Re: Xen 3.4 code freeze

["Zhao, Yu" <yu.zhao@intel.com>]

[-- Attachment #1: Type: text/plain, Size: 707 bytes --]

Hi Keir,

There are some native kernel SR-IOV patches that would be accepted by 
the maintainer soon, and I planed to backport them to Xen/Dom0 then. 
However, it looks they can't be in the native kernel tree before the 
code freeze. I attached the backported patches, if it's possible, can 
you please take them for 3.4?

Thanks,
Yu

Keir Fraser wrote:
> Folks,
> 
> My plan is to take no further large feature patchsets into xen-unstable until 3.4 is branched, the only exception possibly being Dan’s tmem patches. Furthermore, after Friday I’m away for a week and when I get back I intend to shake the trees into better shape and accept bug-fix patches only.
> 
>  -- Keir
> 
> 


[-- Attachment #2: xen.patch --]
[-- Type: text/plain, Size: 6153 bytes --]

# HG changeset patch
# User Yu Zhao <yu.zhao@intel.com>
# Date 1237366099 14400
# Node ID cdf29bb6c74d5e075099855e9ddffb27e633079a
# Parent  9fc957e63f8dc0fdb2400eb424da4c1122b7ac65
Xen: use proper device ID to search VT-d unit for ARI and SR-IOV device

PCIe Alternative Routing-ID Interpretation (ARI) ECN defines the Extended
Function -- a function whose function number is greater than 7 within an
ARI Device. Intel VT-d spec 1.2 section 8.3.2 specifies that the Extended
Function is under the scope of the same remapping unit as the traditional
function. The hypervisor needs to know if a function is Extended Function
so it can find proper DMAR for it.

And section 8.3.3 specifies that the SR-IOV Virtual Function is under the
scope of the same remapping unit as the Physical Function. The hypervisor
also needs to know if a function is the Virtual Function and which Physical
Function it's associated with for same reason.

diff -r 9fc957e63f8d -r cdf29bb6c74d xen/arch/ia64/xen/hypercall.c
--- a/xen/arch/ia64/xen/hypercall.c	Tue Mar 17 15:40:25 2009 +0000
+++ b/xen/arch/ia64/xen/hypercall.c	Wed Mar 18 04:48:19 2009 -0400
@@ -650,6 +650,7 @@
 
     case PHYSDEVOP_manage_pci_add: {
         struct physdev_manage_pci manage_pci;
+        struct pci_dev dev;
         ret = -EPERM;
         if ( !IS_PRIV(current->domain) )
             break;
@@ -657,7 +658,13 @@
         if ( copy_from_guest(&manage_pci, arg, 1) != 0 )
             break;
 
-        ret = pci_add_device(manage_pci.bus, manage_pci.devfn);
+        dev.bus = manage_pci.bus;
+        dev.devfn = manage_pci.devfn;
+        dev.is_extfn = manage_pci.is_extfn;
+        dev.is_virtfn = manage_pci.is_virtfn;
+        dev.physfn.bus = manage_pci.physfn.bus;
+        dev.physfn.devfn = manage_pci.physfn.devfn;
+        ret = pci_add_device(&dev);
             break;
     }
 
diff -r 9fc957e63f8d -r cdf29bb6c74d xen/arch/x86/physdev.c
--- a/xen/arch/x86/physdev.c	Tue Mar 17 15:40:25 2009 +0000
+++ b/xen/arch/x86/physdev.c	Wed Mar 18 04:48:19 2009 -0400
@@ -397,6 +397,7 @@
 
     case PHYSDEVOP_manage_pci_add: {
         struct physdev_manage_pci manage_pci;
+        struct pci_dev dev;
         ret = -EPERM;
         if ( !IS_PRIV(v->domain) )
             break;
@@ -404,7 +405,13 @@
         if ( copy_from_guest(&manage_pci, arg, 1) != 0 )
             break;
 
-        ret = pci_add_device(manage_pci.bus, manage_pci.devfn);
+        dev.bus = manage_pci.bus;
+        dev.devfn = manage_pci.devfn;
+        dev.is_extfn = manage_pci.is_extfn;
+        dev.is_virtfn = manage_pci.is_virtfn;
+        dev.physfn.bus = manage_pci.physfn.bus;
+        dev.physfn.devfn = manage_pci.physfn.devfn;
+        ret = pci_add_device(&dev);
         break;
     }
 
diff -r 9fc957e63f8d -r cdf29bb6c74d xen/drivers/passthrough/pci.c
--- a/xen/drivers/passthrough/pci.c	Tue Mar 17 15:40:25 2009 +0000
+++ b/xen/drivers/passthrough/pci.c	Wed Mar 18 04:48:19 2009 -0400
@@ -43,8 +43,8 @@
         return NULL;
     memset(pdev, 0, sizeof(struct pci_dev));
 
-    *((u8*) &pdev->bus) = bus;
-    *((u8*) &pdev->devfn) = devfn;
+    pdev->bus = bus;
+    pdev->devfn = devfn;
     pdev->domain = NULL;
     INIT_LIST_HEAD(&pdev->msi_list);
     list_add(&pdev->alldevs_list, &alldevs_list);
@@ -92,15 +92,20 @@
     return NULL;
 }
 
-int pci_add_device(u8 bus, u8 devfn)
+int pci_add_device(struct pci_dev *dev)
 {
     struct pci_dev *pdev;
     int ret = -ENOMEM;
 
     spin_lock(&pcidevs_lock);
-    pdev = alloc_pdev(bus, devfn);
+    pdev = alloc_pdev(dev->bus, dev->devfn);
     if ( !pdev )
         goto out;
+
+    pdev->is_extfn = dev->is_extfn;
+    pdev->is_virtfn = dev->is_virtfn;
+    pdev->physfn.bus = dev->physfn.bus;
+    pdev->physfn.devfn = dev->physfn.devfn;
 
     ret = 0;
     if ( !pdev->domain )
@@ -115,8 +120,8 @@
 
 out:
     spin_unlock(&pcidevs_lock);
-    printk(XENLOG_DEBUG "PCI add device %02x:%02x.%x\n", bus,
-           PCI_SLOT(devfn), PCI_FUNC(devfn));
+    printk(XENLOG_DEBUG "PCI add device %02x:%02x.%x\n", dev->bus,
+           PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
     return ret;
 }
 
diff -r 9fc957e63f8d -r cdf29bb6c74d xen/drivers/passthrough/vtd/dmar.c
--- a/xen/drivers/passthrough/vtd/dmar.c	Tue Mar 17 15:40:25 2009 +0000
+++ b/xen/drivers/passthrough/vtd/dmar.c	Wed Mar 18 04:48:19 2009 -0400
@@ -157,6 +157,17 @@
     struct acpi_drhd_unit *drhd;
     struct acpi_drhd_unit *found = NULL, *include_all = NULL;
     int i;
+    struct pci_dev *dev;
+
+    dev = pci_get_pdev(bus, devfn);
+    BUG_ON(!dev);
+
+    if (dev->is_extfn) {
+        devfn = 0;
+    } else if (dev->is_virtfn) {
+        bus = dev->physfn.bus;
+        devfn = PCI_SLOT(dev->physfn.devfn) ? 0 : dev->physfn.devfn;
+    }
 
     list_for_each_entry ( drhd, &acpi_drhd_units, list )
     {
diff -r 9fc957e63f8d -r cdf29bb6c74d xen/include/public/physdev.h
--- a/xen/include/public/physdev.h	Tue Mar 17 15:40:25 2009 +0000
+++ b/xen/include/public/physdev.h	Wed Mar 18 04:48:19 2009 -0400
@@ -178,6 +178,12 @@
     /* IN */
     uint8_t bus;
     uint8_t devfn;
+    unsigned is_extfn:1;
+    unsigned is_virtfn:1;
+    struct {
+        uint8_t bus;
+        uint8_t devfn;
+    } physfn;
 }; 
 
 typedef struct physdev_manage_pci physdev_manage_pci_t;
diff -r 9fc957e63f8d -r cdf29bb6c74d xen/include/xen/pci.h
--- a/xen/include/xen/pci.h	Tue Mar 17 15:40:25 2009 +0000
+++ b/xen/include/xen/pci.h	Wed Mar 18 04:48:19 2009 -0400
@@ -41,8 +41,14 @@
     spinlock_t msix_table_lock;
 
     struct domain *domain;
-    const u8 bus;
-    const u8 devfn;
+    u8 bus;
+    u8 devfn;
+    unsigned is_extfn:1;
+    unsigned is_virtfn:1;
+    struct {
+        u8 bus;
+        u8 devfn;
+    } physfn;
 };
 
 #define for_each_pdev(domain, pdev) \
@@ -62,7 +68,7 @@
 struct pci_dev *pci_lock_domain_pdev(struct domain *d, int bus, int devfn);
 
 void pci_release_devices(struct domain *d);
-int pci_add_device(u8 bus, u8 devfn);
+int pci_add_device(struct pci_dev *dev);
 int pci_remove_device(u8 bus, u8 devfn);
 struct pci_dev *pci_get_pdev(int bus, int devfn);
 struct pci_dev *pci_get_pdev_by_domain(struct domain *d, int bus, int devfn);

[-- Attachment #3: dom0-1.patch --]
[-- Type: text/plain, Size: 12451 bytes --]

# HG changeset patch
# User Yu Zhao <yu.zhao@intel.com>
# Date 1237268585 14400
# Node ID 92730fa710446b2502809faa72bb29fda95ba878
# Parent  e8a9f8910a3f113759906e493eaa211e2c43cd85
PCI: initialize and release SR-IOV capability

If a device has the SR-IOV capability, initialize it (set the ARI
Capable Hierarchy in the lowest numbered PF if necessary; calculate
the System Page Size for the VF MMIO, probe the VF Offset, Stride
and BARs). A lock for the VF bus allocation is also initialized if
a PF is the lowest numbered PF.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>

diff -r e8a9f8910a3f -r 92730fa71044 drivers/pci/Kconfig
--- a/drivers/pci/Kconfig	Fri Mar 13 10:08:22 2009 +0000
+++ b/drivers/pci/Kconfig	Tue Mar 17 01:43:05 2009 -0400
@@ -37,3 +37,12 @@
 	help
 	  Say Y here if you want to reserve PCI device for passthrough.
 
+config PCI_IOV
+	bool "PCI IOV support"
+	depends on PCI
+	help
+	  PCI-SIG I/O Virtualization (IOV) Specifications support.
+	  Single Root IOV: allows the creation of virtual PCI devices
+	  that share the physical resources from a real device.
+
+	  When in doubt, say N.
diff -r e8a9f8910a3f -r 92730fa71044 drivers/pci/Makefile
--- a/drivers/pci/Makefile	Fri Mar 13 10:08:22 2009 +0000
+++ b/drivers/pci/Makefile	Tue Mar 17 01:43:05 2009 -0400
@@ -15,6 +15,8 @@
 
 # Build the PCI Hotplug drivers if we were asked to
 obj-$(CONFIG_HOTPLUG_PCI) += hotplug/
+
+obj-$(CONFIG_PCI_IOV) += iov.o
 
 #
 # Some architectures use the generic PCI setup functions
diff -r e8a9f8910a3f -r 92730fa71044 drivers/pci/iov.c
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/drivers/pci/iov.c	Tue Mar 17 01:43:05 2009 -0400
@@ -0,0 +1,175 @@
+/*
+ * drivers/pci/iov.c
+ *
+ * Copyright (C) 2009 Intel Corporation, Yu Zhao <yu.zhao@intel.com>
+ *
+ * PCI Express I/O Virtualization (IOV) support.
+ *   Single Root IOV 1.0
+ */
+
+#include <linux/pci.h>
+#include <linux/mutex.h>
+#include <linux/string.h>
+#include <linux/delay.h>
+#include "pci.h"
+
+
+static int sriov_init(struct pci_dev *dev, int pos)
+{
+	int i;
+	int rc;
+	int nres;
+	u32 pgsz;
+	u16 ctrl, total, offset, stride;
+	struct pci_sriov *iov;
+	struct resource *res;
+	struct pci_dev *pdev;
+
+	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);
+		ssleep(1);
+	}
+
+	pci_read_config_word(dev, pos + PCI_SRIOV_TOTAL_VF, &total);
+	if (!total)
+		return 0;
+
+	list_for_each_entry(pdev, &dev->bus->devices, bus_list)
+		if (pdev->is_physfn)
+			break;
+	if (list_empty(&dev->bus->devices) || !pdev->is_physfn)
+		pdev = NULL;
+
+	ctrl = 0;
+	if (!pdev && pci_ari_enabled(dev->bus))
+		ctrl |= PCI_SRIOV_CTRL_ARI;
+
+	pci_write_config_word(dev, pos + PCI_SRIOV_CTRL, ctrl);
+	pci_write_config_word(dev, pos + PCI_SRIOV_NUM_VF, total);
+	pci_read_config_word(dev, pos + PCI_SRIOV_VF_OFFSET, &offset);
+	pci_read_config_word(dev, pos + PCI_SRIOV_VF_STRIDE, &stride);
+	if (!offset || (total > 1 && !stride))
+		return -EIO;
+
+	pci_read_config_dword(dev, pos + PCI_SRIOV_SUP_PGSIZE, &pgsz);
+	i = PAGE_SHIFT > 12 ? PAGE_SHIFT - 12 : 0;
+	pgsz &= ~((1 << i) - 1);
+	if (!pgsz)
+		return -EIO;
+
+	pgsz &= ~(pgsz - 1);
+	pci_write_config_dword(dev, pos + PCI_SRIOV_SYS_PGSIZE, pgsz);
+
+	nres = 0;
+	for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) {
+		res = dev->resource + PCI_IOV_RESOURCES + i;
+		i += __pci_read_base(dev, pci_bar_unknown, res,
+				     pos + PCI_SRIOV_BAR + i * 4);
+		if (!res->flags)
+			continue;
+		if ((res->end - res->start + 1) & (PAGE_SIZE - 1)) {
+			rc = -EIO;
+			goto failed;
+		}
+		res->end = res->start + (res->end - res->start + 1) * total - 1;
+		nres++;
+	}
+
+	iov = kzalloc(sizeof(*iov), GFP_KERNEL);
+	if (!iov) {
+		rc = -ENOMEM;
+		goto failed;
+	}
+
+	iov->pos = pos;
+	iov->nres = nres;
+	iov->ctrl = ctrl;
+	iov->total = total;
+	iov->offset = offset;
+	iov->stride = stride;
+	iov->pgsz = pgsz;
+	iov->self = dev;
+	pci_read_config_dword(dev, pos + PCI_SRIOV_CAP, &iov->cap);
+	pci_read_config_byte(dev, pos + PCI_SRIOV_FUNC_LINK, &iov->link);
+
+	if (pdev)
+		iov->dev = pci_dev_get(pdev);
+	else {
+		iov->dev = dev;
+		mutex_init(&iov->lock);
+	}
+
+	dev->sriov = iov;
+	dev->is_physfn = 1;
+
+	return 0;
+
+failed:
+	for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) {
+		res = dev->resource + PCI_IOV_RESOURCES + i;
+		res->flags = 0;
+	}
+
+	return rc;
+}
+
+static void sriov_release(struct pci_dev *dev)
+{
+	if (dev == dev->sriov->dev)
+		mutex_destroy(&dev->sriov->lock);
+	else
+		pci_dev_put(dev->sriov->dev);
+
+	kfree(dev->sriov);
+	dev->sriov = NULL;
+}
+
+/**
+ * pci_iov_init - initialize the IOV capability
+ * @dev: the PCI device
+ *
+ * Returns 0 on success, or negative on failure.
+ */
+int pci_iov_init(struct pci_dev *dev)
+{
+	int pos;
+
+	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_SRIOV);
+	if (pos)
+		return sriov_init(dev, pos);
+
+	return -ENODEV;
+}
+
+/**
+ * pci_iov_release - release resources used by the IOV capability
+ * @dev: the PCI device
+ */
+void pci_iov_release(struct pci_dev *dev)
+{
+	if (dev->is_physfn)
+		sriov_release(dev);
+}
+
+/**
+ * pci_iov_resource_bar - get position of the SR-IOV BAR
+ * @dev: the PCI device
+ * @resno: the resource number
+ * @type: the BAR type to be filled in
+ *
+ * Returns position of the BAR encapsulated in the SR-IOV capability.
+ */
+int pci_iov_resource_bar(struct pci_dev *dev, int resno,
+			 enum pci_bar_type *type)
+{
+	if (resno < PCI_IOV_RESOURCES || resno > PCI_IOV_RESOURCE_END)
+		return 0;
+
+	BUG_ON(!dev->is_physfn);
+
+	*type = pci_bar_unknown;
+
+	return dev->sriov->pos + PCI_SRIOV_BAR +
+		4 * (resno - PCI_IOV_RESOURCES);
+}
diff -r e8a9f8910a3f -r 92730fa71044 drivers/pci/pci.c
--- a/drivers/pci/pci.c	Fri Mar 13 10:08:22 2009 +0000
+++ b/drivers/pci/pci.c	Tue Mar 17 01:43:05 2009 -0400
@@ -1048,12 +1048,19 @@
  */
 int pci_resource_bar(struct pci_dev *dev, int resno, enum pci_bar_type *type)
 {
+	int reg;
+
 	if (resno < PCI_ROM_RESOURCE) {
 		*type = pci_bar_unknown;
 		return PCI_BASE_ADDRESS_0 + 4 * resno;
 	} else if (resno == PCI_ROM_RESOURCE) {
 		*type = pci_bar_mem32;
 		return dev->rom_base_reg;
+	} else if (resno < PCI_BRIDGE_RESOURCES) {
+		/* device specific resource */
+		reg = pci_iov_resource_bar(dev, resno, type);
+		if (reg)
+			return reg;
 	}
 
 	dev_err(&dev->dev, "BAR: invalid resource #%d\n", resno);
diff -r e8a9f8910a3f -r 92730fa71044 drivers/pci/pci.h
--- a/drivers/pci/pci.h	Fri Mar 13 10:08:22 2009 +0000
+++ b/drivers/pci/pci.h	Tue Mar 17 01:43:05 2009 -0400
@@ -132,3 +132,40 @@
 {
 	return bus->self && bus->self->ari_enabled;
 }
+
+/* Single Root I/O Virtualization */
+struct pci_sriov {
+	int pos;		/* capability position */
+	int nres;		/* number of resources */
+	u32 cap;		/* SR-IOV Capabilities */
+	u16 ctrl;		/* SR-IOV Control */
+	u16 total;		/* total VFs associated with the PF */
+	u16 offset;		/* first VF Routing ID offset */
+	u16 stride;		/* following VF stride */
+	u32 pgsz;		/* page size for BAR alignment */
+	u8 link;		/* Function Dependency Link */
+	struct pci_dev *dev;	/* lowest numbered PF */
+	struct pci_dev *self;	/* this PF */
+	struct mutex lock;	/* lock for VF bus */
+};
+
+#ifdef CONFIG_PCI_IOV
+extern int pci_iov_init(struct pci_dev *dev);
+extern void pci_iov_release(struct pci_dev *dev);
+extern int pci_iov_resource_bar(struct pci_dev *dev, int resno,
+				enum pci_bar_type *type);
+#else
+static inline int pci_iov_init(struct pci_dev *dev)
+{
+	return -ENODEV;
+}
+static inline void pci_iov_release(struct pci_dev *dev)
+
+{
+}
+static inline int pci_iov_resource_bar(struct pci_dev *dev, int resno,
+				       enum pci_bar_type *type)
+{
+	return 0;
+}
+#endif /* CONFIG_PCI_IOV */
diff -r e8a9f8910a3f -r 92730fa71044 drivers/pci/probe.c
--- a/drivers/pci/probe.c	Fri Mar 13 10:08:22 2009 +0000
+++ b/drivers/pci/probe.c	Tue Mar 17 01:43:05 2009 -0400
@@ -765,6 +765,9 @@
 	struct pci_dev *pci_dev;
 
 	pci_dev = to_pci_dev(dev);
+
+	pci_iov_release(pci_dev);
+
 	kfree(pci_dev);
 }
 
@@ -891,6 +894,9 @@
 
 	/* Alternative Routing-ID Forwarding */
 	pci_enable_ari(dev);
+
+	/* Single Root I/O Virtualization */
+	pci_iov_init(dev);
 
 	/*
 	 * Add the device to our list of discovered devices
diff -r e8a9f8910a3f -r 92730fa71044 include/linux/pci.h
--- a/include/linux/pci.h	Fri Mar 13 10:08:22 2009 +0000
+++ b/include/linux/pci.h	Tue Mar 17 01:43:05 2009 -0400
@@ -77,6 +77,12 @@
 	/* #6: expansion ROM resource */
 	PCI_ROM_RESOURCE,
 
+	/* device specific resources */
+#ifdef CONFIG_PCI_IOV
+	PCI_IOV_RESOURCES,
+	PCI_IOV_RESOURCE_END = PCI_IOV_RESOURCES + PCI_SRIOV_NUM_BARS - 1,
+#endif
+
 	/* resources assigned to buses behind the bridge */
 #define PCI_BRIDGE_RESOURCE_NUM 4
 
@@ -127,6 +133,8 @@
 	char cap_nr;
 	u32 data[0];
 };
+
+struct pci_sriov;
 
 /*
  * The pci_dev structure is used to describe PCI devices.
@@ -189,13 +197,17 @@
 	unsigned int	broken_parity_status:1;	/* Device generates false positive parity */
 	unsigned int 	msi_enabled:1;
 	unsigned int	msix_enabled:1;
+	unsigned int	ari_enabled:1;	/* ARI forwarding */
+	unsigned int	is_physfn:1;
 
 	u32		saved_config_space[16]; /* config space saved at suspend time */
 	struct hlist_head saved_cap_space;
 	struct bin_attribute *rom_attr; /* attribute descriptor for sysfs ROM entry */
 	int rom_attr_enabled;		/* has display of the rom attribute been enabled? */
 	struct bin_attribute *res_attr[DEVICE_COUNT_RESOURCE]; /* sysfs file for resources */
-	unsigned int	ari_enabled:1;	/* ARI forwarding */
+#ifdef CONFIG_PCI_IOV
+	struct pci_sriov *sriov;	/* SR-IOV capability related */
+#endif
 };
 
 #define pci_dev_g(n) list_entry(n, struct pci_dev, global_list)
diff -r e8a9f8910a3f -r 92730fa71044 include/linux/pci_regs.h
--- a/include/linux/pci_regs.h	Fri Mar 13 10:08:22 2009 +0000
+++ b/include/linux/pci_regs.h	Tue Mar 17 01:43:05 2009 -0400
@@ -332,6 +332,7 @@
 #define  PCI_EXP_TYPE_UPSTREAM	0x5	/* Upstream Port */
 #define  PCI_EXP_TYPE_DOWNSTREAM 0x6	/* Downstream Port */
 #define  PCI_EXP_TYPE_PCI_BRIDGE 0x7	/* PCI/PCI-X Bridge */
+#define  PCI_EXP_TYPE_RC_END	0x9	/* Root Complex Integrated Endpoint */
 #define PCI_EXP_FLAGS_SLOT	0x0100	/* Slot implemented */
 #define PCI_EXP_FLAGS_IRQ	0x3e00	/* Interrupt message number */
 #define PCI_EXP_DEVCAP		4	/* Device capabilities */
@@ -393,6 +394,7 @@
 #define PCI_EXT_CAP_ID_DSN	3
 #define PCI_EXT_CAP_ID_PWR	4
 #define PCI_EXT_CAP_ID_ARI	14
+#define PCI_EXT_CAP_ID_SRIOV	16
 
 /* Advanced Error Reporting */
 #define PCI_ERR_UNCOR_STATUS	4	/* Uncorrectable Error Status */
@@ -478,4 +480,35 @@
 #define  PCI_ARI_CTRL_ACS	0x0002	/* ACS Function Groups Enable */
 #define  PCI_ARI_CTRL_FG(x)	(((x) >> 4) & 7) /* Function Group */
 
+/* Single Root I/O Virtualization */
+#define PCI_SRIOV_CAP		0x04	/* SR-IOV Capabilities */
+#define  PCI_SRIOV_CAP_VFM	0x01	/* VF Migration Capable */
+#define  PCI_SRIOV_CAP_INTR(x)	((x) >> 21) /* Interrupt Message Number */
+#define PCI_SRIOV_CTRL		0x08	/* SR-IOV Control */
+#define  PCI_SRIOV_CTRL_VFE	0x01	/* VF Enable */
+#define  PCI_SRIOV_CTRL_VFM	0x02	/* VF Migration Enable */
+#define  PCI_SRIOV_CTRL_INTR	0x04	/* VF Migration Interrupt Enable */
+#define  PCI_SRIOV_CTRL_MSE	0x08	/* VF Memory Space Enable */
+#define  PCI_SRIOV_CTRL_ARI	0x10	/* ARI Capable Hierarchy */
+#define PCI_SRIOV_STATUS	0x0a	/* SR-IOV Status */
+#define  PCI_SRIOV_STATUS_VFM	0x01	/* VF Migration Status */
+#define PCI_SRIOV_INITIAL_VF	0x0c	/* Initial VFs */
+#define PCI_SRIOV_TOTAL_VF	0x0e	/* Total VFs */
+#define PCI_SRIOV_NUM_VF	0x10	/* Number of VFs */
+#define PCI_SRIOV_FUNC_LINK	0x12	/* Function Dependency Link */
+#define PCI_SRIOV_VF_OFFSET	0x14	/* First VF Offset */
+#define PCI_SRIOV_VF_STRIDE	0x16	/* Following VF Stride */
+#define PCI_SRIOV_VF_DID	0x1a	/* VF Device ID */
+#define PCI_SRIOV_SUP_PGSIZE	0x1c	/* Supported Page Sizes */
+#define PCI_SRIOV_SYS_PGSIZE	0x20	/* System Page Size */
+#define PCI_SRIOV_BAR		0x24	/* VF BAR0 */
+#define  PCI_SRIOV_NUM_BARS	6	/* Number of VF BARs */
+#define PCI_SRIOV_VFM		0x3c	/* VF Migration State Array Offset*/
+#define  PCI_SRIOV_VFM_BIR(x)	((x) & 7)	/* State BIR */
+#define  PCI_SRIOV_VFM_OFFSET(x) ((x) & ~7)	/* State Offset */
+#define  PCI_SRIOV_VFM_UA	0x0	/* Inactive.Unavailable */
+#define  PCI_SRIOV_VFM_MI	0x1	/* Dormant.MigrateIn */
+#define  PCI_SRIOV_VFM_MO	0x2	/* Active.MigrateOut */
+#define  PCI_SRIOV_VFM_AV	0x3	/* Active.Available */
+
 #endif /* LINUX_PCI_REGS_H */

[-- Attachment #4: dom0-2.patch --]
[-- Type: text/plain, Size: 2389 bytes --]

# HG changeset patch
# User Yu Zhao <yu.zhao@intel.com>
# Date 1237268742 14400
# Node ID 2629935bf356bb7118f8691a46e90daed77c3b48
# Parent  92730fa710446b2502809faa72bb29fda95ba878
PCI: restore saved SR-IOV state

Restore the volatile registers in the SR-IOV capability after the
D3->D0 transition.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>

diff -r 92730fa71044 -r 2629935bf356 drivers/pci/iov.c
--- a/drivers/pci/iov.c	Tue Mar 17 01:43:05 2009 -0400
+++ b/drivers/pci/iov.c	Tue Mar 17 01:45:42 2009 -0400
@@ -125,6 +125,25 @@
 	dev->sriov = NULL;
 }
 
+static void sriov_restore_state(struct pci_dev *dev)
+{
+	int i;
+	u16 ctrl;
+	struct pci_sriov *iov = dev->sriov;
+
+	pci_read_config_word(dev, iov->pos + PCI_SRIOV_CTRL, &ctrl);
+	if (ctrl & PCI_SRIOV_CTRL_VFE)
+		return;
+
+	for (i = PCI_IOV_RESOURCES; i <= PCI_IOV_RESOURCE_END; i++)
+		pci_update_resource(dev, i);
+
+	pci_write_config_dword(dev, iov->pos + PCI_SRIOV_SYS_PGSIZE, iov->pgsz);
+	pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl);
+	if (iov->ctrl & PCI_SRIOV_CTRL_VFE)
+		msleep(100);
+}
+
 /**
  * pci_iov_init - initialize the IOV capability
  * @dev: the PCI device
@@ -173,3 +192,13 @@
 	return dev->sriov->pos + PCI_SRIOV_BAR +
 		4 * (resno - PCI_IOV_RESOURCES);
 }
+
+/**
+ * pci_restore_iov_state - restore the state of the IOV capability
+ * @dev: the PCI device
+ */
+void pci_restore_iov_state(struct pci_dev *dev)
+{
+	if (dev->is_physfn)
+		sriov_restore_state(dev);
+}
diff -r 92730fa71044 -r 2629935bf356 drivers/pci/pci.c
--- a/drivers/pci/pci.c	Tue Mar 17 01:43:05 2009 -0400
+++ b/drivers/pci/pci.c	Tue Mar 17 01:45:42 2009 -0400
@@ -562,6 +562,8 @@
 	pci_restore_pcix_state(dev);
 	pci_restore_msi_state(dev);
 	pci_restore_msix_state(dev);
+	pci_restore_iov_state(dev);
+
 	return 0;
 }
 
diff -r 92730fa71044 -r 2629935bf356 drivers/pci/pci.h
--- a/drivers/pci/pci.h	Tue Mar 17 01:43:05 2009 -0400
+++ b/drivers/pci/pci.h	Tue Mar 17 01:45:42 2009 -0400
@@ -154,6 +154,7 @@
 extern void pci_iov_release(struct pci_dev *dev);
 extern int pci_iov_resource_bar(struct pci_dev *dev, int resno,
 				enum pci_bar_type *type);
+extern void pci_restore_iov_state(struct pci_dev *dev);
 #else
 static inline int pci_iov_init(struct pci_dev *dev)
 {
@@ -168,4 +169,7 @@
 {
 	return 0;
 }
+static inline void pci_restore_iov_state(struct pci_dev *dev)
+{
+}
 #endif /* CONFIG_PCI_IOV */

[-- Attachment #5: dom0-3.patch --]
[-- Type: text/plain, Size: 2677 bytes --]

# HG changeset patch
# User Yu Zhao <yu.zhao@intel.com>
# Date 1237268873 14400
# Node ID 6b776c705e444562dda66dc0b33fd80eaceb1bfb
# Parent  2629935bf356bb7118f8691a46e90daed77c3b48
PCI: reserve bus range for SR-IOV device

Reserve the bus number range used by the Virtual Function when
pcibios_assign_all_busses() returns true.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>

diff -r 2629935bf356 -r 6b776c705e44 drivers/pci/iov.c
--- a/drivers/pci/iov.c	Tue Mar 17 01:45:42 2009 -0400
+++ b/drivers/pci/iov.c	Tue Mar 17 01:47:53 2009 -0400
@@ -13,6 +13,18 @@
 #include <linux/delay.h>
 #include "pci.h"
 
+
+static inline u8 virtfn_bus(struct pci_dev *dev, int id)
+{
+	return dev->bus->number + ((dev->devfn + dev->sriov->offset +
+				    dev->sriov->stride * id) >> 8);
+}
+
+static inline u8 virtfn_devfn(struct pci_dev *dev, int id)
+{
+	return (dev->devfn + dev->sriov->offset +
+		dev->sriov->stride * id) & 0xff;
+}
 
 static int sriov_init(struct pci_dev *dev, int pos)
 {
@@ -202,3 +214,27 @@
 	if (dev->is_physfn)
 		sriov_restore_state(dev);
 }
+
+/**
+ * pci_iov_bus_range - find bus range used by Virtual Function
+ * @bus: the PCI bus
+ *
+ * Returns max number of buses (exclude current one) used by Virtual
+ * Functions.
+ */
+int pci_iov_bus_range(struct pci_bus *bus)
+{
+	int max = 0;
+	u8 busnr;
+	struct pci_dev *dev;
+
+	list_for_each_entry(dev, &bus->devices, bus_list) {
+		if (!dev->is_physfn)
+			continue;
+		busnr = virtfn_bus(dev, dev->sriov->total - 1);
+		if (busnr > max)
+			max = busnr;
+	}
+
+	return max ? max - bus->number : 0;
+}
diff -r 2629935bf356 -r 6b776c705e44 drivers/pci/pci.h
--- a/drivers/pci/pci.h	Tue Mar 17 01:45:42 2009 -0400
+++ b/drivers/pci/pci.h	Tue Mar 17 01:47:53 2009 -0400
@@ -155,6 +155,7 @@
 extern int pci_iov_resource_bar(struct pci_dev *dev, int resno,
 				enum pci_bar_type *type);
 extern void pci_restore_iov_state(struct pci_dev *dev);
+extern int pci_iov_bus_range(struct pci_bus *bus);
 #else
 static inline int pci_iov_init(struct pci_dev *dev)
 {
@@ -172,4 +173,8 @@
 static inline void pci_restore_iov_state(struct pci_dev *dev)
 {
 }
+static inline int pci_iov_bus_range(struct pci_bus *bus)
+{
+	return 0;
+}
 #endif /* CONFIG_PCI_IOV */
diff -r 2629935bf356 -r 6b776c705e44 drivers/pci/probe.c
--- a/drivers/pci/probe.c	Tue Mar 17 01:45:42 2009 -0400
+++ b/drivers/pci/probe.c	Tue Mar 17 01:47:53 2009 -0400
@@ -976,6 +976,9 @@
 	for (devfn = 0; devfn < 0x100; devfn += 8)
 		pci_scan_slot(bus, devfn);
 
+	/* Reserve buses for SR-IOV capability. */
+	max += pci_iov_bus_range(bus);
+
 	/*
 	 * After performing arch-dependent fixup of the bus, look behind
 	 * all PCI-to-PCI bridges on this bus.

[-- Attachment #6: dom0-4.patch --]
[-- Type: text/plain, Size: 3758 bytes --]

# HG changeset patch
# User Yu Zhao <yu.zhao@intel.com>
# Date 1237270918 14400
# Node ID 3a2d0f486f533f0ef21267b9a1682997e0caf463
# Parent  6b776c705e444562dda66dc0b33fd80eaceb1bfb
PCI: centralize device setup code

Move the device setup stuff into pci_setup_device() which will be used
to setup the Virtual Function later.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>

diff -r 6b776c705e44 -r 3a2d0f486f53 drivers/pci/pci.h
--- a/drivers/pci/pci.h	Tue Mar 17 01:47:53 2009 -0400
+++ b/drivers/pci/pci.h	Tue Mar 17 02:21:58 2009 -0400
@@ -117,6 +117,7 @@
 	pci_bar_mem64,		/* A 64-bit memory BAR */
 };
 
+extern int pci_setup_device(struct pci_dev *dev);
 extern int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,
 				struct resource *res, unsigned int reg);
 extern int pci_resource_bar(struct pci_dev *dev, int resno,
diff -r 6b776c705e44 -r 3a2d0f486f53 drivers/pci/probe.c
--- a/drivers/pci/probe.c	Tue Mar 17 01:47:53 2009 -0400
+++ b/drivers/pci/probe.c	Tue Mar 17 02:21:58 2009 -0400
@@ -683,13 +683,28 @@
  * Initialize the device structure with information about the device's 
  * vendor,class,memory and IO-space addresses,IRQ lines etc.
  * Called at initialisation of the PCI subsystem and by CardBus services.
- * Returns 0 on success and -1 if unknown type of device (not normal, bridge
- * or CardBus).
+ * Returns 0 on success and negative if unknown type of device (not normal,
+ * bridge or CardBus).
  */
-static int pci_setup_device(struct pci_dev * dev)
+int pci_setup_device(struct pci_dev *dev)
 {
 	u32 class;
+	u8 hdr_type;
 
+	if (pci_read_config_byte(dev, PCI_HEADER_TYPE, &hdr_type))
+		return -EIO;
+
+	dev->sysdata = dev->bus->sysdata;
+	dev->dev.parent = dev->bus->bridge;
+	dev->dev.bus = &pci_bus_type;
+	dev->hdr_type = hdr_type & 0x7f;
+	dev->multifunction = !!(hdr_type & 0x80);
+	dev->cfg_size = pci_cfg_space_size(dev);
+	dev->error_state = pci_channel_io_normal;
+
+	/* Assume 32-bit PCI; let 64-bit PCI cards (which are far rarer)
+	   set this higher, assuming the system even supports it.  */
+	dev->dma_mask = 0xffffffff;
 	sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(dev->bus),
 		dev->bus->number, PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
 
@@ -706,7 +721,6 @@
 
 	/* Early fixups, before probing the BARs */
 	pci_fixup_device(pci_fixup_early, dev);
-	class = dev->class >> 8;
 
 	switch (dev->hdr_type) {		    /* header type */
 	case PCI_HEADER_TYPE_NORMAL:		    /* standard header */
@@ -741,7 +755,7 @@
 	default:				    /* unknown header */
 		printk(KERN_ERR "PCI: device %s has unknown header type %02x, ignoring.\n",
 			pci_name(dev), dev->hdr_type);
-		return -1;
+		return -EIO;
 
 	bad:
 		printk(KERN_ERR "PCI: %s: class %x doesn't match header type %02x. Ignoring class.\n",
@@ -823,7 +837,6 @@
 {
 	struct pci_dev *dev;
 	u32 l;
-	u8 hdr_type;
 	int delay = 1;
 
 	if (pci_bus_read_config_dword(bus, devfn, PCI_VENDOR_ID, &l))
@@ -850,29 +863,16 @@
 		}
 	}
 
-	if (pci_bus_read_config_byte(bus, devfn, PCI_HEADER_TYPE, &hdr_type))
-		return NULL;
-
 	dev = kzalloc(sizeof(struct pci_dev), GFP_KERNEL);
 	if (!dev)
 		return NULL;
 
 	dev->bus = bus;
-	dev->sysdata = bus->sysdata;
-	dev->dev.parent = bus->bridge;
-	dev->dev.bus = &pci_bus_type;
 	dev->devfn = devfn;
-	dev->hdr_type = hdr_type & 0x7f;
-	dev->multifunction = !!(hdr_type & 0x80);
 	dev->vendor = l & 0xffff;
 	dev->device = (l >> 16) & 0xffff;
-	dev->cfg_size = pci_cfg_space_size(dev);
-	dev->error_state = pci_channel_io_normal;
 
-	/* Assume 32-bit PCI; let 64-bit PCI cards (which are far rarer)
-	   set this higher, assuming the system even supports it.  */
-	dev->dma_mask = 0xffffffff;
-	if (pci_setup_device(dev) < 0) {
+	if (pci_setup_device(dev)) {
 		kfree(dev);
 		return NULL;
 	}

[-- Attachment #7: dom0-5.patch --]
[-- Type: text/plain, Size: 10256 bytes --]

# HG changeset patch
# User Yu Zhao <yu.zhao@intel.com>
# Date 1237270982 14400
# Node ID 577169901110eb89ff36f1460e152a5c96297bde
# Parent  3a2d0f486f533f0ef21267b9a1682997e0caf463
PCI: add SR-IOV API for Physical Function driver

Add or remove the Virtual Function when the SR-IOV is enabled or
disabled by the device driver. This can happen anytime rather than
only at the device probe stage.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>

diff -r 3a2d0f486f53 -r 577169901110 drivers/pci/iov.c
--- a/drivers/pci/iov.c	Tue Mar 17 02:21:58 2009 -0400
+++ b/drivers/pci/iov.c	Tue Mar 17 02:23:02 2009 -0400
@@ -13,6 +13,7 @@
 #include <linux/delay.h>
 #include "pci.h"
 
+#define VIRTFN_ID_LEN	16
 
 static inline u8 virtfn_bus(struct pci_dev *dev, int id)
 {
@@ -24,6 +25,267 @@
 {
 	return (dev->devfn + dev->sriov->offset +
 		dev->sriov->stride * id) & 0xff;
+}
+
+static struct pci_bus *virtfn_add_bus(struct pci_bus *bus, int busnr)
+{
+	struct pci_bus *child;
+
+	if (bus->number == busnr)
+		return bus;
+
+	child = pci_find_bus(pci_domain_nr(bus), busnr);
+	if (child)
+		return child;
+
+	child = pci_add_new_bus(bus, NULL, busnr);
+	if (!child)
+		return NULL;
+
+	child->subordinate = busnr;
+
+	return child;
+}
+
+static void virtfn_remove_bus(struct pci_bus *bus, int busnr)
+{
+	struct pci_bus *child;
+
+	if (bus->number == busnr)
+		return;
+
+	child = pci_find_bus(pci_domain_nr(bus), busnr);
+	BUG_ON(!child);
+
+	if (list_empty(&child->devices))
+		pci_remove_bus(child);
+}
+
+static int virtfn_add(struct pci_dev *dev, int id)
+{
+	int i;
+	int rc;
+	u64 size;
+	char buf[VIRTFN_ID_LEN];
+	struct pci_dev *virtfn;
+	struct resource *res;
+	struct pci_sriov *iov = dev->sriov;
+
+	virtfn = kzalloc(sizeof(struct pci_dev), GFP_KERNEL);
+	if (!virtfn)
+		return -ENOMEM;
+
+	mutex_lock(&iov->dev->sriov->lock);
+	virtfn->bus = virtfn_add_bus(dev->bus, virtfn_bus(dev, id));
+	if (!virtfn->bus) {
+		kfree(virtfn);
+		mutex_unlock(&iov->dev->sriov->lock);
+		return -ENOMEM;
+	}
+	virtfn->devfn = virtfn_devfn(dev, id);
+	virtfn->vendor = dev->vendor;
+	pci_read_config_word(dev, iov->pos + PCI_SRIOV_VF_DID, &virtfn->device);
+	pci_setup_device(virtfn);
+	virtfn->dev.parent = dev->dev.parent;
+
+	for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) {
+		res = dev->resource + PCI_IOV_RESOURCES + i;
+		if (!res->parent)
+			continue;
+		virtfn->resource[i].name = pci_name(virtfn);
+		virtfn->resource[i].flags = res->flags;
+		size = res->end - res->start + 1;
+		do_div(size, iov->total);
+		virtfn->resource[i].start = res->start + size * id;
+		virtfn->resource[i].end = virtfn->resource[i].start + size - 1;
+		rc = request_resource(res, &virtfn->resource[i]);
+		BUG_ON(rc);
+	}
+
+	pci_device_add(virtfn, virtfn->bus);
+	mutex_unlock(&iov->dev->sriov->lock);
+
+	virtfn->physfn = pci_dev_get(dev);
+	virtfn->is_virtfn = 1;
+
+	pci_bus_add_device(virtfn);
+	sprintf(buf, "virtfn%u", id);
+	rc = sysfs_create_link(&dev->dev.kobj, &virtfn->dev.kobj, buf);
+	if (rc)
+		goto failed1;
+	rc = sysfs_create_link(&virtfn->dev.kobj, &dev->dev.kobj, "physfn");
+	if (rc)
+		goto failed2;
+
+	kobject_uevent(&virtfn->dev.kobj, KOBJ_CHANGE);
+
+	return 0;
+
+failed2:
+	sysfs_remove_link(&dev->dev.kobj, buf);
+failed1:
+	pci_dev_put(dev);
+	mutex_lock(&iov->dev->sriov->lock);
+	pci_remove_bus_device(virtfn);
+	virtfn_remove_bus(dev->bus, virtfn_bus(dev, id));
+	mutex_unlock(&iov->dev->sriov->lock);
+
+	return rc;
+}
+
+static void virtfn_remove(struct pci_dev *dev, int id)
+{
+	char buf[VIRTFN_ID_LEN];
+	struct pci_bus *bus;
+	struct pci_dev *virtfn;
+	struct pci_sriov *iov = dev->sriov;
+
+	bus = pci_find_bus(pci_domain_nr(dev->bus), virtfn_bus(dev, id));
+	if (!bus)
+		return;
+
+	virtfn = pci_get_slot(bus, virtfn_devfn(dev, id));
+	if (!virtfn)
+		return;
+
+	pci_dev_put(virtfn);
+
+	sprintf(buf, "virtfn%u", id);
+	sysfs_remove_link(&dev->dev.kobj, buf);
+	sysfs_remove_link(&virtfn->dev.kobj, "physfn");
+
+	mutex_lock(&iov->dev->sriov->lock);
+	pci_remove_bus_device(virtfn);
+	virtfn_remove_bus(dev->bus, virtfn_bus(dev, id));
+	mutex_unlock(&iov->dev->sriov->lock);
+
+	pci_dev_put(dev);
+}
+
+static int sriov_enable(struct pci_dev *dev, int nr_virtfn)
+{
+	int rc;
+	int i, j;
+	int nres;
+	u16 offset, stride, initial;
+	struct resource *res;
+	struct pci_dev *pdev;
+	struct pci_sriov *iov = dev->sriov;
+
+	if (!nr_virtfn)
+		return 0;
+
+	if (iov->nr_virtfn)
+		return -EINVAL;
+
+	pci_read_config_word(dev, iov->pos + PCI_SRIOV_INITIAL_VF, &initial);
+	if (initial > iov->total ||
+	    (!(iov->cap & PCI_SRIOV_CAP_VFM) && (initial != iov->total)))
+		return -EIO;
+
+	if (nr_virtfn < 0 || nr_virtfn > iov->total ||
+	    (!(iov->cap & PCI_SRIOV_CAP_VFM) && (nr_virtfn > initial)))
+		return -EINVAL;
+
+	pci_write_config_word(dev, iov->pos + PCI_SRIOV_NUM_VF, nr_virtfn);
+	pci_read_config_word(dev, iov->pos + PCI_SRIOV_VF_OFFSET, &offset);
+	pci_read_config_word(dev, iov->pos + PCI_SRIOV_VF_STRIDE, &stride);
+	if (!offset || (nr_virtfn > 1 && !stride))
+		return -EIO;
+
+	nres = 0;
+	for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) {
+		res = dev->resource + PCI_IOV_RESOURCES + i;
+		if (res->parent)
+			nres++;
+	}
+	if (nres != iov->nres) {
+		dev_err(&dev->dev, "not enough MMIO resources for SR-IOV\n");
+		return -ENOMEM;
+	}
+
+	iov->offset = offset;
+	iov->stride = stride;
+
+	if (virtfn_bus(dev, nr_virtfn - 1) > dev->bus->subordinate) {
+		dev_err(&dev->dev, "SR-IOV: bus number out of range\n");
+		return -ENOMEM;
+	}
+
+	if (iov->link != dev->devfn) {
+		pdev = pci_get_slot(dev->bus, iov->link);
+		if (!pdev)
+			return -ENODEV;
+
+		pci_dev_put(pdev);
+
+		if (!pdev->is_physfn)
+			return -ENODEV;
+
+		rc = sysfs_create_link(&dev->dev.kobj,
+					&pdev->dev.kobj, "dep_link");
+		if (rc)
+			return rc;
+	}
+
+	iov->ctrl |= PCI_SRIOV_CTRL_VFE | PCI_SRIOV_CTRL_MSE;
+	pci_block_user_cfg_access(dev);
+	pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl);
+	msleep(100);
+	pci_unblock_user_cfg_access(dev);
+
+	iov->initial = initial;
+	if (nr_virtfn < initial)
+		initial = nr_virtfn;
+
+	for (i = 0; i < initial; i++) {
+		rc = virtfn_add(dev, i);
+		if (rc)
+			goto failed;
+	}
+
+	kobject_uevent(&dev->dev.kobj, KOBJ_CHANGE);
+	iov->nr_virtfn = nr_virtfn;
+
+	return 0;
+
+failed:
+	for (j = 0; j < i; j++)
+		virtfn_remove(dev, j);
+
+	iov->ctrl &= ~(PCI_SRIOV_CTRL_VFE | PCI_SRIOV_CTRL_MSE);
+	pci_block_user_cfg_access(dev);
+	pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl);
+	ssleep(1);
+	pci_unblock_user_cfg_access(dev);
+
+	if (iov->link != dev->devfn)
+		sysfs_remove_link(&dev->dev.kobj, "dep_link");
+
+	return rc;
+}
+
+static void sriov_disable(struct pci_dev *dev)
+{
+	int i;
+	struct pci_sriov *iov = dev->sriov;
+
+	if (!iov->nr_virtfn)
+		return;
+
+	for (i = 0; i < iov->nr_virtfn; i++)
+		virtfn_remove(dev, i);
+
+	iov->ctrl &= ~(PCI_SRIOV_CTRL_VFE | PCI_SRIOV_CTRL_MSE);
+	pci_block_user_cfg_access(dev);
+	pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl);
+	ssleep(1);
+	pci_unblock_user_cfg_access(dev);
+
+	if (iov->link != dev->devfn)
+		sysfs_remove_link(&dev->dev.kobj, "dep_link");
+
+	iov->nr_virtfn = 0;
 }
 
 static int sriov_init(struct pci_dev *dev, int pos)
@@ -128,6 +390,8 @@
 
 static void sriov_release(struct pci_dev *dev)
 {
+	BUG_ON(dev->sriov->nr_virtfn);
+
 	if (dev == dev->sriov->dev)
 		mutex_destroy(&dev->sriov->lock);
 	else
@@ -151,6 +415,7 @@
 		pci_update_resource(dev, i);
 
 	pci_write_config_dword(dev, iov->pos + PCI_SRIOV_SYS_PGSIZE, iov->pgsz);
+	pci_write_config_word(dev, iov->pos + PCI_SRIOV_NUM_VF, iov->nr_virtfn);
 	pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl);
 	if (iov->ctrl & PCI_SRIOV_CTRL_VFE)
 		msleep(100);
@@ -238,3 +503,35 @@
 
 	return max ? max - bus->number : 0;
 }
+
+/**
+ * pci_enable_sriov - enable the SR-IOV capability
+ * @dev: the PCI device
+ *
+ * Returns 0 on success, or negative on failure.
+ */
+int pci_enable_sriov(struct pci_dev *dev, int nr_virtfn)
+{
+	might_sleep();
+
+	if (!dev->is_physfn)
+		return -ENODEV;
+
+	return sriov_enable(dev, nr_virtfn);
+}
+EXPORT_SYMBOL_GPL(pci_enable_sriov);
+
+/**
+ * pci_disable_sriov - disable the SR-IOV capability
+ * @dev: the PCI device
+ */
+void pci_disable_sriov(struct pci_dev *dev)
+{
+	might_sleep();
+
+	if (!dev->is_physfn)
+		return;
+
+	sriov_disable(dev);
+}
+EXPORT_SYMBOL_GPL(pci_disable_sriov);
diff -r 3a2d0f486f53 -r 577169901110 drivers/pci/pci.h
--- a/drivers/pci/pci.h	Tue Mar 17 02:21:58 2009 -0400
+++ b/drivers/pci/pci.h	Tue Mar 17 02:23:02 2009 -0400
@@ -141,6 +141,8 @@
 	u32 cap;		/* SR-IOV Capabilities */
 	u16 ctrl;		/* SR-IOV Control */
 	u16 total;		/* total VFs associated with the PF */
+	u16 initial;		/* initial VFs associated with the PF */
+	u16 nr_virtfn;		/* number of VFs available */
 	u16 offset;		/* first VF Routing ID offset */
 	u16 stride;		/* following VF stride */
 	u32 pgsz;		/* page size for BAR alignment */
diff -r 3a2d0f486f53 -r 577169901110 include/linux/pci.h
--- a/include/linux/pci.h	Tue Mar 17 02:21:58 2009 -0400
+++ b/include/linux/pci.h	Tue Mar 17 02:23:02 2009 -0400
@@ -199,6 +199,7 @@
 	unsigned int	msix_enabled:1;
 	unsigned int	ari_enabled:1;	/* ARI forwarding */
 	unsigned int	is_physfn:1;
+	unsigned int	is_virtfn:1;
 
 	u32		saved_config_space[16]; /* config space saved at suspend time */
 	struct hlist_head saved_cap_space;
@@ -206,7 +207,10 @@
 	int rom_attr_enabled;		/* has display of the rom attribute been enabled? */
 	struct bin_attribute *res_attr[DEVICE_COUNT_RESOURCE]; /* sysfs file for resources */
 #ifdef CONFIG_PCI_IOV
-	struct pci_sriov *sriov;	/* SR-IOV capability related */
+	union {
+		struct pci_sriov *sriov;	/* SR-IOV capability related */
+		struct pci_dev *physfn;	/* the PF this VF is associated with */
+	};
 #endif
 };
 
@@ -829,5 +833,18 @@
 int pci_is_guestdev(struct pci_dev *dev);
 #endif /* CONFIG_PCI_GUESTDEV */
 
+#ifdef CONFIG_PCI_IOV
+extern int pci_enable_sriov(struct pci_dev *dev, int nr_virtfn);
+extern void pci_disable_sriov(struct pci_dev *dev);
+#else
+static inline int pci_enable_sriov(struct pci_dev *dev, int nr_virtfn)
+{
+	return -ENODEV;
+}
+static inline void pci_disable_sriov(struct pci_dev *dev)
+{
+}
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* LINUX_PCI_H */

[-- Attachment #8: dom0-6.patch --]
[-- Type: text/plain, Size: 2407 bytes --]

# HG changeset patch
# User Yu Zhao <yu.zhao@intel.com>
# Date 1237353055 14400
# Node ID 582ec8e86ffff64834e8c77ef6790774352ddc7a
# Parent  577169901110eb89ff36f1460e152a5c96297bde
PCI: pass ARI and SR-IOV device information to the hypervisor

PCIe Alternative Routing-ID Interpretation (ARI) ECN defines the Extended
Function -- a function whose function number is greater than 7 within an
ARI Device. Intel VT-d spec 1.2 section 8.3.2 specifies that the Extended
Function is under the scope of the same remapping unit as the traditional
function. The hypervisor needs to know if a function is Extended Function
so it can find proper DMAR for it.

And section 8.3.3 specifies that the SR-IOV Virtual Function is under the
scope of the same remapping unit as the Physical Function. The hypervisor
also needs to know if a function is the Virtual Function and which Physical
Function it's associated with for same reason.

diff -r 577169901110 -r 582ec8e86fff drivers/xen/core/pci.c
--- a/drivers/xen/core/pci.c	Tue Mar 17 02:23:02 2009 -0400
+++ b/drivers/xen/core/pci.c	Wed Mar 18 01:10:55 2009 -0400
@@ -6,6 +6,7 @@
 #include <linux/init.h>
 #include <linux/pci.h>
 #include <xen/interface/physdev.h>
+#include "../../pci/pci.h"
 
 static int (*pci_bus_probe)(struct device *dev);
 static int (*pci_bus_remove)(struct device *dev);
@@ -15,8 +16,16 @@
 	int r;
 	struct pci_dev *pci_dev = to_pci_dev(dev);
 	struct physdev_manage_pci manage_pci;
+
+	memset(&manage_pci, 0, sizeof(manage_pci));
 	manage_pci.bus = pci_dev->bus->number;
 	manage_pci.devfn = pci_dev->devfn;
+	if (pci_dev->is_virtfn) {
+		manage_pci.is_virtfn = 1;
+		manage_pci.physfn.bus = pci_dev->physfn->bus->number;
+		manage_pci.physfn.devfn = pci_dev->physfn->devfn;
+	} else if (pci_ari_enabled(pci_dev->bus) && PCI_SLOT(pci_dev->devfn))
+		manage_pci.is_extfn = 1;
 
 	r = HYPERVISOR_physdev_op(PHYSDEVOP_manage_pci_add, &manage_pci);
 	if (r && r != -ENOSYS)
diff -r 577169901110 -r 582ec8e86fff include/xen/interface/physdev.h
--- a/include/xen/interface/physdev.h	Tue Mar 17 02:23:02 2009 -0400
+++ b/include/xen/interface/physdev.h	Wed Mar 18 01:10:55 2009 -0400
@@ -178,6 +178,12 @@
     /* IN */
     uint8_t bus;
     uint8_t devfn;
+    unsigned is_extfn:1;
+    unsigned is_virtfn:1;
+    struct {
+        uint8_t bus;
+        uint8_t devfn;
+    } physfn;
 }; 
 
 typedef struct physdev_manage_pci physdev_manage_pci_t;

[-- Attachment #9: Type: text/plain, Size: 138 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel