xref: /openbmc/qemu/docs/pcie_sriov.txt (revision 9c255cb5)
1PCI SR/IOV EMULATION SUPPORT
2============================
3
4Description
5===========
6SR/IOV (Single Root I/O Virtualization) is an optional extended capability
7of a PCI Express device. It allows a single physical function (PF) to appear as multiple
8virtual functions (VFs) for the main purpose of eliminating software
9overhead in I/O from virtual machines.
10
11QEMU now implements the basic common functionality to enable an emulated device
12to support SR/IOV.
13
14Implementation
15==============
16Implementing emulation of an SR/IOV capable device typically consists of
17implementing support for two types of device classes; the "normal" physical device
18(PF) and the virtual device (VF). From QEMU's perspective, the VFs are just
19like other devices, except that some of their properties are derived from
20the PF.
21
22A virtual function is different from a physical function in that the BAR
23space for all VFs are defined by the BAR registers in the PFs SR/IOV
24capability. All VFs have the same BARs and BAR sizes.
25
26Accesses to these virtual BARs then is computed as
27
28   <VF BAR start> + <VF number> * <BAR sz> + <offset>
29
30From our emulation perspective this means that there is a separate call for
31setting up a BAR for a VF.
32
331) To enable SR/IOV support in the PF, it must be a PCI Express device so
34   you would need to add a PCI Express capability in the normal PCI
35   capability list. You might also want to add an ARI (Alternative
36   Routing-ID Interpretation) capability to indicate that your device
37   supports functions beyond it's "own" function space (0-7),
38   which is necessary to support more than 7 functions, or
39   if functions extends beyond offset 7 because they are placed at an
40   offset > 1 or have stride > 1.
41
42   ...
43   #include "hw/pci/pcie.h"
44   #include "hw/pci/pcie_sriov.h"
45
46   pci_your_pf_dev_realize( ... )
47   {
48      ...
49      int ret = pcie_endpoint_cap_init(d, 0x70);
50      ...
51      pcie_ari_init(d, 0x100);
52      ...
53
54      /* Add and initialize the SR/IOV capability */
55      pcie_sriov_pf_init(d, 0x200, "your_virtual_dev",
56                       vf_devid, initial_vfs, total_vfs,
57                       fun_offset, stride);
58
59      /* Set up individual VF BARs (parameters as for normal BARs) */
60      pcie_sriov_pf_init_vf_bar( ... )
61      ...
62   }
63
64   For cleanup, you simply call:
65
66      pcie_sriov_pf_exit(device);
67
68   which will delete all the virtual functions and associated resources.
69
702) Similarly in the implementation of the virtual function, you need to
71   make it a PCI Express device and add a similar set of capabilities
72   except for the SR/IOV capability. Then you need to set up the VF BARs as
73   subregions of the PFs SR/IOV VF BARs by calling
74   pcie_sriov_vf_register_bar() instead of the normal pci_register_bar() call:
75
76   pci_your_vf_dev_realize( ... )
77   {
78      ...
79      int ret = pcie_endpoint_cap_init(d, 0x60);
80      ...
81      pcie_ari_init(d, 0x100);
82      ...
83      memory_region_init(mr, ... )
84      pcie_sriov_vf_register_bar(d, bar_nr, mr);
85      ...
86   }
87
88Testing on Linux guest
89======================
90The easiest is if your device driver supports sysfs based SR/IOV
91enabling. Support for this was added in kernel v.3.8, so not all drivers
92support it yet.
93
94To enable 4 VFs for a device at 01:00.0:
95
96	modprobe yourdriver
97	echo 4 > /sys/bus/pci/devices/0000:01:00.0/sriov_numvfs
98
99You should now see 4 VFs with lspci.
100To turn SR/IOV off again - the standard requires you to turn it off before you can enable
101another VF count, and the emulation enforces this:
102
103	echo 0 > /sys/bus/pci/devices/0000:01:00.0/sriov_numvfs
104
105Older drivers typically provide a max_vfs module parameter
106to enable it at load time:
107
108	modprobe yourdriver max_vfs=4
109
110To disable the VFs again then, you simply have to unload the driver:
111
112	rmmod yourdriver
113