Merge tag 'pci-v4.6-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci

Pull PCI updates from Bjorn Helgaas:
 "PCI changes for v4.6:

  Enumeration:
   - Disable IO/MEM decoding for devices with non-compliant BARs (Bjorn Helgaas)
   - Mark Broadwell-EP Home Agent & PCU as having non-compliant BARs (Bjorn Helgaas

  Resource management:
   - Mark shadow copy of VGA ROM as IORESOURCE_PCI_FIXED (Bjorn Helgaas)
   - Don't assign or reassign immutable resources (Bjorn Helgaas)
   - Don't enable/disable ROM BAR if we're using a RAM shadow copy (Bjorn Helgaas)
   - Set ROM shadow location in arch code, not in PCI core (Bjorn Helgaas)
   - Remove arch-specific IORESOURCE_ROM_SHADOW size from sysfs (Bjorn Helgaas)
   - ia64: Use ioremap() instead of open-coded equivalent (Bjorn Helgaas)
   - ia64: Keep CPU physical (not virtual) addresses in shadow ROM resource (Bjorn Helgaas)
   - MIPS: Keep CPU physical (not virtual) addresses in shadow ROM resource (Bjorn Helgaas)
   - Remove unused IORESOURCE_ROM_COPY and IORESOURCE_ROM_BIOS_COPY (Bjorn Helgaas)
   - Don't leak memory if sysfs_create_bin_file() fails (Bjorn Helgaas)
   - rcar: Remove PCI_PROBE_ONLY handling (Lorenzo Pieralisi)
   - designware: Remove PCI_PROBE_ONLY handling (Lorenzo Pieralisi)

  Virtualization:
   - Wait for up to 1000ms after FLR reset (Alex Williamson)
   - Support SR-IOV on any function type (Kelly Zytaruk)
   - Add ACS quirk for all Cavium devices (Manish Jaggi)

  AER:
   - Rename pci_ops_aer to aer_inj_pci_ops (Bjorn Helgaas)
   - Restore pci_ops pointer while calling original pci_ops (David Daney)
   - Fix aer_inject error codes (Jean Delvare)
   - Use dev_warn() in aer_inject (Jean Delvare)
   - Log actual error causes in aer_inject (Jean Delvare)
   - Log aer_inject error injections (Jean Delvare)

  VPD:
   - Prevent VPD access for buggy devices (Babu Moger)
   - Move pci_read_vpd() and pci_write_vpd() close to other VPD code (Bjorn Helgaas)
   - Move pci_vpd_release() from header file to pci/access.c (Bjorn Helgaas)
   - Remove struct pci_vpd_ops.release function pointer (Bjorn Helgaas)
   - Rename VPD symbols to remove unnecessary "pci22" (Bjorn Helgaas)
   - Fold struct pci_vpd_pci22 into struct pci_vpd (Bjorn Helgaas)
   - Sleep rather than busy-wait for VPD access completion (Bjorn Helgaas)
   - Update VPD definitions (Hannes Reinecke)
   - Allow access to VPD attributes with size 0 (Hannes Reinecke)
   - Determine actual VPD size on first access (Hannes Reinecke)

  Generic host bridge driver:
   - Move structure definitions to separate header file (David Daney)
   - Add pci_host_common_probe(), based on gen_pci_probe() (David Daney)
   - Expose pci_host_common_probe() for use by other drivers (David Daney)

  Altera host bridge driver:
   - Fix altera_pcie_link_is_up() (Ley Foon Tan)

  Cavium ThunderX host bridge driver:
   - Add PCIe host driver for ThunderX processors (David Daney)
   - Add driver for ThunderX-pass{1,2} on-chip devices (David Daney)

  Freescale i.MX6 host bridge driver:
   - Add DT bindings to configure PHY Tx driver settings (Justin Waters)
   - Move imx6_pcie_reset_phy() near other PHY handling functions (Lucas Stach)
   - Move PHY reset into imx6_pcie_establish_link() (Lucas Stach)
   - Remove broken Gen2 workaround (Lucas Stach)
   - Move link up check into imx6_pcie_wait_for_link() (Lucas Stach)

  Freescale Layerscape host bridge driver:
   - Add "fsl,ls2085a-pcie" compatible ID (Yang Shi)

  Intel VMD host bridge driver:
   - Attach VMD resources to parent domain's resource tree (Jon Derrick)
   - Set bus resource start to 0 (Keith Busch)

  Microsoft Hyper-V host bridge driver:
   - Add fwnode_handle to x86 pci_sysdata (Jake Oshins)
   - Look up IRQ domain by fwnode_handle (Jake Oshins)
   - Add paravirtual PCI front-end for Microsoft Hyper-V VMs (Jake Oshins)

  NVIDIA Tegra host bridge driver:
   - Add pci_ops.{add,remove}_bus() callbacks (Thierry Reding)
   - Implement ->{add,remove}_bus() callbacks (Thierry Reding)
   - Remove unused struct tegra_pcie.num_ports field (Thierry Reding)
   - Track bus -> CPU mapping (Thierry Reding)
   - Remove misleading PHYS_OFFSET (Thierry Reding)

  Renesas R-Car host bridge driver:
   - Depend on ARCH_RENESAS, not ARCH_SHMOBILE (Simon Horman)

  Synopsys DesignWare host bridge driver:
   - ARC: Add PCI support (Joao Pinto)
   - Add generic dw_pcie_wait_for_link() (Joao Pinto)
   - Add default link up check if sub-driver doesn't override (Joao Pinto)
   - Add driver for prototyping kits based on ARC SDP (Joao Pinto)

  TI Keystone host bridge driver:
   - Defer probing if devm_phy_get() returns -EPROBE_DEFER (Shawn Lin)

  Xilinx AXI host bridge driver:
   - Use of_pci_get_host_bridge_resources() to parse DT (Bharat Kumar Gogada)
   - Remove dependency on ARM-specific struct hw_pci (Bharat Kumar Gogada)
   - Don't call pci_fixup_irqs() on Microblaze (Bharat Kumar Gogada)
   - Update Zynq binding with Microblaze node (Bharat Kumar Gogada)
   - microblaze: Support generic Xilinx AXI PCIe Host Bridge IP driver (Bharat Kumar Gogada)

  Xilinx NWL host bridge driver:
   - Add support for Xilinx NWL PCIe Host Controller (Bharat Kumar Gogada)

  Miscellaneous:
   - Check device_attach() return value always (Bjorn Helgaas)
   - Move pci_set_flags() from asm-generic/pci-bridge.h to linux/pci.h (Bjorn Helgaas)
   - Remove includes of empty asm-generic/pci-bridge.h (Bjorn Helgaas)
   - ARM64: Remove generated include of asm-generic/pci-bridge.h (Bjorn Helgaas)
   - Remove empty asm-generic/pci-bridge.h (Bjorn Helgaas)
   - Remove includes of asm/pci-bridge.h (Bjorn Helgaas)
   - Consolidate PCI DMA constants and interfaces in linux/pci-dma-compat.h (Bjorn Helgaas)
   - unicore32: Remove unused HAVE_ARCH_PCI_SET_DMA_MASK definition (Bjorn Helgaas)
   - Cleanup pci/pcie/Kconfig whitespace (Andreas Ziegler)
   - Include pci/hotplug Kconfig directly from pci/Kconfig (Bjorn Helgaas)
   - Include pci/pcie/Kconfig directly from pci/Kconfig (Bogicevic Sasa)
   - frv: Remove stray pci_{alloc,free}_consistent() declaration (Christoph Hellwig)
   - Move pci_dma_* helpers to common code (Christoph Hellwig)
   - Add PCI_CLASS_SERIAL_USB_DEVICE definition (Heikki Krogerus)
   - Add QEMU top-level IDs for (sub)vendor & device (Robin H. Johnson)
   - Fix broken URL for Dell biosdevname (Naga Venkata Sai Indubhaskar Jupudi)"

* tag 'pci-v4.6-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci: (94 commits)
  PCI: Add PCI_CLASS_SERIAL_USB_DEVICE definition
  PCI: designware: Add driver for prototyping kits based on ARC SDP
  PCI: designware: Add default link up check if sub-driver doesn't override
  PCI: designware: Add generic dw_pcie_wait_for_link()
  PCI: Cleanup pci/pcie/Kconfig whitespace
  PCI: Simplify pci_create_attr() control flow
  PCI: Don't leak memory if sysfs_create_bin_file() fails
  PCI: Simplify sysfs ROM cleanup
  PCI: Remove unused IORESOURCE_ROM_COPY and IORESOURCE_ROM_BIOS_COPY
  MIPS: Loongson 3: Keep CPU physical (not virtual) addresses in shadow ROM resource
  MIPS: Loongson 3: Use temporary struct resource * to avoid repetition
  ia64/PCI: Keep CPU physical (not virtual) addresses in shadow ROM resource
  ia64/PCI: Use ioremap() instead of open-coded equivalent
  ia64/PCI: Use temporary struct resource * to avoid repetition
  PCI: Clean up pci_map_rom() whitespace
  PCI: Remove arch-specific IORESOURCE_ROM_SHADOW size from sysfs
  PCI: thunder: Add driver for ThunderX-pass{1,2} on-chip devices
  PCI: thunder: Add PCIe host driver for ThunderX processors
  PCI: generic: Expose pci_host_common_probe() for use by other drivers
  PCI: generic: Add pci_host_common_probe(), based on gen_pci_probe()
  ...

41 files changed, 5079 insertions, 789 deletions

diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig
index 73de4efcbe6e..a1f37db745ab 100644
--- a/drivers/pci/Kconfig
+++ b/drivers/pci/Kconfig
@@ -1,6 +1,9 @@
 #
 # PCI configuration
 #
+
+source "drivers/pci/pcie/Kconfig"
+
 config PCI_BUS_ADDR_T_64BIT
 	def_bool y if (ARCH_DMA_ADDR_T_64BIT || 64BIT)
 	depends on PCI
@@ -118,4 +121,11 @@ config PCI_LABEL
 	def_bool y if (DMI || ACPI)
 	select NLS
 
+config PCI_HYPERV
+        tristate "Hyper-V PCI Frontend"
+        depends on PCI && X86 && HYPERV && PCI_MSI && PCI_MSI_IRQ_DOMAIN && X86_64
+        help
+          The PCI device frontend driver allows the kernel to import arbitrary
+          PCI devices from a PCI backend to support PCI driver domains.
+
 source "drivers/pci/host/Kconfig"
diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile
index be3f631c3f75..2154092ddee8 100644
--- a/drivers/pci/Makefile
+++ b/drivers/pci/Makefile
@@ -32,6 +32,7 @@ obj-$(CONFIG_PCI_IOV) += iov.o
 # Some architectures use the generic PCI setup functions
 #
 obj-$(CONFIG_ALPHA) += setup-irq.o
+obj-$(CONFIG_ARC) += setup-irq.o
 obj-$(CONFIG_ARM) += setup-irq.o
 obj-$(CONFIG_ARM64) += setup-irq.o
 obj-$(CONFIG_UNICORE32) += setup-irq.o
diff --git a/drivers/pci/access.c b/drivers/pci/access.c
index 8c05b5ceeaec..01b9d0a00abc 100644
--- a/drivers/pci/access.c
+++ b/drivers/pci/access.c
@@ -174,38 +174,6 @@ struct pci_ops *pci_bus_set_ops(struct pci_bus *bus, struct pci_ops *ops)
 }
 EXPORT_SYMBOL(pci_bus_set_ops);
 
-/**
- * pci_read_vpd - Read one entry from Vital Product Data
- * @dev:	pci device struct
- * @pos:	offset in vpd space
- * @count:	number of bytes to read
- * @buf:	pointer to where to store result
- *
- */
-ssize_t pci_read_vpd(struct pci_dev *dev, loff_t pos, size_t count, void *buf)
-{
-	if (!dev->vpd || !dev->vpd->ops)
-		return -ENODEV;
-	return dev->vpd->ops->read(dev, pos, count, buf);
-}
-EXPORT_SYMBOL(pci_read_vpd);
-
-/**
- * pci_write_vpd - Write entry to Vital Product Data
- * @dev:	pci device struct
- * @pos:	offset in vpd space
- * @count:	number of bytes to write
- * @buf:	buffer containing write data
- *
- */
-ssize_t pci_write_vpd(struct pci_dev *dev, loff_t pos, size_t count, const void *buf)
-{
-	if (!dev->vpd || !dev->vpd->ops)
-		return -ENODEV;
-	return dev->vpd->ops->write(dev, pos, count, buf);
-}
-EXPORT_SYMBOL(pci_write_vpd);
-
 /*
  * The following routines are to prevent the user from accessing PCI config
  * space when it's unsafe to do so.  Some devices require this during BIST and
@@ -277,15 +245,91 @@ PCI_USER_WRITE_CONFIG(dword, u32)
 
 /* VPD access through PCI 2.2+ VPD capability */
 
-#define PCI_VPD_PCI22_SIZE (PCI_VPD_ADDR_MASK + 1)
+/**
+ * pci_read_vpd - Read one entry from Vital Product Data
+ * @dev:	pci device struct
+ * @pos:	offset in vpd space
+ * @count:	number of bytes to read
+ * @buf:	pointer to where to store result
+ */
+ssize_t pci_read_vpd(struct pci_dev *dev, loff_t pos, size_t count, void *buf)
+{
+	if (!dev->vpd || !dev->vpd->ops)
+		return -ENODEV;
+	return dev->vpd->ops->read(dev, pos, count, buf);
+}
+EXPORT_SYMBOL(pci_read_vpd);
 
-struct pci_vpd_pci22 {
-	struct pci_vpd base;
-	struct mutex lock;
-	u16	flag;
-	bool	busy;
-	u8	cap;
-};
+/**
+ * pci_write_vpd - Write entry to Vital Product Data
+ * @dev:	pci device struct
+ * @pos:	offset in vpd space
+ * @count:	number of bytes to write
+ * @buf:	buffer containing write data
+ */
+ssize_t pci_write_vpd(struct pci_dev *dev, loff_t pos, size_t count, const void *buf)
+{
+	if (!dev->vpd || !dev->vpd->ops)
+		return -ENODEV;
+	return dev->vpd->ops->write(dev, pos, count, buf);
+}
+EXPORT_SYMBOL(pci_write_vpd);
+
+#define PCI_VPD_MAX_SIZE (PCI_VPD_ADDR_MASK + 1)
+
+/**
+ * pci_vpd_size - determine actual size of Vital Product Data
+ * @dev:	pci device struct
+ * @old_size:	current assumed size, also maximum allowed size
+ */
+static size_t pci_vpd_size(struct pci_dev *dev, size_t old_size)
+{
+	size_t off = 0;
+	unsigned char header[1+2];	/* 1 byte tag, 2 bytes length */
+
+	while (off < old_size &&
+	       pci_read_vpd(dev, off, 1, header) == 1) {
+		unsigned char tag;
+
+		if (header[0] & PCI_VPD_LRDT) {
+			/* Large Resource Data Type Tag */
+			tag = pci_vpd_lrdt_tag(header);
+			/* Only read length from known tag items */
+			if ((tag == PCI_VPD_LTIN_ID_STRING) ||
+			    (tag == PCI_VPD_LTIN_RO_DATA) ||
+			    (tag == PCI_VPD_LTIN_RW_DATA)) {
+				if (pci_read_vpd(dev, off+1, 2,
+						 &header[1]) != 2) {
+					dev_warn(&dev->dev,
+						 "invalid large VPD tag %02x size at offset %zu",
+						 tag, off + 1);
+					return 0;
+				}
+				off += PCI_VPD_LRDT_TAG_SIZE +
+					pci_vpd_lrdt_size(header);
+			}
+		} else {
+			/* Short Resource Data Type Tag */
+			off += PCI_VPD_SRDT_TAG_SIZE +
+				pci_vpd_srdt_size(header);
+			tag = pci_vpd_srdt_tag(header);
+		}
+
+		if (tag == PCI_VPD_STIN_END)	/* End tag descriptor */
+			return off;
+
+		if ((tag != PCI_VPD_LTIN_ID_STRING) &&
+		    (tag != PCI_VPD_LTIN_RO_DATA) &&
+		    (tag != PCI_VPD_LTIN_RW_DATA)) {
+			dev_warn(&dev->dev,
+				 "invalid %s VPD tag %02x at offset %zu",
+				 (header[0] & PCI_VPD_LRDT) ? "large" : "short",
+				 tag, off);
+			return 0;
+		}
+	}
+	return 0;
+}
 
 /*
  * Wait for last operation to complete.
@@ -295,55 +339,71 @@ struct pci_vpd_pci22 {
  *
  * Returns 0 on success, negative values indicate error.
  */
-static int pci_vpd_pci22_wait(struct pci_dev *dev)
+static int pci_vpd_wait(struct pci_dev *dev)
 {
-	struct pci_vpd_pci22 *vpd =
-		container_of(dev->vpd, struct pci_vpd_pci22, base);
-	unsigned long timeout = jiffies + HZ/20 + 2;
+	struct pci_vpd *vpd = dev->vpd;
+	unsigned long timeout = jiffies + msecs_to_jiffies(50);
+	unsigned long max_sleep = 16;
 	u16 status;
 	int ret;
 
 	if (!vpd->busy)
 		return 0;
 
-	for (;;) {
+	while (time_before(jiffies, timeout)) {
 		ret = pci_user_read_config_word(dev, vpd->cap + PCI_VPD_ADDR,
 						&status);
 		if (ret < 0)
 			return ret;
 
 		if ((status & PCI_VPD_ADDR_F) == vpd->flag) {
-			vpd->busy = false;
+			vpd->busy = 0;
 			return 0;
 		}
 
-		if (time_after(jiffies, timeout)) {
-			dev_printk(KERN_DEBUG, &dev->dev, "vpd r/w failed.  This is likely a firmware bug on this device.  Contact the card vendor for a firmware update\n");
-			return -ETIMEDOUT;
-		}
 		if (fatal_signal_pending(current))
 			return -EINTR;
-		if (!cond_resched())
-			udelay(10);
+
+		usleep_range(10, max_sleep);
+		if (max_sleep < 1024)
+			max_sleep *= 2;
 	}
+
+	dev_warn(&dev->dev, "VPD access failed.  This is likely a firmware bug on this device.  Contact the card vendor for a firmware update\n");
+	return -ETIMEDOUT;
 }
 
-static ssize_t pci_vpd_pci22_read(struct pci_dev *dev, loff_t pos, size_t count,
-				  void *arg)
+static ssize_t pci_vpd_read(struct pci_dev *dev, loff_t pos, size_t count,
+			    void *arg)
 {
-	struct pci_vpd_pci22 *vpd =
-		container_of(dev->vpd, struct pci_vpd_pci22, base);
+	struct pci_vpd *vpd = dev->vpd;
 	int ret;
 	loff_t end = pos + count;
 	u8 *buf = arg;
 
-	if (pos < 0 || pos > vpd->base.len || end > vpd->base.len)
+	if (pos < 0)
 		return -EINVAL;
 
+	if (!vpd->valid) {
+		vpd->valid = 1;
+		vpd->len = pci_vpd_size(dev, vpd->len);
+	}
+
+	if (vpd->len == 0)
+		return -EIO;
+
+	if (pos > vpd->len)
+		return 0;
+
+	if (end > vpd->len) {
+		end = vpd->len;
+		count = end - pos;
+	}
+
 	if (mutex_lock_killable(&vpd->lock))
 		return -EINTR;
 
-	ret = pci_vpd_pci22_wait(dev);
+	ret = pci_vpd_wait(dev);
 	if (ret < 0)
 		goto out;
 
@@ -355,9 +415,9 @@ static ssize_t pci_vpd_pci22_read(struct pci_dev *dev, loff_t pos, size_t count,
 						 pos & ~3);
 		if (ret < 0)
 			break;
-		vpd->busy = true;
+		vpd->busy = 1;
 		vpd->flag = PCI_VPD_ADDR_F;
-		ret = pci_vpd_pci22_wait(dev);
+		ret = pci_vpd_wait(dev);
 		if (ret < 0)
 			break;
 
@@ -380,22 +440,32 @@ out:
 	return ret ? ret : count;
 }
 
-static ssize_t pci_vpd_pci22_write(struct pci_dev *dev, loff_t pos, size_t count,
-				   const void *arg)
+static ssize_t pci_vpd_write(struct pci_dev *dev, loff_t pos, size_t count,
+			     const void *arg)
 {
-	struct pci_vpd_pci22 *vpd =
-		container_of(dev->vpd, struct pci_vpd_pci22, base);
+	struct pci_vpd *vpd = dev->vpd;
 	const u8 *buf = arg;
 	loff_t end = pos + count;
 	int ret = 0;
 
-	if (pos < 0 || (pos & 3) || (count & 3) || end > vpd->base.len)
+	if (pos < 0 || (pos & 3) || (count & 3))
+		return -EINVAL;
+
+	if (!vpd->valid) {
+		vpd->valid = 1;
+		vpd->len = pci_vpd_size(dev, vpd->len);
+	}
+
+	if (vpd->len == 0)
+		return -EIO;
+
+	if (end > vpd->len)
 		return -EINVAL;
 
 	if (mutex_lock_killable(&vpd->lock))
 		return -EINTR;
 
-	ret = pci_vpd_pci22_wait(dev);
+	ret = pci_vpd_wait(dev);
 	if (ret < 0)
 		goto out;
 
@@ -415,9 +485,9 @@ static ssize_t pci_vpd_pci22_write(struct pci_dev *dev, loff_t pos, size_t count
 		if (ret < 0)
 			break;
 
-		vpd->busy = true;
+		vpd->busy = 1;
 		vpd->flag = 0;
-		ret = pci_vpd_pci22_wait(dev);
+		ret = pci_vpd_wait(dev);
 		if (ret < 0)
 			break;
 
@@ -428,15 +498,9 @@ out:
 	return ret ? ret : count;
 }
 
-static void pci_vpd_pci22_release(struct pci_dev *dev)
-{
-	kfree(container_of(dev->vpd, struct pci_vpd_pci22, base));
-}
-
-static const struct pci_vpd_ops pci_vpd_pci22_ops = {
-	.read = pci_vpd_pci22_read,
-	.write = pci_vpd_pci22_write,
-	.release = pci_vpd_pci22_release,
+static const struct pci_vpd_ops pci_vpd_ops = {
+	.read = pci_vpd_read,
+	.write = pci_vpd_write,
 };
 
 static ssize_t pci_vpd_f0_read(struct pci_dev *dev, loff_t pos, size_t count,
@@ -472,12 +536,11 @@ static ssize_t pci_vpd_f0_write(struct pci_dev *dev, loff_t pos, size_t count,
 static const struct pci_vpd_ops pci_vpd_f0_ops = {
 	.read = pci_vpd_f0_read,
 	.write = pci_vpd_f0_write,
-	.release = pci_vpd_pci22_release,
 };
 
-int pci_vpd_pci22_init(struct pci_dev *dev)
+int pci_vpd_init(struct pci_dev *dev)
 {
-	struct pci_vpd_pci22 *vpd;
+	struct pci_vpd *vpd;
 	u8 cap;
 
 	cap = pci_find_capability(dev, PCI_CAP_ID_VPD);
@@ -488,18 +551,24 @@ int pci_vpd_pci22_init(struct pci_dev *dev)
 	if (!vpd)
 		return -ENOMEM;
 
-	vpd->base.len = PCI_VPD_PCI22_SIZE;
+	vpd->len = PCI_VPD_MAX_SIZE;
 	if (dev->dev_flags & PCI_DEV_FLAGS_VPD_REF_F0)
-		vpd->base.ops = &pci_vpd_f0_ops;
+		vpd->ops = &pci_vpd_f0_ops;
 	else
-		vpd->base.ops = &pci_vpd_pci22_ops;
+		vpd->ops = &pci_vpd_ops;
 	mutex_init(&vpd->lock);
 	vpd->cap = cap;
-	vpd->busy = false;
-	dev->vpd = &vpd->base;
+	vpd->busy = 0;
+	vpd->valid = 0;
+	dev->vpd = vpd;
 	return 0;
 }
 
+void pci_vpd_release(struct pci_dev *dev)
+{
+	kfree(dev->vpd);
+}
+
 /**
  * pci_cfg_access_lock - Lock PCI config reads/writes
  * @dev:	pci device struct
diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c
index 89b3befc7155..f2187d491475 100644
--- a/drivers/pci/bus.c
+++ b/drivers/pci/bus.c
@@ -291,7 +291,12 @@ void pci_bus_add_device(struct pci_dev *dev)
 
 	dev->match_driver = true;
 	retval = device_attach(&dev->dev);
-	WARN_ON(retval < 0);
+	if (retval < 0) {
+		dev_warn(&dev->dev, "device attach failed (%d)\n", retval);
+		pci_proc_detach_device(dev);
+		pci_remove_sysfs_dev_files(dev);
+		return;
+	}
 
 	dev->is_added = 1;
 }
diff --git a/drivers/pci/host/Kconfig b/drivers/pci/host/Kconfig
index d1cdd9c992ac..c5014bf95a20 100644
--- a/drivers/pci/host/Kconfig
+++ b/drivers/pci/host/Kconfig
@@ -17,6 +17,28 @@ config PCI_MVEBU
 	depends on ARM
 	depends on OF
 
+
+config PCIE_XILINX_NWL
+	bool "NWL PCIe Core"
+	depends on ARCH_ZYNQMP
+	select PCI_MSI_IRQ_DOMAIN if PCI_MSI
+	help
+	 Say 'Y' here if you want kernel support for Xilinx
+	 NWL PCIe controller. The controller can act as Root Port
+	 or End Point. The current option selection will only
+	 support root port enabling.
+
+config PCIE_DW_PLAT
+	bool "Platform bus based DesignWare PCIe Controller"
+	select PCIE_DW
+	---help---
+	 This selects the DesignWare PCIe controller support. Select this if
+	 you have a PCIe controller on Platform bus.
+
+	 If you have a controller with this interface, say Y or M here.
+
+	 If unsure, say N.
+
 config PCIE_DW
 	bool
 
@@ -42,7 +64,7 @@ config PCI_TEGRA
 config PCI_RCAR_GEN2
 	bool "Renesas R-Car Gen2 Internal PCI controller"
 	depends on ARM
-	depends on ARCH_SHMOBILE || COMPILE_TEST
+	depends on ARCH_RENESAS || COMPILE_TEST
 	help
 	  Say Y here if you want internal PCI support on R-Car Gen2 SoC.
 	  There are 3 internal PCI controllers available with a single
@@ -50,13 +72,17 @@ config PCI_RCAR_GEN2
 
 config PCI_RCAR_GEN2_PCIE
 	bool "Renesas R-Car PCIe controller"
-	depends on ARCH_SHMOBILE || (ARM && COMPILE_TEST)
+	depends on ARCH_RENESAS || (ARM && COMPILE_TEST)
 	help
 	  Say Y here if you want PCIe controller support on R-Car Gen2 SoCs.
 
+config PCI_HOST_COMMON
+	bool
+
 config PCI_HOST_GENERIC
 	bool "Generic PCI host controller"
 	depends on (ARM || ARM64) && OF
+	select PCI_HOST_COMMON
 	help
 	  Say Y here if you want to support a simple generic PCI host
 	  controller, such as the one emulated by kvmtool.
@@ -82,7 +108,7 @@ config PCI_KEYSTONE
 
 config PCIE_XILINX
 	bool "Xilinx AXI PCIe host bridge support"
-	depends on ARCH_ZYNQ
+	depends on ARCH_ZYNQ || MICROBLAZE
 	help
 	  Say 'Y' here if you want kernel to support the Xilinx AXI PCIe
 	  Host Bridge driver.
@@ -192,4 +218,18 @@ config PCIE_QCOM
 	  PCIe controller uses the Designware core plus Qualcomm-specific
 	  hardware wrappers.
 
+config PCI_HOST_THUNDER_PEM
+	bool "Cavium Thunder PCIe controller to off-chip devices"
+	depends on OF && ARM64
+	select PCI_HOST_COMMON
+	help
+	  Say Y here if you want PCIe support for CN88XX Cavium Thunder SoCs.
+
+config PCI_HOST_THUNDER_ECAM
+	bool "Cavium Thunder ECAM controller to on-chip devices on pass-1.x silicon"
+	depends on OF && ARM64
+	select PCI_HOST_COMMON
+	help
+	  Say Y here if you want ECAM support for CN88XX-Pass-1.x Cavium Thunder SoCs.
+
 endmenu
diff --git a/drivers/pci/host/Makefile b/drivers/pci/host/Makefile
index 7b2f20c6ccc6..d85b5faf9bbc 100644
--- a/drivers/pci/host/Makefile
+++ b/drivers/pci/host/Makefile
@@ -1,15 +1,19 @@
 obj-$(CONFIG_PCIE_DW) += pcie-designware.o
+obj-$(CONFIG_PCIE_DW_PLAT) += pcie-designware-plat.o
 obj-$(CONFIG_PCI_DRA7XX) += pci-dra7xx.o
 obj-$(CONFIG_PCI_EXYNOS) += pci-exynos.o
 obj-$(CONFIG_PCI_IMX6) += pci-imx6.o
+obj-$(CONFIG_PCI_HYPERV) += pci-hyperv.o
 obj-$(CONFIG_PCI_MVEBU) += pci-mvebu.o
 obj-$(CONFIG_PCI_TEGRA) += pci-tegra.o
 obj-$(CONFIG_PCI_RCAR_GEN2) += pci-rcar-gen2.o
 obj-$(CONFIG_PCI_RCAR_GEN2_PCIE) += pcie-rcar.o
+obj-$(CONFIG_PCI_HOST_COMMON) += pci-host-common.o
 obj-$(CONFIG_PCI_HOST_GENERIC) += pci-host-generic.o
 obj-$(CONFIG_PCIE_SPEAR13XX) += pcie-spear13xx.o
 obj-$(CONFIG_PCI_KEYSTONE) += pci-keystone-dw.o pci-keystone.o
 obj-$(CONFIG_PCIE_XILINX) += pcie-xilinx.o
+obj-$(CONFIG_PCIE_XILINX_NWL) += pcie-xilinx-nwl.o
 obj-$(CONFIG_PCI_XGENE) += pci-xgene.o
 obj-$(CONFIG_PCI_XGENE_MSI) += pci-xgene-msi.o
 obj-$(CONFIG_PCI_LAYERSCAPE) += pci-layerscape.o
@@ -22,3 +26,5 @@ obj-$(CONFIG_PCIE_ALTERA) += pcie-altera.o
 obj-$(CONFIG_PCIE_ALTERA_MSI) += pcie-altera-msi.o
 obj-$(CONFIG_PCI_HISI) += pcie-hisi.o
 obj-$(CONFIG_PCIE_QCOM) += pcie-qcom.o
+obj-$(CONFIG_PCI_HOST_THUNDER_ECAM) += pci-thunder-ecam.o
+obj-$(CONFIG_PCI_HOST_THUNDER_PEM) += pci-thunder-pem.o
diff --git a/drivers/pci/host/pci-dra7xx.c b/drivers/pci/host/pci-dra7xx.c
index 923607bdabc5..2ca3a1f30ebf 100644
--- a/drivers/pci/host/pci-dra7xx.c
+++ b/drivers/pci/host/pci-dra7xx.c
@@ -10,7 +10,6 @@
  * published by the Free Software Foundation.
  */
 
-#include <linux/delay.h>
 #include <linux/err.h>
 #include <linux/interrupt.h>
 #include <linux/irq.h>
@@ -108,7 +107,6 @@ static int dra7xx_pcie_establish_link(struct pcie_port *pp)
 {
 	struct dra7xx_pcie *dra7xx = to_dra7xx_pcie(pp);
 	u32 reg;
-	unsigned int retries;
 
 	if (dw_pcie_link_up(pp)) {
 		dev_err(pp->dev, "link is already up\n");
@@ -119,14 +117,7 @@ static int dra7xx_pcie_establish_link(struct pcie_port *pp)
 	reg |= LTSSM_EN;
 	dra7xx_pcie_writel(dra7xx, PCIECTRL_DRA7XX_CONF_DEVICE_CMD, reg);
 
-	for (retries = 0; retries < 1000; retries++) {
-		if (dw_pcie_link_up(pp))
-			return 0;
-		usleep_range(10, 20);
-	}
-
-	dev_err(pp->dev, "link is not up\n");
-	return -EINVAL;
+	return dw_pcie_wait_for_link(pp);
 }
 
 static void dra7xx_pcie_enable_interrupts(struct pcie_port *pp)
diff --git a/drivers/pci/host/pci-exynos.c b/drivers/pci/host/pci-exynos.c
index d997d22d4231..219976103efc 100644
--- a/drivers/pci/host/pci-exynos.c
+++ b/drivers/pci/host/pci-exynos.c
@@ -318,7 +318,6 @@ static int exynos_pcie_establish_link(struct pcie_port *pp)
 {
 	struct exynos_pcie *exynos_pcie = to_exynos_pcie(pp);
 	u32 val;
-	unsigned int retries;
 
 	if (dw_pcie_link_up(pp)) {
 		dev_err(pp->dev, "Link already up\n");
@@ -357,13 +356,8 @@ static int exynos_pcie_establish_link(struct pcie_port *pp)
 			  PCIE_APP_LTSSM_ENABLE);
 
 	/* check if the link is up or not */
-	for (retries = 0; retries < 10; retries++) {
-		if (dw_pcie_link_up(pp)) {
-			dev_info(pp->dev, "Link up\n");
-			return 0;
-		}
-		mdelay(100);
-	}
+	if (!dw_pcie_wait_for_link(pp))
+		return 0;
 
 	while (exynos_phy_readl(exynos_pcie, PCIE_PHY_PLL_LOCKED) == 0) {
 		val = exynos_blk_readl(exynos_pcie, PCIE_PHY_PLL_LOCKED);
@@ -372,8 +366,7 @@ static int exynos_pcie_establish_link(struct pcie_port *pp)
 	/* power off phy */
 	exynos_pcie_power_off_phy(pp);
 
-	dev_err(pp->dev, "PCIe Link Fail\n");
-	return -EINVAL;
+	return -ETIMEDOUT;
 }
 
 static void exynos_pcie_clear_irq_pulse(struct pcie_port *pp)
diff --git a/drivers/pci/host/pci-host-common.c b/drivers/pci/host/pci-host-common.c
new file mode 100644
index 000000000000..e9f850f07968
--- /dev/null
+++ b/drivers/pci/host/pci-host-common.c
@@ -0,0 +1,194 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Copyright (C) 2014 ARM Limited
+ *
+ * Author: Will Deacon <will.deacon@arm.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of_address.h>
+#include <linux/of_pci.h>
+#include <linux/platform_device.h>
+
+#include "pci-host-common.h"
+
+static void gen_pci_release_of_pci_ranges(struct gen_pci *pci)
+{
+	pci_free_resource_list(&pci->resources);
+}
+
+static int gen_pci_parse_request_of_pci_ranges(struct gen_pci *pci)
+{
+	int err, res_valid = 0;
+	struct device *dev = pci->host.dev.parent;
+	struct device_node *np = dev->of_node;
+	resource_size_t iobase;
+	struct resource_entry *win;
+
+	err = of_pci_get_host_bridge_resources(np, 0, 0xff, &pci->resources,
+					       &iobase);
+	if (err)
+		return err;
+
+	resource_list_for_each_entry(win, &pci->resources) {
+		struct resource *parent, *res = win->res;
+
+		switch (resource_type(res)) {
+		case IORESOURCE_IO:
+			parent = &ioport_resource;
+			err = pci_remap_iospace(res, iobase);
+			if (err) {
+				dev_warn(dev, "error %d: failed to map resource %pR\n",
+					 err, res);
+				continue;
+			}
+			break;
+		case IORESOURCE_MEM:
+			parent = &iomem_resource;
+			res_valid |= !(res->flags & IORESOURCE_PREFETCH);
+			break;
+		case IORESOURCE_BUS:
+			pci->cfg.bus_range = res;
+		default:
+			continue;
+		}
+
+		err = devm_request_resource(dev, parent, res);
+		if (err)
+			goto out_release_res;
+	}
+
+	if (!res_valid) {
+		dev_err(dev, "non-prefetchable memory resource required\n");
+		err = -EINVAL;
+		goto out_release_res;
+	}
+
+	return 0;
+
+out_release_res:
+	gen_pci_release_of_pci_ranges(pci);
+	return err;
+}
+
+static int gen_pci_parse_map_cfg_windows(struct gen_pci *pci)
+{
+	int err;
+	u8 bus_max;
+	resource_size_t busn;
+	struct resource *bus_range;
+	struct device *dev = pci->host.dev.parent;
+	struct device_node *np = dev->of_node;
+	u32 sz = 1 << pci->cfg.ops->bus_shift;
+
+	err = of_address_to_resource(np, 0, &pci->cfg.res);
+	if (err) {
+		dev_err(dev, "missing \"reg\" property\n");
+		return err;
+	}
+
+	/* Limit the bus-range to fit within reg */
+	bus_max = pci->cfg.bus_range->start +
+		  (resource_size(&pci->cfg.res) >> pci->cfg.ops->bus_shift) - 1;
+	pci->cfg.bus_range->end = min_t(resource_size_t,
+					pci->cfg.bus_range->end, bus_max);
+
+	pci->cfg.win = devm_kcalloc(dev, resource_size(pci->cfg.bus_range),
+				    sizeof(*pci->cfg.win), GFP_KERNEL);
+	if (!pci->cfg.win)
+		return -ENOMEM;
+
+	/* Map our Configuration Space windows */
+	if (!devm_request_mem_region(dev, pci->cfg.res.start,
+				     resource_size(&pci->cfg.res),
+				     "Configuration Space"))
+		return -ENOMEM;
+
+	bus_range = pci->cfg.bus_range;
+	for (busn = bus_range->start; busn <= bus_range->end; ++busn) {
+		u32 idx = busn - bus_range->start;
+
+		pci->cfg.win[idx] = devm_ioremap(dev,
+						 pci->cfg.res.start + idx * sz,
+						 sz);
+		if (!pci->cfg.win[idx])
+			return -ENOMEM;
+	}
+
+	return 0;
+}
+
+int pci_host_common_probe(struct platform_device *pdev,
+			  struct gen_pci *pci)
+{
+	int err;
+	const char *type;
+	struct device *dev = &pdev->dev;
+	struct device_node *np = dev->of_node;
+	struct pci_bus *bus, *child;
+
+	type = of_get_property(np, "device_type", NULL);
+	if (!type || strcmp(type, "pci")) {
+		dev_err(dev, "invalid \"device_type\" %s\n", type);
+		return -EINVAL;
+	}
+
+	of_pci_check_probe_only();
+
+	pci->host.dev.parent = dev;
+	INIT_LIST_HEAD(&pci->host.windows);
+	INIT_LIST_HEAD(&pci->resources);
+
+	/* Parse our PCI ranges and request their resources */
+	err = gen_pci_parse_request_of_pci_ranges(pci);
+	if (err)
+		return err;
+
+	/* Parse and map our Configuration Space windows */
+	err = gen_pci_parse_map_cfg_windows(pci);
+	if (err) {
+		gen_pci_release_of_pci_ranges(pci);
+		return err;
+	}
+
+	/* Do not reassign resources if probe only */
+	if (!pci_has_flag(PCI_PROBE_ONLY))
+		pci_add_flags(PCI_REASSIGN_ALL_RSRC | PCI_REASSIGN_ALL_BUS);
+
+
+	bus = pci_scan_root_bus(dev, pci->cfg.bus_range->start,
+				&pci->cfg.ops->ops, pci, &pci->resources);
+	if (!bus) {
+		dev_err(dev, "Scanning rootbus failed");
+		return -ENODEV;
+	}
+
+	pci_fixup_irqs(pci_common_swizzle, of_irq_parse_and_map_pci);
+
+	if (!pci_has_flag(PCI_PROBE_ONLY)) {
+		pci_bus_size_bridges(bus);
+		pci_bus_assign_resources(bus);
+
+		list_for_each_entry(child, &bus->children, node)
+			pcie_bus_configure_settings(child);
+	}
+
+	pci_bus_add_devices(bus);
+	return 0;
+}
+
+MODULE_DESCRIPTION("Generic PCI host driver common code");
+MODULE_AUTHOR("Will Deacon <will.deacon@arm.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/pci/host/pci-host-common.h b/drivers/pci/host/pci-host-common.h
new file mode 100644
index 000000000000..09f3fa0a55d7
--- /dev/null
+++ b/drivers/pci/host/pci-host-common.h
@@ -0,0 +1,47 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Copyright (C) 2014 ARM Limited
+ *
+ * Author: Will Deacon <will.deacon@arm.com>
+ */
+
+#ifndef _PCI_HOST_COMMON_H
+#define _PCI_HOST_COMMON_H
+
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+
+struct gen_pci_cfg_bus_ops {
+	u32 bus_shift;
+	struct pci_ops ops;
+};
+
+struct gen_pci_cfg_windows {
+	struct resource				res;
+	struct resource				*bus_range;
+	void __iomem				**win;
+
+	struct gen_pci_cfg_bus_ops		*ops;
+};
+
+struct gen_pci {
+	struct pci_host_bridge			host;
+	struct gen_pci_cfg_windows		cfg;
+	struct list_head			resources;
+};
+
+int pci_host_common_probe(struct platform_device *pdev,
+			  struct gen_pci *pci);
+
+#endif /* _PCI_HOST_COMMON_H */
diff --git a/drivers/pci/host/pci-host-generic.c b/drivers/pci/host/pci-host-generic.c
index 1652bc70b145..e8aa78faa16d 100644
--- a/drivers/pci/host/pci-host-generic.c
+++ b/drivers/pci/host/pci-host-generic.c
@@ -25,24 +25,7 @@
 #include <linux/of_pci.h>
 #include <linux/platform_device.h>
 
-struct gen_pci_cfg_bus_ops {
-	u32 bus_shift;
-	struct pci_ops ops;
-};
-
-struct gen_pci_cfg_windows {
-	struct resource				res;
-	struct resource				*bus_range;
-	void __iomem				**win;
-
-	struct gen_pci_cfg_bus_ops		*ops;
-};
-
-struct gen_pci {
-	struct pci_host_bridge			host;
-	struct gen_pci_cfg_windows		cfg;
-	struct list_head			resources;
-};
+#include "pci-host-common.h"
 
 static void __iomem *gen_pci_map_cfg_bus_cam(struct pci_bus *bus,
 					     unsigned int devfn,
@@ -93,175 +76,19 @@ static const struct of_device_id gen_pci_of_match[] = {
 };
 MODULE_DEVICE_TABLE(of, gen_pci_of_match);
 
-static void gen_pci_release_of_pci_ranges(struct gen_pci *pci)
-{
-	pci_free_resource_list(&pci->resources);
-}
-
-static int gen_pci_parse_request_of_pci_ranges(struct gen_pci *pci)
-{
-	int err, res_valid = 0;
-	struct device *dev = pci->host.dev.parent;
-	struct device_node *np = dev->of_node;
-	resource_size_t iobase;
-	struct resource_entry *win;
-
-	err = of_pci_get_host_bridge_resources(np, 0, 0xff, &pci->resources,
-					       &iobase);
-	if (err)
-		return err;
-
-	resource_list_for_each_entry(win, &pci->resources) {
-		struct resource *parent, *res = win->res;
-
-		switch (resource_type(res)) {
-		case IORESOURCE_IO:
-			parent = &ioport_resource;
-			err = pci_remap_iospace(res, iobase);
-			if (err) {
-				dev_warn(dev, "error %d: failed to map resource %pR\n",
-					 err, res);
-				continue;
-			}
-			break;
-		case IORESOURCE_MEM:
-			parent = &iomem_resource;
-			res_valid |= !(res->flags & IORESOURCE_PREFETCH);
-			break;
-		case IORESOURCE_BUS:
-			pci->cfg.bus_range = res;
-		default:
-			continue;
-		}
-
-		err = devm_request_resource(dev, parent, res);
-		if (err)
-			goto out_release_res;
-	}
-
-	if (!res_valid) {
-		dev_err(dev, "non-prefetchable memory resource required\n");
-		err = -EINVAL;
-		goto out_release_res;
-	}
-
-	return 0;
-
-out_release_res:
-	gen_pci_release_of_pci_ranges(pci);
-	return err;
-}
-
-static int gen_pci_parse_map_cfg_windows(struct gen_pci *pci)
-{
-	int err;
-	u8 bus_max;
-	resource_size_t busn;
-	struct resource *bus_range;
-	struct device *dev = pci->host.dev.parent;
-	struct device_node *np = dev->of_node;
-	u32 sz = 1 << pci->cfg.ops->bus_shift;
-
-	err = of_address_to_resource(np, 0, &pci->cfg.res);
-	if (err) {
-		dev_err(dev, "missing \"reg\" property\n");
-		return err;
-	}
-
-	/* Limit the bus-range to fit within reg */
-	bus_max = pci->cfg.bus_range->start +
-		  (resource_size(&pci->cfg.res) >> pci->cfg.ops->bus_shift) - 1;
-	pci->cfg.bus_range->end = min_t(resource_size_t,
-					pci->cfg.bus_range->end, bus_max);
-
-	pci->cfg.win = devm_kcalloc(dev, resource_size(pci->cfg.bus_range),
-				    sizeof(*pci->cfg.win), GFP_KERNEL);
-	if (!pci->cfg.win)
-		return -ENOMEM;
-
-	/* Map our Configuration Space windows */
-	if (!devm_request_mem_region(dev, pci->cfg.res.start,
-				     resource_size(&pci->cfg.res),
-				     "Configuration Space"))
-		return -ENOMEM;
-
-	bus_range = pci->cfg.bus_range;
-	for (busn = bus_range->start; busn <= bus_range->end; ++busn) {
-		u32 idx = busn - bus_range->start;
-
-		pci->cfg.win[idx] = devm_ioremap(dev,
-						 pci->cfg.res.start + idx * sz,
-						 sz);
-		if (!pci->cfg.win[idx])
-			return -ENOMEM;
-	}
-
-	return 0;
-}
-
 static int gen_pci_probe(struct platform_device *pdev)
 {
-	int err;
-	const char *type;
-	const struct of_device_id *of_id;
 	struct device *dev = &pdev->dev;
-	struct device_node *np = dev->of_node;
+	const struct of_device_id *of_id;
 	struct gen_pci *pci = devm_kzalloc(dev, sizeof(*pci), GFP_KERNEL);
-	struct pci_bus *bus, *child;
 
 	if (!pci)
 		return -ENOMEM;
 
-	type = of_get_property(np, "device_type", NULL);
-	if (!type || strcmp(type, "pci")) {
-		dev_err(dev, "invalid \"device_type\" %s\n", type);
-		return -EINVAL;
-	}
-
-	of_pci_check_probe_only();
-
-	of_id = of_match_node(gen_pci_of_match, np);
+	of_id = of_match_node(gen_pci_of_match, dev->of_node);
 	pci->cfg.ops = (struct gen_pci_cfg_bus_ops *)of_id->data;
-	pci->host.dev.parent = dev;
-	INIT_LIST_HEAD(&pci->host.windows);
-	INIT_LIST_HEAD(&pci->resources);
-
-	/* Parse our PCI ranges and request their resources */
-	err = gen_pci_parse_request_of_pci_ranges(pci);
-	if (err)
-		return err;
-
-	/* Parse and map our Configuration Space windows */
-	err = gen_pci_parse_map_cfg_windows(pci);
-	if (err) {
-		gen_pci_release_of_pci_ranges(pci);
-		return err;
-	}
-
-	/* Do not reassign resources if probe only */
-	if (!pci_has_flag(PCI_PROBE_ONLY))
-		pci_add_flags(PCI_REASSIGN_ALL_RSRC | PCI_REASSIGN_ALL_BUS);
-
-
-	bus = pci_scan_root_bus(dev, pci->cfg.bus_range->start,
-				&pci->cfg.ops->ops, pci, &pci->resources);
-	if (!bus) {
-		dev_err(dev, "Scanning rootbus failed");
-		return -ENODEV;
-	}
-
-	pci_fixup_irqs(pci_common_swizzle, of_irq_parse_and_map_pci);
-
-	if (!pci_has_flag(PCI_PROBE_ONLY)) {
-		pci_bus_size_bridges(bus);
-		pci_bus_assign_resources(bus);
-
-		list_for_each_entry(child, &bus->children, node)
-			pcie_bus_configure_settings(child);
-	}
 
-	pci_bus_add_devices(bus);
-	return 0;
+	return pci_host_common_probe(pdev, pci);
 }
 
 static struct platform_driver gen_pci_driver = {
diff --git a/drivers/pci/host/pci-hyperv.c b/drivers/pci/host/pci-hyperv.c
new file mode 100644
index 000000000000..ed651baa7c50
--- /dev/null
+++ b/drivers/pci/host/pci-hyperv.c
@@ -0,0 +1,2346 @@
+/*
+ * Copyright (c) Microsoft Corporation.
+ *
+ * Author:
+ *   Jake Oshins <jakeo@microsoft.com>
+ *
+ * This driver acts as a paravirtual front-end for PCI Express root buses.
+ * When a PCI Express function (either an entire device or an SR-IOV
+ * Virtual Function) is being passed through to the VM, this driver exposes
+ * a new bus to the guest VM.  This is modeled as a root PCI bus because
+ * no bridges are being exposed to the VM.  In fact, with a "Generation 2"
+ * VM within Hyper-V, there may seem to be no PCI bus at all in the VM
+ * until a device as been exposed using this driver.
+ *
+ * Each root PCI bus has its own PCI domain, which is called "Segment" in
+ * the PCI Firmware Specifications.  Thus while each device passed through
+ * to the VM using this front-end will appear at "device 0", the domain will
+ * be unique.  Typically, each bus will have one PCI function on it, though
+ * this driver does support more than one.
+ *
+ * In order to map the interrupts from the device through to the guest VM,
+ * this driver also implements an IRQ Domain, which handles interrupts (either
+ * MSI or MSI-X) associated with the functions on the bus.  As interrupts are
+ * set up, torn down, or reaffined, this driver communicates with the
+ * underlying hypervisor to adjust the mappings in the I/O MMU so that each
+ * interrupt will be delivered to the correct virtual processor at the right
+ * vector.  This driver does not support level-triggered (line-based)
+ * interrupts, and will report that the Interrupt Line register in the
+ * function's configuration space is zero.
+ *
+ * The rest of this driver mostly maps PCI concepts onto underlying Hyper-V
+ * facilities.  For instance, the configuration space of a function exposed
+ * by Hyper-V is mapped into a single page of memory space, and the
+ * read and write handlers for config space must be aware of this mechanism.
+ * Similarly, device setup and teardown involves messages sent to and from
+ * the PCI back-end driver in Hyper-V.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT.  See the GNU General Public License for more
+ * details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/semaphore.h>
+#include <linux/irqdomain.h>
+#include <asm/irqdomain.h>
+#include <asm/apic.h>
+#include <linux/msi.h>
+#include <linux/hyperv.h>
+#include <asm/mshyperv.h>
+
+/*
+ * Protocol versions. The low word is the minor version, the high word the
+ * major version.
+ */
+
+#define PCI_MAKE_VERSION(major, minor) ((u32)(((major) << 16) | (major)))
+#define PCI_MAJOR_VERSION(version) ((u32)(version) >> 16)
+#define PCI_MINOR_VERSION(version) ((u32)(version) & 0xff)
+
+enum {
+	PCI_PROTOCOL_VERSION_1_1 = PCI_MAKE_VERSION(1, 1),
+	PCI_PROTOCOL_VERSION_CURRENT = PCI_PROTOCOL_VERSION_1_1
+};
+
+#define PCI_CONFIG_MMIO_LENGTH	0x2000
+#define CFG_PAGE_OFFSET 0x1000
+#define CFG_PAGE_SIZE (PCI_CONFIG_MMIO_LENGTH - CFG_PAGE_OFFSET)
+
+#define MAX_SUPPORTED_MSI_MESSAGES 0x400
+
+/*
+ * Message Types
+ */
+
+enum pci_message_type {
+	/*
+	 * Version 1.1
+	 */
+	PCI_MESSAGE_BASE                = 0x42490000,
+	PCI_BUS_RELATIONS               = PCI_MESSAGE_BASE + 0,
+	PCI_QUERY_BUS_RELATIONS         = PCI_MESSAGE_BASE + 1,
+	PCI_POWER_STATE_CHANGE          = PCI_MESSAGE_BASE + 4,
+	PCI_QUERY_RESOURCE_REQUIREMENTS = PCI_MESSAGE_BASE + 5,
+	PCI_QUERY_RESOURCE_RESOURCES    = PCI_MESSAGE_BASE + 6,
+	PCI_BUS_D0ENTRY                 = PCI_MESSAGE_BASE + 7,
+	PCI_BUS_D0EXIT                  = PCI_MESSAGE_BASE + 8,
+	PCI_READ_BLOCK                  = PCI_MESSAGE_BASE + 9,
+	PCI_WRITE_BLOCK                 = PCI_MESSAGE_BASE + 0xA,
+	PCI_EJECT                       = PCI_MESSAGE_BASE + 0xB,
+	PCI_QUERY_STOP                  = PCI_MESSAGE_BASE + 0xC,
+	PCI_REENABLE                    = PCI_MESSAGE_BASE + 0xD,
+	PCI_QUERY_STOP_FAILED           = PCI_MESSAGE_BASE + 0xE,
+	PCI_EJECTION_COMPLETE           = PCI_MESSAGE_BASE + 0xF,
+	PCI_RESOURCES_ASSIGNED          = PCI_MESSAGE_BASE + 0x10,
+	PCI_RESOURCES_RELEASED          = PCI_MESSAGE_BASE + 0x11,
+	PCI_INVALIDATE_BLOCK            = PCI_MESSAGE_BASE + 0x12,
+	PCI_QUERY_PROTOCOL_VERSION      = PCI_MESSAGE_BASE + 0x13,
+	PCI_CREATE_INTERRUPT_MESSAGE    = PCI_MESSAGE_BASE + 0x14,
+	PCI_DELETE_INTERRUPT_MESSAGE    = PCI_MESSAGE_BASE + 0x15,
+	PCI_MESSAGE_MAXIMUM
+};
+
+/*
+ * Structures defining the virtual PCI Express protocol.
+ */
+
+union pci_version {
+	struct {
+		u16 minor_version;
+		u16 major_version;
+	} parts;
+	u32 version;
+} __packed;
+
+/*
+ * Function numbers are 8-bits wide on Express, as interpreted through ARI,
+ * which is all this driver does.  This representation is the one used in
+ * Windows, which is what is expected when sending this back and forth with
+ * the Hyper-V parent partition.
+ */
+union win_slot_encoding {
+	struct {
+		u32	func:8;
+		u32	reserved:24;
+	} bits;
+	u32 slot;
+} __packed;
+
+/*
+ * Pretty much as defined in the PCI Specifications.
+ */
+struct pci_function_description {
+	u16	v_id;	/* vendor ID */
+	u16	d_id;	/* device ID */
+	u8	rev;
+	u8	prog_intf;
+	u8	subclass;
+	u8	base_class;
+	u32	subsystem_id;
+	union win_slot_encoding win_slot;
+	u32	ser;	/* serial number */
+} __packed;
+
+/**
+ * struct hv_msi_desc
+ * @vector:		IDT entry
+ * @delivery_mode:	As defined in Intel's Programmer's
+ *			Reference Manual, Volume 3, Chapter 8.
+ * @vector_count:	Number of contiguous entries in the
+ *			Interrupt Descriptor Table that are
+ *			occupied by this Message-Signaled
+ *			Interrupt. For "MSI", as first defined
+ *			in PCI 2.2, this can be between 1 and
+ *			32. For "MSI-X," as first defined in PCI
+ *			3.0, this must be 1, as each MSI-X table
+ *			entry would have its own descriptor.
+ * @reserved:		Empty space
+ * @cpu_mask:		All the target virtual processors.
+ */
+struct hv_msi_desc {
+	u8	vector;
+	u8	delivery_mode;
+	u16	vector_count;
+	u32	reserved;
+	u64	cpu_mask;
+} __packed;
+
+/**
+ * struct tran_int_desc
+ * @reserved:		unused, padding
+ * @vector_count:	same as in hv_msi_desc
+ * @data:		This is the "data payload" value that is
+ *			written by the device when it generates
+ *			a message-signaled interrupt, either MSI
+ *			or MSI-X.
+ * @address:		This is the address to which the data
+ *			payload is written on interrupt
+ *			generation.
+ */
+struct tran_int_desc {
+	u16	reserved;
+	u16	vector_count;
+	u32	data;
+	u64	address;
+} __packed;
+
+/*
+ * A generic message format for virtual PCI.
+ * Specific message formats are defined later in the file.
+ */
+
+struct pci_message {
+	u32 message_type;
+} __packed;
+
+struct pci_child_message {
+	u32 message_type;
+	union win_slot_encoding wslot;
+} __packed;
+
+struct pci_incoming_message {
+	struct vmpacket_descriptor hdr;
+	struct pci_message message_type;
+} __packed;
+
+struct pci_response {
+	struct vmpacket_descriptor hdr;
+	s32 status;			/* negative values are failures */
+} __packed;
+
+struct pci_packet {
+	void (*completion_func)(void *context, struct pci_response *resp,
+				int resp_packet_size);
+	void *compl_ctxt;
+	struct pci_message message;
+};
+
+/*
+ * Specific message types supporting the PCI protocol.
+ */
+
+/*
+ * Version negotiation message. Sent from the guest to the host.
+ * The guest is free to try different versions until the host
+ * accepts the version.
+ *
+ * pci_version: The protocol version requested.
+ * is_last_attempt: If TRUE, this is the last version guest will request.
+ * reservedz: Reserved field, set to zero.
+ */
+
+struct pci_version_request {
+	struct pci_message message_type;
+	enum pci_message_type protocol_version;
+} __packed;
+
+/*
+ * Bus D0 Entry.  This is sent from the guest to the host when the virtual
+ * bus (PCI Express port) is ready for action.
+ */
+
+struct pci_bus_d0_entry {
+	struct pci_message message_type;
+	u32 reserved;
+	u64 mmio_base;
+} __packed;
+
+struct pci_bus_relations {
+	struct pci_incoming_message incoming;
+	u32 device_count;
+	struct pci_function_description func[1];
+} __packed;
+
+struct pci_q_res_req_response {
+	struct vmpacket_descriptor hdr;
+	s32 status;			/* negative values are failures */
+	u32 probed_bar[6];
+} __packed;
+
+struct pci_set_power {
+	struct pci_message message_type;
+	union win_slot_encoding wslot;
+	u32 power_state;		/* In Windows terms */
+	u32 reserved;
+} __packed;
+
+struct pci_set_power_response {
+	struct vmpacket_descriptor hdr;
+	s32 status;			/* negative values are failures */
+	union win_slot_encoding wslot;
+	u32 resultant_state;		/* In Windows terms */
+	u32 reserved;
+} __packed;
+
+struct pci_resources_assigned {
+	struct pci_message message_type;
+	union win_slot_encoding wslot;
+	u8 memory_range[0x14][6];	/* not used here */
+	u32 msi_descriptors;
+	u32 reserved[4];
+} __packed;
+
+struct pci_create_interrupt {
+	struct pci_message message_type;
+	union win_slot_encoding wslot;
+	struct hv_msi_desc int_desc;
+} __packed;
+
+struct pci_create_int_response {
+	struct pci_response response;
+	u32 reserved;
+	struct tran_int_desc int_desc;
+} __packed;
+
+struct pci_delete_interrupt {
+	struct pci_message message_type;
+	union win_slot_encoding wslot;
+	struct tran_int_desc int_desc;
+} __packed;
+
+struct pci_dev_incoming {
+	struct pci_incoming_message incoming;
+	union win_slot_encoding wslot;
+} __packed;
+
+struct pci_eject_response {
+	u32 message_type;
+	union win_slot_encoding wslot;
+	u32 status;
+} __packed;
+
+static int pci_ring_size = (4 * PAGE_SIZE);
+
+/*
+ * Definitions or interrupt steering hypercall.
+ */
+#define HV_PARTITION_ID_SELF		((u64)-1)
+#define HVCALL_RETARGET_INTERRUPT	0x7e
+
+struct retarget_msi_interrupt {
+	u64	partition_id;		/* use "self" */
+	u64	device_id;
+	u32	source;			/* 1 for MSI(-X) */
+	u32	reserved1;
+	u32	address;
+	u32	data;
+	u64	reserved2;
+	u32	vector;
+	u32	flags;
+	u64	vp_mask;
+} __packed;
+
+/*
+ * Driver specific state.
+ */
+
+enum hv_pcibus_state {
+	hv_pcibus_init = 0,
+	hv_pcibus_probed,
+	hv_pcibus_installed,
+	hv_pcibus_maximum
+};
+
+struct hv_pcibus_device {
+	struct pci_sysdata sysdata;
+	enum hv_pcibus_state state;
+	atomic_t remove_lock;
+	struct hv_device *hdev;
+	resource_size_t low_mmio_space;
+	resource_size_t high_mmio_space;
+	struct resource *mem_config;
+	struct resource *low_mmio_res;
+	struct resource *high_mmio_res;
+	struct completion *survey_event;
+	struct completion remove_event;
+	struct pci_bus *pci_bus;
+	spinlock_t config_lock;	/* Avoid two threads writing index page */
+	spinlock_t device_list_lock;	/* Protect lists below */
+	void __iomem *cfg_addr;
+
+	struct semaphore enum_sem;
+	struct list_head resources_for_children;
+
+	struct list_head children;
+	struct list_head dr_list;
+	struct work_struct wrk;
+
+	struct msi_domain_info msi_info;
+	struct msi_controller msi_chip;
+	struct irq_domain *irq_domain;
+};
+
+/*
+ * Tracks "Device Relations" messages from the host, which must be both
+ * processed in order and deferred so that they don't run in the context
+ * of the incoming packet callback.
+ */
+struct hv_dr_work {
+	struct work_struct wrk;
+	struct hv_pcibus_device *bus;
+};
+
+struct hv_dr_state {
+	struct list_head list_entry;
+	u32 device_count;
+	struct pci_function_description func[1];
+};
+
+enum hv_pcichild_state {
+	hv_pcichild_init = 0,
+	hv_pcichild_requirements,
+	hv_pcichild_resourced,
+	hv_pcichild_ejecting,
+	hv_pcichild_maximum
+};
+
+enum hv_pcidev_ref_reason {
+	hv_pcidev_ref_invalid = 0,
+	hv_pcidev_ref_initial,
+	hv_pcidev_ref_by_slot,
+	hv_pcidev_ref_packet,
+	hv_pcidev_ref_pnp,
+	hv_pcidev_ref_childlist,
+	hv_pcidev_irqdata,
+	hv_pcidev_ref_max
+};
+
+struct hv_pci_dev {
+	/* List protected by pci_rescan_remove_lock */
+	struct list_head list_entry;
+	atomic_t refs;
+	enum hv_pcichild_state state;
+	struct pci_function_description desc;
+	bool reported_missing;
+	struct hv_pcibus_device *hbus;
+	struct work_struct wrk;
+
+	/*
+	 * What would be observed if one wrote 0xFFFFFFFF to a BAR and then
+	 * read it back, for each of the BAR offsets within config space.
+	 */
+	u32 probed_bar[6];
+};
+
+struct hv_pci_compl {
+	struct completion host_event;
+	s32 completion_status;
+};
+
+/**
+ * hv_pci_generic_compl() - Invoked for a completion packet
+ * @context:		Set up by the sender of the packet.
+ * @resp:		The response packet
+ * @resp_packet_size:	Size in bytes of the packet
+ *
+ * This function is used to trigger an event and report status
+ * for any message for which the completion packet contains a
+ * status and nothing else.
+ */
+static
+void
+hv_pci_generic_compl(void *context, struct pci_response *resp,
+		     int resp_packet_size)
+{
+	struct hv_pci_compl *comp_pkt = context;
+
+	if (resp_packet_size >= offsetofend(struct pci_response, status))
+		comp_pkt->completion_status = resp->status;
+	complete(&comp_pkt->host_event);
+}
+
+static struct hv_pci_dev *get_pcichild_wslot(struct hv_pcibus_device *hbus,
+						u32 wslot);
+static void get_pcichild(struct hv_pci_dev *hv_pcidev,
+			 enum hv_pcidev_ref_reason reason);
+static void put_pcichild(struct hv_pci_dev *hv_pcidev,
+			 enum hv_pcidev_ref_reason reason);
+
+static void get_hvpcibus(struct hv_pcibus_device *hv_pcibus);
+static void put_hvpcibus(struct hv_pcibus_device *hv_pcibus);
+
+/**
+ * devfn_to_wslot() - Convert from Linux PCI slot to Windows
+ * @devfn:	The Linux representation of PCI slot
+ *
+ * Windows uses a slightly different representation of PCI slot.
+ *
+ * Return: The Windows representation
+ */
+static u32 devfn_to_wslot(int devfn)
+{
+	union win_slot_encoding wslot;
+
+	wslot.slot = 0;
+	wslot.bits.func = PCI_SLOT(devfn) | (PCI_FUNC(devfn) << 5);
+
+	return wslot.slot;
+}
+
+/**
+ * wslot_to_devfn() - Convert from Windows PCI slot to Linux
+ * @wslot:	The Windows representation of PCI slot
+ *
+ * Windows uses a slightly different representation of PCI slot.
+ *
+ * Return: The Linux representation
+ */
+static int wslot_to_devfn(u32 wslot)
+{
+	union win_slot_encoding slot_no;
+
+	slot_no.slot = wslot;
+	return PCI_DEVFN(0, slot_no.bits.func);
+}
+
+/*
+ * PCI Configuration Space for these root PCI buses is implemented as a pair
+ * of pages in memory-mapped I/O space.  Writing to the first page chooses
+ * the PCI function being written or read.  Once the first page has been
+ * written to, the following page maps in the entire configuration space of
+ * the function.
+ */
+
+/**
+ * _hv_pcifront_read_config() - Internal PCI config read
+ * @hpdev:	The PCI driver's representation of the device
+ * @where:	Offset within config space
+ * @size:	Size of the transfer
+ * @val:	Pointer to the buffer receiving the data
+ */
+static void _hv_pcifront_read_config(struct hv_pci_dev *hpdev, int where,
+				     int size, u32 *val)
+{
+	unsigned long flags;
+	void __iomem *addr = hpdev->hbus->cfg_addr + CFG_PAGE_OFFSET + where;
+
+	/*
+	 * If the attempt is to read the IDs or the ROM BAR, simulate that.
+	 */
+	if (where + size <= PCI_COMMAND) {
+		memcpy(val, ((u8 *)&hpdev->desc.v_id) + where, size);
+	} else if (where >= PCI_CLASS_REVISION && where + size <=
+		   PCI_CACHE_LINE_SIZE) {
+		memcpy(val, ((u8 *)&hpdev->desc.rev) + where -
+		       PCI_CLASS_REVISION, size);
+	} else if (where >= PCI_SUBSYSTEM_VENDOR_ID && where + size <=
+		   PCI_ROM_ADDRESS) {
+		memcpy(val, (u8 *)&hpdev->desc.subsystem_id + where -
+		       PCI_SUBSYSTEM_VENDOR_ID, size);
+	} else if (where >= PCI_ROM_ADDRESS && where + size <=
+		   PCI_CAPABILITY_LIST) {
+		/* ROM BARs are unimplemented */
+		*val = 0;
+	} else if (where >= PCI_INTERRUPT_LINE && where + size <=
+		   PCI_INTERRUPT_PIN) {
+		/*
+		 * Interrupt Line and Interrupt PIN are hard-wired to zero
+		 * because this front-end only supports message-signaled
+		 * interrupts.
+		 */
+		*val = 0;
+	} else if (where + size <= CFG_PAGE_SIZE) {
+		spin_lock_irqsave(&hpdev->hbus->config_lock, flags);
+		/* Choose the function to be read. (See comment above) */
+		writel(hpdev->desc.win_slot.slot, hpdev->hbus->cfg_addr);
+		/* Read from that function's config space. */
+		switch (size) {
+		case 1:
+			*val = readb(addr);
+			break;
+		case 2:
+			*val = readw(addr);
+			break;
+		default:
+			*val = readl(addr);
+			break;
+		}
+		spin_unlock_irqrestore(&hpdev->hbus->config_lock, flags);
+	} else {
+		dev_err(&hpdev->hbus->hdev->device,
+			"Attempt to read beyond a function's config space.\n");
+	}
+}
+
+/**
+ * _hv_pcifront_write_config() - Internal PCI config write
+ * @hpdev:	The PCI driver's representation of the device
+ * @where:	Offset within config space
+ * @size:	Size of the transfer
+ * @val:	The data being transferred
+ */
+static void _hv_pcifront_write_config(struct hv_pci_dev *hpdev, int where,
+				      int size, u32 val)
+{
+	unsigned long flags;
+	void __iomem *addr = hpdev->hbus->cfg_addr + CFG_PAGE_OFFSET + where;
+
+	if (where >= PCI_SUBSYSTEM_VENDOR_ID &&
+	    where + size <= PCI_CAPABILITY_LIST) {
+		/* SSIDs and ROM BARs are read-only */
+	} else if (where >= PCI_COMMAND && where + size <= CFG_PAGE_SIZE) {
+		spin_lock_irqsave(&hpdev->hbus->config_lock, flags);
+		/* Choose the function to be written. (See comment above) */
+		writel(hpdev->desc.win_slot.slot, hpdev->hbus->cfg_addr);
+		/* Write to that function's config space. */
+		switch (size) {
+		case 1:
+			writeb(val, addr);
+			break;
+		case 2:
+			writew(val, addr);
+			break;
+		default:
+			writel(val, addr);
+			break;
+		}
+		spin_unlock_irqrestore(&hpdev->hbus->config_lock, flags);
+	} else {
+		dev_err(&hpdev->hbus->hdev->device,
+			"Attempt to write beyond a function's config space.\n");
+	}
+}
+
+/**
+ * hv_pcifront_read_config() - Read configuration space
+ * @bus: PCI Bus structure
+ * @devfn: Device/function
+ * @where: Offset from base
+ * @size: Byte/word/dword
+ * @val: Value to be read
+ *
+ * Return: PCIBIOS_SUCCESSFUL on success
+ *	   PCIBIOS_DEVICE_NOT_FOUND on failure
+ */
+static int hv_pcifront_read_config(struct pci_bus *bus, unsigned int devfn,
+				   int where, int size, u32 *val)
+{
+	struct hv_pcibus_device *hbus =
+		container_of(bus->sysdata, struct hv_pcibus_device, sysdata);
+	struct hv_pci_dev *hpdev;
+
+	hpdev = get_pcichild_wslot(hbus, devfn_to_wslot(devfn));
+	if (!hpdev)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	_hv_pcifront_read_config(hpdev, where, size, val);
+
+	put_pcichild(hpdev, hv_pcidev_ref_by_slot);
+	return PCIBIOS_SUCCESSFUL;
+}
+
+/**
+ * hv_pcifront_write_config() - Write configuration space
+ * @bus: PCI Bus structure
+ * @devfn: Device/function
+ * @where: Offset from base
+ * @size: Byte/word/dword
+ * @val: Value to be written to device
+ *
+ * Return: PCIBIOS_SUCCESSFUL on success
+ *	   PCIBIOS_DEVICE_NOT_FOUND on failure
+ */
+static int hv_pcifront_write_config(struct pci_bus *bus, unsigned int devfn,
+				    int where, int size, u32 val)
+{
+	struct hv_pcibus_device *hbus =
+	    container_of(bus->sysdata, struct hv_pcibus_device, sysdata);
+	struct hv_pci_dev *hpdev;
+
+	hpdev = get_pcichild_wslot(hbus, devfn_to_wslot(devfn));
+	if (!hpdev)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	_hv_pcifront_write_config(hpdev, where, size, val);
+
+	put_pcichild(hpdev, hv_pcidev_ref_by_slot);
+	return PCIBIOS_SUCCESSFUL;
+}
+
+/* PCIe operations */
+static struct pci_ops hv_pcifront_ops = {
+	.read  = hv_pcifront_read_config,
+	.write = hv_pcifront_write_config,
+};
+
+/* Interrupt management hooks */
+static void hv_int_desc_free(struct hv_pci_dev *hpdev,
+			     struct tran_int_desc *int_desc)
+{
+	struct pci_delete_interrupt *int_pkt;
+	struct {
+		struct pci_packet pkt;
+		u8 buffer[sizeof(struct pci_delete_interrupt) -
+			  sizeof(struct pci_message)];
+	} ctxt;
+
+	memset(&ctxt, 0, sizeof(ctxt));
+	int_pkt = (struct pci_delete_interrupt *)&ctxt.pkt.message;
+	int_pkt->message_type.message_type =
+		PCI_DELETE_INTERRUPT_MESSAGE;
+	int_pkt->wslot.slot = hpdev->desc.win_slot.slot;
+	int_pkt->int_desc = *int_desc;
+	vmbus_sendpacket(hpdev->hbus->hdev->channel, int_pkt, sizeof(*int_pkt),
+			 (unsigned long)&ctxt.pkt, VM_PKT_DATA_INBAND, 0);
+	kfree(int_desc);
+}
+
+/**
+ * hv_msi_free() - Free the MSI.
+ * @domain:	The interrupt domain pointer
+ * @info:	Extra MSI-related context
+ * @irq:	Identifies the IRQ.
+ *
+ * The Hyper-V parent partition and hypervisor are tracking the
+ * messages that are in use, keeping the interrupt redirection
+ * table up to date.  This callback sends a message that frees
+ * the IRT entry and related tracking nonsense.
+ */
+static void hv_msi_free(struct irq_domain *domain, struct msi_domain_info *info,
+			unsigned int irq)
+{
+	struct hv_pcibus_device *hbus;
+	struct hv_pci_dev *hpdev;
+	struct pci_dev *pdev;
+	struct tran_int_desc *int_desc;
+	struct irq_data *irq_data = irq_domain_get_irq_data(domain, irq);
+	struct msi_desc *msi = irq_data_get_msi_desc(irq_data);
+
+	pdev = msi_desc_to_pci_dev(msi);
+	hbus = info->data;
+	hpdev = get_pcichild_wslot(hbus, devfn_to_wslot(pdev->devfn));
+	if (!hpdev)
+		return;
+
+	int_desc = irq_data_get_irq_chip_data(irq_data);
+	if (int_desc) {
+		irq_data->chip_data = NULL;
+		hv_int_desc_free(hpdev, int_desc);
+	}
+
+	put_pcichild(hpdev, hv_pcidev_ref_by_slot);
+}
+
+static int hv_set_affinity(struct irq_data *data, const struct cpumask *dest,
+			   bool force)
+{
+	struct irq_data *parent = data->parent_data;
+
+	return parent->chip->irq_set_affinity(parent, dest, force);
+}
+
+void hv_irq_mask(struct irq_data *data)
+{
+	pci_msi_mask_irq(data);
+}
+
+/**
+ * hv_irq_unmask() - "Unmask" the IRQ by setting its current
+ * affinity.
+ * @data:	Describes the IRQ
+ *
+ * Build new a destination for the MSI and make a hypercall to
+ * update the Interrupt Redirection Table. "Device Logical ID"
+ * is built out of this PCI bus's instance GUID and the function
+ * number of the device.
+ */
+void hv_irq_unmask(struct irq_data *data)
+{
+	struct msi_desc *msi_desc = irq_data_get_msi_desc(data);
+	struct irq_cfg *cfg = irqd_cfg(data);
+	struct retarget_msi_interrupt params;
+	struct hv_pcibus_device *hbus;
+	struct cpumask *dest;
+	struct pci_bus *pbus;
+	struct pci_dev *pdev;
+	int cpu;
+
+	dest = irq_data_get_affinity_mask(data);
+	pdev = msi_desc_to_pci_dev(msi_desc);
+	pbus = pdev->bus;
+	hbus = container_of(pbus->sysdata, struct hv_pcibus_device, sysdata);
+
+	memset(&params, 0, sizeof(params));
+	params.partition_id = HV_PARTITION_ID_SELF;
+	params.source = 1; /* MSI(-X) */
+	params.address = msi_desc->msg.address_lo;
+	params.data = msi_desc->msg.data;
+	params.device_id = (hbus->hdev->dev_instance.b[5] << 24) |
+			   (hbus->hdev->dev_instance.b[4] << 16) |
+			   (hbus->hdev->dev_instance.b[7] << 8) |
+			   (hbus->hdev->dev_instance.b[6] & 0xf8) |
+			   PCI_FUNC(pdev->devfn);
+	params.vector = cfg->vector;
+
+	for_each_cpu_and(cpu, dest, cpu_online_mask)
+		params.vp_mask |= (1ULL << vmbus_cpu_number_to_vp_number(cpu));
+
+	hv_do_hypercall(HVCALL_RETARGET_INTERRUPT, &params, NULL);
+
+	pci_msi_unmask_irq(data);
+}
+
+struct compose_comp_ctxt {
+	struct hv_pci_compl comp_pkt;
+	struct tran_int_desc int_desc;
+};
+
+static void hv_pci_compose_compl(void *context, struct pci_response *resp,
+				 int resp_packet_size)
+{
+	struct compose_comp_ctxt *comp_pkt = context;
+	struct pci_create_int_response *int_resp =
+		(struct pci_create_int_response *)resp;
+
+	comp_pkt->comp_pkt.completion_status = resp->status;
+	comp_pkt->int_desc = int_resp->int_desc;
+	complete(&comp_pkt->comp_pkt.host_event);
+}
+
+/**
+ * hv_compose_msi_msg() - Supplies a valid MSI address/data
+ * @data:	Everything about this MSI
+ * @msg:	Buffer that is filled in by this function
+ *
+ * This function unpacks the IRQ looking for target CPU set, IDT
+ * vector and mode and sends a message to the parent partition
+ * asking for a mapping for that tuple in this partition.  The
+ * response supplies a data value and address to which that data
+ * should be written to trigger that interrupt.
+ */
+static void hv_compose_msi_msg(struct irq_data *data, struct msi_msg *msg)
+{
+	struct irq_cfg *cfg = irqd_cfg(data);
+	struct hv_pcibus_device *hbus;
+	struct hv_pci_dev *hpdev;
+	struct pci_bus *pbus;
+	struct pci_dev *pdev;
+	struct pci_create_interrupt *int_pkt;
+	struct compose_comp_ctxt comp;
+	struct tran_int_desc *int_desc;
+	struct cpumask *affinity;
+	struct {
+		struct pci_packet pkt;
+		u8 buffer[sizeof(struct pci_create_interrupt) -
+			  sizeof(struct pci_message)];
+	} ctxt;
+	int cpu;
+	int ret;
+
+	pdev = msi_desc_to_pci_dev(irq_data_get_msi_desc(data));
+	pbus = pdev->bus;
+	hbus = container_of(pbus->sysdata, struct hv_pcibus_device, sysdata);
+	hpdev = get_pcichild_wslot(hbus, devfn_to_wslot(pdev->devfn));
+	if (!hpdev)
+		goto return_null_message;
+
+	/* Free any previous message that might have already been composed. */
+	if (data->chip_data) {
+		int_desc = data->chip_data;
+		data->chip_data = NULL;
+		hv_int_desc_free(hpdev, int_desc);
+	}
+
+	int_desc = kzalloc(sizeof(*int_desc), GFP_KERNEL);
+	if (!int_desc)
+		goto drop_reference;
+
+	memset(&ctxt, 0, sizeof(ctxt));
+	init_completion(&comp.comp_pkt.host_event);
+	ctxt.pkt.completion_func = hv_pci_compose_compl;
+	ctxt.pkt.compl_ctxt = &comp;
+	int_pkt = (struct pci_create_interrupt *)&ctxt.pkt.message;
+	int_pkt->message_type.message_type = PCI_CREATE_INTERRUPT_MESSAGE;
+	int_pkt->wslot.slot = hpdev->desc.win_slot.slot;
+	int_pkt->int_desc.vector = cfg->vector;
+	int_pkt->int_desc.vector_count = 1;
+	int_pkt->int_desc.delivery_mode =
+		(apic->irq_delivery_mode == dest_LowestPrio) ? 1 : 0;
+
+	/*
+	 * This bit doesn't have to work on machines with more than 64
+	 * processors because Hyper-V only supports 64 in a guest.
+	 */
+	affinity = irq_data_get_affinity_mask(data);
+	for_each_cpu_and(cpu, affinity, cpu_online_mask) {
+		int_pkt->int_desc.cpu_mask |=
+			(1ULL << vmbus_cpu_number_to_vp_number(cpu));
+	}
+
+	ret = vmbus_sendpacket(hpdev->hbus->hdev->channel, int_pkt,
+			       sizeof(*int_pkt), (unsigned long)&ctxt.pkt,
+			       VM_PKT_DATA_INBAND,
+			       VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
+	if (!ret)
+		wait_for_completion(&comp.comp_pkt.host_event);
+
+	if (comp.comp_pkt.completion_status < 0) {
+		dev_err(&hbus->hdev->device,
+			"Request for interrupt failed: 0x%x",
+			comp.comp_pkt.completion_status);
+		goto free_int_desc;
+	}
+
+	/*
+	 * Record the assignment so that this can be unwound later. Using
+	 * irq_set_chip_data() here would be appropriate, but the lock it takes
+	 * is already held.
+	 */
+	*int_desc = comp.int_desc;
+	data->chip_data = int_desc;
+
+	/* Pass up the result. */
+	msg->address_hi = comp.int_desc.address >> 32;
+	msg->address_lo = comp.int_desc.address & 0xffffffff;
+	msg->data = comp.int_desc.data;
+
+	put_pcichild(hpdev, hv_pcidev_ref_by_slot);
+	return;
+
+free_int_desc:
+	kfree(int_desc);
+drop_reference:
+	put_pcichild(hpdev, hv_pcidev_ref_by_slot);
+return_null_message:
+	msg->address_hi = 0;
+	msg->address_lo = 0;
+	msg->data = 0;
+}
+
+/* HW Interrupt Chip Descriptor */
+static struct irq_chip hv_msi_irq_chip = {
+	.name			= "Hyper-V PCIe MSI",
+	.irq_compose_msi_msg	= hv_compose_msi_msg,
+	.irq_set_affinity	= hv_set_affinity,
+	.irq_ack		= irq_chip_ack_parent,
+	.irq_mask		= hv_irq_mask,
+	.irq_unmask		= hv_irq_unmask,
+};
+
+static irq_hw_number_t hv_msi_domain_ops_get_hwirq(struct msi_domain_info *info,
+						   msi_alloc_info_t *arg)
+{
+	return arg->msi_hwirq;
+}
+
+static struct msi_domain_ops hv_msi_ops = {
+	.get_hwirq	= hv_msi_domain_ops_get_hwirq,
+	.msi_prepare	= pci_msi_prepare,
+	.set_desc	= pci_msi_set_desc,
+	.msi_free	= hv_msi_free,
+};
+
+/**
+ * hv_pcie_init_irq_domain() - Initialize IRQ domain
+ * @hbus:	The root PCI bus
+ *
+ * This function creates an IRQ domain which will be used for
+ * interrupts from devices that have been passed through.  These
+ * devices only support MSI and MSI-X, not line-based interrupts
+ * or simulations of line-based interrupts through PCIe's
+ * fabric-layer messages.  Because interrupts are remapped, we
+ * can support multi-message MSI here.
+ *
+ * Return: '0' on success and error value on failure
+ */
+static int hv_pcie_init_irq_domain(struct hv_pcibus_device *hbus)
+{
+	hbus->msi_info.chip = &hv_msi_irq_chip;
+	hbus->msi_info.ops = &hv_msi_ops;
+	hbus->msi_info.flags = (MSI_FLAG_USE_DEF_DOM_OPS |
+		MSI_FLAG_USE_DEF_CHIP_OPS | MSI_FLAG_MULTI_PCI_MSI |
+		MSI_FLAG_PCI_MSIX);
+	hbus->msi_info.handler = handle_edge_irq;
+	hbus->msi_info.handler_name = "edge";
+	hbus->msi_info.data = hbus;
+	hbus->irq_domain = pci_msi_create_irq_domain(hbus->sysdata.fwnode,
+						     &hbus->msi_info,
+						     x86_vector_domain);
+	if (!hbus->irq_domain) {
+		dev_err(&hbus->hdev->device,
+			"Failed to build an MSI IRQ domain\n");
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+/**
+ * get_bar_size() - Get the address space consumed by a BAR
+ * @bar_val:	Value that a BAR returned after -1 was written
+ *              to it.
+ *
+ * This function returns the size of the BAR, rounded up to 1
+ * page.  It has to be rounded up because the hypervisor's page
+ * table entry that maps the BAR into the VM can't specify an
+ * offset within a page.  The invariant is that the hypervisor
+ * must place any BARs of smaller than page length at the
+ * beginning of a page.
+ *
+ * Return:	Size in bytes of the consumed MMIO space.
+ */
+static u64 get_bar_size(u64 bar_val)
+{
+	return round_up((1 + ~(bar_val & PCI_BASE_ADDRESS_MEM_MASK)),
+			PAGE_SIZE);
+}
+
+/**
+ * survey_child_resources() - Total all MMIO requirements
+ * @hbus:	Root PCI bus, as understood by this driver
+ */
+static void survey_child_resources(struct hv_pcibus_device *hbus)
+{
+	struct list_head *iter;
+	struct hv_pci_dev *hpdev;
+	resource_size_t bar_size = 0;
+	unsigned long flags;
+	struct completion *event;
+	u64 bar_val;
+	int i;
+
+	/* If nobody is waiting on the answer, don't compute it. */
+	event = xchg(&hbus->survey_event, NULL);
+	if (!event)
+		return;
+
+	/* If the answer has already been computed, go with it. */
+	if (hbus->low_mmio_space || hbus->high_mmio_space) {
+		complete(event);
+		return;
+	}
+
+	spin_lock_irqsave(&hbus->device_list_lock, flags);
+
+	/*
+	 * Due to an interesting quirk of the PCI spec, all memory regions
+	 * for a child device are a power of 2 in size and aligned in memory,
+	 * so it's sufficient to just add them up without tracking alignment.
+	 */
+	list_for_each(iter, &hbus->children) {
+		hpdev = container_of(iter, struct hv_pci_dev, list_entry);
+		for (i = 0; i < 6; i++) {
+			if (hpdev->probed_bar[i] & PCI_BASE_ADDRESS_SPACE_IO)
+				dev_err(&hbus->hdev->device,
+					"There's an I/O BAR in this list!\n");
+
+			if (hpdev->probed_bar[i] != 0) {
+				/*
+				 * A probed BAR has all the upper bits set that
+				 * can be changed.
+				 */
+
+				bar_val = hpdev->probed_bar[i];
+				if (bar_val & PCI_BASE_ADDRESS_MEM_TYPE_64)
+					bar_val |=
+					((u64)hpdev->probed_bar[++i] << 32);
+				else
+					bar_val |= 0xffffffff00000000ULL;
+
+				bar_size = get_bar_size(bar_val);
+
+				if (bar_val & PCI_BASE_ADDRESS_MEM_TYPE_64)
+					hbus->high_mmio_space += bar_size;
+				else
+					hbus->low_mmio_space += bar_size;
+			}
+		}
+	}
+
+	spin_unlock_irqrestore(&hbus->device_list_lock, flags);
+	complete(event);
+}
+
+/**
+ * prepopulate_bars() - Fill in BARs with defaults
+ * @hbus:	Root PCI bus, as understood by this driver
+ *
+ * The core PCI driver code seems much, much happier if the BARs
+ * for a device have values upon first scan. So fill them in.
+ * The algorithm below works down from large sizes to small,
+ * attempting to pack the assignments optimally. The assumption,
+ * enforced in other parts of the code, is that the beginning of
+ * the memory-mapped I/O space will be aligned on the largest
+ * BAR size.
+ */
+static void prepopulate_bars(struct hv_pcibus_device *hbus)
+{
+	resource_size_t high_size = 0;
+	resource_size_t low_size = 0;
+	resource_size_t high_base = 0;
+	resource_size_t low_base = 0;
+	resource_size_t bar_size;
+	struct hv_pci_dev *hpdev;
+	struct list_head *iter;
+	unsigned long flags;
+	u64 bar_val;
+	u32 command;
+	bool high;
+	int i;
+
+	if (hbus->low_mmio_space) {
+		low_size = 1ULL << (63 - __builtin_clzll(hbus->low_mmio_space));
+		low_base = hbus->low_mmio_res->start;
+	}
+
+	if (hbus->high_mmio_space) {
+		high_size = 1ULL <<
+			(63 - __builtin_clzll(hbus->high_mmio_space));
+		high_base = hbus->high_mmio_res->start;
+	}
+
+	spin_lock_irqsave(&hbus->device_list_lock, flags);
+
+	/* Pick addresses for the BARs. */
+	do {
+		list_for_each(iter, &hbus->children) {
+			hpdev = container_of(iter, struct hv_pci_dev,
+					     list_entry);
+			for (i = 0; i < 6; i++) {
+				bar_val = hpdev->probed_bar[i];
+				if (bar_val == 0)
+					continue;
+				high = bar_val & PCI_BASE_ADDRESS_MEM_TYPE_64;
+				if (high) {
+					bar_val |=
+						((u64)hpdev->probed_bar[i + 1]
+						 << 32);
+				} else {
+					bar_val |= 0xffffffffULL << 32;
+				}
+				bar_size = get_bar_size(bar_val);
+				if (high) {
+					if (high_size != bar_size) {
+						i++;
+						continue;
+					}
+					_hv_pcifront_write_config(hpdev,
+						PCI_BASE_ADDRESS_0 + (4 * i),
+						4,
+						(u32)(high_base & 0xffffff00));
+					i++;
+					_hv_pcifront_write_config(hpdev,
+						PCI_BASE_ADDRESS_0 + (4 * i),
+						4, (u32)(high_base >> 32));
+					high_base += bar_size;
+				} else {
+					if (low_size != bar_size)
+						continue;
+					_hv_pcifront_write_config(hpdev,
+						PCI_BASE_ADDRESS_0 + (4 * i),
+						4,
+						(u32)(low_base & 0xffffff00));
+					low_base += bar_size;
+				}
+			}
+			if (high_size <= 1 && low_size <= 1) {
+				/* Set the memory enable bit. */
+				_hv_pcifront_read_config(hpdev, PCI_COMMAND, 2,
+							 &command);
+				command |= PCI_COMMAND_MEMORY;
+				_hv_pcifront_write_config(hpdev, PCI_COMMAND, 2,
+							  command);
+				break;
+			}
+		}
+
+		high_size >>= 1;
+		low_size >>= 1;
+	}  while (high_size || low_size);
+
+	spin_unlock_irqrestore(&hbus->device_list_lock, flags);
+}
+
+/**
+ * create_root_hv_pci_bus() - Expose a new root PCI bus
+ * @hbus:	Root PCI bus, as understood by this driver
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int create_root_hv_pci_bus(struct hv_pcibus_device *hbus)
+{
+	/* Register the device */
+	hbus->pci_bus = pci_create_root_bus(&hbus->hdev->device,
+					    0, /* bus number is always zero */
+					    &hv_pcifront_ops,
+					    &hbus->sysdata,
+					    &hbus->resources_for_children);
+	if (!hbus->pci_bus)
+		return -ENODEV;
+
+	hbus->pci_bus->msi = &hbus->msi_chip;
+	hbus->pci_bus->msi->dev = &hbus->hdev->device;
+
+	pci_scan_child_bus(hbus->pci_bus);
+	pci_bus_assign_resources(hbus->pci_bus);
+	pci_bus_add_devices(hbus->pci_bus);
+	hbus->state = hv_pcibus_installed;
+	return 0;
+}
+
+struct q_res_req_compl {
+	struct completion host_event;
+	struct hv_pci_dev *hpdev;
+};
+
+/**
+ * q_resource_requirements() - Query Resource Requirements
+ * @context:		The completion context.
+ * @resp:		The response that came from the host.
+ * @resp_packet_size:	The size in bytes of resp.
+ *
+ * This function is invoked on completion of a Query Resource
+ * Requirements packet.
+ */
+static void q_resource_requirements(void *context, struct pci_response *resp,
+				    int resp_packet_size)
+{
+	struct q_res_req_compl *completion = context;
+	struct pci_q_res_req_response *q_res_req =
+		(struct pci_q_res_req_response *)resp;
+	int i;
+
+	if (resp->status < 0) {
+		dev_err(&completion->hpdev->hbus->hdev->device,
+			"query resource requirements failed: %x\n",
+			resp->status);
+	} else {
+		for (i = 0; i < 6; i++) {
+			completion->hpdev->probed_bar[i] =
+				q_res_req->probed_bar[i];
+		}
+	}
+
+	complete(&completion->host_event);
+}
+
+static void get_pcichild(struct hv_pci_dev *hpdev,
+			    enum hv_pcidev_ref_reason reason)
+{
+	atomic_inc(&hpdev->refs);
+}
+
+static void put_pcichild(struct hv_pci_dev *hpdev,
+			    enum hv_pcidev_ref_reason reason)
+{
+	if (atomic_dec_and_test(&hpdev->refs))
+		kfree(hpdev);
+}
+
+/**
+ * new_pcichild_device() - Create a new child device
+ * @hbus:	The internal struct tracking this root PCI bus.
+ * @desc:	The information supplied so far from the host
+ *              about the device.
+ *
+ * This function creates the tracking structure for a new child
+ * device and kicks off the process of figuring out what it is.
+ *
+ * Return: Pointer to the new tracking struct
+ */
+static struct hv_pci_dev *new_pcichild_device(struct hv_pcibus_device *hbus,
+		struct pci_function_description *desc)
+{
+	struct hv_pci_dev *hpdev;
+	struct pci_child_message *res_req;
+	struct q_res_req_compl comp_pkt;
+	union {
+	struct pci_packet init_packet;
+		u8 buffer[0x100];
+	} pkt;
+	unsigned long flags;
+	int ret;
+
+	hpdev = kzalloc(sizeof(*hpdev), GFP_ATOMIC);
+	if (!hpdev)
+		return NULL;
+
+	hpdev->hbus = hbus;
+
+	memset(&pkt, 0, sizeof(pkt));
+	init_completion(&comp_pkt.host_event);
+	comp_pkt.hpdev = hpdev;
+	pkt.init_packet.compl_ctxt = &comp_pkt;
+	pkt.init_packet.completion_func = q_resource_requirements;
+	res_req = (struct pci_child_message *)&pkt.init_packet.message;
+	res_req->message_type = PCI_QUERY_RESOURCE_REQUIREMENTS;
+	res_req->wslot.slot = desc->win_slot.slot;
+
+	ret = vmbus_sendpacket(hbus->hdev->channel, res_req,
+			       sizeof(struct pci_child_message),
+			       (unsigned long)&pkt.init_packet,
+			       VM_PKT_DATA_INBAND,
+			       VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
+	if (ret)
+		goto error;
+
+	wait_for_completion(&comp_pkt.host_event);
+
+	hpdev->desc = *desc;
+	get_pcichild(hpdev, hv_pcidev_ref_initial);
+	get_pcichild(hpdev, hv_pcidev_ref_childlist);
+	spin_lock_irqsave(&hbus->device_list_lock, flags);
+	list_add_tail(&hpdev->list_entry, &hbus->children);
+	spin_unlock_irqrestore(&hbus->device_list_lock, flags);
+	return hpdev;
+
+error:
+	kfree(hpdev);
+	return NULL;
+}
+
+/**
+ * get_pcichild_wslot() - Find device from slot
+ * @hbus:	Root PCI bus, as understood by this driver
+ * @wslot:	Location on the bus
+ *
+ * This function looks up a PCI device and returns the internal
+ * representation of it.  It acquires a reference on it, so that
+ * the device won't be deleted while somebody is using it.  The
+ * caller is responsible for calling put_pcichild() to release
+ * this reference.
+ *
+ * Return:	Internal representation of a PCI device
+ */
+static struct hv_pci_dev *get_pcichild_wslot(struct hv_pcibus_device *hbus,
+					     u32 wslot)
+{
+	unsigned long flags;
+	struct hv_pci_dev *iter, *hpdev = NULL;
+
+	spin_lock_irqsave(&hbus->device_list_lock, flags);
+	list_for_each_entry(iter, &hbus->children, list_entry) {
+		if (iter->desc.win_slot.slot == wslot) {
+			hpdev = iter;
+			get_pcichild(hpdev, hv_pcidev_ref_by_slot);
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&hbus->device_list_lock, flags);
+
+	return hpdev;
+}
+
+/**
+ * pci_devices_present_work() - Handle new list of child devices
+ * @work:	Work struct embedded in struct hv_dr_work
+ *
+ * "Bus Relations" is the Windows term for "children of this
+ * bus."  The terminology is preserved here for people trying to
+ * debug the interaction between Hyper-V and Linux.  This
+ * function is called when the parent partition reports a list
+ * of functions that should be observed under this PCI Express
+ * port (bus).
+ *
+ * This function updates the list, and must tolerate being
+ * called multiple times with the same information.  The typical
+ * number of child devices is one, with very atypical cases
+ * involving three or four, so the algorithms used here can be
+ * simple and inefficient.
+ *
+ * It must also treat the omission of a previously observed device as
+ * notification that the device no longer exists.
+ *
+ * Note that this function is a work item, and it may not be
+ * invoked in the order that it was queued.  Back to back
+ * updates of the list of present devices may involve queuing
+ * multiple work items, and this one may run before ones that
+ * were sent later. As such, this function only does something
+ * if is the last one in the queue.
+ */
+static void pci_devices_present_work(struct work_struct *work)
+{
+	u32 child_no;
+	bool found;
+	struct list_head *iter;
+	struct pci_function_description *new_desc;
+	struct hv_pci_dev *hpdev;
+	struct hv_pcibus_device *hbus;
+	struct list_head removed;
+	struct hv_dr_work *dr_wrk;
+	struct hv_dr_state *dr = NULL;
+	unsigned long flags;
+
+	dr_wrk = container_of(work, struct hv_dr_work, wrk);
+	hbus = dr_wrk->bus;
+	kfree(dr_wrk);
+
+	INIT_LIST_HEAD(&removed);
+
+	if (down_interruptible(&hbus->enum_sem)) {
+		put_hvpcibus(hbus);
+		return;
+	}
+
+	/* Pull this off the queue and process it if it was the last one. */
+	spin_lock_irqsave(&hbus->device_list_lock, flags);
+	while (!list_empty(&hbus->dr_list)) {
+		dr = list_first_entry(&hbus->dr_list, struct hv_dr_state,
+				      list_entry);
+		list_del(&dr->list_entry);
+
+		/* Throw this away if the list still has stuff in it. */
+		if (!list_empty(&hbus->dr_list)) {
+			kfree(dr);
+			continue;
+		}
+	}
+	spin_unlock_irqrestore(&hbus->device_list_lock, flags);
+
+	if (!dr) {
+		up(&hbus->enum_sem);
+		put_hvpcibus(hbus);
+		return;
+	}
+
+	/* First, mark all existing children as reported missing. */
+	spin_lock_irqsave(&hbus->device_list_lock, flags);
+	list_for_each(iter, &hbus->children) {
+			hpdev = container_of(iter, struct hv_pci_dev,
+					     list_entry);
+			hpdev->reported_missing = true;
+	}
+	spin_unlock_irqrestore(&hbus->device_list_lock, flags);
+
+	/* Next, add back any reported devices. */
+	for (child_no = 0; child_no < dr->device_count; child_no++) {
+		found = false;
+		new_desc = &dr->func[child_no];
+
+		spin_lock_irqsave(&hbus->device_list_lock, flags);
+		list_for_each(iter, &hbus->children) {
+			hpdev = container_of(iter, struct hv_pci_dev,
+					     list_entry);
+			if ((hpdev->desc.win_slot.slot ==
+			     new_desc->win_slot.slot) &&
+			    (hpdev->desc.v_id == new_desc->v_id) &&
+			    (hpdev->desc.d_id == new_desc->d_id) &&
+			    (hpdev->desc.ser == new_desc->ser)) {
+				hpdev->reported_missing = false;
+				found = true;
+			}
+		}
+		spin_unlock_irqrestore(&hbus->device_list_lock, flags);
+
+		if (!found) {
+			hpdev = new_pcichild_device(hbus, new_desc);
+			if (!hpdev)
+				dev_err(&hbus->hdev->device,
+					"couldn't record a child device.\n");
+		}
+	}
+
+	/* Move missing children to a list on the stack. */
+	spin_lock_irqsave(&hbus->device_list_lock, flags);
+	do {
+		found = false;
+		list_for_each(iter, &hbus->children) {
+			hpdev = container_of(iter, struct hv_pci_dev,
+					     list_entry);
+			if (hpdev->reported_missing) {
+				found = true;
+				put_pcichild(hpdev, hv_pcidev_ref_childlist);
+				list_del(&hpdev->list_entry);
+				list_add_tail(&hpdev->list_entry, &removed);
+				break;
+			}
+		}
+	} while (found);
+	spin_unlock_irqrestore(&hbus->device_list_lock, flags);
+
+	/* Delete everything that should no longer exist. */
+	while (!list_empty(&removed)) {
+		hpdev = list_first_entry(&removed, struct hv_pci_dev,
+					 list_entry);
+		list_del(&hpdev->list_entry);
+		put_pcichild(hpdev, hv_pcidev_ref_initial);
+	}
+
+	/* Tell the core to rescan bus because there may have been changes. */
+	if (hbus->state == hv_pcibus_installed) {
+		pci_lock_rescan_remove();
+		pci_scan_child_bus(hbus->pci_bus);
+		pci_unlock_rescan_remove();
+	} else {
+		survey_child_resources(hbus);
+	}
+
+	up(&hbus->enum_sem);
+	put_hvpcibus(hbus);
+	kfree(dr);
+}
+
+/**
+ * hv_pci_devices_present() - Handles list of new children
+ * @hbus:	Root PCI bus, as understood by this driver
+ * @relations:	Packet from host listing children
+ *
+ * This function is invoked whenever a new list of devices for
+ * this bus appears.
+ */
+static void hv_pci_devices_present(struct hv_pcibus_device *hbus,
+				   struct pci_bus_relations *relations)
+{
+	struct hv_dr_state *dr;
+	struct hv_dr_work *dr_wrk;
+	unsigned long flags;
+
+	dr_wrk = kzalloc(sizeof(*dr_wrk), GFP_NOWAIT);
+	if (!dr_wrk)
+		return;
+
+	dr = kzalloc(offsetof(struct hv_dr_state, func) +
+		     (sizeof(struct pci_function_description) *
+		      (relations->device_count)), GFP_NOWAIT);
+	if (!dr)  {
+		kfree(dr_wrk);
+		return;
+	}
+
+	INIT_WORK(&dr_wrk->wrk, pci_devices_present_work);
+	dr_wrk->bus = hbus;
+	dr->device_count = relations->device_count;
+	if (dr->device_count != 0) {
+		memcpy(dr->func, relations->func,
+		       sizeof(struct pci_function_description) *
+		       dr->device_count);
+	}
+
+	spin_lock_irqsave(&hbus->device_list_lock, flags);
+	list_add_tail(&dr->list_entry, &hbus->dr_list);
+	spin_unlock_irqrestore(&hbus->device_list_lock, flags);
+
+	get_hvpcibus(hbus);
+	schedule_work(&dr_wrk->wrk);
+}
+
+/**
+ * hv_eject_device_work() - Asynchronously handles ejection
+ * @work:	Work struct embedded in internal device struct
+ *
+ * This function handles ejecting a device.  Windows will
+ * attempt to gracefully eject a device, waiting 60 seconds to
+ * hear back from the guest OS that this completed successfully.
+ * If this timer expires, the device will be forcibly removed.
+ */
+static void hv_eject_device_work(struct work_struct *work)
+{
+	struct pci_eject_response *ejct_pkt;
+	struct hv_pci_dev *hpdev;
+	struct pci_dev *pdev;
+	unsigned long flags;
+	int wslot;
+	struct {
+		struct pci_packet pkt;
+		u8 buffer[sizeof(struct pci_eject_response) -
+			  sizeof(struct pci_message)];
+	} ctxt;
+
+	hpdev = container_of(work, struct hv_pci_dev, wrk);
+
+	if (hpdev->state != hv_pcichild_ejecting) {
+		put_pcichild(hpdev, hv_pcidev_ref_pnp);
+		return;
+	}
+
+	/*
+	 * Ejection can come before or after the PCI bus has been set up, so
+	 * attempt to find it and tear down the bus state, if it exists.  This
+	 * must be done without constructs like pci_domain_nr(hbus->pci_bus)
+	 * because hbus->pci_bus may not exist yet.
+	 */
+	wslot = wslot_to_devfn(hpdev->desc.win_slot.slot);
+	pdev = pci_get_domain_bus_and_slot(hpdev->hbus->sysdata.domain, 0,
+					   wslot);
+	if (pdev) {
+		pci_stop_and_remove_bus_device(pdev);
+		pci_dev_put(pdev);
+	}
+
+	memset(&ctxt, 0, sizeof(ctxt));
+	ejct_pkt = (struct pci_eject_response *)&ctxt.pkt.message;
+	ejct_pkt->message_type = PCI_EJECTION_COMPLETE;
+	ejct_pkt->wslot.slot = hpdev->desc.win_slot.slot;
+	vmbus_sendpacket(hpdev->hbus->hdev->channel, ejct_pkt,
+			 sizeof(*ejct_pkt), (unsigned long)&ctxt.pkt,
+			 VM_PKT_DATA_INBAND, 0);
+
+	spin_lock_irqsave(&hpdev->hbus->device_list_lock, flags);
+	list_del(&hpdev->list_entry);
+	spin_unlock_irqrestore(&hpdev->hbus->device_list_lock, flags);
+
+	put_pcichild(hpdev, hv_pcidev_ref_childlist);
+	put_pcichild(hpdev, hv_pcidev_ref_pnp);
+	put_hvpcibus(hpdev->hbus);
+}
+
+/**
+ * hv_pci_eject_device() - Handles device ejection
+ * @hpdev:	Internal device tracking struct
+ *
+ * This function is invoked when an ejection packet arrives.  It
+ * just schedules work so that we don't re-enter the packet
+ * delivery code handling the ejection.
+ */
+static void hv_pci_eject_device(struct hv_pci_dev *hpdev)
+{
+	hpdev->state = hv_pcichild_ejecting;
+	get_pcichild(hpdev, hv_pcidev_ref_pnp);
+	INIT_WORK(&hpdev->wrk, hv_eject_device_work);
+	get_hvpcibus(hpdev->hbus);
+	schedule_work(&hpdev->wrk);
+}
+
+/**
+ * hv_pci_onchannelcallback() - Handles incoming packets
+ * @context:	Internal bus tracking struct
+ *
+ * This function is invoked whenever the host sends a packet to
+ * this channel (which is private to this root PCI bus).
+ */
+static void hv_pci_onchannelcallback(void *context)
+{
+	const int packet_size = 0x100;
+	int ret;
+	struct hv_pcibus_device *hbus = context;
+	u32 bytes_recvd;
+	u64 req_id;
+	struct vmpacket_descriptor *desc;
+	unsigned char *buffer;
+	int bufferlen = packet_size;
+	struct pci_packet *comp_packet;
+	struct pci_response *response;
+	struct pci_incoming_message *new_message;
+	struct pci_bus_relations *bus_rel;
+	struct pci_dev_incoming *dev_message;
+	struct hv_pci_dev *hpdev;
+
+	buffer = kmalloc(bufferlen, GFP_ATOMIC);
+	if (!buffer)
+		return;
+
+	while (1) {
+		ret = vmbus_recvpacket_raw(hbus->hdev->channel, buffer,
+					   bufferlen, &bytes_recvd, &req_id);
+
+		if (ret == -ENOBUFS) {
+			kfree(buffer);
+			/* Handle large packet */
+			bufferlen = bytes_recvd;
+			buffer = kmalloc(bytes_recvd, GFP_ATOMIC);
+			if (!buffer)
+				return;
+			continue;
+		}
+
+		/*
+		 * All incoming packets must be at least as large as a
+		 * response.
+		 */
+		if (bytes_recvd <= sizeof(struct pci_response)) {
+			kfree(buffer);
+			return;
+		}
+		desc = (struct vmpacket_descriptor *)buffer;
+
+		switch (desc->type) {
+		case VM_PKT_COMP:
+
+			/*
+			 * The host is trusted, and thus it's safe to interpret
+			 * this transaction ID as a pointer.
+			 */
+			comp_packet = (struct pci_packet *)req_id;
+			response = (struct pci_response *)buffer;
+			comp_packet->completion_func(comp_packet->compl_ctxt,
+						     response,
+						     bytes_recvd);
+			kfree(buffer);
+			return;
+
+		case VM_PKT_DATA_INBAND:
+
+			new_message = (struct pci_incoming_message *)buffer;
+			switch (new_message->message_type.message_type) {
+			case PCI_BUS_RELATIONS:
+
+				bus_rel = (struct pci_bus_relations *)buffer;
+				if (bytes_recvd <
+				    offsetof(struct pci_bus_relations, func) +
+				    (sizeof(struct pci_function_description) *
+				     (bus_rel->device_count))) {
+					dev_err(&hbus->hdev->device,
+						"bus relations too small\n");
+					break;
+				}
+
+				hv_pci_devices_present(hbus, bus_rel);
+				break;
+
+			case PCI_EJECT:
+
+				dev_message = (struct pci_dev_incoming *)buffer;
+				hpdev = get_pcichild_wslot(hbus,
+						      dev_message->wslot.slot);
+				if (hpdev) {
+					hv_pci_eject_device(hpdev);
+					put_pcichild(hpdev,
+							hv_pcidev_ref_by_slot);
+				}
+				break;
+
+			default:
+				dev_warn(&hbus->hdev->device,
+					"Unimplemented protocol message %x\n",
+					new_message->message_type.message_type);
+				break;
+			}
+			break;
+
+		default:
+			dev_err(&hbus->hdev->device,
+				"unhandled packet type %d, tid %llx len %d\n",
+				desc->type, req_id, bytes_recvd);
+			break;
+		}
+		break;
+	}
+}
+
+/**
+ * hv_pci_protocol_negotiation() - Set up protocol
+ * @hdev:	VMBus's tracking struct for this root PCI bus
+ *
+ * This driver is intended to support running on Windows 10
+ * (server) and later versions. It will not run on earlier
+ * versions, as they assume that many of the operations which
+ * Linux needs accomplished with a spinlock held were done via
+ * asynchronous messaging via VMBus.  Windows 10 increases the
+ * surface area of PCI emulation so that these actions can take
+ * place by suspending a virtual processor for their duration.
+ *
+ * This function negotiates the channel protocol version,
+ * failing if the host doesn't support the necessary protocol
+ * level.
+ */
+static int hv_pci_protocol_negotiation(struct hv_device *hdev)
+{
+	struct pci_version_request *version_req;
+	struct hv_pci_compl comp_pkt;
+	struct pci_packet *pkt;
+	int ret;
+
+	/*
+	 * Initiate the handshake with the host and negotiate
+	 * a version that the host can support. We start with the
+	 * highest version number and go down if the host cannot
+	 * support it.
+	 */
+	pkt = kzalloc(sizeof(*pkt) + sizeof(*version_req), GFP_KERNEL);
+	if (!pkt)
+		return -ENOMEM;
+
+	init_completion(&comp_pkt.host_event);
+	pkt->completion_func = hv_pci_generic_compl;
+	pkt->compl_ctxt = &comp_pkt;
+	version_req = (struct pci_version_request *)&pkt->message;
+	version_req->message_type.message_type = PCI_QUERY_PROTOCOL_VERSION;
+	version_req->protocol_version = PCI_PROTOCOL_VERSION_CURRENT;
+
+	ret = vmbus_sendpacket(hdev->channel, version_req,
+			       sizeof(struct pci_version_request),
+			       (unsigned long)pkt, VM_PKT_DATA_INBAND,
+			       VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
+	if (ret)
+		goto exit;
+
+	wait_for_completion(&comp_pkt.host_event);
+
+	if (comp_pkt.completion_status < 0) {
+		dev_err(&hdev->device,
+			"PCI Pass-through VSP failed version request %x\n",
+			comp_pkt.completion_status);
+		ret = -EPROTO;
+		goto exit;
+	}
+
+	ret = 0;
+
+exit:
+	kfree(pkt);
+	return ret;
+}
+
+/**
+ * hv_pci_free_bridge_windows() - Release memory regions for the
+ * bus
+ * @hbus:	Root PCI bus, as understood by this driver
+ */
+static void hv_pci_free_bridge_windows(struct hv_pcibus_device *hbus)
+{
+	/*
+	 * Set the resources back to the way they looked when they
+	 * were allocated by setting IORESOURCE_BUSY again.
+	 */
+
+	if (hbus->low_mmio_space && hbus->low_mmio_res) {
+		hbus->low_mmio_res->flags |= IORESOURCE_BUSY;
+		release_mem_region(hbus->low_mmio_res->start,
+				   resource_size(hbus->low_mmio_res));
+	}
+
+	if (hbus->high_mmio_space && hbus->high_mmio_res) {
+		hbus->high_mmio_res->flags |= IORESOURCE_BUSY;
+		release_mem_region(hbus->high_mmio_res->start,
+				   resource_size(hbus->high_mmio_res));
+	}
+}
+
+/**
+ * hv_pci_allocate_bridge_windows() - Allocate memory regions
+ * for the bus
+ * @hbus:	Root PCI bus, as understood by this driver
+ *
+ * This function calls vmbus_allocate_mmio(), which is itself a
+ * bit of a compromise.  Ideally, we might change the pnp layer
+ * in the kernel such that it comprehends either PCI devices
+ * which are "grandchildren of ACPI," with some intermediate bus
+ * node (in this case, VMBus) or change it such that it
+ * understands VMBus.  The pnp layer, however, has been declared
+ * deprecated, and not subject to change.
+ *
+ * The workaround, implemented here, is to ask VMBus to allocate
+ * MMIO space for this bus.  VMBus itself knows which ranges are
+ * appropriate by looking at its own ACPI objects.  Then, after
+ * these ranges are claimed, they're modified to look like they
+ * would have looked if the ACPI and pnp code had allocated
+ * bridge windows.  These descriptors have to exist in this form
+ * in order to satisfy the code which will get invoked when the
+ * endpoint PCI function driver calls request_mem_region() or
+ * request_mem_region_exclusive().
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int hv_pci_allocate_bridge_windows(struct hv_pcibus_device *hbus)
+{
+	resource_size_t align;
+	int ret;
+
+	if (hbus->low_mmio_space) {
+		align = 1ULL << (63 - __builtin_clzll(hbus->low_mmio_space));
+		ret = vmbus_allocate_mmio(&hbus->low_mmio_res, hbus->hdev, 0,
+					  (u64)(u32)0xffffffff,
+					  hbus->low_mmio_space,
+					  align, false);
+		if (ret) {
+			dev_err(&hbus->hdev->device,
+				"Need %#llx of low MMIO space. Consider reconfiguring the VM.\n",
+				hbus->low_mmio_space);
+			return ret;
+		}
+
+		/* Modify this resource to become a bridge window. */
+		hbus->low_mmio_res->flags |= IORESOURCE_WINDOW;
+		hbus->low_mmio_res->flags &= ~IORESOURCE_BUSY;
+		pci_add_resource(&hbus->resources_for_children,
+				 hbus->low_mmio_res);
+	}
+
+	if (hbus->high_mmio_space) {
+		align = 1ULL << (63 - __builtin_clzll(hbus->high_mmio_space));
+		ret = vmbus_allocate_mmio(&hbus->high_mmio_res, hbus->hdev,
+					  0x100000000, -1,
+					  hbus->high_mmio_space, align,
+					  false);
+		if (ret) {
+			dev_err(&hbus->hdev->device,
+				"Need %#llx of high MMIO space. Consider reconfiguring the VM.\n",
+				hbus->high_mmio_space);
+			goto release_low_mmio;
+		}
+
+		/* Modify this resource to become a bridge window. */
+		hbus->high_mmio_res->flags |= IORESOURCE_WINDOW;
+		hbus->high_mmio_res->flags &= ~IORESOURCE_BUSY;
+		pci_add_resource(&hbus->resources_for_children,
+				 hbus->high_mmio_res);
+	}
+
+	return 0;
+
+release_low_mmio:
+	if (hbus->low_mmio_res) {
+		release_mem_region(hbus->low_mmio_res->start,
+				   resource_size(hbus->low_mmio_res));
+	}
+
+	return ret;
+}
+
+/**
+ * hv_allocate_config_window() - Find MMIO space for PCI Config
+ * @hbus:	Root PCI bus, as understood by this driver
+ *
+ * This function claims memory-mapped I/O space for accessing
+ * configuration space for the functions on this bus.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int hv_allocate_config_window(struct hv_pcibus_device *hbus)
+{
+	int ret;
+
+	/*
+	 * Set up a region of MMIO space to use for accessing configuration
+	 * space.
+	 */
+	ret = vmbus_allocate_mmio(&hbus->mem_config, hbus->hdev, 0, -1,
+				  PCI_CONFIG_MMIO_LENGTH, 0x1000, false);
+	if (ret)
+		return ret;
+
+	/*
+	 * vmbus_allocate_mmio() gets used for allocating both device endpoint
+	 * resource claims (those which cannot be overlapped) and the ranges
+	 * which are valid for the children of this bus, which are intended
+	 * to be overlapped by those children.  Set the flag on this claim
+	 * meaning that this region can't be overlapped.
+	 */
+
+	hbus->mem_config->flags |= IORESOURCE_BUSY;
+
+	return 0;
+}
+
+static void hv_free_config_window(struct hv_pcibus_device *hbus)
+{
+	release_mem_region(hbus->mem_config->start, PCI_CONFIG_MMIO_LENGTH);
+}
+
+/**
+ * hv_pci_enter_d0() - Bring the "bus" into the D0 power state
+ * @hdev:	VMBus's tracking struct for this root PCI bus
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int hv_pci_enter_d0(struct hv_device *hdev)
+{
+	struct hv_pcibus_device *hbus = hv_get_drvdata(hdev);
+	struct pci_bus_d0_entry *d0_entry;
+	struct hv_pci_compl comp_pkt;
+	struct pci_packet *pkt;
+	int ret;
+
+	/*
+	 * Tell the host that the bus is ready to use, and moved into the
+	 * powered-on state.  This includes telling the host which region
+	 * of memory-mapped I/O space has been chosen for configuration space
+	 * access.
+	 */
+	pkt = kzalloc(sizeof(*pkt) + sizeof(*d0_entry), GFP_KERNEL);
+	if (!pkt)
+		return -ENOMEM;
+
+	init_completion(&comp_pkt.host_event);
+	pkt->completion_func = hv_pci_generic_compl;
+	pkt->compl_ctxt = &comp_pkt;
+	d0_entry = (struct pci_bus_d0_entry *)&pkt->message;
+	d0_entry->message_type.message_type = PCI_BUS_D0ENTRY;
+	d0_entry->mmio_base = hbus->mem_config->start;
+
+	ret = vmbus_sendpacket(hdev->channel, d0_entry, sizeof(*d0_entry),
+			       (unsigned long)pkt, VM_PKT_DATA_INBAND,
+			       VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
+	if (ret)
+		goto exit;
+
+	wait_for_completion(&comp_pkt.host_event);
+
+	if (comp_pkt.completion_status < 0) {
+		dev_err(&hdev->device,
+			"PCI Pass-through VSP failed D0 Entry with status %x\n",
+			comp_pkt.completion_status);
+		ret = -EPROTO;
+		goto exit;
+	}
+
+	ret = 0;
+
+exit:
+	kfree(pkt);
+	return ret;
+}
+
+/**
+ * hv_pci_query_relations() - Ask host to send list of child
+ * devices
+ * @hdev:	VMBus's tracking struct for this root PCI bus
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int hv_pci_query_relations(struct hv_device *hdev)
+{
+	struct hv_pcibus_device *hbus = hv_get_drvdata(hdev);
+	struct pci_message message;
+	struct completion comp;
+	int ret;
+
+	/* Ask the host to send along the list of child devices */
+	init_completion(&comp);
+	if (cmpxchg(&hbus->survey_event, NULL, &comp))
+		return -ENOTEMPTY;
+
+	memset(&message, 0, sizeof(message));
+	message.message_type = PCI_QUERY_BUS_RELATIONS;
+
+	ret = vmbus_sendpacket(hdev->channel, &message, sizeof(message),
+			       0, VM_PKT_DATA_INBAND, 0);
+	if (ret)
+		return ret;
+
+	wait_for_completion(&comp);
+	return 0;
+}
+
+/**
+ * hv_send_resources_allocated() - Report local resource choices
+ * @hdev:	VMBus's tracking struct for this root PCI bus
+ *
+ * The host OS is expecting to be sent a request as a message
+ * which contains all the resources that the device will use.
+ * The response contains those same resources, "translated"
+ * which is to say, the values which should be used by the
+ * hardware, when it delivers an interrupt.  (MMIO resources are
+ * used in local terms.)  This is nice for Windows, and lines up
+ * with the FDO/PDO split, which doesn't exist in Linux.  Linux
+ * is deeply expecting to scan an emulated PCI configuration
+ * space.  So this message is sent here only to drive the state
+ * machine on the host forward.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int hv_send_resources_allocated(struct hv_device *hdev)
+{
+	struct hv_pcibus_device *hbus = hv_get_drvdata(hdev);
+	struct pci_resources_assigned *res_assigned;
+	struct hv_pci_compl comp_pkt;
+	struct hv_pci_dev *hpdev;
+	struct pci_packet *pkt;
+	u32 wslot;
+	int ret;
+
+	pkt = kmalloc(sizeof(*pkt) + sizeof(*res_assigned), GFP_KERNEL);
+	if (!pkt)
+		return -ENOMEM;
+
+	ret = 0;
+
+	for (wslot = 0; wslot < 256; wslot++) {
+		hpdev = get_pcichild_wslot(hbus, wslot);
+		if (!hpdev)
+			continue;
+
+		memset(pkt, 0, sizeof(*pkt) + sizeof(*res_assigned));
+		init_completion(&comp_pkt.host_event);
+		pkt->completion_func = hv_pci_generic_compl;
+		pkt->compl_ctxt = &comp_pkt;
+		pkt->message.message_type = PCI_RESOURCES_ASSIGNED;
+		res_assigned = (struct pci_resources_assigned *)&pkt->message;
+		res_assigned->wslot.slot = hpdev->desc.win_slot.slot;
+
+		put_pcichild(hpdev, hv_pcidev_ref_by_slot);
+
+		ret = vmbus_sendpacket(
+			hdev->channel, &pkt->message,
+			sizeof(*res_assigned),
+			(unsigned long)pkt,
+			VM_PKT_DATA_INBAND,
+			VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
+		if (ret)
+			break;
+
+		wait_for_completion(&comp_pkt.host_event);
+
+		if (comp_pkt.completion_status < 0) {
+			ret = -EPROTO;
+			dev_err(&hdev->device,
+				"resource allocated returned 0x%x",
+				comp_pkt.completion_status);
+			break;
+		}
+	}
+
+	kfree(pkt);
+	return ret;
+}
+
+/**
+ * hv_send_resources_released() - Report local resources
+ * released
+ * @hdev:	VMBus's tracking struct for this root PCI bus
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int hv_send_resources_released(struct hv_device *hdev)
+{
+	struct hv_pcibus_device *hbus = hv_get_drvdata(hdev);
+	struct pci_child_message pkt;
+	struct hv_pci_dev *hpdev;
+	u32 wslot;
+	int ret;
+
+	for (wslot = 0; wslot < 256; wslot++) {
+		hpdev = get_pcichild_wslot(hbus, wslot);
+		if (!hpdev)
+			continue;
+
+		memset(&pkt, 0, sizeof(pkt));
+		pkt.message_type = PCI_RESOURCES_RELEASED;
+		pkt.wslot.slot = hpdev->desc.win_slot.slot;
+
+		put_pcichild(hpdev, hv_pcidev_ref_by_slot);
+
+		ret = vmbus_sendpacket(hdev->channel, &pkt, sizeof(pkt), 0,
+				       VM_PKT_DATA_INBAND, 0);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static void get_hvpcibus(struct hv_pcibus_device *hbus)
+{
+	atomic_inc(&hbus->remove_lock);
+}
+
+static void put_hvpcibus(struct hv_pcibus_device *hbus)
+{
+	if (atomic_dec_and_test(&hbus->remove_lock))
+		complete(&hbus->remove_event);
+}
+
+/**
+ * hv_pci_probe() - New VMBus channel probe, for a root PCI bus
+ * @hdev:	VMBus's tracking struct for this root PCI bus
+ * @dev_id:	Identifies the device itself
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int hv_pci_probe(struct hv_device *hdev,
+			const struct hv_vmbus_device_id *dev_id)
+{
+	struct hv_pcibus_device *hbus;
+	int ret;
+
+	hbus = kzalloc(sizeof(*hbus), GFP_KERNEL);
+	if (!hbus)
+		return -ENOMEM;
+
+	/*
+	 * The PCI bus "domain" is what is called "segment" in ACPI and
+	 * other specs.  Pull it from the instance ID, to get something
+	 * unique.  Bytes 8 and 9 are what is used in Windows guests, so
+	 * do the same thing for consistency.  Note that, since this code
+	 * only runs in a Hyper-V VM, Hyper-V can (and does) guarantee
+	 * that (1) the only domain in use for something that looks like
+	 * a physical PCI bus (which is actually emulated by the
+	 * hypervisor) is domain 0 and (2) there will be no overlap
+	 * between domains derived from these instance IDs in the same
+	 * VM.
+	 */
+	hbus->sysdata.domain = hdev->dev_instance.b[9] |
+			       hdev->dev_instance.b[8] << 8;
+
+	hbus->hdev = hdev;
+	atomic_inc(&hbus->remove_lock);
+	INIT_LIST_HEAD(&hbus->children);
+	INIT_LIST_HEAD(&hbus->dr_list);
+	INIT_LIST_HEAD(&hbus->resources_for_children);
+	spin_lock_init(&hbus->config_lock);
+	spin_lock_init(&hbus->device_list_lock);
+	sema_init(&hbus->enum_sem, 1);
+	init_completion(&hbus->remove_event);
+
+	ret = vmbus_open(hdev->channel, pci_ring_size, pci_ring_size, NULL, 0,
+			 hv_pci_onchannelcallback, hbus);
+	if (ret)
+		goto free_bus;
+
+	hv_set_drvdata(hdev, hbus);
+
+	ret = hv_pci_protocol_negotiation(hdev);
+	if (ret)
+		goto close;
+
+	ret = hv_allocate_config_window(hbus);
+	if (ret)
+		goto close;
+
+	hbus->cfg_addr = ioremap(hbus->mem_config->start,
+				 PCI_CONFIG_MMIO_LENGTH);
+	if (!hbus->cfg_addr) {
+		dev_err(&hdev->device,
+			"Unable to map a virtual address for config space\n");
+		ret = -ENOMEM;
+		goto free_config;
+	}
+
+	hbus->sysdata.fwnode = irq_domain_alloc_fwnode(hbus);
+	if (!hbus->sysdata.fwnode) {
+		ret = -ENOMEM;
+		goto unmap;
+	}
+
+	ret = hv_pcie_init_irq_domain(hbus);
+	if (ret)
+		goto free_fwnode;
+
+	ret = hv_pci_query_relations(hdev);
+	if (ret)
+		goto free_irq_domain;
+
+	ret = hv_pci_enter_d0(hdev);
+	if (ret)
+		goto free_irq_domain;
+
+	ret = hv_pci_allocate_bridge_windows(hbus);
+	if (ret)
+		goto free_irq_domain;
+
+	ret = hv_send_resources_allocated(hdev);
+	if (ret)
+		goto free_windows;
+
+	prepopulate_bars(hbus);
+
+	hbus->state = hv_pcibus_probed;
+
+	ret = create_root_hv_pci_bus(hbus);
+	if (ret)
+		goto free_windows;
+
+	return 0;
+
+free_windows:
+	hv_pci_free_bridge_windows(hbus);
+free_irq_domain:
+	irq_domain_remove(hbus->irq_domain);
+free_fwnode:
+	irq_domain_free_fwnode(hbus->sysdata.fwnode);
+unmap:
+	iounmap(hbus->cfg_addr);
+free_config:
+	hv_free_config_window(hbus);
+close:
+	vmbus_close(hdev->channel);
+free_bus:
+	kfree(hbus);
+	return ret;
+}
+
+/**
+ * hv_pci_remove() - Remove routine for this VMBus channel
+ * @hdev:	VMBus's tracking struct for this root PCI bus
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int hv_pci_remove(struct hv_device *hdev)
+{
+	int ret;
+	struct hv_pcibus_device *hbus;
+	union {
+		struct pci_packet teardown_packet;
+		u8 buffer[0x100];
+	} pkt;
+	struct pci_bus_relations relations;
+	struct hv_pci_compl comp_pkt;
+
+	hbus = hv_get_drvdata(hdev);
+
+	ret = hv_send_resources_released(hdev);
+	if (ret)
+		dev_err(&hdev->device,
+			"Couldn't send resources released packet(s)\n");
+
+	memset(&pkt.teardown_packet, 0, sizeof(pkt.teardown_packet));
+	init_completion(&comp_pkt.host_event);
+	pkt.teardown_packet.completion_func = hv_pci_generic_compl;
+	pkt.teardown_packet.compl_ctxt = &comp_pkt;
+	pkt.teardown_packet.message.message_type = PCI_BUS_D0EXIT;
+
+	ret = vmbus_sendpacket(hdev->channel, &pkt.teardown_packet.message,
+			       sizeof(struct pci_message),
+			       (unsigned long)&pkt.teardown_packet,
+			       VM_PKT_DATA_INBAND,
+			       VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
+	if (!ret)
+		wait_for_completion_timeout(&comp_pkt.host_event, 10 * HZ);
+
+	if (hbus->state == hv_pcibus_installed) {
+		/* Remove the bus from PCI's point of view. */
+		pci_lock_rescan_remove();
+		pci_stop_root_bus(hbus->pci_bus);
+		pci_remove_root_bus(hbus->pci_bus);
+		pci_unlock_rescan_remove();
+	}
+
+	vmbus_close(hdev->channel);
+
+	/* Delete any children which might still exist. */
+	memset(&relations, 0, sizeof(relations));
+	hv_pci_devices_present(hbus, &relations);
+
+	iounmap(hbus->cfg_addr);
+	hv_free_config_window(hbus);
+	pci_free_resource_list(&hbus->resources_for_children);
+	hv_pci_free_bridge_windows(hbus);
+	irq_domain_remove(hbus->irq_domain);
+	irq_domain_free_fwnode(hbus->sysdata.fwnode);
+	put_hvpcibus(hbus);
+	wait_for_completion(&hbus->remove_event);
+	kfree(hbus);
+	return 0;
+}
+
+static const struct hv_vmbus_device_id hv_pci_id_table[] = {
+	/* PCI Pass-through Class ID */
+	/* 44C4F61D-4444-4400-9D52-802E27EDE19F */
+	{ HV_PCIE_GUID, },
+	{ },
+};
+
+MODULE_DEVICE_TABLE(vmbus, hv_pci_id_table);
+
+static struct hv_driver hv_pci_drv = {
+	.name		= "hv_pci",
+	.id_table	= hv_pci_id_table,
+	.probe		= hv_pci_probe,
+	.remove		= hv_pci_remove,
+};
+
+static void __exit exit_hv_pci_drv(void)
+{
+	vmbus_driver_unregister(&hv_pci_drv);
+}
+
+static int __init init_hv_pci_drv(void)
+{
+	return vmbus_driver_register(&hv_pci_drv);
+}
+
+module_init(init_hv_pci_drv);
+module_exit(exit_hv_pci_drv);
+
+MODULE_DESCRIPTION("Hyper-V PCI");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/pci/host/pci-imx6.c b/drivers/pci/host/pci-imx6.c
index fe600964fa50..eb5a2755a164 100644
--- a/drivers/pci/host/pci-imx6.c
+++ b/drivers/pci/host/pci-imx6.c
@@ -39,6 +39,11 @@ struct imx6_pcie {
 	struct pcie_port	pp;
 	struct regmap		*iomuxc_gpr;
 	void __iomem		*mem_base;
+	u32			tx_deemph_gen1;
+	u32			tx_deemph_gen2_3p5db;
+	u32			tx_deemph_gen2_6db;
+	u32			tx_swing_full;
+	u32			tx_swing_low;
 };
 
 /* PCIe Root Complex registers (memory-mapped) */
@@ -202,6 +207,23 @@ static int pcie_phy_write(void __iomem *dbi_base, int addr, int data)
 	return 0;
 }
 
+static void imx6_pcie_reset_phy(struct pcie_port *pp)
+{
+	u32 tmp;
+
+	pcie_phy_read(pp->dbi_base, PHY_RX_OVRD_IN_LO, &tmp);
+	tmp |= (PHY_RX_OVRD_IN_LO_RX_DATA_EN |
+		PHY_RX_OVRD_IN_LO_RX_PLL_EN);
+	pcie_phy_write(pp->dbi_base, PHY_RX_OVRD_IN_LO, tmp);
+
+	usleep_range(2000, 3000);
+
+	pcie_phy_read(pp->dbi_base, PHY_RX_OVRD_IN_LO, &tmp);
+	tmp &= ~(PHY_RX_OVRD_IN_LO_RX_DATA_EN |
+		  PHY_RX_OVRD_IN_LO_RX_PLL_EN);
+	pcie_phy_write(pp->dbi_base, PHY_RX_OVRD_IN_LO, tmp);
+}
+
 /*  Added for PCI abort handling */
 static int imx6q_pcie_abort_handler(unsigned long addr,
 		unsigned int fsr, struct pt_regs *regs)
@@ -317,32 +339,32 @@ static void imx6_pcie_init_phy(struct pcie_port *pp)
 			IMX6Q_GPR12_LOS_LEVEL, 9 << 4);
 
 	regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR8,
-			IMX6Q_GPR8_TX_DEEMPH_GEN1, 0 << 0);
+			   IMX6Q_GPR8_TX_DEEMPH_GEN1,
+			   imx6_pcie->tx_deemph_gen1 << 0);
 	regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR8,
-			IMX6Q_GPR8_TX_DEEMPH_GEN2_3P5DB, 0 << 6);
+			   IMX6Q_GPR8_TX_DEEMPH_GEN2_3P5DB,
+			   imx6_pcie->tx_deemph_gen2_3p5db << 6);
 	regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR8,
-			IMX6Q_GPR8_TX_DEEMPH_GEN2_6DB, 20 << 12);
+			   IMX6Q_GPR8_TX_DEEMPH_GEN2_6DB,
+			   imx6_pcie->tx_deemph_gen2_6db << 12);
 	regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR8,
-			IMX6Q_GPR8_TX_SWING_FULL, 127 << 18);
+			   IMX6Q_GPR8_TX_SWING_FULL,
+			   imx6_pcie->tx_swing_full << 18);
 	regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR8,
-			IMX6Q_GPR8_TX_SWING_LOW, 127 << 25);
+			   IMX6Q_GPR8_TX_SWING_LOW,
+			   imx6_pcie->tx_swing_low << 25);
 }
 
 static int imx6_pcie_wait_for_link(struct pcie_port *pp)
 {
-	unsigned int retries;
-
-	for (retries = 0; retries < 200; retries++) {
-		if (dw_pcie_link_up(pp))
-			return 0;
-		usleep_range(100, 1000);
-	}
+	/* check if the link is up or not */
+	if (!dw_pcie_wait_for_link(pp))
+		return 0;
 
-	dev_err(pp->dev, "phy link never came up\n");
 	dev_dbg(pp->dev, "DEBUG_R0: 0x%08x, DEBUG_R1: 0x%08x\n",
 		readl(pp->dbi_base + PCIE_PHY_DEBUG_R0),
 		readl(pp->dbi_base + PCIE_PHY_DEBUG_R1));
-	return -EINVAL;
+	return -ETIMEDOUT;
 }
 
 static int imx6_pcie_wait_for_speed_change(struct pcie_port *pp)
@@ -390,8 +412,10 @@ static int imx6_pcie_establish_link(struct pcie_port *pp)
 			IMX6Q_GPR12_PCIE_CTL_2, 1 << 10);
 
 	ret = imx6_pcie_wait_for_link(pp);
-	if (ret)
-		return ret;
+	if (ret) {
+		dev_info(pp->dev, "Link never came up\n");
+		goto err_reset_phy;
+	}
 
 	/* Allow Gen2 mode after the link is up. */
 	tmp = readl(pp->dbi_base + PCIE_RC_LCR);
@@ -410,19 +434,28 @@ static int imx6_pcie_establish_link(struct pcie_port *pp)
 	ret = imx6_pcie_wait_for_speed_change(pp);
 	if (ret) {
 		dev_err(pp->dev, "Failed to bring link up!\n");
-		return ret;
+		goto err_reset_phy;
 	}
 
 	/* Make sure link training is finished as well! */
 	ret = imx6_pcie_wait_for_link(pp);
 	if (ret) {
 		dev_err(pp->dev, "Failed to bring link up!\n");
-		return ret;
+		goto err_reset_phy;
 	}
 
 	tmp = readl(pp->dbi_base + PCIE_RC_LCSR);
 	dev_dbg(pp->dev, "Link up, Gen=%i\n", (tmp >> 16) & 0xf);
+
 	return 0;
+
+err_reset_phy:
+	dev_dbg(pp->dev, "PHY DEBUG_R0=0x%08x DEBUG_R1=0x%08x\n",
+		readl(pp->dbi_base + PCIE_PHY_DEBUG_R0),
+		readl(pp->dbi_base + PCIE_PHY_DEBUG_R1));
+	imx6_pcie_reset_phy(pp);
+
+	return ret;
 }
 
 static void imx6_pcie_host_init(struct pcie_port *pp)
@@ -441,81 +474,10 @@ static void imx6_pcie_host_init(struct pcie_port *pp)
 		dw_pcie_msi_init(pp);
 }
 
-static void imx6_pcie_reset_phy(struct pcie_port *pp)
-{
-	u32 tmp;
-
-	pcie_phy_read(pp->dbi_base, PHY_RX_OVRD_IN_LO, &tmp);
-	tmp |= (PHY_RX_OVRD_IN_LO_RX_DATA_EN |
-		PHY_RX_OVRD_IN_LO_RX_PLL_EN);
-	pcie_phy_write(pp->dbi_base, PHY_RX_OVRD_IN_LO, tmp);
-
-	usleep_range(2000, 3000);
-
-	pcie_phy_read(pp->dbi_base, PHY_RX_OVRD_IN_LO, &tmp);
-	tmp &= ~(PHY_RX_OVRD_IN_LO_RX_DATA_EN |
-		  PHY_RX_OVRD_IN_LO_RX_PLL_EN);
-	pcie_phy_write(pp->dbi_base, PHY_RX_OVRD_IN_LO, tmp);
-}
-
 static int imx6_pcie_link_up(struct pcie_port *pp)
 {
-	u32 rc, debug_r0, rx_valid;
-	int count = 5;
-
-	/*
-	 * Test if the PHY reports that the link is up and also that the LTSSM
-	 * training finished. There are three possible states of the link when
-	 * this code is called:
-	 * 1) The link is DOWN (unlikely)
-	 *     The link didn't come up yet for some reason. This usually means
-	 *     we have a real problem somewhere. Reset the PHY and exit. This
-	 *     state calls for inspection of the DEBUG registers.
-	 * 2) The link is UP, but still in LTSSM training
-	 *     Wait for the training to finish, which should take a very short
-	 *     time. If the training does not finish, we have a problem and we
-	 *     need to inspect the DEBUG registers. If the training does finish,
-	 *     the link is up and operating correctly.
-	 * 3) The link is UP and no longer in LTSSM training
-	 *     The link is up and operating correctly.
-	 */
-	while (1) {
-		rc = readl(pp->dbi_base + PCIE_PHY_DEBUG_R1);
-		if (!(rc & PCIE_PHY_DEBUG_R1_XMLH_LINK_UP))
-			break;
-		if (!(rc & PCIE_PHY_DEBUG_R1_XMLH_LINK_IN_TRAINING))
-			return 1;
-		if (!count--)
-			break;
-		dev_dbg(pp->dev, "Link is up, but still in training\n");
-		/*
-		 * Wait a little bit, then re-check if the link finished
-		 * the training.
-		 */
-		usleep_range(1000, 2000);
-	}
-	/*
-	 * From L0, initiate MAC entry to gen2 if EP/RC supports gen2.
-	 * Wait 2ms (LTSSM timeout is 24ms, PHY lock is ~5us in gen2).
-	 * If (MAC/LTSSM.state == Recovery.RcvrLock)
-	 * && (PHY/rx_valid==0) then pulse PHY/rx_reset. Transition
-	 * to gen2 is stuck
-	 */
-	pcie_phy_read(pp->dbi_base, PCIE_PHY_RX_ASIC_OUT, &rx_valid);
-	debug_r0 = readl(pp->dbi_base + PCIE_PHY_DEBUG_R0);
-
-	if (rx_valid & PCIE_PHY_RX_ASIC_OUT_VALID)
-		return 0;
-
-	if ((debug_r0 & 0x3f) != 0x0d)
-		return 0;
-
-	dev_err(pp->dev, "transition to gen2 is stuck, reset PHY!\n");
-	dev_dbg(pp->dev, "debug_r0=%08x debug_r1=%08x\n", debug_r0, rc);
-
-	imx6_pcie_reset_phy(pp);
-
-	return 0;
+	return readl(pp->dbi_base + PCIE_PHY_DEBUG_R1) &
+			PCIE_PHY_DEBUG_R1_XMLH_LINK_UP;
 }
 
 static struct pcie_host_ops imx6_pcie_host_ops = {
@@ -562,6 +524,7 @@ static int __init imx6_pcie_probe(struct platform_device *pdev)
 	struct imx6_pcie *imx6_pcie;
 	struct pcie_port *pp;
 	struct resource *dbi_base;
+	struct device_node *node = pdev->dev.of_node;
 	int ret;
 
 	imx6_pcie = devm_kzalloc(&pdev->dev, sizeof(*imx6_pcie), GFP_KERNEL);
@@ -614,6 +577,27 @@ static int __init imx6_pcie_probe(struct platform_device *pdev)
 		return PTR_ERR(imx6_pcie->iomuxc_gpr);
 	}
 
+	/* Grab PCIe PHY Tx Settings */
+	if (of_property_read_u32(node, "fsl,tx-deemph-gen1",
+				 &imx6_pcie->tx_deemph_gen1))
+		imx6_pcie->tx_deemph_gen1 = 0;
+
+	if (of_property_read_u32(node, "fsl,tx-deemph-gen2-3p5db",
+				 &imx6_pcie->tx_deemph_gen2_3p5db))
+		imx6_pcie->tx_deemph_gen2_3p5db = 0;
+
+	if (of_property_read_u32(node, "fsl,tx-deemph-gen2-6db",
+				 &imx6_pcie->tx_deemph_gen2_6db))
+		imx6_pcie->tx_deemph_gen2_6db = 20;
+
+	if (of_property_read_u32(node, "fsl,tx-swing-full",
+				 &imx6_pcie->tx_swing_full))
+		imx6_pcie->tx_swing_full = 127;
+
+	if (of_property_read_u32(node, "fsl,tx-swing-low",
+				 &imx6_pcie->tx_swing_low))
+		imx6_pcie->tx_swing_low = 127;
+
 	ret = imx6_add_pcie_port(pp, pdev);
 	if (ret < 0)
 		return ret;
diff --git a/drivers/pci/host/pci-keystone.c b/drivers/pci/host/pci-keystone.c
index 0aa81bd3de12..b71f55bb0315 100644
--- a/drivers/pci/host/pci-keystone.c
+++ b/drivers/pci/host/pci-keystone.c
@@ -97,17 +97,15 @@ static int ks_pcie_establish_link(struct keystone_pcie *ks_pcie)
 		return 0;
 	}
 
-	ks_dw_pcie_initiate_link_train(ks_pcie);
 	/* check if the link is up or not */
-	for (retries = 0; retries < 200; retries++) {
-		if (dw_pcie_link_up(pp))
-			return 0;
-		usleep_range(100, 1000);
+	for (retries = 0; retries < 5; retries++) {
 		ks_dw_pcie_initiate_link_train(ks_pcie);
+		if (!dw_pcie_wait_for_link(pp))
+			return 0;
 	}
 
 	dev_err(pp->dev, "phy link never came up\n");
-	return -EINVAL;
+	return -ETIMEDOUT;
 }
 
 static void ks_pcie_msi_irq_handler(struct irq_desc *desc)
@@ -359,6 +357,9 @@ static int __init ks_pcie_probe(struct platform_device *pdev)
 
 	/* initialize SerDes Phy if present */
 	phy = devm_phy_get(dev, "pcie-phy");
+	if (PTR_ERR_OR_ZERO(phy) == -EPROBE_DEFER)
+		return PTR_ERR(phy);
+
 	if (!IS_ERR_OR_NULL(phy)) {
 		ret = phy_init(phy);
 		if (ret < 0)
diff --git a/drivers/pci/host/pci-layerscape.c b/drivers/pci/host/pci-layerscape.c
index f39961bcf7aa..a21e229d95e0 100644
--- a/drivers/pci/host/pci-layerscape.c
+++ b/drivers/pci/host/pci-layerscape.c
@@ -208,6 +208,7 @@ static const struct of_device_id ls_pcie_of_match[] = {
 	{ .compatible = "fsl,ls1021a-pcie", .data = &ls1021_drvdata },
 	{ .compatible = "fsl,ls1043a-pcie", .data = &ls1043_drvdata },
 	{ .compatible = "fsl,ls2080a-pcie", .data = &ls2080_drvdata },
+	{ .compatible = "fsl,ls2085a-pcie", .data = &ls2080_drvdata },
 	{ },
 };
 MODULE_DEVICE_TABLE(of, ls_pcie_of_match);
diff --git a/drivers/pci/host/pci-tegra.c b/drivers/pci/host/pci-tegra.c
index 30323114c53c..68d1f41b3cbf 100644
--- a/drivers/pci/host/pci-tegra.c
+++ b/drivers/pci/host/pci-tegra.c
@@ -281,6 +281,11 @@ struct tegra_pcie {
 	struct resource prefetch;
 	struct resource busn;
 
+	struct {
+		resource_size_t mem;
+		resource_size_t io;
+	} offset;
+
 	struct clk *pex_clk;
 	struct clk *afi_clk;
 	struct clk *pll_e;
@@ -295,7 +300,6 @@ struct tegra_pcie {
 	struct tegra_msi msi;
 
 	struct list_head ports;
-	unsigned int num_ports;
 	u32 xbar_config;
 
 	struct regulator_bulk_data *supplies;
@@ -426,31 +430,38 @@ free:
 	return ERR_PTR(err);
 }
 
-/*
- * Look up a virtual address mapping for the specified bus number. If no such
- * mapping exists, try to create one.
- */
-static void __iomem *tegra_pcie_bus_map(struct tegra_pcie *pcie,
-					unsigned int busnr)
+static int tegra_pcie_add_bus(struct pci_bus *bus)
 {
-	struct tegra_pcie_bus *bus;
+	struct tegra_pcie *pcie = sys_to_pcie(bus->sysdata);
+	struct tegra_pcie_bus *b;
 
-	list_for_each_entry(bus, &pcie->buses, list)
-		if (bus->nr == busnr)
-			return (void __iomem *)bus->area->addr;
+	b = tegra_pcie_bus_alloc(pcie, bus->number);
+	if (IS_ERR(b))
+		return PTR_ERR(b);
 
-	bus = tegra_pcie_bus_alloc(pcie, busnr);
-	if (IS_ERR(bus))
-		return NULL;
+	list_add_tail(&b->list, &pcie->buses);
 
-	list_add_tail(&bus->list, &pcie->buses);
+	return 0;
+}
 
-	return (void __iomem *)bus->area->addr;
+static void tegra_pcie_remove_bus(struct pci_bus *child)
+{
+	struct tegra_pcie *pcie = sys_to_pcie(child->sysdata);
+	struct tegra_pcie_bus *bus, *tmp;
+
+	list_for_each_entry_safe(bus, tmp, &pcie->buses, list) {
+		if (bus->nr == child->number) {
+			vunmap(bus->area->addr);
+			list_del(&bus->list);
+			kfree(bus);
+			break;
+		}
+	}
 }
 
-static void __iomem *tegra_pcie_conf_address(struct pci_bus *bus,
-					     unsigned int devfn,
-					     int where)
+static void __iomem *tegra_pcie_map_bus(struct pci_bus *bus,
+					unsigned int devfn,
+					int where)
 {
 	struct tegra_pcie *pcie = sys_to_pcie(bus->sysdata);
 	void __iomem *addr = NULL;
@@ -466,7 +477,12 @@ static void __iomem *tegra_pcie_conf_address(struct pci_bus *bus,
 			}
 		}
 	} else {
-		addr = tegra_pcie_bus_map(pcie, bus->number);
+		struct tegra_pcie_bus *b;
+
+		list_for_each_entry(b, &pcie->buses, list)
+			if (b->nr == bus->number)
+				addr = (void __iomem *)b->area->addr;
+
 		if (!addr) {
 			dev_err(pcie->dev,
 				"failed to map cfg. space for bus %u\n",
@@ -481,7 +497,9 @@ static void __iomem *tegra_pcie_conf_address(struct pci_bus *bus,
 }
 
 static struct pci_ops tegra_pcie_ops = {
-	.map_bus = tegra_pcie_conf_address,
+	.add_bus = tegra_pcie_add_bus,
+	.remove_bus = tegra_pcie_remove_bus,
+	.map_bus = tegra_pcie_map_bus,
 	.read = pci_generic_config_read32,
 	.write = pci_generic_config_write32,
 };
@@ -598,6 +616,17 @@ static int tegra_pcie_setup(int nr, struct pci_sys_data *sys)
 	struct tegra_pcie *pcie = sys_to_pcie(sys);
 	int err;
 
+	sys->mem_offset = pcie->offset.mem;
+	sys->io_offset = pcie->offset.io;
+
+	err = devm_request_resource(pcie->dev, &pcie->all, &pcie->io);
+	if (err < 0)
+		return err;
+
+	err = devm_request_resource(pcie->dev, &ioport_resource, &pcie->pio);
+	if (err < 0)
+		return err;
+
 	err = devm_request_resource(pcie->dev, &pcie->all, &pcie->mem);
 	if (err < 0)
 		return err;
@@ -606,6 +635,7 @@ static int tegra_pcie_setup(int nr, struct pci_sys_data *sys)
 	if (err)
 		return err;
 
+	pci_add_resource_offset(&sys->resources, &pcie->pio, sys->io_offset);
 	pci_add_resource_offset(&sys->resources, &pcie->mem, sys->mem_offset);
 	pci_add_resource_offset(&sys->resources, &pcie->prefetch,
 				sys->mem_offset);
@@ -741,7 +771,7 @@ static void tegra_pcie_setup_translations(struct tegra_pcie *pcie)
 	afi_writel(pcie, 0, AFI_FPCI_BAR5);
 
 	/* map all upstream transactions as uncached */
-	afi_writel(pcie, PHYS_OFFSET, AFI_CACHE_BAR0_ST);
+	afi_writel(pcie, 0, AFI_CACHE_BAR0_ST);
 	afi_writel(pcie, 0, AFI_CACHE_BAR0_SZ);
 	afi_writel(pcie, 0, AFI_CACHE_BAR1_ST);
 	afi_writel(pcie, 0, AFI_CACHE_BAR1_SZ);
@@ -1601,6 +1631,9 @@ static int tegra_pcie_parse_dt(struct tegra_pcie *pcie)
 
 		switch (res.flags & IORESOURCE_TYPE_BITS) {
 		case IORESOURCE_IO:
+			/* Track the bus -> CPU I/O mapping offset. */
+			pcie->offset.io = res.start - range.pci_addr;
+
 			memcpy(&pcie->pio, &res, sizeof(res));
 			pcie->pio.name = np->full_name;
 
@@ -1621,6 +1654,14 @@ static int tegra_pcie_parse_dt(struct tegra_pcie *pcie)
 			break;
 
 		case IORESOURCE_MEM:
+			/*
+			 * Track the bus -> CPU memory mapping offset. This
+			 * assumes that the prefetchable and non-prefetchable
+			 * regions will be the last of type IORESOURCE_MEM in
+			 * the ranges property.
+			 * */
+			pcie->offset.mem = res.start - range.pci_addr;
+
 			if (res.flags & IORESOURCE_PREFETCH) {
 				memcpy(&pcie->prefetch, &res, sizeof(res));
 				pcie->prefetch.name = "prefetchable";
diff --git a/drivers/pci/host/pci-thunder-ecam.c b/drivers/pci/host/pci-thunder-ecam.c
new file mode 100644
index 000000000000..d71935cb2678
--- /dev/null
+++ b/drivers/pci/host/pci-thunder-ecam.c
@@ -0,0 +1,403 @@
+/*
+ * 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.
+ *
+ * Copyright (C) 2015, 2016 Cavium, Inc.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/ioport.h>
+#include <linux/of_pci.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+
+#include "pci-host-common.h"
+
+/* Mapping is standard ECAM */
+static void __iomem *thunder_ecam_map_bus(struct pci_bus *bus,
+					  unsigned int devfn,
+					  int where)
+{
+	struct gen_pci *pci = bus->sysdata;
+	resource_size_t idx = bus->number - pci->cfg.bus_range->start;
+
+	return pci->cfg.win[idx] + ((devfn << 12) | where);
+}
+
+static void set_val(u32 v, int where, int size, u32 *val)
+{
+	int shift = (where & 3) * 8;
+
+	pr_debug("set_val %04x: %08x\n", (unsigned)(where & ~3), v);
+	v >>= shift;
+	if (size == 1)
+		v &= 0xff;
+	else if (size == 2)
+		v &= 0xffff;
+	*val = v;
+}
+
+static int handle_ea_bar(u32 e0, int bar, struct pci_bus *bus,
+			 unsigned int devfn, int where, int size, u32 *val)
+{
+	void __iomem *addr;
+	u32 v;
+
+	/* Entries are 16-byte aligned; bits[2,3] select word in entry */
+	int where_a = where & 0xc;
+
+	if (where_a == 0) {
+		set_val(e0, where, size, val);
+		return PCIBIOS_SUCCESSFUL;
+	}
+	if (where_a == 0x4) {
+		addr = bus->ops->map_bus(bus, devfn, bar); /* BAR 0 */
+		if (!addr) {
+			*val = ~0;
+			return PCIBIOS_DEVICE_NOT_FOUND;
+		}
+		v = readl(addr);
+		v &= ~0xf;
+		v |= 2; /* EA entry-1. Base-L */
+		set_val(v, where, size, val);
+		return PCIBIOS_SUCCESSFUL;
+	}
+	if (where_a == 0x8) {
+		u32 barl_orig;
+		u32 barl_rb;
+
+		addr = bus->ops->map_bus(bus, devfn, bar); /* BAR 0 */
+		if (!addr) {
+			*val = ~0;
+			return PCIBIOS_DEVICE_NOT_FOUND;
+		}
+		barl_orig = readl(addr + 0);
+		writel(0xffffffff, addr + 0);
+		barl_rb = readl(addr + 0);
+		writel(barl_orig, addr + 0);
+		/* zeros in unsettable bits */
+		v = ~barl_rb & ~3;
+		v |= 0xc; /* EA entry-2. Offset-L */
+		set_val(v, where, size, val);
+		return PCIBIOS_SUCCESSFUL;
+	}
+	if (where_a == 0xc) {
+		addr = bus->ops->map_bus(bus, devfn, bar + 4); /* BAR 1 */
+		if (!addr) {
+			*val = ~0;
+			return PCIBIOS_DEVICE_NOT_FOUND;
+		}
+		v = readl(addr); /* EA entry-3. Base-H */
+		set_val(v, where, size, val);
+		return PCIBIOS_SUCCESSFUL;
+	}
+	return PCIBIOS_DEVICE_NOT_FOUND;
+}
+
+static int thunder_ecam_p2_config_read(struct pci_bus *bus, unsigned int devfn,
+				       int where, int size, u32 *val)
+{
+	struct gen_pci *pci = bus->sysdata;
+	int where_a = where & ~3;
+	void __iomem *addr;
+	u32 node_bits;
+	u32 v;
+
+	/* EA Base[63:32] may be missing some bits ... */
+	switch (where_a) {
+	case 0xa8:
+	case 0xbc:
+	case 0xd0:
+	case 0xe4:
+		break;
+	default:
+		return pci_generic_config_read(bus, devfn, where, size, val);
+	}
+
+	addr = bus->ops->map_bus(bus, devfn, where_a);
+	if (!addr) {
+		*val = ~0;
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	}
+
+	v = readl(addr);
+
+	/*
+	 * Bit 44 of the 64-bit Base must match the same bit in
+	 * the config space access window.  Since we are working with
+	 * the high-order 32 bits, shift everything down by 32 bits.
+	 */
+	node_bits = (pci->cfg.res.start >> 32) & (1 << 12);
+
+	v |= node_bits;
+	set_val(v, where, size, val);
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int thunder_ecam_config_read(struct pci_bus *bus, unsigned int devfn,
+				    int where, int size, u32 *val)
+{
+	u32 v;
+	u32 vendor_device;
+	u32 class_rev;
+	void __iomem *addr;
+	int cfg_type;
+	int where_a = where & ~3;
+
+	addr = bus->ops->map_bus(bus, devfn, 0xc);
+	if (!addr) {
+		*val = ~0;
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	}
+
+	v = readl(addr);
+
+	/* Check for non type-00 header */
+	cfg_type = (v >> 16) & 0x7f;
+
+	addr = bus->ops->map_bus(bus, devfn, 8);
+	if (!addr) {
+		*val = ~0;
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	}
+
+	class_rev = readl(addr);
+	if (class_rev == 0xffffffff)
+		goto no_emulation;
+
+	if ((class_rev & 0xff) >= 8) {
+		/* Pass-2 handling */
+		if (cfg_type)
+			goto no_emulation;
+		return thunder_ecam_p2_config_read(bus, devfn, where,
+						   size, val);
+	}
+
+	/*
+	 * All BARs have fixed addresses specified by the EA
+	 * capability; they must return zero on read.
+	 */
+	if (cfg_type == 0 &&
+	    ((where >= 0x10 && where < 0x2c) ||
+	     (where >= 0x1a4 && where < 0x1bc))) {
+		/* BAR or SR-IOV BAR */
+		*val = 0;
+		return PCIBIOS_SUCCESSFUL;
+	}
+
+	addr = bus->ops->map_bus(bus, devfn, 0);
+	if (!addr) {
+		*val = ~0;
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	}
+
+	vendor_device = readl(addr);
+	if (vendor_device == 0xffffffff)
+		goto no_emulation;
+
+	pr_debug("%04x:%04x - Fix pass#: %08x, where: %03x, devfn: %03x\n",
+		 vendor_device & 0xffff, vendor_device >> 16, class_rev,
+		 (unsigned) where, devfn);
+
+	/* Check for non type-00 header */
+	if (cfg_type == 0) {
+		bool has_msix;
+		bool is_nic = (vendor_device == 0xa01e177d);
+		bool is_tns = (vendor_device == 0xa01f177d);
+
+		addr = bus->ops->map_bus(bus, devfn, 0x70);
+		if (!addr) {
+			*val = ~0;
+			return PCIBIOS_DEVICE_NOT_FOUND;
+		}
+		/* E_CAP */
+		v = readl(addr);
+		has_msix = (v & 0xff00) != 0;
+
+		if (!has_msix && where_a == 0x70) {
+			v |= 0xbc00; /* next capability is EA at 0xbc */
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+		if (where_a == 0xb0) {
+			addr = bus->ops->map_bus(bus, devfn, where_a);
+			if (!addr) {
+				*val = ~0;
+				return PCIBIOS_DEVICE_NOT_FOUND;
+			}
+			v = readl(addr);
+			if (v & 0xff00)
+				pr_err("Bad MSIX cap header: %08x\n", v);
+			v |= 0xbc00; /* next capability is EA at 0xbc */
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+		if (where_a == 0xbc) {
+			if (is_nic)
+				v = 0x40014; /* EA last in chain, 4 entries */
+			else if (is_tns)
+				v = 0x30014; /* EA last in chain, 3 entries */
+			else if (has_msix)
+				v = 0x20014; /* EA last in chain, 2 entries */
+			else
+				v = 0x10014; /* EA last in chain, 1 entry */
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+		if (where_a >= 0xc0 && where_a < 0xd0)
+			/* EA entry-0. PP=0, BAR0 Size:3 */
+			return handle_ea_bar(0x80ff0003,
+					     0x10, bus, devfn, where,
+					     size, val);
+		if (where_a >= 0xd0 && where_a < 0xe0 && has_msix)
+			 /* EA entry-1. PP=0, BAR4 Size:3 */
+			return handle_ea_bar(0x80ff0043,
+					     0x20, bus, devfn, where,
+					     size, val);
+		if (where_a >= 0xe0 && where_a < 0xf0 && is_tns)
+			/* EA entry-2. PP=0, BAR2, Size:3 */
+			return handle_ea_bar(0x80ff0023,
+					     0x18, bus, devfn, where,
+					     size, val);
+		if (where_a >= 0xe0 && where_a < 0xf0 && is_nic)
+			/* EA entry-2. PP=4, VF_BAR0 (9), Size:3 */
+			return handle_ea_bar(0x80ff0493,
+					     0x1a4, bus, devfn, where,
+					     size, val);
+		if (where_a >= 0xf0 && where_a < 0x100 && is_nic)
+			/* EA entry-3. PP=4, VF_BAR4 (d), Size:3 */
+			return handle_ea_bar(0x80ff04d3,
+					     0x1b4, bus, devfn, where,
+					     size, val);
+	} else if (cfg_type == 1) {
+		bool is_rsl_bridge = devfn == 0x08;
+		bool is_rad_bridge = devfn == 0xa0;
+		bool is_zip_bridge = devfn == 0xa8;
+		bool is_dfa_bridge = devfn == 0xb0;
+		bool is_nic_bridge = devfn == 0x10;
+
+		if (where_a == 0x70) {
+			addr = bus->ops->map_bus(bus, devfn, where_a);
+			if (!addr) {
+				*val = ~0;
+				return PCIBIOS_DEVICE_NOT_FOUND;
+			}
+			v = readl(addr);
+			if (v & 0xff00)
+				pr_err("Bad PCIe cap header: %08x\n", v);
+			v |= 0xbc00; /* next capability is EA at 0xbc */
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+		if (where_a == 0xbc) {
+			if (is_nic_bridge)
+				v = 0x10014; /* EA last in chain, 1 entry */
+			else
+				v = 0x00014; /* EA last in chain, no entries */
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+		if (where_a == 0xc0) {
+			if (is_rsl_bridge || is_nic_bridge)
+				v = 0x0101; /* subordinate:secondary = 1:1 */
+			else if (is_rad_bridge)
+				v = 0x0202; /* subordinate:secondary = 2:2 */
+			else if (is_zip_bridge)
+				v = 0x0303; /* subordinate:secondary = 3:3 */
+			else if (is_dfa_bridge)
+				v = 0x0404; /* subordinate:secondary = 4:4 */
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+		if (where_a == 0xc4 && is_nic_bridge) {
+			/* Enabled, not-Write, SP=ff, PP=05, BEI=6, ES=4 */
+			v = 0x80ff0564;
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+		if (where_a == 0xc8 && is_nic_bridge) {
+			v = 0x00000002; /* Base-L 64-bit */
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+		if (where_a == 0xcc && is_nic_bridge) {
+			v = 0xfffffffe; /* MaxOffset-L 64-bit */
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+		if (where_a == 0xd0 && is_nic_bridge) {
+			v = 0x00008430; /* NIC Base-H */
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+		if (where_a == 0xd4 && is_nic_bridge) {
+			v = 0x0000000f; /* MaxOffset-H */
+			set_val(v, where, size, val);
+			return PCIBIOS_SUCCESSFUL;
+		}
+	}
+no_emulation:
+	return pci_generic_config_read(bus, devfn, where, size, val);
+}
+
+static int thunder_ecam_config_write(struct pci_bus *bus, unsigned int devfn,
+				     int where, int size, u32 val)
+{
+	/*
+	 * All BARs have fixed addresses; ignore BAR writes so they
+	 * don't get corrupted.
+	 */
+	if ((where >= 0x10 && where < 0x2c) ||
+	    (where >= 0x1a4 && where < 0x1bc))
+		/* BAR or SR-IOV BAR */
+		return PCIBIOS_SUCCESSFUL;
+
+	return pci_generic_config_write(bus, devfn, where, size, val);
+}
+
+static struct gen_pci_cfg_bus_ops thunder_ecam_bus_ops = {
+	.bus_shift	= 20,
+	.ops		= {
+		.map_bus        = thunder_ecam_map_bus,
+		.read           = thunder_ecam_config_read,
+		.write          = thunder_ecam_config_write,
+	}
+};
+
+static const struct of_device_id thunder_ecam_of_match[] = {
+	{ .compatible = "cavium,pci-host-thunder-ecam",
+	  .data = &thunder_ecam_bus_ops },
+
+	{ },
+};
+MODULE_DEVICE_TABLE(of, thunder_ecam_of_match);
+
+static int thunder_ecam_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	const struct of_device_id *of_id;
+	struct gen_pci *pci = devm_kzalloc(dev, sizeof(*pci), GFP_KERNEL);
+
+	if (!pci)
+		return -ENOMEM;
+
+	of_id = of_match_node(thunder_ecam_of_match, dev->of_node);
+	pci->cfg.ops = (struct gen_pci_cfg_bus_ops *)of_id->data;
+
+	return pci_host_common_probe(pdev, pci);
+}
+
+static struct platform_driver thunder_ecam_driver = {
+	.driver = {
+		.name = KBUILD_MODNAME,
+		.of_match_table = thunder_ecam_of_match,
+	},
+	.probe = thunder_ecam_probe,
+};
+module_platform_driver(thunder_ecam_driver);
+
+MODULE_DESCRIPTION("Thunder ECAM PCI host driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/pci/host/pci-thunder-pem.c b/drivers/pci/host/pci-thunder-pem.c
new file mode 100644
index 000000000000..cabb92a514ac
--- /dev/null
+++ b/drivers/pci/host/pci-thunder-pem.c
@@ -0,0 +1,346 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Copyright (C) 2015 - 2016 Cavium, Inc.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of_address.h>
+#include <linux/of_pci.h>
+#include <linux/platform_device.h>
+
+#include "pci-host-common.h"
+
+#define PEM_CFG_WR 0x28
+#define PEM_CFG_RD 0x30
+
+struct thunder_pem_pci {
+	struct gen_pci	gen_pci;
+	u32		ea_entry[3];
+	void __iomem	*pem_reg_base;
+};
+
+static void __iomem *thunder_pem_map_bus(struct pci_bus *bus,
+					 unsigned int devfn, int where)
+{
+	struct gen_pci *pci = bus->sysdata;
+	resource_size_t idx = bus->number - pci->cfg.bus_range->start;
+
+	return pci->cfg.win[idx] + ((devfn << 16) | where);
+}
+
+static int thunder_pem_bridge_read(struct pci_bus *bus, unsigned int devfn,
+				   int where, int size, u32 *val)
+{
+	u64 read_val;
+	struct thunder_pem_pci *pem_pci;
+	struct gen_pci *pci = bus->sysdata;
+
+	pem_pci = container_of(pci, struct thunder_pem_pci, gen_pci);
+
+	if (devfn != 0 || where >= 2048) {
+		*val = ~0;
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	}
+
+	/*
+	 * 32-bit accesses only.  Write the address to the low order
+	 * bits of PEM_CFG_RD, then trigger the read by reading back.
+	 * The config data lands in the upper 32-bits of PEM_CFG_RD.
+	 */
+	read_val = where & ~3ull;
+	writeq(read_val, pem_pci->pem_reg_base + PEM_CFG_RD);
+	read_val = readq(pem_pci->pem_reg_base + PEM_CFG_RD);
+	read_val >>= 32;
+
+	/*
+	 * The config space contains some garbage, fix it up.  Also
+	 * synthesize an EA capability for the BAR used by MSI-X.
+	 */
+	switch (where & ~3) {
+	case 0x40:
+		read_val &= 0xffff00ff;
+		read_val |= 0x00007000; /* Skip MSI CAP */
+		break;
+	case 0x70: /* Express Cap */
+		/* PME interrupt on vector 2*/
+		read_val |= (2u << 25);
+		break;
+	case 0xb0: /* MSI-X Cap */
+		/* TableSize=4, Next Cap is EA */
+		read_val &= 0xc00000ff;
+		read_val |= 0x0003bc00;
+		break;
+	case 0xb4:
+		/* Table offset=0, BIR=0 */
+		read_val = 0x00000000;
+		break;
+	case 0xb8:
+		/* BPA offset=0xf0000, BIR=0 */
+		read_val = 0x000f0000;
+		break;
+	case 0xbc:
+		/* EA, 1 entry, no next Cap */
+		read_val = 0x00010014;
+		break;
+	case 0xc0:
+		/* DW2 for type-1 */
+		read_val = 0x00000000;
+		break;
+	case 0xc4:
+		/* Entry BEI=0, PP=0x00, SP=0xff, ES=3 */
+		read_val = 0x80ff0003;
+		break;
+	case 0xc8:
+		read_val = pem_pci->ea_entry[0];
+		break;
+	case 0xcc:
+		read_val = pem_pci->ea_entry[1];
+		break;
+	case 0xd0:
+		read_val = pem_pci->ea_entry[2];
+		break;
+	default:
+		break;
+	}
+	read_val >>= (8 * (where & 3));
+	switch (size) {
+	case 1:
+		read_val &= 0xff;
+		break;
+	case 2:
+		read_val &= 0xffff;
+		break;
+	default:
+		break;
+	}
+	*val = read_val;
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int thunder_pem_config_read(struct pci_bus *bus, unsigned int devfn,
+				   int where, int size, u32 *val)
+{
+	struct gen_pci *pci = bus->sysdata;
+
+	if (bus->number < pci->cfg.bus_range->start ||
+	    bus->number > pci->cfg.bus_range->end)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	/*
+	 * The first device on the bus is the PEM PCIe bridge.
+	 * Special case its config access.
+	 */
+	if (bus->number == pci->cfg.bus_range->start)
+		return thunder_pem_bridge_read(bus, devfn, where, size, val);
+
+	return pci_generic_config_read(bus, devfn, where, size, val);
+}
+
+/*
+ * Some of the w1c_bits below also include read-only or non-writable
+ * reserved bits, this makes the code simpler and is OK as the bits
+ * are not affected by writing zeros to them.
+ */
+static u32 thunder_pem_bridge_w1c_bits(int where)
+{
+	u32 w1c_bits = 0;
+
+	switch (where & ~3) {
+	case 0x04: /* Command/Status */
+	case 0x1c: /* Base and I/O Limit/Secondary Status */
+		w1c_bits = 0xff000000;
+		break;
+	case 0x44: /* Power Management Control and Status */
+		w1c_bits = 0xfffffe00;
+		break;
+	case 0x78: /* Device Control/Device Status */
+	case 0x80: /* Link Control/Link Status */
+	case 0x88: /* Slot Control/Slot Status */
+	case 0x90: /* Root Status */
+	case 0xa0: /* Link Control 2 Registers/Link Status 2 */
+		w1c_bits = 0xffff0000;
+		break;
+	case 0x104: /* Uncorrectable Error Status */
+	case 0x110: /* Correctable Error Status */
+	case 0x130: /* Error Status */
+	case 0x160: /* Link Control 4 */
+		w1c_bits = 0xffffffff;
+		break;
+	default:
+		break;
+	}
+	return w1c_bits;
+}
+
+static int thunder_pem_bridge_write(struct pci_bus *bus, unsigned int devfn,
+				    int where, int size, u32 val)
+{
+	struct gen_pci *pci = bus->sysdata;
+	struct thunder_pem_pci *pem_pci;
+	u64 write_val, read_val;
+	u32 mask = 0;
+
+	pem_pci = container_of(pci, struct thunder_pem_pci, gen_pci);
+
+	if (devfn != 0 || where >= 2048)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	/*
+	 * 32-bit accesses only.  If the write is for a size smaller
+	 * than 32-bits, we must first read the 32-bit value and merge
+	 * in the desired bits and then write the whole 32-bits back
+	 * out.
+	 */
+	switch (size) {
+	case 1:
+		read_val = where & ~3ull;
+		writeq(read_val, pem_pci->pem_reg_base + PEM_CFG_RD);
+		read_val = readq(pem_pci->pem_reg_base + PEM_CFG_RD);
+		read_val >>= 32;
+		mask = ~(0xff << (8 * (where & 3)));
+		read_val &= mask;
+		val = (val & 0xff) << (8 * (where & 3));
+		val |= (u32)read_val;
+		break;
+	case 2:
+		read_val = where & ~3ull;
+		writeq(read_val, pem_pci->pem_reg_base + PEM_CFG_RD);
+		read_val = readq(pem_pci->pem_reg_base + PEM_CFG_RD);
+		read_val >>= 32;
+		mask = ~(0xffff << (8 * (where & 3)));
+		read_val &= mask;
+		val = (val & 0xffff) << (8 * (where & 3));
+		val |= (u32)read_val;
+		break;
+	default:
+		break;
+	}
+
+	/*
+	 * By expanding the write width to 32 bits, we may
+	 * inadvertently hit some W1C bits that were not intended to
+	 * be written.  Calculate the mask that must be applied to the
+	 * data to be written to avoid these cases.
+	 */
+	if (mask) {
+		u32 w1c_bits = thunder_pem_bridge_w1c_bits(where);
+
+		if (w1c_bits) {
+			mask &= w1c_bits;
+			val &= ~mask;
+		}
+	}
+
+	/*
+	 * Low order bits are the config address, the high order 32
+	 * bits are the data to be written.
+	 */
+	write_val = where & ~3ull;
+	write_val |= (((u64)val) << 32);
+	writeq(write_val, pem_pci->pem_reg_base + PEM_CFG_WR);
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int thunder_pem_config_write(struct pci_bus *bus, unsigned int devfn,
+				    int where, int size, u32 val)
+{
+	struct gen_pci *pci = bus->sysdata;
+
+	if (bus->number < pci->cfg.bus_range->start ||
+	    bus->number > pci->cfg.bus_range->end)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	/*
+	 * The first device on the bus is the PEM PCIe bridge.
+	 * Special case its config access.
+	 */
+	if (bus->number == pci->cfg.bus_range->start)
+		return thunder_pem_bridge_write(bus, devfn, where, size, val);
+
+
+	return pci_generic_config_write(bus, devfn, where, size, val);
+}
+
+static struct gen_pci_cfg_bus_ops thunder_pem_bus_ops = {
+	.bus_shift	= 24,
+	.ops		= {
+		.map_bus	= thunder_pem_map_bus,
+		.read		= thunder_pem_config_read,
+		.write		= thunder_pem_config_write,
+	}
+};
+
+static const struct of_device_id thunder_pem_of_match[] = {
+	{ .compatible = "cavium,pci-host-thunder-pem",
+	  .data = &thunder_pem_bus_ops },
+
+	{ },
+};
+MODULE_DEVICE_TABLE(of, thunder_pem_of_match);
+
+static int thunder_pem_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	const struct of_device_id *of_id;
+	resource_size_t bar4_start;
+	struct resource *res_pem;
+	struct thunder_pem_pci *pem_pci;
+
+	pem_pci = devm_kzalloc(dev, sizeof(*pem_pci), GFP_KERNEL);
+	if (!pem_pci)
+		return -ENOMEM;
+
+	of_id = of_match_node(thunder_pem_of_match, dev->of_node);
+	pem_pci->gen_pci.cfg.ops = (struct gen_pci_cfg_bus_ops *)of_id->data;
+
+	/*
+	 * The second register range is the PEM bridge to the PCIe
+	 * bus.  It has a different config access method than those
+	 * devices behind the bridge.
+	 */
+	res_pem = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	if (!res_pem) {
+		dev_err(dev, "missing \"reg[1]\"property\n");
+		return -EINVAL;
+	}
+
+	pem_pci->pem_reg_base = devm_ioremap(dev, res_pem->start, 0x10000);
+	if (!pem_pci->pem_reg_base)
+		return -ENOMEM;
+
+	/*
+	 * The MSI-X BAR for the PEM and AER interrupts is located at
+	 * a fixed offset from the PEM register base.  Generate a
+	 * fragment of the synthesized Enhanced Allocation capability
+	 * structure here for the BAR.
+	 */
+	bar4_start = res_pem->start + 0xf00000;
+	pem_pci->ea_entry[0] = (u32)bar4_start | 2;
+	pem_pci->ea_entry[1] = (u32)(res_pem->end - bar4_start) & ~3u;
+	pem_pci->ea_entry[2] = (u32)(bar4_start >> 32);
+
+	return pci_host_common_probe(pdev, &pem_pci->gen_pci);
+}
+
+static struct platform_driver thunder_pem_driver = {
+	.driver = {
+		.name = KBUILD_MODNAME,
+		.of_match_table = thunder_pem_of_match,
+	},
+	.probe = thunder_pem_probe,
+};
+module_platform_driver(thunder_pem_driver);
+
+MODULE_DESCRIPTION("Thunder PEM PCIe host driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/pci/host/pcie-altera.c b/drivers/pci/host/pcie-altera.c
index 99da549d5d06..dbac6fb3f0bd 100644
--- a/drivers/pci/host/pcie-altera.c
+++ b/drivers/pci/host/pcie-altera.c
@@ -40,6 +40,7 @@
 #define P2A_INT_ENABLE			0x3070
 #define P2A_INT_ENA_ALL			0xf
 #define RP_LTSSM			0x3c64
+#define RP_LTSSM_MASK			0x1f
 #define LTSSM_L0			0xf
 
 /* TLP configuration type 0 and 1 */
@@ -140,7 +141,7 @@ static void tlp_write_tx(struct altera_pcie *pcie,
 
 static bool altera_pcie_link_is_up(struct altera_pcie *pcie)
 {
-	return !!(cra_readl(pcie, RP_LTSSM) & LTSSM_L0);
+	return !!((cra_readl(pcie, RP_LTSSM) & RP_LTSSM_MASK) == LTSSM_L0);
 }
 
 static bool altera_pcie_valid_config(struct altera_pcie *pcie,
diff --git a/drivers/pci/host/pcie-designware-plat.c b/drivers/pci/host/pcie-designware-plat.c
new file mode 100644
index 000000000000..b3500994d08a
--- /dev/null
+++ b/drivers/pci/host/pcie-designware-plat.c
@@ -0,0 +1,138 @@
+/*
+ * PCIe RC driver for Synopsys DesignWare Core
+ *
+ * Copyright (C) 2015-2016 Synopsys, Inc. (www.synopsys.com)
+ *
+ * Authors: Joao Pinto <jpinto@synopsys.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/gpio.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of_gpio.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+#include <linux/resource.h>
+#include <linux/signal.h>
+#include <linux/types.h>
+
+#include "pcie-designware.h"
+
+struct dw_plat_pcie {
+	void __iomem		*mem_base;
+	struct pcie_port	pp;
+};
+
+static irqreturn_t dw_plat_pcie_msi_irq_handler(int irq, void *arg)
+{
+	struct pcie_port *pp = arg;
+
+	return dw_handle_msi_irq(pp);
+}
+
+static void dw_plat_pcie_host_init(struct pcie_port *pp)
+{
+	dw_pcie_setup_rc(pp);
+	dw_pcie_wait_for_link(pp);
+
+	if (IS_ENABLED(CONFIG_PCI_MSI))
+		dw_pcie_msi_init(pp);
+}
+
+static struct pcie_host_ops dw_plat_pcie_host_ops = {
+	.host_init = dw_plat_pcie_host_init,
+};
+
+static int dw_plat_add_pcie_port(struct pcie_port *pp,
+				 struct platform_device *pdev)
+{
+	int ret;
+
+	pp->irq = platform_get_irq(pdev, 1);
+	if (pp->irq < 0)
+		return pp->irq;
+
+	if (IS_ENABLED(CONFIG_PCI_MSI)) {
+		pp->msi_irq = platform_get_irq(pdev, 0);
+		if (pp->msi_irq < 0)
+			return pp->msi_irq;
+
+		ret = devm_request_irq(&pdev->dev, pp->msi_irq,
+					dw_plat_pcie_msi_irq_handler,
+					IRQF_SHARED, "dw-plat-pcie-msi", pp);
+		if (ret) {
+			dev_err(&pdev->dev, "failed to request MSI IRQ\n");
+			return ret;
+		}
+	}
+
+	pp->root_bus_nr = -1;
+	pp->ops = &dw_plat_pcie_host_ops;
+
+	ret = dw_pcie_host_init(pp);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to initialize host\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int dw_plat_pcie_probe(struct platform_device *pdev)
+{
+	struct dw_plat_pcie *dw_plat_pcie;
+	struct pcie_port *pp;
+	struct resource *res;  /* Resource from DT */
+	int ret;
+
+	dw_plat_pcie = devm_kzalloc(&pdev->dev, sizeof(*dw_plat_pcie),
+					GFP_KERNEL);
+	if (!dw_plat_pcie)
+		return -ENOMEM;
+
+	pp = &dw_plat_pcie->pp;
+	pp->dev = &pdev->dev;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res)
+		return -ENODEV;
+
+	dw_plat_pcie->mem_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(dw_plat_pcie->mem_base))
+		return PTR_ERR(dw_plat_pcie->mem_base);
+
+	pp->dbi_base = dw_plat_pcie->mem_base;
+
+	ret = dw_plat_add_pcie_port(pp, pdev);
+	if (ret < 0)
+		return ret;
+
+	platform_set_drvdata(pdev, dw_plat_pcie);
+	return 0;
+}
+
+static const struct of_device_id dw_plat_pcie_of_match[] = {
+	{ .compatible = "snps,dw-pcie", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, dw_plat_pcie_of_match);
+
+static struct platform_driver dw_plat_pcie_driver = {
+	.driver = {
+		.name	= "dw-pcie",
+		.of_match_table = dw_plat_pcie_of_match,
+	},
+	.probe = dw_plat_pcie_probe,
+};
+
+module_platform_driver(dw_plat_pcie_driver);
+
+MODULE_AUTHOR("Joao Pinto <Joao.Pinto@synopsys.com>");
+MODULE_DESCRIPTION("Synopsys PCIe host controller glue platform driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
index 21716827847a..a4cccd356304 100644
--- a/drivers/pci/host/pcie-designware.c
+++ b/drivers/pci/host/pcie-designware.c
@@ -22,6 +22,7 @@
 #include <linux/pci_regs.h>
 #include <linux/platform_device.h>
 #include <linux/types.h>
+#include <linux/delay.h>
 
 #include "pcie-designware.h"
 
@@ -69,6 +70,11 @@
 #define PCIE_ATU_FUNC(x)		(((x) & 0x7) << 16)
 #define PCIE_ATU_UPPER_TARGET		0x91C
 
+/* PCIe Port Logic registers */
+#define PLR_OFFSET			0x700
+#define PCIE_PHY_DEBUG_R1		(PLR_OFFSET + 0x2c)
+#define PCIE_PHY_DEBUG_R1_LINK_UP	0x00000010
+
 static struct pci_ops dw_pcie_ops;
 
 int dw_pcie_cfg_read(void __iomem *addr, int size, u32 *val)
@@ -380,12 +386,33 @@ static struct msi_controller dw_pcie_msi_chip = {
 	.teardown_irq = dw_msi_teardown_irq,
 };
 
+int dw_pcie_wait_for_link(struct pcie_port *pp)
+{
+	int retries;
+
+	/* check if the link is up or not */
+	for (retries = 0; retries < LINK_WAIT_MAX_RETRIES; retries++) {
+		if (dw_pcie_link_up(pp)) {
+			dev_info(pp->dev, "link up\n");
+			return 0;
+		}
+		usleep_range(LINK_WAIT_USLEEP_MIN, LINK_WAIT_USLEEP_MAX);
+	}
+
+	dev_err(pp->dev, "phy link never came up\n");
+
+	return -ETIMEDOUT;
+}
+
 int dw_pcie_link_up(struct pcie_port *pp)
 {
+	u32 val;
+
 	if (pp->ops->link_up)
 		return pp->ops->link_up(pp);
 
-	return 0;
+	val = readl(pp->dbi_base + PCIE_PHY_DEBUG_R1);
+	return val & PCIE_PHY_DEBUG_R1_LINK_UP;
 }
 
 static int dw_pcie_msi_map(struct irq_domain *domain, unsigned int irq,
@@ -517,6 +544,11 @@ int dw_pcie_host_init(struct pcie_port *pp)
 	if (pp->ops->host_init)
 		pp->ops->host_init(pp);
 
+	/*
+	 * If the platform provides ->rd_other_conf, it means the platform
+	 * uses its own address translation component rather than ATU, so
+	 * we should not program the ATU here.
+	 */
 	if (!pp->ops->rd_other_conf)
 		dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX1,
 					  PCIE_ATU_TYPE_MEM, pp->mem_base,
@@ -551,13 +583,11 @@ int dw_pcie_host_init(struct pcie_port *pp)
 	pci_fixup_irqs(pci_common_swizzle, of_irq_parse_and_map_pci);
 #endif
 
-	if (!pci_has_flag(PCI_PROBE_ONLY)) {
-		pci_bus_size_bridges(bus);
-		pci_bus_assign_resources(bus);
+	pci_bus_size_bridges(bus);
+	pci_bus_assign_resources(bus);
 
-		list_for_each_entry(child, &bus->children, node)
-			pcie_bus_configure_settings(child);
-	}
+	list_for_each_entry(child, &bus->children, node)
+		pcie_bus_configure_settings(child);
 
 	pci_bus_add_devices(bus);
 	return 0;
diff --git a/drivers/pci/host/pcie-designware.h b/drivers/pci/host/pcie-designware.h
index 2356d29e8527..f437f9b5be04 100644
--- a/drivers/pci/host/pcie-designware.h
+++ b/drivers/pci/host/pcie-designware.h
@@ -22,6 +22,11 @@
 #define MAX_MSI_IRQS			32
 #define MAX_MSI_CTRLS			(MAX_MSI_IRQS / 32)
 
+/* Parameters for the waiting for link up routine */
+#define LINK_WAIT_MAX_RETRIES		10
+#define LINK_WAIT_USLEEP_MIN		90000
+#define LINK_WAIT_USLEEP_MAX		100000
+
 struct pcie_port {
 	struct device		*dev;
 	u8			root_bus_nr;
@@ -76,6 +81,7 @@ int dw_pcie_cfg_read(void __iomem *addr, int size, u32 *val);
 int dw_pcie_cfg_write(void __iomem *addr, int size, u32 val);
 irqreturn_t dw_handle_msi_irq(struct pcie_port *pp);
 void dw_pcie_msi_init(struct pcie_port *pp);
+int dw_pcie_wait_for_link(struct pcie_port *pp);
 int dw_pcie_link_up(struct pcie_port *pp);
 void dw_pcie_setup_rc(struct pcie_port *pp);
 int dw_pcie_host_init(struct pcie_port *pp);
diff --git a/drivers/pci/host/pcie-qcom.c b/drivers/pci/host/pcie-qcom.c
index e845fba19632..f2f90c50f75d 100644
--- a/drivers/pci/host/pcie-qcom.c
+++ b/drivers/pci/host/pcie-qcom.c
@@ -116,8 +116,6 @@ static irqreturn_t qcom_pcie_msi_irq_handler(int irq, void *arg)
 
 static int qcom_pcie_establish_link(struct qcom_pcie *pcie)
 {
-	struct device *dev = pcie->dev;
-	unsigned int retries = 0;
 	u32 val;
 
 	if (dw_pcie_link_up(&pcie->pp))
@@ -128,15 +126,7 @@ static int qcom_pcie_establish_link(struct qcom_pcie *pcie)
 	val |= PCIE20_ELBI_SYS_CTRL_LT_ENABLE;
 	writel(val, pcie->elbi + PCIE20_ELBI_SYS_CTRL);
 
-	do {
-		if (dw_pcie_link_up(&pcie->pp))
-			return 0;
-		usleep_range(250, 1000);
-	} while (retries < 200);
-
-	dev_warn(dev, "phy link never came up\n");
-
-	return -ETIMEDOUT;
+	return dw_pcie_wait_for_link(&pcie->pp);
 }
 
 static int qcom_pcie_get_resources_v0(struct qcom_pcie *pcie)
diff --git a/drivers/pci/host/pcie-rcar.c b/drivers/pci/host/pcie-rcar.c
index 4edb5181f4e2..35092188039b 100644
--- a/drivers/pci/host/pcie-rcar.c
+++ b/drivers/pci/host/pcie-rcar.c
@@ -390,9 +390,7 @@ static int rcar_pcie_enable(struct rcar_pcie *pcie)
 
 	rcar_pcie_setup(&res, pcie);
 
-	/* Do not reassign resources if probe only */
-	if (!pci_has_flag(PCI_PROBE_ONLY))
-		pci_add_flags(PCI_REASSIGN_ALL_RSRC | PCI_REASSIGN_ALL_BUS);
+	pci_add_flags(PCI_REASSIGN_ALL_RSRC | PCI_REASSIGN_ALL_BUS);
 
 	if (IS_ENABLED(CONFIG_PCI_MSI))
 		bus = pci_scan_root_bus_msi(pcie->dev, pcie->root_bus_nr,
@@ -408,13 +406,11 @@ static int rcar_pcie_enable(struct rcar_pcie *pcie)
 
 	pci_fixup_irqs(pci_common_swizzle, of_irq_parse_and_map_pci);
 
-	if (!pci_has_flag(PCI_PROBE_ONLY)) {
-		pci_bus_size_bridges(bus);
-		pci_bus_assign_resources(bus);
+	pci_bus_size_bridges(bus);
+	pci_bus_assign_resources(bus);
 
-		list_for_each_entry(child, &bus->children, node)
-			pcie_bus_configure_settings(child);
-	}
+	list_for_each_entry(child, &bus->children, node)
+		pcie_bus_configure_settings(child);
 
 	pci_bus_add_devices(bus);
 
diff --git a/drivers/pci/host/pcie-spear13xx.c b/drivers/pci/host/pcie-spear13xx.c
index a6cd8233e8c0..a4060b85ab23 100644
--- a/drivers/pci/host/pcie-spear13xx.c
+++ b/drivers/pci/host/pcie-spear13xx.c
@@ -13,7 +13,6 @@
  */
 
 #include <linux/clk.h>
-#include <linux/delay.h>
 #include <linux/interrupt.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
@@ -149,7 +148,6 @@ static int spear13xx_pcie_establish_link(struct pcie_port *pp)
 	struct spear13xx_pcie *spear13xx_pcie = to_spear13xx_pcie(pp);
 	struct pcie_app_reg *app_reg = spear13xx_pcie->app_base;
 	u32 exp_cap_off = EXP_CAP_ID_OFFSET;
-	unsigned int retries;
 
 	if (dw_pcie_link_up(pp)) {
 		dev_err(pp->dev, "link already up\n");
@@ -200,17 +198,7 @@ static int spear13xx_pcie_establish_link(struct pcie_port *pp)
 			| ((u32)1 << REG_TRANSLATION_ENABLE),
 			&app_reg->app_ctrl_0);
 
-	/* check if the link is up or not */
-	for (retries = 0; retries < 10; retries++) {
-		if (dw_pcie_link_up(pp)) {
-			dev_info(pp->dev, "link up\n");
-			return 0;
-		}
-		mdelay(100);
-	}
-
-	dev_err(pp->dev, "link Fail\n");
-	return -EINVAL;
+	return dw_pcie_wait_for_link(pp);
 }
 
 static irqreturn_t spear13xx_pcie_irq_handler(int irq, void *arg)
diff --git a/drivers/pci/host/pcie-xilinx-nwl.c b/drivers/pci/host/pcie-xilinx-nwl.c
new file mode 100644
index 000000000000..5139e6443bbd
--- /dev/null
+++ b/drivers/pci/host/pcie-xilinx-nwl.c
@@ -0,0 +1,881 @@
+/*
+ * PCIe host controller driver for NWL PCIe Bridge
+ * Based on pcie-xilinx.c, pci-tegra.c
+ *
+ * (C) Copyright 2014 - 2015, Xilinx, Inc.
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/irqdomain.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/msi.h>
+#include <linux/of_address.h>
+#include <linux/of_pci.h>
+#include <linux/of_platform.h>
+#include <linux/of_irq.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+#include <linux/irqchip/chained_irq.h>
+
+/* Bridge core config registers */
+#define BRCFG_PCIE_RX0			0x00000000
+#define BRCFG_INTERRUPT			0x00000010
+#define BRCFG_PCIE_RX_MSG_FILTER	0x00000020
+
+/* Egress - Bridge translation registers */
+#define E_BREG_CAPABILITIES		0x00000200
+#define E_BREG_CONTROL			0x00000208
+#define E_BREG_BASE_LO			0x00000210
+#define E_BREG_BASE_HI			0x00000214
+#define E_ECAM_CAPABILITIES		0x00000220
+#define E_ECAM_CONTROL			0x00000228
+#define E_ECAM_BASE_LO			0x00000230
+#define E_ECAM_BASE_HI			0x00000234
+
+/* Ingress - address translations */
+#define I_MSII_CAPABILITIES		0x00000300
+#define I_MSII_CONTROL			0x00000308
+#define I_MSII_BASE_LO			0x00000310
+#define I_MSII_BASE_HI			0x00000314
+
+#define I_ISUB_CONTROL			0x000003E8
+#define SET_ISUB_CONTROL		BIT(0)
+/* Rxed msg fifo  - Interrupt status registers */
+#define MSGF_MISC_STATUS		0x00000400
+#define MSGF_MISC_MASK			0x00000404
+#define MSGF_LEG_STATUS			0x00000420
+#define MSGF_LEG_MASK			0x00000424
+#define MSGF_MSI_STATUS_LO		0x00000440
+#define MSGF_MSI_STATUS_HI		0x00000444
+#define MSGF_MSI_MASK_LO		0x00000448
+#define MSGF_MSI_MASK_HI		0x0000044C
+
+/* Msg filter mask bits */
+#define CFG_ENABLE_PM_MSG_FWD		BIT(1)
+#define CFG_ENABLE_INT_MSG_FWD		BIT(2)
+#define CFG_ENABLE_ERR_MSG_FWD		BIT(3)
+#define CFG_ENABLE_SLT_MSG_FWD		BIT(5)
+#define CFG_ENABLE_VEN_MSG_FWD		BIT(7)
+#define CFG_ENABLE_OTH_MSG_FWD		BIT(13)
+#define CFG_ENABLE_VEN_MSG_EN		BIT(14)
+#define CFG_ENABLE_VEN_MSG_VEN_INV	BIT(15)
+#define CFG_ENABLE_VEN_MSG_VEN_ID	GENMASK(31, 16)
+#define CFG_ENABLE_MSG_FILTER_MASK	(CFG_ENABLE_PM_MSG_FWD | \
+					CFG_ENABLE_INT_MSG_FWD | \
+					CFG_ENABLE_ERR_MSG_FWD | \
+					CFG_ENABLE_SLT_MSG_FWD | \
+					CFG_ENABLE_VEN_MSG_FWD | \
+					CFG_ENABLE_OTH_MSG_FWD | \
+					CFG_ENABLE_VEN_MSG_EN | \
+					CFG_ENABLE_VEN_MSG_VEN_INV | \
+					CFG_ENABLE_VEN_MSG_VEN_ID)
+
+/* Misc interrupt status mask bits */
+#define MSGF_MISC_SR_RXMSG_AVAIL	BIT(0)
+#define MSGF_MISC_SR_RXMSG_OVER		BIT(1)
+#define MSGF_MISC_SR_SLAVE_ERR		BIT(4)
+#define MSGF_MISC_SR_MASTER_ERR		BIT(5)
+#define MSGF_MISC_SR_I_ADDR_ERR		BIT(6)
+#define MSGF_MISC_SR_E_ADDR_ERR		BIT(7)
+#define MSGF_MISC_SR_UR_DETECT          BIT(20)
+
+#define MSGF_MISC_SR_PCIE_CORE		GENMASK(18, 16)
+#define MSGF_MISC_SR_PCIE_CORE_ERR	GENMASK(31, 22)
+
+#define MSGF_MISC_SR_MASKALL		(MSGF_MISC_SR_RXMSG_AVAIL | \
+					MSGF_MISC_SR_RXMSG_OVER | \
+					MSGF_MISC_SR_SLAVE_ERR | \
+					MSGF_MISC_SR_MASTER_ERR | \
+					MSGF_MISC_SR_I_ADDR_ERR | \
+					MSGF_MISC_SR_E_ADDR_ERR | \
+					MSGF_MISC_SR_UR_DETECT | \
+					MSGF_MISC_SR_PCIE_CORE | \
+					MSGF_MISC_SR_PCIE_CORE_ERR)
+
+/* Legacy interrupt status mask bits */
+#define MSGF_LEG_SR_INTA		BIT(0)
+#define MSGF_LEG_SR_INTB		BIT(1)
+#define MSGF_LEG_SR_INTC		BIT(2)
+#define MSGF_LEG_SR_INTD		BIT(3)
+#define MSGF_LEG_SR_MASKALL		(MSGF_LEG_SR_INTA | MSGF_LEG_SR_INTB | \
+					MSGF_LEG_SR_INTC | MSGF_LEG_SR_INTD)
+
+/* MSI interrupt status mask bits */
+#define MSGF_MSI_SR_LO_MASK		BIT(0)
+#define MSGF_MSI_SR_HI_MASK		BIT(0)
+
+#define MSII_PRESENT			BIT(0)
+#define MSII_ENABLE			BIT(0)
+#define MSII_STATUS_ENABLE		BIT(15)
+
+/* Bridge config interrupt mask */
+#define BRCFG_INTERRUPT_MASK		BIT(0)
+#define BREG_PRESENT			BIT(0)
+#define BREG_ENABLE			BIT(0)
+#define BREG_ENABLE_FORCE		BIT(1)
+
+/* E_ECAM status mask bits */
+#define E_ECAM_PRESENT			BIT(0)
+#define E_ECAM_CR_ENABLE		BIT(0)
+#define E_ECAM_SIZE_LOC			GENMASK(20, 16)
+#define E_ECAM_SIZE_SHIFT		16
+#define ECAM_BUS_LOC_SHIFT		20
+#define ECAM_DEV_LOC_SHIFT		12
+#define NWL_ECAM_VALUE_DEFAULT		12
+
+#define CFG_DMA_REG_BAR			GENMASK(2, 0)
+
+#define INT_PCI_MSI_NR			(2 * 32)
+#define INTX_NUM			4
+
+/* Readin the PS_LINKUP */
+#define PS_LINKUP_OFFSET		0x00000238
+#define PCIE_PHY_LINKUP_BIT		BIT(0)
+#define PHY_RDY_LINKUP_BIT		BIT(1)
+
+/* Parameters for the waiting for link up routine */
+#define LINK_WAIT_MAX_RETRIES          10
+#define LINK_WAIT_USLEEP_MIN           90000
+#define LINK_WAIT_USLEEP_MAX           100000
+
+struct nwl_msi {			/* MSI information */
+	struct irq_domain *msi_domain;
+	unsigned long *bitmap;
+	struct irq_domain *dev_domain;
+	struct mutex lock;		/* protect bitmap variable */
+	int irq_msi0;
+	int irq_msi1;
+};
+
+struct nwl_pcie {
+	struct device *dev;
+	void __iomem *breg_base;
+	void __iomem *pcireg_base;
+	void __iomem *ecam_base;
+	phys_addr_t phys_breg_base;	/* Physical Bridge Register Base */
+	phys_addr_t phys_pcie_reg_base;	/* Physical PCIe Controller Base */
+	phys_addr_t phys_ecam_base;	/* Physical Configuration Base */
+	u32 breg_size;
+	u32 pcie_reg_size;
+	u32 ecam_size;
+	int irq_intx;
+	int irq_misc;
+	u32 ecam_value;
+	u8 last_busno;
+	u8 root_busno;
+	struct nwl_msi msi;
+	struct irq_domain *legacy_irq_domain;
+};
+
+static inline u32 nwl_bridge_readl(struct nwl_pcie *pcie, u32 off)
+{
+	return readl(pcie->breg_base + off);
+}
+
+static inline void nwl_bridge_writel(struct nwl_pcie *pcie, u32 val, u32 off)
+{
+	writel(val, pcie->breg_base + off);
+}
+
+static bool nwl_pcie_link_up(struct nwl_pcie *pcie)
+{
+	if (readl(pcie->pcireg_base + PS_LINKUP_OFFSET) & PCIE_PHY_LINKUP_BIT)
+		return true;
+	return false;
+}
+
+static bool nwl_phy_link_up(struct nwl_pcie *pcie)
+{
+	if (readl(pcie->pcireg_base + PS_LINKUP_OFFSET) & PHY_RDY_LINKUP_BIT)
+		return true;
+	return false;
+}
+
+static int nwl_wait_for_link(struct nwl_pcie *pcie)
+{
+	int retries;
+
+	/* check if the link is up or not */
+	for (retries = 0; retries < LINK_WAIT_MAX_RETRIES; retries++) {
+		if (nwl_phy_link_up(pcie))
+			return 0;
+		usleep_range(LINK_WAIT_USLEEP_MIN, LINK_WAIT_USLEEP_MAX);
+	}
+
+	dev_err(pcie->dev, "PHY link never came up\n");
+	return -ETIMEDOUT;
+}
+
+static bool nwl_pcie_valid_device(struct pci_bus *bus, unsigned int devfn)
+{
+	struct nwl_pcie *pcie = bus->sysdata;
+
+	/* Check link before accessing downstream ports */
+	if (bus->number != pcie->root_busno) {
+		if (!nwl_pcie_link_up(pcie))
+			return false;
+	}
+
+	/* Only one device down on each root port */
+	if (bus->number == pcie->root_busno && devfn > 0)
+		return false;
+
+	return true;
+}
+
+/**
+ * nwl_pcie_map_bus - Get configuration base
+ *
+ * @bus: Bus structure of current bus
+ * @devfn: Device/function
+ * @where: Offset from base
+ *
+ * Return: Base address of the configuration space needed to be
+ *	   accessed.
+ */
+static void __iomem *nwl_pcie_map_bus(struct pci_bus *bus, unsigned int devfn,
+				      int where)
+{
+	struct nwl_pcie *pcie = bus->sysdata;
+	int relbus;
+
+	if (!nwl_pcie_valid_device(bus, devfn))
+		return NULL;
+
+	relbus = (bus->number << ECAM_BUS_LOC_SHIFT) |
+			(devfn << ECAM_DEV_LOC_SHIFT);
+
+	return pcie->ecam_base + relbus + where;
+}
+
+/* PCIe operations */
+static struct pci_ops nwl_pcie_ops = {
+	.map_bus = nwl_pcie_map_bus,
+	.read  = pci_generic_config_read,
+	.write = pci_generic_config_write,
+};
+
+static irqreturn_t nwl_pcie_misc_handler(int irq, void *data)
+{
+	struct nwl_pcie *pcie = data;
+	u32 misc_stat;
+
+	/* Checking for misc interrupts */
+	misc_stat = nwl_bridge_readl(pcie, MSGF_MISC_STATUS) &
+				     MSGF_MISC_SR_MASKALL;
+	if (!misc_stat)
+		return IRQ_NONE;
+
+	if (misc_stat & MSGF_MISC_SR_RXMSG_OVER)
+		dev_err(pcie->dev, "Received Message FIFO Overflow\n");
+
+	if (misc_stat & MSGF_MISC_SR_SLAVE_ERR)
+		dev_err(pcie->dev, "Slave error\n");
+
+	if (misc_stat & MSGF_MISC_SR_MASTER_ERR)
+		dev_err(pcie->dev, "Master error\n");
+
+	if (misc_stat & MSGF_MISC_SR_I_ADDR_ERR)
+		dev_err(pcie->dev,
+			"In Misc Ingress address translation error\n");
+
+	if (misc_stat & MSGF_MISC_SR_E_ADDR_ERR)
+		dev_err(pcie->dev,
+			"In Misc Egress address translation error\n");
+
+	if (misc_stat & MSGF_MISC_SR_PCIE_CORE_ERR)
+		dev_err(pcie->dev, "PCIe Core error\n");
+
+	/* Clear misc interrupt status */
+	nwl_bridge_writel(pcie, misc_stat, MSGF_MISC_STATUS);
+
+	return IRQ_HANDLED;
+}
+
+static void nwl_pcie_leg_handler(struct irq_desc *desc)
+{
+	struct irq_chip *chip = irq_desc_get_chip(desc);
+	struct nwl_pcie *pcie;
+	unsigned long status;
+	u32 bit;
+	u32 virq;
+
+	chained_irq_enter(chip, desc);
+	pcie = irq_desc_get_handler_data(desc);
+
+	while ((status = nwl_bridge_readl(pcie, MSGF_LEG_STATUS) &
+				MSGF_LEG_SR_MASKALL) != 0) {
+		for_each_set_bit(bit, &status, INTX_NUM) {
+			virq = irq_find_mapping(pcie->legacy_irq_domain,
+						bit + 1);
+			if (virq)
+				generic_handle_irq(virq);
+		}
+	}
+
+	chained_irq_exit(chip, desc);
+}
+
+static void nwl_pcie_handle_msi_irq(struct nwl_pcie *pcie, u32 status_reg)
+{
+	struct nwl_msi *msi;
+	unsigned long status;
+	u32 bit;
+	u32 virq;
+
+	msi = &pcie->msi;
+
+	while ((status = nwl_bridge_readl(pcie, status_reg)) != 0) {
+		for_each_set_bit(bit, &status, 32) {
+			nwl_bridge_writel(pcie, 1 << bit, status_reg);
+			virq = irq_find_mapping(msi->dev_domain, bit);
+			if (virq)
+				generic_handle_irq(virq);
+		}
+	}
+}
+
+static void nwl_pcie_msi_handler_high(struct irq_desc *desc)
+{
+	struct irq_chip *chip = irq_desc_get_chip(desc);
+	struct nwl_pcie *pcie = irq_desc_get_handler_data(desc);
+
+	chained_irq_enter(chip, desc);
+	nwl_pcie_handle_msi_irq(pcie, MSGF_MSI_STATUS_HI);
+	chained_irq_exit(chip, desc);
+}
+
+static void nwl_pcie_msi_handler_low(struct irq_desc *desc)
+{
+	struct irq_chip *chip = irq_desc_get_chip(desc);
+	struct nwl_pcie *pcie = irq_desc_get_handler_data(desc);
+
+	chained_irq_enter(chip, desc);
+	nwl_pcie_handle_msi_irq(pcie, MSGF_MSI_STATUS_LO);
+	chained_irq_exit(chip, desc);
+}
+
+static int nwl_legacy_map(struct irq_domain *domain, unsigned int irq,
+			  irq_hw_number_t hwirq)
+{
+	irq_set_chip_and_handler(irq, &dummy_irq_chip, handle_simple_irq);
+	irq_set_chip_data(irq, domain->host_data);
+
+	return 0;
+}
+
+static const struct irq_domain_ops legacy_domain_ops = {
+	.map = nwl_legacy_map,
+};
+
+#ifdef CONFIG_PCI_MSI
+static struct irq_chip nwl_msi_irq_chip = {
+	.name = "nwl_pcie:msi",
+	.irq_enable = unmask_msi_irq,
+	.irq_disable = mask_msi_irq,
+	.irq_mask = mask_msi_irq,
+	.irq_unmask = unmask_msi_irq,
+
+};
+
+static struct msi_domain_info nwl_msi_domain_info = {
+	.flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS |
+		  MSI_FLAG_MULTI_PCI_MSI),
+	.chip = &nwl_msi_irq_chip,
+};
+#endif
+
+static void nwl_compose_msi_msg(struct irq_data *data, struct msi_msg *msg)
+{
+	struct nwl_pcie *pcie = irq_data_get_irq_chip_data(data);
+	phys_addr_t msi_addr = pcie->phys_pcie_reg_base;
+
+	msg->address_lo = lower_32_bits(msi_addr);
+	msg->address_hi = upper_32_bits(msi_addr);
+	msg->data = data->hwirq;
+}
+
+static int nwl_msi_set_affinity(struct irq_data *irq_data,
+				const struct cpumask *mask, bool force)
+{
+	return -EINVAL;
+}
+
+static struct irq_chip nwl_irq_chip = {
+	.name = "Xilinx MSI",
+	.irq_compose_msi_msg = nwl_compose_msi_msg,
+	.irq_set_affinity = nwl_msi_set_affinity,
+};
+
+static int nwl_irq_domain_alloc(struct irq_domain *domain, unsigned int virq,
+				unsigned int nr_irqs, void *args)
+{
+	struct nwl_pcie *pcie = domain->host_data;
+	struct nwl_msi *msi = &pcie->msi;
+	int bit;
+	int i;
+
+	mutex_lock(&msi->lock);
+	bit = bitmap_find_next_zero_area(msi->bitmap, INT_PCI_MSI_NR, 0,
+					 nr_irqs, 0);
+	if (bit >= INT_PCI_MSI_NR) {
+		mutex_unlock(&msi->lock);
+		return -ENOSPC;
+	}
+
+	bitmap_set(msi->bitmap, bit, nr_irqs);
+
+	for (i = 0; i < nr_irqs; i++) {
+		irq_domain_set_info(domain, virq + i, bit + i, &nwl_irq_chip,
+				domain->host_data, handle_simple_irq,
+				NULL, NULL);
+	}
+	mutex_unlock(&msi->lock);
+	return 0;
+}
+
+static void nwl_irq_domain_free(struct irq_domain *domain, unsigned int virq,
+					unsigned int nr_irqs)
+{
+	struct irq_data *data = irq_domain_get_irq_data(domain, virq);
+	struct nwl_pcie *pcie = irq_data_get_irq_chip_data(data);
+	struct nwl_msi *msi = &pcie->msi;
+
+	mutex_lock(&msi->lock);
+	bitmap_clear(msi->bitmap, data->hwirq, nr_irqs);
+	mutex_unlock(&msi->lock);
+}
+
+static const struct irq_domain_ops dev_msi_domain_ops = {
+	.alloc  = nwl_irq_domain_alloc,
+	.free   = nwl_irq_domain_free,
+};
+
+static void nwl_msi_free_irq_domain(struct nwl_pcie *pcie)
+{
+	struct nwl_msi *msi = &pcie->msi;
+
+	if (msi->irq_msi0)
+		irq_set_chained_handler_and_data(msi->irq_msi0, NULL, NULL);
+	if (msi->irq_msi1)
+		irq_set_chained_handler_and_data(msi->irq_msi1, NULL, NULL);
+
+	if (msi->msi_domain)
+		irq_domain_remove(msi->msi_domain);
+	if (msi->dev_domain)
+		irq_domain_remove(msi->dev_domain);
+
+	kfree(msi->bitmap);
+	msi->bitmap = NULL;
+}
+
+static void nwl_pcie_free_irq_domain(struct nwl_pcie *pcie)
+{
+	int i;
+	u32 irq;
+
+	for (i = 0; i < INTX_NUM; i++) {
+		irq = irq_find_mapping(pcie->legacy_irq_domain, i + 1);
+		if (irq > 0)
+			irq_dispose_mapping(irq);
+	}
+	if (pcie->legacy_irq_domain)
+		irq_domain_remove(pcie->legacy_irq_domain);
+
+	nwl_msi_free_irq_domain(pcie);
+}
+
+static int nwl_pcie_init_msi_irq_domain(struct nwl_pcie *pcie)
+{
+#ifdef CONFIG_PCI_MSI
+	struct fwnode_handle *fwnode = of_node_to_fwnode(pcie->dev->of_node);
+	struct nwl_msi *msi = &pcie->msi;
+
+	msi->dev_domain = irq_domain_add_linear(NULL, INT_PCI_MSI_NR,
+						&dev_msi_domain_ops, pcie);
+	if (!msi->dev_domain) {
+		dev_err(pcie->dev, "failed to create dev IRQ domain\n");
+		return -ENOMEM;
+	}
+	msi->msi_domain = pci_msi_create_irq_domain(fwnode,
+						    &nwl_msi_domain_info,
+						    msi->dev_domain);
+	if (!msi->msi_domain) {
+		dev_err(pcie->dev, "failed to create msi IRQ domain\n");
+		irq_domain_remove(msi->dev_domain);
+		return -ENOMEM;
+	}
+#endif
+	return 0;
+}
+
+static int nwl_pcie_init_irq_domain(struct nwl_pcie *pcie)
+{
+	struct device_node *node = pcie->dev->of_node;
+	struct device_node *legacy_intc_node;
+
+	legacy_intc_node = of_get_next_child(node, NULL);
+	if (!legacy_intc_node) {
+		dev_err(pcie->dev, "No legacy intc node found\n");
+		return -EINVAL;
+	}
+
+	pcie->legacy_irq_domain = irq_domain_add_linear(legacy_intc_node,
+							INTX_NUM,
+							&legacy_domain_ops,
+							pcie);
+
+	if (!pcie->legacy_irq_domain) {
+		dev_err(pcie->dev, "failed to create IRQ domain\n");
+		return -ENOMEM;
+	}
+
+	nwl_pcie_init_msi_irq_domain(pcie);
+	return 0;
+}
+
+static int nwl_pcie_enable_msi(struct nwl_pcie *pcie, struct pci_bus *bus)
+{
+	struct platform_device *pdev = to_platform_device(pcie->dev);
+	struct nwl_msi *msi = &pcie->msi;
+	unsigned long base;
+	int ret;
+	int size = BITS_TO_LONGS(INT_PCI_MSI_NR) * sizeof(long);
+
+	mutex_init(&msi->lock);
+
+	msi->bitmap = kzalloc(size, GFP_KERNEL);
+	if (!msi->bitmap)
+		return -ENOMEM;
+
+	/* Get msi_1 IRQ number */
+	msi->irq_msi1 = platform_get_irq_byname(pdev, "msi1");
+	if (msi->irq_msi1 < 0) {
+		dev_err(&pdev->dev, "failed to get IRQ#%d\n", msi->irq_msi1);
+		ret = -EINVAL;
+		goto err;
+	}
+
+	irq_set_chained_handler_and_data(msi->irq_msi1,
+					 nwl_pcie_msi_handler_high, pcie);
+
+	/* Get msi_0 IRQ number */
+	msi->irq_msi0 = platform_get_irq_byname(pdev, "msi0");
+	if (msi->irq_msi0 < 0) {
+		dev_err(&pdev->dev, "failed to get IRQ#%d\n", msi->irq_msi0);
+		ret = -EINVAL;
+		goto err;
+	}
+
+	irq_set_chained_handler_and_data(msi->irq_msi0,
+					 nwl_pcie_msi_handler_low, pcie);
+
+	/* Check for msii_present bit */
+	ret = nwl_bridge_readl(pcie, I_MSII_CAPABILITIES) & MSII_PRESENT;
+	if (!ret) {
+		dev_err(pcie->dev, "MSI not present\n");
+		ret = -EIO;
+		goto err;
+	}
+
+	/* Enable MSII */
+	nwl_bridge_writel(pcie, nwl_bridge_readl(pcie, I_MSII_CONTROL) |
+			  MSII_ENABLE, I_MSII_CONTROL);
+
+	/* Enable MSII status */
+	nwl_bridge_writel(pcie, nwl_bridge_readl(pcie, I_MSII_CONTROL) |
+			  MSII_STATUS_ENABLE, I_MSII_CONTROL);
+
+	/* setup AFI/FPCI range */
+	base = pcie->phys_pcie_reg_base;
+	nwl_bridge_writel(pcie, lower_32_bits(base), I_MSII_BASE_LO);
+	nwl_bridge_writel(pcie, upper_32_bits(base), I_MSII_BASE_HI);
+
+	/*
+	 * For high range MSI interrupts: disable, clear any pending,
+	 * and enable
+	 */
+	nwl_bridge_writel(pcie, (u32)~MSGF_MSI_SR_HI_MASK, MSGF_MSI_MASK_HI);
+
+	nwl_bridge_writel(pcie, nwl_bridge_readl(pcie,  MSGF_MSI_STATUS_HI) &
+			  MSGF_MSI_SR_HI_MASK, MSGF_MSI_STATUS_HI);
+
+	nwl_bridge_writel(pcie, MSGF_MSI_SR_HI_MASK, MSGF_MSI_MASK_HI);
+
+	/*
+	 * For low range MSI interrupts: disable, clear any pending,
+	 * and enable
+	 */
+	nwl_bridge_writel(pcie, (u32)~MSGF_MSI_SR_LO_MASK, MSGF_MSI_MASK_LO);
+
+	nwl_bridge_writel(pcie, nwl_bridge_readl(pcie, MSGF_MSI_STATUS_LO) &
+			  MSGF_MSI_SR_LO_MASK, MSGF_MSI_STATUS_LO);
+
+	nwl_bridge_writel(pcie, MSGF_MSI_SR_LO_MASK, MSGF_MSI_MASK_LO);
+
+	return 0;
+err:
+	kfree(msi->bitmap);
+	msi->bitmap = NULL;
+	return ret;
+}
+
+static int nwl_pcie_bridge_init(struct nwl_pcie *pcie)
+{
+	struct platform_device *pdev = to_platform_device(pcie->dev);
+	u32 breg_val, ecam_val, first_busno = 0;
+	int err;
+
+	breg_val = nwl_bridge_readl(pcie, E_BREG_CAPABILITIES) & BREG_PRESENT;
+	if (!breg_val) {
+		dev_err(pcie->dev, "BREG is not present\n");
+		return breg_val;
+	}
+
+	/* Write bridge_off to breg base */
+	nwl_bridge_writel(pcie, lower_32_bits(pcie->phys_breg_base),
+			  E_BREG_BASE_LO);
+	nwl_bridge_writel(pcie, upper_32_bits(pcie->phys_breg_base),
+			  E_BREG_BASE_HI);
+
+	/* Enable BREG */
+	nwl_bridge_writel(pcie, ~BREG_ENABLE_FORCE & BREG_ENABLE,
+			  E_BREG_CONTROL);
+
+	/* Disable DMA channel registers */
+	nwl_bridge_writel(pcie, nwl_bridge_readl(pcie, BRCFG_PCIE_RX0) |
+			  CFG_DMA_REG_BAR, BRCFG_PCIE_RX0);
+
+	/* Enable Ingress subtractive decode translation */
+	nwl_bridge_writel(pcie, SET_ISUB_CONTROL, I_ISUB_CONTROL);
+
+	/* Enable msg filtering details */
+	nwl_bridge_writel(pcie, CFG_ENABLE_MSG_FILTER_MASK,
+			  BRCFG_PCIE_RX_MSG_FILTER);
+
+	err = nwl_wait_for_link(pcie);
+	if (err)
+		return err;
+
+	ecam_val = nwl_bridge_readl(pcie, E_ECAM_CAPABILITIES) & E_ECAM_PRESENT;
+	if (!ecam_val) {
+		dev_err(pcie->dev, "ECAM is not present\n");
+		return ecam_val;
+	}
+
+	/* Enable ECAM */
+	nwl_bridge_writel(pcie, nwl_bridge_readl(pcie, E_ECAM_CONTROL) |
+			  E_ECAM_CR_ENABLE, E_ECAM_CONTROL);
+
+	nwl_bridge_writel(pcie, nwl_bridge_readl(pcie, E_ECAM_CONTROL) |
+			  (pcie->ecam_value << E_ECAM_SIZE_SHIFT),
+			  E_ECAM_CONTROL);
+
+	nwl_bridge_writel(pcie, lower_32_bits(pcie->phys_ecam_base),
+			  E_ECAM_BASE_LO);
+	nwl_bridge_writel(pcie, upper_32_bits(pcie->phys_ecam_base),
+			  E_ECAM_BASE_HI);
+
+	/* Get bus range */
+	ecam_val = nwl_bridge_readl(pcie, E_ECAM_CONTROL);
+	pcie->last_busno = (ecam_val & E_ECAM_SIZE_LOC) >> E_ECAM_SIZE_SHIFT;
+	/* Write primary, secondary and subordinate bus numbers */
+	ecam_val = first_busno;
+	ecam_val |= (first_busno + 1) << 8;
+	ecam_val |= (pcie->last_busno << E_ECAM_SIZE_SHIFT);
+	writel(ecam_val, (pcie->ecam_base + PCI_PRIMARY_BUS));
+
+	if (nwl_pcie_link_up(pcie))
+		dev_info(pcie->dev, "Link is UP\n");
+	else
+		dev_info(pcie->dev, "Link is DOWN\n");
+
+	/* Get misc IRQ number */
+	pcie->irq_misc = platform_get_irq_byname(pdev, "misc");
+	if (pcie->irq_misc < 0) {
+		dev_err(&pdev->dev, "failed to get misc IRQ %d\n",
+			pcie->irq_misc);
+		return -EINVAL;
+	}
+
+	err = devm_request_irq(pcie->dev, pcie->irq_misc,
+			       nwl_pcie_misc_handler, IRQF_SHARED,
+			       "nwl_pcie:misc", pcie);
+	if (err) {
+		dev_err(pcie->dev, "fail to register misc IRQ#%d\n",
+			pcie->irq_misc);
+		return err;
+	}
+
+	/* Disable all misc interrupts */
+	nwl_bridge_writel(pcie, (u32)~MSGF_MISC_SR_MASKALL, MSGF_MISC_MASK);
+
+	/* Clear pending misc interrupts */
+	nwl_bridge_writel(pcie, nwl_bridge_readl(pcie, MSGF_MISC_STATUS) &
+			  MSGF_MISC_SR_MASKALL, MSGF_MISC_STATUS);
+
+	/* Enable all misc interrupts */
+	nwl_bridge_writel(pcie, MSGF_MISC_SR_MASKALL, MSGF_MISC_MASK);
+
+
+	/* Disable all legacy interrupts */
+	nwl_bridge_writel(pcie, (u32)~MSGF_LEG_SR_MASKALL, MSGF_LEG_MASK);
+
+	/* Clear pending legacy interrupts */
+	nwl_bridge_writel(pcie, nwl_bridge_readl(pcie, MSGF_LEG_STATUS) &
+			  MSGF_LEG_SR_MASKALL, MSGF_LEG_STATUS);
+
+	/* Enable all legacy interrupts */
+	nwl_bridge_writel(pcie, MSGF_LEG_SR_MASKALL, MSGF_LEG_MASK);
+
+	/* Enable the bridge config interrupt */
+	nwl_bridge_writel(pcie, nwl_bridge_readl(pcie, BRCFG_INTERRUPT) |
+			  BRCFG_INTERRUPT_MASK, BRCFG_INTERRUPT);
+
+	return 0;
+}
+
+static int nwl_pcie_parse_dt(struct nwl_pcie *pcie,
+			     struct platform_device *pdev)
+{
+	struct device_node *node = pcie->dev->of_node;
+	struct resource *res;
+	const char *type;
+
+	/* Check for device type */
+	type = of_get_property(node, "device_type", NULL);
+	if (!type || strcmp(type, "pci")) {
+		dev_err(pcie->dev, "invalid \"device_type\" %s\n", type);
+		return -EINVAL;
+	}
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "breg");
+	pcie->breg_base = devm_ioremap_resource(pcie->dev, res);
+	if (IS_ERR(pcie->breg_base))
+		return PTR_ERR(pcie->breg_base);
+	pcie->phys_breg_base = res->start;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "pcireg");
+	pcie->pcireg_base = devm_ioremap_resource(pcie->dev, res);
+	if (IS_ERR(pcie->pcireg_base))
+		return PTR_ERR(pcie->pcireg_base);
+	pcie->phys_pcie_reg_base = res->start;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "cfg");
+	pcie->ecam_base = devm_ioremap_resource(pcie->dev, res);
+	if (IS_ERR(pcie->ecam_base))
+		return PTR_ERR(pcie->ecam_base);
+	pcie->phys_ecam_base = res->start;
+
+	/* Get intx IRQ number */
+	pcie->irq_intx = platform_get_irq_byname(pdev, "intx");
+	if (pcie->irq_intx < 0) {
+		dev_err(&pdev->dev, "failed to get intx IRQ %d\n",
+			pcie->irq_intx);
+		return -EINVAL;
+	}
+
+	irq_set_chained_handler_and_data(pcie->irq_intx,
+					 nwl_pcie_leg_handler, pcie);
+
+	return 0;
+}
+
+static const struct of_device_id nwl_pcie_of_match[] = {
+	{ .compatible = "xlnx,nwl-pcie-2.11", },
+	{}
+};
+
+static int nwl_pcie_probe(struct platform_device *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	struct nwl_pcie *pcie;
+	struct pci_bus *bus;
+	struct pci_bus *child;
+	int err;
+	resource_size_t iobase = 0;
+	LIST_HEAD(res);
+
+	pcie = devm_kzalloc(&pdev->dev, sizeof(*pcie), GFP_KERNEL);
+	if (!pcie)
+		return -ENOMEM;
+
+	pcie->dev = &pdev->dev;
+	pcie->ecam_value = NWL_ECAM_VALUE_DEFAULT;
+
+	err = nwl_pcie_parse_dt(pcie, pdev);
+	if (err) {
+		dev_err(pcie->dev, "Parsing DT failed\n");
+		return err;
+	}
+
+	err = nwl_pcie_bridge_init(pcie);
+	if (err) {
+		dev_err(pcie->dev, "HW Initalization failed\n");
+		return err;
+	}
+
+	err = of_pci_get_host_bridge_resources(node, 0, 0xff, &res, &iobase);
+	if (err) {
+		pr_err("Getting bridge resources failed\n");
+		return err;
+	}
+
+	err = nwl_pcie_init_irq_domain(pcie);
+	if (err) {
+		dev_err(pcie->dev, "Failed creating IRQ Domain\n");
+		return err;
+	}
+
+	bus = pci_create_root_bus(&pdev->dev, pcie->root_busno,
+				  &nwl_pcie_ops, pcie, &res);
+	if (!bus)
+		return -ENOMEM;
+
+	if (IS_ENABLED(CONFIG_PCI_MSI)) {
+		err = nwl_pcie_enable_msi(pcie, bus);
+		if (err < 0) {
+			dev_err(&pdev->dev,
+				"failed to enable MSI support: %d\n", err);
+			return err;
+		}
+	}
+	pci_scan_child_bus(bus);
+	pci_assign_unassigned_bus_resources(bus);
+	list_for_each_entry(child, &bus->children, node)
+		pcie_bus_configure_settings(child);
+	pci_bus_add_devices(bus);
+	platform_set_drvdata(pdev, pcie);
+	return 0;
+}
+
+static int nwl_pcie_remove(struct platform_device *pdev)
+{
+	struct nwl_pcie *pcie = platform_get_drvdata(pdev);
+
+	nwl_pcie_free_irq_domain(pcie);
+	platform_set_drvdata(pdev, NULL);
+	return 0;
+}
+
+static struct platform_driver nwl_pcie_driver = {
+	.driver = {
+		.name = "nwl-pcie",
+		.of_match_table = nwl_pcie_of_match,
+	},
+	.probe = nwl_pcie_probe,
+	.remove = nwl_pcie_remove,
+};
+module_platform_driver(nwl_pcie_driver);
+
+MODULE_AUTHOR("Xilinx, Inc");
+MODULE_DESCRIPTION("NWL PCIe driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/pci/host/pcie-xilinx.c b/drivers/pci/host/pcie-xilinx.c
index 4cfa46360d12..65f0fe0c2eaf 100644
--- a/drivers/pci/host/pcie-xilinx.c
+++ b/drivers/pci/host/pcie-xilinx.c
@@ -94,9 +94,6 @@
 /* Number of MSI IRQs */
 #define XILINX_NUM_MSI_IRQS		128
 
-/* Number of Memory Resources */
-#define XILINX_MAX_NUM_RESOURCES	3
-
 /**
  * struct xilinx_pcie_port - PCIe port information
  * @reg_base: IO Mapped Register Base
@@ -105,7 +102,6 @@
  * @root_busno: Root Bus number
  * @dev: Device pointer
  * @irq_domain: IRQ domain pointer
- * @bus_range: Bus range
  * @resources: Bus Resources
  */
 struct xilinx_pcie_port {
@@ -115,17 +111,11 @@ struct xilinx_pcie_port {
 	u8 root_busno;
 	struct device *dev;
 	struct irq_domain *irq_domain;
-	struct resource bus_range;
 	struct list_head resources;
 };
 
 static DECLARE_BITMAP(msi_irq_in_use, XILINX_NUM_MSI_IRQS);
 
-static inline struct xilinx_pcie_port *sys_to_pcie(struct pci_sys_data *sys)
-{
-	return sys->private_data;
-}
-
 static inline u32 pcie_read(struct xilinx_pcie_port *port, u32 reg)
 {
 	return readl(port->reg_base + reg);
@@ -167,7 +157,7 @@ static void xilinx_pcie_clear_err_interrupts(struct xilinx_pcie_port *port)
  */
 static bool xilinx_pcie_valid_device(struct pci_bus *bus, unsigned int devfn)
 {
-	struct xilinx_pcie_port *port = sys_to_pcie(bus->sysdata);
+	struct xilinx_pcie_port *port = bus->sysdata;
 
 	/* Check if link is up when trying to access downstream ports */
 	if (bus->number != port->root_busno)
@@ -200,7 +190,7 @@ static bool xilinx_pcie_valid_device(struct pci_bus *bus, unsigned int devfn)
 static void __iomem *xilinx_pcie_map_bus(struct pci_bus *bus,
 					 unsigned int devfn, int where)
 {
-	struct xilinx_pcie_port *port = sys_to_pcie(bus->sysdata);
+	struct xilinx_pcie_port *port = bus->sysdata;
 	int relbus;
 
 	if (!xilinx_pcie_valid_device(bus, devfn))
@@ -232,7 +222,7 @@ static void xilinx_pcie_destroy_msi(unsigned int irq)
 
 	if (!test_bit(irq, msi_irq_in_use)) {
 		msi = irq_get_msi_desc(irq);
-		port = sys_to_pcie(msi_desc_to_pci_sysdata(msi));
+		port = msi_desc_to_pci_sysdata(msi);
 		dev_err(port->dev, "Trying to free unused MSI#%d\n", irq);
 	} else {
 		clear_bit(irq, msi_irq_in_use);
@@ -281,7 +271,7 @@ static int xilinx_pcie_msi_setup_irq(struct msi_controller *chip,
 				     struct pci_dev *pdev,
 				     struct msi_desc *desc)
 {
-	struct xilinx_pcie_port *port = sys_to_pcie(pdev->bus->sysdata);
+	struct xilinx_pcie_port *port = pdev->bus->sysdata;
 	unsigned int irq;
 	int hwirq;
 	struct msi_msg msg;
@@ -618,138 +608,6 @@ static void xilinx_pcie_init_port(struct xilinx_pcie_port *port)
 }
 
 /**
- * xilinx_pcie_setup - Setup memory resources
- * @nr: Bus number
- * @sys: Per controller structure
- *
- * Return: '1' on success and error value on failure
- */
-static int xilinx_pcie_setup(int nr, struct pci_sys_data *sys)
-{
-	struct xilinx_pcie_port *port = sys_to_pcie(sys);
-
-	list_splice_init(&port->resources, &sys->resources);
-
-	return 1;
-}
-
-/**
- * xilinx_pcie_scan_bus - Scan PCIe bus for devices
- * @nr: Bus number
- * @sys: Per controller structure
- *
- * Return: Valid Bus pointer on success and NULL on failure
- */
-static struct pci_bus *xilinx_pcie_scan_bus(int nr, struct pci_sys_data *sys)
-{
-	struct xilinx_pcie_port *port = sys_to_pcie(sys);
-	struct pci_bus *bus;
-
-	port->root_busno = sys->busnr;
-
-	if (IS_ENABLED(CONFIG_PCI_MSI))
-		bus = pci_scan_root_bus_msi(port->dev, sys->busnr,
-					    &xilinx_pcie_ops, sys,
-					    &sys->resources,
-					    &xilinx_pcie_msi_chip);
-	else
-		bus = pci_scan_root_bus(port->dev, sys->busnr,
-					&xilinx_pcie_ops, sys, &sys->resources);
-	return bus;
-}
-
-/**
- * xilinx_pcie_parse_and_add_res - Add resources by parsing ranges
- * @port: PCIe port information
- *
- * Return: '0' on success and error value on failure
- */
-static int xilinx_pcie_parse_and_add_res(struct xilinx_pcie_port *port)
-{
-	struct device *dev = port->dev;
-	struct device_node *node = dev->of_node;
-	struct resource *mem;
-	resource_size_t offset;
-	struct of_pci_range_parser parser;
-	struct of_pci_range range;
-	struct resource_entry *win;
-	int err = 0, mem_resno = 0;
-
-	/* Get the ranges */
-	if (of_pci_range_parser_init(&parser, node)) {
-		dev_err(dev, "missing \"ranges\" property\n");
-		return -EINVAL;
-	}
-
-	/* Parse the ranges and add the resources found to the list */
-	for_each_of_pci_range(&parser, &range) {
-
-		if (mem_resno >= XILINX_MAX_NUM_RESOURCES) {
-			dev_err(dev, "Maximum memory resources exceeded\n");
-			return -EINVAL;
-		}
-
-		mem = devm_kmalloc(dev, sizeof(*mem), GFP_KERNEL);
-		if (!mem) {
-			err = -ENOMEM;
-			goto free_resources;
-		}
-
-		of_pci_range_to_resource(&range, node, mem);
-
-		switch (mem->flags & IORESOURCE_TYPE_BITS) {
-		case IORESOURCE_MEM:
-			offset = range.cpu_addr - range.pci_addr;
-			mem_resno++;
-			break;
-		default:
-			err = -EINVAL;
-			break;
-		}
-
-		if (err < 0) {
-			dev_warn(dev, "Invalid resource found %pR\n", mem);
-			continue;
-		}
-
-		err = request_resource(&iomem_resource, mem);
-		if (err)
-			goto free_resources;
-
-		pci_add_resource_offset(&port->resources, mem, offset);
-	}
-
-	/* Get the bus range */
-	if (of_pci_parse_bus_range(node, &port->bus_range)) {
-		u32 val = pcie_read(port, XILINX_PCIE_REG_BIR);
-		u8 last;
-
-		last = (val & XILINX_PCIE_BIR_ECAM_SZ_MASK) >>
-			XILINX_PCIE_BIR_ECAM_SZ_SHIFT;
-
-		port->bus_range = (struct resource) {
-			.name	= node->name,
-			.start	= 0,
-			.end	= last,
-			.flags	= IORESOURCE_BUS,
-		};
-	}
-
-	/* Register bus resource */
-	pci_add_resource(&port->resources, &port->bus_range);
-
-	return 0;
-
-free_resources:
-	release_child_resources(&iomem_resource);
-	resource_list_for_each_entry(win, &port->resources)
-		devm_kfree(dev, win->res);
-	pci_free_resource_list(&port->resources);
-
-	return err;
-}
-
-/**
  * xilinx_pcie_parse_dt - Parse Device tree
  * @port: PCIe port information
  *
@@ -800,9 +658,12 @@ static int xilinx_pcie_parse_dt(struct xilinx_pcie_port *port)
 static int xilinx_pcie_probe(struct platform_device *pdev)
 {
 	struct xilinx_pcie_port *port;
-	struct hw_pci hw;
 	struct device *dev = &pdev->dev;
+	struct pci_bus *bus;
+
 	int err;
+	resource_size_t iobase = 0;
+	LIST_HEAD(res);
 
 	if (!dev->of_node)
 		return -ENODEV;
@@ -827,34 +688,28 @@ static int xilinx_pcie_probe(struct platform_device *pdev)
 		return err;
 	}
 
-	/*
-	 * Parse PCI ranges, configuration bus range and
-	 * request their resources
-	 */
-	INIT_LIST_HEAD(&port->resources);
-	err = xilinx_pcie_parse_and_add_res(port);
+	err = of_pci_get_host_bridge_resources(dev->of_node, 0, 0xff, &res,
+					       &iobase);
 	if (err) {
-		dev_err(dev, "Failed adding resources\n");
+		dev_err(dev, "Getting bridge resources failed\n");
 		return err;
 	}
-
-	platform_set_drvdata(pdev, port);
-
-	/* Register the device */
-	memset(&hw, 0, sizeof(hw));
-	hw = (struct hw_pci) {
-		.nr_controllers	= 1,
-		.private_data	= (void **)&port,
-		.setup		= xilinx_pcie_setup,
-		.map_irq	= of_irq_parse_and_map_pci,
-		.scan		= xilinx_pcie_scan_bus,
-		.ops		= &xilinx_pcie_ops,
-	};
+	bus = pci_create_root_bus(&pdev->dev, 0,
+				  &xilinx_pcie_ops, port, &res);
+	if (!bus)
+		return -ENOMEM;
 
 #ifdef CONFIG_PCI_MSI
 	xilinx_pcie_msi_chip.dev = port->dev;
+	bus->msi = &xilinx_pcie_msi_chip;
 #endif
-	pci_common_init_dev(dev, &hw);
+	pci_scan_child_bus(bus);
+	pci_assign_unassigned_bus_resources(bus);
+#ifndef CONFIG_MICROBLAZE
+	pci_fixup_irqs(pci_common_swizzle, of_irq_parse_and_map_pci);
+#endif
+	pci_bus_add_devices(bus);
+	platform_set_drvdata(pdev, port);
 
 	return 0;
 }
diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c
index 31f31d460fc9..fe4bd0aa91a6 100644
--- a/drivers/pci/iov.c
+++ b/drivers/pci/iov.c
@@ -387,10 +387,6 @@ static int sriov_init(struct pci_dev *dev, int pos)
 	struct resource *res;
 	struct pci_dev *pdev;
 
-	if (pci_pcie_type(dev) != PCI_EXP_TYPE_RC_END &&
-	    pci_pcie_type(dev) != PCI_EXP_TYPE_ENDPOINT)
-		return -ENODEV;
-
 	pci_read_config_word(dev, pos + PCI_SRIOV_CTRL, &ctrl);
 	if (ctrl & PCI_SRIOV_CTRL_VFE) {
 		pci_write_config_word(dev, pos + PCI_SRIOV_CTRL, 0);
diff --git a/drivers/pci/pci-label.c b/drivers/pci/pci-label.c
index 0ae74d96ed85..51357377efbc 100644
--- a/drivers/pci/pci-label.c
+++ b/drivers/pci/pci-label.c
@@ -16,7 +16,7 @@
  * the instance number and string from the type 41 record and exports
  * it to sysfs.
  *
- * Please see http://linux.dell.com/wiki/index.php/Oss/libnetdevname for more
+ * Please see http://linux.dell.com/files/biosdevname/ for more
  * information.
  */
 
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 95d9e7bd933b..e982010f0ed1 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -769,10 +769,12 @@ static ssize_t read_vpd_attr(struct file *filp, struct kobject *kobj,
 {
 	struct pci_dev *dev = to_pci_dev(kobj_to_dev(kobj));
 
-	if (off > bin_attr->size)
-		count = 0;
-	else if (count > bin_attr->size - off)
-		count = bin_attr->size - off;
+	if (bin_attr->size > 0) {
+		if (off > bin_attr->size)
+			count = 0;
+		else if (count > bin_attr->size - off)
+			count = bin_attr->size - off;
+	}
 
 	return pci_read_vpd(dev, off, count, buf);
 }
@@ -783,10 +785,12 @@ static ssize_t write_vpd_attr(struct file *filp, struct kobject *kobj,
 {
 	struct pci_dev *dev = to_pci_dev(kobj_to_dev(kobj));
 
-	if (off > bin_attr->size)
-		count = 0;
-	else if (count > bin_attr->size - off)
-		count = bin_attr->size - off;
+	if (bin_attr->size > 0) {
+		if (off > bin_attr->size)
+			count = 0;
+		else if (count > bin_attr->size - off)
+			count = bin_attr->size - off;
+	}
 
 	return pci_write_vpd(dev, off, count, buf);
 }
@@ -1134,33 +1138,36 @@ static int pci_create_attr(struct pci_dev *pdev, int num, int write_combine)
 	/* allocate attribute structure, piggyback attribute name */
 	int name_len = write_combine ? 13 : 10;
 	struct bin_attribute *res_attr;
+	char *res_attr_name;
 	int retval;
 
 	res_attr = kzalloc(sizeof(*res_attr) + name_len, GFP_ATOMIC);
-	if (res_attr) {
-		char *res_attr_name = (char *)(res_attr + 1);
-
-		sysfs_bin_attr_init(res_attr);
-		if (write_combine) {
-			pdev->res_attr_wc[num] = res_attr;
-			sprintf(res_attr_name, "resource%d_wc", num);
-			res_attr->mmap = pci_mmap_resource_wc;
-		} else {
-			pdev->res_attr[num] = res_attr;
-			sprintf(res_attr_name, "resource%d", num);
-			res_attr->mmap = pci_mmap_resource_uc;
-		}
-		if (pci_resource_flags(pdev, num) & IORESOURCE_IO) {
-			res_attr->read = pci_read_resource_io;
-			res_attr->write = pci_write_resource_io;
-		}
-		res_attr->attr.name = res_attr_name;
-		res_attr->attr.mode = S_IRUSR | S_IWUSR;
-		res_attr->size = pci_resource_len(pdev, num);
-		res_attr->private = &pdev->resource[num];
-		retval = sysfs_create_bin_file(&pdev->dev.kobj, res_attr);
-	} else
-		retval = -ENOMEM;
+	if (!res_attr)
+		return -ENOMEM;
+
+	res_attr_name = (char *)(res_attr + 1);
+
+	sysfs_bin_attr_init(res_attr);
+	if (write_combine) {
+		pdev->res_attr_wc[num] = res_attr;
+		sprintf(res_attr_name, "resource%d_wc", num);
+		res_attr->mmap = pci_mmap_resource_wc;
+	} else {
+		pdev->res_attr[num] = res_attr;
+		sprintf(res_attr_name, "resource%d", num);
+		res_attr->mmap = pci_mmap_resource_uc;
+	}
+	if (pci_resource_flags(pdev, num) & IORESOURCE_IO) {
+		res_attr->read = pci_read_resource_io;
+		res_attr->write = pci_write_resource_io;
+	}
+	res_attr->attr.name = res_attr_name;
+	res_attr->attr.mode = S_IRUSR | S_IWUSR;
+	res_attr->size = pci_resource_len(pdev, num);
+	res_attr->private = &pdev->resource[num];
+	retval = sysfs_create_bin_file(&pdev->dev.kobj, res_attr);
+	if (retval)
+		kfree(res_attr);
 
 	return retval;
 }
@@ -1319,7 +1326,7 @@ static int pci_create_capabilities_sysfs(struct pci_dev *dev)
 			return -ENOMEM;
 
 		sysfs_bin_attr_init(attr);
-		attr->size = dev->vpd->len;
+		attr->size = 0;
 		attr->attr.name = "vpd";
 		attr->attr.mode = S_IRUSR | S_IWUSR;
 		attr->read = read_vpd_attr;
@@ -1356,7 +1363,7 @@ error:
 int __must_check pci_create_sysfs_dev_files(struct pci_dev *pdev)
 {
 	int retval;
-	int rom_size = 0;
+	int rom_size;
 	struct bin_attribute *attr;
 
 	if (!sysfs_initialized)
@@ -1373,12 +1380,8 @@ int __must_check pci_create_sysfs_dev_files(struct pci_dev *pdev)
 	if (retval)
 		goto err_config_file;
 
-	if (pci_resource_len(pdev, PCI_ROM_RESOURCE))
-		rom_size = pci_resource_len(pdev, PCI_ROM_RESOURCE);
-	else if (pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW)
-		rom_size = 0x20000;
-
 	/* If the device has a ROM, try to expose it in sysfs. */
+	rom_size = pci_resource_len(pdev, PCI_ROM_RESOURCE);
 	if (rom_size) {
 		attr = kzalloc(sizeof(*attr), GFP_ATOMIC);
 		if (!attr) {
@@ -1409,7 +1412,7 @@ int __must_check pci_create_sysfs_dev_files(struct pci_dev *pdev)
 	return 0;
 
 err_rom_file:
-	if (rom_size) {
+	if (pdev->rom_attr) {
 		sysfs_remove_bin_file(&pdev->dev.kobj, pdev->rom_attr);
 		kfree(pdev->rom_attr);
 		pdev->rom_attr = NULL;
@@ -1447,8 +1450,6 @@ static void pci_remove_capabilities_sysfs(struct pci_dev *dev)
  */
 void pci_remove_sysfs_dev_files(struct pci_dev *pdev)
 {
-	int rom_size = 0;
-
 	if (!sysfs_initialized)
 		return;
 
@@ -1461,18 +1462,13 @@ void pci_remove_sysfs_dev_files(struct pci_dev *pdev)
 
 	pci_remove_resource_files(pdev);
 
-	if (pci_resource_len(pdev, PCI_ROM_RESOURCE))
-		rom_size = pci_resource_len(pdev, PCI_ROM_RESOURCE);
-	else if (pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW)
-		rom_size = 0x20000;
-
-	if (rom_size && pdev->rom_attr) {
+	if (pdev->rom_attr) {
 		sysfs_remove_bin_file(&pdev->dev.kobj, pdev->rom_attr);
 		kfree(pdev->rom_attr);
+		pdev->rom_attr = NULL;
 	}
 
 	pci_remove_firmware_label_files(pdev);
-
 }
 
 static int __init pci_sysfs_init(void)
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index f89db3af0607..25e0327d4429 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -25,7 +25,6 @@
 #include <linux/device.h>
 #include <linux/pm_runtime.h>
 #include <linux/pci_hotplug.h>
-#include <asm-generic/pci-bridge.h>
 #include <asm/setup.h>
 #include <linux/aer.h>
 #include "pci.h"
@@ -3386,18 +3385,6 @@ bool pci_check_and_unmask_intx(struct pci_dev *dev)
 }
 EXPORT_SYMBOL_GPL(pci_check_and_unmask_intx);
 
-int pci_set_dma_max_seg_size(struct pci_dev *dev, unsigned int size)
-{
-	return dma_set_max_seg_size(&dev->dev, size);
-}
-EXPORT_SYMBOL(pci_set_dma_max_seg_size);
-
-int pci_set_dma_seg_boundary(struct pci_dev *dev, unsigned long mask)
-{
-	return dma_set_seg_boundary(&dev->dev, mask);
-}
-EXPORT_SYMBOL(pci_set_dma_seg_boundary);
-
 /**
  * pci_wait_for_pending_transaction - waits for pending transaction
  * @dev: the PCI device to operate on
@@ -3414,6 +3401,29 @@ int pci_wait_for_pending_transaction(struct pci_dev *dev)
 }
 EXPORT_SYMBOL(pci_wait_for_pending_transaction);
 
+/*
+ * We should only need to wait 100ms after FLR, but some devices take longer.
+ * Wait for up to 1000ms for config space to return something other than -1.
+ * Intel IGD requires this when an LCD panel is attached.  We read the 2nd
+ * dword because VFs don't implement the 1st dword.
+ */
+static void pci_flr_wait(struct pci_dev *dev)
+{
+	int i = 0;
+	u32 id;
+
+	do {
+		msleep(100);
+		pci_read_config_dword(dev, PCI_COMMAND, &id);
+	} while (i++ < 10 && id == ~0);
+
+	if (id == ~0)
+		dev_warn(&dev->dev, "Failed to return from FLR\n");
+	else if (i > 1)
+		dev_info(&dev->dev, "Required additional %dms to return from FLR\n",
+			 (i - 1) * 100);
+}
+
 static int pcie_flr(struct pci_dev *dev, int probe)
 {
 	u32 cap;
@@ -3429,7 +3439,7 @@ static int pcie_flr(struct pci_dev *dev, int probe)
 		dev_err(&dev->dev, "timed out waiting for pending transaction; performing function level reset anyway\n");
 
 	pcie_capability_set_word(dev, PCI_EXP_DEVCTL, PCI_EXP_DEVCTL_BCR_FLR);
-	msleep(100);
+	pci_flr_wait(dev);
 	return 0;
 }
 
@@ -3459,7 +3469,7 @@ static int pci_af_flr(struct pci_dev *dev, int probe)
 		dev_err(&dev->dev, "timed out waiting for pending transaction; performing AF function level reset anyway\n");
 
 	pci_write_config_byte(dev, pos + PCI_AF_CTRL, PCI_AF_CTRL_FLR);
-	msleep(100);
+	pci_flr_wait(dev);
 	return 0;
 }
 
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 9a1660f592ef..d0fb93481573 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -97,21 +97,21 @@ static inline bool pci_has_subordinate(struct pci_dev *pci_dev)
 struct pci_vpd_ops {
 	ssize_t (*read)(struct pci_dev *dev, loff_t pos, size_t count, void *buf);
 	ssize_t (*write)(struct pci_dev *dev, loff_t pos, size_t count, const void *buf);
-	void (*release)(struct pci_dev *dev);
 };
 
 struct pci_vpd {
-	unsigned int len;
 	const struct pci_vpd_ops *ops;
 	struct bin_attribute *attr; /* descriptor for sysfs VPD entry */
+	struct mutex	lock;
+	unsigned int	len;
+	u16		flag;
+	u8		cap;
+	u8		busy:1;
+	u8		valid:1;
 };
 
-int pci_vpd_pci22_init(struct pci_dev *dev);
-static inline void pci_vpd_release(struct pci_dev *dev)
-{
-	if (dev->vpd)
-		dev->vpd->ops->release(dev);
-}
+int pci_vpd_init(struct pci_dev *dev);
+void pci_vpd_release(struct pci_dev *dev);
 
 /* PCI /proc functions */
 #ifdef CONFIG_PROC_FS
diff --git a/drivers/pci/pcie/Kconfig b/drivers/pci/pcie/Kconfig
index e294713c8143..72db7f4209ca 100644
--- a/drivers/pci/pcie/Kconfig
+++ b/drivers/pci/pcie/Kconfig
@@ -44,6 +44,7 @@ config PCIEASPM
 	  /sys/module/pcie_aspm/parameters/policy
 
 	  When in doubt, say Y.
+
 config PCIEASPM_DEBUG
 	bool "Debug PCI Express ASPM"
 	depends on PCIEASPM
@@ -58,20 +59,20 @@ choice
 	depends on PCIEASPM
 
 config PCIEASPM_DEFAULT
-        bool "BIOS default"
+	bool "BIOS default"
 	depends on PCIEASPM
 	help
 	  Use the BIOS defaults for PCI Express ASPM.
 
 config PCIEASPM_POWERSAVE
-        bool "Powersave"
+	bool "Powersave"
 	depends on PCIEASPM
 	help
 	  Enable PCI Express ASPM L0s and L1 where possible, even if the
 	  BIOS did not.
 
 config PCIEASPM_PERFORMANCE
-        bool "Performance"
+	bool "Performance"
 	depends on PCIEASPM
 	help
 	  Disable PCI Express ASPM L0s and L1, even if the BIOS enabled them.
diff --git a/drivers/pci/pcie/aer/aer_inject.c b/drivers/pci/pcie/aer/aer_inject.c
index 20db790465dd..db553dc22c8e 100644
--- a/drivers/pci/pcie/aer/aer_inject.c
+++ b/drivers/pci/pcie/aer/aer_inject.c
@@ -25,6 +25,7 @@
 #include <linux/fs.h>
 #include <linux/uaccess.h>
 #include <linux/stddef.h>
+#include <linux/device.h>
 #include "aerdrv.h"
 
 /* Override the existing corrected and uncorrected error masks */
@@ -124,16 +125,13 @@ static struct pci_ops *__find_pci_bus_ops(struct pci_bus *bus)
 static struct pci_bus_ops *pci_bus_ops_pop(void)
 {
 	unsigned long flags;
-	struct pci_bus_ops *bus_ops = NULL;
+	struct pci_bus_ops *bus_ops;
 
 	spin_lock_irqsave(&inject_lock, flags);
-	if (list_empty(&pci_bus_ops_list))
-		bus_ops = NULL;
-	else {
-		struct list_head *lh = pci_bus_ops_list.next;
-		list_del(lh);
-		bus_ops = list_entry(lh, struct pci_bus_ops, list);
-	}
+	bus_ops = list_first_entry_or_null(&pci_bus_ops_list,
+					   struct pci_bus_ops, list);
+	if (bus_ops)
+		list_del(&bus_ops->list);
 	spin_unlock_irqrestore(&inject_lock, flags);
 	return bus_ops;
 }
@@ -181,14 +179,16 @@ static u32 *find_pci_config_dword(struct aer_error *err, int where,
 	return target;
 }
 
-static int pci_read_aer(struct pci_bus *bus, unsigned int devfn, int where,
-			int size, u32 *val)
+static int aer_inj_read_config(struct pci_bus *bus, unsigned int devfn,
+			       int where, int size, u32 *val)
 {
 	u32 *sim;
 	struct aer_error *err;
 	unsigned long flags;
 	struct pci_ops *ops;
+	struct pci_ops *my_ops;
 	int domain;
+	int rv;
 
 	spin_lock_irqsave(&inject_lock, flags);
 	if (size != sizeof(u32))
@@ -208,19 +208,32 @@ static int pci_read_aer(struct pci_bus *bus, unsigned int devfn, int where,
 	}
 out:
 	ops = __find_pci_bus_ops(bus);
+	/*
+	 * pci_lock must already be held, so we can directly
+	 * manipulate bus->ops.  Many config access functions,
+	 * including pci_generic_config_read() require the original
+	 * bus->ops be installed to function, so temporarily put them
+	 * back.
+	 */
+	my_ops = bus->ops;
+	bus->ops = ops;
+	rv = ops->read(bus, devfn, where, size, val);
+	bus->ops = my_ops;
 	spin_unlock_irqrestore(&inject_lock, flags);
-	return ops->read(bus, devfn, where, size, val);
+	return rv;
 }
 
-static int pci_write_aer(struct pci_bus *bus, unsigned int devfn, int where,
-			 int size, u32 val)
+static int aer_inj_write_config(struct pci_bus *bus, unsigned int devfn,
+				int where, int size, u32 val)
 {
 	u32 *sim;
 	struct aer_error *err;
 	unsigned long flags;
 	int rw1cs;
 	struct pci_ops *ops;
+	struct pci_ops *my_ops;
 	int domain;
+	int rv;
 
 	spin_lock_irqsave(&inject_lock, flags);
 	if (size != sizeof(u32))
@@ -243,13 +256,24 @@ static int pci_write_aer(struct pci_bus *bus, unsigned int devfn, int where,
 	}
 out:
 	ops = __find_pci_bus_ops(bus);
+	/*
+	 * pci_lock must already be held, so we can directly
+	 * manipulate bus->ops.  Many config access functions,
+	 * including pci_generic_config_write() require the original
+	 * bus->ops be installed to function, so temporarily put them
+	 * back.
+	 */
+	my_ops = bus->ops;
+	bus->ops = ops;
+	rv = ops->write(bus, devfn, where, size, val);
+	bus->ops = my_ops;
 	spin_unlock_irqrestore(&inject_lock, flags);
-	return ops->write(bus, devfn, where, size, val);
+	return rv;
 }
 
-static struct pci_ops pci_ops_aer = {
-	.read = pci_read_aer,
-	.write = pci_write_aer,
+static struct pci_ops aer_inj_pci_ops = {
+	.read = aer_inj_read_config,
+	.write = aer_inj_write_config,
 };
 
 static void pci_bus_ops_init(struct pci_bus_ops *bus_ops,
@@ -270,9 +294,9 @@ static int pci_bus_set_aer_ops(struct pci_bus *bus)
 	bus_ops = kmalloc(sizeof(*bus_ops), GFP_KERNEL);
 	if (!bus_ops)
 		return -ENOMEM;
-	ops = pci_bus_set_ops(bus, &pci_ops_aer);
+	ops = pci_bus_set_ops(bus, &aer_inj_pci_ops);
 	spin_lock_irqsave(&inject_lock, flags);
-	if (ops == &pci_ops_aer)
+	if (ops == &aer_inj_pci_ops)
 		goto out;
 	pci_bus_ops_init(bus_ops, bus, ops);
 	list_add(&bus_ops->list, &pci_bus_ops_list);
@@ -334,13 +358,15 @@ static int aer_inject(struct aer_error_inj *einj)
 		return -ENODEV;
 	rpdev = pcie_find_root_port(dev);
 	if (!rpdev) {
+		dev_err(&dev->dev, "aer_inject: Root port not found\n");
 		ret = -ENODEV;
 		goto out_put;
 	}
 
 	pos_cap_err = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR);
 	if (!pos_cap_err) {
-		ret = -EPERM;
+		dev_err(&dev->dev, "aer_inject: Device doesn't support AER\n");
+		ret = -EPROTONOSUPPORT;
 		goto out_put;
 	}
 	pci_read_config_dword(dev, pos_cap_err + PCI_ERR_UNCOR_SEVER, &sever);
@@ -350,7 +376,9 @@ static int aer_inject(struct aer_error_inj *einj)
 
 	rp_pos_cap_err = pci_find_ext_capability(rpdev, PCI_EXT_CAP_ID_ERR);
 	if (!rp_pos_cap_err) {
-		ret = -EPERM;
+		dev_err(&rpdev->dev,
+			"aer_inject: Root port doesn't support AER\n");
+		ret = -EPROTONOSUPPORT;
 		goto out_put;
 	}
 
@@ -397,14 +425,16 @@ static int aer_inject(struct aer_error_inj *einj)
 	if (!aer_mask_override && einj->cor_status &&
 	    !(einj->cor_status & ~cor_mask)) {
 		ret = -EINVAL;
-		printk(KERN_WARNING "The correctable error(s) is masked by device\n");
+		dev_warn(&dev->dev,
+			 "aer_inject: The correctable error(s) is masked by device\n");
 		spin_unlock_irqrestore(&inject_lock, flags);
 		goto out_put;
 	}
 	if (!aer_mask_override && einj->uncor_status &&
 	    !(einj->uncor_status & ~uncor_mask)) {
 		ret = -EINVAL;
-		printk(KERN_WARNING "The uncorrectable error(s) is masked by device\n");
+		dev_warn(&dev->dev,
+			 "aer_inject: The uncorrectable error(s) is masked by device\n");
 		spin_unlock_irqrestore(&inject_lock, flags);
 		goto out_put;
 	}
@@ -457,13 +487,19 @@ static int aer_inject(struct aer_error_inj *einj)
 
 	if (find_aer_device(rpdev, &edev)) {
 		if (!get_service_data(edev)) {
-			printk(KERN_WARNING "AER service is not initialized\n");
-			ret = -EINVAL;
+			dev_warn(&edev->device,
+				 "aer_inject: AER service is not initialized\n");
+			ret = -EPROTONOSUPPORT;
 			goto out_put;
 		}
+		dev_info(&edev->device,
+			 "aer_inject: Injecting errors %08x/%08x into device %s\n",
+			 einj->cor_status, einj->uncor_status, pci_name(dev));
 		aer_irq(-1, edev);
-	} else
-		ret = -EINVAL;
+	} else {
+		dev_err(&rpdev->dev, "aer_inject: AER device not found\n");
+		ret = -ENODEV;
+	}
 out_put:
 	kfree(err_alloc);
 	kfree(rperr_alloc);
diff --git a/drivers/pci/pcie/pme.c b/drivers/pci/pcie/pme.c
index 63fc63911295..1ae4c73e7a3c 100644
--- a/drivers/pci/pcie/pme.c
+++ b/drivers/pci/pcie/pme.c
@@ -396,7 +396,7 @@ static int pcie_pme_suspend(struct pcie_device *srv)
 {
 	struct pcie_pme_service_data *data = get_service_data(srv);
 	struct pci_dev *port = srv->port;
-	bool wakeup;
+	bool wakeup, wake_irq_enabled = false;
 	int ret;
 
 	if (device_may_wakeup(&port->dev)) {
@@ -409,11 +409,12 @@ static int pcie_pme_suspend(struct pcie_device *srv)
 	spin_lock_irq(&data->lock);
 	if (wakeup) {
 		ret = enable_irq_wake(srv->irq);
-		data->suspend_level = PME_SUSPEND_WAKEUP;
+		if (ret == 0) {
+			data->suspend_level = PME_SUSPEND_WAKEUP;
+			wake_irq_enabled = true;
+		}
 	}
-	if (!wakeup || ret) {
-		struct pci_dev *port = srv->port;
-
+	if (!wake_irq_enabled) {
 		pcie_pme_interrupt_enable(port, false);
 		pcie_clear_root_pme_status(port);
 		data->suspend_level = PME_SUSPEND_NOIRQ;
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index 6d7ab9bb0d5a..8004f67c57ec 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -15,7 +15,7 @@
 #include <linux/pci-aspm.h>
 #include <linux/aer.h>
 #include <linux/acpi.h>
-#include <asm-generic/pci-bridge.h>
+#include <linux/irqdomain.h>
 #include "pci.h"
 
 #define CARDBUS_LATENCY_TIMER	176	/* secondary latency timer */
@@ -179,6 +179,9 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,
 	u16 orig_cmd;
 	struct pci_bus_region region, inverted_region;
 
+	if (dev->non_compliant_bars)
+		return 0;
+
 	mask = type ? PCI_ROM_ADDRESS_MASK : ~0;
 
 	/* No printks while decoding is disabled! */
@@ -675,6 +678,20 @@ static struct irq_domain *pci_host_bridge_msi_domain(struct pci_bus *bus)
 	if (!d)
 		d = pci_host_bridge_acpi_msi_domain(bus);
 
+#ifdef CONFIG_PCI_MSI_IRQ_DOMAIN
+	/*
+	 * If no IRQ domain was found via the OF tree, try looking it up
+	 * directly through the fwnode_handle.
+	 */
+	if (!d) {
+		struct fwnode_handle *fwnode = pci_root_bus_fwnode(bus);
+
+		if (fwnode)
+			d = irq_find_matching_fwnode(fwnode,
+						     DOMAIN_BUS_PCI_MSI);
+	}
+#endif
+
 	return d;
 }
 
@@ -758,6 +775,12 @@ add_dev:
 
 	pcibios_add_bus(child);
 
+	if (child->ops->add_bus) {
+		ret = child->ops->add_bus(child);
+		if (WARN_ON(ret < 0))
+			dev_err(&child->dev, "failed to add bus: %d\n", ret);
+	}
+
 	/* Create legacy_io and legacy_mem files for this bus */
 	pci_create_legacy_files(child);
 
@@ -1171,6 +1194,7 @@ static void pci_msi_setup_pci_dev(struct pci_dev *dev)
 int pci_setup_device(struct pci_dev *dev)
 {
 	u32 class;
+	u16 cmd;
 	u8 hdr_type;
 	int pos = 0;
 	struct pci_bus_region region;
@@ -1214,6 +1238,16 @@ int pci_setup_device(struct pci_dev *dev)
 	/* device class may be changed after fixup */
 	class = dev->class >> 8;
 
+	if (dev->non_compliant_bars) {
+		pci_read_config_word(dev, PCI_COMMAND, &cmd);
+		if (cmd & (PCI_COMMAND_IO | PCI_COMMAND_MEMORY)) {
+			dev_info(&dev->dev, "device has non-compliant BARs; disabling IO/MEM decoding\n");
+			cmd &= ~PCI_COMMAND_IO;
+			cmd &= ~PCI_COMMAND_MEMORY;
+			pci_write_config_word(dev, PCI_COMMAND, cmd);
+		}
+	}
+
 	switch (dev->hdr_type) {		    /* header type */
 	case PCI_HEADER_TYPE_NORMAL:		    /* standard header */
 		if (class == PCI_CLASS_BRIDGE_PCI)
@@ -1608,7 +1642,7 @@ static void pci_init_capabilities(struct pci_dev *dev)
 	pci_pm_init(dev);
 
 	/* Vital Product Data */
-	pci_vpd_pci22_init(dev);
+	pci_vpd_init(dev);
 
 	/* Alternative Routing-ID Forwarding */
 	pci_configure_ari(dev);
@@ -1803,6 +1837,13 @@ static int only_one_child(struct pci_bus *bus)
 		return 0;
 	if (pci_pcie_type(parent) == PCI_EXP_TYPE_ROOT_PORT)
 		return 1;
+
+	/*
+	 * PCIe downstream ports are bridges that normally lead to only a
+	 * device 0, but if PCI_SCAN_ALL_PCIE_DEVS is set, scan all
+	 * possible devices, not just device 0.  See PCIe spec r3.0,
+	 * sec 7.3.1.
+	 */
 	if (parent->has_secondary_link &&
 	    !pci_has_flag(PCI_SCAN_ALL_PCIE_DEVS))
 		return 1;
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 0575a1e026b4..8e678027b900 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -438,7 +438,7 @@ static void quirk_amd_nl_class(struct pci_dev *pdev)
 	u32 class = pdev->class;
 
 	/* Use "USB Device (not host controller)" class */
-	pdev->class = (PCI_CLASS_SERIAL_USB << 8) | 0xfe;
+	pdev->class = PCI_CLASS_SERIAL_USB_DEVICE;
 	dev_info(&pdev->dev, "PCI class overridden (%#08x -> %#08x) so dwc3 driver can claim this instead of xhci\n",
 		 class, pdev->class);
 }
@@ -2135,6 +2135,35 @@ static void quirk_via_cx700_pci_parking_caching(struct pci_dev *dev)
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, 0x324e, quirk_via_cx700_pci_parking_caching);
 
 /*
+ * If a device follows the VPD format spec, the PCI core will not read or
+ * write past the VPD End Tag.  But some vendors do not follow the VPD
+ * format spec, so we can't tell how much data is safe to access.  Devices
+ * may behave unpredictably if we access too much.  Blacklist these devices
+ * so we don't touch VPD at all.
+ */
+static void quirk_blacklist_vpd(struct pci_dev *dev)
+{
+	if (dev->vpd) {
+		dev->vpd->len = 0;
+		dev_warn(&dev->dev, FW_BUG "VPD access disabled\n");
+	}
+}
+
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x0060, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x007c, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x0413, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x0078, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x0079, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x0073, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x0071, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x005b, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x002f, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x005d, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_LSI_LOGIC, 0x005f, quirk_blacklist_vpd);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATTANSIC, PCI_ANY_ID,
+		quirk_blacklist_vpd);
+
+/*
  * For Broadcom 5706, 5708, 5709 rev. A nics, any read beyond the
  * VPD end tag will hang the device.  This problem was initially
  * observed when a vpd entry was created in sysfs
@@ -3832,6 +3861,19 @@ static int pci_quirk_amd_sb_acs(struct pci_dev *dev, u16 acs_flags)
 #endif
 }
 
+static int pci_quirk_cavium_acs(struct pci_dev *dev, u16 acs_flags)
+{
+	/*
+	 * Cavium devices matching this quirk do not perform peer-to-peer
+	 * with other functions, allowing masking out these bits as if they
+	 * were unimplemented in the ACS capability.
+	 */
+	acs_flags &= ~(PCI_ACS_SV | PCI_ACS_TB | PCI_ACS_RR |
+		       PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_DT);
+
+	return acs_flags ? 0 : 1;
+}
+
 /*
  * Many Intel PCH root ports do provide ACS-like features to disable peer
  * transactions and validate bus numbers in requests, but do not provide an
@@ -3984,6 +4026,8 @@ static const struct pci_dev_acs_enabled {
 	{ PCI_VENDOR_ID_INTEL, PCI_ANY_ID, pci_quirk_intel_pch_acs },
 	{ 0x19a2, 0x710, pci_quirk_mf_endpoint_acs }, /* Emulex BE3-R */
 	{ 0x10df, 0x720, pci_quirk_mf_endpoint_acs }, /* Emulex Skyhawk-R */
+	/* Cavium ThunderX */
+	{ PCI_VENDOR_ID_CAVIUM, PCI_ANY_ID, pci_quirk_cavium_acs },
 	{ 0 }
 };
 
diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c
index 8a280e9c2ad1..8982026637d5 100644
--- a/drivers/pci/remove.c
+++ b/drivers/pci/remove.c
@@ -7,7 +7,6 @@ static void pci_free_resources(struct pci_dev *dev)
 {
 	int i;
 
-	pci_cleanup_rom(dev);
 	for (i = 0; i < PCI_NUM_RESOURCES; i++) {
 		struct resource *res = dev->resource + i;
 		if (res->parent)
@@ -54,6 +53,10 @@ void pci_remove_bus(struct pci_bus *bus)
 	pci_bus_release_busn_res(bus);
 	up_write(&pci_bus_sem);
 	pci_remove_legacy_files(bus);
+
+	if (bus->ops->remove_bus)
+		bus->ops->remove_bus(bus);
+
 	pcibios_remove_bus(bus);
 	device_unregister(&bus->dev);
 }
diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c
index 9eaca39ef38d..06663d391b39 100644
--- a/drivers/pci/rom.c
+++ b/drivers/pci/rom.c
@@ -24,13 +24,17 @@
  */
 int pci_enable_rom(struct pci_dev *pdev)
 {
-	struct resource *res = pdev->resource + PCI_ROM_RESOURCE;
+	struct resource *res = &pdev->resource[PCI_ROM_RESOURCE];
 	struct pci_bus_region region;
 	u32 rom_addr;
 
 	if (!res->flags)
 		return -1;
 
+	/* Nothing to enable if we're using a shadow copy in RAM */
+	if (res->flags & IORESOURCE_ROM_SHADOW)
+		return 0;
+
 	pcibios_resource_to_bus(pdev->bus, &region, res);
 	pci_read_config_dword(pdev, pdev->rom_base_reg, &rom_addr);
 	rom_addr &= ~PCI_ROM_ADDRESS_MASK;
@@ -49,7 +53,12 @@ EXPORT_SYMBOL_GPL(pci_enable_rom);
  */
 void pci_disable_rom(struct pci_dev *pdev)
 {
+	struct resource *res = &pdev->resource[PCI_ROM_RESOURCE];
 	u32 rom_addr;
+
+	if (res->flags & IORESOURCE_ROM_SHADOW)
+		return;
+
 	pci_read_config_dword(pdev, pdev->rom_base_reg, &rom_addr);
 	rom_addr &= ~PCI_ROM_ADDRESS_ENABLE;
 	pci_write_config_dword(pdev, pdev->rom_base_reg, rom_addr);
@@ -119,43 +128,23 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size)
 	loff_t start;
 	void __iomem *rom;
 
-	/*
-	 * IORESOURCE_ROM_SHADOW set on x86, x86_64 and IA64 supports legacy
-	 * memory map if the VGA enable bit of the Bridge Control register is
-	 * set for embedded VGA.
-	 */
-	if (res->flags & IORESOURCE_ROM_SHADOW) {
-		/* primary video rom always starts here */
-		start = (loff_t)0xC0000;
-		*size = 0x20000; /* cover C000:0 through E000:0 */
-	} else {
-		if (res->flags &
-			(IORESOURCE_ROM_COPY | IORESOURCE_ROM_BIOS_COPY)) {
-			*size = pci_resource_len(pdev, PCI_ROM_RESOURCE);
-			return (void __iomem *)(unsigned long)
-				pci_resource_start(pdev, PCI_ROM_RESOURCE);
-		} else {
-			/* assign the ROM an address if it doesn't have one */
-			if (res->parent == NULL &&
-			    pci_assign_resource(pdev, PCI_ROM_RESOURCE))
-				return NULL;
-			start = pci_resource_start(pdev, PCI_ROM_RESOURCE);
-			*size = pci_resource_len(pdev, PCI_ROM_RESOURCE);
-			if (*size == 0)
-				return NULL;
-
-			/* Enable ROM space decodes */
-			if (pci_enable_rom(pdev))
-				return NULL;
-		}
-	}
+	/* assign the ROM an address if it doesn't have one */
+	if (res->parent == NULL && pci_assign_resource(pdev, PCI_ROM_RESOURCE))
+		return NULL;
+
+	start = pci_resource_start(pdev, PCI_ROM_RESOURCE);
+	*size = pci_resource_len(pdev, PCI_ROM_RESOURCE);
+	if (*size == 0)
+		return NULL;
+
+	/* Enable ROM space decodes */
+	if (pci_enable_rom(pdev))
+		return NULL;
 
 	rom = ioremap(start, *size);
 	if (!rom) {
 		/* restore enable if ioremap fails */
-		if (!(res->flags & (IORESOURCE_ROM_ENABLE |
-				    IORESOURCE_ROM_SHADOW |
-				    IORESOURCE_ROM_COPY)))
+		if (!(res->flags & IORESOURCE_ROM_ENABLE))
 			pci_disable_rom(pdev);
 		return NULL;
 	}
@@ -181,37 +170,15 @@ void pci_unmap_rom(struct pci_dev *pdev, void __iomem *rom)
 {
 	struct resource *res = &pdev->resource[PCI_ROM_RESOURCE];
 
-	if (res->flags & (IORESOURCE_ROM_COPY | IORESOURCE_ROM_BIOS_COPY))
-		return;
-
 	iounmap(rom);
 
-	/* Disable again before continuing, leave enabled if pci=rom */
-	if (!(res->flags & (IORESOURCE_ROM_ENABLE | IORESOURCE_ROM_SHADOW)))
+	/* Disable again before continuing */
+	if (!(res->flags & IORESOURCE_ROM_ENABLE))
 		pci_disable_rom(pdev);
 }
 EXPORT_SYMBOL(pci_unmap_rom);
 
 /**
- * pci_cleanup_rom - free the ROM copy created by pci_map_rom_copy
- * @pdev: pointer to pci device struct
- *
- * Free the copied ROM if we allocated one.
- */
-void pci_cleanup_rom(struct pci_dev *pdev)
-{
-	struct resource *res = &pdev->resource[PCI_ROM_RESOURCE];
-
-	if (res->flags & IORESOURCE_ROM_COPY) {
-		kfree((void *)(unsigned long)res->start);
-		res->flags |= IORESOURCE_UNSET;
-		res->flags &= ~IORESOURCE_ROM_COPY;
-		res->start = 0;
-		res->end = 0;
-	}
-}
-
-/**
  * pci_platform_rom - provides a pointer to any ROM image provided by the
  * platform
  * @pdev: pointer to pci device struct
diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c
index 7796d0a5befa..55641a39a3e9 100644
--- a/drivers/pci/setup-bus.c
+++ b/drivers/pci/setup-bus.c
@@ -25,7 +25,6 @@
 #include <linux/ioport.h>
 #include <linux/cache.h>
 #include <linux/slab.h>
-#include <asm-generic/pci-bridge.h>
 #include "pci.h"
 
 unsigned int pci_flags;
diff --git a/drivers/pci/setup-res.c b/drivers/pci/setup-res.c
index 604011e047d6..66c4d8f42233 100644
--- a/drivers/pci/setup-res.c
+++ b/drivers/pci/setup-res.c
@@ -276,6 +276,9 @@ int pci_assign_resource(struct pci_dev *dev, int resno)
 	resource_size_t align, size;
 	int ret;
 
+	if (res->flags & IORESOURCE_PCI_FIXED)
+		return 0;
+
 	res->flags |= IORESOURCE_UNSET;
 	align = pci_resource_alignment(dev, res);
 	if (!align) {
@@ -321,6 +324,9 @@ int pci_reassign_resource(struct pci_dev *dev, int resno, resource_size_t addsiz
 	resource_size_t new_size;
 	int ret;
 
+	if (res->flags & IORESOURCE_PCI_FIXED)
+		return 0;
+
 	flags = res->flags;
 	res->flags |= IORESOURCE_UNSET;
 	if (!res->parent) {