IP Library › Granted Patent US 9,910,699
Granted Patent B2
US 9,910,699 · App. 14/525,959 · Granted Mar 6, 2018

Virtual processor direct interrupt delivery mechanism

Inventors: Rajesh M. Sankaran (Portland, OR); Gilbert Neiger (Hillsboro, OR)
Assignee: INTEL CORPORATION
G06F9/4812G06F9/45533G06F13/24
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,910,699
App. No.
14/525,959
Granted
Mar 6, 2018
Kind
B2
Abstract

A method comprising is described. The method includes receiving an interrupt targeting a virtual processor, determining a status of the virtual processor and directly delivering the interrupt to the virtual processor upon determining that the virtual processor is operating in a running state.

Claims (15)

1. A method of processing external interrupts for virtual processors using a host processor, the method comprising:

configuring a virtual machine monitor (VMM) in a host processor, the host processor having at least one logical processor and the VMM controlling a plurality of virtual processors for the at least one logical processor, the controlling including:

allocating at least two physical interrupt vectors in the host processor, including a running notification vector (RNV) for posting interrupts to running virtual processors and a blocked notification vector (BNV) for posting interrupts to blocked virtual processors;

masking interrupts posted to preempted virtual processors, the preempted virtual processors including previously blocked virtual processors having at least one posted interrupt and previously running virtual processors; and

enabling delivery of notifications for masked interrupts using the RNV upon transitioning a preempted virtual processor to a running virtual processor.

2. The method of claim 1 , wherein enabling delivery of notifications for masked interrupts using the RNV includes:

issuing a self inter-processor interrupt (self-IPI) to the VMM;

detecting the self-IPI at the host processor upon transitioning a preempted virtual processor into a running virtual processor;

performing posted interrupt processing at the host processor upon detecting the self-IPI, including moving any masked interrupts to the virtual interrupt controller for delivery of notifications.

3. The method of claim 1 , further comprising:

receiving a notification event from the BNV for a blocked virtual processor; and

transitioning the blocked virtual processor into a preempted virtual processor.

4. The method of claim 1 , further comprising:

migrating interrupts posted for preempted virtual processors being migrated to a new destination logical processor, including updating any posted interrupts for the migrated virtual processors to a new destination interrupt controller for the new destination logical processor; and

queueing the migrated virtual processors to the preempted list of preempted virtual processors on the new destination logical processor.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 28, 2014
From: SANKARAN, RAJESH M; NEIGER, GILBERT
To: INTEL CORPORATION
Reel/Frame 034053/0210 →
Continuity (1)
Related Publication 20160117190A1 · Apr 28, 2016