diff --git a/MAINTAINERS b/MAINTAINERS
index ea44a57..6bef70b 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5114,14 +5114,19 @@
 
 KERNEL VIRTUAL MACHINE (KVM) FOR ARM
 M:	Christoffer Dall <christoffer.dall@linaro.org>
+M:	Marc Zyngier <marc.zyngier@arm.com>
+L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
 L:	kvmarm@lists.cs.columbia.edu
 W:	http://systems.cs.columbia.edu/projects/kvm-arm
 S:	Supported
 F:	arch/arm/include/uapi/asm/kvm*
 F:	arch/arm/include/asm/kvm*
 F:	arch/arm/kvm/
+F:	virt/kvm/arm/
+F:	include/kvm/arm_*
 
 KERNEL VIRTUAL MACHINE FOR ARM64 (KVM/arm64)
+M:	Christoffer Dall <christoffer.dall@linaro.org>
 M:	Marc Zyngier <marc.zyngier@arm.com>
 L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
 L:	kvmarm@lists.cs.columbia.edu
diff --git a/arch/arm/kvm/Kconfig b/arch/arm/kvm/Kconfig
index 466bd29..4be5bb1 100644
--- a/arch/arm/kvm/Kconfig
+++ b/arch/arm/kvm/Kconfig
@@ -23,7 +23,7 @@
 	select HAVE_KVM_CPU_RELAX_INTERCEPT
 	select KVM_MMIO
 	select KVM_ARM_HOST
-	depends on ARM_VIRT_EXT && ARM_LPAE
+	depends on ARM_VIRT_EXT && ARM_LPAE && !CPU_BIG_ENDIAN
 	---help---
 	  Support hosting virtualized guest machines. You will also
 	  need to select one or more of the processor modules below.
diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c
index 80bb1e6..16f8049 100644
--- a/arch/arm/kvm/mmu.c
+++ b/arch/arm/kvm/mmu.c
@@ -42,6 +42,8 @@
 static unsigned long hyp_idmap_end;
 static phys_addr_t hyp_idmap_vector;
 
+#define pgd_order get_order(PTRS_PER_PGD * sizeof(pgd_t))
+
 #define kvm_pmd_huge(_x)	(pmd_huge(_x) || pmd_trans_huge(_x))
 
 static void kvm_tlb_flush_vmid_ipa(struct kvm *kvm, phys_addr_t ipa)
@@ -293,14 +295,14 @@
 	if (boot_hyp_pgd) {
 		unmap_range(NULL, boot_hyp_pgd, hyp_idmap_start, PAGE_SIZE);
 		unmap_range(NULL, boot_hyp_pgd, TRAMPOLINE_VA, PAGE_SIZE);
-		kfree(boot_hyp_pgd);
+		free_pages((unsigned long)boot_hyp_pgd, pgd_order);
 		boot_hyp_pgd = NULL;
 	}
 
 	if (hyp_pgd)
 		unmap_range(NULL, hyp_pgd, TRAMPOLINE_VA, PAGE_SIZE);
 
-	kfree(init_bounce_page);
+	free_page((unsigned long)init_bounce_page);
 	init_bounce_page = NULL;
 
 	mutex_unlock(&kvm_hyp_pgd_mutex);
@@ -330,7 +332,7 @@
 		for (addr = VMALLOC_START; is_vmalloc_addr((void*)addr); addr += PGDIR_SIZE)
 			unmap_range(NULL, hyp_pgd, KERN_TO_HYP(addr), PGDIR_SIZE);
 
-		kfree(hyp_pgd);
+		free_pages((unsigned long)hyp_pgd, pgd_order);
 		hyp_pgd = NULL;
 	}
 
@@ -1024,7 +1026,7 @@
 		size_t len = __hyp_idmap_text_end - __hyp_idmap_text_start;
 		phys_addr_t phys_base;
 
-		init_bounce_page = kmalloc(PAGE_SIZE, GFP_KERNEL);
+		init_bounce_page = (void *)__get_free_page(GFP_KERNEL);
 		if (!init_bounce_page) {
 			kvm_err("Couldn't allocate HYP init bounce page\n");
 			err = -ENOMEM;
@@ -1050,8 +1052,9 @@
 			 (unsigned long)phys_base);
 	}
 
-	hyp_pgd = kzalloc(PTRS_PER_PGD * sizeof(pgd_t), GFP_KERNEL);
-	boot_hyp_pgd = kzalloc(PTRS_PER_PGD * sizeof(pgd_t), GFP_KERNEL);
+	hyp_pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, pgd_order);
+	boot_hyp_pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, pgd_order);
+
 	if (!hyp_pgd || !boot_hyp_pgd) {
 		kvm_err("Hyp mode PGD not allocated\n");
 		err = -ENOMEM;
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 1f68c58..33e8c02 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -503,7 +503,7 @@
 				[number##_HIGH] = VMCS12_OFFSET(name)+4
 
 
-static const unsigned long shadow_read_only_fields[] = {
+static unsigned long shadow_read_only_fields[] = {
 	/*
 	 * We do NOT shadow fields that are modified when L0
 	 * traps and emulates any vmx instruction (e.g. VMPTRLD,
@@ -526,10 +526,10 @@
 	GUEST_LINEAR_ADDRESS,
 	GUEST_PHYSICAL_ADDRESS
 };
-static const int max_shadow_read_only_fields =
+static int max_shadow_read_only_fields =
 	ARRAY_SIZE(shadow_read_only_fields);
 
-static const unsigned long shadow_read_write_fields[] = {
+static unsigned long shadow_read_write_fields[] = {
 	GUEST_RIP,
 	GUEST_RSP,
 	GUEST_CR0,
@@ -558,7 +558,7 @@
 	HOST_FS_SELECTOR,
 	HOST_GS_SELECTOR
 };
-static const int max_shadow_read_write_fields =
+static int max_shadow_read_write_fields =
 	ARRAY_SIZE(shadow_read_write_fields);
 
 static const unsigned short vmcs_field_to_offset_table[] = {
@@ -3009,6 +3009,41 @@
 	}
 }
 
+static void init_vmcs_shadow_fields(void)
+{
+	int i, j;
+
+	/* No checks for read only fields yet */
+
+	for (i = j = 0; i < max_shadow_read_write_fields; i++) {
+		switch (shadow_read_write_fields[i]) {
+		case GUEST_BNDCFGS:
+			if (!vmx_mpx_supported())
+				continue;
+			break;
+		default:
+			break;
+		}
+
+		if (j < i)
+			shadow_read_write_fields[j] =
+				shadow_read_write_fields[i];
+		j++;
+	}
+	max_shadow_read_write_fields = j;
+
+	/* shadowed fields guest access without vmexit */
+	for (i = 0; i < max_shadow_read_write_fields; i++) {
+		clear_bit(shadow_read_write_fields[i],
+			  vmx_vmwrite_bitmap);
+		clear_bit(shadow_read_write_fields[i],
+			  vmx_vmread_bitmap);
+	}
+	for (i = 0; i < max_shadow_read_only_fields; i++)
+		clear_bit(shadow_read_only_fields[i],
+			  vmx_vmread_bitmap);
+}
+
 static __init int alloc_kvm_area(void)
 {
 	int cpu;
@@ -3039,6 +3074,8 @@
 		enable_vpid = 0;
 	if (!cpu_has_vmx_shadow_vmcs())
 		enable_shadow_vmcs = 0;
+	if (enable_shadow_vmcs)
+		init_vmcs_shadow_fields();
 
 	if (!cpu_has_vmx_ept() ||
 	    !cpu_has_vmx_ept_4levels()) {
@@ -8803,14 +8840,6 @@
 
 	memset(vmx_vmread_bitmap, 0xff, PAGE_SIZE);
 	memset(vmx_vmwrite_bitmap, 0xff, PAGE_SIZE);
-	/* shadowed read/write fields */
-	for (i = 0; i < max_shadow_read_write_fields; i++) {
-		clear_bit(shadow_read_write_fields[i], vmx_vmwrite_bitmap);
-		clear_bit(shadow_read_write_fields[i], vmx_vmread_bitmap);
-	}
-	/* shadowed read only fields */
-	for (i = 0; i < max_shadow_read_only_fields; i++)
-		clear_bit(shadow_read_only_fields[i], vmx_vmread_bitmap);
 
 	/*
 	 * Allow direct access to the PC debug port (it is often used for I/O
diff --git a/virt/kvm/arm/vgic.c b/virt/kvm/arm/vgic.c
index 47b2983..56ff9be 100644
--- a/virt/kvm/arm/vgic.c
+++ b/virt/kvm/arm/vgic.c
@@ -548,11 +548,10 @@
 	u32 val;
 	u32 *reg;
 
-	offset >>= 1;
 	reg = vgic_bitmap_get_reg(&vcpu->kvm->arch.vgic.irq_cfg,
-				  vcpu->vcpu_id, offset);
+				  vcpu->vcpu_id, offset >> 1);
 
-	if (offset & 2)
+	if (offset & 4)
 		val = *reg >> 16;
 	else
 		val = *reg & 0xffff;
@@ -561,13 +560,13 @@
 	vgic_reg_access(mmio, &val, offset,
 			ACCESS_READ_VALUE | ACCESS_WRITE_VALUE);
 	if (mmio->is_write) {
-		if (offset < 4) {
+		if (offset < 8) {
 			*reg = ~0U; /* Force PPIs/SGIs to 1 */
 			return false;
 		}
 
 		val = vgic_cfg_compress(val);
-		if (offset & 2) {
+		if (offset & 4) {
 			*reg &= 0xffff;
 			*reg |= val << 16;
 		} else {
@@ -916,6 +915,7 @@
 	case 0:
 		if (!target_cpus)
 			return;
+		break;
 
 	case 1:
 		target_cpus = ((1 << nrcpus) - 1) & ~(1 << vcpu_id) & 0xff;
@@ -1667,10 +1667,11 @@
 	if (addr + size < addr)
 		return -EINVAL;
 
+	*ioaddr = addr;
 	ret = vgic_ioaddr_overlap(kvm);
 	if (ret)
-		return ret;
-	*ioaddr = addr;
+		*ioaddr = VGIC_ADDR_UNDEF;
+
 	return ret;
 }
 
diff --git a/virt/kvm/assigned-dev.c b/virt/kvm/assigned-dev.c
index 8db4370..bf06577 100644
--- a/virt/kvm/assigned-dev.c
+++ b/virt/kvm/assigned-dev.c
@@ -395,7 +395,8 @@
 	if (dev->entries_nr == 0)
 		return r;
 
-	r = pci_enable_msix(dev->dev, dev->host_msix_entries, dev->entries_nr);
+	r = pci_enable_msix_exact(dev->dev,
+				  dev->host_msix_entries, dev->entries_nr);
 	if (r)
 		return r;
 
diff --git a/virt/kvm/async_pf.c b/virt/kvm/async_pf.c
index 10df100..06e6401 100644
--- a/virt/kvm/async_pf.c
+++ b/virt/kvm/async_pf.c
@@ -101,7 +101,7 @@
 	if (waitqueue_active(&vcpu->wq))
 		wake_up_interruptible(&vcpu->wq);
 
-	mmdrop(mm);
+	mmput(mm);
 	kvm_put_kvm(vcpu->kvm);
 }
 
@@ -118,7 +118,7 @@
 		flush_work(&work->work);
 #else
 		if (cancel_work_sync(&work->work)) {
-			mmdrop(work->mm);
+			mmput(work->mm);
 			kvm_put_kvm(vcpu->kvm); /* == work->vcpu->kvm */
 			kmem_cache_free(async_pf_cache, work);
 		}
@@ -183,7 +183,7 @@
 	work->addr = hva;
 	work->arch = *arch;
 	work->mm = current->mm;
-	atomic_inc(&work->mm->mm_count);
+	atomic_inc(&work->mm->mm_users);
 	kvm_get_kvm(work->vcpu->kvm);
 
 	/* this can't really happen otherwise gfn_to_pfn_async
@@ -201,7 +201,7 @@
 	return 1;
 retry_sync:
 	kvm_put_kvm(work->vcpu->kvm);
-	mmdrop(work->mm);
+	mmput(work->mm);
 	kmem_cache_free(async_pf_cache, work);
 	return 0;
 }