Updating virtual machine memory by interrupt handler

Patent metadata

Field	Value
Publication number	US-10346330-B2
Application number	US-201414167497-A
Country	US
Kind code	B2
Filing date	Jan 29, 2014
Priority date	Jan 29, 2014
Publication date	Jul 9, 2019
Grant date	Jul 9, 2019

Abstract

Systems and methods for directly updating the virtual machine memory by interrupt handlers. An example method may comprise: receiving, by a computer system, an interrupt triggered by a physical device; receiving, by an interrupt handling routine, a data frame from the physical device; identifying a virtual machine to receive the interrupt; and responsive to determining that an active memory context on the computer system matches a memory context of the virtual machine, writing, by the interrupt handling routine, the data frame into a memory of the virtual machine.

First claim

What is claimed is:

1. A method, comprising:
receiving, by a processor of a host computer system, an interrupt triggered by a physical device;
receiving, by an interrupt handling routine, a data frame from the physical device;
identifying, by the processor, a paravirtualized input/output (I/O) device to receive the interrupt;
responsive to determining that an active memory context associated with the processor matches a memory context of a virtual machine associated with the paravirtualized I/O device, writing the data frame into a memory of the virtual machine; and
responsive to determining that a page fault has been triggered by the writing operation, queuing the data frame for processing by an I/O processing thread and waking up the I/O processing thread.

2. The method of claim 1 , wherein the physical device is provided by one of: a network interface card or a hard disk controller.

3. The method of claim 1 , further comprising:
enabling page fault handling by the host computer system responsive to determining that no page fault has been triggered by the write operation.

4. The method of claim 1 , further comprising:
responsive to determining that the active memory context associated with the processor does not match the memory context of the virtual machine, queuing the data frame for processing by the I/O processing thread.

5. A system comprising:
a memory; and
a processor, operatively coupled to the memory, to:
receive an interrupt triggered by a physical device;
receive, by an interrupt handling routine, a data frame from the physical device;
identify a paravirtualized input/output (I/O) device to receive the interrupt;
responsive to determining that an active memory context associated with the processor matches a memory context of a virtual machine associated with the paravirtualized I/O device, disable page fault handling by the virtual machine and write the data frame into a memory of the virtual machine; and
responsive to determining that the active memory context associated with the processor does not match the memory context of the virtual machine, queue the data frame for processing by an I/O processing thread and wake up the I/O processing thread.

6. The system of claim 5 , wherein the physical device is provided by one of: a network interface card or a hard disk controller.

7. The system of claim 5 , wherein the processor is further to:
enable page fault handling by the virtual machine responsive to determining that no page fault has been triggered by the write operation.

8. A computer-readable non-transitory storage medium comprising executable instructions that, when executed by a processor, cause the processor to:
receive, by the processor, an interrupt triggered by a physical device;
receive, by an interrupt handling routine, a data frame from the physical device;
identify, by the processor, a paravirtualized input/output (I/O) device to receive the interrupt;
responsive to determining that an active memory context associated with the processor matches a memory context of a virtual machine associated with the paravirtualized I/O device, write the data frame into a memory of the virtual machine; and
responsive to determining that a page fault has been triggered by the writing operation, queue the data frame for processing by an I/O processing thread and wake up the I/O processing thread.

9. The computer-readable non-transitory storage medium of claim 8 , wherein the physical device is provided by one of: a network interface card or a hard disk controller.

10. The computer-readable non-transitory storage medium of claim 8 , further comprising executable instructions causing the processor to:
enable page fault handling by the virtual machine responsive to determining that no page fault has been triggered by the write operation.

11. The computer-readable non-transitory storage medium of claim 8 , further comprising executable instructions causing the processor to:
responsive to determining that the active memory context associated with the processor does not match the memory context of the virtual machine, queue the data frame for processing by the I/O processing thread.

12. The method of claim 1 , further comprising:
notifying the virtual machine of the data frame.

13. The method of claim 1 , wherein the active memory context associated with the processor comprises a page table utilized for translating virtual addresses to physical addresses.

14. The method of claim 1 , wherein identifying the virtual machine comprises identifying a network interface card identified by a Media Access Control (MAC) address of the data frame.

15. The method of claim 1 , wherein writing the data frame into a memory of the virtual machine is performed by the interrupt handling routine.

16. The method of claim 1 , further comprising:
responsive to determining that the active memory context associated with the processor matches the memory context of the virtual machine, disabling page fault handling by the virtual machine thus causing the host computer system to handle page faults.

17. The method of claim 1 , further comprising:
responsive to determining that no page fault has been triggered by writing the data frame into the memory of the virtual machine, injecting an interrupt into a virtual processor associated with the virtual machine to notify the virtual machine of the data frame.

18. The computer-readable non-transitory storage medium of claim 8 , further comprising executable instructions to cause the processor to:
responsive to determining that the active memory context associated with the processor matches the memory context of the virtual machine, disable page fault handling by the virtual machine.

19. The computer-readable non-transitory storage medium of claim 8 , further comprising executable instructions to cause the processor to:
inject an interrupt into a virtual processor associated with the virtual machine to notify the virtual machine of the data frame.

20. The computer-readable non-transitory storage medium of claim 8 , wherein the active memory context comprises a page table utilized for translating virtual addresses to physical addresses.

Assignees

• Red Hat Israel Ltd

Inventors

• Tsirkin Michael
• Van Riel Henri Han

Classifications

• G06F2213/2424

  Interrupt packet, e.g. event · CPC title

• G06F9/4812

  by interrupt, e.g. masked · CPC title

• G06F9/5077

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• G06F9/5027

  the resource being a machine, e.g. CPUs, Servers, Terminals · CPC title

• G06F13/34 Primary

  with priority control · CPC title