Merge commit 'konrad/stable/xen-pcifront-0.8.2' into 2.6.36-rc8-initial-domain-v6

8 files changed, 1239 insertions, 227 deletions

diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig
index 34ef70d562b..5b1630e4e9e 100644
--- a/drivers/pci/Kconfig
+++ b/drivers/pci/Kconfig
@@ -40,6 +40,27 @@ config PCI_STUB
 
 	  When in doubt, say N.
 
+config XEN_PCIDEV_FRONTEND
+        tristate "Xen PCI Frontend"
+        depends on PCI && X86 && XEN
+        select HOTPLUG
+        select PCI_XEN
+        default y
+        help
+          The PCI device frontend driver allows the kernel to import arbitrary
+          PCI devices from a PCI backend to support PCI driver domains.
+
+config XEN_PCIDEV_FE_DEBUG
+        bool "Xen PCI Frontend debugging"
+        depends on XEN_PCIDEV_FRONTEND && PCI_DEBUG
+	help
+	  Say Y here if you want the Xen PCI frontend to produce a bunch of debug
+	  messages to the system log.  Select this if you are having a
+	  problem with Xen PCI frontend support and want to see more of what is
+	  going on.
+
+	  When in doubt, say N.
+
 config HT_IRQ
 	bool "Interrupts on hypertransport devices"
 	default y
diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile
index dc1aa092286..d5e27050c4e 100644
--- a/drivers/pci/Makefile
+++ b/drivers/pci/Makefile
@@ -65,6 +65,8 @@ obj-$(CONFIG_PCI_SYSCALL) += syscall.o
 
 obj-$(CONFIG_PCI_STUB) += pci-stub.o
 
+obj-$(CONFIG_XEN_PCIDEV_FRONTEND) += xen-pcifront.o
+
 ifeq ($(CONFIG_PCI_DEBUG),y)
 EXTRA_CFLAGS += -DDEBUG
 endif
diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c
index 7f0af0e9b82..69546e9213d 100644
--- a/drivers/pci/bus.c
+++ b/drivers/pci/bus.c
@@ -299,6 +299,7 @@ void pci_walk_bus(struct pci_bus *top, int (*cb)(struct pci_dev *, void *),
 	}
 	up_read(&pci_bus_sem);
 }
+EXPORT_SYMBOL_GPL(pci_walk_bus);
 
 EXPORT_SYMBOL(pci_bus_alloc_resource);
 EXPORT_SYMBOL_GPL(pci_bus_add_device);
diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 0a19708074c..3de3a436a43 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -1221,9 +1221,9 @@ const char *dmar_get_fault_reason(u8 fault_reason, int *fault_type)
 	}
 }
 
-void dmar_msi_unmask(unsigned int irq)
+void dmar_msi_unmask(struct irq_data *data)
 {
-	struct intel_iommu *iommu = get_irq_data(irq);
+	struct intel_iommu *iommu = irq_data_get_irq_data(data);
 	unsigned long flag;
 
 	/* unmask it */
@@ -1234,10 +1234,10 @@ void dmar_msi_unmask(unsigned int irq)
 	spin_unlock_irqrestore(&iommu->register_lock, flag);
 }
 
-void dmar_msi_mask(unsigned int irq)
+void dmar_msi_mask(struct irq_data *data)
 {
 	unsigned long flag;
-	struct intel_iommu *iommu = get_irq_data(irq);
+	struct intel_iommu *iommu = irq_data_get_irq_data(data);
 
 	/* mask it */
 	spin_lock_irqsave(&iommu->register_lock, flag);
diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c
index 98abf8b9129..834842aa5bb 100644
--- a/drivers/pci/htirq.c
+++ b/drivers/pci/htirq.c
@@ -57,28 +57,22 @@ void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg)
 	*msg = cfg->msg;
 }
 
-void mask_ht_irq(unsigned int irq)
+void mask_ht_irq(struct irq_data *data)
 {
-	struct ht_irq_cfg *cfg;
-	struct ht_irq_msg msg;
-
-	cfg = get_irq_data(irq);
+	struct ht_irq_cfg *cfg = irq_data_get_irq_data(data);
+	struct ht_irq_msg msg = cfg->msg;
 
-	msg = cfg->msg;
 	msg.address_lo |= 1;
-	write_ht_irq_msg(irq, &msg);
+	write_ht_irq_msg(data->irq, &msg);
 }
 
-void unmask_ht_irq(unsigned int irq)
+void unmask_ht_irq(struct irq_data *data)
 {
-	struct ht_irq_cfg *cfg;
-	struct ht_irq_msg msg;
-
-	cfg = get_irq_data(irq);
+	struct ht_irq_cfg *cfg = irq_data_get_irq_data(data);
+	struct ht_irq_msg msg = cfg->msg;
 
-	msg = cfg->msg;
 	msg.address_lo &= ~1;
-	write_ht_irq_msg(irq, &msg);
+	write_ht_irq_msg(data->irq, &msg);
 }
 
 /**
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index fd1d2867cdc..ec87cd66f3e 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -46,109 +46,24 @@ static __init int setup_intremap(char *str)
 }
 early_param("intremap", setup_intremap);
 
-struct irq_2_iommu {
-	struct intel_iommu *iommu;
-	u16 irte_index;
-	u16 sub_handle;
-	u8  irte_mask;
-};
-
-#ifdef CONFIG_GENERIC_HARDIRQS
-static struct irq_2_iommu *get_one_free_irq_2_iommu(int node)
-{
-	struct irq_2_iommu *iommu;
-
-	iommu = kzalloc_node(sizeof(*iommu), GFP_ATOMIC, node);
-	printk(KERN_DEBUG "alloc irq_2_iommu on node %d\n", node);
-
-	return iommu;
-}
-
-static struct irq_2_iommu *irq_2_iommu(unsigned int irq)
-{
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
-
-	if (WARN_ON_ONCE(!desc))
-		return NULL;
-
-	return desc->irq_2_iommu;
-}
-
-static struct irq_2_iommu *irq_2_iommu_alloc(unsigned int irq)
-{
-	struct irq_desc *desc;
-	struct irq_2_iommu *irq_iommu;
-
-	desc = irq_to_desc(irq);
-	if (!desc) {
-		printk(KERN_INFO "can not get irq_desc for %d\n", irq);
-		return NULL;
-	}
-
-	irq_iommu = desc->irq_2_iommu;
-
-	if (!irq_iommu)
-		desc->irq_2_iommu = get_one_free_irq_2_iommu(irq_node(irq));
-
-	return desc->irq_2_iommu;
-}
-
-#else /* !CONFIG_SPARSE_IRQ */
-
-static struct irq_2_iommu irq_2_iommuX[NR_IRQS];
-
-static struct irq_2_iommu *irq_2_iommu(unsigned int irq)
-{
-	if (irq < nr_irqs)
-		return &irq_2_iommuX[irq];
-
-	return NULL;
-}
-static struct irq_2_iommu *irq_2_iommu_alloc(unsigned int irq)
-{
-	return irq_2_iommu(irq);
-}
-#endif
-
 static DEFINE_SPINLOCK(irq_2_ir_lock);
 
-static struct irq_2_iommu *valid_irq_2_iommu(unsigned int irq)
-{
-	struct irq_2_iommu *irq_iommu;
-
-	irq_iommu = irq_2_iommu(irq);
-
-	if (!irq_iommu)
-		return NULL;
-
-	if (!irq_iommu->iommu)
-		return NULL;
-
-	return irq_iommu;
-}
-
-int irq_remapped(int irq)
+static struct irq_2_iommu *irq_2_iommu(unsigned int irq)
 {
-	return valid_irq_2_iommu(irq) != NULL;
+	struct irq_cfg *cfg = get_irq_chip_data(irq);
+	return cfg ? &cfg->irq_2_iommu : NULL;
 }
 
 int get_irte(int irq, struct irte *entry)
 {
-	int index;
-	struct irq_2_iommu *irq_iommu;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	unsigned long flags;
+	int index;
 
-	if (!entry)
+	if (!entry || !irq_iommu)
 		return -1;
 
 	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
 
 	index = irq_iommu->irte_index + irq_iommu->sub_handle;
 	*entry = *(irq_iommu->iommu->ir_table->base + index);
@@ -160,20 +75,14 @@ int get_irte(int irq, struct irte *entry)
 int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 {
 	struct ir_table *table = iommu->ir_table;
-	struct irq_2_iommu *irq_iommu;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	u16 index, start_index;
 	unsigned int mask = 0;
 	unsigned long flags;
 	int i;
 
-	if (!count)
-		return -1;
-
-#ifndef CONFIG_SPARSE_IRQ
-	/* protect irq_2_iommu_alloc later */
-	if (irq >= nr_irqs)
+	if (!count || !irq_iommu)
 		return -1;
-#endif
 
 	/*
 	 * start the IRTE search from index 0.
@@ -214,13 +123,6 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 	for (i = index; i < index + count; i++)
 		table->base[i].present = 1;
 
-	irq_iommu = irq_2_iommu_alloc(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		printk(KERN_ERR "can't allocate irq_2_iommu\n");
-		return -1;
-	}
-
 	irq_iommu->iommu = iommu;
 	irq_iommu->irte_index =  index;
 	irq_iommu->sub_handle = 0;
@@ -244,17 +146,14 @@ static int qi_flush_iec(struct intel_iommu *iommu, int index, int mask)
 
 int map_irq_to_irte_handle(int irq, u16 *sub_handle)
 {
-	int index;
-	struct irq_2_iommu *irq_iommu;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	unsigned long flags;
+	int index;
 
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
+	if (!irq_iommu)
 		return -1;
-	}
 
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 	*sub_handle = irq_iommu->sub_handle;
 	index = irq_iommu->irte_index;
 	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
@@ -263,18 +162,13 @@ int map_irq_to_irte_handle(int irq, u16 *sub_handle)
 
 int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle)
 {
-	struct irq_2_iommu *irq_iommu;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	unsigned long flags;
 
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-
-	irq_iommu = irq_2_iommu_alloc(irq);
-
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		printk(KERN_ERR "can't allocate irq_2_iommu\n");
+	if (!irq_iommu)
 		return -1;
-	}
+
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 
 	irq_iommu->iommu = iommu;
 	irq_iommu->irte_index = index;
@@ -286,43 +180,18 @@ int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle)
 	return 0;
 }
 
-int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index)
-{
-	struct irq_2_iommu *irq_iommu;
-	unsigned long flags;
-
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
-
-	irq_iommu->iommu = NULL;
-	irq_iommu->irte_index = 0;
-	irq_iommu->sub_handle = 0;
-	irq_2_iommu(irq)->irte_mask = 0;
-
-	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-
-	return 0;
-}
-
 int modify_irte(int irq, struct irte *irte_modified)
 {
-	int rc;
-	int index;
-	struct irte *irte;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	struct intel_iommu *iommu;
-	struct irq_2_iommu *irq_iommu;
 	unsigned long flags;
+	struct irte *irte;
+	int rc, index;
 
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
+	if (!irq_iommu)
 		return -1;
-	}
+
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 
 	iommu = irq_iommu->iommu;
 
@@ -339,31 +208,6 @@ int modify_irte(int irq, struct irte *irte_modified)
 	return rc;
 }
 
-int flush_irte(int irq)
-{
-	int rc;
-	int index;
-	struct intel_iommu *iommu;
-	struct irq_2_iommu *irq_iommu;
-	unsigned long flags;
-
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-		return -1;
-	}
-
-	iommu = irq_iommu->iommu;
-
-	index = irq_iommu->irte_index + irq_iommu->sub_handle;
-
-	rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask);
-	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
-
-	return rc;
-}
-
 struct intel_iommu *map_hpet_to_ir(u8 hpet_id)
 {
 	int i;
@@ -420,16 +264,14 @@ static int clear_entries(struct irq_2_iommu *irq_iommu)
 
 int free_irte(int irq)
 {
-	int rc = 0;
-	struct irq_2_iommu *irq_iommu;
+	struct irq_2_iommu *irq_iommu = irq_2_iommu(irq);
 	unsigned long flags;
+	int rc;
 
-	spin_lock_irqsave(&irq_2_ir_lock, flags);
-	irq_iommu = valid_irq_2_iommu(irq);
-	if (!irq_iommu) {
-		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
+	if (!irq_iommu)
 		return -1;
-	}
+
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 
 	rc = clear_entries(irq_iommu);
 
diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 69b7be33b3a..7c24dcef298 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -35,7 +35,12 @@ int arch_msi_check_device(struct pci_dev *dev, int nvec, int type)
 #endif
 
 #ifndef arch_setup_msi_irqs
-int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
+# define arch_setup_msi_irqs default_setup_msi_irqs
+# define HAVE_DEFAULT_MSI_SETUP_IRQS
+#endif
+
+#ifdef HAVE_DEFAULT_MSI_SETUP_IRQS
+int default_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 {
 	struct msi_desc *entry;
 	int ret;
@@ -60,7 +65,12 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 #endif
 
 #ifndef arch_teardown_msi_irqs
-void arch_teardown_msi_irqs(struct pci_dev *dev)
+# define arch_teardown_msi_irqs default_teardown_msi_irqs
+# define HAVE_DEFAULT_MSI_TEARDOWN_IRQS
+#endif
+
+#ifdef HAVE_DEFAULT_MSI_TEARDOWN_IRQS
+void default_teardown_msi_irqs(struct pci_dev *dev)
 {
 	struct msi_desc *entry;
 
@@ -170,33 +180,31 @@ static void msix_mask_irq(struct msi_desc *desc, u32 flag)
 	desc->masked = __msix_mask_irq(desc, flag);
 }
 
-static void msi_set_mask_bit(unsigned irq, u32 flag)
+static void msi_set_mask_bit(struct irq_data *data, u32 flag)
 {
-	struct msi_desc *desc = get_irq_msi(irq);
+	struct msi_desc *desc = irq_data_get_msi(data);
 
 	if (desc->msi_attrib.is_msix) {
 		msix_mask_irq(desc, flag);
 		readl(desc->mask_base);		/* Flush write to device */
 	} else {
-		unsigned offset = irq - desc->dev->irq;
+		unsigned offset = data->irq - desc->dev->irq;
 		msi_mask_irq(desc, 1 << offset, flag << offset);
 	}
 }
 
-void mask_msi_irq(unsigned int irq)
+void mask_msi_irq(struct irq_data *data)
 {
-	msi_set_mask_bit(irq, 1);
+	msi_set_mask_bit(data, 1);
 }
 
-void unmask_msi_irq(unsigned int irq)
+void unmask_msi_irq(struct irq_data *data)
 {
-	msi_set_mask_bit(irq, 0);
+	msi_set_mask_bit(data, 0);
 }
 
-void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
+void __read_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
 {
-	struct msi_desc *entry = get_irq_desc_msi(desc);
-
 	BUG_ON(entry->dev->current_state != PCI_D0);
 
 	if (entry->msi_attrib.is_msix) {
@@ -227,15 +235,13 @@ void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
 
 void read_msi_msg(unsigned int irq, struct msi_msg *msg)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct msi_desc *entry = get_irq_msi(irq);
 
-	read_msi_msg_desc(desc, msg);
+	__read_msi_msg(entry, msg);
 }
 
-void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
+void __get_cached_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
 {
-	struct msi_desc *entry = get_irq_desc_msi(desc);
-
 	/* Assert that the cache is valid, assuming that
 	 * valid messages are not all-zeroes. */
 	BUG_ON(!(entry->msg.address_hi | entry->msg.address_lo |
@@ -246,15 +252,13 @@ void get_cached_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
 
 void get_cached_msi_msg(unsigned int irq, struct msi_msg *msg)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct msi_desc *entry = get_irq_msi(irq);
 
-	get_cached_msi_msg_desc(desc, msg);
+	__get_cached_msi_msg(entry, msg);
 }
 
-void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
+void __write_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
 {
-	struct msi_desc *entry = get_irq_desc_msi(desc);
-
 	if (entry->dev->current_state != PCI_D0) {
 		/* Don't touch the hardware now */
 	} else if (entry->msi_attrib.is_msix) {
@@ -292,9 +296,9 @@ void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg)
 
 void write_msi_msg(unsigned int irq, struct msi_msg *msg)
 {
-	struct irq_desc *desc = irq_to_desc(irq);
+	struct msi_desc *entry = get_irq_msi(irq);
 
-	write_msi_msg_desc(desc, msg);
+	__write_msi_msg(entry, msg);
 }
 
 static void free_msi_irqs(struct pci_dev *dev)
diff --git a/drivers/pci/xen-pcifront.c b/drivers/pci/xen-pcifront.c
new file mode 100644
index 00000000000..a87c4985326
--- /dev/null
+++ b/drivers/pci/xen-pcifront.c
@@ -0,0 +1,1148 @@
+/*
+ * Xen PCI Frontend.
+ *
+ *   Author: Ryan Wilson <hap9@epoch.ncsc.mil>
+ */
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <xen/xenbus.h>
+#include <xen/events.h>
+#include <xen/grant_table.h>
+#include <xen/page.h>
+#include <linux/spinlock.h>
+#include <linux/pci.h>
+#include <linux/msi.h>
+#include <xen/xenbus.h>
+#include <xen/interface/io/pciif.h>
+#include <asm/xen/pci.h>
+#include <linux/interrupt.h>
+#include <asm/atomic.h>
+#include <linux/workqueue.h>
+#include <linux/bitops.h>
+#include <linux/time.h>
+
+#define INVALID_GRANT_REF (0)
+#define INVALID_EVTCHN    (-1)
+
+struct pci_bus_entry {
+	struct list_head list;
+	struct pci_bus *bus;
+};
+
+#define _PDEVB_op_active		(0)
+#define PDEVB_op_active			(1 << (_PDEVB_op_active))
+
+struct pcifront_device {
+	struct xenbus_device *xdev;
+	struct list_head root_buses;
+
+	int evtchn;
+	int gnt_ref;
+
+	int irq;
+
+	/* Lock this when doing any operations in sh_info */
+	spinlock_t sh_info_lock;
+	struct xen_pci_sharedinfo *sh_info;
+	struct work_struct op_work;
+	unsigned long flags;
+
+};
+
+struct pcifront_sd {
+	int domain;
+	struct pcifront_device *pdev;
+};
+
+static inline struct pcifront_device *
+pcifront_get_pdev(struct pcifront_sd *sd)
+{
+	return sd->pdev;
+}
+
+static inline void pcifront_init_sd(struct pcifront_sd *sd,
+				    unsigned int domain, unsigned int bus,
+				    struct pcifront_device *pdev)
+{
+	sd->domain = domain;
+	sd->pdev = pdev;
+}
+
+static DEFINE_SPINLOCK(pcifront_dev_lock);
+static struct pcifront_device *pcifront_dev;
+
+static int verbose_request;
+module_param(verbose_request, int, 0644);
+
+static int errno_to_pcibios_err(int errno)
+{
+	switch (errno) {
+	case XEN_PCI_ERR_success:
+		return PCIBIOS_SUCCESSFUL;
+
+	case XEN_PCI_ERR_dev_not_found:
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	case XEN_PCI_ERR_invalid_offset:
+	case XEN_PCI_ERR_op_failed:
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	case XEN_PCI_ERR_not_implemented:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+
+	case XEN_PCI_ERR_access_denied:
+		return PCIBIOS_SET_FAILED;
+	}
+	return errno;
+}
+
+static inline void schedule_pcifront_aer_op(struct pcifront_device *pdev)
+{
+	if (test_bit(_XEN_PCIB_active, (unsigned long *)&pdev->sh_info->flags)
+		&& !test_and_set_bit(_PDEVB_op_active, &pdev->flags)) {
+		dev_dbg(&pdev->xdev->dev, "schedule aer frontend job\n");
+		schedule_work(&pdev->op_work);
+	}
+}
+
+static int do_pci_op(struct pcifront_device *pdev, struct xen_pci_op *op)
+{
+	int err = 0;
+	struct xen_pci_op *active_op = &pdev->sh_info->op;
+	unsigned long irq_flags;
+	evtchn_port_t port = pdev->evtchn;
+	unsigned irq = pdev->irq;
+	s64 ns, ns_timeout;
+	struct timeval tv;
+
+	spin_lock_irqsave(&pdev->sh_info_lock, irq_flags);
+
+	memcpy(active_op, op, sizeof(struct xen_pci_op));
+
+	/* Go */
+	wmb();
+	set_bit(_XEN_PCIF_active, (unsigned long *)&pdev->sh_info->flags);
+	notify_remote_via_evtchn(port);
+
+	/*
+	 * We set a poll timeout of 3 seconds but give up on return after
+	 * 2 seconds. It is better to time out too late rather than too early
+	 * (in the latter case we end up continually re-executing poll() with a
+	 * timeout in the past). 1s difference gives plenty of slack for error.
+	 */
+	do_gettimeofday(&tv);
+	ns_timeout = timeval_to_ns(&tv) + 2 * (s64)NSEC_PER_SEC;
+
+	xen_clear_irq_pending(irq);
+
+	while (test_bit(_XEN_PCIF_active,
+			(unsigned long *)&pdev->sh_info->flags)) {
+		xen_poll_irq_timeout(irq, jiffies + 3*HZ);
+		xen_clear_irq_pending(irq);
+		do_gettimeofday(&tv);
+		ns = timeval_to_ns(&tv);
+		if (ns > ns_timeout) {
+			dev_err(&pdev->xdev->dev,
+				"pciback not responding!!!\n");
+			clear_bit(_XEN_PCIF_active,
+				  (unsigned long *)&pdev->sh_info->flags);
+			err = XEN_PCI_ERR_dev_not_found;
+			goto out;
+		}
+	}
+
+	/*
+	* We might lose backend service request since we
+	* reuse same evtchn with pci_conf backend response. So re-schedule
+	* aer pcifront service.
+	*/
+	if (test_bit(_XEN_PCIB_active,
+			(unsigned long *)&pdev->sh_info->flags)) {
+		dev_err(&pdev->xdev->dev,
+			"schedule aer pcifront service\n");
+		schedule_pcifront_aer_op(pdev);
+	}
+
+	memcpy(op, active_op, sizeof(struct xen_pci_op));
+
+	err = op->err;
+out:
+	spin_unlock_irqrestore(&pdev->sh_info_lock, irq_flags);
+	return err;
+}
+
+/* Access to this function is spinlocked in drivers/pci/access.c */
+static int pcifront_bus_read(struct pci_bus *bus, unsigned int devfn,
+			     int where, int size, u32 *val)
+{
+	int err = 0;
+	struct xen_pci_op op = {
+		.cmd    = XEN_PCI_OP_conf_read,
+		.domain = pci_domain_nr(bus),
+		.bus    = bus->number,
+		.devfn  = devfn,
+		.offset = where,
+		.size   = size,
+	};
+	struct pcifront_sd *sd = bus->sysdata;
+	struct pcifront_device *pdev = pcifront_get_pdev(sd);
+
+	if (verbose_request)
+		dev_info(&pdev->xdev->dev,
+			 "read dev=%04x:%02x:%02x.%01x - offset %x size %d\n",
+			 pci_domain_nr(bus), bus->number, PCI_SLOT(devfn),
+			 PCI_FUNC(devfn), where, size);
+
+	err = do_pci_op(pdev, &op);
+
+	if (likely(!err)) {
+		if (verbose_request)
+			dev_info(&pdev->xdev->dev, "read got back value %x\n",
+				 op.value);
+
+		*val = op.value;
+	} else if (err == -ENODEV) {
+		/* No device here, pretend that it just returned 0 */
+		err = 0;
+		*val = 0;
+	}
+
+	return errno_to_pcibios_err(err);
+}
+
+/* Access to this function is spinlocked in drivers/pci/access.c */
+static int pcifront_bus_write(struct pci_bus *bus, unsigned int devfn,
+			      int where, int size, u32 val)
+{
+	struct xen_pci_op op = {
+		.cmd    = XEN_PCI_OP_conf_write,
+		.domain = pci_domain_nr(bus),
+		.bus    = bus->number,
+		.devfn  = devfn,
+		.offset = where,
+		.size   = size,
+		.value  = val,
+	};
+	struct pcifront_sd *sd = bus->sysdata;
+	struct pcifront_device *pdev = pcifront_get_pdev(sd);
+
+	if (verbose_request)
+		dev_info(&pdev->xdev->dev,
+			 "write dev=%04x:%02x:%02x.%01x - "
+			 "offset %x size %d val %x\n",
+			 pci_domain_nr(bus), bus->number,
+			 PCI_SLOT(devfn), PCI_FUNC(devfn), where, size, val);
+
+	return errno_to_pcibios_err(do_pci_op(pdev, &op));
+}
+
+struct pci_ops pcifront_bus_ops = {
+	.read = pcifront_bus_read,
+	.write = pcifront_bus_write,
+};
+
+#ifdef CONFIG_PCI_MSI
+static int pci_frontend_enable_msix(struct pci_dev *dev,
+				    int **vector, int nvec)
+{
+	int err;
+	int i;
+	struct xen_pci_op op = {
+		.cmd    = XEN_PCI_OP_enable_msix,
+		.domain = pci_domain_nr(dev->bus),
+		.bus = dev->bus->number,
+		.devfn = dev->devfn,
+		.value = nvec,
+	};
+	struct pcifront_sd *sd = dev->bus->sysdata;
+	struct pcifront_device *pdev = pcifront_get_pdev(sd);
+	struct msi_desc *entry;
+
+	if (nvec > SH_INFO_MAX_VEC) {
+		dev_err(&dev->dev, "too much vector for pci frontend: %x."
+				   " Increase SH_INFO_MAX_VEC.\n", nvec);
+		return -EINVAL;
+	}
+
+	i = 0;
+	list_for_each_entry(entry, &dev->msi_list, list) {
+		op.msix_entries[i].entry = entry->msi_attrib.entry_nr;
+		/* Vector is useless at this point. */
+		op.msix_entries[i].vector = -1;
+		i++;
+	}
+
+	err = do_pci_op(pdev, &op);
+
+	if (likely(!err)) {
+		if (likely(!op.value)) {
+			/* we get the result */
+			for (i = 0; i < nvec; i++)
+				*(*vector+i) = op.msix_entries[i].vector;
+			return 0;
+		} else {
+			printk(KERN_DEBUG "enable msix get value %x\n",
+				op.value);
+			return op.value;
+		}
+	} else {
+		dev_err(&dev->dev, "enable msix get err %x\n", err);
+		return err;
+	}
+}
+
+static void pci_frontend_disable_msix(struct pci_dev *dev)
+{
+	int err;
+	struct xen_pci_op op = {
+		.cmd    = XEN_PCI_OP_disable_msix,
+		.domain = pci_domain_nr(dev->bus),
+		.bus = dev->bus->number,
+		.devfn = dev->devfn,
+	};
+	struct pcifront_sd *sd = dev->bus->sysdata;
+	struct pcifront_device *pdev = pcifront_get_pdev(sd);
+
+	err = do_pci_op(pdev, &op);
+
+	/* What should do for error ? */
+	if (err)
+		dev_err(&dev->dev, "pci_disable_msix get err %x\n", err);
+}
+
+static int pci_frontend_enable_msi(struct pci_dev *dev, int **vector)
+{
+	int err;
+	struct xen_pci_op op = {
+		.cmd    = XEN_PCI_OP_enable_msi,
+		.domain = pci_domain_nr(dev->bus),
+		.bus = dev->bus->number,
+		.devfn = dev->devfn,
+	};
+	struct pcifront_sd *sd = dev->bus->sysdata;
+	struct pcifront_device *pdev = pcifront_get_pdev(sd);
+
+	err = do_pci_op(pdev, &op);
+	if (likely(!err)) {
+		*(*vector) = op.value;
+	} else {
+		dev_err(&dev->dev, "pci frontend enable msi failed for dev "
+				    "%x:%x\n", op.bus, op.devfn);
+		err = -EINVAL;
+	}
+	return err;
+}
+
+static void pci_frontend_disable_msi(struct pci_dev *dev)
+{
+	int err;
+	struct xen_pci_op op = {
+		.cmd    = XEN_PCI_OP_disable_msi,
+		.domain = pci_domain_nr(dev->bus),
+		.bus = dev->bus->number,
+		.devfn = dev->devfn,
+	};
+	struct pcifront_sd *sd = dev->bus->sysdata;
+	struct pcifront_device *pdev = pcifront_get_pdev(sd);
+
+	err = do_pci_op(pdev, &op);
+	if (err == XEN_PCI_ERR_dev_not_found) {
+		/* XXX No response from backend, what shall we do? */
+		printk(KERN_DEBUG "get no response from backend for disable MSI\n");
+		return;
+	}
+	if (err)
+		/* how can pciback notify us fail? */
+		printk(KERN_DEBUG "get fake response frombackend\n");
+}
+
+static struct xen_pci_frontend_ops pci_frontend_ops = {
+	.enable_msi = pci_frontend_enable_msi,
+	.disable_msi = pci_frontend_disable_msi,
+	.enable_msix = pci_frontend_enable_msix,
+	.disable_msix = pci_frontend_disable_msix,
+};
+
+static void pci_frontend_registrar(int enable)
+{
+	if (enable)
+		xen_pci_frontend = &pci_frontend_ops;
+	else
+		xen_pci_frontend = NULL;
+};
+#else
+static inline void pci_frontend_registrar(int enable) { };
+#endif /* CONFIG_PCI_MSI */
+
+/* Claim resources for the PCI frontend as-is, backend won't allow changes */
+static int pcifront_claim_resource(struct pci_dev *dev, void *data)
+{
+	struct pcifront_device *pdev = data;
+	int i;
+	struct resource *r;
+
+	for (i = 0; i < PCI_NUM_RESOURCES; i++) {
+		r = &dev->resource[i];
+
+		if (!r->parent && r->start && r->flags) {
+			dev_info(&pdev->xdev->dev, "claiming resource %s/%d\n",
+				pci_name(dev), i);
+			if (pci_claim_resource(dev, i)) {
+				dev_err(&pdev->xdev->dev, "Could not claim "
+					"resource %s/%d! Device offline. Try "
+					"giving less than 4GB to domain.\n",
+					pci_name(dev), i);
+			}
+		}
+	}
+
+	return 0;
+}
+
+static int __devinit pcifront_scan_bus(struct pcifront_device *pdev,
+				unsigned int domain, unsigned int bus,
+				struct pci_bus *b)
+{
+	struct pci_dev *d;
+	unsigned int devfn;
+
+	/* Scan the bus for functions and add.
+	 * We omit handling of PCI bridge attachment because pciback prevents
+	 * bridges from being exported.
+	 */
+	for (devfn = 0; devfn < 0x100; devfn++) {
+		d = pci_get_slot(b, devfn);
+		if (d) {
+			/* Device is already known. */
+			pci_dev_put(d);
+			continue;
+		}
+
+		d = pci_scan_single_device(b, devfn);
+		if (d)
+			dev_info(&pdev->xdev->dev, "New device on "
+				 "%04x:%02x:%02x.%02x found.\n", domain, bus,
+				 PCI_SLOT(devfn), PCI_FUNC(devfn));
+	}
+
+	return 0;
+}
+
+static int __devinit pcifront_scan_root(struct pcifront_device *pdev,
+				 unsigned int domain, unsigned int bus)
+{
+	struct pci_bus *b;
+	struct pcifront_sd *sd = NULL;
+	struct pci_bus_entry *bus_entry = NULL;
+	int err = 0;
+
+#ifndef CONFIG_PCI_DOMAINS
+	if (domain != 0) {
+		dev_err(&pdev->xdev->dev,
+			"PCI Root in non-zero PCI Domain! domain=%d\n", domain);
+		dev_err(&pdev->xdev->dev,
+			"Please compile with CONFIG_PCI_DOMAINS\n");
+		err = -EINVAL;
+		goto err_out;
+	}
+#endif
+
+	dev_info(&pdev->xdev->dev, "Creating PCI Frontend Bus %04x:%02x\n",
+		 domain, bus);
+
+	bus_entry = kmalloc(sizeof(*bus_entry), GFP_KERNEL);
+	sd = kmalloc(sizeof(*sd), GFP_KERNEL);
+	if (!bus_entry || !sd) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+	pcifront_init_sd(sd, domain, bus, pdev);
+
+	b = pci_scan_bus_parented(&pdev->xdev->dev, bus,
+				  &pcifront_bus_ops, sd);
+	if (!b) {
+		dev_err(&pdev->xdev->dev,
+			"Error creating PCI Frontend Bus!\n");
+		err = -ENOMEM;
+		goto err_out;
+	}
+
+	bus_entry->bus = b;
+
+	list_add(&bus_entry->list, &pdev->root_buses);
+
+	/* pci_scan_bus_parented skips devices which do not have a have
+	* devfn==0. The pcifront_scan_bus enumerates all devfn. */
+	err = pcifront_scan_bus(pdev, domain, bus, b);
+
+	/* Claim resources before going "live" with our devices */
+	pci_walk_bus(b, pcifront_claim_resource, pdev);
+
+	/* Create SysFS and notify udev of the devices. Aka: "going live" */
+	pci_bus_add_devices(b);
+
+	return err;
+
+err_out:
+	kfree(bus_entry);
+	kfree(sd);
+
+	return err;
+}
+
+static int __devinit pcifront_rescan_root(struct pcifront_device *pdev,
+				   unsigned int domain, unsigned int bus)
+{
+	int err;
+	struct pci_bus *b;
+
+#ifndef CONFIG_PCI_DOMAINS
+	if (domain != 0) {
+		dev_err(&pdev->xdev->dev,
+			"PCI Root in non-zero PCI Domain! domain=%d\n", domain);
+		dev_err(&pdev->xdev->dev,
+			"Please compile with CONFIG_PCI_DOMAINS\n");
+		return -EINVAL;
+	}
+#endif
+
+	dev_info(&pdev->xdev->dev, "Rescanning PCI Frontend Bus %04x:%02x\n",
+		 domain, bus);
+
+	b = pci_find_bus(domain, bus);
+	if (!b)
+		/* If the bus is unknown, create it. */
+		return pcifront_scan_root(pdev, domain, bus);
+
+	err = pcifront_scan_bus(pdev, domain, bus, b);
+
+	/* Claim resources before going "live" with our devices */
+	pci_walk_bus(b, pcifront_claim_resource, pdev);
+
+	/* Create SysFS and notify udev of the devices. Aka: "going live" */
+	pci_bus_add_devices(b);
+
+	return err;
+}
+
+static void free_root_bus_devs(struct pci_bus *bus)
+{
+	struct pci_dev *dev;
+
+	while (!list_empty(&bus->devices)) {
+		dev = container_of(bus->devices.next, struct pci_dev,
+				   bus_list);
+		dev_dbg(&dev->dev, "removing device\n");
+		pci_remove_bus_device(dev);
+	}
+}
+
+static void pcifront_free_roots(struct pcifront_device *pdev)
+{
+	struct pci_bus_entry *bus_entry, *t;
+
+	dev_dbg(&pdev->xdev->dev, "cleaning up root buses\n");
+
+	list_for_each_entry_safe(bus_entry, t, &pdev->root_buses, list) {
+		list_del(&bus_entry->list);
+
+		free_root_bus_devs(bus_entry->bus);
+
+		kfree(bus_entry->bus->sysdata);
+
+		device_unregister(bus_entry->bus->bridge);
+		pci_remove_bus(bus_entry->bus);
+
+		kfree(bus_entry);
+	}
+}
+
+static pci_ers_result_t pcifront_common_process(int cmd,
+						struct pcifront_device *pdev,
+						pci_channel_state_t state)
+{
+	pci_ers_result_t result;
+	struct pci_driver *pdrv;
+	int bus = pdev->sh_info->aer_op.bus;
+	int devfn = pdev->sh_info->aer_op.devfn;
+	struct pci_dev *pcidev;
+	int flag = 0;
+
+	dev_dbg(&pdev->xdev->dev,
+		"pcifront AER process: cmd %x (bus:%x, devfn%x)",
+		cmd, bus, devfn);
+	result = PCI_ERS_RESULT_NONE;
+
+	pcidev = pci_get_bus_and_slot(bus, devfn);
+	if (!pcidev || !pcidev->driver) {
+		dev_err(&pcidev->dev,
+			"device or driver is NULL\n");
+		return result;
+	}
+	pdrv = pcidev->driver;
+
+	if (get_driver(&pdrv->driver)) {
+		if (pdrv->err_handler && pdrv->err_handler->error_detected) {
+			dev_dbg(&pcidev->dev,
+				"trying to call AER service\n");
+			if (pcidev) {
+				flag = 1;
+				switch (cmd) {
+				case XEN_PCI_OP_aer_detected:
+					result = pdrv->err_handler->
+						 error_detected(pcidev, state);
+					break;
+				case XEN_PCI_OP_aer_mmio:
+					result = pdrv->err_handler->
+						 mmio_enabled(pcidev);
+					break;
+				case XEN_PCI_OP_aer_slotreset:
+					result = pdrv->err_handler->
+						 slot_reset(pcidev);
+					break;
+				case XEN_PCI_OP_aer_resume:
+					pdrv->err_handler->resume(pcidev);
+					break;
+				default:
+					dev_err(&pdev->xdev->dev,
+						"bad request in aer recovery "
+						"operation!\n");
+
+				}
+			}
+		}
+		put_driver(&pdrv->driver);
+	}
+	if (!flag)
+		result = PCI_ERS_RESULT_NONE;
+
+	return result;
+}
+
+
+static void pcifront_do_aer(struct work_struct *data)
+{
+	struct pcifront_device *pdev =
+		container_of(data, struct pcifront_device, op_work);
+	int cmd = pdev->sh_info->aer_op.cmd;
+	pci_channel_state_t state =
+		(pci_channel_state_t)pdev->sh_info->aer_op.err;
+
+	/*If a pci_conf op is in progress,
+		we have to wait until it is done before service aer op*/
+	dev_dbg(&pdev->xdev->dev,
+		"pcifront service aer bus %x devfn %x\n",
+		pdev->sh_info->aer_op.bus, pdev->sh_info->aer_op.devfn);
+
+	pdev->sh_info->aer_op.err = pcifront_common_process(cmd, pdev, state);
+
+	/* Post the operation to the guest. */
+	wmb();
+	clear_bit(_XEN_PCIB_active, (unsigned long *)&pdev->sh_info->flags);
+	notify_remote_via_evtchn(pdev->evtchn);
+
+	/*in case of we lost an aer request in four lines time_window*/
+	smp_mb__before_clear_bit();
+	clear_bit(_PDEVB_op_active, &pdev->flags);
+	smp_mb__after_clear_bit();
+
+	schedule_pcifront_aer_op(pdev);
+
+}
+
+static irqreturn_t pcifront_handler_aer(int irq, void *dev)
+{
+	struct pcifront_device *pdev = dev;
+	schedule_pcifront_aer_op(pdev);
+	return IRQ_HANDLED;
+}
+static int pcifront_connect(struct pcifront_device *pdev)
+{
+	int err = 0;
+
+	spin_lock(&pcifront_dev_lock);
+
+	if (!pcifront_dev) {
+		dev_info(&pdev->xdev->dev, "Installing PCI frontend\n");
+		pcifront_dev = pdev;
+	} else {
+		dev_err(&pdev->xdev->dev, "PCI frontend already installed!\n");
+		err = -EEXIST;
+	}
+
+	spin_unlock(&pcifront_dev_lock);
+
+	return err;
+}
+
+static void pcifront_disconnect(struct pcifront_device *pdev)
+{
+	spin_lock(&pcifront_dev_lock);
+
+	if (pdev == pcifront_dev) {
+		dev_info(&pdev->xdev->dev,
+			 "Disconnecting PCI Frontend Buses\n");
+		pcifront_dev = NULL;
+	}
+
+	spin_unlock(&pcifront_dev_lock);
+}
+static struct pcifront_device *alloc_pdev(struct xenbus_device *xdev)
+{
+	struct pcifront_device *pdev;
+
+	pdev = kzalloc(sizeof(struct pcifront_device), GFP_KERNEL);
+	if (pdev == NULL)
+		goto out;
+
+	pdev->sh_info =
+	    (struct xen_pci_sharedinfo *)__get_free_page(GFP_KERNEL);
+	if (pdev->sh_info == NULL) {
+		kfree(pdev);
+		pdev = NULL;
+		goto out;
+	}
+	pdev->sh_info->flags = 0;
+
+	/*Flag for registering PV AER handler*/
+	set_bit(_XEN_PCIB_AERHANDLER, (void *)&pdev->sh_info->flags);
+
+	dev_set_drvdata(&xdev->dev, pdev);
+	pdev->xdev = xdev;
+
+	INIT_LIST_HEAD(&pdev->root_buses);
+
+	spin_lock_init(&pdev->sh_info_lock);
+
+	pdev->evtchn = INVALID_EVTCHN;
+	pdev->gnt_ref = INVALID_GRANT_REF;
+	pdev->irq = -1;
+
+	INIT_WORK(&pdev->op_work, pcifront_do_aer);
+
+	dev_dbg(&xdev->dev, "Allocated pdev @ 0x%p pdev->sh_info @ 0x%p\n",
+		pdev, pdev->sh_info);
+out:
+	return pdev;
+}
+
+static void free_pdev(struct pcifront_device *pdev)
+{
+	dev_dbg(&pdev->xdev->dev, "freeing pdev @ 0x%p\n", pdev);
+
+	pcifront_free_roots(pdev);
+
+	/*For PCIE_AER error handling job*/
+	flush_scheduled_work();
+
+	if (pdev->irq >= 0)
+		unbind_from_irqhandler(pdev->irq, pdev);
+
+	if (pdev->evtchn != INVALID_EVTCHN)
+		xenbus_free_evtchn(pdev->xdev, pdev->evtchn);
+
+	if (pdev->gnt_ref != INVALID_GRANT_REF)
+		gnttab_end_foreign_access(pdev->gnt_ref, 0 /* r/w page */,
+					  (unsigned long)pdev->sh_info);
+	else
+		free_page((unsigned long)pdev->sh_info);
+
+	dev_set_drvdata(&pdev->xdev->dev, NULL);
+
+	kfree(pdev);
+}
+
+static int pcifront_publish_info(struct pcifront_device *pdev)
+{
+	int err = 0;
+	struct xenbus_transaction trans;
+
+	err = xenbus_grant_ring(pdev->xdev, virt_to_mfn(pdev->sh_info));
+	if (err < 0)
+		goto out;
+
+	pdev->gnt_ref = err;
+
+	err = xenbus_alloc_evtchn(pdev->xdev, &pdev->evtchn);
+	if (err)
+		goto out;
+
+	err = bind_evtchn_to_irqhandler(pdev->evtchn, pcifront_handler_aer,
+		0, "pcifront", pdev);
+
+	if (err < 0)
+		return err;
+
+	pdev->irq = err;
+
+do_publish:
+	err = xenbus_transaction_start(&trans);
+	if (err) {
+		xenbus_dev_fatal(pdev->xdev, err,
+				 "Error writing configuration for backend "
+				 "(start transaction)");
+		goto out;
+	}
+
+	err = xenbus_printf(trans, pdev->xdev->nodename,
+			    "pci-op-ref", "%u", pdev->gnt_ref);
+	if (!err)
+		err = xenbus_printf(trans, pdev->xdev->nodename,
+				    "event-channel", "%u", pdev->evtchn);
+	if (!err)
+		err = xenbus_printf(trans, pdev->xdev->nodename,
+				    "magic", XEN_PCI_MAGIC);
+
+	if (err) {
+		xenbus_transaction_end(trans, 1);
+		xenbus_dev_fatal(pdev->xdev, err,
+				 "Error writing configuration for backend");
+		goto out;
+	} else {
+		err = xenbus_transaction_end(trans, 0);
+		if (err == -EAGAIN)
+			goto do_publish;
+		else if (err) {
+			xenbus_dev_fatal(pdev->xdev, err,
+					 "Error completing transaction "
+					 "for backend");
+			goto out;
+		}
+	}
+
+	xenbus_switch_state(pdev->xdev, XenbusStateInitialised);
+
+	dev_dbg(&pdev->xdev->dev, "publishing successful!\n");
+
+out:
+	return err;
+}
+
+static int __devinit pcifront_try_connect(struct pcifront_device *pdev)
+{
+	int err = -EFAULT;
+	int i, num_roots, len;
+	char str[64];
+	unsigned int domain, bus;
+
+
+	/* Only connect once */
+	if (xenbus_read_driver_state(pdev->xdev->nodename) !=
+	    XenbusStateInitialised)
+		goto out;
+
+	err = pcifront_connect(pdev);
+	if (err) {
+		xenbus_dev_fatal(pdev->xdev, err,
+				 "Error connecting PCI Frontend");
+		goto out;
+	}
+
+	err = xenbus_scanf(XBT_NIL, pdev->xdev->otherend,
+			   "root_num", "%d", &num_roots);
+	if (err == -ENOENT) {
+		xenbus_dev_error(pdev->xdev, err,
+				 "No PCI Roots found, trying 0000:00");
+		err = pcifront_scan_root(pdev, 0, 0);
+		num_roots = 0;
+	} else if (err != 1) {
+		if (err == 0)
+			err = -EINVAL;
+		xenbus_dev_fatal(pdev->xdev, err,
+				 "Error reading number of PCI roots");
+		goto out;
+	}
+
+	for (i = 0; i < num_roots; i++) {
+		len = snprintf(str, sizeof(str), "root-%d", i);
+		if (unlikely(len >= (sizeof(str) - 1))) {
+			err = -ENOMEM;
+			goto out;
+		}
+
+		err = xenbus_scanf(XBT_NIL, pdev->xdev->otherend, str,
+				   "%x:%x", &domain, &bus);
+		if (err != 2) {
+			if (err >= 0)
+				err = -EINVAL;
+			xenbus_dev_fatal(pdev->xdev, err,
+					 "Error reading PCI root %d", i);
+			goto out;
+		}
+
+		err = pcifront_scan_root(pdev, domain, bus);
+		if (err) {
+			xenbus_dev_fatal(pdev->xdev, err,
+					 "Error scanning PCI root %04x:%02x",
+					 domain, bus);
+			goto out;
+		}
+	}
+
+	err = xenbus_switch_state(pdev->xdev, XenbusStateConnected);
+
+out:
+	return err;
+}
+
+static int pcifront_try_disconnect(struct pcifront_device *pdev)
+{
+	int err = 0;
+	enum xenbus_state prev_state;
+
+
+	prev_state = xenbus_read_driver_state(pdev->xdev->nodename);
+
+	if (prev_state >= XenbusStateClosing)
+		goto out;
+
+	if (prev_state == XenbusStateConnected) {
+		pcifront_free_roots(pdev);
+		pcifront_disconnect(pdev);
+	}
+
+	err = xenbus_switch_state(pdev->xdev, XenbusStateClosed);
+
+out:
+
+	return err;
+}
+
+static int __devinit pcifront_attach_devices(struct pcifront_device *pdev)
+{
+	int err = -EFAULT;
+	int i, num_roots, len;
+	unsigned int domain, bus;
+	char str[64];
+
+	if (xenbus_read_driver_state(pdev->xdev->nodename) !=
+	    XenbusStateReconfiguring)
+		goto out;
+
+	err = xenbus_scanf(XBT_NIL, pdev->xdev->otherend,
+			   "root_num", "%d", &num_roots);
+	if (err == -ENOENT) {
+		xenbus_dev_error(pdev->xdev, err,
+				 "No PCI Roots found, trying 0000:00");
+		err = pcifront_rescan_root(pdev, 0, 0);
+		num_roots = 0;
+	} else if (err != 1) {
+		if (err == 0)
+			err = -EINVAL;
+		xenbus_dev_fatal(pdev->xdev, err,
+				 "Error reading number of PCI roots");
+		goto out;
+	}
+
+	for (i = 0; i < num_roots; i++) {
+		len = snprintf(str, sizeof(str), "root-%d", i);
+		if (unlikely(len >= (sizeof(str) - 1))) {
+			err = -ENOMEM;
+			goto out;
+		}
+
+		err = xenbus_scanf(XBT_NIL, pdev->xdev->otherend, str,
+				   "%x:%x", &domain, &bus);
+		if (err != 2) {
+			if (err >= 0)
+				err = -EINVAL;
+			xenbus_dev_fatal(pdev->xdev, err,
+					 "Error reading PCI root %d", i);
+			goto out;
+		}
+
+		err = pcifront_rescan_root(pdev, domain, bus);
+		if (err) {
+			xenbus_dev_fatal(pdev->xdev, err,
+					 "Error scanning PCI root %04x:%02x",
+					 domain, bus);
+			goto out;
+		}
+	}
+
+	xenbus_switch_state(pdev->xdev, XenbusStateConnected);
+
+out:
+	return err;
+}
+
+static int pcifront_detach_devices(struct pcifront_device *pdev)
+{
+	int err = 0;
+	int i, num_devs;
+	unsigned int domain, bus, slot, func;
+	struct pci_bus *pci_bus;
+	struct pci_dev *pci_dev;
+	char str[64];
+
+	if (xenbus_read_driver_state(pdev->xdev->nodename) !=
+	    XenbusStateConnected)
+		goto out;
+
+	err = xenbus_scanf(XBT_NIL, pdev->xdev->otherend, "num_devs", "%d",
+			   &num_devs);
+	if (err != 1) {
+		if (err >= 0)
+			err = -EINVAL;
+		xenbus_dev_fatal(pdev->xdev, err,
+				 "Error reading number of PCI devices");
+		goto out;
+	}
+
+	/* Find devices being detached and remove them. */
+	for (i = 0; i < num_devs; i++) {
+		int l, state;
+		l = snprintf(str, sizeof(str), "state-%d", i);
+		if (unlikely(l >= (sizeof(str) - 1))) {
+			err = -ENOMEM;
+			goto out;
+		}
+		err = xenbus_scanf(XBT_NIL, pdev->xdev->otherend, str, "%d",
+				   &state);
+		if (err != 1)
+			state = XenbusStateUnknown;
+
+		if (state != XenbusStateClosing)
+			continue;
+
+		/* Remove device. */
+		l = snprintf(str, sizeof(str), "vdev-%d", i);
+		if (unlikely(l >= (sizeof(str) - 1))) {
+			err = -ENOMEM;
+			goto out;
+		}
+		err = xenbus_scanf(XBT_NIL, pdev->xdev->otherend, str,
+				   "%x:%x:%x.%x", &domain, &bus, &slot, &func);
+		if (err != 4) {
+			if (err >= 0)
+				err = -EINVAL;
+			xenbus_dev_fatal(pdev->xdev, err,
+					 "Error reading PCI device %d", i);
+			goto out;
+		}
+
+		pci_bus = pci_find_bus(domain, bus);
+		if (!pci_bus) {
+			dev_dbg(&pdev->xdev->dev, "Cannot get bus %04x:%02x\n",
+				domain, bus);
+			continue;
+		}
+		pci_dev = pci_get_slot(pci_bus, PCI_DEVFN(slot, func));
+		if (!pci_dev) {
+			dev_dbg(&pdev->xdev->dev,
+				"Cannot get PCI device %04x:%02x:%02x.%02x\n",
+				domain, bus, slot, func);
+			continue;
+		}
+		pci_remove_bus_device(pci_dev);
+		pci_dev_put(pci_dev);
+
+		dev_dbg(&pdev->xdev->dev,
+			"PCI device %04x:%02x:%02x.%02x removed.\n",
+			domain, bus, slot, func);
+	}
+
+	err = xenbus_switch_state(pdev->xdev, XenbusStateReconfiguring);
+
+out:
+	return err;
+}
+
+static void __init_refok pcifront_backend_changed(struct xenbus_device *xdev,
+						  enum xenbus_state be_state)
+{
+	struct pcifront_device *pdev = dev_get_drvdata(&xdev->dev);
+
+	switch (be_state) {
+	case XenbusStateUnknown:
+	case XenbusStateInitialising:
+	case XenbusStateInitWait:
+	case XenbusStateInitialised:
+	case XenbusStateClosed:
+		break;
+
+	case XenbusStateConnected:
+		pcifront_try_connect(pdev);
+		break;
+
+	case XenbusStateClosing:
+		dev_warn(&xdev->dev, "backend going away!\n");
+		pcifront_try_disconnect(pdev);
+		break;
+
+	case XenbusStateReconfiguring:
+		pcifront_detach_devices(pdev);
+		break;
+
+	case XenbusStateReconfigured:
+		pcifront_attach_devices(pdev);
+		break;
+	}
+}
+
+static int pcifront_xenbus_probe(struct xenbus_device *xdev,
+				 const struct xenbus_device_id *id)
+{
+	int err = 0;
+	struct pcifront_device *pdev = alloc_pdev(xdev);
+
+	if (pdev == NULL) {
+		err = -ENOMEM;
+		xenbus_dev_fatal(xdev, err,
+				 "Error allocating pcifront_device struct");
+		goto out;
+	}
+
+	err = pcifront_publish_info(pdev);
+	if (err)
+		free_pdev(pdev);
+
+out:
+	return err;
+}
+
+static int pcifront_xenbus_remove(struct xenbus_device *xdev)
+{
+	struct pcifront_device *pdev = dev_get_drvdata(&xdev->dev);
+	if (pdev)
+		free_pdev(pdev);
+
+	return 0;
+}
+
+static const struct xenbus_device_id xenpci_ids[] = {
+	{"pci"},
+	{""},
+};
+
+static struct xenbus_driver xenbus_pcifront_driver = {
+	.name			= "pcifront",
+	.owner			= THIS_MODULE,
+	.ids			= xenpci_ids,
+	.probe			= pcifront_xenbus_probe,
+	.remove			= pcifront_xenbus_remove,
+	.otherend_changed	= pcifront_backend_changed,
+};
+
+static int __init pcifront_init(void)
+{
+	if (!xen_pv_domain() || xen_initial_domain())
+		return -ENODEV;
+
+	pci_frontend_registrar(1 /* enable */);
+
+	return xenbus_register_frontend(&xenbus_pcifront_driver);
+}
+
+static void __exit pcifront_cleanup(void)
+{
+	xenbus_unregister_driver(&xenbus_pcifront_driver);
+	pci_frontend_registrar(0 /* disable */);
+}
+module_init(pcifront_init);
+module_exit(pcifront_cleanup);
+
+MODULE_DESCRIPTION("Xen PCI passthrough frontend.");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("xen:pci");