Merge tag 'pci-for-3.5' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci

Pull PCI changes from Bjorn Helgaas:
 - Host bridge cleanups from Yinghai
 - Disable Bus Master bit on PCI device shutdown (kexec-related)
 - Stratus ftServer fix
 - pci_dev_reset() locking fix
 - IvyBridge graphics erratum workaround

* tag 'pci-for-3.5' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci: (21 commits)
  microblaze/PCI: fix "io_offset undeclared" error
  x86/PCI: only check for spinlock being held in SMP kernels
  resources: add resource_overlaps()
  PCI: fix uninitialized variable 'cap_mask'
  MAINTAINERS: update PCI git tree and patchwork
  PCI: disable Bus Master on PCI device shutdown
  PCI: work around IvyBridge internal graphics FLR erratum
  x86/PCI: fix unused variable warning in amd_bus.c
  PCI: move mutex locking out of pci_dev_reset function
  PCI: work around Stratus ftServer broken PCIe hierarchy
  x86/PCI: merge pcibios_scan_root() and pci_scan_bus_on_node()
  x86/PCI: dynamically allocate pci_root_info for native host bridge drivers
  x86/PCI: embed pci_sysdata into pci_root_info on ACPI path
  x86/PCI: embed name into pci_root_info struct
  x86/PCI: add host bridge resource release for _CRS path
  x86/PCI: refactor get_current_resources()
  PCI: add host bridge release support
  PCI: add generic device into pci_host_bridge struct
  PCI: rename pci_host_bridge() to find_pci_root_bridge()
  x86/PCI: fix memleak with get_current_resources()
  ...

Documentation/kernel-parameters.txt

@@ -2161,6 +2161,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 				on: Turn realloc on
 		realloc		same as realloc=on
 		noari		do not use PCIe ARI.
+		pcie_scan_all	Scan all possible PCIe devices.  Otherwise we
+				only look for one device below a PCIe downstream
+				port.
 
 	pcie_aspm=	[PCIE] Forcibly enable or disable PCIe Active State Power
 			Management.

MAINTAINERS

@@ -5128,19 +5128,13 @@ F:	Documentation/powerpc/eeh-pci-error-recovery.txt
 PCI SUBSYSTEM
 M:	Bjorn Helgaas <bhelgaas@google.com>
 L:	linux-pci@vger.kernel.org
-Q:	http://patchwork.kernel.org/project/linux-pci/list/
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci.git
+Q:	http://patchwork.ozlabs.org/project/linux-pci/list/
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/linux.git
 S:	Supported
 F:	Documentation/PCI/
 F:	drivers/pci/
 F:	include/linux/pci*
 
-PCI HOTPLUG
-M:	Bjorn Helgaas <bhelgaas@google.com>
-L:	linux-pci@vger.kernel.org
-S:	Supported
-F:	drivers/pci/hotplug
-
 PCMCIA SUBSYSTEM
 P:	Linux PCMCIA Team
 L:	linux-pcmcia@lists.infradead.org

arch/microblaze/pci/pci-common.c

@@ -1422,6 +1422,7 @@ int pcibios_enable_device(struct pci_dev *dev, int mask)
 
 static void __devinit pcibios_setup_phb_resources(struct pci_controller *hose, struct list_head *resources)
 {
+	unsigned long io_offset;
 	struct resource *res;
 	int i;
 

arch/x86/pci/acpi.c

@@ -9,11 +9,11 @@
 
 struct pci_root_info {
 	struct acpi_device *bridge;
-	char *name;
+	char name[16];
 	unsigned int res_num;
 	struct resource *res;
-	struct list_head *resources;
 	int busnum;
+	struct pci_sysdata sd;
 };
 
 static bool pci_use_crs = true;
@@ -245,13 +245,6 @@ setup_resource(struct acpi_resource *acpi_res, void *data)
 	return AE_OK;
 }
 
-static bool resource_contains(struct resource *res, resource_size_t point)
-{
-	if (res->start <= point && point <= res->end)
-		return true;
-	return false;
-}
-
 static void coalesce_windows(struct pci_root_info *info, unsigned long type)
 {
 	int i, j;
@@ -272,10 +265,7 @@ static void coalesce_windows(struct pci_root_info *info, unsigned long type)
 			 * our resources no longer match the ACPI _CRS, but
 			 * the kernel resource tree doesn't allow overlaps.
 			 */
-			if (resource_contains(res1, res2->start) ||
-			    resource_contains(res1, res2->end) ||
-			    resource_contains(res2, res1->start) ||
-			    resource_contains(res2, res1->end)) {
+			if (resource_overlaps(res1, res2)) {
 				res1->start = min(res1->start, res2->start);
 				res1->end = max(res1->end, res2->end);
 				dev_info(&info->bridge->dev,
@@ -287,7 +277,8 @@ static void coalesce_windows(struct pci_root_info *info, unsigned long type)
 	}
 }
 
-static void add_resources(struct pci_root_info *info)
+static void add_resources(struct pci_root_info *info,
+			  struct list_head *resources)
 {
 	int i;
 	struct resource *res, *root, *conflict;
@@ -311,53 +302,74 @@ static void add_resources(struct pci_root_info *info)
 				 "ignoring host bridge window %pR (conflicts with %s %pR)\n",
 				 res, conflict->name, conflict);
 		else
-			pci_add_resource(info->resources, res);
+			pci_add_resource(resources, res);
 	}
 }
 
+static void free_pci_root_info_res(struct pci_root_info *info)
+{
+	kfree(info->res);
+	info->res = NULL;
+	info->res_num = 0;
+}
+
+static void __release_pci_root_info(struct pci_root_info *info)
+{
+	int i;
+	struct resource *res;
+
+	for (i = 0; i < info->res_num; i++) {
+		res = &info->res[i];
+
+		if (!res->parent)
+			continue;
+
+		if (!(res->flags & (IORESOURCE_MEM | IORESOURCE_IO)))
+			continue;
+
+		release_resource(res);
+	}
+
+	free_pci_root_info_res(info);
+
+	kfree(info);
+}
+static void release_pci_root_info(struct pci_host_bridge *bridge)
+{
+	struct pci_root_info *info = bridge->release_data;
+
+	__release_pci_root_info(info);
+}
+
 static void
-get_current_resources(struct acpi_device *device, int busnum,
-		      int domain, struct list_head *resources)
+probe_pci_root_info(struct pci_root_info *info, struct acpi_device *device,
+		    int busnum, int domain)
 {
-	struct pci_root_info info;
 	size_t size;
 
-	info.bridge = device;
-	info.res_num = 0;
-	info.resources = resources;
+	info->bridge = device;
+	info->res_num = 0;
 	acpi_walk_resources(device->handle, METHOD_NAME__CRS, count_resource,
-				&info);
-	if (!info.res_num)
+				info);
+	if (!info->res_num)
 		return;
 
-	size = sizeof(*info.res) * info.res_num;
-	info.res = kmalloc(size, GFP_KERNEL);
-	if (!info.res)
+	size = sizeof(*info->res) * info->res_num;
+	info->res_num = 0;
+	info->res = kmalloc(size, GFP_KERNEL);
+	if (!info->res)
 		return;
 
-	info.name = kasprintf(GFP_KERNEL, "PCI Bus %04x:%02x", domain, busnum);
-	if (!info.name)
-		goto name_alloc_fail;
+	sprintf(info->name, "PCI Bus %04x:%02x", domain, busnum);
 
-	info.res_num = 0;
 	acpi_walk_resources(device->handle, METHOD_NAME__CRS, setup_resource,
-				&info);
-
-	if (pci_use_crs) {
-		add_resources(&info);
-
-		return;
-	}
-
-	kfree(info.name);
-
-name_alloc_fail:
-	kfree(info.res);
+				info);
 }
 
 struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_pci_root *root)
 {
 	struct acpi_device *device = root->device;
+	struct pci_root_info *info = NULL;
 	int domain = root->segment;
 	int busnum = root->secondary.start;
 	LIST_HEAD(resources);
@@ -389,17 +401,14 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_pci_root *root)
 	if (node != -1 && !node_online(node))
 		node = -1;
 
-	/* Allocate per-root-bus (not per bus) arch-specific data.
-	 * TODO: leak; this memory is never freed.
-	 * It's arguable whether it's worth the trouble to care.
-	 */
-	sd = kzalloc(sizeof(*sd), GFP_KERNEL);
-	if (!sd) {
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (!info) {
 		printk(KERN_WARNING "pci_bus %04x:%02x: "
 		       "ignored (out of memory)\n", domain, busnum);
 		return NULL;
 	}
 
+	sd = &info->sd;
 	sd->domain = domain;
 	sd->node = node;
 	/*
@@ -413,22 +422,32 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_pci_root *root)
 		 * be replaced by sd.
 		 */
 		memcpy(bus->sysdata, sd, sizeof(*sd));
-		kfree(sd);
+		kfree(info);
 	} else {
-		get_current_resources(device, busnum, domain, &resources);
+		probe_pci_root_info(info, device, busnum, domain);
 
 		/*
 		 * _CRS with no apertures is normal, so only fall back to
 		 * defaults or native bridge info if we're ignoring _CRS.
 		 */
-		if (!pci_use_crs)
+		if (pci_use_crs)
+			add_resources(info, &resources);
+		else {
+			free_pci_root_info_res(info);
 			x86_pci_root_bus_resources(busnum, &resources);
+		}
+
 		bus = pci_create_root_bus(NULL, busnum, &pci_root_ops, sd,
 					  &resources);
-		if (bus)
+		if (bus) {
 			bus->subordinate = pci_scan_child_bus(bus);
-		else
+			pci_set_host_bridge_release(
+				to_pci_host_bridge(bus->bridge),
+				release_pci_root_info, info);
+		} else {
 			pci_free_resource_list(&resources);
+			__release_pci_root_info(info);
+		}
 	}
 
 	/* After the PCI-E bus has been walked and all devices discovered,
@@ -445,9 +464,6 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_pci_root *root)
 		}
 	}
 
-	if (!bus)
-		kfree(sd);
-
 	if (bus && node != -1) {
 #ifdef CONFIG_ACPI_NUMA
 		if (pxm >= 0)

arch/x86/pci/amd_bus.c

@@ -32,6 +32,27 @@ static struct pci_hostbridge_probe pci_probes[] __initdata = {
 
 #define RANGE_NUM 16
 
+static struct pci_root_info __init *find_pci_root_info(int node, int link)
+{
+	struct pci_root_info *info;
+
+	/* find the position */
+	list_for_each_entry(info, &pci_root_infos, list)
+		if (info->node == node && info->link == link)
+			return info;
+
+	return NULL;
+}
+
+static void __init set_mp_bus_range_to_node(int min_bus, int max_bus, int node)
+{
+#ifdef CONFIG_NUMA
+	int j;
+
+	for (j = min_bus; j <= max_bus; j++)
+		set_mp_bus_to_node(j, node);
+#endif
+}
 /**
  * early_fill_mp_bus_to_node()
  * called before pcibios_scan_root and pci_scan_bus
@@ -41,7 +62,6 @@ static struct pci_hostbridge_probe pci_probes[] __initdata = {
 static int __init early_fill_mp_bus_info(void)
 {
 	int i;
-	int j;
 	unsigned bus;
 	unsigned slot;
 	int node;
@@ -50,7 +70,6 @@ static int __init early_fill_mp_bus_info(void)
 	int def_link;
 	struct pci_root_info *info;
 	u32 reg;
-	struct resource *res;
 	u64 start;
 	u64 end;
 	struct range range[RANGE_NUM];
@@ -86,7 +105,6 @@ static int __init early_fill_mp_bus_info(void)
 	if (!found)
 		return 0;
 
-	pci_root_num = 0;
 	for (i = 0; i < 4; i++) {
 		int min_bus;
 		int max_bus;
@@ -99,19 +117,11 @@ static int __init early_fill_mp_bus_info(void)
 		min_bus = (reg >> 16) & 0xff;
 		max_bus = (reg >> 24) & 0xff;
 		node = (reg >> 4) & 0x07;
-#ifdef CONFIG_NUMA
-		for (j = min_bus; j <= max_bus; j++)
-			set_mp_bus_to_node(j, node);
-#endif
+		set_mp_bus_range_to_node(min_bus, max_bus, node);
 		link = (reg >> 8) & 0x03;
 
-		info = &pci_root_info[pci_root_num];
-		info->bus_min = min_bus;
-		info->bus_max = max_bus;
-		info->node = node;
-		info->link = link;
+		info = alloc_pci_root_info(min_bus, max_bus, node, link);
 		sprintf(info->name, "PCI Bus #%02x", min_bus);
-		pci_root_num++;
 	}
 
 	/* get the default node and link for left over res */
@@ -134,16 +144,10 @@ static int __init early_fill_mp_bus_info(void)
 		link = (reg >> 4) & 0x03;
 		end = (reg & 0xfff000) | 0xfff;
 
-		/* find the position */
-		for (j = 0; j < pci_root_num; j++) {
-			info = &pci_root_info[j];
-			if (info->node == node && info->link == link)
-				break;
-		}
-		if (j == pci_root_num)
+		info = find_pci_root_info(node, link);
+		if (!info)
 			continue; /* not found */
 
-		info = &pci_root_info[j];
 		printk(KERN_DEBUG "node %d link %d: io port [%llx, %llx]\n",
 		       node, link, start, end);
 
@@ -155,13 +159,8 @@ static int __init early_fill_mp_bus_info(void)
 	}
 	/* add left over io port range to def node/link, [0, 0xffff] */
 	/* find the position */
-	for (j = 0; j < pci_root_num; j++) {
-		info = &pci_root_info[j];
-		if (info->node == def_node && info->link == def_link)
-			break;
-	}
-	if (j < pci_root_num) {
-		info = &pci_root_info[j];
+	info = find_pci_root_info(def_node, def_link);
+	if (info) {
 		for (i = 0; i < RANGE_NUM; i++) {
 			if (!range[i].end)
 				continue;
@@ -214,16 +213,10 @@ static int __init early_fill_mp_bus_info(void)
 		end <<= 8;
 		end |= 0xffff;
 
-		/* find the position */
-		for (j = 0; j < pci_root_num; j++) {
-			info = &pci_root_info[j];
-			if (info->node == node && info->link == link)
-				break;
-		}
-		if (j == pci_root_num)
-			continue; /* not found */
+		info = find_pci_root_info(node, link);
 
-		info = &pci_root_info[j];
+		if (!info)
+			continue;
 
 		printk(KERN_DEBUG "node %d link %d: mmio [%llx, %llx]",
 		       node, link, start, end);
@@ -291,14 +284,8 @@ static int __init early_fill_mp_bus_info(void)
 	 * add left over mmio range to def node/link ?
 	 * that is tricky, just record range in from start_min to 4G
 	 */
-	for (j = 0; j < pci_root_num; j++) {
-		info = &pci_root_info[j];
-		if (info->node == def_node && info->link == def_link)
-			break;
-	}
-	if (j < pci_root_num) {
-		info = &pci_root_info[j];
-
+	info = find_pci_root_info(def_node, def_link);
+	if (info) {
 		for (i = 0; i < RANGE_NUM; i++) {
 			if (!range[i].end)
 				continue;
@@ -309,20 +296,16 @@ static int __init early_fill_mp_bus_info(void)
 		}
 	}
 
-	for (i = 0; i < pci_root_num; i++) {
-		int res_num;
+	list_for_each_entry(info, &pci_root_infos, list) {
 		int busnum;
+		struct pci_root_res *root_res;
 
-		info = &pci_root_info[i];
-		res_num = info->res_num;
 		busnum = info->bus_min;
 		printk(KERN_DEBUG "bus: [%02x, %02x] on node %x link %x\n",
 		       info->bus_min, info->bus_max, info->node, info->link);
-		for (j = 0; j < res_num; j++) {
-			res = &info->res[j];
-			printk(KERN_DEBUG "bus: %02x index %x %pR\n",
-				       busnum, j, res);
-		}
+		list_for_each_entry(root_res, &info->resources, list)
+			printk(KERN_DEBUG "bus: %02x %pR\n",
+				       busnum, &root_res->res);
 	}
 
 	return 0;

arch/x86/pci/broadcom_bus.c

@@ -22,19 +22,15 @@
 static void __init cnb20le_res(u8 bus, u8 slot, u8 func)
 {
 	struct pci_root_info *info;
+	struct pci_root_res *root_res;
 	struct resource res;
 	u16 word1, word2;
 	u8 fbus, lbus;
-	int i;
-
-	info = &pci_root_info[pci_root_num];
-	pci_root_num++;
 
 	/* read the PCI bus numbers */
 	fbus = read_pci_config_byte(bus, slot, func, 0x44);
 	lbus = read_pci_config_byte(bus, slot, func, 0x45);
-	info->bus_min = fbus;
-	info->bus_max = lbus;
+	info = alloc_pci_root_info(fbus, lbus, 0, 0);
 
 	/*
 	 * Add the legacy IDE ports on bus 0
@@ -86,8 +82,8 @@ static void __init cnb20le_res(u8 bus, u8 slot, u8 func)
 	res.flags = IORESOURCE_BUS;
 	printk(KERN_INFO "CNB20LE PCI Host Bridge (domain 0000 %pR)\n", &res);
 
-	for (i = 0; i < info->res_num; i++)
-		printk(KERN_INFO "host bridge window %pR\n", &info->res[i]);
+	list_for_each_entry(root_res, &info->resources, list)
+		printk(KERN_INFO "host bridge window %pR\n", &root_res->res);
 }
 
 static int __init broadcom_postcore_init(void)

arch/x86/pci/bus_numa.c

@@ -4,35 +4,38 @@
 
 #include "bus_numa.h"
 
-int pci_root_num;
-struct pci_root_info pci_root_info[PCI_ROOT_NR];
+LIST_HEAD(pci_root_infos);
 
-void x86_pci_root_bus_resources(int bus, struct list_head *resources)
+static struct pci_root_info *x86_find_pci_root_info(int bus)
 {
-	int i;
-	int j;
 	struct pci_root_info *info;
 
-	if (!pci_root_num)
-		goto default_resources;
+	if (list_empty(&pci_root_infos))
+		return NULL;
 
-	for (i = 0; i < pci_root_num; i++) {
-		if (pci_root_info[i].bus_min == bus)
-			break;
-	}
+	list_for_each_entry(info, &pci_root_infos, list)
+		if (info->bus_min == bus)
+			return info;
+
+	return NULL;
+}
 
-	if (i == pci_root_num)
+void x86_pci_root_bus_resources(int bus, struct list_head *resources)
+{
+	struct pci_root_info *info = x86_find_pci_root_info(bus);
+	struct pci_root_res *root_res;
+
+	if (!info)
 		goto default_resources;
 
 	printk(KERN_DEBUG "PCI: root bus %02x: hardware-probed resources\n",
 	       bus);
 
-	info = &pci_root_info[i];
-	for (j = 0; j < info->res_num; j++) {
+	list_for_each_entry(root_res, &info->resources, list) {
 		struct resource *res;
 		struct resource *root;
 
-		res = &info->res[j];
+		res = &root_res->res;
 		pci_add_resource(resources, res);
 		if (res->flags & IORESOURCE_IO)
 			root = &ioport_resource;
@@ -53,11 +56,32 @@ void x86_pci_root_bus_resources(int bus, struct list_head *resources)
 	pci_add_resource(resources, &iomem_resource);
 }
 
+struct pci_root_info __init *alloc_pci_root_info(int bus_min, int bus_max,
+						 int node, int link)
+{
+	struct pci_root_info *info;
+
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+
+	if (!info)
+		return info;
+
+	INIT_LIST_HEAD(&info->resources);
+	info->bus_min = bus_min;
+	info->bus_max = bus_max;
+	info->node = node;
+	info->link = link;
+
+	list_add_tail(&info->list, &pci_root_infos);
+
+	return info;
+}
+
 void __devinit update_res(struct pci_root_info *info, resource_size_t start,
 			  resource_size_t end, unsigned long flags, int merge)
 {
-	int i;
 	struct resource *res;
+	struct pci_root_res *root_res;
 
 	if (start > end)
 		return;
@@ -69,11 +93,11 @@ void __devinit update_res(struct pci_root_info *info, resource_size_t start,
 		goto addit;
 
 	/* try to merge it with old one */
-	for (i = 0; i < info->res_num; i++) {
+	list_for_each_entry(root_res, &info->resources, list) {
 		resource_size_t final_start, final_end;
 		resource_size_t common_start, common_end;
 
-		res = &info->res[i];
+		res = &root_res->res;
 		if (res->flags != flags)
 			continue;
 
@@ -93,14 +117,15 @@ void __devinit update_res(struct pci_root_info *info, resource_size_t start,
 addit:
 
 	/* need to add that */
-	if (info->res_num >= RES_NUM)
+	root_res = kzalloc(sizeof(*root_res), GFP_KERNEL);
+	if (!root_res)
 		return;
 
-	res = &info->res[info->res_num];
+	res = &root_res->res;
 	res->name = info->name;
 	res->flags = flags;
 	res->start = start;
 	res->end = end;
-	res->child = NULL;
-	info->res_num++;
+
+	list_add_tail(&root_res->list, &info->resources);
 }

arch/x86/pci/bus_numa.h

@@ -4,22 +4,24 @@
  * sub bus (transparent) will use entres from 3 to store extra from
  * root, so need to make sure we have enough slot there.
  */
-#define RES_NUM 16
+struct pci_root_res {
+	struct list_head list;
+	struct resource res;
+};
+
 struct pci_root_info {
+	struct list_head list;
 	char name[12];
-	unsigned int res_num;
-	struct resource res[RES_NUM];
+	struct list_head resources;
 	int bus_min;
 	int bus_max;
 	int node;
 	int link;
 };
 
-/* 4 at this time, it may become to 32 */
-#define PCI_ROOT_NR 4
-extern int pci_root_num;
-extern struct pci_root_info pci_root_info[PCI_ROOT_NR];
-
+extern struct list_head pci_root_infos;
+struct pci_root_info *alloc_pci_root_info(int bus_min, int bus_max,
+						int node, int link);
 extern void update_res(struct pci_root_info *info, resource_size_t start,
 		      resource_size_t end, unsigned long flags, int merge);
 #endif

arch/x86/pci/common.c

@@ -11,6 +11,7 @@
 #include <linux/dmi.h>
 #include <linux/slab.h>
 
+#include <asm-generic/pci-bridge.h>
 #include <asm/acpi.h>
 #include <asm/segment.h>
 #include <asm/io.h>
@@ -229,6 +230,14 @@ static int __devinit assign_all_busses(const struct dmi_system_id *d)
 }
 #endif
 
+static int __devinit set_scan_all(const struct dmi_system_id *d)
+{
+	printk(KERN_INFO "PCI: %s detected, enabling pci=pcie_scan_all\n",
+	       d->ident);
+	pci_add_flags(PCI_SCAN_ALL_PCIE_DEVS);
+	return 0;
+}
+
 static const struct dmi_system_id __devinitconst pciprobe_dmi_table[] = {
 #ifdef __i386__
 /*
@@ -420,6 +429,13 @@ static const struct dmi_system_id __devinitconst pciprobe_dmi_table[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL585 G2"),
 		},
 	},
+	{
+		.callback = set_scan_all,
+		.ident = "Stratus/NEC ftServer",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ftServer"),
+		},
+	},
 	{}
 };
 
@@ -430,9 +446,7 @@ void __init dmi_check_pciprobe(void)
 
 struct pci_bus * __devinit pcibios_scan_root(int busnum)
 {
-	LIST_HEAD(resources);
 	struct pci_bus *bus = NULL;
-	struct pci_sysdata *sd;
 
 	while ((bus = pci_find_next_bus(bus)) != NULL) {
 		if (bus->number == busnum) {
@@ -441,28 +455,10 @@ struct pci_bus * __devinit pcibios_scan_root(int busnum)
 		}
 	}
 
-	/* Allocate per-root-bus (not per bus) arch-specific data.
-	 * TODO: leak; this memory is never freed.
-	 * It's arguable whether it's worth the trouble to care.
-	 */
-	sd = kzalloc(sizeof(*sd), GFP_KERNEL);
-	if (!sd) {
-		printk(KERN_ERR "PCI: OOM, not probing PCI bus %02x\n", busnum);
-		return NULL;
-	}
-
-	sd->node = get_mp_bus_to_node(busnum);
-
-	printk(KERN_DEBUG "PCI: Probing PCI hardware (bus %02x)\n", busnum);
-	x86_pci_root_bus_resources(busnum, &resources);
-	bus = pci_scan_root_bus(NULL, busnum, &pci_root_ops, sd, &resources);
-	if (!bus) {
-		pci_free_resource_list(&resources);
-		kfree(sd);
-	}
-
-	return bus;
+	return pci_scan_bus_on_node(busnum, &pci_root_ops,
+					get_mp_bus_to_node(busnum));
 }
+
 void __init pcibios_set_cache_line_size(void)
 {
 	struct cpuinfo_x86 *c = &boot_cpu_data;
@@ -656,6 +652,7 @@ struct pci_bus * __devinit pci_scan_bus_on_node(int busno, struct pci_ops *ops,
 	}
 	sd->node = node;
 	x86_pci_root_bus_resources(busno, &resources);
+	printk(KERN_DEBUG "PCI: Probing PCI hardware (bus %02x)\n", busno);
 	bus = pci_scan_root_bus(NULL, busno, ops, sd, &resources);
 	if (!bus) {
 		pci_free_resource_list(&resources);

arch/x86/pci/i386.c

@@ -57,7 +57,7 @@ static struct pcibios_fwaddrmap *pcibios_fwaddrmap_lookup(struct pci_dev *dev)
 {
 	struct pcibios_fwaddrmap *map;
 
-	WARN_ON(!spin_is_locked(&pcibios_fwaddrmap_lock));
+	WARN_ON_SMP(!spin_is_locked(&pcibios_fwaddrmap_lock));
 
 	list_for_each_entry(map, &pcibios_fwaddrmappings, list)
 		if (map->dev == dev)

drivers/pci/Makefile

@@ -2,7 +2,7 @@
 # Makefile for the PCI bus specific drivers.
 #
 
-obj-y		+= access.o bus.o probe.o remove.o pci.o \
+obj-y		+= access.o bus.o probe.o host-bridge.o remove.o pci.o \
 			pci-driver.o search.o pci-sysfs.o rom.o setup-res.o \
 			irq.o vpd.o
 obj-$(CONFIG_PROC_FS) += proc.o

drivers/pci/host-bridge.c

+/*
+ * host bridge related code
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/module.h>
+
+#include "pci.h"
+
+static struct pci_bus *find_pci_root_bus(struct pci_dev *dev)
+{
+	struct pci_bus *bus;
+
+	bus = dev->bus;
+	while (bus->parent)
+		bus = bus->parent;
+
+	return bus;
+}
+
+static struct pci_host_bridge *find_pci_host_bridge(struct pci_dev *dev)
+{
+	struct pci_bus *bus = find_pci_root_bus(dev);
+
+	return to_pci_host_bridge(bus->bridge);
+}
+
+void pci_set_host_bridge_release(struct pci_host_bridge *bridge,
+				 void (*release_fn)(struct pci_host_bridge *),
+				 void *release_data)
+{
+	bridge->release_fn = release_fn;
+	bridge->release_data = release_data;
+}
+
+static bool resource_contains(struct resource *res1, struct resource *res2)
+{
+	return res1->start <= res2->start && res1->end >= res2->end;
+}
+
+void pcibios_resource_to_bus(struct pci_dev *dev, struct pci_bus_region *region,
+			     struct resource *res)
+{
+	struct pci_host_bridge *bridge = find_pci_host_bridge(dev);
+	struct pci_host_bridge_window *window;
+	resource_size_t offset = 0;
+
+	list_for_each_entry(window, &bridge->windows, list) {
+		if (resource_type(res) != resource_type(window->res))
+			continue;
+
+		if (resource_contains(window->res, res)) {
+			offset = window->offset;
+			break;
+		}
+	}
+
+	region->start = res->start - offset;
+	region->end = res->end - offset;
+}
+EXPORT_SYMBOL(pcibios_resource_to_bus);
+
+static bool region_contains(struct pci_bus_region *region1,
+			    struct pci_bus_region *region2)
+{
+	return region1->start <= region2->start && region1->end >= region2->end;
+}
+
+void pcibios_bus_to_resource(struct pci_dev *dev, struct resource *res,
+			     struct pci_bus_region *region)
+{
+	struct pci_host_bridge *bridge = find_pci_host_bridge(dev);
+	struct pci_host_bridge_window *window;
+	resource_size_t offset = 0;
+
+	list_for_each_entry(window, &bridge->windows, list) {
+		struct pci_bus_region bus_region;
+
+		if (resource_type(res) != resource_type(window->res))
+			continue;
+
+		bus_region.start = window->res->start - window->offset;
+		bus_region.end = window->res->end - window->offset;
+
+		if (region_contains(&bus_region, region)) {
+			offset = window->offset;
+			break;
+		}
+	}
+
+	res->start = region->start + offset;
+	res->end = region->end + offset;
+}
+EXPORT_SYMBOL(pcibios_bus_to_resource);

drivers/pci/pci-driver.c

@@ -420,6 +420,12 @@ static void pci_device_shutdown(struct device *dev)
 	pci_msi_shutdown(pci_dev);
 	pci_msix_shutdown(pci_dev);
 
+	/*
+	 * Turn off Bus Master bit on the device to tell it to not
+	 * continue to do DMA
+	 */
+	pci_disable_device(pci_dev);
+
 	/*
 	 * Devices may be enabled to wake up by runtime PM, but they need not
 	 * be supposed to wake up the system from its "power off" state (e.g.

drivers/pci/pci.c

@@ -22,6 +22,7 @@
 #include <linux/interrupt.h>
 #include <linux/device.h>
 #include <linux/pm_runtime.h>
+#include <asm-generic/pci-bridge.h>
 #include <asm/setup.h>
 #include "pci.h"
 
@@ -3164,18 +3165,12 @@ static int pci_parent_bus_reset(struct pci_dev *dev, int probe)
 	return 0;
 }
 
-static int pci_dev_reset(struct pci_dev *dev, int probe)
+static int __pci_dev_reset(struct pci_dev *dev, int probe)
 {
 	int rc;
 
 	might_sleep();
 
-	if (!probe) {
-		pci_cfg_access_lock(dev);
-		/* block PM suspend, driver probe, etc. */
-		device_lock(&dev->dev);
-	}
-
 	rc = pci_dev_specific_reset(dev, probe);
 	if (rc != -ENOTTY)
 		goto done;
@@ -3194,14 +3189,27 @@ static int pci_dev_reset(struct pci_dev *dev, int probe)
 
 	rc = pci_parent_bus_reset(dev, probe);
 done:
+	return rc;
+}
+
+static int pci_dev_reset(struct pci_dev *dev, int probe)
+{
+	int rc;
+
+	if (!probe) {
+		pci_cfg_access_lock(dev);
+		/* block PM suspend, driver probe, etc. */
+		device_lock(&dev->dev);
+	}
+
+	rc = __pci_dev_reset(dev, probe);
+
 	if (!probe) {
 		device_unlock(&dev->dev);
 		pci_cfg_access_unlock(dev);
 	}
-
 	return rc;
 }
-
 /**
  * __pci_reset_function - reset a PCI device function
  * @dev: PCI device to reset
@@ -3246,7 +3254,7 @@ EXPORT_SYMBOL_GPL(__pci_reset_function);
  */
 int __pci_reset_function_locked(struct pci_dev *dev)
 {
-	return pci_dev_reset(dev, 1);
+	return __pci_dev_reset(dev, 0);
 }
 EXPORT_SYMBOL_GPL(__pci_reset_function_locked);
 
@@ -3893,6 +3901,8 @@ static int __init pci_setup(char *str)
 				pcie_bus_config = PCIE_BUS_PERFORMANCE;
 			} else if (!strncmp(str, "pcie_bus_peer2peer", 18)) {
 				pcie_bus_config = PCIE_BUS_PEER2PEER;
+			} else if (!strncmp(str, "pcie_scan_all", 13)) {
+				pci_add_flags(PCI_SCAN_ALL_PCIE_DEVS);
 			} else {
 				printk(KERN_ERR "PCI: Unknown option `%s'\n",
 						str);

drivers/pci/pcie/portdrv_core.c

@@ -249,7 +249,7 @@ static int get_port_device_capability(struct pci_dev *dev)
 	int services = 0, pos;
 	u16 reg16;
 	u32 reg32;
-	int cap_mask;
+	int cap_mask = 0;
 	int err;
 
 	if (pcie_ports_disabled)

drivers/pci/probe.c

@@ -10,18 +10,16 @@
 #include <linux/module.h>
 #include <linux/cpumask.h>
 #include <linux/pci-aspm.h>
+#include <asm-generic/pci-bridge.h>
 #include "pci.h"
 
 #define CARDBUS_LATENCY_TIMER	176	/* secondary latency timer */
 #define CARDBUS_RESERVE_BUSNR	3
 
-static LIST_HEAD(pci_host_bridges);
-
 /* Ugh.  Need to stop exporting this to modules. */
 LIST_HEAD(pci_root_buses);
 EXPORT_SYMBOL(pci_root_buses);
 
-
 static int find_anything(struct device *dev, void *data)
 {
 	return 1;
@@ -44,82 +42,6 @@ int no_pci_devices(void)
 }
 EXPORT_SYMBOL(no_pci_devices);
 
-static struct pci_host_bridge *pci_host_bridge(struct pci_dev *dev)
-{
-	struct pci_bus *bus;
-	struct pci_host_bridge *bridge;
-
-	bus = dev->bus;
-	while (bus->parent)
-		bus = bus->parent;
-
-	list_for_each_entry(bridge, &pci_host_bridges, list) {
-		if (bridge->bus == bus)
-			return bridge;
-	}
-
-	return NULL;
-}
-
-static bool resource_contains(struct resource *res1, struct resource *res2)
-{
-	return res1->start <= res2->start && res1->end >= res2->end;
-}
-
-void pcibios_resource_to_bus(struct pci_dev *dev, struct pci_bus_region *region,
-			     struct resource *res)
-{
-	struct pci_host_bridge *bridge = pci_host_bridge(dev);
-	struct pci_host_bridge_window *window;
-	resource_size_t offset = 0;
-
-	list_for_each_entry(window, &bridge->windows, list) {
-		if (resource_type(res) != resource_type(window->res))
-			continue;
-
-		if (resource_contains(window->res, res)) {
-			offset = window->offset;
-			break;
-		}
-	}
-
-	region->start = res->start - offset;
-	region->end = res->end - offset;
-}
-EXPORT_SYMBOL(pcibios_resource_to_bus);
-
-static bool region_contains(struct pci_bus_region *region1,
-			    struct pci_bus_region *region2)
-{
-	return region1->start <= region2->start && region1->end >= region2->end;
-}
-
-void pcibios_bus_to_resource(struct pci_dev *dev, struct resource *res,
-			     struct pci_bus_region *region)
-{
-	struct pci_host_bridge *bridge = pci_host_bridge(dev);
-	struct pci_host_bridge_window *window;
-	struct pci_bus_region bus_region;
-	resource_size_t offset = 0;
-
-	list_for_each_entry(window, &bridge->windows, list) {
-		if (resource_type(res) != resource_type(window->res))
-			continue;
-
-		bus_region.start = window->res->start - window->offset;
-		bus_region.end = window->res->end - window->offset;
-
-		if (region_contains(&bus_region, region)) {
-			offset = window->offset;
-			break;
-		}
-	}
-
-	res->start = region->start + offset;
-	res->end = region->end + offset;
-}
-EXPORT_SYMBOL(pcibios_bus_to_resource);
-
 /*
  * PCI Bus Class
  */
@@ -501,6 +423,19 @@ static struct pci_bus * pci_alloc_bus(void)
 	return b;
 }
 
+static struct pci_host_bridge *pci_alloc_host_bridge(struct pci_bus *b)
+{
+	struct pci_host_bridge *bridge;
+
+	bridge = kzalloc(sizeof(*bridge), GFP_KERNEL);
+	if (bridge) {
+		INIT_LIST_HEAD(&bridge->windows);
+		bridge->bus = b;
+	}
+
+	return bridge;
+}
+
 static unsigned char pcix_bus_speed[] = {
 	PCI_SPEED_UNKNOWN,		/* 0 */
 	PCI_SPEED_66MHz_PCIX,		/* 1 */
@@ -1201,7 +1136,14 @@ int pci_cfg_space_size(struct pci_dev *dev)
 
 static void pci_release_bus_bridge_dev(struct device *dev)
 {
-	kfree(dev);
+	struct pci_host_bridge *bridge = to_pci_host_bridge(dev);
+
+	if (bridge->release_fn)
+		bridge->release_fn(bridge);
+
+	pci_free_resource_list(&bridge->windows);
+
+	kfree(bridge);
 }
 
 struct pci_dev *alloc_pci_dev(void)
@@ -1395,10 +1337,13 @@ static unsigned no_next_fn(struct pci_dev *dev, unsigned fn)
 static int only_one_child(struct pci_bus *bus)
 {
 	struct pci_dev *parent = bus->self;
+
 	if (!parent || !pci_is_pcie(parent))
 		return 0;
-	if (parent->pcie_type == PCI_EXP_TYPE_ROOT_PORT ||
-	    parent->pcie_type == PCI_EXP_TYPE_DOWNSTREAM)
+	if (parent->pcie_type == PCI_EXP_TYPE_ROOT_PORT)
+		return 1;
+	if (parent->pcie_type == PCI_EXP_TYPE_DOWNSTREAM &&
+	    !pci_has_flag(PCI_SCAN_ALL_PCIE_DEVS))
 		return 1;
 	return 0;
 }
@@ -1650,28 +1595,19 @@ struct pci_bus *pci_create_root_bus(struct device *parent, int bus,
 	int error;
 	struct pci_host_bridge *bridge;
 	struct pci_bus *b, *b2;
-	struct device *dev;
 	struct pci_host_bridge_window *window, *n;
 	struct resource *res;
 	resource_size_t offset;
 	char bus_addr[64];
 	char *fmt;
 
-	bridge = kzalloc(sizeof(*bridge), GFP_KERNEL);
-	if (!bridge)
-		return NULL;
 
 	b = pci_alloc_bus();
 	if (!b)
-		goto err_bus;
-
-	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
-	if (!dev)
-		goto err_dev;
+		return NULL;
 
 	b->sysdata = sysdata;
 	b->ops = ops;
-
 	b2 = pci_find_bus(pci_domain_nr(b), bus);
 	if (b2) {
 		/* If we already got to this bus through a different bridge, ignore it */
@@ -1679,13 +1615,17 @@ struct pci_bus *pci_create_root_bus(struct device *parent, int bus,
 		goto err_out;
 	}
 
-	dev->parent = parent;
-	dev->release = pci_release_bus_bridge_dev;
-	dev_set_name(dev, "pci%04x:%02x", pci_domain_nr(b), bus);
-	error = device_register(dev);
+	bridge = pci_alloc_host_bridge(b);
+	if (!bridge)
+		goto err_out;
+
+	bridge->dev.parent = parent;
+	bridge->dev.release = pci_release_bus_bridge_dev;
+	dev_set_name(&bridge->dev, "pci%04x:%02x", pci_domain_nr(b), bus);
+	error = device_register(&bridge->dev);
 	if (error)
-		goto dev_reg_err;
-	b->bridge = get_device(dev);
+		goto bridge_dev_reg_err;
+	b->bridge = get_device(&bridge->dev);
 	device_enable_async_suspend(b->bridge);
 	pci_set_bus_of_node(b);
 
@@ -1704,9 +1644,6 @@ struct pci_bus *pci_create_root_bus(struct device *parent, int bus,
 
 	b->number = b->secondary = bus;
 
-	bridge->bus = b;
-	INIT_LIST_HEAD(&bridge->windows);
-
 	if (parent)
 		dev_info(parent, "PCI host bridge to bus %s\n", dev_name(&b->dev));
 	else
@@ -1732,25 +1669,18 @@ struct pci_bus *pci_create_root_bus(struct device *parent, int bus,
 	}
 
 	down_write(&pci_bus_sem);
-	list_add_tail(&bridge->list, &pci_host_bridges);
 	list_add_tail(&b->node, &pci_root_buses);
 	up_write(&pci_bus_sem);
 
 	return b;
 
 class_dev_reg_err:
-	device_unregister(dev);
-dev_reg_err:
-	down_write(&pci_bus_sem);
-	list_del(&bridge->list);
-	list_del(&b->node);
-	up_write(&pci_bus_sem);
+	put_device(&bridge->dev);
+	device_unregister(&bridge->dev);
+bridge_dev_reg_err:
+	kfree(bridge);
 err_out:
-	kfree(dev);
-err_dev:
 	kfree(b);
-err_bus:
-	kfree(bridge);
 	return NULL;
 }
 

drivers/pci/quirks.c

@@ -3097,16 +3097,74 @@ static int reset_intel_82599_sfp_virtfn(struct pci_dev *dev, int probe)
 	return 0;
 }
 
+#include "../gpu/drm/i915/i915_reg.h"
+#define MSG_CTL			0x45010
+#define NSDE_PWR_STATE		0xd0100
+#define IGD_OPERATION_TIMEOUT	10000     /* set timeout 10 seconds */
+
+static int reset_ivb_igd(struct pci_dev *dev, int probe)
+{
+	void __iomem *mmio_base;
+	unsigned long timeout;
+	u32 val;
+
+	if (probe)
+		return 0;
+
+	mmio_base = pci_iomap(dev, 0, 0);
+	if (!mmio_base)
+		return -ENOMEM;
+
+	iowrite32(0x00000002, mmio_base + MSG_CTL);
+
+	/*
+	 * Clobbering SOUTH_CHICKEN2 register is fine only if the next
+	 * driver loaded sets the right bits. However, this's a reset and
+	 * the bits have been set by i915 previously, so we clobber
+	 * SOUTH_CHICKEN2 register directly here.
+	 */
+	iowrite32(0x00000005, mmio_base + SOUTH_CHICKEN2);
+
+	val = ioread32(mmio_base + PCH_PP_CONTROL) & 0xfffffffe;
+	iowrite32(val, mmio_base + PCH_PP_CONTROL);
+
+	timeout = jiffies + msecs_to_jiffies(IGD_OPERATION_TIMEOUT);
+	do {
+		val = ioread32(mmio_base + PCH_PP_STATUS);
+		if ((val & 0xb0000000) == 0)
+			goto reset_complete;
+		msleep(10);
+	} while (time_before(jiffies, timeout));
+	dev_warn(&dev->dev, "timeout during reset\n");
+
+reset_complete:
+	iowrite32(0x00000002, mmio_base + NSDE_PWR_STATE);
+
+	pci_iounmap(dev, mmio_base);
+	return 0;
+}
+
 #define PCI_DEVICE_ID_INTEL_82599_SFP_VF   0x10ed
+#define PCI_DEVICE_ID_INTEL_IVB_M_VGA      0x0156
+#define PCI_DEVICE_ID_INTEL_IVB_M2_VGA     0x0166
 
 static const struct pci_dev_reset_methods pci_dev_reset_methods[] = {
 	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82599_SFP_VF,
 		 reset_intel_82599_sfp_virtfn },
+	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IVB_M_VGA,
+		reset_ivb_igd },
+	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IVB_M2_VGA,
+		reset_ivb_igd },
 	{ PCI_VENDOR_ID_INTEL, PCI_ANY_ID,
 		reset_intel_generic_dev },
 	{ 0 }
 };
 
+/*
+ * These device-specific reset methods are here rather than in a driver
+ * because when a host assigns a device to a guest VM, the host may need
+ * to reset the device but probably doesn't have a driver for it.
+ */
 int pci_dev_specific_reset(struct pci_dev *dev, int probe)
 {
 	const struct pci_dev_reset_methods *i;

include/asm-generic/pci-bridge.h

@@ -30,6 +30,12 @@ enum {
 	PCI_ENABLE_PROC_DOMAINS	= 0x00000010,
 	/* ... except for domain 0 */
 	PCI_COMPAT_DOMAIN_0	= 0x00000020,
+
+	/* PCIe downstream ports are bridges that normally lead to only a
+	 * device 0, but if this is set, we scan all possible devices, not
+	 * just device 0.
+	 */
+	PCI_SCAN_ALL_PCIE_DEVS	= 0x00000040,
 };
 
 #ifdef CONFIG_PCI

include/linux/ioport.h

@@ -223,5 +223,12 @@ extern int
 walk_system_ram_range(unsigned long start_pfn, unsigned long nr_pages,
 		void *arg, int (*func)(unsigned long, unsigned long, void *));
 
+/* True if any part of r1 overlaps r2 */
+static inline bool resource_overlaps(struct resource *r1, struct resource *r2)
+{
+       return (r1->start <= r2->end && r1->end >= r2->start);
+}
+
+
 #endif /* __ASSEMBLY__ */
 #endif	/* _LINUX_IOPORT_H */

include/linux/pci.h

@@ -375,11 +375,18 @@ struct pci_host_bridge_window {
 };
 
 struct pci_host_bridge {
-	struct list_head list;
+	struct device dev;
 	struct pci_bus *bus;		/* root bus */
 	struct list_head windows;	/* pci_host_bridge_windows */
+	void (*release_fn)(struct pci_host_bridge *);
+	void *release_data;
 };
 
+#define	to_pci_host_bridge(n) container_of(n, struct pci_host_bridge, dev)
+void pci_set_host_bridge_release(struct pci_host_bridge *bridge,
+		     void (*release_fn)(struct pci_host_bridge *),
+		     void *release_data);
+
 /*
  * The first PCI_BRIDGE_RESOURCE_NUM PCI bus resources (those that correspond
  * to P2P or CardBus bridge windows) go in a table.  Additional ones (for