From 00d879937512b24db576a39fc94c78b427f8916d Mon Sep 17 00:00:00 2001
Message-Id: <00d879937512b24db576a39fc94c78b427f8916d@dist-git>
From: Martin Kletzander <mkletzan@redhat.com>
Date: Mon, 23 May 2016 18:10:04 +0200
Subject: [PATCH] nodedev: Expose PCI header type
RHEL-7.2.z: https://bugzilla.redhat.com/show_bug.cgi?id=1331328
Upstream: https://bugzilla.redhat.com/show_bug.cgi?id=1317531
If we expose this information, which is one byte in every PCI config
file, we let all mgmt apps know whether the device itself is an endpoint
or not so it's easier for them to decide whether such device can be
passed through into a VM (endpoint) or not (*-bridge).
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
(cherry picked from commit d77ffb6876e87a5c6f4c74c49cf0d89ade4f8326)
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
---
docs/formatnode.html.in | 36 ++++++++++++++--------
docs/schemas/nodedev.rng | 11 +++++++
src/conf/node_device_conf.c | 20 ++++++++++++
src/conf/node_device_conf.h | 1 +
src/libvirt_private.syms | 3 ++
src/node_device/node_device_udev.c | 3 ++
src/util/virpci.c | 33 ++++++++++++++++++++
src/util/virpci.h | 12 ++++++++
.../pci_0000_00_02_0_header_type.xml | 15 +++++++++
.../pci_0000_00_1c_0_header_type.xml | 20 ++++++++++++
tests/nodedevxml2xmltest.c | 2 ++
11 files changed, 144 insertions(+), 12 deletions(-)
create mode 100644 tests/nodedevschemadata/pci_0000_00_02_0_header_type.xml
create mode 100644 tests/nodedevschemadata/pci_0000_00_1c_0_header_type.xml
diff --git a/docs/formatnode.html.in b/docs/formatnode.html.in
index 3ff1bef..a14bcfd 100644
--- a/docs/formatnode.html.in
+++ b/docs/formatnode.html.in
@@ -97,18 +97,30 @@
<dd>
This optional element can occur multiple times. If it
exists, it has a mandatory <code>type</code> attribute
- which will be set to
- either <code>physical_function</code>
- or <code>virtual_functions</code>. If the type
- is <code>physical_function</code>, there will be a
- single <code>address</code> subelement which contains
- the PCI address of the SRIOV Physical Function (PF)
- that is the parent of this device (and this device is,
- by implication, an SRIOV Virtual Function (VF)). If
- the type is <code>virtual_functions</code>, then this
- device is an SRIOV PF, and the capability element will
- have a list of <code>address</code> subelements, one
- for each VF on this PF.
+ which will be set to:
+ <dl>
+ <dt><code>physical_function</code></dt>
+ <dd>
+ That means there will be a single <code>address</code>
+ subelement which contains the PCI address of the SRIOV
+ Physical Function (PF) that is the parent of this device
+ (and this device is, by implication, an SRIOV Virtual
+ Function (VF)).
+ </dd>
+ <dt><code>virtual_function</code></dt>
+ <dd>
+ In this case this device is an SRIOV PF, and the capability
+ element will have a list of <code>address</code>
+ subelements, one for each VF on this PF.
+ </dd>
+ <dt><code>pci-bridge</code> or <code>cardbus-bridge</code></dt>
+ <dd>
+ This shows merely that the lower 7 bits of PCI header type
+ have either value of 1 or 2 respectively. Usually this
+ means such device cannot be used for PCI passthrough.
+ <span class="since">Since 1.3.3</span>
+ </dd>
+ </dl>
</dd>
<dt><code>numa</code></dt>
<dd>
diff --git a/docs/schemas/nodedev.rng b/docs/schemas/nodedev.rng
index 744dccd..949811c 100644
--- a/docs/schemas/nodedev.rng
+++ b/docs/schemas/nodedev.rng
@@ -169,6 +169,17 @@
</optional>
<optional>
+ <element name='capability'>
+ <attribute name='type'>
+ <choice>
+ <value>pci-bridge</value>
+ <value>cardbus-bridge</value>
+ </choice>
+ </attribute>
+ </element>
+ </optional>
+
+ <optional>
<element name='pci-express'>
<zeroOrMore>
<element name='link'>
diff --git a/src/conf/node_device_conf.c b/src/conf/node_device_conf.c
index e6f3f27..feefb9a 100644
--- a/src/conf/node_device_conf.c
+++ b/src/conf/node_device_conf.c
@@ -394,6 +394,12 @@ char *virNodeDeviceDefFormat(const virNodeDeviceDef *def)
if (data->pci_dev.numa_node >= 0)
virBufferAsprintf(&buf, "<numa node='%d'/>\n",
data->pci_dev.numa_node);
+
+ if (data->pci_dev.hdrType) {
+ virBufferAsprintf(&buf, "<capability type='%s'/>\n",
+ virPCIHeaderTypeToString(data->pci_dev.hdrType));
+ }
+
if (data->pci_dev.flags & VIR_NODE_DEV_CAP_FLAG_PCIE)
virPCIEDeviceInfoFormat(&buf, data->pci_dev.pci_express);
break;
@@ -1264,6 +1270,7 @@ virNodeDevCapPCIDevParseXML(xmlXPathContextPtr ctxt,
xmlNodePtr orignode, iommuGroupNode, pciExpress;
int ret = -1;
virPCIEDeviceInfoPtr pci_express = NULL;
+ char *tmp = NULL;
orignode = ctxt->node;
ctxt->node = node;
@@ -1321,6 +1328,18 @@ virNodeDevCapPCIDevParseXML(xmlXPathContextPtr ctxt,
_("invalid NUMA node ID supplied for '%s'")) < 0)
goto out;
+ if ((tmp = virXPathString("string(./capability[1]/@type)", ctxt))) {
+ int hdrType = virPCIHeaderTypeFromString(tmp);
+
+ if (hdrType <= 0) {
+ virReportError(VIR_ERR_INTERNAL_ERROR,
+ _("Unknown PCI header type '%s'"), tmp);
+ goto out;
+ }
+
+ data->pci_dev.hdrType = hdrType;
+ }
+
if ((pciExpress = virXPathNode("./pci-express[1]", ctxt))) {
if (VIR_ALLOC(pci_express) < 0)
goto out;
@@ -1335,6 +1354,7 @@ virNodeDevCapPCIDevParseXML(xmlXPathContextPtr ctxt,
ret = 0;
out:
+ VIR_FREE(tmp);
virPCIEDeviceInfoFree(pci_express);
ctxt->node = orignode;
return ret;
diff --git a/src/conf/node_device_conf.h b/src/conf/node_device_conf.h
index 7dd39ca..73d2a7f 100644
--- a/src/conf/node_device_conf.h
+++ b/src/conf/node_device_conf.h
@@ -118,6 +118,7 @@ typedef struct _virNodeDevCapData {
unsigned int iommuGroupNumber;
int numa_node;
virPCIEDeviceInfoPtr pci_express;
+ int hdrType; /* enum virPCIHeaderType or -1 */
} pci_dev;
struct {
unsigned int bus;
diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms
index 86909c1..0db30be 100644
--- a/src/libvirt_private.syms
+++ b/src/libvirt_private.syms
@@ -1976,11 +1976,14 @@ virPCIDeviceSetUsedBy;
virPCIDeviceUnbind;
virPCIDeviceWaitForCleanup;
virPCIEDeviceInfoFree;
+virPCIGetHeaderType;
virPCIGetNetName;
virPCIGetPhysicalFunction;
virPCIGetVirtualFunctionIndex;
virPCIGetVirtualFunctionInfo;
virPCIGetVirtualFunctions;
+virPCIHeaderTypeFromString;
+virPCIHeaderTypeToString;
virPCIIsVirtualFunction;
diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c
index aaee0e5..6bff5ba 100644
--- a/src/node_device/node_device_udev.c
+++ b/src/node_device/node_device_udev.c
@@ -506,6 +506,9 @@ static int udevProcessPCI(struct udev_device *device,
/* We need to be root to read PCI device configs */
if (priv->privileged) {
+ if (virPCIGetHeaderType(pciDev, &data->pci_dev.hdrType) < 0)
+ goto out;
+
if (virPCIDeviceIsPCIExpress(pciDev) > 0) {
if (VIR_ALLOC(pci_express) < 0)
goto out;
diff --git a/src/util/virpci.c b/src/util/virpci.c
index 35b1459..d98db3f 100644
--- a/src/util/virpci.c
+++ b/src/util/virpci.c
@@ -55,6 +55,12 @@ VIR_LOG_INIT("util.pci");
VIR_ENUM_IMPL(virPCIELinkSpeed, VIR_PCIE_LINK_SPEED_LAST,
"", "2.5", "5", "8")
+VIR_ENUM_IMPL(virPCIHeader, VIR_PCI_HEADER_LAST,
+ "endpoint",
+ "pci-bridge",
+ "cardbus-bridge",
+);
+
struct _virPCIDevice {
unsigned int domain;
unsigned int bus;
@@ -2872,6 +2878,33 @@ virPCIDeviceGetLinkCapSta(virPCIDevicePtr dev,
}
+int virPCIGetHeaderType(virPCIDevicePtr dev, int *hdrType)
+{
+ int fd;
+ uint8_t type;
+
+ *hdrType = -1;
+
+ if ((fd = virPCIDeviceConfigOpen(dev, true)) < 0)
+ return -1;
+
+ type = virPCIDeviceRead8(dev, fd, PCI_HEADER_TYPE);
+
+ virPCIDeviceConfigClose(dev, fd);
+
+ type &= PCI_HEADER_TYPE_MASK;
+ if (type >= VIR_PCI_HEADER_LAST) {
+ virReportError(VIR_ERR_INTERNAL_ERROR,
+ _("Unknown PCI header type '%d'"), type);
+ return -1;
+ }
+
+ *hdrType = type;
+
+ return 0;
+}
+
+
void
virPCIEDeviceInfoFree(virPCIEDeviceInfoPtr dev)
{
diff --git a/src/util/virpci.h b/src/util/virpci.h
index 64b9e96..1ed59dd 100644
--- a/src/util/virpci.h
+++ b/src/util/virpci.h
@@ -52,6 +52,16 @@ typedef enum {
VIR_ENUM_DECL(virPCIELinkSpeed)
+typedef enum {
+ VIR_PCI_HEADER_ENDPOINT = 0,
+ VIR_PCI_HEADER_PCI_BRIDGE,
+ VIR_PCI_HEADER_CARDBUS_BRIDGE,
+
+ VIR_PCI_HEADER_LAST
+} virPCIHeaderType;
+
+VIR_ENUM_DECL(virPCIHeader)
+
typedef struct _virPCIELink virPCIELink;
typedef virPCIELink *virPCIELinkPtr;
struct _virPCIELink {
@@ -214,6 +224,8 @@ int virPCIDeviceGetLinkCapSta(virPCIDevicePtr dev,
unsigned int *sta_speed,
unsigned int *sta_width);
+int virPCIGetHeaderType(virPCIDevicePtr dev, int *hdrType);
+
void virPCIEDeviceInfoFree(virPCIEDeviceInfoPtr dev);
#endif /* __VIR_PCI_H__ */
diff --git a/tests/nodedevschemadata/pci_0000_00_02_0_header_type.xml b/tests/nodedevschemadata/pci_0000_00_02_0_header_type.xml
new file mode 100644
index 0000000..5150fd1
--- /dev/null
+++ b/tests/nodedevschemadata/pci_0000_00_02_0_header_type.xml
@@ -0,0 +1,15 @@
+<device>
+ <name>pci_0000_00_02_0</name>
+ <parent>computer</parent>
+ <capability type='pci'>
+ <domain>0</domain>
+ <bus>0</bus>
+ <slot>2</slot>
+ <function>0</function>
+ <product id='0x0416'>4th Gen Core Processor Integrated Graphics Controller</product>
+ <vendor id='0x8086'>Intel Corporation</vendor>
+ <iommuGroup number='1'>
+ <address domain='0x0000' bus='0x00' slot='0x02' function='0x0'/>
+ </iommuGroup>
+ </capability>
+</device>
diff --git a/tests/nodedevschemadata/pci_0000_00_1c_0_header_type.xml b/tests/nodedevschemadata/pci_0000_00_1c_0_header_type.xml
new file mode 100644
index 0000000..dea5f05
--- /dev/null
+++ b/tests/nodedevschemadata/pci_0000_00_1c_0_header_type.xml
@@ -0,0 +1,20 @@
+<device>
+ <name>pci_0000_00_1c_0</name>
+ <parent>computer</parent>
+ <capability type='pci'>
+ <domain>0</domain>
+ <bus>0</bus>
+ <slot>28</slot>
+ <function>0</function>
+ <product id='0x8c10'>8 Series/C220 Series Chipset Family PCI Express Root Port #1</product>
+ <vendor id='0x8086'>Intel Corporation</vendor>
+ <iommuGroup number='8'>
+ <address domain='0x0000' bus='0x00' slot='0x1c' function='0x0'/>
+ </iommuGroup>
+ <capability type='pci-bridge'/>
+ <pci-express>
+ <link validity='cap' port='1' speed='5' width='1'/>
+ <link validity='sta' speed='2.5' width='1'/>
+ </pci-express>
+ </capability>
+</device>
diff --git a/tests/nodedevxml2xmltest.c b/tests/nodedevxml2xmltest.c
index a37d729..d2cd444 100644
--- a/tests/nodedevxml2xmltest.c
+++ b/tests/nodedevxml2xmltest.c
@@ -91,6 +91,8 @@ mymain(void)
DO_TEST("usb_device_1d6b_1_0000_00_1d_0");
DO_TEST("pci_8086_4238_pcie_wireless");
DO_TEST("pci_8086_0c0c_snd_hda_intel");
+ DO_TEST("pci_0000_00_02_0_header_type");
+ DO_TEST("pci_0000_00_1c_0_header_type");
return ret == 0 ? EXIT_SUCCESS : EXIT_FAILURE;
}
--
2.8.3