cuda.core.system.PciInfo#

class cuda.core.system.PciInfo(pci_info_ext: nvml.PciInfoExt_v1, int handle: int)#

PCI information about a GPU device.

Methods

__init__(*args, **kwargs)#

Retrieve the current PCIe link generation.

For Fermi™ or newer fully supported devices.

Retrieve the current PCIe link width.

For Fermi™ or newer fully supported devices.

Retrieve the maximum PCIe link generation supported by this GPU device.

For Fermi™ or newer fully supported devices.

Retrieve the maximum PCIe link generation possible with this device and system.

For Fermi™ or newer fully supported devices.

For example, for a generation 2 PCIe device attached to a generation 1 PCIe bus, the max link generation this function will report is generation 1.

Retrieve the maximum PCIe link width possible with this device and system.

For Fermi™ or newer fully supported devices.

For example, for a device with a 16x PCIe bus width attached to a 8x PCIe system bus this function will report a max link width of 8.

get_pcie_replay_counter(self) → int#

Retrieve the PCIe replay counter.

For Kepler™ or newer fully supported devices.

get_pcie_throughput(self, counter: PcieUtilCounter) → int#

Retrieve PCIe utilization information, in KB/s.

This function is querying a byte counter over a 20ms interval, and thus is the PCIe throughput over that interval.

For Maxwell™ or newer fully supported devices.

This method is not supported in virtual machines running virtual GPU (vGPU).

Attributes

base_class#

The 8-bit PCI base class code

bus#

The bus on which the device resides, 0 to 255

bus_id#

The tuple domain:bus:device.function PCI identifier string

device#

The device’s id on the bus, 0 to 31

device_id#

The PCI device id of the device

domain#

The PCI domain on which the device’s bus resides, 0 to 0xffffffff

sub_class#

The 8-bit PCI sub class code

subsystem_id#

The subsystem device ID

vendor_id#

The PCI vendor id of the device