초록 ▼

Various operations are disclosed for improving the operational efficiency of interrupt handling in a virtualized environment. A virtualized interrupt controller may obviate the need for an explicit end-of-interrupt command by providing an automatic EOI capability even when a physical interrupt cont

Various operations are disclosed for improving the operational efficiency of interrupt handling in a virtualized environment. A virtualized interrupt controller may obviate the need for an explicit end-of-interrupt command by providing an automatic EOI capability even when a physical interrupt controller offers no such mechanism. The use of a message pending bit for inter-partition communications facilitates avoiding an EOI command of inter-processor interrupts used in inter-partition communications whenever no further messages are cued for a particular message slot. A virtualized interrupt controller facilitates the selective EOI of an interrupt even when it is not the highest priority in-service interrupt irrespective of whether a physical interrupt controller provides such functionality.

대표청구항 ▼

What is claimed: 1. A method of processing interrupts for a virtualized computing system, the virtualized computing system comprising a virtual machine monitor and at least one guest operating system, the method comprising: receiving a first interrupt request at a virtualized interrupt controller;

What is claimed: 1. A method of processing interrupts for a virtualized computing system, the virtualized computing system comprising a virtual machine monitor and at least one guest operating system, the method comprising: receiving a first interrupt request at a virtualized interrupt controller; and clearing an interrupt service flag corresponding to the first interrupt request substantially immediately after the first interrupt request is received. 2. The method of claim 1, further comprising performing an action prescribed by the first interrupt and wherein clearing an interrupt service flag corresponding to the first interrupt request substantially immediately after the first interrupt request is received comprises clearing the interrupt service flag corresponding to the first interrupt request before the action prescribed by the first interrupt has been completely performed. 3. The method of claim 1, further comprising: performing an action prescribed by the first interrupt; and permitting the delivery of other interrupts while performing the action prescribed by the first interrupt. 4. The method of claim 3, wherein performing an action prescribed by the first interrupt is mediated by a virtual machine monitor. 5. The method of claim 1, wherein the first interrupt request corresponds to a request for a first inter-partition message. 6. The method of claim 5, further comprising: providing at least one message slot, each said message slot being associated with a virtual processor; and setting a message-pending flag associated with the first inter-partition message if a second inter-partition message is queued for the message slot associated with a processor of the first interrupt request before processing of the first inter-partition message has been completed. 7. The method of claim 6, further comprising: sending an end-of-message command to the virtualized interrupt controller after processing an inter-partition message is completed if a message-pending flag associated with the inter-partition message is set. 8. A method of processing interrupts for a virtualized computing system, the virtualized computing system comprising a virtual machine monitor and at least one guest operating system, the method comprising: receiving an end-of-interrupt command from a guest operating system; determining that the end-of-interrupt command does not correspond to a highest priority interrupt flagged as being in-service; recording information identifying the interrupt corresponding to the received end-of-interrupt command; and processing the end-of-interrupt command after an end-of-interrupt command for the highest priority interrupt flagged as being in-service is received. 9. The method of claim 8, further comprising: processing end-of-interrupt commands for interrupts corresponding to the recorded information upon completion of processing an end-of-interrupt command for a highest-priority in-service interrupt. 10. A system, comprising: a virtual machine monitor; and a virtualized interrupt controller for clearing an interrupt service flag corresponding to an interrupt request substantially immediately after the interrupt request is received. 11. The system of claim 10, wherein the virtualized interrupt controller clears an interrupt service flag corresponding to an interrupt request before an action prescribed by the interrupt request is completely performed. 12. The system of claim 10, wherein the virtual machine monitor flags a first inter-partition message when a second inter-partition message is queued before processing of the first inter-partition message is completed and wherein the virtualized interrupt controller is signaled when processing of the first inter-partition message is completed only if the first inter-partition message is flagged. 13. A system comprising: a virtual machine monitor; and a virtualized interrupt controller, wherein the virtualized interrupt controller records information identifying an interrupt corresponding to a received end-of-interrupt command to enable processing the received end-of-interrupt command after processing of a higher priority in-service interrupt is completed. 14. A computer readable medium comprising computer executable instructions for managing interrupts, the instructions comprising instructions for clearing an interrupt service flag corresponding to an interrupt service request substantially immediately after the interrupt service request is received by a virtualized interrupt controller. 15. The computer readable medium of claim 14, further comprising computer executable instructions for performing an action prescribed by an interrupt service request while permitting the delivery of other interrupt service requests. 16. The computer readable medium of claim 14, further comprising computer executable instructions for facilitating inter-processor communication comprising: instructions for setting a message-pending flag associated with a first inter-processor message if a second inter-processor message is queued for a processor receiving the first inter-processor message before processing of the first inter-processor message is completed. 17. The computer readable medium of claim 16, wherein the message-pending flag associated with a first inter-processor message is embodied as a bit in a header of the first inter-processor message. 18. The computer readable medium of claim 16, further comprising computer executable instructions for signaling a virtualized interrupt controller after processing an inter-partition message is completed if a message-pending flag associated with the inter-partition message has been set. 19. A computer readable medium comprising computer executable instructions for managing interrupts, the instructions comprising instructions for queuing an end-of-interrupt command for later processing if the end-of-interrupt command does not correspond to a highest priority interrupt that is being processed when the end-of-interrupt command is received. 20. The computer readable medium of claim 19, further comprising instructions for processing a queued end-of-interrupt command after processing an end-of-interrupt command corresponding to an interrupt of a higher priority than the interrupt corresponding to the queued end-of-interrupt command.