On Thu, 2016-01-21 at 07:51 +0000, Tian, Kevin wrote: > > From: Alex Williamson [mailto:alex.william...@redhat.com] > > Sent: Thursday, January 21, 2016 2:07 AM > > > > vfio-pci currently requires a host= parameter, which comes in the > > form of a PCI address in [domain:] notation. We > > expect to find a matching entry in sysfs for that under > > /sys/bus/pci/devices/. vfio-platform takes a similar approach, but > > defines the host= parameter to be a string, which can be matched > > directly under /sys/bus/platform/devices/. On the PCI side, we have > > some interest in using vfio to expose vGPU devices. These are not > > actual discrete PCI devices, so they don't have a compatible host PCI > > bus address or a device link where QEMU wants to look for it. There's > > also really no requirement that vfio can only be used to expose > > physical devices, a new vfio bus and iommu driver could expose a > > completely emulated device. To fit within the vfio framework, it > > would need a kernel struct device and associated IOMMU group, but > > those are easy constraints to manage. > > > > To support such devices, which would include vGPUs, that honor the > > VFIO PCI programming API, but are not necessarily backed by a unique > > PCI address, add support for specifying any device in sysfs. The > > vfio API already has support for probing the device type to ensure > > compatibility with either vfio-pci or vfio-platform. > > > > With this, a vfio-pci device could either be specified as: > > > > -device vfio-pci,host=02:00.0 > > > > or > > > > -device vfio-pci,sysfsdev=/sys/devices/pci0000:00/0000:00:1c.0/0000:02:00.0 > > > > or even > > > > -device vfio-pci,sysfsdev=/sys/bus/pci/devices/0000:02:00.0 > > > > When vGPU support comes along, this might look something more like: > > > > -device vfio-pci,sysfsdev=/sys/devices/virtual/intel-vgpu/vgpu0@0000:00:02.0 > > > > NB - This is only a made up example path, but it should be noted that > > the device namespace is global for vfio, a virtual device cannot > > overlap with existing namespaces and should not create a name prone to > > conflict, such as a simple instance number. > > > > Thanks Alex! It's a good improvement to support coming vgpu feature. > Just curious. Does the virtual device name has to include a BDF format > or it can be random strings (e.g. just "vgpu0")? In the latter case, then > overlapping chance would be small.
Hi Kevin, Yeah, looking at the vfio code again (vfio_device_get_from_name), as long as the name is unique within the IOMMU group, I think we'll be fine. I expect that vGPUs will create singleton groups, so the namespace constraints I mention above are maybe not a concern. For vendors that can support multiple GPUs, each with vGPUs, userspace will probably want some way to determine the source of a vGPU for load balancing and locality purpose, but that's better handled through parent/child device links in sysfs rather than embedding it in the device name. Thanks, Alex
