7 files changed, 181 insertions, 83 deletions

diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
index f420f1f..4781f80 100644
--- a/drivers/xen/Makefile
+++ b/drivers/xen/Makefile
@@ -4,21 +4,21 @@ obj-y	+= xenbus/
 nostackp := $(call cc-option, -fno-stack-protector)
 CFLAGS_features.o			:= $(nostackp)
 
-obj-$(CONFIG_BLOCK)		+= biomerge.o
-obj-$(CONFIG_HOTPLUG_CPU)	+= cpu_hotplug.o
-obj-$(CONFIG_XEN_XENCOMM)	+= xencomm.o
-obj-$(CONFIG_XEN_BALLOON)	+= xen-balloon.o
-obj-$(CONFIG_XEN_DEV_EVTCHN)	+= xen-evtchn.o
-obj-$(CONFIG_XEN_GNTDEV)	+= xen-gntdev.o
+obj-$(CONFIG_BLOCK)			+= biomerge.o
+obj-$(CONFIG_HOTPLUG_CPU)		+= cpu_hotplug.o
+obj-$(CONFIG_XEN_XENCOMM)		+= xencomm.o
+obj-$(CONFIG_XEN_BALLOON)		+= xen-balloon.o
+obj-$(CONFIG_XEN_DEV_EVTCHN)		+= xen-evtchn.o
+obj-$(CONFIG_XEN_GNTDEV)		+= xen-gntdev.o
 obj-$(CONFIG_XEN_GRANT_DEV_ALLOC)	+= xen-gntalloc.o
-obj-$(CONFIG_XENFS)		+= xenfs/
+obj-$(CONFIG_XENFS)			+= xenfs/
 obj-$(CONFIG_XEN_SYS_HYPERVISOR)	+= sys-hypervisor.o
-obj-$(CONFIG_XEN_PLATFORM_PCI)	+= xen-platform-pci.o
-obj-$(CONFIG_SWIOTLB_XEN)	+= swiotlb-xen.o
-obj-$(CONFIG_XEN_DOM0)		+= pci.o
+obj-$(CONFIG_XEN_PLATFORM_PCI)		+= xen-platform-pci.o
+obj-$(CONFIG_SWIOTLB_XEN)		+= swiotlb-xen.o
+obj-$(CONFIG_XEN_DOM0)			+= pci.o
 
-xen-evtchn-y			:= evtchn.o
+xen-evtchn-y				:= evtchn.o
 xen-gntdev-y				:= gntdev.o
 xen-gntalloc-y				:= gntalloc.o
 
-xen-platform-pci-y		:= platform-pci.o
+xen-platform-pci-y			:= platform-pci.o
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index 043af8a..f54290b 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -114,7 +114,6 @@ static void __balloon_append(struct page *page)
 	if (PageHighMem(page)) {
 		list_add_tail(&page->lru, &ballooned_pages);
 		balloon_stats.balloon_high++;
-		dec_totalhigh_pages();
 	} else {
 		list_add(&page->lru, &ballooned_pages);
 		balloon_stats.balloon_low++;
@@ -124,6 +123,8 @@ static void __balloon_append(struct page *page)
 static void balloon_append(struct page *page)
 {
 	__balloon_append(page);
+	if (PageHighMem(page))
+		dec_totalhigh_pages();
 	totalram_pages--;
 }
 
@@ -193,7 +194,7 @@ static enum bp_state update_schedule(enum bp_state state)
 	return BP_EAGAIN;
 }
 
-static unsigned long current_target(void)
+static long current_credit(void)
 {
 	unsigned long target = balloon_stats.target_pages;
 
@@ -202,7 +203,7 @@ static unsigned long current_target(void)
 		     balloon_stats.balloon_low +
 		     balloon_stats.balloon_high);
 
-	return target;
+	return target - balloon_stats.current_pages;
 }
 
 static enum bp_state increase_reservation(unsigned long nr_pages)
@@ -246,7 +247,7 @@ static enum bp_state increase_reservation(unsigned long nr_pages)
 		set_phys_to_machine(pfn, frame_list[i]);
 
 		/* Link back into the page tables if not highmem. */
-		if (!xen_hvm_domain() && pfn < max_low_pfn) {
+		if (xen_pv_domain() && !PageHighMem(page)) {
 			int ret;
 			ret = HYPERVISOR_update_va_mapping(
 				(unsigned long)__va(pfn << PAGE_SHIFT),
@@ -293,7 +294,7 @@ static enum bp_state decrease_reservation(unsigned long nr_pages, gfp_t gfp)
 
 		scrub_page(page);
 
-		if (!xen_hvm_domain() && !PageHighMem(page)) {
+		if (xen_pv_domain() && !PageHighMem(page)) {
 			ret = HYPERVISOR_update_va_mapping(
 				(unsigned long)__va(pfn << PAGE_SHIFT),
 				__pte_ma(0), 0);
@@ -337,7 +338,7 @@ static void balloon_process(struct work_struct *work)
 	mutex_lock(&balloon_mutex);
 
 	do {
-		credit = current_target() - balloon_stats.current_pages;
+		credit = current_credit();
 
 		if (credit > 0)
 			state = increase_reservation(credit);
@@ -420,7 +421,7 @@ void free_xenballooned_pages(int nr_pages, struct page** pages)
 	}
 
 	/* The balloon may be too large now. Shrink it if needed. */
-	if (current_target() != balloon_stats.current_pages)
+	if (current_credit())
 		schedule_delayed_work(&balloon_worker, 0);
 
 	mutex_unlock(&balloon_mutex);
@@ -429,7 +430,7 @@ EXPORT_SYMBOL(free_xenballooned_pages);
 
 static int __init balloon_init(void)
 {
- 	unsigned long pfn, nr_pages, extra_pfn_end;
+	unsigned long pfn, extra_pfn_end;
 	struct page *page;
 
 	if (!xen_domain())
@@ -437,11 +438,7 @@ static int __init balloon_init(void)
 
 	pr_info("xen/balloon: Initialising balloon driver.\n");
 
- 	if (xen_pv_domain())
- 		nr_pages = xen_start_info->nr_pages;
- 	else
- 		nr_pages = max_pfn;
- 	balloon_stats.current_pages = min(nr_pages, max_pfn);
+	balloon_stats.current_pages = xen_pv_domain() ? min(xen_start_info->nr_pages, max_pfn) : max_pfn;
 	balloon_stats.target_pages  = balloon_stats.current_pages;
 	balloon_stats.balloon_low   = 0;
 	balloon_stats.balloon_high  = 0;
@@ -466,7 +463,7 @@ static int __init balloon_init(void)
 	     pfn < extra_pfn_end;
 	     pfn++) {
 		page = pfn_to_page(pfn);
-		/* totalram_pages doesn't include the boot-time
+		/* totalram_pages and totalhigh_pages do not include the boot-time
 		   balloon extension, so don't subtract from it. */
 		__balloon_append(page);
 	}
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 33167b4..3ff822b 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -101,6 +101,7 @@ struct irq_info
 			unsigned short gsi;
 			unsigned char vector;
 			unsigned char flags;
+			uint16_t domid;
 		} pirq;
 	} u;
 };
@@ -118,6 +119,8 @@ static DEFINE_PER_CPU(unsigned long [NR_EVENT_CHANNELS/BITS_PER_LONG],
 static struct irq_chip xen_dynamic_chip;
 static struct irq_chip xen_percpu_chip;
 static struct irq_chip xen_pirq_chip;
+static void enable_dynirq(struct irq_data *data);
+static void disable_dynirq(struct irq_data *data);
 
 /* Get info for IRQ */
 static struct irq_info *info_for_irq(unsigned irq)
@@ -184,6 +187,7 @@ static void xen_irq_info_pirq_init(unsigned irq,
 				   unsigned short pirq,
 				   unsigned short gsi,
 				   unsigned short vector,
+				   uint16_t domid,
 				   unsigned char flags)
 {
 	struct irq_info *info = info_for_irq(irq);
@@ -193,6 +197,7 @@ static void xen_irq_info_pirq_init(unsigned irq,
 	info->u.pirq.pirq = pirq;
 	info->u.pirq.gsi = gsi;
 	info->u.pirq.vector = vector;
+	info->u.pirq.domid = domid;
 	info->u.pirq.flags = flags;
 }
 
@@ -473,16 +478,6 @@ static void xen_free_irq(unsigned irq)
 	irq_free_desc(irq);
 }
 
-static void pirq_unmask_notify(int irq)
-{
-	struct physdev_eoi eoi = { .irq = pirq_from_irq(irq) };
-
-	if (unlikely(pirq_needs_eoi(irq))) {
-		int rc = HYPERVISOR_physdev_op(PHYSDEVOP_eoi, &eoi);
-		WARN_ON(rc);
-	}
-}
-
 static void pirq_query_unmask(int irq)
 {
 	struct physdev_irq_status_query irq_status;
@@ -506,6 +501,29 @@ static bool probing_irq(int irq)
 	return desc && desc->action == NULL;
 }
 
+static void eoi_pirq(struct irq_data *data)
+{
+	int evtchn = evtchn_from_irq(data->irq);
+	struct physdev_eoi eoi = { .irq = pirq_from_irq(data->irq) };
+	int rc = 0;
+
+	irq_move_irq(data);
+
+	if (VALID_EVTCHN(evtchn))
+		clear_evtchn(evtchn);
+
+	if (pirq_needs_eoi(data->irq)) {
+		rc = HYPERVISOR_physdev_op(PHYSDEVOP_eoi, &eoi);
+		WARN_ON(rc);
+	}
+}
+
+static void mask_ack_pirq(struct irq_data *data)
+{
+	disable_dynirq(data);
+	eoi_pirq(data);
+}
+
 static unsigned int __startup_pirq(unsigned int irq)
 {
 	struct evtchn_bind_pirq bind_pirq;
@@ -539,7 +557,7 @@ static unsigned int __startup_pirq(unsigned int irq)
 
 out:
 	unmask_evtchn(evtchn);
-	pirq_unmask_notify(irq);
+	eoi_pirq(irq_get_irq_data(irq));
 
 	return 0;
 }
@@ -579,18 +597,7 @@ static void enable_pirq(struct irq_data *data)
 
 static void disable_pirq(struct irq_data *data)
 {
-}
-
-static void ack_pirq(struct irq_data *data)
-{
-	int evtchn = evtchn_from_irq(data->irq);
-
-	irq_move_irq(data);
-
-	if (VALID_EVTCHN(evtchn)) {
-		mask_evtchn(evtchn);
-		clear_evtchn(evtchn);
-	}
+	disable_dynirq(data);
 }
 
 static int find_irq_by_gsi(unsigned gsi)
@@ -639,9 +646,6 @@ int xen_bind_pirq_gsi_to_irq(unsigned gsi,
 	if (irq < 0)
 		goto out;
 
-	irq_set_chip_and_handler_name(irq, &xen_pirq_chip, handle_level_irq,
-				      name);
-
 	irq_op.irq = irq;
 	irq_op.vector = 0;
 
@@ -655,9 +659,35 @@ int xen_bind_pirq_gsi_to_irq(unsigned gsi,
 		goto out;
 	}
 
-	xen_irq_info_pirq_init(irq, 0, pirq, gsi, irq_op.vector,
+	xen_irq_info_pirq_init(irq, 0, pirq, gsi, irq_op.vector, DOMID_SELF,
 			       shareable ? PIRQ_SHAREABLE : 0);
 
+	pirq_query_unmask(irq);
+	/* We try to use the handler with the appropriate semantic for the
+	 * type of interrupt: if the interrupt doesn't need an eoi
+	 * (pirq_needs_eoi returns false), we treat it like an edge
+	 * triggered interrupt so we use handle_edge_irq.
+	 * As a matter of fact this only happens when the corresponding
+	 * physical interrupt is edge triggered or an msi.
+	 *
+	 * On the other hand if the interrupt needs an eoi (pirq_needs_eoi
+	 * returns true) we treat it like a level triggered interrupt so we
+	 * use handle_fasteoi_irq like the native code does for this kind of
+	 * interrupts.
+	 * Depending on the Xen version, pirq_needs_eoi might return true
+	 * not only for level triggered interrupts but for edge triggered
+	 * interrupts too. In any case Xen always honors the eoi mechanism,
+	 * not injecting any more pirqs of the same kind if the first one
+	 * hasn't received an eoi yet. Therefore using the fasteoi handler
+	 * is the right choice either way.
+	 */
+	if (pirq_needs_eoi(irq))
+		irq_set_chip_and_handler_name(irq, &xen_pirq_chip,
+				handle_fasteoi_irq, name);
+	else
+		irq_set_chip_and_handler_name(irq, &xen_pirq_chip,
+				handle_edge_irq, name);
+
 out:
 	spin_unlock(&irq_mapping_update_lock);
 
@@ -680,7 +710,8 @@ int xen_allocate_pirq_msi(struct pci_dev *dev, struct msi_desc *msidesc)
 }
 
 int xen_bind_pirq_msi_to_irq(struct pci_dev *dev, struct msi_desc *msidesc,
-			     int pirq, int vector, const char *name)
+			     int pirq, int vector, const char *name,
+			     domid_t domid)
 {
 	int irq, ret;
 
@@ -690,10 +721,10 @@ int xen_bind_pirq_msi_to_irq(struct pci_dev *dev, struct msi_desc *msidesc,
 	if (irq == -1)
 		goto out;
 
-	irq_set_chip_and_handler_name(irq, &xen_pirq_chip, handle_level_irq,
-				      name);
+	irq_set_chip_and_handler_name(irq, &xen_pirq_chip, handle_edge_irq,
+			name);
 
-	xen_irq_info_pirq_init(irq, 0, pirq, 0, vector, 0);
+	xen_irq_info_pirq_init(irq, 0, pirq, 0, vector, domid, 0);
 	ret = irq_set_msi_desc(irq, msidesc);
 	if (ret < 0)
 		goto error_irq;
@@ -722,9 +753,16 @@ int xen_destroy_irq(int irq)
 
 	if (xen_initial_domain()) {
 		unmap_irq.pirq = info->u.pirq.pirq;
-		unmap_irq.domid = DOMID_SELF;
+		unmap_irq.domid = info->u.pirq.domid;
 		rc = HYPERVISOR_physdev_op(PHYSDEVOP_unmap_pirq, &unmap_irq);
-		if (rc) {
+		/* If another domain quits without making the pci_disable_msix
+		 * call, the Xen hypervisor takes care of freeing the PIRQs
+		 * (free_domain_pirqs).
+		 */
+		if ((rc == -ESRCH && info->u.pirq.domid != DOMID_SELF))
+			printk(KERN_INFO "domain %d does not have %d anymore\n",
+				info->u.pirq.domid, info->u.pirq.pirq);
+		else if (rc) {
 			printk(KERN_WARNING "unmap irq failed %d\n", rc);
 			goto out;
 		}
@@ -759,6 +797,12 @@ out:
 	return irq;
 }
 
+
+int xen_pirq_from_irq(unsigned irq)
+{
+	return pirq_from_irq(irq);
+}
+EXPORT_SYMBOL_GPL(xen_pirq_from_irq);
 int bind_evtchn_to_irq(unsigned int evtchn)
 {
 	int irq;
@@ -773,7 +817,7 @@ int bind_evtchn_to_irq(unsigned int evtchn)
 			goto out;
 
 		irq_set_chip_and_handler_name(irq, &xen_dynamic_chip,
-					      handle_fasteoi_irq, "event");
+					      handle_edge_irq, "event");
 
 		xen_irq_info_evtchn_init(irq, evtchn);
 	}
@@ -1179,9 +1223,6 @@ static void __xen_evtchn_do_upcall(void)
 				port = (word_idx * BITS_PER_LONG) + bit_idx;
 				irq = evtchn_to_irq[port];
 
-				mask_evtchn(port);
-				clear_evtchn(port);
-
 				if (irq != -1) {
 					desc = irq_to_desc(irq);
 					if (desc)
@@ -1337,10 +1378,16 @@ static void ack_dynirq(struct irq_data *data)
 {
 	int evtchn = evtchn_from_irq(data->irq);
 
-	irq_move_masked_irq(data);
+	irq_move_irq(data);
 
 	if (VALID_EVTCHN(evtchn))
-		unmask_evtchn(evtchn);
+		clear_evtchn(evtchn);
+}
+
+static void mask_ack_dynirq(struct irq_data *data)
+{
+	disable_dynirq(data);
+	ack_dynirq(data);
 }
 
 static int retrigger_dynirq(struct irq_data *data)
@@ -1502,6 +1549,18 @@ void xen_poll_irq(int irq)
 	xen_poll_irq_timeout(irq, 0 /* no timeout */);
 }
 
+/* Check whether the IRQ line is shared with other guests. */
+int xen_test_irq_shared(int irq)
+{
+	struct irq_info *info = info_for_irq(irq);
+	struct physdev_irq_status_query irq_status = { .irq = info->u.pirq.pirq };
+
+	if (HYPERVISOR_physdev_op(PHYSDEVOP_irq_status_query, &irq_status))
+		return 0;
+	return !(irq_status.flags & XENIRQSTAT_shared);
+}
+EXPORT_SYMBOL_GPL(xen_test_irq_shared);
+
 void xen_irq_resume(void)
 {
 	unsigned int cpu, evtchn;
@@ -1535,7 +1594,9 @@ static struct irq_chip xen_dynamic_chip __read_mostly = {
 	.irq_mask		= disable_dynirq,
 	.irq_unmask		= enable_dynirq,
 
-	.irq_eoi		= ack_dynirq,
+	.irq_ack		= ack_dynirq,
+	.irq_mask_ack		= mask_ack_dynirq,
+
 	.irq_set_affinity	= set_affinity_irq,
 	.irq_retrigger		= retrigger_dynirq,
 };
@@ -1545,14 +1606,15 @@ static struct irq_chip xen_pirq_chip __read_mostly = {
 
 	.irq_startup		= startup_pirq,
 	.irq_shutdown		= shutdown_pirq,
-
 	.irq_enable		= enable_pirq,
-	.irq_unmask		= enable_pirq,
-
 	.irq_disable		= disable_pirq,
-	.irq_mask		= disable_pirq,
 
-	.irq_ack		= ack_pirq,
+	.irq_mask		= disable_dynirq,
+	.irq_unmask		= enable_dynirq,
+
+	.irq_ack		= eoi_pirq,
+	.irq_eoi		= eoi_pirq,
+	.irq_mask_ack		= mask_ack_pirq,
 
 	.irq_set_affinity	= set_affinity_irq,
 
diff --git a/drivers/xen/gntalloc.c b/drivers/xen/gntalloc.c
index a7ffdfe..f6832f4 100644
--- a/drivers/xen/gntalloc.c
+++ b/drivers/xen/gntalloc.c
@@ -427,6 +427,17 @@ static long gntalloc_ioctl(struct file *filp, unsigned int cmd,
 	return 0;
 }
 
+static void gntalloc_vma_open(struct vm_area_struct *vma)
+{
+	struct gntalloc_gref *gref = vma->vm_private_data;
+	if (!gref)
+		return;
+
+	spin_lock(&gref_lock);
+	gref->users++;
+	spin_unlock(&gref_lock);
+}
+
 static void gntalloc_vma_close(struct vm_area_struct *vma)
 {
 	struct gntalloc_gref *gref = vma->vm_private_data;
@@ -441,6 +452,7 @@ static void gntalloc_vma_close(struct vm_area_struct *vma)
 }
 
 static struct vm_operations_struct gntalloc_vmops = {
+	.open = gntalloc_vma_open,
 	.close = gntalloc_vma_close,
 };
 
@@ -471,8 +483,6 @@ static int gntalloc_mmap(struct file *filp, struct vm_area_struct *vma)
 	vma->vm_private_data = gref;
 
 	vma->vm_flags |= VM_RESERVED;
-	vma->vm_flags |= VM_DONTCOPY;
-	vma->vm_flags |= VM_PFNMAP | VM_PFN_AT_MMAP;
 
 	vma->vm_ops = &gntalloc_vmops;
 
diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c
index b0f9e8f..f914b26 100644
--- a/drivers/xen/gntdev.c
+++ b/drivers/xen/gntdev.c
@@ -330,17 +330,26 @@ static int unmap_grant_pages(struct grant_map *map, int offset, int pages)
 
 /* ------------------------------------------------------------------ */
 
+static void gntdev_vma_open(struct vm_area_struct *vma)
+{
+	struct grant_map *map = vma->vm_private_data;
+
+	pr_debug("gntdev_vma_open %p\n", vma);
+	atomic_inc(&map->users);
+}
+
 static void gntdev_vma_close(struct vm_area_struct *vma)
 {
 	struct grant_map *map = vma->vm_private_data;
 
-	pr_debug("close %p\n", vma);
+	pr_debug("gntdev_vma_close %p\n", vma);
 	map->vma = NULL;
 	vma->vm_private_data = NULL;
 	gntdev_put_map(map);
 }
 
 static struct vm_operations_struct gntdev_vmops = {
+	.open = gntdev_vma_open,
 	.close = gntdev_vma_close,
 };
 
@@ -652,7 +661,10 @@ static int gntdev_mmap(struct file *flip, struct vm_area_struct *vma)
 
 	vma->vm_ops = &gntdev_vmops;
 
-	vma->vm_flags |= VM_RESERVED|VM_DONTCOPY|VM_DONTEXPAND|VM_PFNMAP;
+	vma->vm_flags |= VM_RESERVED|VM_DONTEXPAND;
+
+	if (use_ptemod)
+		vma->vm_flags |= VM_DONTCOPY|VM_PFNMAP;
 
 	vma->vm_private_data = map;
 
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index 3745a31..fd725cd 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -466,13 +466,30 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
 		if (map_ops[i].status)
 			continue;
 
-		/* m2p override only supported for GNTMAP_contains_pte mappings */
-		if (!(map_ops[i].flags & GNTMAP_contains_pte))
-			continue;
-		pte = (pte_t *) (mfn_to_virt(PFN_DOWN(map_ops[i].host_addr)) +
+		if (map_ops[i].flags & GNTMAP_contains_pte) {
+			pte = (pte_t *) (mfn_to_virt(PFN_DOWN(map_ops[i].host_addr)) +
 				(map_ops[i].host_addr & ~PAGE_MASK));
-		mfn = pte_mfn(*pte);
-		ret = m2p_add_override(mfn, pages[i]);
+			mfn = pte_mfn(*pte);
+		} else {
+			/* If you really wanted to do this:
+			 * mfn = PFN_DOWN(map_ops[i].dev_bus_addr);
+			 *
+			 * The reason we do not implement it is b/c on the
+			 * unmap path (gnttab_unmap_refs) we have no means of
+			 * checking whether the page is !GNTMAP_contains_pte.
+			 *
+			 * That is without some extra data-structure to carry
+			 * the struct page, bool clear_pte, and list_head next
+			 * tuples and deal with allocation/delallocation, etc.
+			 *
+			 * The users of this API set the GNTMAP_contains_pte
+			 * flag so lets just return not supported until it
+			 * becomes neccessary to implement.
+			 */
+			return -EOPNOTSUPP;
+		}
+		ret = m2p_add_override(mfn, pages[i],
+				       map_ops[i].flags & GNTMAP_contains_pte);
 		if (ret)
 			return ret;
 	}
@@ -494,7 +511,7 @@ int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
 		return ret;
 
 	for (i = 0; i < count; i++) {
-		ret = m2p_remove_override(pages[i]);
+		ret = m2p_remove_override(pages[i], true /* clear the PTE */);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/xen/sys-hypervisor.c b/drivers/xen/sys-hypervisor.c
index 60f1827..1e0fe01 100644
--- a/drivers/xen/sys-hypervisor.c
+++ b/drivers/xen/sys-hypervisor.c
@@ -215,7 +215,7 @@ static struct attribute_group xen_compilation_group = {
 	.attrs = xen_compile_attrs,
 };
 
-int __init static xen_compilation_init(void)
+static int __init xen_compilation_init(void)
 {
 	return sysfs_create_group(hypervisor_kobj, &xen_compilation_group);
 }