---

yaml
---
r: 343275
b: refs/heads/master
c: edb1daa
h: refs/heads/master
i:
  343273: cc23fd4
  343271: 27352ba
v: v3

[refs]

@@ -1,2 +1,2 @@
 ---
-refs/heads/master: 967577b062417b4e4b8e27b711220f4124f5153a
+refs/heads/master: edb1daab8e91338b7e2a6c41faec695891ccda35

trunk/Documentation/ABI/testing/sysfs-bus-pci

@@ -222,3 +222,37 @@ Description:
 		satisfied too.  Reading this attribute will show the current
 		value of d3cold_allowed bit.  Writing this attribute will set
 		the value of d3cold_allowed bit.
+
+What:		/sys/bus/pci/devices/.../sriov_totalvfs
+Date:		November 2012
+Contact:	Donald Dutile <ddutile@redhat.com>
+Description:
+		This file appears when a physical PCIe device supports SR-IOV.
+		Userspace applications can read this file to determine the
+		maximum number of Virtual Functions (VFs) a PCIe physical
+		function (PF) can support. Typically, this is the value reported
+		in the PF's SR-IOV extended capability structure's TotalVFs
+		element.  Drivers have the ability at probe time to reduce the
+		value read from this file via the pci_sriov_set_totalvfs()
+		function.
+
+What:		/sys/bus/pci/devices/.../sriov_numvfs
+Date:		November 2012
+Contact:	Donald Dutile <ddutile@redhat.com>
+Description:
+		This file appears when a physical PCIe device supports SR-IOV.
+		Userspace applications can read and write to this file to
+		determine and control the enablement or disablement of Virtual
+		Functions (VFs) on the physical function (PF). A read of this
+		file will return the number of VFs that are enabled on this PF.
+		A number written to this file will enable the specified
+		number of VFs. A userspace application would typically read the
+		file and check that the value is zero, and then write the number
+		of VFs that should be enabled on the PF; the value written
+		should be less than or equal to the value in the sriov_totalvfs
+		file. A userspace application wanting to disable the VFs would
+		write a zero to this file. The core ensures that valid values
+		are written to this file, and returns errors when values are not
+		valid.  For example, writing a 2 to this file when sriov_numvfs
+		is not 0 and not 2 already will return an error. Writing a 10
+		when the value of sriov_totalvfs is 8 will return an error.

trunk/Documentation/PCI/pci-iov-howto.txt

@@ -2,6 +2,9 @@
 		Copyright (C) 2009 Intel Corporation
 		    Yu Zhao <yu.zhao@intel.com>
 
+		Update: November 2012
+			-- sysfs-based SRIOV enable-/disable-ment
+		Donald Dutile <ddutile@redhat.com>
 
 1. Overview
 
@@ -24,10 +27,21 @@ real existing PCI device.
 
 2.1 How can I enable SR-IOV capability
 
-The device driver (PF driver) will control the enabling and disabling
-of the capability via API provided by SR-IOV core. If the hardware
-has SR-IOV capability, loading its PF driver would enable it and all
-VFs associated with the PF.
+Multiple methods are available for SR-IOV enablement.
+In the first method, the device driver (PF driver) will control the
+enabling and disabling of the capability via API provided by SR-IOV core.
+If the hardware has SR-IOV capability, loading its PF driver would
+enable it and all VFs associated with the PF.  Some PF drivers require
+a module parameter to be set to determine the number of VFs to enable.
+In the second method, a write to the sysfs file sriov_numvfs will
+enable and disable the VFs associated with a PCIe PF.  This method
+enables per-PF, VF enable/disable values versus the first method,
+which applies to all PFs of the same device.  Additionally, the
+PCI SRIOV core support ensures that enable/disable operations are
+valid to reduce duplication in multiple drivers for the same
+checks, e.g., check numvfs == 0 if enabling VFs, ensure
+numvfs <= totalvfs.
+The second method is the recommended method for new/future VF devices.
 
 2.2 How can I use the Virtual Functions
 
@@ -40,13 +54,22 @@ requires device driver that is same as a normal PCI device's.
 3.1 SR-IOV API
 
 To enable SR-IOV capability:
+(a) For the first method, in the driver:
 	int pci_enable_sriov(struct pci_dev *dev, int nr_virtfn);
 	'nr_virtfn' is number of VFs to be enabled.
+(b) For the second method, from sysfs:
+	echo 'nr_virtfn' > \
+        /sys/bus/pci/devices/<DOMAIN:BUS:DEVICE.FUNCTION>/sriov_numvfs
 
 To disable SR-IOV capability:
+(a) For the first method, in the driver:
 	void pci_disable_sriov(struct pci_dev *dev);
+(b) For the second method, from sysfs:
+	echo  0 > \
+        /sys/bus/pci/devices/<DOMAIN:BUS:DEVICE.FUNCTION>/sriov_numvfs
 
 To notify SR-IOV core of Virtual Function Migration:
+(a) In the driver:
 	irqreturn_t pci_sriov_migration(struct pci_dev *dev);
 
 3.2 Usage example
@@ -88,6 +111,22 @@ static void dev_shutdown(struct pci_dev *dev)
 	...
 }
 
+static int dev_sriov_configure(struct pci_dev *dev, int numvfs)
+{
+	if (numvfs > 0) {
+		...
+		pci_enable_sriov(dev, numvfs);
+		...
+		return numvfs;
+	}
+	if (numvfs == 0) {
+		....
+		pci_disable_sriov(dev);
+		...
+		return 0;
+	}
+}
+
 static struct pci_driver dev_driver = {
 	.name =		"SR-IOV Physical Function driver",
 	.id_table =	dev_id_table,
@@ -96,4 +135,5 @@ static struct pci_driver dev_driver = {
 	.suspend =	dev_suspend,
 	.resume =	dev_resume,
 	.shutdown =	dev_shutdown,
+	.sriov_configure = dev_sriov_configure,
 };

trunk/arch/x86/pci/acpi.c

@@ -12,6 +12,7 @@ struct pci_root_info {
 	char name[16];
 	unsigned int res_num;
 	struct resource *res;
+	resource_size_t *res_offset;
 	struct pci_sysdata sd;
 #ifdef	CONFIG_PCI_MMCONFIG
 	bool mcfg_added;
@@ -22,6 +23,7 @@ struct pci_root_info {
 };
 
 static bool pci_use_crs = true;
+static bool pci_ignore_seg = false;
 
 static int __init set_use_crs(const struct dmi_system_id *id)
 {
@@ -35,7 +37,14 @@ static int __init set_nouse_crs(const struct dmi_system_id *id)
 	return 0;
 }
 
-static const struct dmi_system_id pci_use_crs_table[] __initconst = {
+static int __init set_ignore_seg(const struct dmi_system_id *id)
+{
+	printk(KERN_INFO "PCI: %s detected: ignoring ACPI _SEG\n", id->ident);
+	pci_ignore_seg = true;
+	return 0;
+}
+
+static const struct dmi_system_id pci_crs_quirks[] __initconst = {
 	/* http://bugzilla.kernel.org/show_bug.cgi?id=14183 */
 	{
 		.callback = set_use_crs,
@@ -98,6 +107,16 @@ static const struct dmi_system_id pci_use_crs_table[] __initconst = {
 			DMI_MATCH(DMI_BIOS_VERSION, "6JET85WW (1.43 )"),
 		},
 	},
+
+	/* https://bugzilla.kernel.org/show_bug.cgi?id=15362 */
+	{
+		.callback = set_ignore_seg,
+		.ident = "HP xw9300",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP xw9300 Workstation"),
+		},
+	},
 	{}
 };
 
@@ -108,7 +127,7 @@ void __init pci_acpi_crs_quirks(void)
 	if (dmi_get_date(DMI_BIOS_DATE, &year, NULL, NULL) && year < 2008)
 		pci_use_crs = false;
 
-	dmi_check_system(pci_use_crs_table);
+	dmi_check_system(pci_crs_quirks);
 
 	/*
 	 * If the user specifies "pci=use_crs" or "pci=nocrs" explicitly, that
@@ -305,6 +324,7 @@ setup_resource(struct acpi_resource *acpi_res, void *data)
 	res->flags = flags;
 	res->start = start;
 	res->end = end;
+	info->res_offset[info->res_num] = addr.translation_offset;
 
 	if (!pci_use_crs) {
 		dev_printk(KERN_DEBUG, &info->bridge->dev,
@@ -374,14 +394,17 @@ 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(resources, res);
+			pci_add_resource_offset(resources, res,
+					info->res_offset[i]);
 	}
 }
 
 static void free_pci_root_info_res(struct pci_root_info *info)
 {
 	kfree(info->res);
 	info->res = NULL;
+	kfree(info->res_offset);
+	info->res_offset = NULL;
 	info->res_num = 0;
 }
 
@@ -432,10 +455,20 @@ probe_pci_root_info(struct pci_root_info *info, struct acpi_device *device,
 		return;
 
 	size = sizeof(*info->res) * info->res_num;
-	info->res_num = 0;
 	info->res = kzalloc(size, GFP_KERNEL);
-	if (!info->res)
+	if (!info->res) {
+		info->res_num = 0;
+		return;
+	}
+
+	size = sizeof(*info->res_offset) * info->res_num;
+	info->res_num = 0;
+	info->res_offset = kzalloc(size, GFP_KERNEL);
+	if (!info->res_offset) {
+		kfree(info->res);
+		info->res = NULL;
 		return;
+	}
 
 	acpi_walk_resources(device->handle, METHOD_NAME__CRS, setup_resource,
 				info);
@@ -455,6 +488,9 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_pci_root *root)
 	int pxm;
 #endif
 
+	if (pci_ignore_seg)
+		domain = 0;
+
 	if (domain && !pci_domains_supported) {
 		printk(KERN_WARNING "pci_bus %04x:%02x: "
 		       "ignored (multiple domains not supported)\n",

trunk/arch/x86/pci/common.c

@@ -626,7 +626,7 @@ void pcibios_disable_device (struct pci_dev *dev)
 		pcibios_disable_irq(dev);
 }
 
-int pci_ext_cfg_avail(struct pci_dev *dev)
+int pci_ext_cfg_avail(void)
 {
 	if (raw_pci_ext_ops)
 		return 1;

trunk/drivers/acpi/pci_bind.c

@@ -45,11 +45,12 @@ static int acpi_pci_unbind(struct acpi_device *device)
 
 	device_set_run_wake(&dev->dev, false);
 	pci_acpi_remove_pm_notifier(device);
+	acpi_power_resource_unregister_device(&dev->dev, device->handle);
 
 	if (!dev->subordinate)
 		goto out;
 
-	acpi_pci_irq_del_prt(dev->subordinate);
+	acpi_pci_irq_del_prt(pci_domain_nr(dev->bus), dev->subordinate->number);
 
 	device->ops.bind = NULL;
 	device->ops.unbind = NULL;
@@ -63,14 +64,15 @@ static int acpi_pci_bind(struct acpi_device *device)
 {
 	acpi_status status;
 	acpi_handle handle;
-	struct pci_bus *bus;
+	unsigned char bus;
 	struct pci_dev *dev;
 
 	dev = acpi_get_pci_dev(device->handle);
 	if (!dev)
 		return 0;
 
 	pci_acpi_add_pm_notifier(device, dev);
+	acpi_power_resource_register_device(&dev->dev, device->handle);
 	if (device->wakeup.flags.run_wake)
 		device_set_run_wake(&dev->dev, true);
 
@@ -100,11 +102,11 @@ static int acpi_pci_bind(struct acpi_device *device)
 		goto out;
 
 	if (dev->subordinate)
-		bus = dev->subordinate;
+		bus = dev->subordinate->number;
 	else
-		bus = dev->bus;
+		bus = dev->bus->number;
 
-	acpi_pci_irq_add_prt(device->handle, bus);
+	acpi_pci_irq_add_prt(device->handle, pci_domain_nr(dev->bus), bus);
 
 out:
 	pci_dev_put(dev);

trunk/drivers/acpi/pci_irq.c

@@ -184,7 +184,7 @@ static void do_prt_fixups(struct acpi_prt_entry *entry,
 	}
 }
 
-static int acpi_pci_irq_add_entry(acpi_handle handle, struct pci_bus *bus,
+static int acpi_pci_irq_add_entry(acpi_handle handle, int segment, int bus,
 				  struct acpi_pci_routing_table *prt)
 {
 	struct acpi_prt_entry *entry;
@@ -198,8 +198,8 @@ static int acpi_pci_irq_add_entry(acpi_handle handle, struct pci_bus *bus,
 	 * 1=INTA, 2=INTB.  We use the PCI encoding throughout, so convert
 	 * it here.
 	 */
-	entry->id.segment = pci_domain_nr(bus);
-	entry->id.bus = bus->number;
+	entry->id.segment = segment;
+	entry->id.bus = bus;
 	entry->id.device = (prt->address >> 16) & 0xFFFF;
 	entry->pin = prt->pin + 1;
 
@@ -244,7 +244,7 @@ static int acpi_pci_irq_add_entry(acpi_handle handle, struct pci_bus *bus,
 	return 0;
 }
 
-int acpi_pci_irq_add_prt(acpi_handle handle, struct pci_bus *bus)
+int acpi_pci_irq_add_prt(acpi_handle handle, int segment, int bus)
 {
 	acpi_status status;
 	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
@@ -273,7 +273,7 @@ int acpi_pci_irq_add_prt(acpi_handle handle, struct pci_bus *bus)
 
 	entry = buffer.pointer;
 	while (entry && (entry->length > 0)) {
-		acpi_pci_irq_add_entry(handle, bus, entry);
+		acpi_pci_irq_add_entry(handle, segment, bus, entry);
 		entry = (struct acpi_pci_routing_table *)
 		    ((unsigned long)entry + entry->length);
 	}
@@ -282,17 +282,16 @@ int acpi_pci_irq_add_prt(acpi_handle handle, struct pci_bus *bus)
 	return 0;
 }
 
-void acpi_pci_irq_del_prt(struct pci_bus *bus)
+void acpi_pci_irq_del_prt(int segment, int bus)
 {
 	struct acpi_prt_entry *entry, *tmp;
 
 	printk(KERN_DEBUG
 	       "ACPI: Delete PCI Interrupt Routing Table for %04x:%02x\n",
-	       pci_domain_nr(bus), bus->number);
+	       segment, bus);
 	spin_lock(&acpi_prt_lock);
 	list_for_each_entry_safe(entry, tmp, &acpi_prt_list, list) {
-		if (pci_domain_nr(bus) == entry->id.segment
-			&& bus->number == entry->id.bus) {
+		if (segment == entry->id.segment && bus == entry->id.bus) {
 			list_del(&entry->list);
 			kfree(entry);
 		}