US 7,454,756 B2Granted

Method, apparatus and system for seamlessly sharing devices amongst virtual machines

AssigneeIntel Corporation

InventorsPhilip R. Lantz, Michael A. Goldsmith, David J. Cowperthwaite, Kiran S. Panesar

Life of this patent

Estimating the term…

Loading patent…

Abstract

A method, apparatus and system are described for seamlessly sharing I/O devices amongst multiple virtual machines (“VMs”) on a host computer. Specifically, according to one embodiment of the invention, the virtual machine manager (“VMM”) on the host cycles access to the I/O devices amongst the VMs according to a round robin or other such allocation scheme. In order to provide direct access to the devices, the VMM may save the device state pertaining to the currently active VM, store the state in a memory region allocated to the currently active VM, retrieve a device state for a new VM from its memory region and restore the device using the retrieved device state, thus providing the illusion that each VM has direct, full-speed, exclusive access to the I/O device.

Claims12

Dependent claims

A method for sharing a physical device amongst at least first and second virtual machines (“VMs”) in a processing system, the method comprising:

configuring the processing system to enable the first VM to directly access the physical device and to utilize full functionality of the physical device;
a predetermined time after configuring the processing system to enable the first VM to directly access the physical device and to utilize full functionality of the physical device, performing operations comprising:
idling the physical device by causing the physical device to complete all pending operations for the first VM and temporarily preventing further interactions between the first VM and the physical device,
saving device state pertaining to the first VM from registers in the physical device to a first memory region associated with the first VM,
loading device state pertaining to the second VM from a second memory region associated with the second VM to the registers in the physical device, and
restarting the physical device to enable the second VM to directly access the physical device and to utilize full functionality of the physical device;
a predetermined time after restarting the physical device for the second VM, performing operations comprising:
idling the physical device,
saving device state pertaining to the second VM from registers in the physical device to the second memory region associated with the second VM,
reloading the saved device state pertaining to the first VM from the first memory region to the registers in the physical device, and
restarting the physical device to enable the first VM to directly access the physical device and to utilize full functionality of the physical device;
suspending the first VM prior to loading the device state for the second VM;
suspending the second VM prior to loading the device state for the first VM;
and
periodically repeating the operations of idling the physical device, saving device state for a current VM, reloading device state for a next VM, and restarting the physical device to enable the next VM to directly access the physical device and to utilize full functionality of the physical device, according to a round robin schedule for the first and second VMs, wherein the predetermined times correspond to a configurable rate for switching between VMs.

12 more elements

An apparatus, comprising:

a tangible, machine-accessible medium;
and
instructions in the machine-accessible medium, wherein the instructions, when executed by a processing system, result in operations comprising:
configuring the processing system to enable a first virtual machine (VM) in the processing system to directly access a physical device in the processing system and to utilize full functionality of the physical device;
a predetermined time after configuring the processing system to enable the first VM to directly access the physical device and to utilize full functionality of the physical device, performing operations comprising:
idling the physical device by causing the physical device to complete all pending operations for the first VM and temporarily preventing further interactions between the first VM and the physical device,
saving device state pertaining to the first VM from registers in the physical device to a first memory region associated with the first VM,
loading device state pertaining to a second VM from a second memory region associated with the second VM to the registers in the physical device, and
restarting the physical device to enable the second VM to directly access the physical device and to utilize full functionality of the physical device;
a predetermined time after restarting the physical device for the second VM, performing operations comprising:
idling the physical device,
saving device state pertaining to the second VM from registers in the physical device to the second memory region associated with the second VM,
reloading the saved device state pertaining to the first VM from the first memory region to the registers in the physical device, and
restarting the physical device to enable the first VM to directly access the physical device and to utilize full functionality of the physical device;
suspending the first VM prior to loading the device state for the second VM;
suspending the second VM prior to loading the device state for the first VM;
and
periodically repeating the operations of idling the physical device, saving device state for a current VM, reloading device state for a next VM, and restarting the physical device to enable the next VM to directly access the physical device and to utilize full functionality of the physical device, according to a round robin schedule for the first and second VMs, wherein the predetermined times correspond to a configurable rate for switching between VMs.

15 more elements

A processing system, comprising:

a processor;
a physical device responsive to the processor;
a tangible, machine-accessible medium responsive to the processor;
and
instructions in the machine-accessible medium, wherein the instructions, when executed by the processing system, result in operations comprising:
configuring the processing system to enable a first virtual machine (VM) in the processing system to directly access the physical device and to utilize full functionality of the physical device;
a predetermined time after configuring the processing system to enable the first VM to directly access the physical device and to utilize full functionality of the physical device, performing operations comprising:
idling the physical device by causing the physical device to complete all pending operations for the first VM and temporarily preventing further interactions between the first VM and the physical device,
saving device state pertaining to the first VM from registers in the physical device to a first memory region associated with the first VM,
loading device state pertaining to a second VM from a second memory region associated with the second VM to the registers in the physical device, and
restarting the physical device to enable the second VM to directly access the physical device and to utilize full functionality of the physical device;
a predetermined time after restarting the physical device for the second VM, performing operations comprising:
idling the physical device,
saving device state pertaining to the second VM from registers in the physical device to the second memory region associated with the second VM,
reloading the saved device state pertaining to the first VM from the first memory region to the registers in the physical device, and
restarting the physical device to enable the first VM to directly access the physical device and to utilize full functionality of the physical device;
suspending the first VM prior to loading the device state for the second VM;
suspending the second VM prior to loading the device state for the first VM;
and
periodically repeating the operations of idling the physical device, saving device state for a current VM, reloading device state for a next VM, and restarting the physical device to enable the next VM to directly access the physical device and to utilize full functionality of the physical device, according to a round robin schedule for the first and second VMs, wherein the predetermined times correspond to a configurable rate for switching between VMs.

17 more elements

Dependencies are inferred from the claim text.