Merge branch 'master' into for-2.6.35

Conflicts:
	fs/ext3/fsync.c

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

3 files changed, 107 insertions, 38 deletions

diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c
index 03a5eb22da2b..7c79c1d76d0c 100644
--- a/virt/kvm/ioapic.c
+++ b/virt/kvm/ioapic.c
@@ -197,7 +197,7 @@ int kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level)
 	union kvm_ioapic_redirect_entry entry;
 	int ret = 1;
 
-	mutex_lock(&ioapic->lock);
+	spin_lock(&ioapic->lock);
 	if (irq >= 0 && irq < IOAPIC_NUM_PINS) {
 		entry = ioapic->redirtbl[irq];
 		level ^= entry.fields.polarity;
@@ -214,7 +214,7 @@ int kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level)
 		}
 		trace_kvm_ioapic_set_irq(entry.bits, irq, ret == 0);
 	}
-	mutex_unlock(&ioapic->lock);
+	spin_unlock(&ioapic->lock);
 
 	return ret;
 }
@@ -238,9 +238,9 @@ static void __kvm_ioapic_update_eoi(struct kvm_ioapic *ioapic, int vector,
 		 * is dropped it will be put into irr and will be delivered
 		 * after ack notifier returns.
 		 */
-		mutex_unlock(&ioapic->lock);
+		spin_unlock(&ioapic->lock);
 		kvm_notify_acked_irq(ioapic->kvm, KVM_IRQCHIP_IOAPIC, i);
-		mutex_lock(&ioapic->lock);
+		spin_lock(&ioapic->lock);
 
 		if (trigger_mode != IOAPIC_LEVEL_TRIG)
 			continue;
@@ -259,9 +259,9 @@ void kvm_ioapic_update_eoi(struct kvm *kvm, int vector, int trigger_mode)
 	smp_rmb();
 	if (!test_bit(vector, ioapic->handled_vectors))
 		return;
-	mutex_lock(&ioapic->lock);
+	spin_lock(&ioapic->lock);
 	__kvm_ioapic_update_eoi(ioapic, vector, trigger_mode);
-	mutex_unlock(&ioapic->lock);
+	spin_unlock(&ioapic->lock);
 }
 
 static inline struct kvm_ioapic *to_ioapic(struct kvm_io_device *dev)
@@ -287,7 +287,7 @@ static int ioapic_mmio_read(struct kvm_io_device *this, gpa_t addr, int len,
 	ASSERT(!(addr & 0xf));	/* check alignment */
 
 	addr &= 0xff;
-	mutex_lock(&ioapic->lock);
+	spin_lock(&ioapic->lock);
 	switch (addr) {
 	case IOAPIC_REG_SELECT:
 		result = ioapic->ioregsel;
@@ -301,7 +301,7 @@ static int ioapic_mmio_read(struct kvm_io_device *this, gpa_t addr, int len,
 		result = 0;
 		break;
 	}
-	mutex_unlock(&ioapic->lock);
+	spin_unlock(&ioapic->lock);
 
 	switch (len) {
 	case 8:
@@ -338,7 +338,7 @@ static int ioapic_mmio_write(struct kvm_io_device *this, gpa_t addr, int len,
 	}
 
 	addr &= 0xff;
-	mutex_lock(&ioapic->lock);
+	spin_lock(&ioapic->lock);
 	switch (addr) {
 	case IOAPIC_REG_SELECT:
 		ioapic->ioregsel = data;
@@ -356,7 +356,7 @@ static int ioapic_mmio_write(struct kvm_io_device *this, gpa_t addr, int len,
 	default:
 		break;
 	}
-	mutex_unlock(&ioapic->lock);
+	spin_unlock(&ioapic->lock);
 	return 0;
 }
 
@@ -386,7 +386,7 @@ int kvm_ioapic_init(struct kvm *kvm)
 	ioapic = kzalloc(sizeof(struct kvm_ioapic), GFP_KERNEL);
 	if (!ioapic)
 		return -ENOMEM;
-	mutex_init(&ioapic->lock);
+	spin_lock_init(&ioapic->lock);
 	kvm->arch.vioapic = ioapic;
 	kvm_ioapic_reset(ioapic);
 	kvm_iodevice_init(&ioapic->dev, &ioapic_mmio_ops);
@@ -419,9 +419,9 @@ int kvm_get_ioapic(struct kvm *kvm, struct kvm_ioapic_state *state)
 	if (!ioapic)
 		return -EINVAL;
 
-	mutex_lock(&ioapic->lock);
+	spin_lock(&ioapic->lock);
 	memcpy(state, ioapic, sizeof(struct kvm_ioapic_state));
-	mutex_unlock(&ioapic->lock);
+	spin_unlock(&ioapic->lock);
 	return 0;
 }
 
@@ -431,9 +431,9 @@ int kvm_set_ioapic(struct kvm *kvm, struct kvm_ioapic_state *state)
 	if (!ioapic)
 		return -EINVAL;
 
-	mutex_lock(&ioapic->lock);
+	spin_lock(&ioapic->lock);
 	memcpy(ioapic, state, sizeof(struct kvm_ioapic_state));
 	update_handled_vectors(ioapic);
-	mutex_unlock(&ioapic->lock);
+	spin_unlock(&ioapic->lock);
 	return 0;
 }
diff --git a/virt/kvm/ioapic.h b/virt/kvm/ioapic.h
index 8a751b78a430..0b190c34ccc3 100644
--- a/virt/kvm/ioapic.h
+++ b/virt/kvm/ioapic.h
@@ -45,7 +45,7 @@ struct kvm_ioapic {
 	struct kvm_io_device dev;
 	struct kvm *kvm;
 	void (*ack_notifier)(void *opaque, int irq);
-	struct mutex lock;
+	spinlock_t lock;
 	DECLARE_BITMAP(handled_vectors, 256);
 };
 
diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c
index 80fd3ad3b2de..11692b9e8830 100644
--- a/virt/kvm/iommu.c
+++ b/virt/kvm/iommu.c
@@ -32,12 +32,30 @@ static int kvm_iommu_unmap_memslots(struct kvm *kvm);
 static void kvm_iommu_put_pages(struct kvm *kvm,
 				gfn_t base_gfn, unsigned long npages);
 
+static pfn_t kvm_pin_pages(struct kvm *kvm, struct kvm_memory_slot *slot,
+			   gfn_t gfn, unsigned long size)
+{
+	gfn_t end_gfn;
+	pfn_t pfn;
+
+	pfn     = gfn_to_pfn_memslot(kvm, slot, gfn);
+	end_gfn = gfn + (size >> PAGE_SHIFT);
+	gfn    += 1;
+
+	if (is_error_pfn(pfn))
+		return pfn;
+
+	while (gfn < end_gfn)
+		gfn_to_pfn_memslot(kvm, slot, gfn++);
+
+	return pfn;
+}
+
 int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot)
 {
-	gfn_t gfn = slot->base_gfn;
-	unsigned long npages = slot->npages;
+	gfn_t gfn, end_gfn;
 	pfn_t pfn;
-	int i, r = 0;
+	int r = 0;
 	struct iommu_domain *domain = kvm->arch.iommu_domain;
 	int flags;
 
@@ -45,31 +63,62 @@ int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot)
 	if (!domain)
 		return 0;
 
+	gfn     = slot->base_gfn;
+	end_gfn = gfn + slot->npages;
+
 	flags = IOMMU_READ | IOMMU_WRITE;
 	if (kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY)
 		flags |= IOMMU_CACHE;
 
-	for (i = 0; i < npages; i++) {
-		/* check if already mapped */
-		if (iommu_iova_to_phys(domain, gfn_to_gpa(gfn)))
+
+	while (gfn < end_gfn) {
+		unsigned long page_size;
+
+		/* Check if already mapped */
+		if (iommu_iova_to_phys(domain, gfn_to_gpa(gfn))) {
+			gfn += 1;
+			continue;
+		}
+
+		/* Get the page size we could use to map */
+		page_size = kvm_host_page_size(kvm, gfn);
+
+		/* Make sure the page_size does not exceed the memslot */
+		while ((gfn + (page_size >> PAGE_SHIFT)) > end_gfn)
+			page_size >>= 1;
+
+		/* Make sure gfn is aligned to the page size we want to map */
+		while ((gfn << PAGE_SHIFT) & (page_size - 1))
+			page_size >>= 1;
+
+		/*
+		 * Pin all pages we are about to map in memory. This is
+		 * important because we unmap and unpin in 4kb steps later.
+		 */
+		pfn = kvm_pin_pages(kvm, slot, gfn, page_size);
+		if (is_error_pfn(pfn)) {
+			gfn += 1;
 			continue;
+		}
 
-		pfn = gfn_to_pfn_memslot(kvm, slot, gfn);
-		r = iommu_map_range(domain,
-				    gfn_to_gpa(gfn),
-				    pfn_to_hpa(pfn),
-				    PAGE_SIZE, flags);
+		/* Map into IO address space */
+		r = iommu_map(domain, gfn_to_gpa(gfn), pfn_to_hpa(pfn),
+			      get_order(page_size), flags);
 		if (r) {
 			printk(KERN_ERR "kvm_iommu_map_address:"
 			       "iommu failed to map pfn=%lx\n", pfn);
 			goto unmap_pages;
 		}
-		gfn++;
+
+		gfn += page_size >> PAGE_SHIFT;
+
+
 	}
+
 	return 0;
 
 unmap_pages:
-	kvm_iommu_put_pages(kvm, slot->base_gfn, i);
+	kvm_iommu_put_pages(kvm, slot->base_gfn, gfn);
 	return r;
 }
 
@@ -189,27 +238,47 @@ out_unmap:
 	return r;
 }
 
+static void kvm_unpin_pages(struct kvm *kvm, pfn_t pfn, unsigned long npages)
+{
+	unsigned long i;
+
+	for (i = 0; i < npages; ++i)
+		kvm_release_pfn_clean(pfn + i);
+}
+
 static void kvm_iommu_put_pages(struct kvm *kvm,
 				gfn_t base_gfn, unsigned long npages)
 {
-	gfn_t gfn = base_gfn;
+	struct iommu_domain *domain;
+	gfn_t end_gfn, gfn;
 	pfn_t pfn;
-	struct iommu_domain *domain = kvm->arch.iommu_domain;
-	unsigned long i;
 	u64 phys;
 
+	domain  = kvm->arch.iommu_domain;
+	end_gfn = base_gfn + npages;
+	gfn     = base_gfn;
+
 	/* check if iommu exists and in use */
 	if (!domain)
 		return;
 
-	for (i = 0; i < npages; i++) {
+	while (gfn < end_gfn) {
+		unsigned long unmap_pages;
+		int order;
+
+		/* Get physical address */
 		phys = iommu_iova_to_phys(domain, gfn_to_gpa(gfn));
-		pfn = phys >> PAGE_SHIFT;
-		kvm_release_pfn_clean(pfn);
-		gfn++;
-	}
+		pfn  = phys >> PAGE_SHIFT;
+
+		/* Unmap address from IO address space */
+		order       = iommu_unmap(domain, gfn_to_gpa(gfn), PAGE_SIZE);
+		unmap_pages = 1ULL << order;
 
-	iommu_unmap_range(domain, gfn_to_gpa(base_gfn), PAGE_SIZE * npages);
+		/* Unpin all pages we just unmapped to not leak any memory */
+		kvm_unpin_pages(kvm, pfn, unmap_pages);
+
+		gfn += unmap_pages;
+	}
 }
 
 static int kvm_iommu_unmap_memslots(struct kvm *kvm)