1, Cacheable, noncoherent, write-back, write allocate */
+	kvm_write_c0_guest_config(cop0, MIPS_CONF_M | (0x3 << CP0C0_K0) |
+				  (0x1 << CP0C0_AR) |
 				  (MMU_TYPE_R4000 << CP0C0_MT));
 
 	/* Read the cache characteristics from the host Config1 Register */
@@ -368,10 +484,18 @@ static int kvm_trap_emul_vcpu_setup(struct kvm_vcpu *vcpu)
 	      (1 << CP0C1_WR) | (1 << CP0C1_CA));
 	kvm_write_c0_guest_config1(cop0, config1);
 
-	kvm_write_c0_guest_config2(cop0, MIPS_CONFIG2);
-	/* MIPS_CONFIG2 | (read_c0_config2() & 0xfff) */
-	kvm_write_c0_guest_config3(cop0, MIPS_CONFIG3 | (0 << CP0C3_VInt) |
-					 (1 << CP0C3_ULRI));
+	/* Have config3, no tertiary/secondary caches implemented */
+	kvm_write_c0_guest_config2(cop0, MIPS_CONF_M);
+	/* MIPS_CONF_M | (read_c0_config2() & 0xfff) */
+
+	/* Have config4, UserLocal */
+	kvm_write_c0_guest_config3(cop0, MIPS_CONF_M | MIPS_CONF3_ULRI);
+
+	/* Have config5 */
+	kvm_write_c0_guest_config4(cop0, MIPS_CONF_M);
+
+	/* No config6 */
+	kvm_write_c0_guest_config5(cop0, 0);
 
 	/* Set Wait IE/IXMT Ignore in Config7, IAR, AR */
 	kvm_write_c0_guest_config7(cop0, (MIPS_CONF7_WII) | (1 << 10));
@@ -416,6 +540,7 @@ static int kvm_trap_emul_set_one_reg(struct kvm_vcpu *vcpu,
 {
 	struct mips_coproc *cop0 = vcpu->arch.cop0;
 	int ret = 0;
+	unsigned int cur, change;
 
 	switch (reg->id) {
 	case KVM_REG_MIPS_CP0_COUNT:
@@ -444,6 +569,44 @@ static int kvm_trap_emul_set_one_reg(struct kvm_vcpu *vcpu,
 			kvm_write_c0_guest_cause(cop0, v);
 		}
 		break;
+	case KVM_REG_MIPS_CP0_CONFIG:
+		/* read-only for now */
+		break;
+	case KVM_REG_MIPS_CP0_CONFIG1:
+		cur = kvm_read_c0_guest_config1(cop0);
+		change = (cur ^ v) & kvm_mips_config1_wrmask(vcpu);
+		if (change) {
+			v = cur ^ change;
+			kvm_write_c0_guest_config1(cop0, v);
+		}
+		break;
+	case KVM_REG_MIPS_CP0_CONFIG2:
+		/* read-only for now */
+		break;
+	case KVM_REG_MIPS_CP0_CONFIG3:
+		cur = kvm_read_c0_guest_config3(cop0);
+		change = (cur ^ v) & kvm_mips_config3_wrmask(vcpu);
+		if (change) {
+			v = cur ^ change;
+			kvm_write_c0_guest_config3(cop0, v);
+		}
+		break;
+	case KVM_REG_MIPS_CP0_CONFIG4:
+		cur = kvm_read_c0_guest_config4(cop0);
+		change = (cur ^ v) & kvm_mips_config4_wrmask(vcpu);
+		if (change) {
+			v = cur ^ change;
+			kvm_write_c0_guest_config4(cop0, v);
+		}
+		break;
+	case KVM_REG_MIPS_CP0_CONFIG5:
+		cur = kvm_read_c0_guest_config5(cop0);
+		change = (cur ^ v) & kvm_mips_config5_wrmask(vcpu);
+		if (change) {
+			v = cur ^ change;
+			kvm_write_c0_guest_config5(cop0, v);
+		}
+		break;
 	case KVM_REG_MIPS_COUNT_CTL:
 		ret = kvm_mips_set_count_ctl(vcpu, v);
 		break;
@@ -459,6 +622,18 @@ static int kvm_trap_emul_set_one_reg(struct kvm_vcpu *vcpu,
 	return ret;
 }
 
+static int kvm_trap_emul_vcpu_get_regs(struct kvm_vcpu *vcpu)
+{
+	kvm_lose_fpu(vcpu);
+
+	return 0;
+}
+
+static int kvm_trap_emul_vcpu_set_regs(struct kvm_vcpu *vcpu)
+{
+	return 0;
+}
+
 static struct kvm_mips_callbacks kvm_trap_emul_callbacks = {
 	/* exit handlers */
 	.handle_cop_unusable = kvm_trap_emul_handle_cop_unusable,
@@ -470,6 +645,10 @@ static struct kvm_mips_callbacks kvm_trap_emul_callbacks = {
 	.handle_syscall = kvm_trap_emul_handle_syscall,
 	.handle_res_inst = kvm_trap_emul_handle_res_inst,
 	.handle_break = kvm_trap_emul_handle_break,
+	.handle_trap = kvm_trap_emul_handle_trap,
+	.handle_msa_fpe = kvm_trap_emul_handle_msa_fpe,
+	.handle_fpe = kvm_trap_emul_handle_fpe,
+	.handle_msa_disabled = kvm_trap_emul_handle_msa_disabled,
 
 	.vm_init = kvm_trap_emul_vm_init,
 	.vcpu_init = kvm_trap_emul_vcpu_init,
@@ -483,6 +662,8 @@ static struct kvm_mips_callbacks kvm_trap_emul_callbacks = {
 	.irq_clear = kvm_mips_irq_clear_cb,
 	.get_one_reg = kvm_trap_emul_get_one_reg,
 	.set_one_reg = kvm_trap_emul_set_one_reg,
+	.vcpu_get_regs = kvm_trap_emul_vcpu_get_regs,
+	.vcpu_set_regs = kvm_trap_emul_vcpu_set_regs,
 };
 
 int kvm_mips_emulation_init(struct kvm_mips_callbacks **install_callbacks)

arch/powerpc/kvm/mpic.c

@@ -34,7 +34,7 @@
 #include <asm/kvm_para.h>
 #include <asm/kvm_host.h>
 #include <asm/kvm_ppc.h>
-#include "iodev.h"
+#include <kvm/iodev.h>
 
 #define MAX_CPU     32
 #define MAX_SRC     256
@@ -289,11 +289,6 @@ static inline void IRQ_resetbit(struct irq_queue *q, int n_IRQ)
 	clear_bit(n_IRQ, q->queue);
 }
 
-static inline int IRQ_testbit(struct irq_queue *q, int n_IRQ)
-{
-	return test_bit(n_IRQ, q->queue);
-}
-
 static void IRQ_check(struct openpic *opp, struct irq_queue *q)
 {
 	int irq = -1;
@@ -1374,8 +1369,9 @@ static int kvm_mpic_write_internal(struct openpic *opp, gpa_t addr, u32 val)
 	return -ENXIO;
 }
 
-static int kvm_mpic_read(struct kvm_io_device *this, gpa_t addr,
-			 int len, void *ptr)
+static int kvm_mpic_read(struct kvm_vcpu *vcpu,
+			 struct kvm_io_device *this,
+			 gpa_t addr, int len, void *ptr)
 {
 	struct openpic *opp = container_of(this, struct openpic, mmio);
 	int ret;
@@ -1415,8 +1411,9 @@ static int kvm_mpic_read(struct kvm_io_device *this, gpa_t addr,
 	return ret;
 }
 
-static int kvm_mpic_write(struct kvm_io_device *this, gpa_t addr,
-			  int len, const void *ptr)
+static int kvm_mpic_write(struct kvm_vcpu *vcpu,
+			  struct kvm_io_device *this,
+			  gpa_t addr, int len, const void *ptr)
 {
 	struct openpic *opp = container_of(this, struct openpic, mmio);
 	int ret;

arch/powerpc/kvm/powerpc.c

@@ -807,7 +807,7 @@ int kvmppc_handle_load(struct kvm_run *run, struct kvm_vcpu *vcpu,
 
 	idx = srcu_read_lock(&vcpu->kvm->srcu);
 
-	ret = kvm_io_bus_read(vcpu->kvm, KVM_MMIO_BUS, run->mmio.phys_addr,
+	ret = kvm_io_bus_read(vcpu, KVM_MMIO_BUS, run->mmio.phys_addr,
 			      bytes, &run->mmio.data);
 
 	srcu_read_unlock(&vcpu->kvm->srcu, idx);
@@ -880,7 +880,7 @@ int kvmppc_handle_store(struct kvm_run *run, struct kvm_vcpu *vcpu,
 
 	idx = srcu_read_lock(&vcpu->kvm->srcu);
 
-	ret = kvm_io_bus_write(vcpu->kvm, KVM_MMIO_BUS, run->mmio.phys_addr,
+	ret = kvm_io_bus_write(vcpu, KVM_MMIO_BUS, run->mmio.phys_addr,
 			       bytes, &run->mmio.data);
 
 	srcu_read_unlock(&vcpu->kvm->srcu, idx);

arch/s390/include/asm/kvm_host.h

@@ -172,7 +172,9 @@ struct kvm_s390_sie_block {
 	__u32	fac;			/* 0x01a0 */
 	__u8	reserved1a4[20];	/* 0x01a4 */
 	__u64	cbrlo;			/* 0x01b8 */
-	__u8	reserved1c0[30];	/* 0x01c0 */
+	__u8	reserved1c0[8];		/* 0x01c0 */
+	__u32	ecd;			/* 0x01c8 */
+	__u8	reserved1cc[18];	/* 0x01cc */
 	__u64	pp;			/* 0x01de */
 	__u8	reserved1e6[2];		/* 0x01e6 */
 	__u64	itdba;			/* 0x01e8 */
@@ -183,11 +185,17 @@ struct kvm_s390_itdb {
 	__u8	data[256];
 } __packed;
 
+struct kvm_s390_vregs {
+	__vector128 vrs[32];
+	__u8	reserved200[512];	/* for future vector expansion */
+} __packed;
+
 struct sie_page {
 	struct kvm_s390_sie_block sie_block;
 	__u8 reserved200[1024];		/* 0x0200 */
 	struct kvm_s390_itdb itdb;	/* 0x0600 */
-	__u8 reserved700[2304];		/* 0x0700 */
+	__u8 reserved700[1280];		/* 0x0700 */
+	struct kvm_s390_vregs vregs;	/* 0x0c00 */
 } __packed;
 
 struct kvm_vcpu_stat {
@@ -238,6 +246,7 @@ struct kvm_vcpu_stat {
 	u32 instruction_sigp_stop;
 	u32 instruction_sigp_stop_store_status;
 	u32 instruction_sigp_store_status;
+	u32 instruction_sigp_store_adtl_status;
 	u32 instruction_sigp_arch;
 	u32 instruction_sigp_prefix;
 	u32 instruction_sigp_restart;
@@ -270,6 +279,7 @@ struct kvm_vcpu_stat {
 #define PGM_SPECIAL_OPERATION		0x13
 #define PGM_OPERAND			0x15
 #define PGM_TRACE_TABEL			0x16
+#define PGM_VECTOR_PROCESSING		0x1b
 #define PGM_SPACE_SWITCH		0x1c
 #define PGM_HFP_SQUARE_ROOT		0x1d
 #define PGM_PC_TRANSLATION_SPEC		0x1f
@@ -334,6 +344,11 @@ enum irq_types {
 	IRQ_PEND_COUNT
 };
 
+/* We have 2M for virtio device descriptor pages. Smallest amount of
+ * memory per page is 24 bytes (1 queue), so (2048*1024) / 24 = 87381
+ */
+#define KVM_S390_MAX_VIRTIO_IRQS 87381
+
 /*
  * Repressible (non-floating) machine check interrupts
  * subclass bits in MCIC
@@ -411,13 +426,32 @@ struct kvm_s390_local_interrupt {
 	unsigned long pending_irqs;
 };
 
+#define FIRQ_LIST_IO_ISC_0 0
+#define FIRQ_LIST_IO_ISC_1 1
+#define FIRQ_LIST_IO_ISC_2 2
+#define FIRQ_LIST_IO_ISC_3 3
+#define FIRQ_LIST_IO_ISC_4 4
+#define FIRQ_LIST_IO_ISC_5 5
+#define FIRQ_LIST_IO_ISC_6 6
+#define FIRQ_LIST_IO_ISC_7 7
+#define FIRQ_LIST_PFAULT   8
+#define FIRQ_LIST_VIRTIO   9
+#define FIRQ_LIST_COUNT   10
+#define FIRQ_CNTR_IO       0
+#define FIRQ_CNTR_SERVICE  1
+#define FIRQ_CNTR_VIRTIO   2
+#define FIRQ_CNTR_PFAULT   3
+#define FIRQ_MAX_COUNT     4
+
 struct kvm_s390_float_interrupt {
+	unsigned long pending_irqs;
 	spinlock_t lock;
-	struct list_head list;
-	atomic_t active;
+	struct list_head lists[FIRQ_LIST_COUNT];
+	int counters[FIRQ_MAX_COUNT];
+	struct kvm_s390_mchk_info mchk;
+	struct kvm_s390_ext_info srv_signal;
 	int next_rr_cpu;
 	unsigned long idle_mask[BITS_TO_LONGS(KVM_MAX_VCPUS)];
-	unsigned int irq_count;
 };
 
 struct kvm_hw_wp_info_arch {
@@ -465,6 +499,7 @@ struct kvm_vcpu_arch {
 	s390_fp_regs      host_fpregs;
 	unsigned int      host_acrs[NUM_ACRS];
 	s390_fp_regs      guest_fpregs;
+	struct kvm_s390_vregs	*host_vregs;
 	struct kvm_s390_local_interrupt local_int;
 	struct hrtimer    ckc_timer;
 	struct kvm_s390_pgm_info pgm;
@@ -553,6 +588,7 @@ struct kvm_arch{
 	int use_cmma;
 	int user_cpu_state_ctrl;
 	int user_sigp;
+	int user_stsi;
 	struct s390_io_adapter *adapters[MAX_S390_IO_ADAPTERS];
 	wait_queue_head_t ipte_wq;
 	int ipte_lock_count;

arch/s390/include/uapi/asm/kvm.h

@@ -150,6 +150,7 @@ struct kvm_guest_debug_arch {
 #define KVM_SYNC_CRS    (1UL << 3)
 #define KVM_SYNC_ARCH0  (1UL << 4)
 #define KVM_SYNC_PFAULT (1UL << 5)
+#define KVM_SYNC_VRS    (1UL << 6)
 /* definition of registers in kvm_run */
 struct kvm_sync_regs {
 	__u64 prefix;	/* prefix register */
@@ -164,6 +165,9 @@ struct kvm_sync_regs {
 	__u64 pft;	/* pfault token [PFAULT] */
 	__u64 pfs;	/* pfault select [PFAULT] */
 	__u64 pfc;	/* pfault compare [PFAULT] */
+	__u64 vrs[32][2];	/* vector registers */
+	__u8  reserved[512];	/* for future vector expansion */
+	__u32 fpc;	/* only valid with vector registers */
 };
 
 #define KVM_REG_S390_TODPR	(KVM_REG_S390 | KVM_REG_SIZE_U32 | 0x1)

arch/s390/include/uapi/asm/sie.h

@@ -230,7 +230,7 @@
  * and returns a key, which can be used to find a mnemonic name
  * of the instruction in the icpt_insn_codes table.
  */
-#define icpt_insn_decoder(insn)			\
+#define icpt_insn_decoder(insn) (		\
 	INSN_DECODE_IPA0(0x01, insn, 48, 0xff)	\
 	INSN_DECODE_IPA0(0xaa, insn, 48, 0x0f)	\
 	INSN_DECODE_IPA0(0xb2, insn, 48, 0xff)	\
@@ -239,6 +239,6 @@
 	INSN_DECODE_IPA0(0xe5, insn, 48, 0xff)	\
 	INSN_DECODE_IPA0(0xeb, insn, 16, 0xff)	\
 	INSN_DECODE_IPA0(0xc8, insn, 48, 0x0f)	\
-	INSN_DECODE(insn)
+	INSN_DECODE(insn))
 
 #endif /* _UAPI_ASM_S390_SIE_H */

arch/s390/kernel/asm-offsets.c

@@ -171,6 +171,7 @@ int main(void)
 #else /* CONFIG_32BIT */
 	DEFINE(__LC_DATA_EXC_CODE, offsetof(struct _lowcore, data_exc_code));
 	DEFINE(__LC_MCCK_FAIL_STOR_ADDR, offsetof(struct _lowcore, failing_storage_address));
+	DEFINE(__LC_VX_SAVE_AREA_ADDR, offsetof(struct _lowcore, vector_save_area_addr));
 	DEFINE(__LC_EXT_PARAMS2, offsetof(struct _lowcore, ext_params2));
 	DEFINE(SAVE_AREA_BASE, offsetof(struct _lowcore, floating_pt_save_area));
 	DEFINE(__LC_PASTE, offsetof(struct _lowcore, paste));

arch/s390/kvm/diag.c

@@ -77,7 +77,7 @@ static int __diag_page_ref_service(struct kvm_vcpu *vcpu)
 
 	if (vcpu->run->s.regs.gprs[rx] & 7)
 		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
-	rc = read_guest(vcpu, vcpu->run->s.regs.gprs[rx], &parm, sizeof(parm));
+	rc = read_guest(vcpu, vcpu->run->s.regs.gprs[rx], rx, &parm, sizeof(parm));
 	if (rc)
 		return kvm_s390_inject_prog_cond(vcpu, rc);
 	if (parm.parm_version != 2 || parm.parm_len < 5 || parm.code != 0x258)
@@ -213,7 +213,7 @@ static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
 	 * - gpr 3 contains the virtqueue index (passed as datamatch)
 	 * - gpr 4 contains the index on the bus (optionally)
 	 */
-	ret = kvm_io_bus_write_cookie(vcpu->kvm, KVM_VIRTIO_CCW_NOTIFY_BUS,
+	ret = kvm_io_bus_write_cookie(vcpu, KVM_VIRTIO_CCW_NOTIFY_BUS,
 				      vcpu->run->s.regs.gprs[2] & 0xffffffff,
 				      8, &vcpu->run->s.regs.gprs[3],
 				      vcpu->run->s.regs.gprs[4]);
@@ -230,7 +230,7 @@ static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
 
 int kvm_s390_handle_diag(struct kvm_vcpu *vcpu)
 {
-	int code = kvm_s390_get_base_disp_rs(vcpu) & 0xffff;
+	int code = kvm_s390_get_base_disp_rs(vcpu, NULL) & 0xffff;
 
 	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
 		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);

arch/s390/kvm/gaccess.h

@@ -156,9 +156,11 @@ int read_guest_lc(struct kvm_vcpu *vcpu, unsigned long gra, void *data,
 }
 
 int guest_translate_address(struct kvm_vcpu *vcpu, unsigned long gva,
-			    unsigned long *gpa, int write);
+			    ar_t ar, unsigned long *gpa, int write);
+int check_gva_range(struct kvm_vcpu *vcpu, unsigned long gva, ar_t ar,
+		    unsigned long length, int is_write);
 
-int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, void *data,
+int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data,
 		 unsigned long len, int write);
 
 int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra,
@@ -168,6 +170,7 @@ int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra,
  * write_guest - copy data from kernel space to guest space
  * @vcpu: virtual cpu
  * @ga: guest address
+ * @ar: access register
  * @data: source address in kernel space
  * @len: number of bytes to copy
  *
@@ -176,8 +179,7 @@ int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra,
  * If DAT is off data will be copied to guest real or absolute memory.
  * If DAT is on data will be copied to the address space as specified by
  * the address space bits of the PSW:
- * Primary, secondory or home space (access register mode is currently not
- * implemented).
+ * Primary, secondary, home space or access register mode.
  * The addressing mode of the PSW is also inspected, so that address wrap
  * around is taken into account for 24-, 31- and 64-bit addressing mode,
  * if the to be copied data crosses page boundaries in guest address space.
@@ -210,16 +212,17 @@ int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra,
  *	 if data has been changed in guest space in case of an exception.
  */
 static inline __must_check
-int write_guest(struct kvm_vcpu *vcpu, unsigned long ga, void *data,
+int write_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data,
 		unsigned long len)
 {
-	return access_guest(vcpu, ga, data, len, 1);
+	return access_guest(vcpu, ga, ar, data, len, 1);
 }
 
 /**
  * read_guest - copy data from guest space to kernel space
  * @vcpu: virtual cpu
  * @ga: guest address
+ * @ar: access register
  * @data: destination address in kernel space
  * @len: number of bytes to copy
  *
@@ -229,10 +232,10 @@ int write_guest(struct kvm_vcpu *vcpu, unsigned long ga, void *data,
  * data will be copied from guest space to kernel space.
  */
 static inline __must_check
-int read_guest(struct kvm_vcpu *vcpu, unsigned long ga, void *data,
+int read_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data,
 	       unsigned long len)
 {
-	return access_guest(vcpu, ga, data, len, 0);
+	return access_guest(vcpu, ga, ar, data, len, 0);
 }
 
 /**
@@ -330,6 +333,6 @@ int read_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, void *data,
 void ipte_lock(struct kvm_vcpu *vcpu);
 void ipte_unlock(struct kvm_vcpu *vcpu);
 int ipte_lock_held(struct kvm_vcpu *vcpu);
-int kvm_s390_check_low_addr_protection(struct kvm_vcpu *vcpu, unsigned long ga);
+int kvm_s390_check_low_addr_prot_real(struct kvm_vcpu *vcpu, unsigned long gra);
 
 #endif /* __KVM_S390_GACCESS_H */

arch/s390/kvm/guestdbg.c

@@ -191,7 +191,7 @@ static int __import_wp_info(struct kvm_vcpu *vcpu,
 	if (!wp_info->old_data)
 		return -ENOMEM;
 	/* try to backup the original value */
-	ret = read_guest(vcpu, wp_info->phys_addr, wp_info->old_data,
+	ret = read_guest_abs(vcpu, wp_info->phys_addr, wp_info->old_data,
 			     wp_info->len);
 	if (ret) {
 		kfree(wp_info->old_data);
@@ -362,7 +362,7 @@ static struct kvm_hw_wp_info_arch *any_wp_changed(struct kvm_vcpu *vcpu)
 			continue;
 
 		/* refetch the wp data and compare it to the old value */
-		if (!read_guest(vcpu, wp_info->phys_addr, temp,
+		if (!read_guest_abs(vcpu, wp_info->phys_addr, temp,
 				    wp_info->len)) {
 			if (memcmp(temp, wp_info->old_data, wp_info->len)) {
 				kfree(temp);

arch/s390/kvm/intercept.c

@@ -165,6 +165,7 @@ static void __extract_prog_irq(struct kvm_vcpu *vcpu,
 		pgm_info->mon_class_nr = vcpu->arch.sie_block->mcn;
 		pgm_info->mon_code = vcpu->arch.sie_block->tecmc;
 		break;
+	case PGM_VECTOR_PROCESSING:
 	case PGM_DATA:
 		pgm_info->data_exc_code = vcpu->arch.sie_block->dxc;
 		break;
@@ -319,7 +320,7 @@ static int handle_mvpg_pei(struct kvm_vcpu *vcpu)
 
 	/* Make sure that the source is paged-in */
 	rc = guest_translate_address(vcpu, vcpu->run->s.regs.gprs[reg2],
-				     &srcaddr, 0);
+				     reg2, &srcaddr, 0);
 	if (rc)
 		return kvm_s390_inject_prog_cond(vcpu, rc);
 	rc = kvm_arch_fault_in_page(vcpu, srcaddr, 0);
@@ -328,7 +329,7 @@ static int handle_mvpg_pei(struct kvm_vcpu *vcpu)
 
 	/* Make sure that the destination is paged-in */
 	rc = guest_translate_address(vcpu, vcpu->run->s.regs.gprs[reg1],
-				     &dstaddr, 1);
+				     reg1, &dstaddr, 1);
 	if (rc)
 		return kvm_s390_inject_prog_cond(vcpu, rc);
 	rc = kvm_arch_fault_in_page(vcpu, dstaddr, 1);

arch/s390/kvm/sigp.c

@@ -393,6 +393,9 @@ static int handle_sigp_order_in_user_space(struct kvm_vcpu *vcpu, u8 order_code)
 	case SIGP_STORE_STATUS_AT_ADDRESS:
 		vcpu->stat.instruction_sigp_store_status++;
 		break;
+	case SIGP_STORE_ADDITIONAL_STATUS:
+		vcpu->stat.instruction_sigp_store_adtl_status++;
+		break;
 	case SIGP_SET_PREFIX:
 		vcpu->stat.instruction_sigp_prefix++;
 		break;
@@ -431,7 +434,7 @@ int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu)
 	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
 		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);
 
-	order_code = kvm_s390_get_base_disp_rs(vcpu);
+	order_code = kvm_s390_get_base_disp_rs(vcpu, NULL);
 	if (handle_sigp_order_in_user_space(vcpu, order_code))
 		return -EOPNOTSUPP;
 
@@ -473,7 +476,7 @@ int kvm_s390_handle_sigp_pei(struct kvm_vcpu *vcpu)
 	int r3 = vcpu->arch.sie_block->ipa & 0x000f;
 	u16 cpu_addr = vcpu->run->s.regs.gprs[r3];
 	struct kvm_vcpu *dest_vcpu;
-	u8 order_code = kvm_s390_get_base_disp_rs(vcpu);
+	u8 order_code = kvm_s390_get_base_disp_rs(vcpu, NULL);
 
 	trace_kvm_s390_handle_sigp_pei(vcpu, order_code, cpu_addr);
 

arch/x86/include/asm/kvm_para.h

@@ -115,7 +115,7 @@ static inline void kvm_spinlock_init(void)
 
 static inline bool kvm_para_available(void)
 {
-	return 0;
+	return false;
 }
 
 static inline unsigned int kvm_arch_para_features(void)

arch/x86/include/asm/pvclock.h

@@ -95,6 +95,7 @@ unsigned __pvclock_read_cycles(const struct pvclock_vcpu_time_info *src,
 
 struct pvclock_vsyscall_time_info {
 	struct pvclock_vcpu_time_info pvti;
+	u32 migrate_count;
 } __attribute__((__aligned__(SMP_CACHE_BYTES)));
 
 #define PVTI_SIZE sizeof(struct pvclock_vsyscall_time_info)